Donate
Independent, objective, nonpartisan research

English as a Second Language in Californias Community Colleges

Database

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

View live version

object(Timber\Post)#3710 (45) { ["ImageClass"]=> string(12) "Timber\Image" ["PostClass"]=> string(11) "Timber\Post" ["TermClass"]=> string(11) "Timber\Term" ["object_type"]=> string(4) "post" ["custom"]=> array(6) { ["_wp_attached_file"]=> string(66) "english-as-a-second-language-in-californias-community-colleges.pdf" ["wpmf_size"]=> string(6) "862772" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(109594) "APRIL 2019 Olga Rodriguez, Sarah Bohn, Laura Hill, and Bonnie Brooks with research support from Sergio Sanchez and Stephanie Pulles Supported with funding from Bill & Melinda Gates Foundation, the College Futures Foundation, and the Sutton Family Fund English as a Second Language in California’s Community Colleges © 2019 Public Policy Institute of California PPIC is a public charity. It does not take or support positions on any ballot measures or on any local, state, or federal legislation, nor does it endorse, support, or oppose any political parties or candidates for public office. Short sections of text, not to exceed three paragraphs, may be quoted without written permission provided that full attribution is given to the source. Research publications reflect the views of the authors and do not necessarily reflect the views of our funders or of the staff, officers, advisory councils, or board of directors of the Public Policy Institute of California. SUMMARY CONTENTS Introduction 5 Current Landscape of ESL Programs 7 Degree-seeking ESL Students: Demographics and Outcomes 9 How Do ESL Reforms Affect Student Achievement? 17 ESL Reforms in Response to AB 705 21 Conclusion and Recommendations 27 The technical appendices to this report are available on the PPIC website. Due in large part to their accessibility and affordability, California’s community colleges (CCCs) play a central role in educating students who are not fully proficient in English. While there are many motivations for learning English, the economic benefits of English language skills are clear: in an economy that increasingly demands skilled workers, effective ESL programs can facilitate social and economic mobility for non-native speakers and their families. This report aims to fill the gap in research on ESL programs at California’s community colleges and the effectiveness of reforms aimed at improving student success. A number of colleges have implemented curricular reforms in recent years—accelerating coursework, integrating skills courses, and the like. These reforms are likely to become more widespread in the wake of Assembly Bill (AB) 705, signed into law in 2017. In addition to mandating that community colleges amend how students are placed in remedial courses, AB 705 requires that credit ESL sequences—which offer credit-bearing courses for which students pay tuition—maximize the probability that students enter and complete transfer-level English coursework within three years. Our research suggests that while many colleges are moving in the right direction, more could be done. Specifically, we find that:  Most degree-seeking ESL students do not successfully complete transfer-level English; those who begin at lower proficiency levels are especially likely to drop off the pathway. We find that 34 percent of all degree-seeking ESL students who first enrolled between 2010 and 2012 successfully completed transfer-level English within six years—but this share includes 56 percent of those who began one level below transfer-level English and just 9 percent of those who began eight levels below transfer.  Nearly half of colleges offer no more than five levels of ESL, which would theoretically allow students to complete transfer-level English in the sixth term. Still, only 20 percent of students who started five levels below transfer completed this course within six years and even fewer did so within three years.  Colleges have made great strides in moving away from a traditional approach—which teaches language skills such as reading and writing separately—to an integrated approach that teaches multiple language skills in a single course. Our findings suggest that all students benefit from the integration of language skills in ESL coursework.  Many colleges offer ESL courses that are transferable to UC and CSU. Students are benefiting from the opportunity to make progress toward degree and transfer goals while gaining proficiency in academic English. Recent efforts to secure humanities credit for advanced ESL courses may further boost the impact of transferable ESL coursework. PPIC.ORG English as a Second Language in California’s Community Colleges 3  Many colleges still offer ESL sequences that require students to complete developmental English coursework prior to enrolling to transfer-level English. Our findings indicate that these colleges would see a boost in outcomes if ESL sequences led directly to transfer-level English. This structural change is supported by AB 705, which recognizes that instruction in ESL is distinct from remediation in English. Our analysis suggests that current ESL program reforms hold promise. As colleges across the state move toward compliance with AB 705, more research is needed to determine whether new approaches to ESL assessment and placement and acceleration help improve the outcomes of diverse groups of students. PPIC.ORG English as a Second Language in California’s Community Colleges 4 Introduction Community college ESL programs serve a large and diverse mix of students, including young adults who attended California’s K–12 schools, immigrants with high school, college, or graduate degrees from their home countries, and working-age immigrants in California’s labor force. In the 2016–17 academic year alone, more than 58,000 students enrolled in one or more English as a Second Language (ESL) courses at community colleges across the state.1 Some ESL students need to improve their English in order to progress toward associate degrees, career certificates, or transfers to four-year schools. Others enroll in ESL for personal reasons or to improve their job prospects. The range of motivations and goals present a challenge in ensuring successful outcomes for all students.2 But in an economy where job opportunity is tied to skills and higher education (Johnson, Cuellar Mejia, and Bohn 2018), research has consistently shown that acquiring higher levels of English proficiency improves labor market outcomes (Bleakly and Chin 2004; Chiswick and Miller 1995; Gonzales 2000; Rivera-Batiz 1990). In addition to this diverse student population and other complexities, data constraints make it difficult to assess ESL programs. But a better understanding is critical, especially now. California’s community colleges (CCCs) are in the midst of reforming many programs that aim to prepare students for transfer-level coursework. A number of colleges have implemented curricular reforms in recent years to improve student outcomes. In addition, Assembly Bill 705, signed into law in 2017, requires community colleges to amend how students are assessed and placed and to maximize the probability that students who start on ESL pathways will complete transfer-level English coursework within three years (see text box).3 Previous PPIC research on developmental math and English reforms found policies aimed at accelerating developmental sequences and refining course placement have the potential to markedly improve student outcomes (Cuellar Mejia, Rodriguez, and Johnson 2018; Rodriguez, Cuellar Mejia, and Johnson 2018, 2017). This report offers a similar analysis of ESL programs in light of reforms that are being considered or are already under way. We use qualitative and quantitative data to shed light on the breadth of students in ESL programs and how colleges aim to meet student needs. Longitudinal student records collected by the Chancellor’s Office through its management information system (MIS) form the foundation for our analysis. But this data cannot provide course information that is critical to understanding how changes in ESL programs affect student outcomes—accurate information on the skills taught in each course, whether a course is a prerequisite for transfer-level English, or how many levels below transfer-level English it is. For this reason, we collected detailed information on the characteristics of ESL courses and programs from course catalogs and faculty input. Our analysis is informed by semi-structured interviews we conducted with ESL department chairs, faculty, and division deans at 13 community colleges in November and December 2018. We are grateful for the insights of our interviewees, many of whom are evaluating how to revise their programs to accomplish the goals of AB 705 (see Technical Appendix A for details). The bulk of this report examines the pathways of students who intend to continue their community college education beyond ESL.4 We focus on these students and their outcomes for two main reasons. AB 705 aims to 1 This statistic is based on an unduplicated count of students enrolled in ESL courses at a given college in academic year 2016-2017. Students who enrolled in ESL courses at multiple colleges in that year may be counted twice. During this year, 12,098 sections of ESL were offered across the system. We estimate that out of all math, English and ESL students in the CCC, those in ESL represent about 17 percent of the entire population (i.e., not restricted to degree-seeking students). 2 For background see Almon 2012; Bailey and Weininger 2002; Bunch and Endris 2012; Bunch et al 2011; Suarez-Orozco, Suarez-Orozco and Todorova 2008. 3 The equivalent timeframe for math and English is one year. 4 We estimate that degree-seeking ESL students represented about 8 percent of all degree-seeking math, English, and ESL students in the CCC between the 2009–10 and 2014–15 academic years. The share of ESL students in the degree-seeking sample is lower than the ESL share in the overall population because English learners are less likely than their math and English peers to enter the CCC with an intent to pursue a degree or transfer. PPIC.ORG English as a Second Language in California’s Community Colleges 5 AB 705 and credit-bearing ESL Assembly Bill 705 was signed into law by Governor Brown in October 2017. It is probably best known for its reforms to developmental education: it requires community colleges to maximize the probability that students will enter and complete transfer-level coursework in English and mathematics and mandates that colleges use high school records (e.g., coursework, grades, and/or grade point average) as the primary criteria for placement recommendations (California Legislative Information 2017). However, the bill also acknowledges that ESL is distinct from remediation in English, in that it serves foreign language learners seeking proficiency in an additional language. AB 705 requires colleges to maximize the probability that students who enter credit-bearing ESL course sequences will “complete degree and transfer requirements in English within three years.” The three-year timeframe begins when a student declares an intent to pursue a degree or transfer to a four-year institution. Importantly, the law only applies to credit-bearing ESL sequences. To help colleges meet the compliance deadline of fall 2020, the Chancellor’s Office and the Academic Senate are developing recommendations. Memos released in July 2018 and April 2019 provide guidance on using evidence-based rules for placing students into credit ESL to maximize the probability that they will complete transfer-level English within three years (six semesters or nine quarters). The memos also offer a series of curricular recommendations. The memos also suggest that colleges use high school GPA for placement, integrate ESL skills courses, ensure that ESL sequences lead directly to transfer-level English instead of developmental English, and the like. Additional guidance is expected to inform placement within the ESL sequence, especially for students who did not graduate from US high schools and those with no high school records—groups that can comprise sizable shares of the ESL population (Perez and Stanskas 2018; Perez 2019). Our data suggest that two-thirds of degree-seeking ESL students in California do not have a US high school diploma; a similar share are noncitizens (e.g., permanent residents, students with visas, refugees/asylees, etc.) and about a third are non-traditional-age (see technical appendix Table B6). improve outcomes for students intending to complete transfer-level English, a course that is required for students intending to earn an associate degree or transfer to a four-year university. Second, it is difficult to obtain data related to the success of ESL students who learn English for other reasons. For example, learning English might improve an individual’s job prospects, but although we are fortunate to have access to longitudinal student-level data through the Chancellor’s Office, information on career outcomes is not available. We start with a detailed picture of the current landscape of ESL course offerings that highlights the variation across colleges in terms of sequence length, type of sequence, and sequence end point. Next, we identify degreeseeking ESL students and examine their demographic characteristics and progress toward transfer-level English. We then explore how ESL program features affect student achievement, including the outcomes of ESL students from historically underrepresented backgrounds. Finally, we turn to the changes currently taking place in light of AB 705, examining key issues that colleges should consider. We conclude with several policy recommendations. PPIC.ORG English as a Second Language in California’s Community Colleges 6 Current Landscape of ESL Programs ESL programs are typically designed to teach a range of language skills—listening, speaking, reading, grammar, writing, vocabulary—at a variety of proficiency levels. Traditionally, each language skill was taught discretely in a standalone course. For example, a student might enroll concurrently in a reading course and a writing course, and the two courses would have largely unrelated content. More and more, colleges are offering integrated sequences, which teach more than one skill in a single course—for example, a student with mid-range English skills might enroll in an intermediate reading and writing course. A handful of colleges have adopted new course sequences that allow students to skip levels of ESL, depending their proficiency (these are known as accordion sequences), compress multiple levels of ESL into a single course, and support ESL students enrolled in transferlevel English courses. These recent reforms attempt to shorten the pathways from ESL to transfer-level English— with the aim of reducing attrition and improving student completion of ESL sequences.5 As evidence in