Donate
Independent, objective, nonpartisan research

R 407AGR

Authors

R 407AGR

Tagged with:

Publication PDFs

Database

This is the content currently stored in the post and postmeta tables.

View live version

object(Timber\Post)#3742 (44) { ["ImageClass"]=> string(12) "Timber\Image" ["PostClass"]=> string(11) "Timber\Post" ["TermClass"]=> string(11) "Timber\Term" ["object_type"]=> string(4) "post" ["custom"]=> array(5) { ["_wp_attached_file"]=> string(12) "R_407AGR.pdf" ["wpmf_size"]=> string(7) "1277235" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(137366) "Public Policy Institute of California California’s Commitment to Adult English Learners: Caught Between Funding and Need ••• Arturo Gonzalez California’s Commitment to Adult English Learners: Caught Between Funding and Need ••• Arturo Gonzalez with research support from Rebecca Steinbach and Shannon McConville 2007 Library of Congress Cataloging-in-Publication Data Gonzalez, Arturo. California’s commitment to adult English learners: caught between funding and need / Arturo Gonzalez. p. cm. Includes bibliographical references. ISBN 978-1-58213-123-8 1. Adult education—California—Finance. 2. English language— Study and teaching—Foreign speakers. I. Title. LC5252.C2G66 2007 374.9794—dc22 2007007564 Copyright © 2007 by Public Policy Institute of California All rights reserved San Francisco, CA Short sections of text, not to exceed three paragraphs, may be quoted without written permission provided that full attribution is given to the source and the above copyright notice is included. PPIC does not take or support positions on any ballot measure or on any local, state, or federal legislation, nor does it endorse, support, or oppose any political parties or candidates for public office. Research publications reflect the views of the authors and do not necessarily reflect the views of the staff, officers, or Board of Directors of the Public Policy Institute of California. Summary The ability to speak English is a fundamental prerequisite for the integration of immigrants into California’s society. However, an unprecedented number of immigrants may be relying on English as a Second Language (ESL) courses provided free by the state, suggesting a growing need for English instruction. These courses are the primary means by which the state helps integrate immigrants, and free English courses for adults are provided as part of the state’s adult education mission. They have been offered by various agencies since the 1850s on the belief that an English-proficient immigrant population is valuable from a social, cultural, and economic perspective, both for immigrants and for the state. Although providing classes is a statewide mission, it is one carried out at the local level by adult schools, community colleges, libraries, and community organizations, among others. All of these agencies administer their programs independently and receive funding in drastically different ways. Adult schools, administered by local school districts, are the largest providers of free ESL classes in the state, teaching more than 75 percent of the state’s ESL students, and they receive their funding directly from the state general fund. Adult schools face two important financial constraints: revenue limits and a 2.5 percent limit on the growth of funded enrollment. The 2.5 percent growth limit has been in place since the 1970s, although the linguistic landscape has changed dramatically since then. For instance, between 1980 and 2000, the number of adult immigrants who could benefit from ESL courses increased by nearly 6 percent per year. Local school districts strive to serve the mission set forth by the state, but when the demand for English courses exceeds their funding level, their adult schools must either turn away students or absorb the cost of enrolling the excess. This report examines whether the funding formula for adult schools can meet the changing language needs of the state’s immigrant population. In particular, the report explores the inconsistency between the challenges faced by adult schools wanting to teach a growing number of English learners and a funding formula that may limit the fulfillment of this mission. How do adult schools cope with these competing policies? iii This report focuses on the three components of this potential conflict: the educational commitment of local providers, particularly adult schools; the English-language needs of immigrants; and the funding formula’s effect on adult school enrollments. The report also considers the consequences resulting from this misalignment of policies, the efforts to reform the funding formula, and the tradeoffs associated with increasing resources to the local providers of ESL courses. Main Findings ESL courses are primarily provided by adult schools and community colleges. This dual-provider system makes one agency—the school district or the community college district—the main provider of adult education in a local area, although libraries and community organizations also play an important, although lesser, role. However, adult schools teach 75 percent of all ESL students statewide, and ESL students make up over 40 percent of all adult school students. Demographic trends suggest a growing need for English instruction but one that varies throughout the state. The ESL target population—adults ages 18 and older not proficient in English and not enrolled in school— increased from 900,000 in 1980 to 2.7 million in 2000, an annual growth rate of 5.7 percent per year. The ESL target population grew faster outside Los Angeles County, nearly 7 percent per year. The Los Angeles perimeter counties (Orange, Riverside, San Bernardino, and Ventura) had the highest growth of any region (nearly 8%) over this 20-year period. Predicted levels of enrollment reveal that immigrants outside Los Angeles County have the highest likelihood of enrollment in ESL. The predicted statewide annual growth in enrollment is 5 percent, excluding Los Angeles County, and 3.5 percent including it. The 2.5 percent limit on the growth of funding hinders the majority of adult school programs and, therefore, the number of ESL classes offered. Close to 60 percent of adult schools in California exceed their funding limit. In 2004–05, the dollar value of enrollment in excess of state funding totaled $15.7 million. And among adult schools that do overenroll students, nearly 80 percent exceed their funding limit by over 2.5 percent. iv Policy Implications The fundamental flaw in the funding formula for adult schools is its inability to keep pace with and adapt to the changing English-language needs of communities throughout the state. A recent reform of the adult school funding formula (Assembly Bill 23), which redistributes unused adult education funds to high-demand districts, is a first step toward remedying the variation in need for ESL in the state. Yet, the 2.5 percent limit on the growth of adult school funding reduces the net expenditure per student. This, in turn, reduces the quality of adult education programs and the ability of adult schools to reach a greater number of students. Relying on adult schools to be the primary ESL provider in a region may not be as effective as increasing the resources of all providers to meet the varied and growing needs of communities.  Contents Summary.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii Figures.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix Tables.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi Acknowledgments.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii I. INTRODUCTION.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Research Goals and Questions .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Outline of the Report.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 4 4 2. POLICY CONTEXT FOR ADULT EDUCATION IN CALIFORNIA.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Public and Private Providers in the State.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Dual-Provider System.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Funding Issues Facing Adult Schools.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 3. The English-Language Proficiency of California’s Immigrants.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 The ESL Target Population in California. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Predicted Enrollment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 4. Enrollment in ESL Programs.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 ESL Enrollment: Regional Variation.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Trends in Enrollment in Public Providers.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 ESL Enrollment and ADA Units.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Where Is Enrollment Constrained?.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Teaching for Free?.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Summary.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 5. Policy Implications and Conclusion.. . . . . . . . . . . . . . . . . . . 45 Policy Consequences of Overenrolled Courses. . . . . . . . . . . . . . . . . . . . . . . . . . 45 Partial Reform of the Adult School Funding Formula. . . . . . . . . . . . . . . . . 47 How Much Does Partial Reform of the Funding Formula Help?.. . . . 49 Policy Implications.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 Summary.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 vii Appendix A. Notes on Data and Methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 B. Estimates and Results.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 References.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 About the Author. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 Related PPIC Publications.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 viii Figures 2.1. Private Institutions Offering English Courses. . . . . . . . . . . . . . . . . . . 8 2.2. Start Dates for Adult Education Programs in California.. . . . . . 11 3.1. Size of ESL Target Population, by Region, 2000.. . . . . . . . . . . . . . . 17 3.2. English Ability of Adult Immigrants, by Country of Birth, 2000.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 3.3. Predicted Probability of Enrollment in ESL Courses, 1980, 1990, and 2000.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 4.1. Regional Enrollment in ESL Programs, by Type of Provider, 2002–03.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 4.2. Total Enrollment in ESL Courses in Adult Schools and Community Colleges, 1999–00 to 2002–03. . . . . . . . . . . . . . . . . . . . 32 4.3. ESL Enrollment in Adult School Programs, 1999–00 to 2002–03.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 4.4. Regional Enrollment in Noncredit Community College ESL Courses, 1997–98 to 2003–04.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 4.5. Percentage of All Districts Exceeding Their ADA Enrollment Cap, 1998–99 to 2004–05.. . . . . . . . . . . . . . . . . . . . . . . . . . 38 4.6. Percentage by Which Funded ADA Is Exceeded, 2004–05. . . 39 4.7. Percentage Change in Dollar Value of Funded ADA Needed to Reach Parity with Reported ADA, 1998–99 to 2004–05.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 4.8. Difference Between Dollar Value of Funded ADA and Reported ADA, 1998–99 to 2004–05.. . . . . . . . . . . . . . . . . . . . . . . . . . . 42 5.1. Redistributed ADAs Under AB 23, by Region, 2005–06.. . . . . 49 5.2. Increase in ESL Enrollment as a Result of Redistribution of ADAs for Different Instructional Goals.. . . . . . . . . . . . . . . . . . . . . . 50 ix Tables 3.1. 3.2. 4.1. B.1. B.2. B.3. B.4. B.5. B.6. Size and Growth of the ESL Target Population, 1980–2000. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Size and Growth of Predicted ESL Enrollment, by Region, 1980, 1990, and 2000.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 ESL ADAs and ESL Enrollment, by Goal.. . . . . . . . . . . . . . . . . . . . . . . 37 Logit Regression Results of the Probability of ESL Enrollment, NHES.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 Mean Characteristics of Immigrant Samples in Census and NHES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 Mean Characteristics of Enrolled ESL Students.. . . . . . . . . . . . . . . . 64 Predicted Enrollment Probabilities and Levels of Enrollment for Adult School or Community College ESL Enrollees. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 Effect on Noncredit ESL Enrollment of Districts Exceeding Their ADA Cap.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 Data for Figures 4.7 and 4.8. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 xi Acknowledgments This report would not have been possible without the help, input, and contributions of many people. At the top of the list is Rebecca Steinbach, who provided superb assistance in every stage of this report, including data management and analysis of multiple datasets. I would like to also thank José Ortega, Halena Le, and Randi Knutson at the California Department of Education, and Willard Hom, Tonia Lu, Elias Regalado, Ed Monroe, and Chris Yatooma in the California Community College Chancellor’s Office. The report benefited from conversations with many experts throughout the state, including Ed Morris in the Los Angeles Unified School District; Bob Harper at Pajaro Valley Unified School District; Paul Warren and Rachel Ehlers in the Legislative Analyst’s Office; Luís Esparza, former consultant to Select Committee on Adult Education; Bruce Hamlet in Assemblymember Carol Liu’s Office; Warren Quann in the office of Assemblymember Mervyn Dymally; Robert Perez in the office of Senator Martha Escutia; and Ireri Valenzuela-Vergara at California Tomorrow. This report benefited from detailed reviews and comments by Christopher Jepsen, Marie Mora, Patricia de Cos, and Richard Greene. Finally, I would like to thank Deborah Reed at PPIC for reading through the multiple versions of this report and giving invaluable suggestions and direction. Research publications reflect the views of the authors and do not necessarily reflect the views of the staff, officers, or Board of Directors of the Public Policy Institute of California. xiii 1. Introduction California continues to be the destination for many immigrants from different countries and backgrounds, and the fast growth in the number of immigrants has focused attention on the pace of their integration. The ability to speak English is the most fundamental prerequisite for this process to take place. However, lack of English skills is a hurdle facing a growing number of immigrants because many arrive with little knowledge of English, and learning English is an often difficult and time-consuming endeavor. To help immigrants integrate into society, and for other social and economic reasons, California provides free English as a Second Language (ESL) courses to those who cannot speak English (California Department of Education, 2004; West, 1995). In the words of the Joint Board Task Force on Noncredit and Adult Education (1998, p. 5), there are multiple motives for this policy: “The mission of ESL programs for adults in California is to equip students with the language and cultural proficiencies required for the eventual fulfillment of personal, vocational, academic, and citizenship goals so that they may participate fully in American society.” Indeed, immigrants benefit in many ways from improving their English communication skills. For adults, these benefits include greater participation in civic activities and in their children’s education and increased likelihood of naturalization (Capps et al., 2003; Johnson et al., 1999; Little Hoover Commission, 2002; Martinez and Wang, 2005; Ramakrishnan and Baldassare, 2004). English-proficient immigrants also do better in the labor market, with better job prospects, higher wages, and occupational mobility (see, for example, Berman, Lang, and Siniver, 2003; Bloom and Grenier, 1996; Carliner, 2000; Chiswick and Miller, 1995; Dustmann and Van Soest, 2002; Gonzalez, 2000). For instance, a common finding is that immigrants who can speak English proficiently earn at least 10 percent more than those who cannot. For these and other reasons, the majority of Hispanic immigrants—by far the largest immigrant group in the state—believe that they need to speak English to be part of American society and are eager participants in ESL courses (Pew Hispanic Center, 2006).  California has a long-standing commitment to providing free English instruction. Free ESL courses were first provided by individual school districts in the mid-1850s, but they are now provided as part of the state’s adult education mission (de Cos, 2004; West, 1995). Although adult education is delivered by various agencies, including community colleges and public libraries, adult schools are the largest providers in the state. At the same time, adult schools must deal with funding policies that hamper their ability to meet their mission of teaching English learners. Since adult schools account for more than 75 percent of all public adult education students and funding, the issues affecting adult schools disproportionately affect the state (de Cos, 2004; Joint Board Task Force on Noncredit and Adult Education, 1998). For these reasons, this report focuses on adult schools, although the policy implications are generalized to other public providers. Adult schools receive direct funding from the state’s general apportionment fund based on the levels of funding the local school district received in 1977–78 and this amount can grow a maximum of 2.5 percent from the previous year’s funded level. However, this policy may be out of step with the changes in the immigrant population experienced by California since the 1970s. California now has more immigrants than any state in the country; many of them cannot speak English effectively. Among recently arrived adult immigrants (ages 18 and older), 52 percent report not being able to speak English proficiently. This percentage is higher among those  It is important to emphasize that this report focuses on adult immigrants not enrolled in school, since they are the target population of the adult education system, rather than child immigrants or adult immigrants on student visas (Young, 1995). Limited-English-proficient (LEP) adults enrolled in formal degree programs potentially include foreign-born persons on student visas and hence are in the state only temporarily. These students are not likely affected by adult education policies. The policy implications will differ depending on the population of LEPs considered. School-age LEPs learn English in school, and issues such as failure to enroll, type of English program, and length of time spent in the program have received much attention (Hill, 2004; Jepsen and de Alth, 2005). Another line of research considers natives (usually the second generation) and their English-language ability (McManus, 1985; Portes and Rumbaut, 1996). This report uses the terms “not proficient” and “limited English proficient” interchangeably.  In 1993, adult education legislation permitted the establishment of new adult education programs and also established new revenue limits to eliminate differences in apportionment between districts.  not in school (56%) or who are from Mexico (74%). In 2000, the ESL target population—foreign-born adults classified as LEP—numbered 2.7 million, or 35 percent of California’s seven million adult immigrants. In contrast, this population was less than 900,000 in 1980 and has increased annually by nearly 6 percent since then. Although the large ESL target population generates much heated debate, especially when it concerns the pace of their assimilation, what is overlooked is that the restriction on adult school funding challenges the ability of adult schools to accelerate this process. This is especially true in districts that did not have a large program in 1979 when the base levels for growth were set but have experienced significant growth since then in the target ESL population. Faced with enrollment pressures, educators have to choose between abiding by the cap on enrollment or—if they loosely interpret the state education code mandating the establishment of ESL programs—enrolling more students than the funding level permits. Educators in districts that are experiencing growing demand for ESL courses are more likely to be at odds with adult school funding policies (Joint Board Task Force on Noncredit and Adult Education, 1998). Thus, it is possible that these policies are inconsistent with the continued growth in the number of LEP immigrants in many regions and the mission to  English-language information comes from the Census questions, “Does this person speak a language other than English at home?” and for respondents answering yes, “How well does this person speak English?” Proficient is defined as speaking English only, very well, or well. Not proficient is defined as speaking English not well or not at all. Language questions are asked of persons ages five and older.  According to the U.S. Department of Labor, “An individual with LEP is one who has limited ability in speaking, reading, writing or understanding the English language and (a) whose native language is a language other than English or (b) who lives in a family or community environment where a language other than English is the dominant language” (http://www.doleta.gov/usworkforce/lep/glossary/English_Only.cfm).  The 10 largest countries/regions of origin for immigrants ages 18 and older in the 2000 Census are Mexico (3,384,574), Central America (660,088), the Philippines (614,829), China (535,579), Vietnam (389,384), Korea (247,074), India (239,408), South America (188,714), Iran (149,138), and the former Soviet Union/Russia (148,052). English is one of the official languages in the Philippines and is the most important language in India (http://www.cia.gov/cia/publications/factbook).  Specifically, sections 52540 and 52542 of the education code stipulate that adult schools establish ESL courses if more than 20 persons solicit an ESL course or if the school board decides to establish an ESL program (California Department of Education, 2005).  teach English to those who seek to enroll in ESL classes (de Cos, 2004; Moore, Schulock, and Lang, 2004). Without a change in the funding formula, the conflict between funding and mission may result in a long-term decline in the quality of instruction and diminish the pace of immigrant integration. Research Goals and Questions This report examines the current funding formula for adult schools, California’s mission to English learners, and the demographic changes that challenge the ability of adult schools simultaneously to fulfill their mission while complying with their funding constraints. The report also considers efforts to reform the funding formula. In addition, the report considers the implications of not fully funding public agencies such as community colleges and libraries that teach English to those who seek to learn it. This report addresses the following questions: . What is the policy background for the provision of ESL courses in California? What distinguishes adult schools from other providers? 2. How has the ESL target population changed throughout the state since 1980? What is the level of predicted enrollment and what demographic changes affect enrollment in ESL courses? 3. What are the trends in the provision of ESL courses by adult schools and community colleges statewide and in the different regions of the state? Does the adult school funding formula limit adult school enrollment? To what extent do adult school districts exceed their level of funding? 4. What do adult schools that exceed their funding limit forgo in terms of quality of adult education classes and future growth of adult education programs? How much does the redistribution of unused funding alleviate the challenges facing high-demand regions? How would adult education providers benefit from increased funding? Outline of the Report Chapter 2 considers the policy context for the current provision of ESL courses in the state, including the various providers in the state and the funding issues facing adult schools. Chapter 3 analyzes the ESL  target population from 1980 to 2000 at the regional and state levels, and examines how predicted enrollment responds to changes in the size and characteristics of the immigrant population. Chapter 4 presents actual enrollment levels, the degree to which adult schools overenroll students, and the fiscal effect of the 2.5 percent limit on funding growth. Finally, Chapter 5 considers the effect of AB 23, passed in October 2005 to reform the adult school funding formula, and the importance of increasing public funding to all providers to remedy the conflict between the funding and the mission of adult education policies.  2. Policy Context for Adult Education in California California’s adult education system is not only the largest in the nation, it is also very complex. The system consists of adult schools, community colleges, public libraries, community- and faith-based organizations, the California Conservation Corps, and prisons. Each has different governance and funding mechanisms (de Cos, 2004). This multiple-provider system is the result of a combination of policies that support the type of initial provider in an area. Because of past educational policies, the adult school system is by far the largest provider in the state, but the community college system also teaches a significant share of students in adult education. Therefore, the adult education system in California is, for all intents and purposes, a dual-provider system. For adult schools, the most important policy is the funding formula that limits year-to-year growth of funded enrollment to 2.5 percent. This chapter discusses the public and, briefly, the private, adult education system in the state, expands on the roles played by community colleges and adult schools in delivering ESL courses, and focuses on the funding mechanism for adult schools. Public and Private Providers in the State ESL courses can be provided by private and public institutions, and there are arguments for the role played by each type of provider. As discussed in Chapter 1, a common finding is that increased English ability is rewarded in the labor market. As this gain accrues to the individual, he or she is willing to pay to acquire more English skills as long as there is an economic incentive to do so. This willingness to pay generates a demand for ESL courses that private institutions can provide for a fee no greater than the value of the skill. In fact, private institutions that teach English exist throughout the state, as is seen in Figure 2.1. In total, 105 state-licensed institutions offer  /POEFHSFFQSPWJEFST 0UIFSQSPWJEFST   /VNCFSPGJOTUJUVUJPOT      $FOUSBM 7BMMFZ #BZ -PT"OHFMFT -PT"OHFMFT 4BO%JFHP 3FTUPG "SFB $PVOUZ QFSJNFUFS $PVOUZ UIFTUBUF 4063$&#VSFBVGPS1SJWBUF1PTUTFDPOEBSZBOE7PDBUJPOBM&EVDBUJPO %FQBSUNFOUPG $POTVNFS"GGBJST /05&45IFTFJOTUJUVUJPOTPGGFSOPOEFHSFFPSEFHSFFPUIFS&4-BOETJNJMBSMZ EFTDSJCFEDPVSTFT5IFTBNFJOTUJUVUJPOJTJODMVEFENPSFUIBOPODFJGJUQSPWJEFT JOTUSVDUJPOJOEJGGFSFOUDJUJFT3FHJPOEFGJOJUJPOT#BZ"SFB‡"MBNFEB $POUSB$PTUB  4BO'SBODJTDP 4BOUB$MBSB BOE4BO.BUFP$PVOUJFT$FOUSBM7BMMFZ‡'SFTOP ,FSO  .FSDFE BOE5VMBSF$PVOUJFT-PT"OHFMFTQFSJNFUFS‡0SBOHF 3JWFSTJEF 4BO #FSOBSEJOP BOE7FOUVSB$PVOUJFTSFTUPGUIFTUBUF‡BMMPUIFSDPVOUJFT Figure 2.1—Private Ins'tJiHtuVStFions Offering English Courses nondegree ESL courses and 66 institutions offer degrees or other ESL courses. The bulk of the programs are in the Los Angeles region with the Bay Area also having a high number of private institutions offering ESL courses for a fee. The Central Valley has very few, if any, such institutions.  The regions used in this report do not conform to most definitions of the state’s regions; all data in this report are aggregated according to the definitions used by the Comprehensive Adult Student Assessment System (CASAS) (except as noted) for the sake of consistency with the enrollment data provided by CASAS. See Figure 2.1 for these definitions.  Data are derived from the Directory of Degree, Non-Degree and Registered Institutions and Programs gathered by the Bureau for Private Postsecondary and Vocational Education  Unfortunately, enrollment information is not available for such institutions, so it is not possible to compare enrollment levels in public and private institutions in California. However, estimates from the 2001 National Household Education Survey (NHES) indicate that private institutions in California represent less than 1 percent of all persons who took an ESL course in the previous 12 months. Of those in public institutions, 83 percent of recent ESL students enrolled in adult schools and 14 percent in community colleges. Another 3 percent enrolled in public libraries and other public providers, whereas less than 1 percent enrolled in community- and religious-based organizations. Since there is a private gain to learning English, why does the state provide this skill free? One reason is because this is how it has always been. The tradition of not charging for adult education originated with the first adult school in the state, founded in 1856 in San Francisco (West, 1995). Also, immigrants may not have the resources to pay or borrow for school. Even more important, not all English learners are interested in using that skill for work. Parents out of the labor force, for instance, might want to learn English to help in their children’s education. Immigrants might also want to learn English to become citizens. Increasing the English ability of immigrants is also in the public interest. Since it is not possible to charge only those who would benefit in the labor market, and to promote English acquisition for the overall public good, the state has made a commitment to provide ESL courses and other forms of adult education free. This report makes no judgments about the pros and cons of providing free ESL courses but recognizes that private institutions are part of the larger system of English-language instruction in the state. Dual-Provider System California delivers adult education via this dual-provider system by design, although adult schools and community colleges overlap (BPPVE) at the Department of Consumer Affairs (https://app.dca.ca.gov/bppve/schoolsearch/default.htm). A keyword search was conducted, using terms such as “English as a Second Language” to identify providers.  Private institutions account for 10 percent of ESL students in states outside California.  significantly in their goals and types of programs offered. In addition, adult education has traditionally been a local education function and, depending on the origins of the school, was the responsibility of either the local junior college or the school district. Adult schools offer 10 types of adult education programs and the community college system offers nine, as part of its noncredit “basic skills” mission (California Department of Education, 2005). Adult schools are the primary providers (80%) of adult education and community colleges teach more than 10 percent (California Department of Education, 2006, p. A-1). Several important policies shaped the current adult education system. The first is the creation of adult schools (sometimes called evening or night schools) by local education agencies. The creation and funding of adult education programs rested on the importance placed by local residents on adult education. Figure 2.2 shows the chronology of the founding of adult education programs in the state. Most were founded between 1910 and 1949, with the bulk in the 1930s, partly to provide jobs for teachers (West, 1995). Following the precedent set by St. Mary’s Cathedral in San Francisco, and to stress the importance of “Americanization,” these schools provided courses free (West, 1995). The trends from the 1970s on reflect the effect of an important policy change in education created by Proposition 13—a change that has meant that only a few new programs were funded between then and 1992. In 1993, new legislation provided partial funding for new adult schools, and 43 were founded shortly thereafter. Because local communities had taken the primary responsibility of establishing and funding adult education programs, it was also up to them to integrate their programs into the school district or into the community college system. In 1960, community colleges were integrated into the state’s postsecondary school system as part of the state’s Master Plan for Higher Education and, subsequently, certain adult education programs became part of the noncredit program of the higher education system. With the absorption of community colleges into the postsecondary school system, school districts and community colleges established agreements regarding which would be the main providers in an area. The dual- 10    /VNCFSPGQSPHSBNT                m m m m m m m m m m m            4063$&IUUQXXXDBBEVMUFEIJTUPSZPSH /05&45IFTFEBUBFYDMVEFQSPWJEFSTXJUIVOLOPXOTUBSUEBUFT5IFEBUBBMTPEP OPUJODMVEFBEVMUTDIPPMEJTUSJDUT TJYPGXIJDIBSFOPXDMPTFE UIBUBSFGPVOEJOUIF $BMJGPSOJB"EVMU&EVDBUJPO1SPWJEFS%JSFDUPSZ Figure 2.2—Start Dates for A'JdHuVSlFt Education Programs in California provider system was firmly established at this time (Moore, Schulock, and Lang, 2004; West, 1995). Funding Issues Facing Adult Schools The decision by local communities to place their adult education program in either the K–12 or the community college system had important consequences after the passage of Proposition 13 in 1978, when the state assumed funding responsibility for all K–12 education, including adult schools that were within the local district. Community college  Formal delineation of function agreements were established in 1972 (Senate Bill 94), which gave preference to adult schools as the primary providers of adult education (unless they agreed to give up this right to community colleges). 