developmental English and math suggests, shortening the pathway to transfer-level coursework and strengthening the alignment with the transfer-level course improves the odds that students enroll and pass transfer-level courses (Cuellar Mejia, Rodriguez, and Johnson 2018; Rodriguez, Cuellar Mejia, and Johnson 2018, 2017). ESL Pathways to Transfer-Level English Roughly 27 percent of ESL course sections are prerequisites for transfer-level English, or “sequence courses.” These course offerings reflect the needs of students at a given community college, but are also shaped by alternative ESL providers in the community, such as adult schools.6 ESL sequence courses vary substantially. Most (89%) confer credit but only roughly less than in four offer credit that is degree applicable or transferable to a four-year institution (see technical appendix Table B3 for more details). Sequence courses that do not confer credit are typically many levels below transfer-level English and intended for beginning English speakers at colleges in communities with few other ESL providers. A few colleges also offer noncredit ESL sequences—which offer tuition-free courses and are not subject to the AB 705 timeframe—that mirror their credit ESL pathways. It may seem odd that some ESL courses offer transfer credits even though they are prerequisites to transfer-level English.7 However, transferable ESL courses reflect the reality that learning English is not “remedial”; it is equivalent to a native English speaker learning a foreign language. AB 705 acknowledges this reality; it states that “ESL is distinct from remediation in English” and that “students enrolled in ESL credit coursework are foreign language learners.” Faculty we spoke to agreed that transferable ESL courses make sense because their rigor is comparable to that of a foreign language course. A secondary reason for making ESL credits transferable is to allow students to make faster progress toward their degree or transfer goals while learning English. As is 5 Recent legislation aims to do something similar in the K–12 system—that is, to increase English Learners’ access to college preparatory courses. AB 2735 requires that school districts allow English Learners full access to all of a school’s standard instructional program, including a–g courses as well as AP classes. AB 2121 requires districts to permit a fifth year of high school for newcomers and students in the Migrant Education Program in order to provide more opportunities to earn required graduation credits. 6 Adult education ESL and community college-based ESL often work in tandem to meet the needs of English learners in their community. English learners, especially those with very little or no English proficiency, may start coursework outside of community college, even if they intend to seek transfer-level English and/or college credentials. If these students continue on the pathway, we only observe them once they have entered a community college ESL sequence course. In our analysis, we assume that courses at a given level below transfer-level English are comparable, so that although we cannot observe prior English training in any detail, students are at similar levels of English proficiency. Six community college districts include adult education institutions, meaning that we can observe enrollment in adult education– based ESL. In those four districts alone, it is clear that there is a great demand for ESL even beyond students who enroll in community college. For example, in 2013, 79 percent of ESL students enrolled in adult education–based ESL in the San Diego Community College district, compared to 21 percent in the colleges in that district. Unfortunately, there are many more adult education schools that are not included in the MIS data. 7 See technical appendix Table A5 for more information on the types of ESL sequences that use transferable ESL courses. PPIC.ORG English as a Second Language in California’s Community Colleges 7 discussed in a later section, there are important developments with respect to the CSU/UC transferability of ESL in response to AB 705.8 Sequence courses in ESL typically focus on reading (54%) and writing (80%) skills, or integrate the two. Relatively few of these courses teach listening, speaking, and/or grammar, but most students (53%) who enroll in sequence courses at some point in their community college career also enroll in non-sequence courses that offer these additional skills. ESL Pathway Characteristics While ESL pathways take different forms, traditionally ESL sequences to transfer-level English offer sequential, separate skills courses as shown in Figure 1. As of the 2016–17 academic year, 29 colleges offer a traditional ESL sequence. FIGURE 1 A sample of a traditional ESL sequence SOURCE: Author’s chart based on PPIC ESL program database. The most common number of levels—offered at 25 percent of colleges—is six (Figure 2). AB 705 indicates that a five-level sequence would theoretically allow students to complete transfer-level English in six terms. Almost half of colleges, 48 percent, offer a sequence that is five or fewer levels,9 but relatively few students enroll at the lowest levels: 82 percent of students begin at level five or higher. This suggests that it may not be extraordinarily difficult to shorten sequences. 8 It is also important to point out that during the timeframe we study, five colleges also offered transferable ESL courses that were equivalent to freshman composition offered in the English Department in that they meet graduation requirements. In fact, 8 percent of students in our sample completed the freshman composition requirement via the ESL version of this course. In this study we opted not to highlight these courses because they do not transfer to all colleges and universities—and even when they do transfer, they may only transfer to certain programs. In fact, we find that 58 percent of students who enrolled in an ESL freshman composition course also took it a second time in the English Department, which contributes to the problem of excess credits. 9 Keep in mind that this only includes ESL sequence courses; some students may have taken non-sequence ESL courses in college or enrolled in ESL courses at adult schools or other venues. PPIC.ORG English as a Second Language in California’s Community Colleges 8 FIGURE 2 Some ESL sequences are lengthy, but most students start ESL only a few levels below transfer-level English Percent 30% 25% 20% 15% 10% 5% 0% 6% 1% 9+ 8% 2% 14% 4% 25% 10% 17% 13% 19% 16% 21% 11% 21% 2% Colleges Degree-seeking students 8% 2% 87654321 Levels below transfer-level English SOURCE: Author calculations from Chancellor’s Office MIS data and PPIC ESL program database. NOTES: Degree-seeking students are coded according to their first sequence ESL course, for students who first enrolled between the 2009–10 and 2016–17 academic years (N=120,365). Colleges are coded according to the longest sequence offered in the 2016–17 academic year. If ESL coursework leads to developmental English, the developmental English course(s) is counted in the sequence. See technical appendix Table B4 for raw numbers. In 2016–17 (the most recent academic year for which we have data), completing the ESL sequence was not sufficient for enrolling in transfer-level English at 40 colleges.10 At these colleges, students also needed to complete one or more developmental English courses. ESL students at these colleges would enter the sequence to transfer-level English at least two levels below transfer English. Faculty told us that the developmental English requirement was intended to help ESL students acquire the language skills needed for academic English courses. It may have been partly driven by a perception that only prerequisites taught in an English department can help students acquire the proficiency needed for transfer-level English. Current Chancellor’s Office guidance on AB 705 cites ESL-to-developmental-English pathways as a focus of reform. Of the 51 colleges that had ESL sequences leading to developmental English in the 2009–10 school year, 14 had changed their sequences to lead directly to transfer-level English as of 2016–17.11 These data suggest that many colleges are already in alignment with AB 705’s requirement that ESL students be able to complete transfer-level English within three years, or six terms. But whether students actually do complete transfer-level English within three years is another matter. As we show in the remainder of this report, a relatively small percentage of degree-seeking ESL students successfully complete transfer-level English even within six years. Degree-seeking ESL Students: Demographics and Outcomes We can identify degree/transfer seeking ESL students by looking at their course-taking behavior. Our degreeseeking cohort includes any ESL student who took at least one degree-applicable course outside of ESL or transfer-level English and/or at least one developmental English/math course at some point during their college 10 See technical appendix Table A5 for more information on the types of sequences that feed into transfer-level English and developmental English. 11 Some colleges introduce ESL programs/sequences after the 2009–10 academic year that are included for in the 40 colleges we observe as still offering ESL sequences that feed into developmental English coursework as of 2016–17, but are not part of the original 51 that offered this type of sequence in 2009–10. We derive 14 by asking the original 51 colleges how many changed their ESL sequences to feed directly into transfer-level English. PPIC.ORG English as a Second Language in California’s Community Colleges 9 journey.12 One-third of the 356,209 students who took at least one ESL course between 2009–10 and 2014–15 also took at least one ESL course that is a prerequisite for transfer-level English (i.e., a sequence class).13 As Figure 3 shows, the majority of ESL students (53%) only took ESL courses that are not prerequisites for transferlevel English courses (i.e., a non-sequence class).14 And the remaining 13 percent took at least one ESL sequence course but did not take any other courses consistent with degree/transfer-seeking.15 FIGURE 3 A third of ESL students are seeking degrees 53% 47% 34% non-sequence sequence, non-degree sequence, degree-seeking 13% SOURCE: Authors’ calculations based on MIS data and college catalog scan. Because our research focus is on ESL students seeking degrees or transfers, we omit the two-thirds of community college students who enrolled in ESL but did not demonstrate course-taking consistent with an intent to pursue a degree or transfer. In technical appendix Table B9, we compare demographic characteristics of degree-seeking ESL students and other ESL students. Overall, we find that degree-seeking ESL students are more likely to be Asian, to have a foreign high school diploma, to be a permanent resident, to be under the age of 25 (traditional college age), and to be a recent high school graduate than non-degree-seeking ESL students. We find that ESL students seeking degrees typically enter an ESL sequence three or four levels below transferlevel English, and 62 percent are traditionally college-aged. The most common prior educational attainment is a foreign secondary school diploma (41%), while 33 percent hold a US high school diploma.16 The most common immigration status is legal permanent resident (38%), followed by student visa holder (25%). US citizens make up 23 percent of degree-seeking ESL students. Nearly half of degree-seeking ESL students are Asian (47%); 23 percent are Latino, 19 percent are white, and 2 percent are black. There are some important educational and immigration-related differences across racial/ethnic groups that may be linked to students’ ESL experiences. Latino students are much more likely to have been US educated than other 12 Our definition of degree and/or transfer-seeking student relies on course taking behavior. CCCCO data administrators cautioned us against using the educational goal data element collected during our sample timeframe because it is considered an “uninformed” educational goal which is collected during the application process. Starting summer 2014, this data element is considered an “informed” educational goal because it is updated after a student has reviewed assessment results and received orientation and other student services. This educational goals variable is now being used by the CCCCO Student Success Metrics to define the degree-seeking population. See Technical Appendix A for more information. 13 Enrolled for the first time in an ESL course within this timeframe. We limit our sample to this timeframe in order to have at least a three-year follow-up for outcomes. 14 Non-prerequisite courses are advisory or elective courses that commonly teach listening/speaking, grammar, and some reading skills. See technical appendix Table B3 for more detail on the characteristics of these courses. 15 Some of these ESL students not seeking a college degree may have pursued or obtained a certificate. 16 It is possible to graduate from a US high school and still be considered an English Learner. High school graduation rates are lower for current English Learners than for students who have been reclassified or were never English Learners. PPIC.ORG English as a Second Language in California’s Community Colleges 10 students (Figure 4); 62 percent have either a high school diploma or GED. Only one-third of Asian students, 39 percent of black students, and 29 percent of white students have US diplomas or GEDs. The majority of white students (55%) hold a foreign secondary school diploma, as do 44 percent of Asian students and 42 percent of black students, compared to only 18 percent of Latino students. Black and white ESL students start four levels below transfer-level English, while Latino and Asian students start three levels below, on average. FIGURE 4 Latino degree-seeking ESL students are much more likely to be US-educated 100 90 14 10 16 11 11 80 18 Unknown Percent 70 41 60 50 40 5 30 20 33 10 06 Overall 9 53 9 Latino 44 5 29 5 Asian 42 9 31 7 Black 55 4 25 3 White Foreign secondary school diploma GDEDE// HS equivalency/ CA HS proficiency certificate High school diploma Adult school Less than HS SOURCES: Authors’ calculation based on MIS data. NOTE: Sample includes students first enrolled in ESL in the CCC system between the 2009–10 and 2014–15 academic years. See technical appendix Tables B1 and B2 for more details. While more than half of white ESL students are permanent residents, as are approximately 40 percent of black and Asian ESL students, only 29 percent of Latino ESL students are legal permanent residents (Figure 5). More than one-third of Asian ESL students are visa holders, compared to only 4 percent of Latino ESL students. Among ESL students, we find that half of Latinos are US citizens, but citizenship rates are much lower for other racial/ethnic groups. FIGURE 5 Half of Latino degree-seeking ESL students are US citizens Percent 100 7 90 80 25 70 6 60 50 40 38 30 20 10 23 0 Overall 14 4 29 50 Latino 5 34 3 41 15 Asian 2 12 9 40 31 Black 3 9 21 52 14 White Other/Unknown Student Visa holder (F1 or M1) Refugee/Asylee Temporal resident Permanent resident U.S. Citizen SOURCES: Authors’ calculation based on MIS data. NOTE: Sample includes students first enrolled in ESL in the CCC system between the 2009–10 and 2014–15 academic years. See technical appendix Tables B1 and B2 for more details. PPIC.ORG English as a Second Language in California’s Community Colleges 11 ESL students with either a GED or US high school diploma start three levels below transfer-level English, on average, whereas all other educational groups start four levels below (see technical appendix Table B2 for more detail). More than three-quarters of high school diploma holders are traditionally college-age; they make up the youngest group of ESL students. These varying characteristics can help us understand how different groups of students enter and progress through ESL sequences. ESL Progression to Transfer-level English Completing transfer-level English is one of the foundational general education requirements at California’s community colleges. For students intending to obtain an associate degree or transfer to a four-year university this requirement is a key milestone. Yet many CCC students fail to achieve this outcome, and ESL students are even less likely to do so. We estimate that only 36 percent of ESL degree-seeking students who first enrolled during the 2011–12 academic year successfully completed transfer-level English within six years (Table 1) compared to 48 percent among those who did not receive ESL instruction (CCCCO 2018a). While this number has improved over time, Figure 6 shows it is apparent that starting at a lower level of ESL coursework is correlated with lower completion rates of transfer-level English. TABLE 1 Successful completion of transfer-level English by year first enrolled in CCC system (%) Overall 2009–10 2010–11 2011–12 2012–13 2013–14 2014–15 3-year 26 22 23 25 27 28 30 6-year 34 33 34 36 n/a n/a n/a SOURCE: Author calculations from Chancellor’s Office MIS data and PPIC ESL program database. NOTES: Students who first enrolled between 2009–10 and 2014–15 are tracked for three years (N=120,365); Students who first enrolled between 2009–10 and 2011–12 are tracked for six years (N=64,527). See technical appendix Table A3 for cohort sample sizes. Figure 6 charts student progress through an ESL sequence depending on the level of entry. One thing is clear: as is the case in traditional developmental math and English sequences, the more levels students must traverse, the more exit points there are, and the less likely students are to complete a transfer-level course (Bailey, Jeong, and Cho 2010; Cuellar-Mejia et al. 2016). To illustrate this, we highlight two ESL sequence starting points: (1) two levels below transfer English, from which theoretically a student could complete transfer-level English in three terms or 1.5 years, and (2) five levels below transfer English, from which a student could theoretically complete a transfer-level course in the AB 705 timeframe of six terms or three years. We see that among the 13,212 students in our sample who started in ESL two levels below transfer-level English between the 2009–10 and 2011–12 academic years, 86 percent successfully completed this level, 62 percent subsequently took the next course in the sequence (one level below) and 56 percent successfully completed the ESL sequence. Ultimately, 56 percent of students starting in ESL two levels below enrolled in a transfer-level English course within six years and 51 percent successfully completed the course. Among the 8,509 students who started five levels below transfer we find that only 20 percent completed it within six years and only 9 percent achieved this outcome within the AB 705-mandated three-year timeframe.17 17 This statistic is presented for comparison purposes to align with the three-year AB 705 timeline. However, given length of ESL sequences our tables and figures focus on six-year completion of transfer-level English. Table with three-year completion of transfer-level English by ESL starting level is available upon request. PPIC.ORG English as a Second Language in California’s Community Colleges 12 FIGURE 6 Attrition is high in ESL sequences ESL Cohort (2010–2012): Six Year Progression in ESL Sequence Percent 100 100 100 90 85 80 86 70 69 60 60 50 Six Levels Below Transfer (6,981) 54 46 62 56 56 51 40 Five Levels Below Transfer (8,509) 30 Four Levels Below Transfer (12,702) 41 35 32 28 20 Three Levels Below (12,793) 23 20 Two Levels Below (13,212) 10 One-Level Below (5,122) 0 SOURCE: Author calculations from Chancellor’s Office MIS data and PPIC ESL program database. NOTES: Includes students who first enrolled during the 2009–10 and 2011–12 academic years in one through six levels below transfer (LBT) English. Numbers in parentheses indicate number of students starting at each level. See technical appendix Table B5 for underlying data for these and other levels below transfer. Figures for attrition by ESL sequence type (feeding directly into transfer-level English or developmental English) and characteristics (traditional, integrated, mixed, and accordion sequences) are available on request. There are a few points to highlight. First, the high first-course pass rates (above 80%) indicate that a student who enrolls in an ESL course is likely to successfully complete it. Second, many students who successfully complete a first course do not enroll in the next course in the sequence. This may not always be a negative outcome. As we track and interpret student progression though the ESL sequence, it is important to note that some colleges have designed their sequences or instituted departmental policies that enable students to skip levels based on achievement. For example, colleges might have ESL sequence structures intended to accelerate student progress, first-day-of-class writing assessments, or departmental rules that allow students to bypass developmental English.18 We can see that 12 percent of students who began two levels below transfer English and successfully completed that course actually skipped that level and moved directly to transfer-level English. However, a larger share (17%) of those who successfully completed ESL two levels below transfer failed to show up again in an ESL course or in transfer-level English. Our faculty interviews suggest that high attrition rates at upper levels may partly be driven by ESL sequence structures that require some ESL students to complete one or more levels of developmental English prior to enrolling in transfer-level English. Given the link between successful completion of transfer-level English and the level at which a degree-seeking ESL student starts, it seems important to look at differences across races/ethnicities and levels of prior education. 18 Several faculty we spoke to indicated that it is not uncommon for ESL faculty to give students an in-class writing assessment during the first weeks of class, which may result in students moving up levels. Additionally, one college shared that although their sequence officially leads to developmental English, up to two-thirds of its students are cleared move directly to transfer-level English based on demonstrated achievement. PPIC.ORG English as a Second Language in California’s Community Colleges 13 At almost every level of ESL entry, Asian degree-seeking students are the most likely and Latino students the least likely to complete a transfer-level course (Figure 7). Recall that black and white students start four levels below transfer-level English on average, while Asian and Latino students start three levels below. We find that white students starting four levels below are more likely than Latino students starting three levels below to complete transfer-level English (30% versus 24%). Focusing on racial/ethnic differences among students starting two levels below (the largest category), we find even more dramatic variation in completion rates at the extremes: 62 percent for Asian ESL students and 33 percent for Latino ESL students. FIGURE 7 Equity gaps persist in completion for students who enter ESL sequences at all levels 70 62 60 % completing transfer-level English from each starting level 50 40 33 30 20 14 10 45 42 36 27 17 15 15 1 LBT 2 LBT 3 LBT 4 LBT 5 LBT 6 LBT 0 Latino Asian Black Two or More White SOURCE: Author calculations from Chancellor’s Office MIS data and PPIC ESL program database. NOTES: Sample includes degree-seeking students who first enrolled in ESL at a California Community College between the 2009–10 and 2011–12 academic years and are tracked for six years, through the 2016–17 academic year. LBT refers to level below transfer English. In sum, black students, a relatively small percentage of all degree-seeking ESL students, start at low levels and have lower throughput rates at every level. Latino students start at high levels (a greater proportion have US high school diplomas) but have the lowest completion rates. Analyses of degree-seeking ESL students by prior education suggest that foreign high school graduates have the highest completion rates no matter where they start in the sequence, with one notable exception: adult school students who start one level below transfer-level English have the highest completion rates.19 Students who report a US high school diploma or a GED have roughly equal transfer-level English completion rates across starting levels. These findings are described in more detail in technical appendix Figure B2. This descriptive examination of student progression makes it clear that the overwhelming majority of degreeintending ESL students are not reaching a key milestone needed to achieve their educational goals, even after six years. It also reveals important differences across races/ethnicities and prior education levels. Ultimately the high attrition in ESL sequences highlights the importance of redesigning sequences to meet AB 705 goals. Early Reform Efforts in ESL Programs Given system-wide information on relatively low transition rates of ESL students to transfer-level English, as well as the evolution of curricular models for teaching ESL, a number of ESL programs have been adapting their course models in recent years (CCCCO 2018b; Bunch 2008; Henson and Hern 2018; Hodara 2015; Park 2018; 19 It should be noted that this group comprises a small sample (N=61). PPIC.ORG English as a Second Language in California’s Community Colleges 14 Rodriguez 2013). Some colleges have shortened or compressed their ESL pathways to transfer-level English (technical appendix Figure B1). However, most ESL reforms have focused on how English skills are taught. Since 2010, ESL sequence courses that teach integrated language skills have overtaken those that teach skills in the traditional, standalone manner. In academic year 2009–10, roughly 31,000 students enrolled in traditional sequence courses;20 by 2016–17, that number had fallen to around 22,000 (Figure 8). Over that same period, 81 colleges included traditional courses in their ESL sequences and 91 offered integrated courses. ESL sequences leading to transfer-level English often include a mix of traditional and integrated courses; in recent years, students have increasingly been choosing integrated courses over stand-alone offerings. FIGURE 8 Overall ESL enrollment is holding steady, but traditional ESL enrollment is declining 70,000 Students enrolled in ESL sequence courses 60,000 50,000 40,000 30,000 20,000 TrRaAdiDtiIoTnIaOlNAL InNtTeEgrGatReAdTED ToOtaTlAL 10,000 0 2010 2011 2012 2013 2014 2015 2016 2017 SOURCE: Author calculations from Chancellor’s Office MIS data and PPIC ESL program database. NOTES: Enrollment for students who enrolled in ESL sequence courses over time where 2010 refers to 2009–10, and so on. Students are counted once in a given academic year, according to the type of ESL course(s) he or she enrolled in. If the student enrolled in both traditional and integrated coursework in a given academic year, they are counted in both lines. Otherwise students are unduplicated within college but may be duplicated across colleges if students enrolled in multiple colleges at any point in their college career. See Table B7 in Appendix B for raw numbers. ESL courses have traditionally taken a discrete skills approach for a variety of reasons. One faculty member told us that discrete skills were taught at her college partly because faculty were accustomed to discrete skill structures or because they specialized in particular skills. Some faculty believed that students needed to learn skills separately to be successful in academic English settings. For instance, one faculty member said that having separate reading and writing courses is important because strategies that reading instructors feel are critical for success might not be covered thoroughly in an integrated reading and writing course. These comments illustrate the challenges of integrating ESL skills and reinforce the notion that true integration requires more than simply combining discrete skills into a single, longer course (Bickerstaff and Raufman 2017). They also underscore the importance of professional development to help ESL faculty move toward an integrated approach. However, faculty who transitioned from discrete to integrated skills courses believed that some degree of integration had already been happening. Given that reading and grammar is necessary for academic writing, it 20 It is important to note that even when a college does offer a traditional ESL sequence with separate writing, reading, grammar and listening/speaking courses, it may require students to take only the writing sequence to advance to college English—other skill areas are often offered as electives. PPIC.ORG English as a Second Language in California’s Community Colleges 15 was natural to embed grammar exercises into a writing course—for example, by identifying error patterns and providing “just-in-time” support. Similarly, students in integrated courses read a variety of material—including novels, short stories, and articles sometimes based on a theme—to support their writing and develop their vocabulary. Integrated courses were often described as being designed with the transfer-level English course objectives in mind. Some faculty also indicated that they used the California Acceleration Project’s (CAP’s) principles of design in integrated courses—which includes “just in time” interventions, low-stakes collaborative practice, group work, and an emphasis on addressing students’ affective needs. One faculty member noted that the traditional approach made it difficult to evaluate student preparedness in English because skills were taught in separate courses and not all skills were required. Integration helps address this problem. On the logistical front, faculty indicated that enrollment difficulties and faculty availability were motivations for integrating courses—it was not uncommon for courses in some skill areas to be cancelled due to low enrollment, leaving students stranded and faculty without courses to teach. Integrated ESL sequences are now the dominant approach across California community colleges (Figure 9). In addition, a handful of colleges have implemented course sequences that allow students to make faster progress toward transfer-level English. Only 3 percent of colleges have adopted reforms that shorten sequences by compressing courses and only 4 percent allow students who make significant progress to skip ahead. Both approaches eliminate exit points in the sequence and allow students to progress at a faster rate. However, because only a few colleges and a few students have adopted these reforms, it is not possible to assess their impact on student outcomes.21 FIGURE 9 The integrated sequence has become the most common type 100 90 80 Percent of colleges 70 60 50 44 40 29 30 20 21 10 4 3 0 Traditional Integrated Mixed ESL sequence type Accordion Compressed SOURCE: PPIC ESL program database. NOTES: Sequence structures of colleges offering ESL sequences as of the 2016–2017 academic year. Our study includes a total of 105 colleges. The nine colleges not included in our study did not have readily available or comprehensible information about their respective ESL sequences, and/or could not be reached for more information. In the timeframe of our analytical sample, the total number of colleges in each category fluctuates, as some colleges established ESL programs in different years, and colleges have changed their structure and sequence types between 2010 and 2017. See Technical Appendix A Glossary for more details. 