11 districts continued to maintain local authority over their adult education programs. The fundamentals of the current funding formula for adult schools were set forth in legislation passed in 1979 and had three important characteristics. First, funding for adult schools was a separate funding category from the state general apportionment and could be spent only on adult education. (Until the passage of AB 23 in 2005, unspent adult education funding could not be redistributed to other school districts but instead was returned to the state for redistribution to other programs.) Second, the level of funding was calculated to be the product of the number of average daily attendance (ADA) units and the baseline revenue limit per ADA. (One ADA unit is equal to 525 hours of instruction.) Third, growth of funded ADA was limited to 2.5 percent per year. Moreover, the baseline point for calculating this growth was based on the enrollment levels of 1977–78 (de Cos, 2004; Moore, Schulock, and Lang, 2004; West, 1995). The exact origins of the 2.5 percent growth limit are not clear, but it is likely that this value may not be arbitrary (Moore, Schulock, and Lang, 2004) in the sense that a 5 percent cap on ADA had already been established in 1975 because of significant increases in enrollment, especially in ESL courses (West, 1995). In addition to state funding, adult schools and community colleges that teach ESL can apply for federal funding from the Workforce Investment Act (WIA) Title II program. In 2005–06, adult schools received $621 million from the state and over $53 million from WIA Title II. Unlike adult school funding, funding for adult education programs in community colleges is determined locally. Community college districts receive funding from the state’s general fund, determined by a programbased funding formula, which in turn depends on full-time-equivalent-  Cost-of-living adjustments (COLAs) were also included in the funding growth, up to a maximum of 6 percent, but are not guaranteed.  Each unit of an ADA was funded at $2,292 for adult schools. (These data are from recertified first principal apportionment (P-1) fiscal year 2005–06, available at http://165.74.253.241/ias/Exhibits/pasummary2005p1recert.xls.) In 2005–06, 185 adult school agencies and County Offices of Education received WIA Title II funds (http:// www.cde.ca.gov/fg/fo/r17/documents/ae05awards.xls). 12 student (FTES) hours. Each community college district allocates adult education funding depending on community interest, need, and the level of priority assigned to noncredit programs by administrators. In 2004–05, the noncredit rate per FTES was $1,612. The noncredit baseline revenue funding (noncredit FTES*$1,612) plus COLA (2.41%) totaled $172 million plus around $10 million from WIA Title II funds. The unique funding conditions facing adult schools pose specific challenges to them as they strive to meet the commitment to adults and to English learners in particular. Regardless of differences in demographic need, all adult schools face the same constraint on funding growth. Yet school districts are not limited in the level of enrollment; they are free to exceed enrollment growth beyond 2.5 percent. Those that exceed their enrollment growth, however, absorb the full cost of overenrollment because the net level of per-ADA compensation is lower than the state-approved level. Such a policy can reduce the quality of education and development of new programs. On the other hand, adult schools that turn away students because they are already at their cap underserve a motivated group of immigrants who would otherwise learn English. The level of funded ADA is not based on current demographic conditions but on 1977–78 funded enrollment levels. Since that year, of course, local districts have undergone substantial changes in their immigrant population. Adult schools that served few immigrants in the late 1970s but have experienced significant growth since are hampered today by the initially low levels of funded ADA. Conversely, districts that had an adult education program in the 1970s continue to receive the base amount even if they have continually experienced enrollment declines.  The program-based funding formula for each district is based on FTES, maintenance and operations costs, and other factors. Districts with noncredit programs have their allocations for maintenance and operations and instructional support computed to explicitly account for noncredit FTES. The final rate for noncredit FTES deducts the noncredit allocation for maintenance and operations and institutional support (California Community Colleges Chancellor’s Office, 1999, p. 14).  This estimate includes only the amount attributable to noncredit enrollment. I would like to thank Ed Monroe and Elias Regalado who provided the financial data (http://www.cccco.edu/divisions/cffp/fiscal/allocations/links/apportionment/04_05_ simulated_recal/2004-05PostedSimulatedRecal.pdf). WIA Title II information comes from the 2005–06 Grants Awards (http://www.cde.ca.gov/fg/fo/r17/ documents/ae05awards.xls). 13 In other words, the distribution of funded ADAs in any given year is not linked to any current measure of demand. This shortcoming in the funding formula curbs the ability of school districts to adjust their programs (beyond the 2.5% limit) to better meet their community’s needs. At the state level, ADAs are inappropriately distributed. The shortcomings of these two aspects of the funding formula are considered in the chapters below, which examine the extent to which adult school enrollment is restricted by this limit and the growth rate in the population that would benefit from ESL courses. 14 3. The English-Language Proficiency of California’s Immigrants The provision of public ESL courses is conditioned on the need for them. However, the state’s adult schools receive increases in annual funding only up to a maximum of 2.5 percent even if their enrollment growth is 2.5 percent or greater. Since enrollment is potentially constrained, there is no direct measure of need to ascertain whether the mission of teaching English learners is being met. Reported enrollment is not a good measure of the demand for adult education because some local districts cut off enrollment when the 2.5 percent limit is reached, whereas others enroll students beyond this limit. This chapter presents two alternative measures of need based on Census and survey data about ESL course-taking behavior. The ESL Target Population in California Because adult education programs serve adults not enrolled in an academic program, the target population for ESL programs consists of LEP adults not currently enrolled in school. This definition is used for multiple reasons. First, full- or part-time students are likely to enroll in credit courses or, if they are younger than age 18, in the K–12 system. Second, adults served by ESL programs are likely unable to succeed in most academic or certificate programs because of language barriers, although they may later enroll in such programs. Third, adult immigrants who are not English-proficient are the most likely of any group of immigrants to benefit from adult schools. This set of immigrants then  Hill (2004) notes that immigrants arriving in their late teens do not enroll in the K– 12 system. An LEP immigrant younger than age 18 cannot enroll in an adult school unless he or she is concurrently enrolled in high school and then only under certain circumstances (but he or she is not defined as an adult for funding purposes) (California Department of Education, 2004, p. 1). 15 represents a reasonable approximation of the population that is relevant for policymakers and the adult education system. However, for several reasons the ESL target population is only an approximation of the actual number of people wishing to take ESL courses. First, it overestimates the potential number of ESL students because not all the ESL target population wishes to enroll in public ESL classes. Some have no desire to learn English at the moment; others may do so on their own or through private institutions. Second, it is likely that the ESL target population differs from that actually enrolled in ESL classes, since the latter consists of a self-selected group of individuals who have decided to seek out ESL courses at a particular point in time (Lazear, 1999). Third, this population represents only the stock of need for ESL rather than the flow; the former provides information about the need for ESL courses at one point in time (at the time of the Census), and the latter describes the change in need for ESL courses from one year to the next. Nevertheless, such information from the Census is a popular measure because there are few alternatives. Regional information for the ESL target population in 2000 is shown in Figure 3.1. For consistency with data from CASAS presented throughout this report, counties are grouped according to the definitions used by CASAS. Statewide, there were nearly 2.7 million immigrants in the ESL target population in 2000. Much of the state’s ESL target population in 2000 was concentrated in several parts of the state. Los Angeles County had by far the largest ESL target population, numbering over 1.1 million immigrants or 43 percent of the state total. Around 510,000 persons in the ESL target population lived in the Los Angeles perimeter—the counties in Southern California outside Los Angeles County. The Bay Area (380,000) had the third-largest ESL target population of any region (14%), followed by the rest of the state with  The participation rate estimated by Young (1995) using Public Use Microdata Sample (PUMS) data is also not ideal, since that source considers only first-time students and, more important, does not account for the inflow and outflow of immigrants into the target population. It also assumes that all LEP students are the same in their need for ESL instruction, which may not be the case.  Although we made repeated attempts to gain access to district-level enrollment data, these data were not made available to us for analysis. Hence, despite its shortcomings, we have used the regions defined by CASAS in this report. 16 Del Norte Siskiyou Modoc Trinity Humboldt Shasta Lassen Tehama Plumas Yuba Mendocino Glenn Butte Sierra Colusa Sutter Lake Placer Nevada Sonoma Napa Yolo El Dorado Amador Alpine Sacramento Marin Solano Calaveras San Francisco Contra San Costa Joaquin Tuolumne Mono San Mateo Alameda Santa Stanislaus Mariposa Clara Merced Santa Cruz Madera San Benito Monterey Fresno Kings Tulare 4BO%JFHP$PVOUZ   $FOUSBM7BMMFZ   3FTUPGUIFTUBUF   #BZ"SFB   -PT"OHFMFTQFSJNFUFS   -PT"OHFMFT$PVOUZ    Inyo San Luis Obispo Kern Santa Barbara Ventura Los Angeles San Bernardino Orange Riverside San Diego Imperial 4063$&1VCMJD6TF.JDSPEBUB4BNQMF 16.4  /05&45IF&4-UBSHFUQPQVMBUJPODPOTJTUTPGGPSFJHOCPSOPVUPGTDIPPMBEVMUT BHFT BOEPMEFS XIPTQFBL&OHMJTIiOPUBUBMMuPSiOPUXFMMu4FF'JHVSFGPSSFHJPOEFGJOJUJPOT Figure 3.1—Size of ESL Ta'rJgHeVtSFPopulation, by Region, 2000 320,000 (12%). The Central Valley had 6 percent of the state’s ESL target population in 2000, or about 170,000 persons. Since many of the state’s immigrant communities were in Los Angeles County, the Los Angeles perimeter, and the Bay Area, this distribution of the target population is 17 consistent with other analyses of the English ability and settlement patterns of immigrants (Bauer, Epstein, and Gang, 2005; Chiswick and Miller, 2002). To better compare the demographic changes in the state with the growth constraints on adult school enrollment, Table 3.1 presents the growth in the ESL target population from 1980 to 2000. Los Angeles County stands out from other regions but in a manner contrary to expectations given its large immigrant population: It had the smallest peryear growth of the ESL target population of any region during this 20year period—4.4 percent. This rate is nearly 2 percentage points lower than that of the Bay Area, the next slowest-growing region (6.1%). The region with the fastest growth over this 20-year period was the Los Angeles perimeter (7.7%), followed by the rest of the state (7.1%). Since Los Angeles County contained 43 percent of the state’s ESL target population in 2000, the statewide percentage was less than 6 percent from 1980 to 2000. Without Los Angeles County, the statewide annual growth rate was 6.9 percent from 1980 to 2000. The annualized growth rates based on two 10-year periods (1980–1990, and 1990–2000) are also shown in Table 3.1 and reveal a decrease in the growth rate of the ESL target population. In the 1980s, the ESL target population grew by nearly 10 percent per year in the counties constituting the Los Angeles perimeter, followed by less than 8 percent in the Central Valley and San Diego County regions. The other regions grew between 5.7 percent (Los Angeles County) and 7.1 percent (rest of the state) in the 1980s. However, in the 1990s, a redistribution of the state’s ESL target population took place. First, the growth rates declined in all regions except the rest of the state in the 1990s. Second, Los Angeles County’s share of the growth of target ESL population declined from 44 to 31 percent in the 1980s. During this time, the Bay Area and the rest of the state increased their shares of the ESL target population; these two regions accounted for  The constant annual growth where LEPxand LEPy are the LEP rpaotepuislactaiolcnulianteydeabryx[a(LnEdPyxe /aLr Ey,Prye )s1p/(ex–cyt)iv–e1ly].,  In 1980, 54 percent of the ESL target population resided in Los Angeles County. By 2000, this figure had declined to 43 percent. In contrast, the share of the ESL target population grew in the Los Angeles perimeter and rest of the state regions, from 13 and 9 18 Table 3.1 Size and Growth of the ESL Target Population, 1980–2000 19 ESL Target Population   1980 1990 2000 Central Valley 47,649 100,452 166,750 Bay Area 116,282 214,323 382,744 Los Angeles County 479,225 835,554 1,131,623 Los Angeles perimeter 114,291 283,480 508,297 San Diego County 40,620 86,068 142,445 Rest of the state 81,900 163,191 320,868 Total 879,967 1,683,068 2,652,727 Total minus Los Angeles County 400,742 847,514 1,521,104 SOURCE: 2000, 1990, and 1980 5% PUMS. NOTE: See Figure 2.1 for sample and definitions. Average Annual Increase and Annualized Growth Rate 1980–1990 1990–2000 1980–2000 5,280 7.7% 6,630 5.2% 5,955 6.5% 9,804 6.3% 16,842 6.0% 13,323 6.1% 35,633 5.7% 29,607 3.1% 32,620 4.4% 16,919 9.5% 22,482 6.0% 19,700 7.7% 4,545 7.8% 5,638 5.2% 5,091 6.5% 8,129 7.1% 15,768 7.0% 11,948 7.1% 80,310 6.7% 96,966 4.7% 88,638 5.7% 44,677 7.8% 67,359 6.0% 56,018 6.9% 34 percent of the growth in the ESL target population in the 1990s—an increase from 22 percent in the 1980s. Meanwhile, the share in all other regions increased slightly from the 1980s, with the Los Angeles perimeter’s share representing one-fifth of the state’s growth in the ESL target population. Since the Los Angeles perimeter has one of the fastest and most sustained growth rates in the state, it is likely that the continued decline in the growth rate in Los Angeles County results perhaps because more immigrants choose to live in nearby counties. Yet, despite Los Angeles County’s having the lowest (and declining) growth rate of the ESL target population, it continues to have both the largest ESL target immigrant population in the state—1.1 million in 2000—and the largest share of the ESL target population of any region of the state—39 percent. There is also a great deal of variation in the foreign-born adult population by country of birth. Figure 3.2 categorizes adult immigrants by country of birth and English-language ability in 2000. Around 55 percent of Mexican immigrants (1.7 million) reported not being proficient in English—the largest group in either absolute or percentage terms. Immigrants from the rest of Latin America (37%) and Southeast Asia (40%) also had a relatively high percentage of LEP immigrants. The combined Latin American and Mexican immigrant population not only had the largest percentage of immigrants who could not speak English well, but this population (two million in 2000) was two-thirds the size of the total immigrant population from other countries (3.1 million). In aggregate, Latino LEP immigrants made up 72 percent of the LEP population in California in 2000. About 23 percent of other Asians (nearly 400,000) reported not being proficient in English, second to Mexico in the number of LEP immigrants. The remaining LEP immigrants from percent in 1980 to 19 and 12 percent, respectively in 2000. The share for all other regions remained constant.  Ed Morris of the Los Angeles Unified School District (LAUSD) Adult Education Office noted the urban-to-rural migration as a contributing factor to the decline in enrollment in ESL courses there, from a high of 43,285 ADAs in 2001–02 to 40,313 in 2004–05.  Over half of the Central Valley’s adult immigrants are LEP, followed by Los Angeles County at 41 percent. Los Angeles County’s share of the LEP population is similar to its distribution of the adult immigrant population (39%). 20 &4-UBSHFUQPQVMBUJPO &OHMJTIQSPGJDJFOU   1FSDFOUBHF-&1BOE&OHMJTIQSPGJDJFOU                             .FYJDP 3FTUPG 4PVUIFBTU -BUJO"NFSJDB "TJB "TJB       &VSPQF   "GSJDB PUIFS 4063$&16.4GPS$BMJGPSOJB /05&44FF'JHVSFGPSTBNQMF$BOBEB /FX;FBMBOE BOE"VTUSBMJBBSFJODMVEFE JOUIF&VSPQFDBUFHPSZ Figure 3.2—English Ability of Ad'uJHlVtSFImm igrants, by Country of Birth, 2000 Europe, Africa, and other countries totaled over 85,000. More than 90 percent of persons from these regions of the world reported being proficient in English. The growth rate of the ESL target population is outpacing the stateimposed growth limit on adult schools by more than a 2-to-1 ratio statewide and by nearly 3-to-1 in certain regions of the state. Despite the caveats in Table 3.1, these growth rates support the contention that the funding formula does not recognize the demographic reality. Predicted Enrollment An alternative measure of the need for ESL courses can be derived from the 1999 and 2001 NHES, which asked individuals if they had enrolled 21 in an ESL class in the previous 12 months. From these national data, it is possible to predict the factors affecting the probability of whether an individual in California enrolls in an ESL course. The decision to learn English depends on many factors, including accessibility, availability, convenience, and economic incentives (such as higher wages). The variables chosen to estimate this behavioral model capture the various costs and incentives leading to the decision on whether to enroll in ESL courses (Crandall and Sheppard, 2004). The variables are age, age at arrival in this country, highest level of education completed, Hispanic origin, country of origin, family size, marital status, whether female, and whether living in California. Once the effect of these variables is estimated using the NHES data, these values are combined with the demographic characteristics of the appropriate sample of immigrants from 1980, 1990, and 2000 Census data (5% PUMS) to obtain individual probabilities of enrollment in ESL courses. These probabilities are then averaged by region and multiplied by the total population size that the immigrant sample represents. The estimated enrollment levels reflect enrollment in all ESL providers in California at the time of the Census. This measure of the need for ESL courses has several advantages over others. First, it is based on a model of enrollment during a 12-month period and thus measures the flow, rather than the stock, of the need for ESL courses. Second, the predictions are a function of factors affecting the decision to enroll in ESL courses. Since the decision to learn English depends on multiple factors, such as age, years of schooling, country of origin, marital status, and so on, changes in these factors will alter the likelihood that immigrants enroll in ESL courses (Crandall and Sheppard, 2004; National Center for Family Literacy and National Center for ESL Literacy Education at the Center for Applied Linguistics, 2006; Young, 1995). Increases in the size of the population, holding composition the  Appendix A discusses the NHES sample restriction, consisting of persons ages 18 and older who did not enroll in an ESL college program. Because of sample size considerations, the sample includes enrollees in all ESL providers, not just adult schools or community colleges. Restricting the ESL student sample to adult school and community college enrollees does not change the sign or the magnitude of the statistically significant variables. See Appendix Table B.1. 22 same, also increase the number of ESL students. The estimates from this measure can be used to gauge regional and statewide changes in predicted enrollment resulting from changes in the composition and size of immigrant groups to assess the overall provision of ESL courses. Table 3.2 presents predicted ESL enrollment for different regions of the state. That enrollment in 2000 is around 620,000. Los Angeles County has the largest level of predicted enrollment, around 250,000. Los Angeles County’s share of the predicted level of enrollment, 40 percent, is similar to the ESL target population in Table 3.1 (43%). The Bay Area and Los Angeles perimeter have the second- and third-largest number of predicted ESL students, nearly 110,000 and 120,000, respectively. The counties in the rest of the state category have 76,000 students predicted to enroll in an ESL course, San Diego County has 37,000 students, and the Central Valley has a predicted enrollment of 32,000. Around 75 percent of predicted enrollment is distributed in the Bay Area, Los Angeles County, and Los Angeles perimeter. The annualized regional growth rates reveal differential patterns of ESL enrollment over time. The level of predicted enrollment in 2000 is 30 percent greater than in 1990 and 100 percent higher than in 1980. The statewide levels in those years based on the model are around 315,000 and 475,000, respectively. Los Angeles County stands out from all other regions of the state because the predicted number of students is flat between 1990 and 2000. Elsewhere, the annual growth rate in predicted enrollment exceeds 2.5 percent, ranging from 2.8 to 5.8 percent in the 1990s. The regions with the greatest growth in predicted enrollment in the 1990s are the Bay Area and the rest of the state with rates of 5.8 percent. The Central Valley and the Los Angeles perimeter have similar annual growths during this period, 3.7 and 3.2 percent, respectively. San Diego County’s annualized growth rate is also relatively modest, at 2.8 percent. The statewide growth is 2.7 percent but without Los Angeles, the annualized growth rate is 4.4 percent. The differences in estimates are due to two factors: the characteristics and the number of immigrants in each Census year. Figure 3.3 presents the average probability of enrollment in ESL courses in each year by region. These probabilities depend on the characteristics of immigrants in each region, and any change over time is due to differences in the composition 23 Table 3.2 Size and Growth of Predicted ESL Enrollment, by Region Predicted Enrollment Annualized Growth (%)   1980 1990 2000   1980–1990 1990–2000 1980–2000 Central Valley 12,607 22,375 32,317 5.9 3.7 4.8 Bay Area 44,810 62,267 109,138 3.3 5.8 4.6 Los Angeles County 173,781 233,046 249,170 3.0 0.7 1.8 Los Angeles perimeter 39,944 85,419 117,591 7.9 3.2 5.5 San Diego County 15,921 28,365 37,368 5.9 2.8 4.4 Rest of the state 27,406 43,594 76,291 4.8 5.8 5.3 Total 314,468 475,065 621,876 4.2 2.7 3.5 Total minus Los Angeles County 140,688 242,019 372,706   5.6 4.4 5.0 SOURCE: Author’s calculations from a model estimated using 1999/2001 NHES data and applied to 1980, 1990, and 2000 Census 5% PUMS. NOTES: These estimates are based on a logit regression of ESL enrollment from the NHES data and applied to the PUMS data. See Appendix Table B.1 for estimates. 24   1FSDFOUBHFQSPCBCJMJUZPGFOSPMMNFOU     $FOUSBM 7BMMFZ #BZ -PT"OHFMFT -PT"OHFMFT 4BO%JFHP 3FTUPG "SFB $PVOUZ QFSJNFUFS $PVOUZ UIFTUBUF 4063$&"VUIPSTDBMDVMBUJPOT /05&45IFTFFTUJNBUFTBSFCBTFEPOBMPHJUSFHSFTTJPOPG&4-FOSPMMNFOUGSPNUIF BOE/)&4BOEBQQMJFEUPUIF  BOEGPS$BMJGPSOJB16.4 Figure 3.3—Predicted Probab'ilJiHtVySFofEnrollment in ESL Courses, 1980, 1990, and 2000 of the ESL target population rather than to changes in size. In 2000, statewide, these probabilities are around 11 percent, compared to 15 percent in 1990 and 1980. From 1990 to 2000, these probabilities declined in all regions of the state, but the percentage decline was greatest in Los Angeles County and the Los Angeles perimeter (over 35%), slightly greater than the 30 percent decline in San Diego County and the Central Valley. In contrast, the mean probability of enrollment declined in the Bay Area and the rest of the state by around 18 percent. The 6 percentage point decline in enrollment probability in Los Angeles County, the Los Angeles perimeter, and the Central Valley equalized enrollment probabilities in all the regions by 2000 to 10–11 percent. In other words, these regions had higher enrollment 25 probabilities, on average, than other regions in previous decades, but not in 2000. The changes in average probability of enrollment are due to changes in the demographic characteristics of immigrants. The three most important factors leading to a decline in the probability of enrollment from 1990 to 2000 were the statewide increase in average age (40.9 to 42.4), the decrease in average age at arrival (28.8 to 25.1), and the increase in percentage Mexican (43 to 47). In the case of Los Angeles County, the relatively large decline in enrollment probability is explained by an increase of 2.7 years in the average age of immigrants, compared to a 0.6 year increase in mean age in the non–Los Angeles counties (see Appendix Table B.2). Enrollment also depends on the size of the eligible immigrant population. As Table 3.1 shows, the ESL target population increased slightly in Los Angeles County. This explains why, despite the significant decline in enrollment probability in Los Angeles County, the predicted level of enrollment increased by a small amount during the 1990s. These two factors—change in characteristics and change in target population— explain the levels of enrollment in Table 3.2. These predicted growth rates are also reflected in the statewide population growth rates reported by Johnson (2002). He reports that the Inland Empire, San Joaquin Valley, and the Sacramento Metro areas had the largest population growth rates in the 1990s. International migration accounted for at least 30 percent of the population growth in Los Angeles, Orange, San Diego, and Ventura Counties and the Bay Area and 27 percent in the Inland Empire (San Bernardino and Riverside Counties). At the same time, he notes that there was substantial out-migration from Los Angeles County during this time. These trends suggest that the same areas also experienced large growth in ESL enrollment. Summary This chapter considered two ways to assess the level of need for ESL courses in California. The broadest but perhaps least satisfactory measure is the size of the ESL target population in the state and its growth from 1980 to 2000. The second is derived from a national survey of participation in ESL courses in the previous 12 months. Both measures reveal similar patterns with regard to the need for ESL since the 1980s. First, a large (and 26 growing) population would clearly benefit from enrolling in ESL courses. Second, this population is unevenly distributed throughout the state. The counties around Los Angeles and the Bay Area in particular have the largest number of potential ESL enrollees. Third, the growth in each measure of need (except for predicted enrollment in Los Angeles County) exceeds 2.5 percent in every region, and in certain regions (such as the rest of the state and the Bay Area), the growth in need is more than double this amount. Enrollment in ESL courses depends on demographic factors that have changed since the 1970s and differ throughout the state. For instance, an increase in the average age and the percentage Mexican and a decrease in age at arrival explain the lower statewide probability of enrollment in the 1990s. These findings are compelling evidence that the level of provision of ESL courses in any given year falls short of the need. Furthermore, the variability in need for ESL courses throughout the state points to another fundamental flaw in the adult school funding formula: It does not recognize changes in immigrant demographics in different regions of the state. 27 4. Enrollment in ESL Programs Adult schools and community colleges are the most important providers of ESL courses in the state, currently enrolling more than 600,000 students per year, with adult schools teaching more than 75 percent of all ESL students. The level of enrollment is the outcome of the interplay between the demand for ESL courses and their provision. Up to the 2.5 percent growth limit, adult schools are not constrained from enrolling all potential students. Additional students, however, if allowed to enroll, are not funded by the state; some adult schools strive to align actual enrollment with funded enrollment, whereas others enroll more students than they receive funding for. The two measures of demand imply that the need for ESL courses exceeds the growth of funding for adult schools. And since adult schools are the major providers of adult education in the state, it is likely that the number of potential English learners exceeds the public provision in many parts of the state. When the need for ESL courses grows by more than 2.5 percent per year, adult schools may be caught in the middle of two competing policies: the 2.5 percent limit on funding growth and the mission to teach those who seek to learn English. This chapter examines which districts abide by the funding constraint, which disregard this limit (either as a matter of policy or by happenstance), and the extent to which adult schools are underfunded. Although adult schools are the primary providers of adult education, community colleges are an important component in the state’s adult education system. In some areas, such as San Francisco and San Diego County, they are either the sole or major provider. For this reason, this chapter presents enrollment information for both providers, although the emphasis is on how the adult school system is affected by the funding formula. ESL Enrollment: Regional Variation Enrollment information for adult schools is obtained from various reports published by CASAS for the 1999–00 to 2002–03 fiscal years. Adult school enrollment is obtained from the CASAS reports on adult schools, and community college enrollment information is obtained from data provided 29 by the California Community College Chancellor’s Office (CCCCO). All noncredit nonduplicated students with positive enrollment hours are included, although students who may have also enrolled in a credit class are not excluded. Appendix Table B.3 gives a brief demographic portrait of students enrolled in adult schools and community colleges and shows that students in both providers are similar. Figure 4.1 presents the geographic diversity in total enrollment (adult schools plus community colleges) in 2002–03, the last year for which enrollment information by region is available. Los Angeles County accounted for more than 40 percent of total ESL enrollment in the state, with 272,000 students enrolled. This enrollment was more than twice the size of the combined next two largest regions, the Bay Area and the Los Angeles perimeter, each with around 110,000 enrolled students. The rest of the state followed with 80,000 students. San Diego County and the agricultural counties of the Central Valley enrolled 50,000 and 21,000 ESL students, respectively, that year. In 2002–03, there was a great deal of variation in the percentage enrolled in adult schools across regions. At the high end, adult schools accounted for over 90 percent of ESL enrollment in Los Angeles County and the Central Valley and in the low end for 40 percent in San Diego County and 57 percent in the Los Angeles perimeter. In the Bay Area, adult schools accounted for 73 percent of all enrolled ESL students, a value similar to that for the rest of the state (81%). Regardless of these percentages, the contribution of community colleges to adult education should not to be minimized, since they are either the sole providers in an area or supplement the provision of ESL courses through agreements with local adult schools. Trends in Enrollment in Public Providers Figure 4.2 shows the total ESL student population, split between adult schools and community colleges for the years in which the available data overlap (1999–00 to 2002–03). The total number of students increased from 496,000 in 1999–00 to 652,000 in 2002–03. Yet, this increase in total  Only around 60 percent of all community colleges are surveyed by the CASAS WIA Title II data. The data for adult schools exclude agencies that did not receive funding and did not meet the minimum reporting requirements (CASAS, 2004). 