21 Given the newness of accordion and compression models, we are unable to track student outcomes for six years. However, emerging evidence from colleges suggests that these approaches can work in ESL. Faculty at Cuyamaca College shared that after compressing the top levels and incorporating an accordion model the ESL sequence was reduced from seven levels to five; this resulted in all students “going faster”—about one-third of students completed transfer-level English in four terms compared to fewer than 10 percent under the old model. Future research should assess the efficacy of these models. PPIC.ORG English as a Second Language in California’s Community Colleges 16 How Do ESL Reforms Affect Student Achievement? Our analysis of student course-taking data and academic outcomes covers the period prior to the passage and implementation of AB 705. However, as we have seen, a number of colleges were either attempting reforms or already had structures similar to those recommended in the AB 705 guidance (including integrated courses, direct pathways to transfer English, and transferable ESL credits). In this section, we use statistical analyses to assess the relationship between some of the reformed ESL course features and academic outcomes for degree-seeking ESL students. Our analysis focuses on four primary outcomes: completion of transfer-level English, transferable credits awarded, community college degree completion, and transfer to a four-year college. This part of the analysis focuses on students whose outcomes of interest we could track for three or six years. To align with the AB 705 timeframe, we track completing transfer-level English (C or better) within three and six years; given the lengthy ESL sequences, however, we limit tracking subsequent outcomes—accruing transfer credits and obtaining a degree or transferring—to six years. Assessing Individual ESL Sequence Features To help inform AB 705 reforms, we now examine the role of different ESL sequence features in promoting improved student outcomes. We focus on three features: (1) integrated sequences, which teach two or more language skills per class; (2) sequences that lead directly to transfer-level English; and (3) transferable ESL courses, which award credits that transfer to a CSU and/or UC school.22 In this phase of the research, we used statistical models to control for differences in student, college and cohort characteristics that may also influence student outcomes (see Technical Appendix A for more detail). After controlling for student characteristics and college and cohort effects, including the level at which a student enters the ESL sequence, we find evidence that each of these ESL sequence features helps students progress, and that some help more than others. Holding all else constant, we find that taking a transferable ESL course has the biggest impact on a student’s likelihood of completing transfer-level English within three and six years: between 16 and 20 percentage points (Figure 10). Students taking integrated courses are 11 to 12 percentage points more likely to complete transfer-level English within three and six years, respectively. A sequence that leads directly to transfer-level English is also associated with a higher likelihood of completion of that course within three and six years (between 7 and 12 percentage points). While our research finds evidence to support what we saw in the ESL sequence progression charts (Figure 6)— that the lower the level at which students enter, the less likely they are to complete transfer-level English—we also find that students starting at the lowest levels of ESL benefit less from integration and from sequences that lead directly to transfer-level English compared to those who start at the top two levels. Interestingly, students starting at levels three or four appear to benefit more from transferable ESL in a six-year timeframe, compared to their peers who begin in the top two levels (see technical appendix Tables B11–12). These findings suggest that students who start at different levels have varying experiences with sequence features and that more attention should be given to their differing needs. Our research also found some evidence that ESL sequence structures support longer-term student success—that is, the proportion of transferable credits obtained, degree completion, and transfer to a four-year college within six 22 See technical appendix Table A5 for more information on the how these different sequence features overlap (i.e., 55% of integrated sequences feed directly into transfer-level English). PPIC.ORG English as a Second Language in California’s Community Colleges 17 years. All else being equal, as would be expected, we find that taking a transferable ESL course increases the proportion of transfer FIGURE 10 ESL sequence features increase the likelihood of completing transfer-level English Transferable ESL Integration 11.2 11.8 16.1 20.4 6 Years 3 Years Direct to TLE 11.5 7.4 0 5 10 15 20 25 Change in likelihood of completing transfer-level English (percentage point) SOURCE: Author calculations from Chancellor’s Office MIS data and PPIC ESL program database. Full results available in technical appendix Tables B11 and B12. NOTES: Results from individual-level regression models. Our sample includes degree-seeking first-time college students entering one of California’s community colleges between 2009–10 and 2014–15 who are tracked through the 2016–17 academic year. All results are statistically significant at the 90% level or above. credits earned by 9 percentage points, and that students entering three to four levels below the sequence benefit more from taking a transferable ESL course than their peers who enter in the top two levels. We find that the only ESL sequence feature that is associated with obtaining a community college degree is a transferable ESL course— which improves the likelihood of earning a degree by almost 13 percentage points. Integration was also found to be beneficial, but it was only associated with the share of transferable credits awarded (4 percentage points more than traditional ESL courses). Finally, we find little evidence that ESL sequence features improve a student’s likelihood of transferring after six years; taking a transferable course improves this likelihood marginally (by 3 percentage points). Across the board, we find that students entering at the lowest levels of ESL are much less likely to have positive academic outcomes over the six-year timeframe (see technical appendix Tables B13–B15). Our ESL faculty interviews shed light on our findings. They indicate that transferable ESL courses consistently emerge as a strong correlate of student outcomes because these courses “hold students to high a standard.” Faculty we spoke to confirmed that offering transferable ESL courses as part of the sequence served to emphasize their rigor as university-level foreign language work. The opportunity to earn transferable credits while they are learning English allows students to advance toward their degree or transfer goals—and this, in turn, probably boosts aspirations, especially among those who begin in lower levels. We also learned that integration supports student success partly due to the fact that in an integrated course, students are more likely to receive “just in time” help with skills that are necessary to progress through the sequence (e.g., grammar and reading comprehension associated with a given writing assignment). We also learned that integrated sequences that are designed with transfer-level English course objectives in mind and taught with a theme-based approach that aligns reading and writing assignments can be especially helpful in facilitating student success. These design features mimic the type of reading and writing that students do in transfer-level English and other content courses outside of ESL. By contrast, in a traditional sequence, all of these skills would be addressed separately, so that the material in a reading course does not always align with what students are learning in the writing course. PPIC.ORG English as a Second Language in California’s Community Colleges 18 Do ESL Sequence Reforms Help All Student Groups? Our descriptive analyses have already demonstrated differences in outcomes across groups of degree-seeking ESL students. In particular, we see lower sequence completion rates for Latino and black students. Here, we examine whether there are statistically significant differences in the impact of ESL sequence features on the likelihood of degree-seeking students across racial/ethnic groups completing sequences, controlling for student characteristics (including a student’s level of entry), college characteristics, and cohort effects. Taking transferable ESL courses is associated with higher rates of throughput within three and six years for white, Latino, and Asian students (Figure 11). Such course-taking is associated with 13 to 19 percentage point increases in three-year throughput and 17 to 25 percentage point increases in six-year throughput. Taking a transferable ESL course does not appear to be associated with higher throughput for black students. FIGURE 11 ESL sequence features increase the likelihood of completing transfer-level English, but varies by race/ethnicity Three-Year Throughput Transferable ESL Feed Direct TLE Integration 0 12.7 18.8 16.3 7.1 15.1 11.9 7.3 12.5 5 10 15 20.8 20 25 Black White Latino Asian Six-Year Throughput 32 30 35 Transferable ESL Feed Direct TLE 11.5 16.8 20.5 18.4 24.9 36.1 Integration 0 5.9 15.2 5 10 15 20 25 30 35 40 Black White Latino Asian SOURCE: Author calculations from Chancellor's Office MIS data and PPIC ESL program database. Full results available in technical appendix Tables B16 and B17. NOTES: Results from individual-level regression models, estimated separately by race. Our sample includes degree-seeking first-time college students entering one of California’s community colleges between 2009–10 and 2014–15 who are tracked through the 2016–17 academic year. Only statistically significant results displayed. While our sample size for black students is small, our coefficients are estimated with sufficient precision to be certain of differences between black ESL and other ESL students on many measures. Only statistically significant results at the 90% level or above are shown. We also test for statistically significant differences between race/ethnic groups. In three-year throughput, all differences between black and other racial/ethnic groups reported are statistically significant, with the exception of the black/white difference in the association between integration and throughput. For six-year throughput, all differences between black and other racial/ethnic groups and between Asian and other racial/ethnic groups were statistically significant for feeding directly into transfer-level English. Racial/ethnic group differences were not statistically significant in six-year throughput for transferable ESL or integration. See technical appendix Table B20 for a summary of the tests for between group statistical significance. PPIC.ORG English as a Second Language in California’s Community Colleges 19 We find that Asian students appear to benefit from integrated ESL courses: they are 13 percentage points more likely to complete transfer-level English within three years and 15 percentages points more likely to do so within six years if they have ever enrolled in an integrated course. Latino ESL students also appear to benefit from integrated course structures, but the effect is smaller (7 percentage points in the three-year timeframe and 6 percentage points in the six-year time horizon). Integrated ESL course structure also improves successful completion within three years for white and black students, increasing throughput by 12 and 21 percentage points, respectively. Integration is the only ESL feature that appears to increase throughput for all racial/ethnic groups. Being at a college where the ESL pathway feeds directly into transfer-level English appears to be particularly beneficial to black students, who are 32 and 36 percentage points more likely to reach transfer English within three and six years, respectively, than their peers in colleges where ESL feeds into developmental English. We also find a statistically significant association for white students—an 18 percentage point increase in six-year completion. Black and white students start sequences at lower levels, on average, than do Asian or Latino students. While Latino students typically start at higher levels than black and white students, they also have higher throughput rates at colleges where ESL feeds into transfer-level English (a 12 percentage point increase in six-year completion and a 15 percentage point increase in three- year completion). It is also important to consider how a number of successful longer-term academic outcomes across racial/ethnic groups are related to ESL course structure, controlling for all else. In particular, we examine the association between the percentages of attempted transfer credits earned, transfers to a four-year college, and degree attainment within six years and our three ESL design features: integrated courses, sequences that feed directly into transfer-level English, and transferable courses. We find that ESL sequences ending in transfer-level English are most beneficial to Latino students when it comes to percentage of attempted credits completed (technical appendix Table B18).23 For Latino students, these sequences are associated with obtaining community college degrees (technical appendix Table B19).24 These sequences do not appear to be associated with transfers to four-year colleges for any racial/ethnic group (technical appendix Table B20). As with successful completion of an ESL sequences, there is no positive association for Asian ESL students between sequences that feed into transfer-level English and longer-term outcomes. In general we do not find strong evidence that integrated sequences are particularly helpful when it comes to successful longer-term outcomes, with just a couple of exceptions: Latino and black students at colleges with integrated sequences are more likely to obtain a community college degree (4 and 10 percentage points, respectively).25 It is not surprising that transferable ESL is associated with the percentage of attempted transfer credits earned. However, this feature seems to be especially helpful for black students, who experience a 17 percentage point increase in credits earned, compared to the other racial/ethnic groups (whose increases range from 8 to 12 percentage points), although differences between black and other ESL students are not statistically significant. And while we did not find that transferable ESL is linked to black ESL students completing transfer-level English, it is associated with helping black students earn more college credits. In addition, taking transferable ESL is positively associated with completing a community college degree for all race/ethnic groups, but particularly for Asian and white ESL students.26 Transferable ESL credits appear associated with modest increases in transferring to four- year colleges for Latino and white students but not for Asian or black students. 23 Differences between Latinos and whites and Latinos and Asians are statistically significant. 24 Differences between Latinos and Asians are statistically significant. 25 Differences between black and white ESL students are statistically significant. 26 Differences between Asian and Latino and Asian and white students are statistically significant. PPIC.ORG English as a Second Language in California’s Community Colleges 20 In short, our analysis suggests that integration increases the odds of all groups of students completing transferlevel English and improves several longer-term outcomes. As they did with our overall findings, faculty interviews suggest that these effects are likely driven by the key principals of integration itself—in particular, content designed to lead students toward the goal of transfer-level English; the ability to provide “just in time” support on skills that students need to progress through the sequence (e.g., helping students with grammar, in in the context of a writing assignment); and a thematic approach that is relevant to students’ lives (e.g., social justice). Interestingly, we find that a direct transition from ESL to transfer-level English has differential impacts across student groups: Latino and black students benefit more from going directly into transfer-level English, while Asian students at colleges that offered an ESL sequence that led directly to transfer English earned a smaller share of attempted credits. Faculty we spoke to indicated that an ESL sequence leading to transfer-level English provides language development support throughout a student’s pathway to transfer English. Still, it appears that Asian students may benefit from transitioning to the English department sooner via an ESL sequence that leads to developmental English. Finally, transferable ESL credit was important for a variety of student subgroups. This suggests that the opportunity to take more rigorous ESL courses and be considered foreign language learners may be raising educational aspirations among traditionally underserved groups. ESL Reforms in Response to AB 705 We have shown that few degree-seeking students in traditional ESL programs achieve key educational milestones, but that integrated ESL courses, direct pathways from ESL to transfer-level English, and transferable ESL coursework are associated with an improved likelihood of achieving key academic outcomes. Many colleges are already implementing these structures, but many others are not—and, until recently, few questioned the assessment and placement system that assigned students to an ESL sequence. AB 705 aims to boost outcomes for ESL students by requiring all colleges to rethink their assessment and placement systems and traditional ESL course structures. Interviews with ESL department chairs, faculty, and administrators at 13 colleges across the CCC yielded insights into the changes colleges are making in response to AB 705. In this section, we highlight the most salient themes. Changes to ESL Assessment and Placement In earlier research on assessment and placement in California’s community colleges, we found that all colleges reported using standardized placement tests in ESL during the 2015–16 academic year. We also found that the use of other measures was very limited. The most common multiple measure, used at 24 percent of colleges, was an instructor/counselor recommendation (Rodriguez et al. 2016). A recent AB 705 implementation survey confirmed that the majority of colleges continue to use standardized tests to assess and place students in ESL, even those with four years of high school data (Multiple Measures Assessment Project 2019).27 The inclusion of ESL programs in AB 705 has completely upended this status quo—indeed, one ESL faculty member described the law as “the biggest change that has happened at the community colleges.” 27 This is partly due to the fact that a high share of students in ESL have not graduated from a US high school. In our sample of degree-seeking students, only one-third have a US high school diploma. This small share is not surprising: a recent estimate using Cal-PASS plus data finds that 87 percent of ESL students who graduated from a California high school and enrolled in community college did not take ESL; instead, they enrolled in English for native speakers (Willett 2017). PPIC.ORG English as a Second Language in California’s Community Colleges 21 Starting in fall 2020, colleges must use “evidence-based multiple measures” for placement into ESL coursework (California Legislative Information 2017; CCCCO 2018c, 2018d).28 High school records are the default evidencebased measure. If high school records are unavailable, the law considers guided self-placement (GSP) to be a viable method. The AB 705 ESL workgroup recently provided colleges with guidance on evidence-based rules for placing students into credit ESL (Perez and Stanskas 2018). To begin, the workgroup outlined default placement rules for ESL students with four years of high school records (Table 2). The default placement guidelines mirror the recommendations for English placement—i.e., that all students can be placed in transfer-level English, and that students with lower GPAs should receive concurrent support. Importantly, the workgroup clearly states that this does not mean that all recent high school graduates will be ready for transfer-level English—some students may still need to enroll in ESL to acquire academic language proficiency. Additional guidance is expected on placement within the ESL sequence, especially for students with fewer than four years of high school records (e.g., older immigrants, refugees, and foreign students). Still, several faculty voiced concern about the default placement rules making it difficult for students to take ESL—as one faculty member noted, choosing ESL over a co-requisite could require extra time and money. TABLE 2 Chancellor’s Office AB 705 default placement rules for ESL High school performance metric Recommended AB 705 placement Throughput for students enrolling directly in transfer-level English GPA >=2.6 Transfer-level English composition No additional academic and concurrent support required 78.6% GPA 1.9to 2.6 Transfer-level English composition Additional academic and concurrent support recommended GPA <1.9 Transfer-level English composition Additional academic and concurrent support strongly recommended SOURCE: Perez and Stanskas 2018. 57.7% 42.6% As of fall 2018, 10 of the 13 colleges we spoke to had begun to plan for or had already implemented changes in response to AB 705. Changes to assessment and placement were partly driven by the elimination of standardized placement tests in English as a result of AB 705—some colleges use the English assessment for ESL, and others have been using ESL placement tests that they now think they will need to eliminate. New guidance issued in April 2019 clarified that AB 705 does not require that colleges eliminate placement tests for ESL at this time— noting that some tests can still be used at least through July 2021 (Perez 2019).The colleges that did not report making changes indicated that they were waiting for more guidance from the Chancellor’s Office. It is anticipated that as new guidance and information is released, an increasing number of colleges will be implementing changes in response to AB 705. A recent AB 705 guidance memo indicated that all colleges will be asked to submit AB 705 adoption plans for ESL by July 2020 (Perez 2019). Below we summarize the most commonly described changes to ESL assessment and placement policies in response to AB 705. 28 In December 2018, the CCCCO and the Academic Senate published a frequently asked questions document in which they clarify that the three-year timeline for AB 705 “begins at the point that a student enrolled in a credit ESL sequence declares intent for degree or transfer as determined by their educational goal, education plan, and/or major declaration” (CCCCO, 2018c). In math and English, the one-year timeline begins when a student first enrolls in a math or English course—the assumption is that all students who enroll in math or English courses intend to pursue a degree or transfer. PPIC.ORG English as a Second Language in California’s Community Colleges 22 Guided Self-placement In our interviews with ESL faculty we frequently heard that colleges would be adopting a guided self-placement (GSP) approach in addition to or instead of a standardized placement test.29 As part of a GSP process, a college provides students with course information that is intended to help them evaluate their readiness for a given course level. This information includes course descriptions, sample assignments, projects, and exams. The GSP process is meant to help students select courses that align with their educational goals and abilities (CCC Academic Senate 2018). GSP is an attractive placement option for a large group of ESL students (e.g., older adults, refugees, or those who went to foreign high schools) who may not have high school records from a US high school. ESL faculty described a typical GSP process as providing students with reading and/or writing samples for different levels of ESL and allowing them to choose the example that aligns most closely with their reading/writing abilities. This in turn helps students select the most appropriate ESL placement level. Some examples of the guiding questions in the GSP process are “Which of these writings is most like the one you would do? What level do you think you should be in?” or “Do I understand the reading? Can I write this well?” Changes to assessment and placement at Cuyamaca College Prior to fall 2018, Cuyamaca College primarily relied on standardized placement tests and writing samples. For instance, applicants would be asked a few questions to determine whether they need to be referred to ESL (e.g., is English used in the home, is English your first language, etc.), then they would take the Combined English Language Skills Assessment (CELSA) and respond to a writing prompt that was graded by faculty. Because this approach had low predictive value, the college completely transformed the way it assessed and placed students into ESL. The college has eliminated the placement test and instead uses the “best of” three measures: Self-reported high school grade point average—which use the MMAP rule set for placement into the different levels of ESL (MMAP 2017). The ESL placement survey developed by Irvine Valley College. Guided self-placement: students are given writing samples from different levels of ESL and asked to pick the level that best matches the type of writing they would do. The department chair believes the new assessment and placement approach has been working well. Students are good at placing themselves and appear to be more satisfied with this approach. It is important to highlight that in addition to the changes described above, Cuyamaca College has also transformed their ESL sequence over the past several years by incorporating acceleration strategies. In making these changes to the course structures the college reduced the amount of time students spend in ESL from 3.5 to 2 years. After the changes to the ESL sequence and assessment and placement, the college believes it is ahead of the game and in compliance with the law. In fact, the ESL department chair believes they are “grinning about AB 705” – noting that the policy “caught up to us” not the other way around. Multiple Measures Faculty also reported that their colleges are moving toward multiple measures in ESL. Several reported using a multiple measures ESL survey that was developed by Irvine Valley College (IVC). The IVC survey gathers 29 A recent AB 705 implementation survey conducted by the Multiple Measures Assessment Project (MMAP) finds that of 104 colleges that responded, 13 are currently using GSP and 44 plan to use it in the future (Multiple Measures Assessment Project 2019). PPIC.ORG English as a Second Language in California’s Community Colleges 23 information about a student’s academic attainment, self-evaluated English proficiency, and use of English in everyday life (e.g., phone language settings). Researchers at IVC have begun to assess survey data from other colleges to assess the survey’s predictive utility. A couple of colleges reported using the Multiple Measures Assessment Project rules to determine ESL placements. For example, Cuyamaca College has used self-reported high school GPA as part of a disjunctive, or “best of,” placement process that also included the IVC survey and a guided self-placement tool (see text box). Changes to ESL Course Structure Ten of the thirteen colleges we spoke to were in the process or had already made changes to their ESL course structures.30 Changes to course structures are primarily driven by AB 705’s mandate that colleges maximize the probability that ESL students complete transfer English within three years. The guidance provided by the Chancellor’s Office also provided a series of curricular recommendations for colleges as they prepare for the fall 2020 compliance deadline. Colleges are