30 "EVMUTDIPPMT $PNNVOJUZDPMMFHFT    /VNCFSFOSPMMFE  T         $FOUSBM 7BMMFZ   #BZ "SFB      -PT"OHFMFT -PT"OHFMFT 4BO%JFHP $PVOUZ QFSJNFUFS $PVOUZ  3FTUPG UIFTUBUF 4063$&$"4"4%ZOB3FQPSUT"EVMU4DIPPMEBUBBOE$$$$0FOSPMMNFOUEBUB /05&45IFBEVMUTDIPPMEBUBBMTPJODMVEF&4-DJUJ[FOTIJQDPVSTFT*OUIF$$$$0EBUB  POMZSFTQPOEFOUTXJUIQPTJUJWFFOSPMMNFOUIPVSTBOEOPOEVQMJDBUFEBOOVBMFOSPMMNFOU JOGPSNBUJPOBSFJODMVEFE Figure 4.1—Regional Enrollm'enJHtViSnFESL Programs, by Type of Provider, 2002–03 enrollment masks the fact that the share of students in community colleges declined from a high of 29 percent in 1999–00 to 23 percent in 2002–03. The number of students served by community colleges also was lower in real terms after 2001–02. In fiscal year 2001–02, there were substantial midyear cutbacks resulting in stagnant or negative growth in adult education for both providers. The budget problems continued in subsequent years, but the state dealt with the cutbacks by deferring payments to adult schools and community colleges from one fiscal year to the next (July 1). This accounting tactic allowed agencies to continue offering services at a level that was at least similar to that of previous years. This rolling forward of 31 "EVMUTDIPPMT $PNNVOJUZDPMMFHFT  /VNCFSFOSPMMFE  T                m m m m /05&4&4-BMTPJODMVEFT&4-DJUJ[FOTIJQDPVSTFT4FF'JHVSFGPSPUIFSOPUFT Figure 4.2—Total Enrollmen'tJHiVnSFESL Courses in Adult Schools and Community Colleges, 1999–00 to 2002–03 payments has been used since then to limit the effect of cutbacks in the current fiscal year. For this reason, the only significant decline in funding is noticeable in the 2002–03 fiscal year. In addition to the deferrals in 2002–03, there were midyear reductions in funding for the K–12 and community college systems. The effect of this reduction is seen in the lower enrollment levels in community colleges in that year in Figure 4.2. From 1999–00 through 2002–03, adult schools enrolled nearly 1.8 million students in ESL courses (see Figure 4.3), and ESL accounted for  The 2005–06 budget deferred $45.9 million of adult school funding to 2006–07 (see http://www.cde.ca.gov/fg/fr/eb/documents/budgetreport05.pdf).  Adult school funding was reduced by $74 million and total community college funding was reduced by $231 million (http://www.lao.ca.gov/analysis_2003/education/ ed_2_cc_midyr_prop98_anl03.htm). 32 /VNCFSFOSPMMFE  T   &4-DPVSTFT 0UIFSQSPHSBNT                  m m 4063$&$"4"4%ZOB3FQPSUT"EVMU4DIPPMEBUB /05&&4-BMTPJODMVEFT&4-DJUJ[FOTIJQDPVSTFT m m Figure 4.3—ESL Enrollment in A'JdHuVSlFt School Programs, 1999–00 to 2002–03 42 to 44 percent of all adult school programs. Although the share of ESL students in adult schools was rather stable over this period, the number of students in ESL programs steadily increased from about 350,000 in 1999–00 to 500,000 in 2002–03. The corresponding year-to-year growth rates in ESL enrollment were 15, 23, and 1 percent. Enrollment growth slowed considerably in the 2002–03 program year (1%), not just compared to previous years but also relative to the growth in non-ESL programs (11%) in that same year. This is likely a result of the funding problems noted above. Yet, before 2002–03, ESL and non-ESL programs grew at an identical pace, but this was not the case in 2002–03 (1% versus 11%). In any given year, total enrollment in community colleges is significantly less than enrollment in adult schools. Figure 4.4 shows that enrollment in noncredit ESL courses in community colleges peaked at 33 /VNCFSFOSPMMFE  T #BZ"SFB-PT"OHFMFT$PVOUZ-PT"OHFMFTQFSJNFUFS 4BO%JFHP$PVOUZ3FTUPGUIFTUBUF$FOUSBM7BMMFZ          m m m m m m m 4063$&"VUIPSTDBMDVMBUJPOTGSPNEBUBQSPWJEFECZ$$$$0 /05&5IFEBUBSFGFSUPVOEVQMJDBUFE XJUIJODPMMFHF TUVEFOUDPVOUTJOUIFTBNF BDBEFNJDZFBS  Figure 4.4—Regional Enrollme'nJHtViSnFNoncredit Community College ESL Courses, 1997–98 to 2003–04 160,000 in 2001–02. This pattern is similar to that of adult schools in that enrollment slowed significantly after 2001–02 (although adult school enrollment leveled off). The reason for this decline is the same as that for adult schools: decreased funding as a result of the state’s budget problems beginning in 2001–02. The regional variation in the level of enrollment in noncredit ESL courses in Figure 4.4 is a result of multiple factors, such as the level of need in a region, but perhaps more important are the historical factors determining the main provider in a region. Community colleges are the main providers of adult education in an area if they were the main providers before the 1960s, if no current adult school exists in the area, or if community colleges reach a delineation of function agreement with the local school district to provide adult education classes. In any given 34 year, enrollment in noncredit ESL courses is greatest in the Los Angeles perimeter and tends to surpass 40,000 per year. Enrollment in San Diego County and the Bay Area is normally around 30,000 per year. Los Angeles County is the state’s fourth-largest provider of noncredit ESL courses at over 20,000 students each year. The community colleges in the rest of the state enroll around 15,000 students in most years whereas those in the Central Valley enroll fewer than 1,000 students. ESL Enrollment and ADA Units This section focuses on adult school enrollment but distinguishes between actual enrollment levels in adult education programs and funded enrollment. The former is the number of students that adult schools enroll, and the latter is enrollment within the state-approved level of funding. However, the state funds enrollment based on hours of attendance rather than number of students. Specifically, adult school funding is based on the total number of ADA units taught. Because funding is measured in ADAs, this chapter presents enrollment and funding in terms of ADA units and the dollar value of these units. Because the hours of instruction differ across enrollees, the levels of enrollment and ESL target population previously presented can be translated into ADAs only after making assumptions regarding the hours of instruction needed by each ESL student. Most of the estimates regarding the appropriate amount of instruction depend on the goal of instruction. At the high end of such estimates, one report suggests that it takes between 500 and 1,000 hours of instruction to achieve functional proficiency in English (Crandall and Sheppard, 2004). Others estimate that it takes 100 hours at a minimum for a one-grade-level increase in reading comprehension tests but suggest 150 hours to significantly increase the chances that students attain this goal (National Center for the Study of Adult Learning and Literacy, 2002, p. 85). Another evaluation of adult education programs found that the median number of hours of instruction for students who successfully completed beginning ESL is 216 hours, 136 for intermediate ESL class, and 372 for a combined beginning and advanced ESL class (Young, 1995, p. 43). The average and median number of hours of attendance for ESL in California is 113 (CASAS, 2004; Young, 1995). 35 Table 4.1 presents the implied number of ADA units for the recommended hours of instruction associated with three different goals. For comparison, the total reported adult school ADAs in 2002–03 are also provided. Goals that require more hours of instruction require a greater share of a region’s total ADA. To attain the goal of providing the statewide average level of instruction, adult schools in San Diego County and the Central Valley use the least number of ESL ADAs in the state, 4,300 each, or less than one-third of their adult school’s total reported ADA. The Los Angeles perimeter and Bay Area dedicate more of their ADAs than any other region to attain any of the three goals. For instance, to complete a beginning ESL course, 86 to 89 percent of reported ADAs in those regions is required. Where Is Enrollment Constrained? The levels of funding and enrollment for ESL courses are likely mismatched in school districts that are already at the maximum of their adult education enrollment cap. Unfortunately, ADA information for ESL programs is not available separately, but since these programs account for more than 40 percent of adult education programs statewide, it is likely that a binding ADA enrollment cap affects ESL programs the most. What percentage of districts is affected by the limit on the growth of adult education ADA? Figure 4.5 examines the percentage of school districts in a region that are over their ADA enrollment cap over time. Reported ADA is the total units of instruction that the district actually provides, and funded ADA is the ADA cap that each district is allocated according to the funding formula (based on the previous year’s enrollment level, growth adjustments, and COLA). This measure makes it possible to compare differences across regions regardless of the size of their programs. The Bay Area, Los Angeles  In 1998–99, eight districts were at their cap, four in 1999–00, six in 2000–01, five in 2001–02 and 2002–03, one in 2003–04, and none in 2004–05. Most regions that are below their cap follow the same overall pattern over time. Between 1998–99 and 2000– 01, the number of school districts under their ADA enrollment cap increased and then, after 2000–01, the number fell, consistent with the state budget problems experienced after 2001. The rest of the state, Los Angeles County, and the Bay Area in particular experience the greatest numerical declines between 2001–02 and 2003–04. 36 Table 4.1 ESL ADAs and ESL Enrollment, by Goal Average Hours of Increase Reading Score Complete Beginning Attendance One Grade Level ESL Course Reported Adult % ESL % ESL % ESL   School ADAs ESL ADAs ADAs   ESL ADAs ADAs   ESL ADAs ADAs Central Valley 16,647 4,327 26.0 5,743 34.5 8,271 49.7 Bay Area 40,527 18,298 45.1 24,289 59.9 34,976 86.3 Los Angeles County 134,319 53,210 39.6 70,633 52.6 101,712 75.7 Los Angeles perimeter 29,561 13,821 46.8 18,346 62.1 26,419 89.4 San Diego County 13,074 4,306 32.9 5,716 43.7 8,231 63.0 Rest of the state 37,777 13,985 37.0 18,564 49.1 26,732 70.8 Total 271,905 107,947 39.7   143,292 52.7   206,341 75.9 NOTE: Based on 2002–03 ESL enrollment levels reported in Figure 4.1. 37 $FOUSBM7BMMFZ #BZ"SFB -PT"OHFMFT$PVOUZ -PT"OHFMFTQFSJNFUFS 4BO%JFHP$PVOUZ 3FTUPGUIFTUBUF     1FSDFOUBHF      m m m m m m m 4063$&$BMJGPSOJB%FQBSUNFOUPG&EVDBUJPO 1SJODJQBM"QQPSUJPONFOU6OJU  Figure 4.5—Percentage of All Dis'trJHicVtSFsExceeding Their ADA Enrollment Cap, 1998–99 to 2004–05 County, and Los Angeles perimeter seem the most similar in terms of the percentage of the school districts over their ADA cap. From 2001–02 on, more than 60 percent of adult schools in these regions were over their ADA cap. Similarly, between 40 and 50 percent of San Diego County school districts exceeded their cap during this seven-year period. The rest of the state is the least affected region in all years except 2002–03. By 2003–04, all regions began to converge, partly because the percentage of districts over their cap fell significantly in the Bay Area and in all Southern California counties. Each bar in Figure 4.6 represents the percentage difference between reported and funded ADA, focusing on districts that exceed their adult education ADA cap. Over 60 percent of overcap districts in Los Angeles County exceeded their cap by 2.5 percent or less. However, this is not 38 1FSDFOUBHF      mm                              -PT"OHFMFT -PT"OHFMFT $PVOUZ QFSJNFUFS #BZ "SFB     $FOUSBM 4BO%JFHP 3FTUPG $BMJGPSOJB 7BMMFZ $PVOUZ UIFTUBUF 4063$&$BMJGPSOJB%FQBSUNFOUPG&EVDBUJPO 1SJODJQBM"QQPSUJPONFOU6OJU /05&5IFBOBMZTJTJTSFTUSJDUFEUPBEVMUTDIPPMTXJUISFQPSUFE"%"BUMFBTUFRVBMUPUIFJS "%"DBQ Figure 4.6—Percentage by W'JhHiVcShFFunded ADA Is Exceeded, 2004–05 the case elsewhere. For instance, all adult schools in the Central Valley enrolled students in excess of the 2.5 percent funding limit. Statewide, nearly 80 percent of adult schools exceeded the 2.5 percent limit. Furthermore, one-third of all districts outside Los Angeles County overenrolled by 20 percent or more. A significant number of districts that exceed their cap ADA of school districts, therefore, are not fully funded under the current funding formula.  Since certain community colleges also offer ESL courses, it is possible to examine whether the size or growth of the community college ESL population is correlated with the percentage of school districts that exceeded their ADA cap in the same region from 1998– 99 to 2003–04. A positive but insignificant effect of the percentage of adult schools over their ADA cap on the level of provision by community colleges is found (after controlling regional and yearly factors). One interpretation of a positive coefficient is that both providers respond to the same factors driving increases in enrollment not already captured by regional and yearly factors. Another interpretation is that community colleges increase 39 Teaching for Free? The extent to which the ADA cap affects the ability of school districts to provide enough high-quality instruction hours to students depends on whether adult schools receive the “appropriate” level of funding. One measure of what is appropriate is the dollar value of the difference between reported ADA and funded ADA. The dollar value of each ADA is the baseline revenue limit (BRL), and the product of the BRL and funded ADA is roughly the total funding that a school district receives for adult education that year. The product of BRL and reported ADA is the funding that the school district needs to be fully compensated for the instructional load it undertakes that year. Figure 4.7 graphs the percentage by which the value of funded ADA needs to increase to equal the value of reported ADA from 1998 to 2004. The Central Valley and Los Angeles perimeter are the two regions that consistently would have needed their funding levels to increase the most during this period. From 1998–99 to 2002–03, these two regions tended to overenroll students whose dollar value was 10 percent or more above the funding formula amount. This difference declined to about 8 percent from 2003–04 to 2004–05. Although school districts in San Diego County do not have a consistent pattern, in the early 2000s the dollar value of their overenrollment was between 11 and 15 percent above their funded level. For adult schools in Los Angeles County generally, the value of overenrollment was between 1 and 3 percent above their funded levels. This could be because adult schools there do not face as much demand for ESL courses as noted above or they chose to limit enrollment to or near their funded ADA cap. The Bay Area and the rest of the state are more like Los Angeles County than are the other three regions. Yet adult schools there still exceed their funding levels by 4 to 6 percent, about twice as much as the funding growth permits. In 2001–02, the funded amount for adult ADAs fell significantly short of the dollar value of enrollment, causing the graphed differences to spike. This gap was driven by a combination of enrollment growth (10% increase in 2001–02 from the previous year) and midyear cuts to adult education their level of provision when adult schools face pressure on their ADA. See Appendix Table B.5. 40 $FOUSBM7BMMFZ #BZ"SFB -PT"OHFMFT$PVOUZ -PT"OHFMFTQFSJNFUFS 4BO%JFHP$PVOUZ 3FTUPGUIFTUBUF   1FSDFOUBHF    m m m m m m m 4063$&$BMJGPSOJB%FQBSUNFOUPG&EVDBUJPO 1SJODJQBM"QQPSUJPONFOU6OJU  Figure 4.7—Percentage Change 'inJHVDSFollar Value of Funded ADA Needed to Reach Parity with Reported ADA, 1998–99 to 2004–05 (among other programs). After 2001–02, however, the percentage difference between the dollar value of enrollment and funding decreased, perhaps because adult schools cut back on enrollment. Nevertheless, even with the convergence between reported and funded ADA, adult schools continue to receive insufficient funding. Figure 4.8 shows the difference in dollar value between funded and actual ADA to better reflect the size of the adult education program (as well as the policy response to demand for ESL). Despite the previously documented relatively lower growth in the need for ESL, Los Angeles County experienced the largest dollar value of enrollment in excess of funding, perhaps because, given the large size of the districts there (such as LAUSD), a small percentage increase translates into a large dollar value. It is also possible that adult schools with greater physical capacity can 41 $FOUSBM7BMMFZ #BZ"SFB -PT"OHFMFT$PVOUZ -PT"OHFMFTQFSJNFUFS 4BO%JFHP$PVOUZ 3FTUPGUIFTUBUF    7BMVF NJMMJPOT     m m m m m m m 4063$&$BMJGPSOJB%FQBSUNFOUPG&EVDBUJPO 1SJODJQBM"QQPSUJPONFOU6OJU  'JHVSF Figure 4.8—Difference Between Dollar Value of Funded ADA and Reported ADA, 1998–99 to 2004–05 overenroll more students than smaller adult schools. At its peak, the deficit in Los Angeles County was $10 million. The Los Angeles perimeter consistently has a difference of more than $5 million. The difference in the Central Valley and the rest of the state is similar to the amount in the Los Angeles perimeter in most years ($3 million to $4 million), whereas the Bay Area’s difference is $2 million to $4 million in most years. The accumulated amount that adult schools are underfunded is $172 million during this seven-year period (see Appendix Table B.6). The continued budget problems since 2001 have been dealt with by deferring payments to the next fiscal year, starting in 2002–03. Yet in 2003–04, total funding fell by more than 2 percent, and enrollment declined by slightly less than 2 percent. In 2004–05, funding increased to $591 million, surpassing the previous high set in 2002–03. This restoration 42 of funding is indicated in Figure 4.8 by the decline in funding shortfall in 2004–05. Summary More ESL students enroll in adult schools than in community colleges or any other providers. Yearly, adult schools teach English to over 500,000 students, a trend that seems to be correlated with state funding. The recent budget problems have disproportionately reduced the growth in ESL enrollment. Because ESL programs make up more than 40 percent of adult school enrollment, this cap on adult education growth especially constrains ESL enrollment. In 2004–05, nearly 60 percent of all adult school districts exceeded their ADA enrollment growth cap as determined by the funding formula. Of districts outside Los Angeles County that exceeded their cap ADA, more than 70 percent of them exceeded the 2.5 percent funding growth limit. This finding and the finding that 60 percent of Los Angeles County adult schools have enrollment within the 0–2.5 percent range are consistent with the state’s demographic trends. That the bulk of the state’s adult schools exceed the 2.5 percent limit is indicative of the decision by adult schools to carry out their education mission without the appropriate level of funding. The dollar value of exceeding the 2.5 percent limit on ADA growth is more than $15 million per year, possibly because districts are caught between serving students who wish to learn English and the financial constraint the state places on them. 43 5. Policy Implications and Conclusion Adult schools must deal with two contradictory policies regarding ESL programs in the state when faced with increasing levels of need in excess of the 2.5 percent growth cap. The evidence presented in Chapter 4 supports the view that the majority of local agencies do not abide by the 2.5 percent growth limit when local demand for adult education exceeds the level of funding. In practice, the cap on ADA growth may affect only the funds received by school districts rather than the level of instruction they provide. What are the implications for policymakers resulting from the confluence of the demographic changes, the antiquated funding formula, and the education mission that local educators are charged with carrying out? Because what matters is how educators and local districts carry out their adult education mission, this chapter examines the consequences resulting from the difference between reported and capped ADA funding as described in the previous chapter. Even though many adult schools are enrolling more students than the funding formula would suggest, there are short- and long-term consequences. Specifically, this chapter considers how the quality of instruction, the ability to implement new programs, and outreach efforts suffer when adult education funding does not correspond to the actual level of instruction provided. Adult schools that exceed their ADA cap are likely forced to make tradeoffs in the level and type of services they provide as well as in their investment in innovative programs. Other providers likely experience the same issues. Although the discussion below focuses on the funding for adult schools, it is applicable to other providers facing funding constraints, such as community colleges and public libraries. Policy Consequences of Overenrolled Courses In 2004–05, nearly 60 percent of all districts in the state exceeded their enrollment caps, and the difference between what all districts received from the state ($591 million) and the dollar value of actual enrollment ($607 45 million) is $15.7 million. There are two possible interpretations of this gap. The optimistic view is that despite the apparent shortcoming, in practice the funding formula does not hinder the ability of school districts to fully meet the English instruction needs of immigrants, since so many school districts behave in a manner inconsistent with the funding formula’s limits. Therefore, the mission of adult schools may not be adversely affected by the funding formula, especially in the short run or when capacity is exceeded by enrollment of only a few students. Yet, even if local agencies still meet the needs of most English learners, by absorbing the marginal cost of this service, the quality of instruction suffers and the ability of adult schools to grow and better serve English learners is hindered. Thus, the less optimistic interpretation suggests that the funding formula has an adverse effect on adult schools. This occurs because the state is under no obligation to fund any ADA in excess of 2.5 percent of the previous year’s funded ADA, resulting in a net ADA rate that is less than the baseline revenue limit per ADA. For instance, in 2004–05, San Leandro Unified in the Bay Area received $1,188,972 in funding for 518.69 ADAs but reported enrolling 577.93 ADAs; although the baseline revenue limit is $2,292.26, the net or effective per-ADA rate is $2,057.29 ($1,188,972 ÷ 577.93). In the short run, adult schools accommodate unfunded students by increasing classroom size or mixing students of different levels into one classroom. The short-run adjustments (i.e., continuously overenrolling classes) may negatively affect learning because of diminishing returns to scale with regard to teaching quality. Adult schools may also respond by offering fewer advanced classes, scheduling fewer classes that meet during off-peak hours, or eliminating marketing or community outreach efforts that increase awareness of and interest in the program. In the long run, it is also possible that the inability of adult schools to expand their instructional capability limits not just the number of students who can enroll but also the development of innovative programs and curriculum. For instance, affected adult schools may also be restricted in the ways they can target new funds to accommodate emerging needs,  See the discussion in Chapter 1 about the interpretation of the state’s mission to English learners. 46 such as illiterate immigrants. Inevitably, the inability to expand may deter potential students from attempting to enroll or adult schools may have to turn away students. Partial Reform of the Adult School Funding Formula Many of those involved in studying or drafting policy for adult education describe the current funding formula as “antiquated” (Assembly Select Committee on Adult Education, 2003; de Cos, 2004; Levine Sherriff, 2003). This section examines the effects of AB 23, which reformed one aspect of the funding formula. It also describes the extent to which the funding formula fails to meet the funding needs of adult schools. The current level of funded ADA for each adult school district is a direct result of the levels established after Proposition 13. Many communities have experienced growth, but others have experienced declines in the need for adult education courses since the funding formula was implemented. Until the passage of AB 23 in 2005, adult schools that experienced growth exceeding 2.5 percent were not eligible to receive additional funding despite the fact that other low-demand districts did not use all of their allocated ADAs. Adult schools that did not earn their allocated ADAs had to refund unused ADA revenue to the Proposition 98 reversion fund (de Cos, 2004). The funding formula was not structured to incorporate regional variations in need; AB 23 addresses this shortcoming. The law creates a formula that permits high-demand districts to receive funds allocated to districts that experience continued enrollment declines. AB 23 specifies that, beginning in 2006–07, school districts that do not reach their ADA cap for two consecutive years will have the cap permanently reduced by an amount equal to half of the lowest unearned ADA from these two years. Since the redistribution of unearned funded ADAs cannot finance all unfunded ADAs, the bill gives priority to small districts (fewer than 100 ADAs), although the mechanism also ensures that larger districts (100 or more ADAs) receive additional funding.  An unearned ADA is equal to the difference between a district’s ADA cap and funded ADA.  Details of the bill are available at http://www.leginfo.ca.gov/pub/bill/asm/ab_00010050/ab_23_bill_20051004_chaptered.pdf. 47 The first distribution of additional ADAs will take place in 2006–07, but the two most recently certified data, the “R-1” version, are used to consider the effect of this bill. Hence, the simulated redistribution of ADAs in Figure 5.1 pertains to the 2005–06 fiscal year using 2003–04 and 2004–05 data (school-district-level ESL ADA was not made available for this report). This policy makes 3,326 ADAs available for distribution, with the bulk (2,603) coming from large districts. The rest of the state (720) and Los Angeles County (1,050) are by far the total largest contributors of unearned ADAs in the state. The Los Angeles perimeter receives the most additional funded ADAs of any region (1,271), double the amount received by Los Angeles County adult schools (626). To the extent that overenrolled districts face demand for ESL in excess of 2.5 percent, Figure 5.1 suggests that there is some variability in need for ESL courses within regions. For instance, in the Central Valley, small districts lose and gain nearly the same amount of ADAs. Nevertheless, the main effect of this reform comes from the redistribution of ADAs from one region to another. The net gain in ADAs from AB 23 is positive only for the Central Valley (208), the Los Angeles perimeter (858), and the rest of the state (23). Adult schools in the Bay Area (–391), Los Angeles County (–404), and San Diego County (–301) lose more than they gain. However, using reported ADAs to factor in the size of the adult education program gives insights into the relative effect of the partial reform. This approach reveals that the Los Angeles perimeter benefits the most of any region. Its gain represents a 3 percent increase in funded ADAs, whereas the Central Valley gains an additional 1.25 percent in ADAs, and the rest of the state gains less than 0.1 percent. San Diego County experiences the most negative effect, as it permanently loses 2.75 percent of its funded ADAs, according to this simulation. The Bay Area and Los Angeles County lose 1 and 0.3 percent of their allocated ADAs, respectively—relatively modest percentages considering that adult schools  One potential effect not considered is if school districts that would potentially be affected attempt to increase or maintain steady enrollment levels.  In 2005–06 and 2006–07, several districts are ineligible for any growth, except for a 0.5 percent increase solely for nursing programs. This potential small growth is assumed to be zero for simplicity and because this restriction is valid for only two fiscal years. 48 /VNCFSPG"%"VOJUT   3FEVDUJPO TNBMMEJTUSJDUT  3FEVDUJPO MBSHFEJTUSJDUT  *ODSFBTF TNBMMEJTUSJDUT  *ODSFBTF MBSHFEJTUSJDUT         m m m m m m  m  $FOUSBM 7BMMFZ #BZ -PT"OHFMFT -PT"OHFMFT 4BO%JFHP 3FTUPG "SFB $PVOUZ QFSJNFUFS $PVOUZ UIFTUBUF 4063$&"%"JOGPSNBUJPOGPSm 3 BOEm 3 JTQSPWJEFECZUIF $BMJGPSOJB%FQBSUNFOUPG&EVDBUJPO 1SJODJQBM"QQPSUJPONFOU6OJU /05&5IFGJHVSFJODMVEFTBGFXEJTUSJDUTOPUFMJHJCMFGPSSFEJTUSJCVUJPOQFS"# Figure 5.1—Redistributed A'JDHVASFsUnder AB 23, by Region, 2005–06 in these regions account for half of the unearned ADAs. Because districts that permanently lose ADAs are not using them, these reductions should not have an adverse effect on the adult education programs of these schools. How Much Does Partial Reform of the Funding Formula Help? These additional ADAs, however, do not directly measure the increase in the number of students that are now funded. In Table 4.1, it was shown that California ESL students spend an average of 113 hours in an ESL class, but they need 216 hours of instruction to successfully complete a beginning ESL course. Under the first assumption, 4.7 additional students can be taught and 2.4 additional students per ADA under the 49 /VNCFSFOSPMMFE second assumption. In total, 15,453 and 8,084 more students are funded statewide, respectively. Figure 5.2 shows the regional distribution of these students as a consequence of AB 23. Adult schools in the Los Angeles perimeter teach about 5,900 and 3,100 more students under the first, and then second, assumptions. Estimates under the first assumption are 3,500 and 3,000 for the rest of the state and Los Angeles County and 1,800 and 1,600, respectively, under the second assumption. The Bay Area and Central Valley would teach about 1,400 more students for 113 hours of instruction   IPVSTPGJOTUSVDUJPO IPVSTPGJOTUSVDUJPO            $FOUSBM 7BMMFZ /05&4FF'JHVSF #BZ -PT"OHFMFT -PT"OHFMFT 4BO%JFHP 3FTUPG "SFB $PVOUZ QFSJNFUFS $PVOUZ UIFTUBUF Figure 5.2—Increase in ESL Enro'lJHlmVSeFntas a Result of Redistribution of ADAs for Different Instructional Goals  Each ADA is 525 student-hours, so 525/113 = 4.65 students and 525/216 = 2.43 students. Multiplying these values by the ADAs yields the estimates.  Using the share of ESL enrollment in the region yields estimates very similar to the assumption of 216 hours of instruction. 50 and about 750 hours for 216 hours of instruction. San Diego County adult schools would teach 259 and 136 additional students under these two assumptions. Even the largest estimate of the additional funded students represents a small share of the unmet need as estimated in Chapter 3 or in the number of districts that exceed their ADA cap. The dollar value of the additional ADAs resulting from AB 23 is $7.6 million, whereas the estimated amount of unfunded ADA in 2004–05 is $15.7 million. In other words, even if the difference between what adult schools receive and the level of instruction they provide represents all of the unmet need for ESL instruction, AB 23 addresses less than half of this need. But it is more than likely that this gap does not represent the full level of need for ESL courses in the state. Policy Implications The low level of funding for adult education is considered the most important issue facing all adult education providers, including adult schools, community colleges, community-based organizations, religious organizations, and libraries (Crandall and Sheppard, 2004, p. 12). Any comprehensive strategy to increase the provision of ESL courses to adult immigrants through additional funding must include all providers, not just adult schools. Even though this report’s focus is on adult schools, the policy discussions pertaining to the findings are also relevant for other providers. The funding formula effectively compensates adult school enrollment at a lower per-pupil level than the baseline revenue limit. Adult schools that do not receive funding equal to the actual level of enrollment face tradeoffs between providing instruction in a given year or expanding instruction in subsequent years, as well as providing other services that encourage enrollment, such as child care. As already mentioned, the quality of instruction is lowered as a consequence of not fully funding the level of instruction in districts that exceed their ADA cap. This effect occurs in both the short and long run. Yet focusing on current needs potentially reduces the availability of funds for investment in new curricula, facilities, or technology to improve the learning environment. Two examples of innovative curriculum or teaching methodology that may be affected are distance learning and workplace ESL instruction. 