advised to:  integrate the teaching of language skills  develop direct pathways from ESL to transfer-level English  design ESL courses from the vantage point of transfer-level English  allow credit ESL faculty to teach transfer-level English or create an ESL equivalent  increase professional development opportunities for credit and transfer-level English faculty  seek CSU humanities transfer credit for advanced transfer-level ESL courses Several colleges we spoke to said that their college already had acted on transferable ESL, designing sequences from the vantage point of transfer-level English, sequences that were five levels or less, and direct pathways to transfer-level English. During our interviews we heard that colleges were planning to change their ESL course structures in other ways. Below we summarize what we learned. Non-credit ESL Non-credit ESL, which was not included in the Chancellor’s Office recommendations, was being considered or implemented at six colleges. Colleges described using non-credit ESL in two ways: (1) to replace the lowest levels of credit ESL and (2) to mirror the credit ESL sequence. Both approaches focus on students who need more than three years to reach transfer-level English. As one faculty member noted, the three-year timeframe could drive some students toward the non-credit pathway—especially those at lower levels. Several faculty at colleges with lengthy sequences spoke about turning the lowest levels into non-credit ESL. Others spoke about developing non-credit courses that mirror the credit ESL sequence—so that all students could benefit from the same ESL course whether it offers credit or not. Faculty shared that in addition to being outside a three-year timeframe, noncredit courses are free and repeatable. Several faculty also noted that a non-credit option would be helpful for students who do not intend to obtain a degree or transfer to a four-year school. ESL Sequences that Lead Directly to Transfer-level English Faculty at colleges with ESL sequences that feed into developmental English emphasized the urgent need for ESL sequences that lead directly to transfer-level English.31 This need is driven by AB 705, which is prompting colleges to replace developmental English with co-requisite remediation, and which has made a clear distinction 30 The MMAP survey of AB 705 implementation finds that 13 campuses are currently piloting curricular changes to credit ESL in 2018–19; 27 plan to pilot curricular changes in 2019–20. 31 The MMAP (2019) survey found that eight colleges are developing a credit pathway from ESL to transfer-level English or an ESL version of this course. PPIC.ORG English as a Second Language in California’s Community Colleges 24 between ESL—a program that involves advanced foreign-language learning—and developmental English. ESL faculty spoke about developing additional ESL courses that would parallel developmental English. Our research suggests this will reduce attrition among ESL students. One faculty member noted that it was not uncommon for students to do well ESL but fail to progress in developmental English. One possible explanation is that because developmental English is taught from “an aspect of teaching English speakers,” ESL errors are penalized and there is no language support. Accelerated ESL Completion Several colleges indicated that they were planning to accelerate ESL sequences by compressing, integrating, using an accordion model, or having faculty certified in both ESL and freshman composition teach transfer-level English and a linked co-requisite course.32 Some of these colleges began their acceleration efforts prior to AB 705; at other colleges, the policy created a heightened sense of urgency. The primary motivation for accelerating ESL pathways was data showing that few students were completing transfer-level English. At two colleges that had been offering seven-level sequences, fewer than 10 percent of students were making it to transfer-level English. ESL faculty members stressed that these outcomes were “horrible” and “not good enough.” Irvine Valley College has or is in the process of implementing several reforms (see text box). Interestingly, two colleges implemented acceleration strategies that did not work for them. One college introduced an accordion model and then realized that most students transitioned from “A to A,” so there was little need for a “B” level. Moreover, the accordion model had “always been confusing for data collection purposes” because the system considers it to be twice as long. Starting fall 2019, this college decided to allow students who need additional support—which would have otherwise been provided in a “B” level course—to take a non-credit version of the credit ESL course—the hope is that adopting the mirrored sequence will result in faculty electing to drop the “B” level course. Another college introduced a co-requisite but quickly learned that having English faculty teach the transfer-level English course and ESL faculty teach the co-requisite was not going to work. Having different instructors was “logistically impossible” because students could go through five sections of freshman composition with five different sets of readings and assignments. It is important to highlight that some colleges are not considering reducing the number of ESL levels even though their sequences include six or more levels. Faculty cited the need to provide longer sequences because they serve student populations with very low English skills—and because they believe that many are only there to learn English, not to pursue a degree or transfer. They noted that AB 705 does not outlaw sequences with more than five levels; it prohibits the college from placing students any lower than level five, but students could place themselves at level six or lower. Other Innovative Changes in ESL Colleges we spoke to are exploring ways to support the progression of ESL students more broadly in college and beyond. 32 On acceleration front, the MMAP (2019) survey found that eight colleges are working on creating an ESL program that lasts three semesters. A similar number of colleges identified plans to pilot concurrent support models in ESL. PPIC.ORG English as a Second Language in California’s Community Colleges 25 Changes to ESL course structures at Irvine Valley College The ESL transformation at Irvine Valley College (IVC) exemplifies several of the approaches described above: IVC implemented an accordion sequence, created a direct pathway to transfer-level English, designed a noncredit adult ESL program, and expects to soon offer a co-requisite paired with transfer-level English taught by ESL instructors certified in composition. IVC launched its reform effort in 2012 in response to low throughput percentages, particularly at the transition point from ESL to developmental English. In 2014, IVC redesigned their sequence so that it would lead directly to transfer-level English. In doing so it designed the top two levels of ESL writing/reading courses to mirror the developmental writing sequence, and brought the curriculum in line with California Acceleration Project (CAP) design principles. However, these changes did not improve throughput. IVC staff believed that one of the reasons for this was that many ESL students did not want to pursue a degree or transfer—after surveying their students they found only about half were pursuing these goals. In 2016 they implemented two pathways: academic ESL (credit) and adult ESL (non-credit). The credit ESL path also includes an accordion model, which has been in place for many years. With the implementation of AB 705, IVC recognized that it can no longer offer six or seven levels of credit ESL courses. It plans to offer only the top four levels in credit ESL; lower-level courses will be part of the non-credit pathway, which also includes intermediate and advanced courses to meet the needs of the noncredit student population. Adult ESL students who intend to transfer or pursue a degree will be able to transition to the credit ESL sequence by enrolling in the first level of academic writing or higher, depending on their ability and needs. Additionally, starting in fall 2019, the college plans to offer a co-requisite linked to transfer-level English that will be taught by an ESL instructor certified to teach freshman composition. The co-requisite will provide the language support needed to achieve the objectives of transfer-level English. In addition to changing its ESL course structures, the college has been transforming its ESL assessment and placement system. In fall 2019, the college will replace the standardized placement test with a new in-house survey, while continuing to use a writing sample. In addition, credit ESL students will be able to use the English guided self-placement tool. ESL Certificates Recognizing the benefits of aligning its student success efforts with two major statewide reforms—AB 705 and Guided Pathways—Cypress College has created a Guided Pathways ESL Milestone certificate of achievement. The certificate recognizes that by successfully completing the highest levels of ESL—which currently transfer to a CSU and UC as elective units and fulfill Category C2 Humanities requirements for the local degree—and completing an introductory or core course in a meta major pathway, a student has demonstrated academic English reading and writing skills at an advanced, postsecondary level and achieved success in a pathway toward a degree, certificate, or general education transfer. The college has developed the certificates in various programs and plans to begin awarding them in fall 2019 (see text box). Humanities Credits Several faculty we spoke to were considering or had already submitted advanced ESL courses that offer not just elective credit but humanities credit. The goal is to have advanced ESL coursework fulfill CSU C2/IGETC 3B Humanities requirements. One faculty member emphasized that this is an equity issue: ESL students should PPIC.ORG English as a Second Language in California’s Community Colleges 26 receive credit for being foreign language learners able to engage in rich cultural content inherent to ESL courses. Several faculty indicated that the language competencies in top-level ESL courses are at a higher level than those in Spanish 1. If this change were to happen, faculty argued that students could potentially save up to 15 units of coursework. This would align with the AB 705 goal for that students to make quicker progress toward degree and transfer without excess credits. Our findings suggest this change could further boost the impact of transferable ESL courses on student outcomes, and could be especially helpful for students entering ESL sequences at lower levels. Guided Pathways ESL Milestone Certificates at Cypress College Starting in fall 2019, Cypress College will award certificates to students who master advanced transfer-level academic English skills and complete at least one course in a Guided Pathways meta major. The idea is that these students have achieved academic English reading and writing skills at an advanced postsecondary level and taken prerequisite or introductory coursework in a degree, certificate, or general education pathway. Most certificate programs recommend but do not require that students take an educational planning elective. To determine which meta majors to prioritize for the 2018–19 credit ESL Milestone certificates, institutional researchers gathered data on the top awards and/or certificates earned by ESL students at the college. This rich collaboration between research and faculty produced a list of pathways:  Business and Computer Information Systems (CIS)  Career Technical Education/Aviation and Travel Careers  Dental Hygiene/Nursing/Psychiatric Technology  Diagnostic Medical Sonography, Health Information Technology, Radiologic Technology  Kinesiology  Language Arts/Oral Communication  Language Arts/Written Communication  Science Engineering and Math (SEM)  Social Sciences For example, to earn the GP ESL Milestone certificate for completing the Science Engineering and Math (SEM) pathway (the most commonly completed pathway), students would need 10 units of the top two levels of ESL (ESL 185 and 186) 4 units of College Algebra; and 4–5 units of Preparation for General Chemistry, General Chemistry I, or Elementary Physics. Students would also be encouraged to take an educational planning course (COUN 140) for 1 unit. As of April 2019, the certificate program had been approved by the college, district, and Chancellor’s office. Conclusion and Recommendations One of the most striking findings in this report is the simple fact that a large share of degree-intending students drop off the ESL pathway to transfer-level English: only one-third of all degree-seeking ESL students who first enrolled between 2010 and 2012 successfully completed transfer-level English within six years. Moreover, this share includes a high of 56 percent of those who started out one level below transfer-level English to just 9 percent of those who begin eight levels below. And while nearly half of colleges offer no more than five levels of ESL, which PPIC.ORG English as a Second Language in California’s Community Colleges 27 would theoretically allow students to complete transfer-level ESL in six terms, only 20 percent of students starting at this level complete transfer-level English within six years and even fewer did so within three years. Many community colleges have already begun structuring their ESL sequences in ways that are aligned with the AB 705 implementation guidelines provided by the Chancellor’s Office and Academic Senate (Perez and Stanskas 2018). In particular, colleges have made great strides in integrating the teaching of multiple language skills, structuring sequences to lead directly to transfer-level English, incorporating transferable ESL coursework into the sequence, and adopting other strategies that have the potential to reduce the number of pre-requisite courses ESL students need to take. Still, there is room for improvement at many colleges that continue to offer lengthy sequences using a traditional approach. Drawing on our research findings, we offer the following recommendations, several of which support the recommendations provided by the AB 705 ESL subcommittee. Colleges should consider shortening ESL sequences. Given our finding that students who enter the sequence at lower levels are less likely to complete transfer-level English within six years, we recommend that colleges shorten their sequences to help eliminate exit points and boost student outcomes. AB 705 has led some colleges to shorten their sequences already, either by incorporating accordion features or by compressing levels. While these acceleration strategies are relatively new and not to scale in many instances, we were unable to assess whether students do indeed progress to transfer-level English at a faster pace. Future research will need to examine whether acceleration strategies help students and whether