51 These types of programs are particularly well-suited to address various issues in adult ESL. They aim to make the ESL curriculum more effective and to target hard-to-reach groups, such as those with inconsistent work schedules, lack of transportation, or hectic lifestyles. From the student’s point of view, these types of programs are attractive because they are also flexible with regard to pace and scheduling and because they focus the curriculum on topics most useful or of most interest to them. From the provider’s point of view, they reduce the physical constraints associated with expanding programs at the adult school site as well as potentially improving the program. It thus makes it easier for the provider to expand without fear that the quality of instruction suffers or that it may crowd out other programs. Distance learning (DL) is a technology-based delivery system that connects students with instructors even though they may be in different locations or not present at the same time. The motivations for complementing traditional in-class instruction with distance learning differ. In most cases, distance learning is considered an effective means by which to reach students who may otherwise participate less because of personal or other obstacles. For instance, the LAUSD application for program approval states, “Based on information gathered at the school sites, much of our adult ESL population had scheduling, travel or child care issues which made traditional school difficult to attend. DL also allowed ambitious adults to learn English faster. . . . Schools also reach hard-to-serve adults by setting up DL classes at businesses, factories, retail centers, churches, recreation centers and apartment complexes.” Since 1993, adult schools have been able to use no more than 5 percent of their funding for distance  The formal definition of DL is: “An instructional delivery system that connects learners with educational resources. DL provides educational access to learners not enrolled in educational institutions and can augment the learning opportunities of current students. The implementation of DL is a process that uses available resources and will evolve to incorporate emerging technologies” (http://www.cdlponline.org/index.cfm?fuse action=whatis&pg=33). See the state’s Innovation and Alternative Instructional Delivery Program website for full program description (http://www3.scoe.net/fivepercent/login. cfm?fuseaction=desc).  http://www3.scoe.net/fivepercent/pdfs/applications/2005/2005_2361_10112005_ 20512.pdf, p. 8. 52 learning programs and, presently, 99 adult school agencies participate in this program, mostly large- and medium-sized adult schools.10 Vocational ESL and workplace ESL are programs that aim to link the classroom with the workplace in a very direct way. Instruction in workplace ESL programs in fact takes place not in the classroom but at the English learner’s place of work.11 This program is generally provided in partnership with employers who provide the space and other resources, and the school system provides the curricular materials and the instructor. Employees attend classes during work hours or when they are not working. The content has immediate returns for the student and the employer and, because instruction is conveniently located and scheduled, students may be more likely to persist and complete more hours of instruction than they would in traditional in-class programs. This saves the district money in the long run because students are less likely to continually drop in and out of classes before completing their goal. Adult schools are not the only providers of these types of innovative programs. Nonprofit organizations, churches, public libraries, and community colleges also have courses that are geared toward the community’s particular needs (Crandall and Sheppard, 2004). These organizations may be able to provide smaller classes or may be able to foster greater trust and confidence from the community because the provider, such as a local church, is well known to them. Besides the effect on investment in new and effective programs, an insufficient level of funding may force adult schools to provide more ESL courses at the expense of other adult education programs. Schools faced with substantial demand for ESL can accommodate more students without substantially exceeding their enrollment cap by reducing the funding spent on other programs. For example, under a budget-neutral scheme, adult schools could shift more spending to ESL programs to teach an additional 100,000 students by proportionately reducing all other adult education programs. Under this scenario, 72 percent of this shift would come from 10 See the state’s Innovation and Alternative Instructional Delivery Program website for full program description (http://www3.scoe.net/fivepercent/login.cfm?fuseaction=desc). 11 Outreach and Technical Assistance Network (OTAN) provides significant background on vocational ESL programs (http://www.otan.us/wia/0607/pdfs/vesldoc.pdf). 53 reductions in vocational education, adult secondary education, and older adult programs.12 It is possible that some agencies are already practicing such a policy.13 Some policymakers, already facing budget constraints, may be reluctant to consider a policy of increased funding for all adult education programs. However, as other options require making tradeoffs between various adult education programs, policymakers may find it difficult to find a budgetneutral policy that benefits one constituency at the expense of another. The additional funding would permit affected providers to invest in teaching methods or facilities, such as distance learning, that enhance their ability to serve more students and reach out to students who are not being currently served. The additional funding would not just focus on adult schools but would also fund community colleges, public libraries, and nonprofit organizations. In the long run, such investment is likely to provide many positive benefits for the state and the immigrants wanting to learn English. Summary This chapter considered the policy implications of the findings from this report. The dollar value of exceeding the 2.5 percent limit on ADA growth is more than $15 million per year, possibly because districts choose to serve students who wish to learn English rather than turn them away. Yet, districts that do not receive full funding for all the ADAs they teach are likely constrained from expanding or adjusting to meet the needs of immigrants in the short and long term, since they incur the cost of exceeding their 2.5 percent growth cap. Assembly Bill 23 is the first attempt to deal with one of the shortcomings of the adult school funding formula. The bill makes it possible to redistribute unspent adult education funding to districts that exceed their ADA cap. This reform is a first step toward increasing the number of available ESL courses, but it fills less than 50 percent of the gap of the dollar value of the difference between enrollment and state funding. Nevertheless, certain adult schools in the state, especially those in the Los 12 This assumes a one-to-one exchange between programs. The average number of hours of instruction in ESL classes is 13 to 30 hours greater than in these programs. 13 I would like to thank Patricia de Cos for pointing out this possibility. 54 Angeles perimeter, benefit the most from this reform. Because redistributed ADAs would otherwise go unused, this benefit comes at a minimal cost to adult schools with declining enrollment. The inflexibility of the adult school funding formula continues to raise important questions about whether adult schools are sufficiently funded to accomplish their mission of teaching English learners. Other public providers of ESL courses also face funding shortages, and any policy must include these providers to better reach the population in need of ESL instruction. Consequently, adult education programs may be making difficult choices in accommodating the needs of present students. They may be sacrificing investing in new programs that would better accommodate future cohorts of students, or they may be reducing the availability of resources for other groups of adult learners. State policymakers must weigh whether increasing funding for adult education programs in the face of growing demand for ESL programs is a policy worth exploring, while they also consider the alternatives. Yet given the public interest over the pace of immigrant acquisition of English, policymakers must also recognize that the lack of available ESL courses may perpetuate the linguistic divide in California. 55 Appendix A Notes on Data and Methods Census Data The 1980, 1990, and 2000 5% PUMS datasets provide valuable information regarding the English-language ability of immigrants in California. The sample consists of foreign-born persons ages 18 years and older who are not U.S. citizens, although the sample also includes persons born in Puerto Rico. An LEP is defined as a person speaking English “not well” or not speaking any English. See the NHES discussion below for further sample restrictions made to estimate enrollment. The regional areas are constructed from the Census data using the most consistent grouping of counties across the three PUMS datasets, which do not overlap completely in each year. For this reason, certain counties are included in geographic regions that generally would not include them. For instance, San Benito County is included in the San Joaquin Valley group even though this county is not considered part of the Central Valley. The regional definitions used in this report conform to those used by CASAS in its reports. The regions are Los Angeles County; Bay Area—Alameda, Contra Costa, San Francisco, Santa Clara, and San Mateo Counties; Central Valley—Fresno, Kern, Merced, and Tulare Counties; Los Angeles perimeter—Orange, Riverside, San Bernardino, and Ventura Counties; San Diego County; and the rest of the state—all other counties. Estimates of Enrollment Data The two data sources used to estimate the probability of ESL enrollment are the NHES for 1999 and 2001. A virtue of these datasets is that they are nationally representative (with a small California sample) of persons ages 16 and older not enrolled in elementary or secondary school (the surveys are conducted in Spanish or English). The extensive language questions of these data provide valuable information on the nation’s 57 immigrant groups that participated in the surveys. The NHES asked eligible adults whether they enrolled in an ESL course in the previous 12 months. The sample drawn from these data includes persons born outside the United States and with an identifiable country of origin (Puerto Ricans could not be identified from the data and so anyone born in U.S. territories are exclude from the sample), at least age 18, and not enrolled in an ESL college program. The predicted enrollment estimates yield the size of the population with the characteristics described above. More information about the NHES can be found at http://nces.ed.gov/nhes. Since this report focuses on the policy issues dealing with adult schools and community colleges, the sample should ideally consist of those who enrolled only in these agencies. However, because of the small sample size of ESL enrollees, the sample includes persons who enrolled in adult schools, community colleges, community-based organizations, religious organizations, other public providers (such as libraries), private institutions, and other agencies. Of the final number who enrolled (145) in an ESL class in the previous 12 months, 112 enrolled in adult schools or community colleges, 17 in community-based or religious organizations, six in other public providers, nine in private business, and one did not specify. The estimates represent the total enrollment of all providers in the state, not just adult schools and community colleges. See Appendix Table B.4 for results of the estimates using only adult school and community college attendees. Another important limitation is that the NHES surveys were conducted in either Spanish or English, meaning that non-Spanish speakers who do not speak English well enough to participate in the survey may not be included, and therefore the estimates may not reflect the role of this population. A necessary assumption of the model is that those not surveyed regarding their ESL enrollment status because of low English skills would have answered the ESL question the same, on average, as Spanish speakers with low English skills. These individuals may more likely be enumerated in the Census, since lack of English is not a basis for exclusion. A minor issue with the 2001 NHES is also worth discussing. The 2001 NHES is the most recent NHES dataset to include questions on ESL  For consistency with previous years’ ESL question, the 2001 NHES ESL question was recoded to missing if the respondent’s first language was not English. Only nine cases were recoded to missing. 58 attendance. In both years, persons whose first language learned was not English were asked about the language they speak at home. Those speaking a language other than English were asked whether they had taken an ESL course in the past 12 months. However, in 2001, but not in 1999, persons who currently speak only English were also asked for ESL enrollment information if their first language was not English. Thus, for consistency across data, in 2001, responses for nine English-only speakers were recoded to missing. Empirical Model The following logit model is estimated for the combined 2001/1999 sample: ESLi = Xiβ + Califi′D + ui , (A.1) where ESL equals 1 if individual i took an ESL course and 0 otherwise. Xi consists of age, age at arrival, whether Hispanic, country of origin, highest education level completed, marital status, whether female, family size, and a California dummy variable. Our sample predictions for California are obtained by estimating β and using the values of Xi from the 5% PUMS for each year (1980, 1990, and 2000). Several sample restrictions have to be made to the Census data to make them compatible with the NHES sample: age greater than 17, not enrolled in school, not born in Puerto Rico (because it is not identified in the NHES data), not an English-only speaker at home, and not born in North America, Australia, or New Zealand. Using PUMS weights yields individual representative probability of enrollment and averaging within a region provides the expected number of persons who enroll in all ESL classes. The marginal effects are estimated at the means of the variables for the logit results, provided in Appendix Table B.1. However, to align as closely as possible the 2000 enrollment estimates with the 1999–00 fiscal year enrollment levels, it is necessary to apply the 2001–02 regional distribution of ESL programs. Regional information is not available for 1999–00. The regional distributions are stable, however, as the 2002–03 distribution is very similar to the 2001–02 distribution. It is unlikely, therefore, that the imputed regional distribution of ESL enrollment in 1999–00 suffers from significant measurement error. 59 Adult School and Community College Data This section provides more detailed information about adult school and community college data sources. For adult schools: 1. Annual enrollment counts and demographic information for ESL students in adult schools were obtained from the CASAS DynaReports website, http://www.casas.org/dynareps/dynareps2002.cfm. 2. ADA information and finance information for adult schools were obtained from Halena Le at the California Department of Education, Principal Apportionment Unit. For community colleges: 1. Community college data were obtained from the Management Information System department of the California Community College Chancellor’s Office. These data contain enrollment information at the class level. 2. Unduplicated student counts are used, as measured by the agencyspecific student ID. Thus, the same person may enroll in a community college or adult school. The enrollment data are based on student counts in noncredit courses and as such the sample does not exclude students that potentially were also enrolled in credit courses. 3. Finance data were provided by Ed Monroe and Elias Regalado at the Fiscal Services division at the CCCCO. 60 Appendix B Estimates and Results Table B.1 Logit Regression Results of the Probability of ESL Enrollment, NHES   Age Age at arrival Hispanic Mexico South and Central America Caribbean Asia Middle East, Africa, and other Southeast Asia Grades 0–3 Grades 9–11 Grades 12–15 Grades 16+ Married, spouse present Female All Providers Coefficient Marginal Effect   –0.100** (0.017) 0.073** (0.021) 2.022** (0.494) –2.146** (0.806) –2.049* (0.827) –1.400+ (0.821) –0.671 (0.739) –1.860* (0.762) –1.060 (0.832) –1.221+ (0.641) –0.819+ (0.475) –0.276 (0.349) –0.212 (0.409) –0.067 (0.327) –0.506+ (0.277) –0.005** (0.001) 0.004** (0.001) 0.089** (0.022) –0.120* (0.052) –0.065** (0.018) –0.046** (0.017) –0.029 (0.026) –0.056** (0.013) –0.039+ (0.020) –0.042** (0.014) –0.035* (0.017) –0.014 (0.017) –0.011 (0.020) –0.004 (0.018) –0.027+ (0.015) Adult Schools and Community Collegesa Coefficient Marginal Effect –0.096** (0.019) 0.077** (0.022) 2.448** (0.530) –0.758 (1.159) –0.652 (1.174) –0.008 (1.183) 0.503 (1.207) –0.465 (1.158) 0.176 (1.334) –1.103 (0.705) –0.677 (0.521) –0.111 (0.372) –0.058 (0.504) –0.303 (0.369) –0.594* (0.293) –0.003** (0.001) 0.003** (0.001) 0.066** (0.016) –0.025 (0.039) –0.018 (0.027) –0.000 (0.039) 0.020 (0.057) –0.013 (0.028) 0.006 (0.051) –0.024* (0.010) –0.019 (0.012) –0.004 (0.012) –0.002 (0.016) –0.011 (0.013) –0.020+ (0.010) 61 Table B.1 (continued) All Providers Adult Schools and Community Collegesa   Coefficient Marginal Effect   Coefficient Marginal Effect Family size –0.041 –0.002 0.024 0.001 (0.085) (0.005) (0.088) (0.003) California 0.673* 0.040* 1.030** 0.041** (0.280) (0.018) (0.328) (0.014) Constant 0.181 –2.440+ (0.938) (1.348) No. of observations 1,083 1,083   1,050 1,050 SOURCE: Author’s calculations from the 1999/2001 NHES. NOTES: Standard errors are in parentheses. The sample is restricted to persons who are age 18 or older and did not enroll in an ESL college program. Marginal effects are estimated at the means of the variables. aRestricted to respondents indicating type of agency attended as adult school or community college. +Significant at 10 percent. *Significant at 5 percent. **Significant at 1 percent. 62 63 Table B.2 Mean Characteristics of Immigrant Samples in Census and NHES Census 1980 1990 Age 41.68 40.88 Age at arrival 30.27 28.76 Hispanic 0.56 0.58 Mexico 0.43 0.43 South and Central America 0.08 0.11 Caribbean 0.02 0.01 Europe 0.16 0.09 Asia 0.14 0.15 Middle East, Africa, and other 0.06 0.06 Southeast Asia 0.11 0.14 Grades 0–3 0.18 0.19 Grades 4–8 0.26 0.18 Grades 9–11 0.12 0.11 Grades 12–15 0.31 0.38 Grades 16+ 0.13 0.14 Married, spouse present 0.65 0.58 Female 0.52 0.50 Family size 3.93 4.15 No. of observations 104,739 159,178 Population size 2,096,226 3,236,796 NOTE: See Appendix A for sample selection. 2000 42.41 25.12 0.58 0.47 0.11 0.01 0.06 0.17 0.03 0.16 0.14 0.18 0.11 0.39 0.18 0.67 0.50 4.32 306,744 6,189,076 NHES 1999/2001 37.93 23.20 0.68 0.45 0.15 0.07 0.06 0.12 0.09 0.06 0.06 0.26 0.18 0.31 0.19 0.69 0.47 4.04 1,083 26,808,454 Los Angeles County Census Sample 1990 2000 39.97 42.63 28.80 25.25 0.65 0.63 0.43 0.46 0.17 0.17 0.02 0.01 0.07 0.06 0.15 0.16 0.07 0.03 0.10 0.11 0.18 0.14 0.13 0.16 0.20 0.19 0.12 0.12 0.37 0.39 0.55 0.64 0.50 0.50 4.18 4.26 72,507 122,332 1,478,753 2,486,533 Non–Los Angeles County Census Sample 1990 2000 41.66 28.73 0.53 0.42 0.07 0.01 0.10 0.16 0.06 0.18 0.19 0.17 0.10 0.39 0.15 0.60 0.50 4.13 86,671 1,758,043 42.26 25.03 0.54 0.48 0.07 0.00 0.07 0.17 0.03 0.19 0.14 0.17 0.10 0.40 0.19 0.69 0.50 4.35 184,412 3,702,543 Table B.3 Mean Characteristics of Enrolled ESL Students   Adult Community Schools Colleges Female 0.56 0.56 Male 0.44 0.44 No high school diploma High school diploma or GEDa 0.60 0.25 0.66 0.29 Post–high school 0.15 0.05 Ages 16–20 0.11 0.13 Ages 21–30 0.35 0.36 Ages 31–40 0.27 0.25 Ages 41–50 0.14 0.14 Ages 51–60 0.08 0.06 Ages 61+ 0.04 0.05 White 0.06 0.08 Hispanic 0.73 0.67 Asian 0.17 0.22 Filipino or Pacific Islander 0.02 0.01 Other 0.03 0.02 SOURCES: CASAS DynaReports Adult School data for 1999–00 to 2003–04, and CCCCO for 1997–98 to 2003–04. NOTE: The CCCCO sample includes unduplicated students with nonmissing information for that variable. aGED is general equivalency diploma. 64 Table B.4 Predicted Enrollment Probabilities and Levels of Enrollment for Adult School or Community College ESL Enrollees Probability of Enrollment (%) Predicted Enrollment Central Valley Bay Area Los Angeles County Los Angeles perimeter San Diego County 1980 12.5 7.9 14.6 13.1 10.6 1990 2000 14.9 10.7 8.7 7.0 14.4 8.8 14.8 9.4 11.5 7.9 1980 10,686 30,601 149,696 34,086 12,817 1990 20,753 47,834 212,483 78,348 24,489 2000 30,445 81,488 219,907 107,974 32,117 Rest of the state 10.0 11.6 9.5 21,178 37,804 66,855 Total 12.4 13.0 8.7   259,064 421,710 538,785 NOTE: Estimates are based on the second regression in Table B.1. 65 Table B.5 Effect on Noncredit ESL Enrollment of Districts Exceeding Their ADA Cap Coefficient Estimate Coefficient Estimate Percentage of adult schools over their ADA cap 43,564.615** 6,280.917 (15,684.726) (5,639.659) Year = 1999 2,308.697+ (1,175.183) Year = 2000 3,158.999* (1,315.961) Year = 2001 4,693.076** (1,131.226) Year = 2002 3,045.705* (1,116.420) Year = 2003 –648.739 (1,466.803) Bay Area 14,558.988** (1,688.464) Central Valley –13,796.300** (1,127.439) Los Angeles perimeter 28,290.014** (1,574.876) Los Angeles County 8,634.902** (1,429.417) San Diego County 15,568.116** (1,215.532) Constant 2,405.690 9,999.777** (8,095.254) (2,463.964) No. of observations 36 36 R-squared 0.18 0.99 SOURCES: Data are from CCCCO and California Department of Education, Principal Apportionment Unit, 1998–99 to 2003–04. NOTES: The dependent variable is the level of ESL provision by community colleges. Standard errors are in parentheses. +Significant at 10 percent. *Significant at 5 percent. **Significant at 1 percent. 66 Table B.6 Data for Figures 4.7 and 4.8   1998–99 1999–00 2000–01 2001–02 2002–03 2003–04 2004–05 Value of Reported ADAs ($ millions ) Central Valley 35.1 36.5 37.2 41.3 41.0 39.4 39.4 Bay Area 73.4 73.9 76.8 88.6 90.9 89.0 85.8 Los Angeles County 241.3 245.5 256.0 290.1 301.2 297.5 298.8 Los Angeles perimeter 55.8 57.5 59.4 64.7 66.3 64.4 65.5 San Diego County 24.6 24.1 25.2 28.0 29.3 26.3 25.2 Rest of the state 71.2 73.4 74.3 82.4 84.7 82.4 92.4 Total 501.5 510.9 528.8 595.1 613.4 599.1 607.1 Value of Funded ADAs ($ millions) Central Valley 30.7 32.3 33.3 36.0 36.9 36.3 36.6 Bay Area 70.2 71.8 75.2 84.0 86.6 85.4 83.4 Los Angeles County 237.8 241.9 252.5 281.6 290.6 286.8 296.5 Los Angeles perimeter 51.1 52.6 54.4 58.4 60.4 59.4 61.3 San Diego County 22.5 22.8 22.7 24.3 25.9 24.8 24.7 Rest of the state 67.2 69.0 71.0 77.3 80.7 78.3 88.8 Total 479.4 490.2 509.3 561.6 581.1 571.1 591.3 Difference (Value of Reported ADA – Value of Funded ADA) ($ millions) Central Valley 4.4 4.2 3.9 5.2 4.1 3.1 2.8 Bay Area 3.2 2.1 1.5 4.6 4.3 3.6 2.4 Los Angeles County 3.5 3.6 3.5 8.5 10.6 10.7 2.3 Los Angeles perimeter 4.7 4.9 4.9 6.3 5.8 4.9 4.1 San Diego County 2.1 1.3 2.4 3.7 3.4 1.5 0.6 Rest of the state 4.0 4.4 3.3 5.2 4.0 4.1 3.6 Total 22.1 20.6 19.6 33.4 32.3 28.0 15.7 Percentage Difference Between Reported and Funded ADA Central Valley 14.4 13.1 11.7 14.6 11.2 8.4 7.5 Bay Area 4.5 3.0 2.0 5.4 5.0 4.3 2.9 Los Angeles County 1.5 1.5 1.4 3.0 3.6 3.7 0.8 Los Angeles perimeter 9.3 9.2 9.0 10.7 9.7 8.3 6.7 San Diego County 9.5 5.8 10.8 15.3 13.3 6.0 2.3 Rest of the state 6.0 6.4 4.7 6.7 5.0 5.3 4.0 Total 4.6 4.2 3.8 6.0 5.6 4.9 2.7 SOURCE: California Department of Education, Principal Apportionment Unit. 67 References Assembly Select Committee on Adult Education, Assessing and Addressing Need: A Local Provider’s Perspective Hearing, Testimony Provided by Santiago Jackson, Assistant Superintendent Division of Adult and Career Education, Los Angeles Unified School District, Sacramento, California, 2003. Bauer, Thomas, Gil. S. Epstein, and Ira N. Gang, “The Role of the Family in Immigrants’ Destination Language Acquisition,” Journal of Population Economics, Vol. 18, No. 4, 2005, pp. 631–647. Berman, Eli, Kevin Lang, and Erez Siniver, “Language-Skill Complementarity: Returns to Immigrant Language Acquisition,” Labour Economics, Vol. 10, No. 3, 2003, pp. 265–290. Bloom, David E., and Gilles Grenier, “Language, Employment, and Earnings in the United States: Spanish-English Differentials from 1970 to 1990,” International Journal of the Sociology of Language, Vol. 121, 1996, pp. 45–68. California Community Colleges Chancellor’s Office, Understanding Funding, Finance, and Budgeting: A Manager’s Handbook, Sacramento, California, 1999. California Department of Education, Adult Education Office, Adult Education & Family Literacy Act California State Plan 1999–2004 Extended Through 6/30/05, Sacramento, California, 2004. California Department of Education, Adult Education Office, Adult Education Handbook for California, 2005 Edition, Sacramento, California, 2005. California Department of Education, Adult Education Office, California Annual Performance Report: Federally Funded Workforce Investment Act Title II Programs Program Year July 1, 2004–June 30, 2005, Sacramento, California, 2006. Capps, Randolph, Michael E. Fix, Jeffrey S. Passel, Jason Ost, and Dan Perez-Lopez, A Profile of the Low-Wage Immigrant Workforce, Brief No. 4. The Urban Institute, Washington, D.C., 2003. Carliner, Geoffrey, “The Language Ability of U.S. Immigrants: Assimilation and Cohort Effects,” International Migration Review, Vol. 34, No. 1, 2000, pp. 158–182. 69 CASAS, Student Progress and Goal Attainment Report: Adult School Programs in California 2002–03, San Diego, California, 2004. Chiswick, Barry, and Paul W. Miller, Do Enclaves Matter in Immigrant Adjustment? IZA DP No. 449. IZA (Institute for the Study of Labor), Bonn, 2002. Chiswick, Barry R., and Paul W. Miller, “The Endogeneity between Language and Earnings: International Analyses,” Journal of Labor Economics, Vol. 13, No. 2, 1995, pp. 246–288. Crandall, JoAnn, and Ken Sheppard, Adult ESL and the Community College, Council for Advancement of Adult Literacy, New York, New York, 2004. de Cos, Patricia L., Educational Opportunities for Adults in California, California Research Bureau, California State Library, Sacramento, California, 2004. Dustmann, Christian, and Arthur Van Soest, “Language and the Earnings of Immigrants,” Industrial and Labor Relations Review, Vol. 55, No. 3, 2002, pp. 473–492. Gonzalez, Arturo, “The Acquisition and Labor Market Value of Four English Skills: New Evidence from NALS,” Contemporary Economic Policy, Vol. 18, No. 3, 2000, pp. 259–269. Hill, Laura E., The Socioeconomic Well-Being of California’s Immigrant Youth, Public Policy Institute of California, San Francisco, California, 2004. Jepsen, Christopher, and Shelley de Alth, English Learners in California Schools, Public Policy Institute of California, San Francisco, California, 2005. Johnson, Hans P., “A State of Diversity: Demographic Trends in California’s Regions,” California Counts: Population Trends and Profiles, Vol. 3, No. 5, Public Policy Institute of California, San Francisco, California, 2002. Johnson, Hans P., Belinda I. Reyes, Laura Mameesh, and Elisa Barbour, Taking the Oath: An Analysis of Naturalization in California and the United States, Public Policy Institute of California, San Francisco, California, 1999. Joint Board Task Force on Noncredit and Adult Education, Challenges Opportunities Changes, California State Board of Education and the Board of Governors of the California Community Colleges, Sacramento, California, 1998. Lazear, Edward P., “Culture and Language,” Journal of Political Economy, Vol. 107, No. 6, Part 2, 1999, pp. S95–S126. 70 Levine Sherriff, Rona, Adult Education: Will It Meet the Challenges of the Future? Senate Office of Research, Sacramento, California, 2003. Little Hoover Commission, We the People: Helping Newcomers Become Californians, Little Hoover Commission, Sacramento, California, 2002. Martinez, Tia Elena, and Ted Wang, Supporting English Language Acquisition: Opportunities for Foundations to Strengthen the Social and Economic Well-Being of Immigrant Families, The Annie E. Casey Foundation and Grantmakers Concerned with Immigrants and Refugees, Baltimore, Maryland, 2005. McManus, Walter S., “Labor Market Costs of Language Disparity: An Interpretation of Hispanic Earnings Differences,” American Economic Review, Vol. 75, No. 4, 1985, pp. 818–827. Moore, Colleen, Nancy Schulock, and David Lang, Funding Adult Education: Does California Put the Money Where the Needs Are? Faculty Research Fellows Program, Center for California Studies, California State University, Sacramento, California, 2004. National Center for Family Literacy and National Center for ESL Literacy Education at the Center for Applied Linguistics, “Practitioner Toolkit: Working with Adult English Language Learners,” available at http://www.cal.org/CAELA/tools/program_development/elltoolkit/ CombinedFilesl.pdf, accessed December 11, 2006. National Center for the Study of Adult Learning and Literacy, The First Five Years, Harvard University Graduate School of Education, Cambridge, Massachusetts 2002. Pew Hispanic Center, Hispanic Attitudes Toward Learning English, Washington, D.C., 2006. Portes, Alejandro, and Ruben G. Rumbaut, Immigrant America: A Portrait, University of California Press, Berkeley, California, 1996. Ramakrishnan, S. Karthick, and Mark Baldassare, The Ties That Bind: Changing Demographics and Civic Engagement in California, Public Policy Institute of California, San Francisco, California, 2004. West, Linda L., Meeting the Challenge: A History of Adult Education in California from the Beginnings to the 1990s, California Department of Education, Sacramento, California, 1995. Young, Malcolm B., National Evaluation of Adult Education Programs: Executive Summary, Development Associates, Arlington, Virginia, 1995. 71 About the Author ARTURO GONZALEZ Arturo Gonzalez is a research fellow at the Public Policy Institute of California. He was previously an associate professor at the University of Arizona and a visiting professor at the Carlos III University in Madrid. His research interests include the acquisition of human capital by lowskilled workers and immigrants, including studies on the effect of the Job Corps on the earnings of Hispanics and non-Hispanics, and the role of English-as-a-second-language classes in improving the English ability of adult immigrants. His publications include Mexican Americans & the U.S. Economy: Quest for Buenos Días (2002). He received a Ph.D. in economics from the University of California, Santa Barbara. 73 Related PPIC Publications Educational Progress Across Immigrant Generations in California (2005) Deborah Reed, Laura E. Hill, Christopher Jepsen, Hans P. Johnson English Learners in California Schools (2005) Christopher Jepsen, Shelley de Alth The Socioeconomic Well-Being of California’s Immigrant Youth (2004) Laura E. Hill The Ties That Bind: Changing Demographics and Civic Engagement in California (2004) S. Karthick Ramakrishnan, Mark Baldassare 75" } ["___content":protected]=> string(102) "

R 407AGR