all students benefit equally. Colleges should consider offering transferable ESL courses. Our study found that all ESL students benefit from the transferability of ESL coursework, and those who begin at lower levels benefit to a greater extent. While we cannot attribute causality to the findings, our analysis suggests that colleges with fewer (or no) transferable ESL courses have the potential to improve student success with this reform. Offering transferability for ESL allows students to make quicker progress toward their degree or transfer goals while learning English; in addition, transferability for ESL affirms the academic rigor and value of the coursework. In response to AB 705, colleges are also seeking humanities credit (not just elective credit) for ESL coursework; this will further boost the impact of transferable ESL coursework. Colleges should adopt an integrated approach to teaching English skills. Our analysis suggests that colleges that currently offer courses that focus on one skill (teaching writing and reading in separate courses) have the potential to improve student success by transitioning to an integrated approach (teaching reading and writing in the same course). Faculty indicated that in an integrated course, students are more likely to receive “just in-time” support on skills that are necessary to progress through the sequence (e.g., grammar support associated with a writing assignment). Integrated courses are also often are designed with the transfer-level English course objectives in mind and assign readings and associated writing exercises that are linked to a given theme. As a result, ESL students are exposed to the type of reading and writing they will need to do in transfer-level English and other courses—possibly gaining higher levels of comfort and confidence vis-à-vis course materials outside of ESL. However, simply offering an integrated course is not sufficient to promote improved outcomes: professional development that supports faculty in teaching integrated skills is critical. Many students can benefit from a direct pathway from ESL to transfer-level English. As colleges search for strategies to support the success of traditionally underrepresented students, our research found that a direct pathway from ESL to transfer-level English is especially beneficial for black and Latino students. Faculty we spoke to indicated that this kind of direct pathway provides language development support throughout a student’s journey to transfer English. However, Asian students may benefit from a faster transition to the English department via an ESL sequence that leads to developmental English. Therefore, each college should take account of its ESL student population when implementing this change. PPIC.ORG English as a Second Language in California’s Community Colleges 28 Non-credit offerings should seamlessly connect with credit ESL. Our research finds that colleges are offering non-credit ESL coursework at the lowest levels of ESL (some in partnership with adult schools) or parallel/mirroring credit ESL sequences. If this becomes an important strategy for shortening sequences, it will be imperative that colleges connect non-credit with credit ESL sequences. Our finding that Latinos are much less likely to be degree-seeking ESL students suggests that they may be more likely to enroll in these non-credit pathways. Therefore, strengthening connections between non-credit and credit ESL pathways can also help promote more equitable outcomes. New assessment and placement policies should be monitored to ensure accuracy, effectiveness, and equity. Many colleges indicated that they were making changes to assessment and placement in response to AB 705. Guided self-placement (GSP) figured prominently in these discussions. While the movement away from a reliance on standardized test scores is supported by research, it is unclear if the move toward GSP will promote improved outcomes. While GSP has been called the most socially just form of placement (Toth 2018) because it assumes that students are autonomous and can honestly determine their skill levels, it may also contribute to greater inequality due to its heavy reliance on information that is often strongly linked to social and cultural capital. It appears that some colleges may be using GSP to allow students to place themselves at lower ESL levels. One faculty member we spoke to voiced hesitation about using GSP in ESL because she questioned whether students low skill levels would result in poor accuracy and have an impact on equity. It will be critical to monitor student placements under GSP policies to ensure they are effectively and equitably placing students. Non-sequence students should be encouraged to pursue a degree or transfer. Given that Latino students are underrepresented among associate degree earners and transfer students, the high number of Latinos taking nonsequence courses presents an opportunity for the system. However, it is also possible that a growth in noncredit offerings could pull these students further off the degree-seeking ESL sequence track. It will be important to ensure that students are not unfairly tracked into these pathways. It will also be important to help students learn about and enroll in sequence courses and other degree applicable courses, which can potentially raise aspirations and lead to college degrees. Stronger student supports, including advising, placement, and information about degree and certificates available at the college might help more underrepresented students advance toward college degrees. The Guided Pathways ESL Milestone certificates at Cypress College is a great opportunity on this front. If non-sequence students learn about the prospect of earning a credential while learning English, they may be motivated to pursue further education. AB 705 is motivating colleges across the system to reform ESL programs and has initiated conversations about evidence-based practices that could improve long-term outcomes for ESL students. Careful implementation of assessment and placement and curricular reforms will be critical to ensuring that the goals of AB 705 are realized. To ensure equity in access and outcomes, it will also be important to include ESL students in broader college and system-wide initiatives such as the Vision for Success and Guided Pathways. PPIC.ORG English as a Second Language in California’s Community Colleges 29 REFERENCES Almon, C. 2012. “Retention of English Language Learners at a Community College.” In Linguistic Minority Students Go to College: Preparation, Access, and Persistence, ed. Y. Kanno and L. Harklau (New York: Routledge). Bailey, T., D. W Jeong, and S. W. Cho. 2010. “Referral, Enrollment, and Completion in Developmental Education Sequences in Community Colleges.” Economics of Education Review 29 (2): 255–70. Bailey, Thomas and E. Weininger. 2002. “Performance, Graduation, and Transfer Rates of Immigrants and Natives at City University of New York Community Colleges.” Educational Evaluation and Policy Analysis 24 (4) 359–77. Betts, Julian, Sam M. Young, Andrew C. Zau, and Karen Volz Bachofer. 2016. College Prep for All: Will San Diego Students Meet Challenging New Graduation Requirements? Public Policy Institute of California. Bickerstaff, Susan, and Julia Raufman. 2017. From "Additive" to "Integrative": Experiences of Faculty Teaching Developmental Integrated Reading and Writing Courses. Columbia University Teachers College, Community College Research Center. Bleakley, H. and Chin. A. 2004. “Language Skills and Earnings: Evidence from Childhood Immigrants.” Review of Economics and Statistics 86 (2): 481-496. Bunch, George, C. 2008. “Language Minority Students in California Community Colleges: Current Issues and Future Directions.” C4 eJournal, Issue 1. UC Riverside: California Community College Collaborative (C4). Bunch, George C., Ann Endris, Dora Panayotova, Michelle Romero, and Lorena Llosa. 2011. Mapping the Terrain: Language Testing and Placement for US-Educated Language Minority Students in California’s Community Colleges. Report prepared for the William and Flora Hewlett Foundation. Bunch, George and Ann Endris. 2012. Navigating “Open Access” Community Colleges. In Linguistic Minority Students Go to College: Preparation, Access, and Persistence, ed. Y. Kanno and L. Harklau (New York: Routledge). California Community Colleges Academic Senate. 2018. An Introduction to Guided Self-Placement (GSP). California Community College Chancellor’s Office. California Community Colleges Chancellor’s Office. 2018a. Statewide Student Success Scorecard, 2018. California Community College Chancellor’s Office, accessed January 15, 2019. California Community Colleges Chancellor’s Office. 2018b. Chancellor’s Office MIS Basic Skills Cohort Tracker. California Community College Chancellor’s Office. California Community Colleges Chancellor’s Office. 2018c. “Frequently Asked Questions about AB 705 for Credit English as a Second Language (ESL).” California Community College Chancellor’s Office. California Community Colleges Chancellor’s Office. 2018d. “AB 705 Credit ESL Implementation Timeline.” California Community College Chancellor’s Office. California Legislative Information. 2017. “AB-705 Seymour-Campbell Student Success Act of 2012.” California Legislative Information. Chiswick, B. R., and P. Miller. 1995. “The Endogeneity between Language and Earnings: International Analyses.” Journal of Labor Economics 13 (2): 246–88. Cuellar Mejia, Marisol, Olga Rodriguez, Hans Johnson, and Bonnie Brooks. 2018. Reforming English Pathways in California’s Community Colleges. Public Policy Institute of California. Gonzalez, A. 2000. “The Acquisition and Labor Market Value of Four English Skills: New Evidence from NALS.” Contemporary Economic Policy 18 (3): 259-269. Henson, Leslie and Katie Hern. 2018. A Seat at the Table: Supporting Student and Teacher Capacity in Corequisite English Remediation and Accelerated ESL. The California Acceleration Project. Hern, Katie. 2016. Sample Accelerated, Integrated Reading and Writing Course. Chabot College. Hodara, Michelle. 2015. “The Effects of English as a Second Language Courses on Language Minority Community Colleges Students.” Educational Evaluation and Policy Analysis, 37 (2): 1-24. Johnson, Hans, Marisol Cuellar Mejia, and Sarah Bohn. 2018. Higher Education as a Driver of Economic Mobility. Public Policy Institute of California. Kuehner, Alison V. and Jennifer Hurley. Forthcoming. “How Integrating Reading and Writing Supports Student Success.” Working Paper. Fremont, CA: Ohlone College. PPIC.ORG English as a Second Language in California’s Community Colleges 30 Merriam, Sharan. 1998. Qualitative Research and Case Study Applications in Education. (San Francisco: Jossey-Bass). Multiple Measures Assessment Project. 2017. Multiple Measures High School Variables Model Summary, Phase 2—Revised February 2017. Research and Planning Group for California Community Colleges. Multiple Measures Assessment Project. 2018. AB 705 Implementation Survey: ESL Focus. Research and Planning Group for California Community Colleges. Park, Elizabeth. Forthcoming. “Examining Community College Students’ Progression through the English as a Second Language Sequence.” University of Southern California. Perez, Alice. 2019. “Assembly Bill 705 and 1805 Spring 2019 Guidance Language for Credit English as a Second Language (ESL).” California Community College Chancellor’s Office. Perez, Alice, and John Stanskas. 2018. “Assembly Bill 705 Initial Guidance Language for Credit English as a Second Language.” California Community College Chancellor’s Office. Rivera-Batiz, F. L. 1990. “English Language Proficiency and the Economic Progress of Immigrants.” Economics Letters 34(3): 295–300. Rodriguez, Olga. 2013. “Examining the Effects of Academic English as a Second Language Pathways at the Community College: A Mixed Methods Study.” Doctoral Dissertation, Columbia University. Rodriguez, Olga, Marisol Cuellar Mejia, and Hans Johnson. 2016. Determining College Readiness in California’s Community Colleges: A Survey of Assessment and Placement Policies. Public Policy Institute of California. Rodriguez, Olga, Marisol Cuellar Mejia, and Hans Johnson. 2018. Remedial Education Reforms at Community Colleges: Early Evidence on Placement and Curricular Reforms. Public Policy Institute of California. Rodriguez, Olga, Hans Johnson, Marisol Cuellar Mejia, and Bonnie Brooks. 2017. Reforming Math Pathways at California’s Community Colleges. Public Policy Institute of California. Suárez-Orozco, Carola, Marcelo Suárez-Orozco, and Irina Todorova. 2008. Learning a New land: Immigrant Students in American Society. (Boston: Harvard University Press). Toth, C. 2018. “Directed Self-Placement at ‘Democracy’s Open Door’: Writing Placement and Social Justice in Community Colleges.” In Writing Assessment, Social Justice, and the Advancement of Opportunity, ed. Mya Poe, Asao B. Inoue, and Norbert Elliot. (Fort Collins, CO: University Press of Colorado). Willett, Terrence. 