" ["_permalink":protected]=> string(123) "https://www.ppic.org/publication/californias-commitment-to-adult-english-learners-caught-between-funding-and-need/r_407agr/" ["_next":protected]=> array(0) { } ["_prev":protected]=> array(0) { } ["_css_class":protected]=> NULL ["id"]=> int(8463) ["ID"]=> int(8463) ["post_author"]=> string(1) "1" ["post_content"]=> string(0) "" ["post_date"]=> string(19) "2017-05-20 02:37:49" ["post_excerpt"]=> string(0) "" ["post_parent"]=> int(3669) ["post_status"]=> string(7) "inherit" ["post_title"]=> string(8) "R 407AGR" ["post_type"]=> string(10) "attachment" ["slug"]=> string(8) "r_407agr" ["__type":protected]=> NULL ["_wp_attached_file"]=> string(12) "R_407AGR.pdf" ["wpmf_size"]=> string(7) "1277235" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(137366) "Public Policy Institute of California California’s Commitment to Adult English Learners: Caught Between Funding and Need ••• Arturo Gonzalez California’s Commitment to Adult English Learners: Caught Between Funding and Need ••• Arturo Gonzalez with research support from Rebecca Steinbach and Shannon McConville 2007 Library of Congress Cataloging-in-Publication Data Gonzalez, Arturo. California’s commitment to adult English learners: caught between funding and need / Arturo Gonzalez. p. cm. Includes bibliographical references. ISBN 978-1-58213-123-8 1. Adult education—California—Finance. 2. English language— Study and teaching—Foreign speakers. I. Title. LC5252.C2G66 2007 374.9794—dc22 2007007564 Copyright © 2007 by Public Policy Institute of California All rights reserved San Francisco, CA Short sections of text, not to exceed three paragraphs, may be quoted without written permission provided that full attribution is given to the source and the above copyright notice is included. PPIC does not take or support positions on any ballot measure or on any local, state, or federal legislation, nor does it endorse, support, or oppose any political parties or candidates for public office. Research publications reflect the views of the authors and do not necessarily reflect the views of the staff, officers, or Board of Directors of the Public Policy Institute of California. Summary The ability to speak English is a fundamental prerequisite for the integration of immigrants into California’s society. However, an unprecedented number of immigrants may be relying on English as a Second Language (ESL) courses provided free by the state, suggesting a growing need for English instruction. These courses are the primary means by which the state helps integrate immigrants, and free English courses for adults are provided as part of the state’s adult education mission. They have been offered by various agencies since the 1850s on the belief that an English-proficient immigrant population is valuable from a social, cultural, and economic perspective, both for immigrants and for the state. Although providing classes is a statewide mission, it is one carried out at the local level by adult schools, community colleges, libraries, and community organizations, among others. All of these agencies administer their programs independently and receive funding in drastically different ways. Adult schools, administered by local school districts, are the largest providers of free ESL classes in the state, teaching more than 75 percent of the state’s ESL students, and they receive their funding directly from the state general fund. Adult schools face two important financial constraints: revenue limits and a 2.5 percent limit on the growth of funded enrollment. The 2.5 percent growth limit has been in place since the 1970s, although the linguistic landscape has changed dramatically since then. For instance, between 1980 and 2000, the number of adult immigrants who could benefit from ESL courses increased by nearly 6 percent per year. Local school districts strive to serve the mission set forth by the state, but when the demand for English courses exceeds their funding level, their adult schools must either turn away students or absorb the cost of enrolling the excess. This report examines whether the funding formula for adult schools can meet the changing language needs of the state’s immigrant population. In particular, the report explores the inconsistency between the challenges faced by adult schools wanting to teach a growing number of English learners and a funding formula that may limit the fulfillment of this mission. How do adult schools cope with these competing policies? iii This report focuses on the three components of this potential conflict: the educational commitment of local providers, particularly adult schools; the English-language needs of immigrants; and the funding formula’s effect on adult school enrollments. The report also considers the consequences resulting from this misalignment of policies, the efforts to reform the funding formula, and the tradeoffs associated with increasing resources to the local providers of ESL courses. Main Findings ESL courses are primarily provided by adult schools and community colleges. This dual-provider system makes one agency—the school district or the community college district—the main provider of adult education in a local area, although libraries and community organizations also play an important, although lesser, role. However, adult schools teach 75 percent of all ESL students statewide, and ESL students make up over 40 percent of all adult school students. Demographic trends suggest a growing need for English instruction but one that varies throughout the state. The ESL target population—adults ages 18 and older not proficient in English and not enrolled in school— increased from 900,000 in 1980 to 2.7 million in 2000, an annual growth rate of 5.7 percent per year. The ESL target population grew faster outside Los Angeles County, nearly 7 percent per year. The Los Angeles perimeter counties (Orange, Riverside, San Bernardino, and Ventura) had the highest growth of any region (nearly 8%) over this 20-year period. Predicted levels of enrollment reveal that immigrants outside Los Angeles County have the highest likelihood of enrollment in ESL. The predicted statewide annual growth in enrollment is 5 percent, excluding Los Angeles County, and 3.5 percent including it. The 2.5 percent limit on the growth of funding hinders the majority of adult school programs and, therefore, the number of ESL classes offered. Close to 60 percent of adult schools in California exceed their funding limit. In 2004–05, the dollar value of enrollment in excess of state funding totaled $15.7 million. And among adult schools that do overenroll students, nearly 80 percent exceed their funding limit by over 2.5 percent. iv Policy Implications The fundamental flaw in the funding formula for adult schools is its inability to keep pace with and adapt to the changing English-language needs of communities throughout the state. A recent reform of the adult school funding formula (Assembly Bill 23), which redistributes unused adult education funds to high-demand districts, is a first step toward remedying the variation in need for ESL in the state. Yet, the 2.5 percent limit on the growth of adult school funding reduces the net expenditure per student. This, in turn, reduces the quality of adult education programs and the ability of adult schools to reach a greater number of students. Relying on adult schools to be the primary ESL provider in a region may not be as effective as increasing the resources of all providers to meet the varied and growing needs of communities.  Contents Summary.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii Figures.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix Tables.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi Acknowledgments.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii I. INTRODUCTION.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Research Goals and Questions .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Outline of the Report.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 4 4 2. POLICY CONTEXT FOR ADULT EDUCATION IN CALIFORNIA.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Public and Private Providers in the State.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Dual-Provider System.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Funding Issues Facing Adult Schools.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 3. The English-Language Proficiency of California’s Immigrants.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 The ESL Target Population in California. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Predicted Enrollment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 4. Enrollment in ESL Programs.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 ESL Enrollment: Regional Variation.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Trends in Enrollment in Public Providers.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 ESL Enrollment and ADA Units.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Where Is Enrollment Constrained?.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Teaching for Free?.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Summary.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 5. Policy Implications and Conclusion.. . . . . . . . . . . . . . . . . . . 45 Policy Consequences of Overenrolled Courses. . . . . . . . . . . . . . . . . . . . . . . . . . 45 Partial Reform of the Adult School Funding Formula. . . . . . . . . . . . . . . . . 47 How Much Does Partial Reform of the Funding Formula Help?.. . . . 49 Policy Implications.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 Summary.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 vii Appendix A. Notes on Data and Methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 B. Estimates and Results.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 References.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 About the Author. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 Related PPIC Publications.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 viii Figures 2.1. Private Institutions Offering English Courses. . . . . . . . . . . . . . . . . . . 8 2.2. Start Dates for Adult Education Programs in California.. . . . . . 11 3.1. Size of ESL Target Population, by Region, 2000.. . . . . . . . . . . . . . . 17 3.2. English Ability of Adult Immigrants, by Country of Birth, 2000.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 3.3. Predicted Probability of Enrollment in ESL Courses, 1980, 1990, and 2000.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 4.1. Regional Enrollment in ESL Programs, by Type of Provider, 2002–03.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 4.2. Total Enrollment in ESL Courses in Adult Schools and Community Colleges, 1999–00 to 2002–03. . . . . . . . . . . . . . . . . . . . 32 4.3. ESL Enrollment in Adult School Programs, 1999–00 to 2002–03.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 4.4. Regional Enrollment in Noncredit Community College ESL Courses, 1997–98 to 2003–04.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 4.5. Percentage of All Districts Exceeding Their ADA Enrollment Cap, 1998–99 to 2004–05.. . . . . . . . . . . . . . . . . . . . . . . . . . 38 4.6. Percentage by Which Funded ADA Is Exceeded, 2004–05. . . 39 4.7. Percentage Change in Dollar Value of Funded ADA Needed to Reach Parity with Reported ADA, 1998–99 to 2004–05.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 4.8. Difference Between Dollar Value of Funded ADA and Reported ADA, 1998–99 to 2004–05.. . . . . . . . . . . . . . . . . . . . . . . . . . . 42 5.1. Redistributed ADAs Under AB 23, by Region, 2005–06.. . . . . 49 5.2. Increase in ESL Enrollment as a Result of Redistribution of ADAs for Different Instructional Goals.. . . . . . . . . . . . . . . . . . . . . . 50 ix Tables 3.1. 3.2. 4.1. B.1. B.2. B.3. B.4. B.5. B.6. Size and Growth of the ESL Target Population, 1980–2000. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Size and Growth of Predicted ESL Enrollment, by Region, 1980, 1990, and 2000.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 ESL ADAs and ESL Enrollment, by Goal.. . . . . . . . . . . . . . . . . . . . . . . 37 Logit Regression Results of the Probability of ESL Enrollment, NHES.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 Mean Characteristics of Immigrant Samples in Census and NHES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 Mean Characteristics of Enrolled ESL Students.. . . . . . . . . . . . . . . . 64 Predicted Enrollment Probabilities and Levels of Enrollment for Adult School or Community College ESL Enrollees. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 Effect on Noncredit ESL Enrollment of Districts Exceeding Their ADA Cap.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 Data for Figures 4.7 and 4.8. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 xi Acknowledgments This report would not have been possible without the help, input, and contributions of many people. At the top of the list is Rebecca Steinbach, who provided superb assistance in every stage of this report, including data management and analysis of multiple datasets. I would like to also thank José Ortega, Halena Le, and Randi Knutson at the California Department of Education, and Willard Hom, Tonia Lu, Elias Regalado, Ed Monroe, and Chris Yatooma in the California Community College Chancellor’s Office. The report benefited from conversations with many experts throughout the state, including Ed Morris in the Los Angeles Unified School District; Bob Harper at Pajaro Valley Unified School District; Paul Warren and Rachel Ehlers in the Legislative Analyst’s Office; Luís Esparza, former consultant to Select Committee on Adult Education; Bruce Hamlet in Assemblymember Carol Liu’s Office; Warren Quann in the office of Assemblymember Mervyn Dymally; Robert Perez in the office of Senator Martha Escutia; and Ireri Valenzuela-Vergara at California Tomorrow. This report benefited from detailed reviews and comments by Christopher Jepsen, Marie Mora, Patricia de Cos, and Richard Greene. Finally, I would like to thank Deborah Reed at PPIC for reading through the multiple versions of this report and giving invaluable suggestions and direction. Research publications reflect the views of the authors and do not necessarily reflect the views of the staff, officers, or Board of Directors of the Public Policy Institute of California. xiii 1. Introduction California continues to be the destination for many immigrants from different countries and backgrounds, and the fast growth in the number of immigrants has focused attention on the pace of their integration. The ability to speak English is the most fundamental prerequisite for this process to take place. However, lack of English skills is a hurdle facing a growing number of immigrants because many arrive with little knowledge of English, and learning English is an often difficult and time-consuming endeavor. To help immigrants integrate into society, and for other social and economic reasons, California provides free English as a Second Language (ESL) courses to those who cannot speak English (California Department of Education, 2004; West, 1995). In the words of the Joint Board Task Force on Noncredit and Adult Education (1998, p. 5), there are multiple motives for this policy: “The mission of ESL programs for adults in California is to equip students with the language and cultural proficiencies required for the eventual fulfillment of personal, vocational, academic, and citizenship goals so that they may participate fully in American society.” Indeed, immigrants benefit in many ways from improving their English communication skills. For adults, these benefits include greater participation in civic activities and in their children’s education and increased likelihood of naturalization (Capps et al., 2003; Johnson et al., 1999; Little Hoover Commission, 2002; Martinez and Wang, 2005; Ramakrishnan and Baldassare, 2004). English-proficient immigrants also do better in the labor market, with better job prospects, higher wages, and occupational mobility (see, for example, Berman, Lang, and Siniver, 2003; Bloom and Grenier, 1996; Carliner, 2000; Chiswick and Miller, 1995; Dustmann and Van Soest, 2002; Gonzalez, 2000). For instance, a common finding is that immigrants who can speak English proficiently earn at least 10 percent more than those who cannot. For these and other reasons, the majority of Hispanic immigrants—by far the largest immigrant group in the state—believe that they need to speak English to be part of American society and are eager participants in ESL courses (Pew Hispanic Center, 2006).  California has a long-standing commitment to providing free English instruction. Free ESL courses were first provided by individual school districts in the mid-1850s, but they are now provided as part of the state’s adult education mission (de Cos, 2004; West, 1995). Although adult education is delivered by various agencies, including community colleges and public libraries, adult schools are the largest providers in the state. At the same time, adult schools must deal with funding policies that hamper their ability to meet their mission of teaching English learners. Since adult schools account for more than 75 percent of all public adult education students and funding, the issues affecting adult schools disproportionately affect the state (de Cos, 2004; Joint Board Task Force on Noncredit and Adult Education, 1998). For these reasons, this report focuses on adult schools, although the policy implications are generalized to other public providers. Adult schools receive direct funding from the state’s general apportionment fund based on the levels of funding the local school district received in 1977–78 and this amount can grow a maximum of 2.5 percent from the previous year’s funded level. However, this policy may be out of step with the changes in the immigrant population experienced by California since the 1970s. California now has more immigrants than any state in the country; many of them cannot speak English effectively. Among recently arrived adult immigrants (ages 18 and older), 52 percent report not being able to speak English proficiently. This percentage is higher among those  It is important to emphasize that this report focuses on adult immigrants not enrolled in school, since they are the target population of the adult education system, rather than child immigrants or adult immigrants on student visas (Young, 1995). Limited-English-proficient (LEP) adults enrolled in formal degree programs potentially include foreign-born persons on student visas and hence are in the state only temporarily. These students are not likely affected by adult education policies. The policy implications will differ depending on the population of LEPs considered. School-age LEPs learn English in school, and issues such as failure to enroll, type of English program, and length of time spent in the program have received much attention (Hill, 2004; Jepsen and de Alth, 2005). Another line of research considers natives (usually the second generation) and their English-language ability (McManus, 1985; Portes and Rumbaut, 1996). This report uses the terms “not proficient” and “limited English proficient” interchangeably.  In 1993, adult education legislation permitted the establishment of new adult education programs and also established new revenue limits to eliminate differences in apportionment between districts.  not in school (56%) or who are from Mexico (74%). In 2000, the ESL target population—foreign-born adults classified as LEP—numbered 2.7 million, or 35 percent of California’s seven million adult immigrants. In contrast, this population was less than 900,000 in 1980 and has increased annually by nearly 6 percent since then. Although the large ESL target population generates much heated debate, especially when it concerns the pace of their assimilation, what is overlooked is that the restriction on adult school funding challenges the ability of adult schools to accelerate this process. This is especially true in districts that did not have a large program in 1979 when the base levels for growth were set but have experienced significant growth since then in the target ESL population. Faced with enrollment pressures, educators have to choose between abiding by the cap on enrollment or—if they loosely interpret the state education code mandating the establishment of ESL programs—enrolling more students than the funding level permits. Educators in districts that are experiencing growing demand for ESL courses are more likely to be at odds with adult school funding policies (Joint Board Task Force on Noncredit and Adult Education, 1998). Thus, it is possible that these policies are inconsistent with the continued growth in the number of LEP immigrants in many regions and the mission to  English-language information comes from the Census questions, “Does this person speak a language other than English at home?” and for respondents answering yes, “How well does this person speak English?” Proficient is defined as speaking English only, very well, or well. Not proficient is defined as speaking English not well or not at all. Language questions are asked of persons ages five and older.  According to the U.S. Department of Labor, “An individual with LEP is one who has limited ability in speaking, reading, writing or understanding the English language and (a) whose native language is a language other than English or (b) who lives in a family or community environment where a language other than English is the dominant language” (http://www.doleta.gov/usworkforce/lep/glossary/English_Only.cfm).  The 10 largest countries/regions of origin for immigrants ages 18 and older in the 2000 Census are Mexico (3,384,574), Central America (660,088), the Philippines (614,829), China (535,579), Vietnam (389,384), Korea (247,074), India (239,408), South America (188,714), Iran (149,138), and the former Soviet Union/Russia (148,052). English is one of the official languages in the Philippines and is the most important language in India (http://www.cia.gov/cia/publications/factbook).  Specifically, sections 52540 and 52542 of the education code stipulate that adult schools establish ESL courses if more than 20 persons solicit an ESL course or if the school board decides to establish an ESL program (California Department of Education, 2005).  teach English to those who seek to enroll in ESL classes (de Cos, 2004; Moore, Schulock, and Lang, 2004). Without a change in the funding formula, the conflict between funding and mission may result in a long-term decline in the quality of instruction and diminish the pace of immigrant integration. Research Goals and Questions This report examines the current funding formula for adult schools, California’s mission to English learners, and the demographic changes that challenge the ability of adult schools simultaneously to fulfill their mission while complying with their funding constraints. The report also considers efforts to reform the funding formula. In addition, the report considers the implications of not fully funding public agencies such as community colleges and libraries that teach English to those who seek to learn it. This report addresses the following questions: . What is the policy background for the provision of ESL courses in California? What distinguishes adult schools from other providers? 2. How has the ESL target population changed throughout the state since 1980? What is the level of predicted enrollment and what demographic changes affect enrollment in ESL courses? 3. What are the trends in the provision of ESL courses by adult schools and community colleges statewide and in the different regions of the state? Does the adult school funding formula limit adult school enrollment? To what extent do adult school districts exceed their level of funding? 4. What do adult schools that exceed their funding limit forgo in terms of quality of adult education classes and future growth of adult education programs? How much does the redistribution of unused funding alleviate the challenges facing high-demand regions? How would adult education providers benefit from increased funding? Outline of the Report Chapter 2 considers the policy context for the current provision of ESL courses in the state, including the various providers in the state and the funding issues facing adult schools. Chapter 3 analyzes the ESL  target population from 1980 to 2000 at the regional and state levels, and examines how predicted enrollment responds to changes in the size and characteristics of the immigrant population. Chapter 4 presents actual enrollment levels, the degree to which adult schools overenroll students, and the fiscal effect of the 2.5 percent limit on funding growth. Finally, Chapter 5 considers the effect of AB 23, passed in October 2005 to reform the adult school funding formula, and the importance of increasing public funding to all providers to remedy the conflict between the funding and the mission of adult education policies.  2. Policy Context for Adult Education in California California’s adult education system is not only the largest in the nation, it is also very complex. The system consists of adult schools, community colleges, public libraries, community- and faith-based organizations, the California Conservation Corps, and prisons. Each has different governance and funding mechanisms (de Cos, 2004). This multiple-provider system is the result of a combination of policies that support the type of initial provider in an area. Because of past educational policies, the adult school system is by far the largest provider in the state, but the community college system also teaches a significant share of students in adult education. Therefore, the adult education system in California is, for all intents and purposes, a dual-provider system. For adult schools, the most important policy is the funding formula that limits year-to-year growth of funded enrollment to 2.5 percent. This chapter discusses the public and, briefly, the private, adult education system in the state, expands on the roles played by community colleges and adult schools in delivering ESL courses, and focuses on the funding mechanism for adult schools. Public and Private Providers in the State ESL courses can be provided by private and public institutions, and there are arguments for the role played by each type of provider. As discussed in Chapter 1, a common finding is that increased English ability is rewarded in the labor market. As this gain accrues to the individual, he or she is willing to pay to acquire more English skills as long as there is an economic incentive to do so. This willingness to pay generates a demand for ESL courses that private institutions can provide for a fee no greater than the value of the skill. In fact, private institutions that teach English exist throughout the state, as is seen in Figure 2.1. In total, 105 state-licensed institutions offer  /POEFHSFFQSPWJEFST 0UIFSQSPWJEFST   /VNCFSPGJOTUJUVUJPOT      $FOUSBM 7BMMFZ #BZ -PT"OHFMFT -PT"OHFMFT 4BO%JFHP 3FTUPG "SFB $PVOUZ QFSJNFUFS $PVOUZ UIFTUBUF 4063$&#VSFBVGPS1SJWBUF1PTUTFDPOEBSZBOE7PDBUJPOBM&EVDBUJPO %FQBSUNFOUPG $POTVNFS"GGBJST /05&45IFTFJOTUJUVUJPOTPGGFSOPOEFHSFFPSEFHSFFPUIFS&4-BOETJNJMBSMZ EFTDSJCFEDPVSTFT5IFTBNFJOTUJUVUJPOJTJODMVEFENPSFUIBOPODFJGJUQSPWJEFT JOTUSVDUJPOJOEJGGFSFOUDJUJFT3FHJPOEFGJOJUJPOT#BZ"SFB‡"MBNFEB $POUSB$PTUB  4BO'SBODJTDP 4BOUB$MBSB BOE4BO.BUFP$PVOUJFT$FOUSBM7BMMFZ‡'SFTOP ,FSO  .FSDFE BOE5VMBSF$PVOUJFT-PT"OHFMFTQFSJNFUFS‡0SBOHF 3JWFSTJEF 4BO #FSOBSEJOP BOE7FOUVSB$PVOUJFTSFTUPGUIFTUBUF‡BMMPUIFSDPVOUJFT Figure 2.1—Private Ins'tJiHtuVStFions Offering English Courses nondegree ESL courses and 66 institutions offer degrees or other ESL courses. The bulk of the programs are in the Los Angeles region with the Bay Area also having a high number of private institutions offering ESL courses for a fee. The Central Valley has very few, if any, such institutions.  The regions used in this report do not conform to most definitions of the state’s regions; all data in this report are aggregated according to the definitions used by the Comprehensive Adult Student Assessment System (CASAS) (except as noted) for the sake of consistency with the enrollment data provided by CASAS. See Figure 2.1 for these definitions.  Data are derived from the Directory of Degree, Non-Degree and Registered Institutions and Programs gathered by the Bureau for Private Postsecondary and Vocational Education  Unfortunately, enrollment information is not available for such institutions, so it is not possible to compare enrollment levels in public and private institutions in California. However, estimates from the 2001 National Household Education Survey (NHES) indicate that private institutions in California represent less than 1 percent of all persons who took an ESL course in the previous 12 months. Of those in public institutions, 83 percent of recent ESL students enrolled in adult schools and 14 percent in community colleges. Another 3 percent enrolled in public libraries and other public providers, whereas less than 1 percent enrolled in community- and religious-based organizations. Since there is a private gain to learning English, why does the state provide this skill free? One reason is because this is how it has always been. The tradition of not charging for adult education originated with the first adult school in the state, founded in 1856 in San Francisco (West, 1995). Also, immigrants may not have the resources to pay or borrow for school. Even more important, not all English learners are interested in using that skill for work. Parents out of the labor force, for instance, might want to learn English to help in their children’s education. Immigrants might also want to learn English to become citizens. Increasing the English ability of immigrants is also in the public interest. Since it is not possible to charge only those who would benefit in the labor market, and to promote English acquisition for the overall public good, the state has made a commitment to provide ESL courses and other forms of adult education free. This report makes no judgments about the pros and cons of providing free ESL courses but recognizes that private institutions are part of the larger system of English-language instruction in the state. Dual-Provider System California delivers adult education via this dual-provider system by design, although adult schools and community colleges overlap (BPPVE) at the Department of Consumer Affairs (https://app.dca.ca.gov/bppve/schoolsearch/default.htm). A keyword search was conducted, using terms such as “English as a Second Language” to identify providers.  Private institutions account for 10 percent of ESL students in states outside California.  significantly in their goals and types of programs offered. In addition, adult education has traditionally been a local education function and, depending on the origins of the school, was the responsibility of either the local junior college or the school district. Adult schools offer 10 types of adult education programs and the community college system offers nine, as part of its noncredit “basic skills” mission (California Department of Education, 2005). Adult schools are the primary providers (80%) of adult education and community colleges teach more than 10 percent (California Department of Education, 2006, p. A-1). Several important policies shaped the current adult education system. The first is the creation of adult schools (sometimes called evening or night schools) by local education agencies. The creation and funding of adult education programs rested on the importance placed by local residents on adult education. Figure 2.2 shows the chronology of the founding of adult education programs in the state. Most were founded between 1910 and 1949, with the bulk in the 1930s, partly to provide jobs for teachers (West, 1995). Following the precedent set by St. Mary’s Cathedral in San Francisco, and to stress the importance of “Americanization,” these schools provided courses free (West, 1995). The trends from the 1970s on reflect the effect of an important policy change in education created by Proposition 13—a change that has meant that only a few new programs were funded between then and 1992. In 1993, new legislation provided partial funding for new adult schools, and 43 were founded shortly thereafter. Because local communities had taken the primary responsibility of establishing and funding adult education programs, it was also up to them to integrate their programs into the school district or into the community college system. In 1960, community colleges were integrated into the state’s postsecondary school system as part of the state’s Master Plan for Higher Education and, subsequently, certain adult education programs became part of the noncredit program of the higher education system. With the absorption of community colleges into the postsecondary school system, school districts and community colleges established agreements regarding which would be the main providers in an area. The dual- 10    /VNCFSPGQSPHSBNT                m m m m m m m m m m m            4063$&IUUQXXXDBBEVMUFEIJTUPSZPSH /05&45IFTFEBUBFYDMVEFQSPWJEFSTXJUIVOLOPXOTUBSUEBUFT5IFEBUBBMTPEP OPUJODMVEFBEVMUTDIPPMEJTUSJDUT TJYPGXIJDIBSFOPXDMPTFE UIBUBSFGPVOEJOUIF $BMJGPSOJB"EVMU&EVDBUJPO1SPWJEFS%JSFDUPSZ Figure 2.2—Start Dates for A'JdHuVSlFt Education Programs in California provider system was firmly established at this time (Moore, Schulock, and Lang, 2004; West, 1995). Funding Issues Facing Adult Schools The decision by local communities to place their adult education program in either the K–12 or the community college system had important consequences after the passage of Proposition 13 in 1978, when the state assumed funding responsibility for all K–12 education, including adult schools that were within the local district. Community college  Formal delineation of function agreements were established in 1972 (Senate Bill 94), which gave preference to adult schools as the primary providers of adult education (unless they agreed to give up this right to community colleges). 11 districts continued to maintain local authority over their adult education programs. The fundamentals of the current funding formula for adult schools were set forth in legislation passed in 1979 and had three important characteristics. First, funding for adult schools was a separate funding category from the state general apportionment and could be spent only on adult education. (Until the passage of AB 23 in 2005, unspent adult education funding could not be redistributed to other school districts but instead was returned to the state for redistribution to other programs.) Second, the level of funding was calculated to be the product of the number of average daily attendance (ADA) units and the baseline revenue limit per ADA. (One ADA unit is equal to 525 hours of instruction.) Third, growth of funded ADA was limited to 2.5 percent per year. Moreover, the baseline point for calculating this growth was based on the enrollment levels of 1977–78 (de Cos, 2004; Moore, Schulock, and Lang, 2004; West, 1995). The exact origins of the 2.5 percent growth limit are not clear, but it is likely that this value may not be arbitrary (Moore, Schulock, and Lang, 2004) in the sense that a 5 percent cap on ADA had already been established in 1975 because of significant increases in enrollment, especially in ESL courses (West, 1995). In addition to state funding, adult schools and community colleges that teach ESL can apply for federal funding from the Workforce Investment Act (WIA) Title II program. In 2005–06, adult schools received $621 million from the state and over $53 million from WIA Title II. Unlike adult school funding, funding for adult education programs in community colleges is determined locally. Community college districts receive funding from the state’s general fund, determined by a programbased funding formula, which in turn depends on full-time-equivalent-  Cost-of-living adjustments (COLAs) were also included in the funding growth, up to a maximum of 6 percent, but are not guaranteed.  Each unit of an ADA was funded at $2,292 for adult schools. (These data are from recertified first principal apportionment (P-1) fiscal year 2005–06, available at http://165.74.253.241/ias/Exhibits/pasummary2005p1recert.xls.) In 2005–06, 185 adult school agencies and County Offices of Education received WIA Title II funds (http:// www.cde.ca.gov/fg/fo/r17/documents/ae05awards.xls). 12 student (FTES) hours. Each community college district allocates adult education funding depending on community interest, need, and the level of priority assigned to noncredit programs by administrators. In 2004–05, the noncredit rate per FTES was $1,612. The noncredit baseline revenue funding (noncredit FTES*$1,612) plus COLA (2.41%) totaled $172 million plus around $10 million from WIA Title II funds. The unique funding conditions facing adult schools pose specific challenges to them as they strive to meet the commitment to adults and to English learners in particular. Regardless of differences in demographic need, all adult schools face the same constraint on funding growth. Yet school districts are not limited in the level of enrollment; they are free to exceed enrollment growth beyond 2.5 percent. Those that exceed their enrollment growth, however, absorb the full cost of overenrollment because the net level of per-ADA compensation is lower than the state-approved level. Such a policy can reduce the quality of education and development of new programs. On the other hand, adult schools that turn away students because they are already at their cap underserve a motivated group of immigrants who would otherwise learn English. The level of funded ADA is not based on current demographic conditions but on 1977–78 funded enrollment levels. Since that year, of course, local districts have undergone substantial changes in their immigrant population. Adult schools that served few immigrants in the late 1970s but have experienced significant growth since are hampered today by the initially low levels of funded ADA. Conversely, districts that had an adult education program in the 1970s continue to receive the base amount even if they have continually experienced enrollment declines.  The program-based funding formula for each district is based on FTES, maintenance and operations costs, and other factors. Districts with noncredit programs have their allocations for maintenance and operations and instructional support computed to explicitly account for noncredit FTES. The final rate for noncredit FTES deducts the noncredit allocation for maintenance and operations and institutional support (California Community Colleges Chancellor’s Office, 1999, p. 14).  This estimate includes only the amount attributable to noncredit enrollment. I would like to thank Ed Monroe and Elias Regalado who provided the financial data (http://www.cccco.edu/divisions/cffp/fiscal/allocations/links/apportionment/04_05_ simulated_recal/2004-05PostedSimulatedRecal.pdf). WIA Title II information comes from the 2005–06 Grants Awards (http://www.cde.ca.gov/fg/fo/r17/ documents/ae05awards.xls). 13 In other words, the distribution of funded ADAs in any given year is not linked to any current measure of demand. This shortcoming in the funding formula curbs the ability of school districts to adjust their programs (beyond the 2.5% limit) to better meet their community’s needs. At the state level, ADAs are inappropriately distributed. The shortcomings of these two aspects of the funding formula are considered in the chapters below, which examine the extent to which adult school enrollment is restricted by this limit and the growth rate in the population that would benefit from ESL courses. 14 3. The English-Language Proficiency of California’s Immigrants The provision of public ESL courses is conditioned on the need for them. However, the state’s adult schools receive increases in annual funding only up to a maximum of 2.5 percent even if their enrollment growth is 2.5 percent or greater. Since enrollment is potentially constrained, there is no direct measure of need to ascertain whether the mission of teaching English learners is being met. Reported enrollment is not a good measure of the demand for adult education because some local districts cut off enrollment when the 2.5 percent limit is reached, whereas others enroll students beyond this limit. This chapter presents two alternative measures of need based on Census and survey data about ESL course-taking behavior. The ESL Target Population in California Because adult education programs serve adults not enrolled in an academic program, the target population for ESL programs consists of LEP adults not currently enrolled in school. This definition is used for multiple reasons. First, full- or part-time students are likely to enroll in credit courses or, if they are younger than age 18, in the K–12 system. Second, adults served by ESL programs are likely unable to succeed in most academic or certificate programs because of language barriers, although they may later enroll in such programs. Third, adult immigrants who are not English-proficient are the most likely of any group of immigrants to benefit from adult schools. This set of immigrants then  Hill (2004) notes that immigrants arriving in their late teens do not enroll in the K– 12 system. An LEP immigrant younger than age 18 cannot enroll in an adult school unless he or she is concurrently enrolled in high school and then only under certain circumstances (but he or she is not defined as an adult for funding purposes) (California Department of Education, 2004, p. 1). 15 represents a reasonable approximation of the population that is relevant for policymakers and the adult education system. However, for several reasons the ESL target population is only an approximation of the actual number of people wishing to take ESL courses. First, it overestimates the potential number of ESL students because not all the ESL target population wishes to enroll in public ESL classes. Some have no desire to learn English at the moment; others may do so on their own or through private institutions. Second, it is likely that the ESL target population differs from that actually enrolled in ESL classes, since the latter consists of a self-selected group of individuals who have decided to seek out ESL courses at a particular point in time (Lazear, 1999). Third, this population represents only the stock of need for ESL rather than the flow; the former provides information about the need for ESL courses at one point in time (at the time of the Census), and the latter describes the change in need for ESL courses from one year to the next. Nevertheless, such information from the Census is a popular measure because there are few alternatives. Regional information for the ESL target population in 2000 is shown in Figure 3.1. For consistency with data from CASAS presented throughout this report, counties are grouped according to the definitions used by CASAS. Statewide, there were nearly 2.7 million immigrants in the ESL target population in 2000. Much of the state’s ESL target population in 2000 was concentrated in several parts of the state. Los Angeles County had by far the largest ESL target population, numbering over 1.1 million immigrants or 43 percent of the state total. Around 510,000 persons in the ESL target population lived in the Los Angeles perimeter—the counties in Southern California outside Los Angeles County. The Bay Area (380,000) had the third-largest ESL target population of any region (14%), followed by the rest of the state with  The participation rate estimated by Young (1995) using Public Use Microdata Sample (PUMS) data is also not ideal, since that source considers only first-time students and, more important, does not account for the inflow and outflow of immigrants into the target population. It also assumes that all LEP students are the same in their need for ESL instruction, which may not be the case.  Although we made repeated attempts to gain access to district-level enrollment data, these data were not made available to us for analysis. Hence, despite its shortcomings, we have used the regions defined by CASAS in this report. 16 Del Norte Siskiyou Modoc Trinity Humboldt Shasta Lassen Tehama Plumas Yuba Mendocino Glenn Butte Sierra Colusa Sutter Lake Placer Nevada Sonoma Napa Yolo El Dorado Amador Alpine Sacramento Marin Solano Calaveras San Francisco Contra San Costa Joaquin Tuolumne Mono San Mateo Alameda Santa Stanislaus Mariposa Clara Merced Santa Cruz Madera San Benito Monterey Fresno Kings Tulare 4BO%JFHP$PVOUZ   $FOUSBM7BMMFZ   3FTUPGUIFTUBUF   #BZ"SFB   -PT"OHFMFTQFSJNFUFS   -PT"OHFMFT$PVOUZ    Inyo San Luis Obispo Kern Santa Barbara Ventura Los Angeles San Bernardino Orange Riverside San Diego Imperial 4063$&1VCMJD6TF.JDSPEBUB4BNQMF 16.4  /05&45IF&4-UBSHFUQPQVMBUJPODPOTJTUTPGGPSFJHOCPSOPVUPGTDIPPMBEVMUT BHFT BOEPMEFS XIPTQFBL&OHMJTIiOPUBUBMMuPSiOPUXFMMu4FF'JHVSFGPSSFHJPOEFGJOJUJPOT Figure 3.1—Size of ESL Ta'rJgHeVtSFPopulation, by Region, 2000 320,000 (12%). The Central Valley had 6 percent of the state’s ESL target population in 2000, or about 170,000 persons. Since many of the state’s immigrant communities were in Los Angeles County, the Los Angeles perimeter, and the Bay Area, this distribution of the target population is 17 consistent with other analyses of the English ability and settlement patterns of immigrants (Bauer, Epstein, and Gang, 2005; Chiswick and Miller, 2002). To better compare the demographic changes in the state with the growth constraints on adult school enrollment, Table 3.1 presents the growth in the ESL target population from 1980 to 2000. Los Angeles County stands out from other regions but in a manner contrary to expectations given its large immigrant population: It had the smallest peryear growth of the ESL target population of any region during this 20year period—4.4 percent. This rate is nearly 2 percentage points lower than that of the Bay Area, the next slowest-growing region (6.1%). The region with the fastest growth over this 20-year period was the Los Angeles perimeter (7.7%), followed by the rest of the state (7.1%). Since Los Angeles County contained 43 percent of the state’s ESL target population in 2000, the statewide percentage was less than 6 percent from 1980 to 2000. Without Los Angeles County, the statewide annual growth rate was 6.9 percent from 1980 to 2000. The annualized growth rates based on two 10-year periods (1980–1990, and 1990–2000) are also shown in Table 3.1 and reveal a decrease in the growth rate of the ESL target population. In the 1980s, the ESL target population grew by nearly 10 percent per year in the counties constituting the Los Angeles perimeter, followed by less than 8 percent in the Central Valley and San Diego County regions. The other regions grew between 5.7 percent (Los Angeles County) and 7.1 percent (rest of the state) in the 1980s. However, in the 1990s, a redistribution of the state’s ESL target population took place. First, the growth rates declined in all regions except the rest of the state in the 1990s. Second, Los Angeles County’s share of the growth of target ESL population declined from 44 to 31 percent in the 1980s. During this time, the Bay Area and the rest of the state increased their shares of the ESL target population; these two regions accounted for  The constant annual growth where LEPxand LEPy are the LEP rpaotepuislactaiolcnulianteydeabryx[a(LnEdPyxe /aLr Ey,Prye )s1p/(ex–cyt)iv–e1ly].,  In 1980, 54 percent of the ESL target population resided in Los Angeles County. By 2000, this figure had declined to 43 percent. In contrast, the share of the ESL target population grew in the Los Angeles perimeter and rest of the state regions, from 13 and 9 18 Table 3.1 Size and Growth of the ESL Target Population, 1980–2000 19 ESL Target Population   1980 1990 2000 Central Valley 47,649 100,452 166,750 Bay Area 116,282 214,323 382,744 Los Angeles County 479,225 835,554 1,131,623 Los Angeles perimeter 114,291 283,480 508,297 San Diego County 40,620 86,068 142,445 Rest of the state 81,900 163,191 320,868 Total 879,967 1,683,068 2,652,727 Total minus Los Angeles County 400,742 847,514 1,521,104 SOURCE: 2000, 1990, and 1980 5% PUMS. NOTE: See Figure 2.1 for sample and definitions. Average Annual Increase and Annualized Growth Rate 1980–1990 1990–2000 1980–2000 5,280 7.7% 6,630 5.2% 5,955 6.5% 9,804 6.3% 16,842 6.0% 13,323 6.1% 35,633 5.7% 29,607 3.1% 32,620 4.4% 16,919 9.5% 22,482 6.0% 19,700 7.7% 4,545 7.8% 5,638 5.2% 5,091 6.5% 8,129 7.1% 15,768 7.0% 11,948 7.1% 80,310 6.7% 96,966 4.7% 88,638 5.7% 44,677 7.8% 67,359 6.0% 56,018 6.9% 34 percent of the growth in the ESL target population in the 1990s—an increase from 22 percent in the 1980s. Meanwhile, the share in all other regions increased slightly from the 1980s, with the Los Angeles perimeter’s share representing one-fifth of the state’s growth in the ESL target population. Since the Los Angeles perimeter has one of the fastest and most sustained growth rates in the state, it is likely that the continued decline in the growth rate in Los Angeles County results perhaps because more immigrants choose to live in nearby counties. Yet, despite Los Angeles County’s having the lowest (and declining) growth rate of the ESL target population, it continues to have both the largest ESL target immigrant population in the state—1.1 million in 2000—and the largest share of the ESL target population of any region of the state—39 percent. There is also a great deal of variation in the foreign-born adult population by country of birth. Figure 3.2 categorizes adult immigrants by country of birth and English-language ability in 2000. Around 55 percent of Mexican immigrants (1.7 million) reported not being proficient in English—the largest group in either absolute or percentage terms. Immigrants from the rest of Latin America (37%) and Southeast Asia (40%) also had a relatively high percentage of LEP immigrants. The combined Latin American and Mexican immigrant population not only had the largest percentage of immigrants who could not speak English well, but this population (two million in 2000) was two-thirds the size of the total immigrant population from other countries (3.1 million). In aggregate, Latino LEP immigrants made up 72 percent of the LEP population in California in 2000. About 23 percent of other Asians (nearly 400,000) reported not being proficient in English, second to Mexico in the number of LEP immigrants. The remaining LEP immigrants from percent in 1980 to 19 and 12 percent, respectively in 2000. The share for all other regions remained constant.  Ed Morris of the Los Angeles Unified School District (LAUSD) Adult Education Office noted the urban-to-rural migration as a contributing factor to the decline in enrollment in ESL courses there, from a high of 43,285 ADAs in 2001–02 to 40,313 in 2004–05.  Over half of the Central Valley’s adult immigrants are LEP, followed by Los Angeles County at 41 percent. Los Angeles County’s share of the LEP population is similar to its distribution of the adult immigrant population (39%). 20 &4-UBSHFUQPQVMBUJPO &OHMJTIQSPGJDJFOU   1FSDFOUBHF-&1BOE&OHMJTIQSPGJDJFOU                             .FYJDP 3FTUPG 4PVUIFBTU -BUJO"NFSJDB "TJB "TJB       &VSPQF   "GSJDB PUIFS 4063$&16.4GPS$BMJGPSOJB /05&44FF'JHVSFGPSTBNQMF$BOBEB /FX;FBMBOE BOE"VTUSBMJBBSFJODMVEFE JOUIF&VSPQFDBUFHPSZ Figure 3.2—English Ability of Ad'uJHlVtSFImm igrants, by Country of Birth, 2000 Europe, Africa, and other countries totaled over 85,000. More than 90 percent of persons from these regions of the world reported being proficient in English. The growth rate of the ESL target population is outpacing the stateimposed growth limit on adult schools by more than a 2-to-1 ratio statewide and by nearly 3-to-1 in certain regions of the state. Despite the caveats in Table 3.1, these growth rates support the contention that the funding formula does not recognize the demographic reality. Predicted Enrollment An alternative measure of the need for ESL courses can be derived from the 1999 and 2001 NHES, which asked individuals if they had enrolled 21 in an ESL class in the previous 12 months. From these national data, it is possible to predict the factors affecting the probability of whether an individual in California enrolls in an ESL course. The decision to learn English depends on many factors, including accessibility, availability, convenience, and economic incentives (such as higher wages). The variables chosen to estimate this behavioral model capture the various costs and incentives leading to the decision on whether to enroll in ESL courses (Crandall and Sheppard, 2004). The variables are age, age at arrival in this country, highest level of education completed, Hispanic origin, country of origin, family size, marital status, whether female, and whether living in California. Once the effect of these variables is estimated using the NHES data, these values are combined with the demographic characteristics of the appropriate sample of immigrants from 1980, 1990, and 2000 Census data (5% PUMS) to obtain individual probabilities of enrollment in ESL courses. These probabilities are then averaged by region and multiplied by the total population size that the immigrant sample represents. The estimated enrollment levels reflect enrollment in all ESL providers in California at the time of the Census. This measure of the need for ESL courses has several advantages over others. First, it is based on a model of enrollment during a 12-month period and thus measures the flow, rather than the stock, of the need for ESL courses. Second, the predictions are a function of factors affecting the decision to enroll in ESL courses. Since the decision to learn English depends on multiple factors, such as age, years of schooling, country of origin, marital status, and so on, changes in these factors will alter the likelihood that immigrants enroll in ESL courses (Crandall and Sheppard, 2004; National Center for Family Literacy and National Center for ESL Literacy Education at the Center for Applied Linguistics, 2006; Young, 1995). Increases in the size of the population, holding composition the  Appendix A discusses the NHES sample restriction, consisting of persons ages 18 and older who did not enroll in an ESL college program. Because of sample size considerations, the sample includes enrollees in all ESL providers, not just adult schools or community colleges. Restricting the ESL student sample to adult school and community college enrollees does not change the sign or the magnitude of the statistically significant variables. See Appendix Table B.1. 22 same, also increase the number of ESL students. The estimates from this measure can be used to gauge regional and statewide changes in predicted enrollment resulting from changes in the composition and size of immigrant groups to assess the overall provision of ESL courses. Table 3.2 presents predicted ESL enrollment for different regions of the state. That enrollment in 2000 is around 620,000. Los Angeles County has the largest level of predicted enrollment, around 250,000. Los Angeles County’s share of the predicted level of enrollment, 40 percent, is similar to the ESL target population in Table 3.1 (43%). The Bay Area and Los Angeles perimeter have the second- and third-largest number of predicted ESL students, nearly 110,000 and 120,000, respectively. The counties in the rest of the state category have 76,000 students predicted to enroll in an ESL course, San Diego County has 37,000 students, and the Central Valley has a predicted enrollment of 32,000. Around 75 percent of predicted enrollment is distributed in the Bay Area, Los Angeles County, and Los Angeles perimeter. The annualized regional growth rates reveal differential patterns of ESL enrollment over time. The level of predicted enrollment in 2000 is 30 percent greater than in 1990 and 100 percent higher than in 1980. The statewide levels in those years based on the model are around 315,000 and 475,000, respectively. Los Angeles County stands out from all other regions of the state because the predicted number of students is flat between 1990 and 2000. Elsewhere, the annual growth rate in predicted enrollment exceeds 2.5 percent, ranging from 2.8 to 5.8 percent in the 1990s. The regions with the greatest growth in predicted enrollment in the 1990s are the Bay Area and the rest of the state with rates of 5.8 percent. The Central Valley and the Los Angeles perimeter have similar annual growths during this period, 3.7 and 3.2 percent, respectively. San Diego County’s annualized growth rate is also relatively modest, at 2.8 percent. The statewide growth is 2.7 percent but without Los Angeles, the annualized growth rate is 4.4 percent. The differences in estimates are due to two factors: the characteristics and the number of immigrants in each Census year. Figure 3.3 presents the average probability of enrollment in ESL courses in each year by region. These probabilities depend on the characteristics of immigrants in each region, and any change over time is due to differences in the composition 23 Table 3.2 Size and Growth of Predicted ESL Enrollment, by Region Predicted Enrollment Annualized Growth (%)   1980 1990 2000   1980–1990 1990–2000 1980–2000 Central Valley 12,607 22,375 32,317 5.9 3.7 4.8 Bay Area 44,810 62,267 109,138 3.3 5.8 4.6 Los Angeles County 173,781 233,046 249,170 3.0 0.7 1.8 Los Angeles perimeter 39,944 85,419 117,591 7.9 3.2 5.5 San Diego County 15,921 28,365 37,368 5.9 2.8 4.4 Rest of the state 27,406 43,594 76,291 4.8 5.8 5.3 Total 314,468 475,065 621,876 4.2 2.7 3.5 Total minus Los Angeles County 140,688 242,019 372,706   5.6 4.4 5.0 SOURCE: Author’s calculations from a model estimated using 1999/2001 NHES data and applied to 1980, 1990, and 2000 Census 5% PUMS. NOTES: These estimates are based on a logit regression of ESL enrollment from the NHES data and applied to the PUMS data. See Appendix Table B.1 for estimates. 24   1FSDFOUBHFQSPCBCJMJUZPGFOSPMMNFOU     $FOUSBM 7BMMFZ #BZ -PT"OHFMFT -PT"OHFMFT 4BO%JFHP 3FTUPG "SFB $PVOUZ QFSJNFUFS $PVOUZ UIFTUBUF 4063$&"VUIPSTDBMDVMBUJPOT /05&45IFTFFTUJNBUFTBSFCBTFEPOBMPHJUSFHSFTTJPOPG&4-FOSPMMNFOUGSPNUIF BOE/)&4BOEBQQMJFEUPUIF  BOEGPS$BMJGPSOJB16.4 Figure 3.3—Predicted Probab'ilJiHtVySFofEnrollment in ESL Courses, 1980, 1990, and 2000 of the ESL target population rather than to changes in size. In 2000, statewide, these probabilities are around 11 percent, compared to 15 percent in 1990 and 1980. From 1990 to 2000, these probabilities declined in all regions of the state, but the percentage decline was greatest in Los Angeles County and the Los Angeles perimeter (over 35%), slightly greater than the 30 percent decline in San Diego County and the Central Valley. In contrast, the mean probability of enrollment declined in the Bay Area and the rest of the state by around 18 percent. The 6 percentage point decline in enrollment probability in Los Angeles County, the Los Angeles perimeter, and the Central Valley equalized enrollment probabilities in all the regions by 2000 to 10–11 percent. In other words, these regions had higher enrollment 25 probabilities, on average, than other regions in previous decades, but not in 2000. The changes in average probability of enrollment are due to changes in the demographic characteristics of immigrants. The three most important factors leading to a decline in the probability of enrollment from 1990 to 2000 were the statewide increase in average age (40.9 to 42.4), the decrease in average age at arrival (28.8 to 25.1), and the increase in percentage Mexican (43 to 47). In the case of Los Angeles County, the relatively large decline in enrollment probability is explained by an increase of 2.7 years in the average age of immigrants, compared to a 0.6 year increase in mean age in the non–Los Angeles counties (see Appendix Table B.2). Enrollment also depends on the size of the eligible immigrant population. As Table 3.1 shows, the ESL target population increased slightly in Los Angeles County. This explains why, despite the significant decline in enrollment probability in Los Angeles County, the predicted level of enrollment increased by a small amount during the 1990s. These two factors—change in characteristics and change in target population— explain the levels of enrollment in Table 3.2. These predicted growth rates are also reflected in the statewide population growth rates reported by Johnson (2002). He reports that the Inland Empire, San Joaquin Valley, and the Sacramento Metro areas had the largest population growth rates in the 1990s. International migration accounted for at least 30 percent of the population growth in Los Angeles, Orange, San Diego, and Ventura Counties and the Bay Area and 27 percent in the Inland Empire (San Bernardino and Riverside Counties). At the same time, he notes that there was substantial out-migration from Los Angeles County during this time. These trends suggest that the same areas also experienced large growth in ESL enrollment. Summary This chapter considered two ways to assess the level of need for ESL courses in California. The broadest but perhaps least satisfactory measure is the size of the ESL target population in the state and its growth from 1980 to 2000. The second is derived from a national survey of participation in ESL courses in the previous 12 months. Both measures reveal similar patterns with regard to the need for ESL since the 1980s. First, a large (and 26 growing) population would clearly benefit from enrolling in ESL courses. Second, this population is unevenly distributed throughout the state. The counties around Los Angeles and the Bay Area in particular have the largest number of potential ESL enrollees. Third, the growth in each measure of need (except for predicted enrollment in Los Angeles County) exceeds 2.5 percent in every region, and in certain regions (such as the rest of the state and the Bay Area), the growth in need is more than double this amount. Enrollment in ESL courses depends on demographic factors that have changed since the 1970s and differ throughout the state. For instance, an increase in the average age and the percentage Mexican and a decrease in age at arrival explain the lower statewide probability of enrollment in the 1990s. These findings are compelling evidence that the level of provision of ESL courses in any given year falls short of the need. Furthermore, the variability in need for ESL courses throughout the state points to another fundamental flaw in the adult school funding formula: It does not recognize changes in immigrant demographics in different regions of the state. 27 4. Enrollment in ESL Programs Adult schools and community colleges are the most important providers of ESL courses in the state, currently enrolling more than 600,000 students per year, with adult schools teaching more than 75 percent of all ESL students. The level of enrollment is the outcome of the interplay between the demand for ESL courses and their provision. Up to the 2.5 percent growth limit, adult schools are not constrained from enrolling all potential students. Additional students, however, if allowed to enroll, are not funded by the state; some adult schools strive to align actual enrollment with funded enrollment, whereas others enroll more students than they receive funding for. The two measures of demand imply that the need for ESL courses exceeds the growth of funding for adult schools. And since adult schools are the major providers of adult education in the state, it is likely that the number of potential English learners exceeds the public provision in many parts of the state. When the need for ESL courses grows by more than 2.5 percent per year, adult schools may be caught in the middle of two competing policies: the 2.5 percent limit on funding growth and the mission to teach those who seek to learn English. This chapter examines which districts abide by the funding constraint, which disregard this limit (either as a matter of policy or by happenstance), and the extent to which adult schools are underfunded. Although adult schools are the primary providers of adult education, community colleges are an important component in the state’s adult education system. In some areas, such as San Francisco and San Diego County, they are either the sole or major provider. For this reason, this chapter presents enrollment information for both providers, although the emphasis is on how the adult school system is affected by the funding formula. ESL Enrollment: Regional Variation Enrollment information for adult schools is obtained from various reports published by CASAS for the 1999–00 to 2002–03 fiscal years. Adult school enrollment is obtained from the CASAS reports on adult schools, and community college enrollment information is obtained from data provided 29 by the California Community College Chancellor’s Office (CCCCO). All noncredit nonduplicated students with positive enrollment hours are included, although students who may have also enrolled in a credit class are not excluded. Appendix Table B.3 gives a brief demographic portrait of students enrolled in adult schools and community colleges and shows that students in both providers are similar. Figure 4.1 presents the geographic diversity in total enrollment (adult schools plus community colleges) in 2002–03, the last year for which enrollment information by region is available. Los Angeles County accounted for more than 40 percent of total ESL enrollment in the state, with 272,000 students enrolled. This enrollment was more than twice the size of the combined next two largest regions, the Bay Area and the Los Angeles perimeter, each with around 110,000 enrolled students. The rest of the state followed with 80,000 students. San Diego County and the agricultural counties of the Central Valley enrolled 50,000 and 21,000 ESL students, respectively, that year. In 2002–03, there was a great deal of variation in the percentage enrolled in adult schools across regions. At the high end, adult schools accounted for over 90 percent of ESL enrollment in Los Angeles County and the Central Valley and in the low end for 40 percent in San Diego County and 57 percent in the Los Angeles perimeter. In the Bay Area, adult schools accounted for 73 percent of all enrolled ESL students, a value similar to that for the rest of the state (81%). Regardless of these percentages, the contribution of community colleges to adult education should not to be minimized, since they are either the sole providers in an area or supplement the provision of ESL courses through agreements with local adult schools. Trends in Enrollment in Public Providers Figure 4.2 shows the total ESL student population, split between adult schools and community colleges for the years in which the available data overlap (1999–00 to 2002–03). The total number of students increased from 496,000 in 1999–00 to 652,000 in 2002–03. Yet, this increase in total  Only around 60 percent of all community colleges are surveyed by the CASAS WIA Title II data. The data for adult schools exclude agencies that did not receive funding and did not meet the minimum reporting requirements (CASAS, 2004). 30 "EVMUTDIPPMT $PNNVOJUZDPMMFHFT    /VNCFSFOSPMMFE  T         $FOUSBM 7BMMFZ   #BZ "SFB      -PT"OHFMFT -PT"OHFMFT 4BO%JFHP $PVOUZ QFSJNFUFS $PVOUZ  3FTUPG UIFTUBUF 4063$&$"4"4%ZOB3FQPSUT"EVMU4DIPPMEBUBBOE$$$$0FOSPMMNFOUEBUB /05&45IFBEVMUTDIPPMEBUBBMTPJODMVEF&4-DJUJ[FOTIJQDPVSTFT*OUIF$$$$0EBUB  POMZSFTQPOEFOUTXJUIQPTJUJWFFOSPMMNFOUIPVSTBOEOPOEVQMJDBUFEBOOVBMFOSPMMNFOU JOGPSNBUJPOBSFJODMVEFE Figure 4.1—Regional Enrollm'enJHtViSnFESL Programs, by Type of Provider, 2002–03 enrollment masks the fact that the share of students in community colleges declined from a high of 29 percent in 1999–00 to 23 percent in 2002–03. The number of students served by community colleges also was lower in real terms after 2001–02. In fiscal year 2001–02, there were substantial midyear cutbacks resulting in stagnant or negative growth in adult education for both providers. The budget problems continued in subsequent years, but the state dealt with the cutbacks by deferring payments to adult schools and community colleges from one fiscal year to the next (July 1). This accounting tactic allowed agencies to continue offering services at a level that was at least similar to that of previous years. This rolling forward of 31 "EVMUTDIPPMT $PNNVOJUZDPMMFHFT  /VNCFSFOSPMMFE  T                m m m m /05&4&4-BMTPJODMVEFT&4-DJUJ[FOTIJQDPVSTFT4FF'JHVSFGPSPUIFSOPUFT Figure 4.2—Total Enrollmen'tJHiVnSFESL Courses in Adult Schools and Community Colleges, 1999–00 to 2002–03 payments has been used since then to limit the effect of cutbacks in the current fiscal year. For this reason, the only significant decline in funding is noticeable in the 2002–03 fiscal year. In addition to the deferrals in 2002–03, there were midyear reductions in funding for the K–12 and community college systems. The effect of this reduction is seen in the lower enrollment levels in community colleges in that year in Figure 4.2. From 1999–00 through 2002–03, adult schools enrolled nearly 1.8 million students in ESL courses (see Figure 4.3), and ESL accounted for  The 2005–06 budget deferred $45.9 million of adult school funding to 2006–07 (see http://www.cde.ca.gov/fg/fr/eb/documents/budgetreport05.pdf).  Adult school funding was reduced by $74 million and total community college funding was reduced by $231 million (http://www.lao.ca.gov/analysis_2003/education/ ed_2_cc_midyr_prop98_anl03.htm). 32 /VNCFSFOSPMMFE  T   &4-DPVSTFT 0UIFSQSPHSBNT                  m m 4063$&$"4"4%ZOB3FQPSUT"EVMU4DIPPMEBUB /05&&4-BMTPJODMVEFT&4-DJUJ[FOTIJQDPVSTFT m m Figure 4.3—ESL Enrollment in A'JdHuVSlFt School Programs, 1999–00 to 2002–03 42 to 44 percent of all adult school programs. Although the share of ESL students in adult schools was rather stable over this period, the number of students in ESL programs steadily increased from about 350,000 in 1999–00 to 500,000 in 2002–03. The corresponding year-to-year growth rates in ESL enrollment were 15, 23, and 1 percent. Enrollment growth slowed considerably in the 2002–03 program year (1%), not just compared to previous years but also relative to the growth in non-ESL programs (11%) in that same year. This is likely a result of the funding problems noted above. Yet, before 2002–03, ESL and non-ESL programs grew at an identical pace, but this was not the case in 2002–03 (1% versus 11%). In any given year, total enrollment in community colleges is significantly less than enrollment in adult schools. Figure 4.4 shows that enrollment in noncredit ESL courses in community colleges peaked at 33 /VNCFSFOSPMMFE  T #BZ"SFB-PT"OHFMFT$PVOUZ-PT"OHFMFTQFSJNFUFS 4BO%JFHP$PVOUZ3FTUPGUIFTUBUF$FOUSBM7BMMFZ          m m m m m m m 4063$&"VUIPSTDBMDVMBUJPOTGSPNEBUBQSPWJEFECZ$$$$0 /05&5IFEBUBSFGFSUPVOEVQMJDBUFE XJUIJODPMMFHF TUVEFOUDPVOUTJOUIFTBNF BDBEFNJDZFBS  Figure 4.4—Regional Enrollme'nJHtViSnFNoncredit Community College ESL Courses, 1997–98 to 2003–04 160,000 in 2001–02. This pattern is similar to that of adult schools in that enrollment slowed significantly after 2001–02 (although adult school enrollment leveled off). The reason for this decline is the same as that for adult schools: decreased funding as a result of the state’s budget problems beginning in 2001–02. The regional variation in the level of enrollment in noncredit ESL courses in Figure 4.4 is a result of multiple factors, such as the level of need in a region, but perhaps more important are the historical factors determining the main provider in a region. Community colleges are the main providers of adult education in an area if they were the main providers before the 1960s, if no current adult school exists in the area, or if community colleges reach a delineation of function agreement with the local school district to provide adult education classes. In any given 34 year, enrollment in noncredit ESL courses is greatest in the Los Angeles perimeter and tends to surpass 40,000 per year. Enrollment in San Diego County and the Bay Area is normally around 30,000 per year. Los Angeles County is the state’s fourth-largest provider of noncredit ESL courses at over 20,000 students each year. The community colleges in the rest of the state enroll around 15,000 students in most years whereas those in the Central Valley enroll fewer than 1,000 students. ESL Enrollment and ADA Units This section focuses on adult school enrollment but distinguishes between actual enrollment levels in adult education programs and funded enrollment. The former is the number of students that adult schools enroll, and the latter is enrollment within the state-approved level of funding. However, the state funds enrollment based on hours of attendance rather than number of students. Specifically, adult school funding is based on the total number of ADA units taught. Because funding is measured in ADAs, this chapter presents enrollment and funding in terms of ADA units and the dollar value of these units. Because the hours of instruction differ across enrollees, the levels of enrollment and ESL target population previously presented can be translated into ADAs only after making assumptions regarding the hours of instruction needed by each ESL student. Most of the estimates regarding the appropriate amount of instruction depend on the goal of instruction. At the high end of such estimates, one report suggests that it takes between 500 and 1,000 hours of instruction to achieve functional proficiency in English (Crandall and Sheppard, 2004). Others estimate that it takes 100 hours at a minimum for a one-grade-level increase in reading comprehension tests but suggest 150 hours to significantly increase the chances that students attain this goal (National Center for the Study of Adult Learning and Literacy, 2002, p. 85). Another evaluation of adult education programs found that the median number of hours of instruction for students who successfully completed beginning ESL is 216 hours, 136 for intermediate ESL class, and 372 for a combined beginning and advanced ESL class (Young, 1995, p. 43). The average and median number of hours of attendance for ESL in California is 113 (CASAS, 2004; Young, 1995). 35 Table 4.1 presents the implied number of ADA units for the recommended hours of instruction associated with three different goals. For comparison, the total reported adult school ADAs in 2002–03 are also provided. Goals that require more hours of instruction require a greater share of a region’s total ADA. To attain the goal of providing the statewide average level of instruction, adult schools in San Diego County and the Central Valley use the least number of ESL ADAs in the state, 4,300 each, or less than one-third of their adult school’s total reported ADA. The Los Angeles perimeter and Bay Area dedicate more of their ADAs than any other region to attain any of the three goals. For instance, to complete a beginning ESL course, 86 to 89 percent of reported ADAs in those regions is required. Where Is Enrollment Constrained? The levels of funding and enrollment for ESL courses are likely mismatched in school districts that are already at the maximum of their adult education enrollment cap. Unfortunately, ADA information for ESL programs is not available separately, but since these programs account for more than 40 percent of adult education programs statewide, it is likely that a binding ADA enrollment cap affects ESL programs the most. What percentage of districts is affected by the limit on the growth of adult education ADA? Figure 4.5 examines the percentage of school districts in a region that are over their ADA enrollment cap over time. Reported ADA is the total units of instruction that the district actually provides, and funded ADA is the ADA cap that each district is allocated according to the funding formula (based on the previous year’s enrollment level, growth adjustments, and COLA). This measure makes it possible to compare differences across regions regardless of the size of their programs. The Bay Area, Los Angeles  In 1998–99, eight districts were at their cap, four in 1999–00, six in 2000–01, five in 2001–02 and 2002–03, one in 2003–04, and none in 2004–05. Most regions that are below their cap follow the same overall pattern over time. Between 1998–99 and 2000– 01, the number of school districts under their ADA enrollment cap increased and then, after 2000–01, the number fell, consistent with the state budget problems experienced after 2001. The rest of the state, Los Angeles County, and the Bay Area in particular experience the greatest numerical declines between 2001–02 and 2003–04. 36 Table 4.1 ESL ADAs and ESL Enrollment, by Goal Average Hours of Increase Reading Score Complete Beginning Attendance One Grade Level ESL Course Reported Adult % ESL % ESL % ESL   School ADAs ESL ADAs ADAs   ESL ADAs ADAs   ESL ADAs ADAs Central Valley 16,647 4,327 26.0 5,743 34.5 8,271 49.7 Bay Area 40,527 18,298 45.1 24,289 59.9 34,976 86.3 Los Angeles County 134,319 53,210 39.6 70,633 52.6 101,712 75.7 Los Angeles perimeter 29,561 13,821 46.8 18,346 62.1 26,419 89.4 San Diego County 13,074 4,306 32.9 5,716 43.7 8,231 63.0 Rest of the state 37,777 13,985 37.0 18,564 49.1 26,732 70.8 Total 271,905 107,947 39.7   143,292 52.7   206,341 75.9 NOTE: Based on 2002–03 ESL enrollment levels reported in Figure 4.1. 37 $FOUSBM7BMMFZ #BZ"SFB -PT"OHFMFT$PVOUZ -PT"OHFMFTQFSJNFUFS 4BO%JFHP$PVOUZ 3FTUPGUIFTUBUF     1FSDFOUBHF      m m m m m m m 4063$&$BMJGPSOJB%FQBSUNFOUPG&EVDBUJPO 1SJODJQBM"QQPSUJPONFOU6OJU  Figure 4.5—Percentage of All Dis'trJHicVtSFsExceeding Their ADA Enrollment Cap, 1998–99 to 2004–05 County, and Los Angeles perimeter seem the most similar in terms of the percentage of the school districts over their ADA cap. From 2001–02 on, more than 60 percent of adult schools in these regions were over their ADA cap. Similarly, between 40 and 50 percent of San Diego County school districts exceeded their cap during this seven-year period. The rest of the state is the least affected region in all years except 2002–03. By 2003–04, all regions began to converge, partly because the percentage of districts over their cap fell significantly in the Bay Area and in all Southern California counties. Each bar in Figure 4.6 represents the percentage difference between reported and funded ADA, focusing on districts that exceed their adult education ADA cap. Over 60 percent of overcap districts in Los Angeles County exceeded their cap by 2.5 percent or less. However, this is not 38 1FSDFOUBHF      mm                              -PT"OHFMFT -PT"OHFMFT $PVOUZ QFSJNFUFS #BZ "SFB     $FOUSBM 4BO%JFHP 3FTUPG $BMJGPSOJB 7BMMFZ $PVOUZ UIFTUBUF 4063$&$BMJGPSOJB%FQBSUNFOUPG&EVDBUJPO 1SJODJQBM"QQPSUJPONFOU6OJU /05&5IFBOBMZTJTJTSFTUSJDUFEUPBEVMUTDIPPMTXJUISFQPSUFE"%"BUMFBTUFRVBMUPUIFJS "%"DBQ Figure 4.6—Percentage by W'JhHiVcShFFunded ADA Is Exceeded, 2004–05 the case elsewhere. For instance, all adult schools in the Central Valley enrolled students in excess of the 2.5 percent funding limit. Statewide, nearly 80 percent of adult schools exceeded the 2.5 percent limit. Furthermore, one-third of all districts outside Los Angeles County overenrolled by 20 percent or more. A significant number of districts that exceed their cap ADA of school districts, therefore, are not fully funded under the current funding formula.  Since certain community colleges also offer ESL courses, it is possible to examine whether the size or growth of the community college ESL population is correlated with the percentage of school districts that exceeded their ADA cap in the same region from 1998– 99 to 2003–04. A positive but insignificant effect of the percentage of adult schools over their ADA cap on the level of provision by community colleges is found (after controlling regional and yearly factors). One interpretation of a positive coefficient is that both providers respond to the same factors driving increases in enrollment not already captured by regional and yearly factors. Another interpretation is that community colleges increase 39 Teaching for Free? The extent to which the ADA cap affects the ability of school districts to provide enough high-quality instruction hours to students depends on whether adult schools receive the “appropriate” level of funding. One measure of what is appropriate is the dollar value of the difference between reported ADA and funded ADA. The dollar value of each ADA is the baseline revenue limit (BRL), and the product of the BRL and funded ADA is roughly the total funding that a school district receives for adult education that year. The product of BRL and reported ADA is the funding that the school district needs to be fully compensated for the instructional load it undertakes that year. Figure 4.7 graphs the percentage by which the value of funded ADA needs to increase to equal the value of reported ADA from 1998 to 2004. The Central Valley and Los Angeles perimeter are the two regions that consistently would have needed their funding levels to increase the most during this period. From 1998–99 to 2002–03, these two regions tended to overenroll students whose dollar value was 10 percent or more above the funding formula amount. This difference declined to about 8 percent from 2003–04 to 2004–05. Although school districts in San Diego County do not have a consistent pattern, in the early 2000s the dollar value of their overenrollment was between 11 and 15 percent above their funded level. For adult schools in Los Angeles County generally, the value of overenrollment was between 1 and 3 percent above their funded levels. This could be because adult schools there do not face as much demand for ESL courses as noted above or they chose to limit enrollment to or near their funded ADA cap. The Bay Area and the rest of the state are more like Los Angeles County than are the other three regions. Yet adult schools there still exceed their funding levels by 4 to 6 percent, about twice as much as the funding growth permits. In 2001–02, the funded amount for adult ADAs fell significantly short of the dollar value of enrollment, causing the graphed differences to spike. This gap was driven by a combination of enrollment growth (10% increase in 2001–02 from the previous year) and midyear cuts to adult education their level of provision when adult schools face pressure on their ADA. See Appendix Table B.5. 40 $FOUSBM7BMMFZ #BZ"SFB -PT"OHFMFT$PVOUZ -PT"OHFMFTQFSJNFUFS 4BO%JFHP$PVOUZ 3FTUPGUIFTUBUF   1FSDFOUBHF    m m m m m m m 4063$&$BMJGPSOJB%FQBSUNFOUPG&EVDBUJPO 1SJODJQBM"QQPSUJPONFOU6OJU  Figure 4.7—Percentage Change 'inJHVDSFollar Value of Funded ADA Needed to Reach Parity with Reported ADA, 1998–99 to 2004–05 (among other programs). After 2001–02, however, the percentage difference between the dollar value of enrollment and funding decreased, perhaps because adult schools cut back on enrollment. Nevertheless, even with the convergence between reported and funded ADA, adult schools continue to receive insufficient funding. Figure 4.8 shows the difference in dollar value between funded and actual ADA to better reflect the size of the adult education program (as well as the policy response to demand for ESL). Despite the previously documented relatively lower growth in the need for ESL, Los Angeles County experienced the largest dollar value of enrollment in excess of funding, perhaps because, given the large size of the districts there (such as LAUSD), a small percentage increase translates into a large dollar value. It is also possible that adult schools with greater physical capacity can 41 $FOUSBM7BMMFZ #BZ"SFB -PT"OHFMFT$PVOUZ -PT"OHFMFTQFSJNFUFS 4BO%JFHP$PVOUZ 3FTUPGUIFTUBUF    7BMVF NJMMJPOT     m m m m m m m 4063$&$BMJGPSOJB%FQBSUNFOUPG&EVDBUJPO 1SJODJQBM"QQPSUJPONFOU6OJU  'JHVSF Figure 4.8—Difference Between Dollar Value of Funded ADA and Reported ADA, 1998–99 to 2004–05 overenroll more students than smaller adult schools. At its peak, the deficit in Los Angeles County was $10 million. The Los Angeles perimeter consistently has a difference of more than $5 million. The difference in the Central Valley and the rest of the state is similar to the amount in the Los Angeles perimeter in most years ($3 million to $4 million), whereas the Bay Area’s difference is $2 million to $4 million in most years. The accumulated amount that adult schools are underfunded is $172 million during this seven-year period (see Appendix Table B.6). The continued budget problems since 2001 have been dealt with by deferring payments to the next fiscal year, starting in 2002–03. Yet in 2003–04, total funding fell by more than 2 percent, and enrollment declined by slightly less than 2 percent. In 2004–05, funding increased to $591 million, surpassing the previous high set in 2002–03. This restoration 42 of funding is indicated in Figure 4.8 by the decline in funding shortfall in 2004–05. Summary More ESL students enroll in adult schools than in community colleges or any other providers. Yearly, adult schools teach English to over 500,000 students, a trend that seems to be correlated with state funding. The recent budget problems have disproportionately reduced the growth in ESL enrollment. Because ESL programs make up more than 40 percent of adult school enrollment, this cap on adult education growth especially constrains ESL enrollment. In 2004–05, nearly 60 percent of all adult school districts exceeded their ADA enrollment growth cap as determined by the funding formula. Of districts outside Los Angeles County that exceeded their cap ADA, more than 70 percent of them exceeded the 2.5 percent funding growth limit. This finding and the finding that 60 percent of Los Angeles County adult schools have enrollment within the 0–2.5 percent range are consistent with the state’s demographic trends. That the bulk of the state’s adult schools exceed the 2.5 percent limit is indicative of the decision by adult schools to carry out their education mission without the appropriate level of funding. The dollar value of exceeding the 2.5 percent limit on ADA growth is more than $15 million per year, possibly because districts are caught between serving students who wish to learn English and the financial constraint the state places on them. 43 5. Policy Implications and Conclusion Adult schools must deal with two contradictory policies regarding ESL programs in the state when faced with increasing levels of need in excess of the 2.5 percent growth cap. The evidence presented in Chapter 4 supports the view that the majority of local agencies do not abide by the 2.5 percent growth limit when local demand for adult education exceeds the level of funding. In practice, the cap on ADA growth may affect only the funds received by school districts rather than the level of instruction they provide. What are the implications for policymakers resulting from the confluence of the demographic changes, the antiquated funding formula, and the education mission that local educators are charged with carrying out? Because what matters is how educators and local districts carry out their adult education mission, this chapter examines the consequences resulting from the difference between reported and capped ADA funding as described in the previous chapter. Even though many adult schools are enrolling more students than the funding formula would suggest, there are short- and long-term consequences. Specifically, this chapter considers how the quality of instruction, the ability to implement new programs, and outreach efforts suffer when adult education funding does not correspond to the actual level of instruction provided. Adult schools that exceed their ADA cap are likely forced to make tradeoffs in the level and type of services they provide as well as in their investment in innovative programs. Other providers likely experience the same issues. Although the discussion below focuses on the funding for adult schools, it is applicable to other providers facing funding constraints, such as community colleges and public libraries. Policy Consequences of Overenrolled Courses In 2004–05, nearly 60 percent of all districts in the state exceeded their enrollment caps, and the difference between what all districts received from the state ($591 million) and the dollar value of actual enrollment ($607 45 million) is $15.7 million. There are two possible interpretations of this gap. The optimistic view is that despite the apparent shortcoming, in practice the funding formula does not hinder the ability of school districts to fully meet the English instruction needs of immigrants, since so many school districts behave in a manner inconsistent with the funding formula’s limits. Therefore, the mission of adult schools may not be adversely affected by the funding formula, especially in the short run or when capacity is exceeded by enrollment of only a few students. Yet, even if local agencies still meet the needs of most English learners, by absorbing the marginal cost of this service, the quality of instruction suffers and the ability of adult schools to grow and better serve English learners is hindered. Thus, the less optimistic interpretation suggests that the funding formula has an adverse effect on adult schools. This occurs because the state is under no obligation to fund any ADA in excess of 2.5 percent of the previous year’s funded ADA, resulting in a net ADA rate that is less than the baseline revenue limit per ADA. For instance, in 2004–05, San Leandro Unified in the Bay Area received $1,188,972 in funding for 518.69 ADAs but reported enrolling 577.93 ADAs; although the baseline revenue limit is $2,292.26, the net or effective per-ADA rate is $2,057.29 ($1,188,972 ÷ 577.93). In the short run, adult schools accommodate unfunded students by increasing classroom size or mixing students of different levels into one classroom. The short-run adjustments (i.e., continuously overenrolling classes) may negatively affect learning because of diminishing returns to scale with regard to teaching quality. Adult schools may also respond by offering fewer advanced classes, scheduling fewer classes that meet during off-peak hours, or eliminating marketing or community outreach efforts that increase awareness of and interest in the program. In the long run, it is also possible that the inability of adult schools to expand their instructional capability limits not just the number of students who can enroll but also the development of innovative programs and curriculum. For instance, affected adult schools may also be restricted in the ways they can target new funds to accommodate emerging needs,  See the discussion in Chapter 1 about the interpretation of the state’s mission to English learners. 46 such as illiterate immigrants. Inevitably, the inability to expand may deter potential students from attempting to enroll or adult schools may have to turn away students. Partial Reform of the Adult School Funding Formula Many of those involved in studying or drafting policy for adult education describe the current funding formula as “antiquated” (Assembly Select Committee on Adult Education, 2003; de Cos, 2004; Levine Sherriff, 2003). This section examines the effects of AB 23, which reformed one aspect of the funding formula. It also describes the extent to which the funding formula fails to meet the funding needs of adult schools. The current level of funded ADA for each adult school district is a direct result of the levels established after Proposition 13. Many communities have experienced growth, but others have experienced declines in the need for adult education courses since the funding formula was implemented. Until the passage of AB 23 in 2005, adult schools that experienced growth exceeding 2.5 percent were not eligible to receive additional funding despite the fact that other low-demand districts did not use all of their allocated ADAs. Adult schools that did not earn their allocated ADAs had to refund unused ADA revenue to the Proposition 98 reversion fund (de Cos, 2004). The funding formula was not structured to incorporate regional variations in need; AB 23 addresses this shortcoming. The law creates a formula that permits high-demand districts to receive funds allocated to districts that experience continued enrollment declines. AB 23 specifies that, beginning in 2006–07, school districts that do not reach their ADA cap for two consecutive years will have the cap permanently reduced by an amount equal to half of the lowest unearned ADA from these two years. Since the redistribution of unearned funded ADAs cannot finance all unfunded ADAs, the bill gives priority to small districts (fewer than 100 ADAs), although the mechanism also ensures that larger districts (100 or more ADAs) receive additional funding.  An unearned ADA is equal to the difference between a district’s ADA cap and funded ADA.  Details of the bill are available at http://www.leginfo.ca.gov/pub/bill/asm/ab_00010050/ab_23_bill_20051004_chaptered.pdf. 47 The first distribution of additional ADAs will take place in 2006–07, but the two most recently certified data, the “R-1” version, are used to consider the effect of this bill. Hence, the simulated redistribution of ADAs in Figure 5.1 pertains to the 2005–06 fiscal year using 2003–04 and 2004–05 data (school-district-level ESL ADA was not made available for this report). This policy makes 3,326 ADAs available for distribution, with the bulk (2,603) coming from large districts. The rest of the state (720) and Los Angeles County (1,050) are by far the total largest contributors of unearned ADAs in the state. The Los Angeles perimeter receives the most additional funded ADAs of any region (1,271), double the amount received by Los Angeles County adult schools (626). To the extent that overenrolled districts face demand for ESL in excess of 2.5 percent, Figure 5.1 suggests that there is some variability in need for ESL courses within regions. For instance, in the Central Valley, small districts lose and gain nearly the same amount of ADAs. Nevertheless, the main effect of this reform comes from the redistribution of ADAs from one region to another. The net gain in ADAs from AB 23 is positive only for the Central Valley (208), the Los Angeles perimeter (858), and the rest of the state (23). Adult schools in the Bay Area (–391), Los Angeles County (–404), and San Diego County (–301) lose more than they gain. However, using reported ADAs to factor in the size of the adult education program gives insights into the relative effect of the partial reform. This approach reveals that the Los Angeles perimeter benefits the most of any region. Its gain represents a 3 percent increase in funded ADAs, whereas the Central Valley gains an additional 1.25 percent in ADAs, and the rest of the state gains less than 0.1 percent. San Diego County experiences the most negative effect, as it permanently loses 2.75 percent of its funded ADAs, according to this simulation. The Bay Area and Los Angeles County lose 1 and 0.3 percent of their allocated ADAs, respectively—relatively modest percentages considering that adult schools  One potential effect not considered is if school districts that would potentially be affected attempt to increase or maintain steady enrollment levels.  In 2005–06 and 2006–07, several districts are ineligible for any growth, except for a 0.5 percent increase solely for nursing programs. This potential small growth is assumed to be zero for simplicity and because this restriction is valid for only two fiscal years. 48 /VNCFSPG"%"VOJUT   3FEVDUJPO TNBMMEJTUSJDUT  3FEVDUJPO MBSHFEJTUSJDUT  *ODSFBTF TNBMMEJTUSJDUT  *ODSFBTF MBSHFEJTUSJDUT         m m m m m m  m  $FOUSBM 7BMMFZ #BZ -PT"OHFMFT -PT"OHFMFT 4BO%JFHP 3FTUPG "SFB $PVOUZ QFSJNFUFS $PVOUZ UIFTUBUF 4063$&"%"JOGPSNBUJPOGPSm 3 BOEm 3 JTQSPWJEFECZUIF $BMJGPSOJB%FQBSUNFOUPG&EVDBUJPO 1SJODJQBM"QQPSUJPONFOU6OJU /05&5IFGJHVSFJODMVEFTBGFXEJTUSJDUTOPUFMJHJCMFGPSSFEJTUSJCVUJPOQFS"# Figure 5.1—Redistributed A'JDHVASFsUnder AB 23, by Region, 2005–06 in these regions account for half of the unearned ADAs. Because districts that permanently lose ADAs are not using them, these reductions should not have an adverse effect on the adult education programs of these schools. How Much Does Partial Reform of the Funding Formula Help? These additional ADAs, however, do not directly measure the increase in the number of students that are now funded. In Table 4.1, it was shown that California ESL students spend an average of 113 hours in an ESL class, but they need 216 hours of instruction to successfully complete a beginning ESL course. Under the first assumption, 4.7 additional students can be taught and 2.4 additional students per ADA under the 49 /VNCFSFOSPMMFE second assumption. In total, 15,453 and 8,084 more students are funded statewide, respectively. Figure 5.2 shows the regional distribution of these students as a consequence of AB 23. Adult schools in the Los Angeles perimeter teach about 5,900 and 3,100 more students under the first, and then second, assumptions. Estimates under the first assumption are 3,500 and 3,000 for the rest of the state and Los Angeles County and 1,800 and 1,600, respectively, under the second assumption. The Bay Area and Central Valley would teach about 1,400 more students for 113 hours of instruction   IPVSTPGJOTUSVDUJPO IPVSTPGJOTUSVDUJPO            $FOUSBM 7BMMFZ /05&4FF'JHVSF #BZ -PT"OHFMFT -PT"OHFMFT 4BO%JFHP 3FTUPG "SFB $PVOUZ QFSJNFUFS $PVOUZ UIFTUBUF Figure 5.2—Increase in ESL Enro'lJHlmVSeFntas a Result of Redistribution of ADAs for Different Instructional Goals  Each ADA is 525 student-hours, so 525/113 = 4.65 students and 525/216 = 2.43 students. Multiplying these values by the ADAs yields the estimates.  Using the share of ESL enrollment in the region yields estimates very similar to the assumption of 216 hours of instruction. 50 and about 750 hours for 216 hours of instruction. San Diego County adult schools would teach 259 and 136 additional students under these two assumptions. Even the largest estimate of the additional funded students represents a small share of the unmet need as estimated in Chapter 3 or in the number of districts that exceed their ADA cap. The dollar value of the additional ADAs resulting from AB 23 is $7.6 million, whereas the estimated amount of unfunded ADA in 2004–05 is $15.7 million. In other words, even if the difference between what adult schools receive and the level of instruction they provide represents all of the unmet need for ESL instruction, AB 23 addresses less than half of this need. But it is more than likely that this gap does not represent the full level of need for ESL courses in the state. Policy Implications The low level of funding for adult education is considered the most important issue facing all adult education providers, including adult schools, community colleges, community-based organizations, religious organizations, and libraries (Crandall and Sheppard, 2004, p. 12). Any comprehensive strategy to increase the provision of ESL courses to adult immigrants through additional funding must include all providers, not just adult schools. Even though this report’s focus is on adult schools, the policy discussions pertaining to the findings are also relevant for other providers. The funding formula effectively compensates adult school enrollment at a lower per-pupil level than the baseline revenue limit. Adult schools that do not receive funding equal to the actual level of enrollment face tradeoffs between providing instruction in a given year or expanding instruction in subsequent years, as well as providing other services that encourage enrollment, such as child care. As already mentioned, the quality of instruction is lowered as a consequence of not fully funding the level of instruction in districts that exceed their ADA cap. This effect occurs in both the short and long run. Yet focusing on current needs potentially reduces the availability of funds for investment in new curricula, facilities, or technology to improve the learning environment. Two examples of innovative curriculum or teaching methodology that may be affected are distance learning and workplace ESL instruction. 51 These types of programs are particularly well-suited to address various issues in adult ESL. They aim to make the ESL curriculum more effective and to target hard-to-reach groups, such as those with inconsistent work schedules, lack of transportation, or hectic lifestyles. From the student’s point of view, these types of programs are attractive because they are also flexible with regard to pace and scheduling and because they focus the curriculum on topics most useful or of most interest to them. From the provider’s point of view, they reduce the physical constraints associated with expanding programs at the adult school site as well as potentially improving the program. It thus makes it easier for the provider to expand without fear that the quality of instruction suffers or that it may crowd out other programs. Distance learning (DL) is a technology-based delivery system that connects students with instructors even though they may be in different locations or not present at the same time. The motivations for complementing traditional in-class instruction with distance learning differ. In most cases, distance learning is considered an effective means by which to reach students who may otherwise participate less because of personal or other obstacles. For instance, the LAUSD application for program approval states, “Based on information gathered at the school sites, much of our adult ESL population had scheduling, travel or child care issues which made traditional school difficult to attend. DL also allowed ambitious adults to learn English faster. . . . Schools also reach hard-to-serve adults by setting up DL classes at businesses, factories, retail centers, churches, recreation centers and apartment complexes.” Since 1993, adult schools have been able to use no more than 5 percent of their funding for distance  The formal definition of DL is: “An instructional delivery system that connects learners with educational resources. DL provides educational access to learners not enrolled in educational institutions and can augment the learning opportunities of current students. The implementation of DL is a process that uses available resources and will evolve to incorporate emerging technologies” (http://www.cdlponline.org/index.cfm?fuse action=whatis&pg=33). See the state’s Innovation and Alternative Instructional Delivery Program website for full program description (http://www3.scoe.net/fivepercent/login. cfm?fuseaction=desc).  http://www3.scoe.net/fivepercent/pdfs/applications/2005/2005_2361_10112005_ 20512.pdf, p. 8. 52 learning programs and, presently, 99 adult school agencies participate in this program, mostly large- and medium-sized adult schools.10 Vocational ESL and workplace ESL are programs that aim to link the classroom with the workplace in a very direct way. Instruction in workplace ESL programs in fact takes place not in the classroom but at the English learner’s place of work.11 This program is generally provided in partnership with employers who provide the space and other resources, and the school system provides the curricular materials and the instructor. Employees attend classes during work hours or when they are not working. The content has immediate returns for the student and the employer and, because instruction is conveniently located and scheduled, students may be more likely to persist and complete more hours of instruction than they would in traditional in-class programs. This saves the district money in the long run because students are less likely to continually drop in and out of classes before completing their goal. Adult schools are not the only providers of these types of innovative programs. Nonprofit organizations, churches, public libraries, and community colleges also have courses that are geared toward the community’s particular needs (Crandall and Sheppard, 2004). These organizations may be able to provide smaller classes or may be able to foster greater trust and confidence from the community because the provider, such as a local church, is well known to them. Besides the effect on investment in new and effective programs, an insufficient level of funding may force adult schools to provide more ESL courses at the expense of other adult education programs. Schools faced with substantial demand for ESL can accommodate more students without substantially exceeding their enrollment cap by reducing the funding spent on other programs. For example, under a budget-neutral scheme, adult schools could shift more spending to ESL programs to teach an additional 100,000 students by proportionately reducing all other adult education programs. Under this scenario, 72 percent of this shift would come from 10 See the state’s Innovation and Alternative Instructional Delivery Program website for full program description (http://www3.scoe.net/fivepercent/login.cfm?fuseaction=desc). 11 Outreach and Technical Assistance Network (OTAN) provides significant background on vocational ESL programs (http://www.otan.us/wia/0607/pdfs/vesldoc.pdf). 53 reductions in vocational education, adult secondary education, and older adult programs.12 It is possible that some agencies are already practicing such a policy.13 Some policymakers, already facing budget constraints, may be reluctant to consider a policy of increased funding for all adult education programs. However, as other options require making tradeoffs between various adult education programs, policymakers may find it difficult to find a budgetneutral policy that benefits one constituency at the expense of another. The additional funding would permit affected providers to invest in teaching methods or facilities, such as distance learning, that enhance their ability to serve more students and reach out to students who are not being currently served. The additional funding would not just focus on adult schools but would also fund community colleges, public libraries, and nonprofit organizations. In the long run, such investment is likely to provide many positive benefits for the state and the immigrants wanting to learn English. Summary This chapter considered the policy implications of the findings from this report. The dollar value of exceeding the 2.5 percent limit on ADA growth is more than $15 million per year, possibly because districts choose to serve students who wish to learn English rather than turn them away. Yet, districts that do not receive full funding for all the ADAs they teach are likely constrained from expanding or adjusting to meet the needs of immigrants in the short and long term, since they incur the cost of exceeding their 2.5 percent growth cap. Assembly Bill 23 is the first attempt to deal with one of the shortcomings of the adult school funding formula. The bill makes it possible to redistribute unspent adult education funding to districts that exceed their ADA cap. This reform is a first step toward increasing the number of available ESL courses, but it fills less than 50 percent of the gap of the dollar value of the difference between enrollment and state funding. Nevertheless, certain adult schools in the state, especially those in the Los 12 This assumes a one-to-one exchange between programs. The average number of hours of instruction in ESL classes is 13 to 30 hours greater than in these programs. 13 I would like to thank Patricia de Cos for pointing out this possibility. 54 Angeles perimeter, benefit the most from this reform. Because redistributed ADAs would otherwise go unused, this benefit comes at a minimal cost to adult schools with declining enrollment. The inflexibility of the adult school funding formula continues to raise important questions about whether adult schools are sufficiently funded to accomplish their mission of teaching English learners. Other public providers of ESL courses also face funding shortages, and any policy must include these providers to better reach the population in need of ESL instruction. Consequently, adult education programs may be making difficult choices in accommodating the needs of present students. They may be sacrificing investing in new programs that would better accommodate future cohorts of students, or they may be reducing the availability of resources for other groups of adult learners. State policymakers must weigh whether increasing funding for adult education programs in the face of growing demand for ESL programs is a policy worth exploring, while they also consider the alternatives. Yet given the public interest over the pace of immigrant acquisition of English, policymakers must also recognize that the lack of available ESL courses may perpetuate the linguistic divide in California. 55 Appendix A Notes on Data and Methods Census Data The 1980, 1990, and 2000 5% PUMS datasets provide valuable information regarding the English-language ability of immigrants in California. The sample consists of foreign-born persons ages 18 years and older who are not U.S. citizens, although the sample also includes persons born in Puerto Rico. An LEP is defined as a person speaking English “not well” or not speaking any English. See the NHES discussion below for further sample restrictions made to estimate enrollment. The regional areas are constructed from the Census data using the most consistent grouping of counties across the three PUMS datasets, which do not overlap completely in each year. For this reason, certain counties are included in geographic regions that generally would not include them. For instance, San Benito County is included in the San Joaquin Valley group even though this county is not considered part of the Central Valley. The regional definitions used in this report conform to those used by CASAS in its reports. The regions are Los Angeles County; Bay Area—Alameda, Contra Costa, San Francisco, Santa Clara, and San Mateo Counties; Central Valley—Fresno, Kern, Merced, and Tulare Counties; Los Angeles perimeter—Orange, Riverside, San Bernardino, and Ventura Counties; San Diego County; and the rest of the state—all other counties. Estimates of Enrollment Data The two data sources used to estimate the probability of ESL enrollment are the NHES for 1999 and 2001. A virtue of these datasets is that they are nationally representative (with a small California sample) of persons ages 16 and older not enrolled in elementary or secondary school (the surveys are conducted in Spanish or English). The extensive language questions of these data provide valuable information on the nation’s 57 immigrant groups that participated in the surveys. The NHES asked eligible adults whether they enrolled in an ESL course in the previous 12 months. The sample drawn from these data includes persons born outside the United States and with an identifiable country of origin (Puerto Ricans could not be identified from the data and so anyone born in U.S. territories are exclude from the sample), at least age 18, and not enrolled in an ESL college program. The predicted enrollment estimates yield the size of the population with the characteristics described above. More information about the NHES can be found at http://nces.ed.gov/nhes. Since this report focuses on the policy issues dealing with adult schools and community colleges, the sample should ideally consist of those who enrolled only in these agencies. However, because of the small sample size of ESL enrollees, the sample includes persons who enrolled in adult schools, community colleges, community-based organizations, religious organizations, other public providers (such as libraries), private institutions, and other agencies. Of the final number who enrolled (145) in an ESL class in the previous 12 months, 112 enrolled in adult schools or community colleges, 17 in community-based or religious organizations, six in other public providers, nine in private business, and one did not specify. The estimates represent the total enrollment of all providers in the state, not just adult schools and community colleges. See Appendix Table B.4 for results of the estimates using only adult school and community college attendees. Another important limitation is that the NHES surveys were conducted in either Spanish or English, meaning that non-Spanish speakers who do not speak English well enough to participate in the survey may not be included, and therefore the estimates may not reflect the role of this population. A necessary assumption of the model is that those not surveyed regarding their ESL enrollment status because of low English skills would have answered the ESL question the same, on average, as Spanish speakers with low English skills. These individuals may more likely be enumerated in the Census, since lack of English is not a basis for exclusion. A minor issue with the 2001 NHES is also worth discussing. The 2001 NHES is the most recent NHES dataset to include questions on ESL  For consistency with previous years’ ESL question, the 2001 NHES ESL question was recoded to missing if the respondent’s first language was not English. Only nine cases were recoded to missing. 58 attendance. In both years, persons whose first language learned was not English were asked about the language they speak at home. Those speaking a language other than English were asked whether they had taken an ESL course in the past 12 months. However, in 2001, but not in 1999, persons who currently speak only English were also asked for ESL enrollment information if their first language was not English. Thus, for consistency across data, in 2001, responses for nine English-only speakers were recoded to missing. Empirical Model The following logit model is estimated for the combined 2001/1999 sample: ESLi = Xiβ + Califi′D + ui , (A.1) where ESL equals 1 if individual i took an ESL course and 0 otherwise. Xi consists of age, age at arrival, whether Hispanic, country of origin, highest education level completed, marital status, whether female, family size, and a California dummy variable. Our sample predictions for California are obtained by estimating β and using the values of Xi from the 5% PUMS for each year (1980, 1990, and 2000). Several sample restrictions have to be made to the Census data to make them compatible with the NHES sample: age greater than 17, not enrolled in school, not born in Puerto Rico (because it is not identified in the NHES data), not an English-only speaker at home, and not born in North America, Australia, or New Zealand. Using PUMS weights yields individual representative probability of enrollment and averaging within a region provides the expected number of persons who enroll in all ESL classes. The marginal effects are estimated at the means of the variables for the logit results, provided in Appendix Table B.1. However, to align as closely as possible the 2000 enrollment estimates with the 1999–00 fiscal year enrollment levels, it is necessary to apply the 2001–02 regional distribution of ESL programs. Regional information is not available for 1999–00. The regional distributions are stable, however, as the 2002–03 distribution is very similar to the 2001–02 distribution. It is unlikely, therefore, that the imputed regional distribution of ESL enrollment in 1999–00 suffers from significant measurement error. 59 Adult School and Community College Data This section provides more detailed information about adult school and community college data sources. For adult schools: 1. Annual enrollment counts and demographic information for ESL students in adult schools were obtained from the CASAS DynaReports website, http://www.casas.org/dynareps/dynareps2002.cfm. 2. ADA information and finance information for adult schools were obtained from Halena Le at the California Department of Education, Principal Apportionment Unit. For community colleges: 1. Community college data were obtained from the Management Information System department of the California Community College Chancellor’s Office. These data contain enrollment information at the class level. 2. Unduplicated student counts are used, as measured by the agencyspecific student ID. Thus, the same person may enroll in a community college or adult school. The enrollment data are based on student counts in noncredit courses and as such the sample does not exclude students that potentially were also enrolled in credit courses. 3. Finance data were provided by Ed Monroe and Elias Regalado at the Fiscal Services division at the CCCCO. 60 Appendix B Estimates and Results Table B.1 Logit Regression Results of the Probability of ESL Enrollment, NHES   Age Age at arrival Hispanic Mexico South and Central America Caribbean Asia Middle East, Africa, and other Southeast Asia Grades 0–3 Grades 9–11 Grades 12–15 Grades 16+ Married, spouse present Female All Providers Coefficient Marginal Effect   –0.100** (0.017) 0.073** (0.021) 2.022** (0.494) –2.146** (0.806) –2.049* (0.827) –1.400+ (0.821) –0.671 (0.739) –1.860* (0.762) –1.060 (0.832) –1.221+ (0.641) –0.819+ (0.475) –0.276 (0.349) –0.212 (0.409) –0.067 (0.327) –0.506+ (0.277) –0.005** (0.001) 0.004** (0.001) 0.089** (0.022) –0.120* (0.052) –0.065** (0.018) –0.046** (0.017) –0.029 (0.026) –0.056** (0.013) –0.039+ (0.020) –0.042** (0.014) –0.035* (0.017) –0.014 (0.017) –0.011 (0.020) –0.004 (0.018) –0.027+ (0.015) Adult Schools and Community Collegesa Coefficient Marginal Effect –0.096** (0.019) 0.077** (0.022) 2.448** (0.530) –0.758 (1.159) –0.652 (1.174) –0.008 (1.183) 0.503 (1.207) –0.465 (1.158) 0.176 (1.334) –1.103 (0.705) –0.677 (0.521) –0.111 (0.372) –0.058 (0.504) –0.303 (0.369) –0.594* (0.293) –0.003** (0.001) 0.003** (0.001) 0.066** (0.016) –0.025 (0.039) –0.018 (0.027) –0.000 (0.039) 0.020 (0.057) –0.013 (0.028) 0.006 (0.051) –0.024* (0.010) –0.019 (0.012) –0.004 (0.012) –0.002 (0.016) –0.011 (0.013) –0.020+ (0.010) 61 Table B.1 (continued) All Providers Adult Schools and Community Collegesa   Coefficient Marginal Effect   Coefficient Marginal Effect Family size –0.041 –0.002 0.024 0.001 (0.085) (0.005) (0.088) (0.003) California 0.673* 0.040* 1.030** 0.041** (0.280) (0.018) (0.328) (0.014) Constant 0.181 –2.440+ (0.938) (1.348) No. of observations 1,083 1,083   1,050 1,050 SOURCE: Author’s calculations from the 1999/2001 NHES. NOTES: Standard errors are in parentheses. The sample is restricted to persons who are age 18 or older and did not enroll in an ESL college program. Marginal effects are estimated at the means of the variables. aRestricted to respondents indicating type of agency attended as adult school or community college. +Significant at 10 percent. *Significant at 5 percent. **Significant at 1 percent. 62 63 Table B.2 Mean Characteristics of Immigrant Samples in Census and NHES Census 1980 1990 Age 41.68 40.88 Age at arrival 30.27 28.76 Hispanic 0.56 0.58 Mexico 0.43 0.43 South and Central America 0.08 0.11 Caribbean 0.02 0.01 Europe 0.16 0.09 Asia 0.14 0.15 Middle East, Africa, and other 0.06 0.06 Southeast Asia 0.11 0.14 Grades 0–3 0.18 0.19 Grades 4–8 0.26 0.18 Grades 9–11 0.12 0.11 Grades 12–15 0.31 0.38 Grades 16+ 0.13 0.14 Married, spouse present 0.65 0.58 Female 0.52 0.50 Family size 3.93 4.15 No. of observations 104,739 159,178 Population size 2,096,226 3,236,796 NOTE: See Appendix A for sample selection. 2000 42.41 25.12 0.58 0.47 0.11 0.01 0.06 0.17 0.03 0.16 0.14 0.18 0.11 0.39 0.18 0.67 0.50 4.32 306,744 6,189,076 NHES 1999/2001 37.93 23.20 0.68 0.45 0.15 0.07 0.06 0.12 0.09 0.06 0.06 0.26 0.18 0.31 0.19 0.69 0.47 4.04 1,083 26,808,454 Los Angeles County Census Sample 1990 2000 39.97 42.63 28.80 25.25 0.65 0.63 0.43 0.46 0.17 0.17 0.02 0.01 0.07 0.06 0.15 0.16 0.07 0.03 0.10 0.11 0.18 0.14 0.13 0.16 0.20 0.19 0.12 0.12 0.37 0.39 0.55 0.64 0.50 0.50 4.18 4.26 72,507 122,332 1,478,753 2,486,533 Non–Los Angeles County Census Sample 1990 2000 41.66 28.73 0.53 0.42 0.07 0.01 0.10 0.16 0.06 0.18 0.19 0.17 0.10 0.39 0.15 0.60 0.50 4.13 86,671 1,758,043 42.26 25.03 0.54 0.48 0.07 0.00 0.07 0.17 0.03 0.19 0.14 0.17 0.10 0.40 0.19 0.69 0.50 4.35 184,412 3,702,543 Table B.3 Mean Characteristics of Enrolled ESL Students   Adult Community Schools Colleges Female 0.56 0.56 Male 0.44 0.44 No high school diploma High school diploma or GEDa 0.60 0.25 0.66 0.29 Post–high school 0.15 0.05 Ages 16–20 0.11 0.13 Ages 21–30 0.35 0.36 Ages 31–40 0.27 0.25 Ages 41–50 0.14 0.14 Ages 51–60 0.08 0.06 Ages 61+ 0.04 0.05 White 0.06 0.08 Hispanic 0.73 0.67 Asian 0.17 0.22 Filipino or Pacific Islander 0.02 0.01 Other 0.03 0.02 SOURCES: CASAS DynaReports Adult School data for 1999–00 to 2003–04, and CCCCO for 1997–98 to 2003–04. NOTE: The CCCCO sample includes unduplicated students with nonmissing information for that variable. aGED is general equivalency diploma. 64 Table B.4 Predicted Enrollment Probabilities and Levels of Enrollment for Adult School or Community College ESL Enrollees Probability of Enrollment (%) Predicted Enrollment Central Valley Bay Area Los Angeles County Los Angeles perimeter San Diego County 1980 12.5 7.9 14.6 13.1 10.6 1990 2000 14.9 10.7 8.7 7.0 14.4 8.8 14.8 9.4 11.5 7.9 1980 10,686 30,601 149,696 34,086 12,817 1990 20,753 47,834 212,483 78,348 24,489 2000 30,445 81,488 219,907 107,974 32,117 Rest of the state 10.0 11.6 9.5 21,178 37,804 66,855 Total 12.4 13.0 8.7   259,064 421,710 538,785 NOTE: Estimates are based on the second regression in Table B.1. 65 Table B.5 Effect on Noncredit ESL Enrollment of Districts Exceeding Their ADA Cap Coefficient Estimate Coefficient Estimate Percentage of adult schools over their ADA cap 43,564.615** 6,280.917 (15,684.726) (5,639.659) Year = 1999 2,308.697+ (1,175.183) Year = 2000 3,158.999* (1,315.961) Year = 2001 4,693.076** (1,131.226) Year = 2002 3,045.705* (1,116.420) Year = 2003 –648.739 (1,466.803) Bay Area 14,558.988** (1,688.464) Central Valley –13,796.300** (1,127.439) Los Angeles perimeter 28,290.014** (1,574.876) Los Angeles County 8,634.902** (1,429.417) San Diego County 15,568.116** (1,215.532) Constant 2,405.690 9,999.777** (8,095.254) (2,463.964) No. of observations 36 36 R-squared 0.18 0.99 SOURCES: Data are from CCCCO and California Department of Education, Principal Apportionment Unit, 1998–99 to 2003–04. NOTES: The dependent variable is the level of ESL provision by community colleges. Standard errors are in parentheses. +Significant at 10 percent. *Significant at 5 percent. **Significant at 1 percent. 66 Table B.6 Data for Figures 4.7 and 4.8   1998–99 1999–00 2000–01 2001–02 2002–03 2003–04 2004–05 Value of Reported ADAs ($ millions ) Central Valley 35.1 36.5 37.2 41.3 41.0 39.4 39.4 Bay Area 73.4 73.9 76.8 88.6 90.9 89.0 85.8 Los Angeles County 241.3 245.5 256.0 290.1 301.2 297.5 298.8 Los Angeles perimeter 55.8 57.5 59.4 64.7 66.3 64.4 65.5 San Diego County 24.6 24.1 25.2 28.0 29.3 26.3 25.2 Rest of the state 71.2 73.4 74.3 82.4 84.7 82.4 92.4 Total 501.5 510.9 528.8 595.1 613.4 599.1 607.1 Value of Funded ADAs ($ millions) Central Valley 30.7 32.3 33.3 36.0 36.9 36.3 36.6 Bay Area 70.2 71.8 75.2 84.0 86.6 85.4 83.4 Los Angeles County 237.8 241.9 252.5 281.6 290.6 286.8 296.5 Los Angeles perimeter 51.1 52.6 54.4 58.4 60.4 59.4 61.3 San Diego County 22.5 22.8 22.7 24.3 25.9 24.8 24.7 Rest of the state 67.2 69.0 71.0 77.3 80.7 78.3 88.8 Total 479.4 490.2 509.3 561.6 581.1 571.1 591.3 Difference (Value of Reported ADA – Value of Funded ADA) ($ millions) Central Valley 4.4 4.2 3.9 5.2 4.1 3.1 2.8 Bay Area 3.2 2.1 1.5 4.6 4.3 3.6 2.4 Los Angeles County 3.5 3.6 3.5 8.5 10.6 10.7 2.3 Los Angeles perimeter 4.7 4.9 4.9 6.3 5.8 4.9 4.1 San Diego County 2.1 1.3 2.4 3.7 3.4 1.5 0.6 Rest of the state 4.0 4.4 3.3 5.2 4.0 4.1 3.6 Total 22.1 20.6 19.6 33.4 32.3 28.0 15.7 Percentage Difference Between Reported and Funded ADA Central Valley 14.4 13.1 11.7 14.6 11.2 8.4 7.5 Bay Area 4.5 3.0 2.0 5.4 5.0 4.3 2.9 Los Angeles County 1.5 1.5 1.4 3.0 3.6 3.7 0.8 Los Angeles perimeter 9.3 9.2 9.0 10.7 9.7 8.3 6.7 San Diego County 9.5 5.8 10.8 15.3 13.3 6.0 2.3 Rest of the state 6.0 6.4 4.7 6.7 5.0 5.3 4.0 Total 4.6 4.2 3.8 6.0 5.6 4.9 2.7 SOURCE: California Department of Education, Principal Apportionment Unit. 67 References Assembly Select Committee on Adult Education, Assessing and Addressing Need: A Local Provider’s Perspective Hearing, Testimony Provided by Santiago Jackson, Assistant Superintendent Division of Adult and Career Education, Los Angeles Unified School District, Sacramento, California, 2003. Bauer, Thomas, Gil. S. Epstein, and Ira N. Gang, “The Role of the Family in Immigrants’ Destination Language Acquisition,” Journal of Population Economics, Vol. 18, No. 4, 2005, pp. 631–647. Berman, Eli, Kevin Lang, and Erez Siniver, “Language-Skill Complementarity: Returns to Immigrant Language Acquisition,” Labour Economics, Vol. 10, No. 3, 2003, pp. 265–290. Bloom, David E., and Gilles Grenier, “Language, Employment, and Earnings in the United States: Spanish-English Differentials from 1970 to 1990,” International Journal of the Sociology of Language, Vol. 121, 1996, pp. 45–68. California Community Colleges Chancellor’s Office, Understanding Funding, Finance, and Budgeting: A Manager’s Handbook, Sacramento, California, 1999. California Department of Education, Adult Education Office, Adult Education & Family Literacy Act California State Plan 1999–2004 Extended Through 6/30/05, Sacramento, California, 2004. California Department of Education, Adult Education Office, Adult Education Handbook for California, 2005 Edition, Sacramento, California, 2005. California Department of Education, Adult Education Office, California Annual Performance Report: Federally Funded Workforce Investment Act Title II Programs Program Year July 1, 2004–June 30, 2005, Sacramento, California, 2006. Capps, Randolph, Michael E. Fix, Jeffrey S. Passel, Jason Ost, and Dan Perez-Lopez, A Profile of the Low-Wage Immigrant Workforce, Brief No. 4. The Urban Institute, Washington, D.C., 2003. Carliner, Geoffrey, “The Language Ability of U.S. Immigrants: Assimilation and Cohort Effects,” International Migration Review, Vol. 34, No. 1, 2000, pp. 158–182. 69 CASAS, Student Progress and Goal Attainment Report: Adult School Programs in California 2002–03, San Diego, California, 2004. Chiswick, Barry, and Paul W. Miller, Do Enclaves Matter in Immigrant Adjustment? IZA DP No. 449. IZA (Institute for the Study of Labor), Bonn, 2002. Chiswick, Barry R., and Paul W. Miller, “The Endogeneity between Language and Earnings: International Analyses,” Journal of Labor Economics, Vol. 13, No. 2, 1995, pp. 246–288. Crandall, JoAnn, and Ken Sheppard, Adult ESL and the Community College, Council for Advancement of Adult Literacy, New York, New York, 2004. de Cos, Patricia L., Educational Opportunities for Adults in California, California Research Bureau, California State Library, Sacramento, California, 2004. Dustmann, Christian, and Arthur Van Soest, “Language and the Earnings of Immigrants,” Industrial and Labor Relations Review, Vol. 55, No. 3, 2002, pp. 473–492. Gonzalez, Arturo, “The Acquisition and Labor Market Value of Four English Skills: New Evidence from NALS,” Contemporary Economic Policy, Vol. 18, No. 3, 2000, pp. 259–269. Hill, Laura E., The Socioeconomic Well-Being of California’s Immigrant Youth, Public Policy Institute of California, San Francisco, California, 2004. Jepsen, Christopher, and Shelley de Alth, English Learners in California Schools, Public Policy Institute of California, San Francisco, California, 2005. Johnson, Hans P., “A State of Diversity: Demographic Trends in California’s Regions,” California Counts: Population Trends and Profiles, Vol. 3, No. 5, Public Policy Institute of California, San Francisco, California, 2002. Johnson, Hans P., Belinda I. Reyes, Laura Mameesh, and Elisa Barbour, Taking the Oath: An Analysis of Naturalization in California and the United States, Public Policy Institute of California, San Francisco, California, 1999. Joint Board Task Force on Noncredit and Adult Education, Challenges Opportunities Changes, California State Board of Education and the Board of Governors of the California Community Colleges, Sacramento, California, 1998. Lazear, Edward P., “Culture and Language,” Journal of Political Economy, Vol. 107, No. 6, Part 2, 1999, pp. S95–S126. 70 Levine Sherriff, Rona, Adult Education: Will It Meet the Challenges of the Future? Senate Office of Research, Sacramento, California, 2003. Little Hoover Commission, We the People: Helping Newcomers Become Californians, Little Hoover Commission, Sacramento, California, 2002. Martinez, Tia Elena, and Ted Wang, Supporting English Language Acquisition: Opportunities for Foundations to Strengthen the Social and Economic Well-Being of Immigrant Families, The Annie E. Casey Foundation and Grantmakers Concerned with Immigrants and Refugees, Baltimore, Maryland, 2005. McManus, Walter S., “Labor Market Costs of Language Disparity: An Interpretation of Hispanic Earnings Differences,” American Economic Review, Vol. 75, No. 4, 1985, pp. 818–827. Moore, Colleen, Nancy Schulock, and David Lang, Funding Adult Education: Does California Put the Money Where the Needs Are? Faculty Research Fellows Program, Center for California Studies, California State University, Sacramento, California, 2004. National Center for Family Literacy and National Center for ESL Literacy Education at the Center for Applied Linguistics, “Practitioner Toolkit: Working with Adult English Language Learners,” available at http://www.cal.org/CAELA/tools/program_development/elltoolkit/ CombinedFilesl.pdf, accessed December 11, 2006. National Center for the Study of Adult Learning and Literacy, The First Five Years, Harvard University Graduate School of Education, Cambridge, Massachusetts 2002. Pew Hispanic Center, Hispanic Attitudes Toward Learning English, Washington, D.C., 2006. Portes, Alejandro, and Ruben G. Rumbaut, Immigrant America: A Portrait, University of California Press, Berkeley, California, 1996. Ramakrishnan, S. Karthick, and Mark Baldassare, The Ties That Bind: Changing Demographics and Civic Engagement in California, Public Policy Institute of California, San Francisco, California, 2004. West, Linda L., Meeting the Challenge: A History of Adult Education in California from the Beginnings to the 1990s, California Department of Education, Sacramento, California, 1995. Young, Malcolm B., National Evaluation of Adult Education Programs: Executive Summary, Development Associates, Arlington, Virginia, 1995. 71 About the Author ARTURO GONZALEZ Arturo Gonzalez is a research fellow at the Public Policy Institute of California. He was previously an associate professor at the University of Arizona and a visiting professor at the Carlos III University in Madrid. His research interests include the acquisition of human capital by lowskilled workers and immigrants, including studies on the effect of the Job Corps on the earnings of Hispanics and non-Hispanics, and the role of English-as-a-second-language classes in improving the English ability of adult immigrants. His publications include Mexican Americans & the U.S. Economy: Quest for Buenos Días (2002). He received a Ph.D. in economics from the University of California, Santa Barbara. 73 Related PPIC Publications Educational Progress Across Immigrant Generations in California (2005) Deborah Reed, Laura E. Hill, Christopher Jepsen, Hans P. Johnson English Learners in California Schools (2005) Christopher Jepsen, Shelley de Alth The Socioeconomic Well-Being of California’s Immigrant Youth (2004) Laura E. Hill The Ties That Bind: Changing Demographics and Civic Engagement in California (2004) S. Karthick Ramakrishnan, Mark Baldassare 75" ["post_date_gmt"]=> string(19) "2017-05-20 09:37:49" ["comment_status"]=> string(4) "open" ["ping_status"]=> string(6) "closed" ["post_password"]=> string(0) "" ["post_name"]=> string(8) "r_407agr" ["to_ping"]=> string(0) "" ["pinged"]=> string(0) "" ["post_modified"]=> string(19) "2017-05-20 02:37:49" ["post_modified_gmt"]=> string(19) "2017-05-20 09:37:49" ["post_content_filtered"]=> string(0) "" ["guid"]=> string(50) "http://148.62.4.17/wp-content/uploads/R_407AGR.pdf" ["menu_order"]=> int(0) ["post_mime_type"]=> string(15) "application/pdf" ["comment_count"]=> string(1) "0" ["filter"]=> string(3) "raw" ["status"]=> string(7) "inherit" ["attachment_authors"]=> bool(false) }