2017. English as a Second Language Multiple Measure Development. Research and Planning Group for California Community Colleges. PPIC.ORG English as a Second Language in California’s Community Colleges 31 ABOUT THE AUTHORS Olga Rodriguez is a research fellow at the PPIC Higher Education Center. She conducts research on the impact of programs and policies on student outcomes, with a particular focus on college access and success among underserved students. Her recent research focuses on statewide developmental education reform, assessment and placement systems, and place-based efforts to help students get into and through college. Before joining PPIC, she was a postdoctoral research associate at the Community College Research Center at Teachers College, Columbia University. She holds a PhD in economics and education from Columbia University. Laura Hill is a policy director and senior fellow at the Public Policy Institute of California. Her areas of expertise are K–12 education and immigration. She is currently researching English Learners in California schools, special education, and ESL students in community colleges. Her recent publications examine student achievement on the state’s K–12 assessments and the link between language reclassification policies and student success. Prior to joining PPIC, she was a research associate at the SPHERE Institute and a National Institute of Aging postdoctoral fellow. She holds a PhD in demography and an MA in economics from the University of California, Berkeley. Sarah Bohn is director of research and a senior fellow at the Public Policy Institute of California, where she holds the John and Louise Bryson Chair in Policy Research. She is also a member of the PPIC Higher Education Center. As director of research, she works with PPIC staff to bring high-quality, nonpartisan research to important policy issues in California. Her own research focuses on the role of social safety net policy and education policy in alleviating poverty and enhancing economic mobility. Her other areas of expertise include immigration policy, the workforce skills gap, and California’s economy. Her work has been published in major academic journals, including the American Economic Review, Demography, American Economic Journal: Economic Policy, and The Review of Economics and Statistics. She holds a PhD in economics from the University of Maryland, College Park Bonnie Brooks is a research associate at the PPIC Higher Education Center. Her work focuses on developmental education and career technical education at community colleges. Before joining PPIC, she was a data and research associate at the Los Angeles Promise Zone Initiative, where she measured the initiative’s impact on areas like public safety and education in local neighborhoods. She also worked at Grinnell College’s Data Analysis and Social Inquiry Lab, where she helped students and faculty integrate data into research and classroom work. She holds a BA in economics with a concentration in global development studies from Grinnell College. ACKNOWLEDGMENTS The authors would like to thank the community college ESL faculty who generously shared their time and valuable insights with us. We also thank faculty leaders from CATESOL and the Academic Senate who helped us better understand the complexity of ESL sequences and course data. We are especially thankful to Kathy Wada, Nikki Edgecombe, Niu Gao, Marisol Cuellar Mejia, and Lynette Ubois who provided valuable comments and reviews of earlier drafts. We also appreciate the critical editorial support provided by Mary Severance and government affairs support provided by Kellie Longo Flores and Deborah Gonzalez. This research would not have been possible without the data provided by the California Community Colleges Chancellor’s Office. Any errors are our own. PPIC.ORG English as a Second Language in California’s Community Colleges 32 PUBLIC POLICY INSTITUTE OF CALIFORNIA Board of Directors Steven A. Merksamer, Chair Senior Partner Nielsen Merksamer Parrinello Gross & Leoni LLP Mark Baldassare President and CEO Public Policy Institute of California Ruben Barrales Senior Vice President, External Relations Wells Fargo Bank María Blanco Executive Director University of California Immigrant Legal Services Center Louise Henry Bryson Chair Emerita, Board of Trustees J. Paul Getty Trust A. Marisa Chun Partner Crowell & Moring LLP Chet Hewitt President and CEO Sierra Health Foundation Phil Isenberg Former Chair Delta Stewardship Council Donna Lucas Chief Executive Officer Lucas Public Affairs Mas Masumoto Author and Farmer Leon E. Panetta Chairman The Panetta Institute for Public Policy Gerald L. Parsky Chairman Aurora Capital Group Kim Polese Chairman ClearStreet, Inc. Gaddi H. Vasquez Retired Senior Vice President, Government Affairs Edison International Southern California Edison PPIC.ORG English as a Second Language in California’s Community Colleges 33 The Public Policy Institute of California is dedicated to informing and improving public policy in California through independent, objective, nonpartisan research. Public Policy Institute of California 500 Washington Street, Suite 600 San Francisco, CA 94111 T: 415.291.4400 F: 415.291.4401 PPIC.ORG PPIC Sacramento Center Senator Office Building 1121 L Street, Suite 801 Sacramento, CA 95814 T: 916.440.1120 F: 916.440.1121" ["_edit_lock"]=> string(12) "1557248182:1" } ["___content":protected]=> string(210) "

English as a Second Language in Californias Community Colleges

" ["_permalink":protected]=> string(161) "https://www.ppic.org/publication/english-as-a-second-language-in-californias-community-colleges/english-as-a-second-language-in-californias-community-colleges-2/" ["_next":protected]=> array(0) { } ["_prev":protected]=> array(0) { } ["_css_class":protected]=> NULL ["id"]=> int(19209) ["ID"]=> int(19209) ["post_author"]=> string(1) "4" ["post_content"]=> string(0) "" ["post_date"]=> string(19) "2019-04-29 10:56:27" ["post_excerpt"]=> string(0) "" ["post_parent"]=> int(19134) ["post_status"]=> string(7) "inherit" ["post_title"]=> string(62) "English as a Second Language in Californias Community Colleges" ["post_type"]=> string(10) "attachment" ["slug"]=> string(64) "english-as-a-second-language-in-californias-community-colleges-2" ["__type":protected]=> NULL ["_wp_attached_file"]=> string(66) "english-as-a-second-language-in-californias-community-colleges.pdf" ["wpmf_size"]=> string(6) "862772" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(109594) "APRIL 2019 Olga Rodriguez, Sarah Bohn, Laura Hill, and Bonnie Brooks with research support from Sergio Sanchez and Stephanie Pulles Supported with funding from Bill & Melinda Gates Foundation, the College Futures Foundation, and the Sutton Family Fund English as a Second Language in California’s Community Colleges © 2019 Public Policy Institute of California PPIC is a public charity. It does not take or support positions on any ballot measures or on any local, state, or federal legislation, nor does it endorse, support, or oppose any political parties or candidates for public office. Short sections of text, not to exceed three paragraphs, may be quoted without written permission provided that full attribution is given to the source. Research publications reflect the views of the authors and do not necessarily reflect the views of our funders or of the staff, officers, advisory councils, or board of directors of the Public Policy Institute of California. SUMMARY CONTENTS Introduction 5 Current Landscape of ESL Programs 7 Degree-seeking ESL Students: Demographics and Outcomes 9 How Do ESL Reforms Affect Student Achievement? 17 ESL Reforms in Response to AB 705 21 Conclusion and Recommendations 27 The technical appendices to this report are available on the PPIC website. Due in large part to their accessibility and affordability, California’s community colleges (CCCs) play a central role in educating students who are not fully proficient in English. While there are many motivations for learning English, the economic benefits of English language skills are clear: in an economy that increasingly demands skilled workers, effective ESL programs can facilitate social and economic mobility for non-native speakers and their families. This report aims to fill the gap in research on ESL programs at California’s community colleges and the effectiveness of reforms aimed at improving student success. A number of colleges have implemented curricular reforms in recent years—accelerating coursework, integrating skills courses, and the like. These reforms are likely to become more widespread in the wake of Assembly Bill (AB) 705, signed into law in 2017. In addition to mandating that community colleges amend how students are placed in remedial courses, AB 705 requires that credit ESL sequences—which offer credit-bearing courses for which students pay tuition—maximize the probability that students enter and complete transfer-level English coursework within three years. Our research suggests that while many colleges are moving in the right direction, more could be done. Specifically, we find that:  Most degree-seeking ESL students do not successfully complete transfer-level English; those who begin at lower proficiency levels are especially likely to drop off the pathway. We find that 34 percent of all degree-seeking ESL students who first enrolled between 2010 and 2012 successfully completed transfer-level English within six years—but this share includes 56 percent of those who began one level below transfer-level English and just 9 percent of those who began eight levels below transfer.  Nearly half of colleges offer no more than five levels of ESL, which would theoretically allow students to complete transfer-level English in the sixth term. Still, only 20 percent of students who started five levels below transfer completed this course within six years and even fewer did so within three years.  Colleges have made great strides in moving away from a traditional approach—which teaches language skills such as reading and writing separately—to an integrated approach that teaches multiple language skills in a single course. Our findings suggest that all students benefit from the integration of language skills in ESL coursework.  Many colleges offer ESL courses that are transferable to UC and CSU. Students are benefiting from the opportunity to make progress toward degree and transfer goals while gaining proficiency in academic English. Recent efforts to secure humanities credit for advanced ESL courses may further boost the impact of transferable ESL coursework. PPIC.ORG English as a Second Language in California’s Community Colleges 3  Many colleges still offer ESL sequences that require students to complete developmental English coursework prior to enrolling to transfer-level English. Our findings indicate that these colleges would see a boost in outcomes if ESL sequences led directly to transfer-level English. This structural change is supported by AB 705, which recognizes that instruction in ESL is distinct from remediation in English. Our analysis suggests that current ESL program reforms hold promise. As colleges across the state move toward compliance with AB 705, more research is needed to determine whether new approaches to ESL assessment and placement and acceleration help improve the outcomes of diverse groups of students. PPIC.ORG English as a Second Language in California’s Community Colleges 4 Introduction Community college ESL programs serve a large and diverse mix of students, including young adults who attended California’s K–12 schools, immigrants with high school, college, or graduate degrees from their home countries, and working-age immigrants in California’s labor force. In the 2016–17 academic year alone, more than 58,000 students enrolled in one or more English as a Second Language (ESL) courses at community colleges across the state.1 Some ESL students need to improve their English in order to progress toward associate degrees, career certificates, or transfers to four-year schools. Others enroll in ESL for personal reasons or to improve their job prospects. The range of motivations and goals present a challenge in ensuring successful outcomes for all students.2 But in an economy where job opportunity is tied to skills and higher education (Johnson, Cuellar Mejia, and Bohn 2018), research has consistently shown that acquiring higher levels of English proficiency improves labor market outcomes (Bleakly and Chin 2004; Chiswick and Miller 1995; Gonzales 2000; Rivera-Batiz 1990). In addition to this diverse student population and other complexities, data constraints make it difficult to assess ESL programs. But a better understanding is critical, especially now. California’s community colleges (CCCs) are in the midst of reforming many programs that aim to prepare students for transfer-level coursework. A number of colleges have implemented curricular reforms in recent years to improve student outcomes. In addition, Assembly Bill 705, signed into law in 2017, requires community colleges to amend how students are assessed and placed and to maximize the probability that students who start on ESL pathways will complete transfer-level English coursework within three years (see text box).3 Previous PPIC research on developmental math and English reforms found policies aimed at accelerating developmental sequences and refining course placement have the potential to markedly improve student outcomes (Cuellar Mejia, Rodriguez, and Johnson 2018; Rodriguez, Cuellar Mejia, and Johnson 2018, 2017). This report offers a similar analysis of ESL programs in light of reforms that are being considered or are already under way. We use qualitative and quantitative data to shed light on the breadth of students in ESL programs and how colleges aim to meet student needs. Longitudinal student records collected by the Chancellor’s Office through its management information system (MIS) form the foundation for our analysis. But this data cannot provide course information that is critical to understanding how changes in ESL programs affect student outcomes—accurate information on the skills taught in each course, whether a course is a prerequisite for transfer-level English, or how many levels below transfer-level English it is. For this reason, we collected detailed information on the characteristics of ESL courses and programs from course catalogs and faculty input. Our analysis is informed by semi-structured interviews we conducted with ESL department chairs, faculty, and division deans at 13 community colleges in November and December 2018. We are grateful for the insights of our interviewees, many of whom are evaluating how to revise their programs to accomplish the goals of AB 705 (see Technical Appendix A for details). The bulk of this report examines the pathways of students who intend to continue their community college education beyond ESL.4 We focus on these students and their outcomes for two main reasons. AB 705 aims to 1 This statistic is based on an unduplicated count of students enrolled in ESL courses at a given college in academic year 2016-2017. Students who enrolled in ESL courses at multiple colleges in that year may be counted twice. During this year, 12,098 sections of ESL were offered across the system. We estimate that out of all math, English and ESL students in the CCC, those in ESL represent about 17 percent of the entire population (i.e., not restricted to degree-seeking students). 2 For background see Almon 2012; Bailey and Weininger 2002; Bunch and Endris 2012; Bunch et al 2011; Suarez-Orozco, Suarez-Orozco and Todorova 2008. 3 The equivalent timeframe for math and English is one year. 4 We estimate that degree-seeking ESL students represented about 8 percent of all degree-seeking math, English, and ESL students in the CCC between the 2009–10 and 2014–15 academic years. The share of ESL students in the degree-seeking sample is lower than the ESL share in the overall population because English learners are less likely than their math and English peers to enter the CCC with an intent to pursue a degree or transfer. PPIC.ORG English as a Second Language in California’s Community Colleges 5 AB 705 and credit-bearing ESL Assembly Bill 705 was signed into law by Governor Brown in October 2017. It is probably best known for its reforms to developmental education: it requires community colleges to maximize the probability that students will enter and complete transfer-level coursework in English and mathematics and mandates that colleges use high school records (e.g., coursework, grades, and/or grade point average) as the primary criteria for placement recommendations (California Legislative Information 2017). However, the bill also acknowledges that ESL is distinct from remediation in English, in that it serves foreign language learners seeking proficiency in an additional language. AB 705 requires colleges to maximize the probability that students who enter credit-bearing ESL course sequences will “complete degree and transfer requirements in English within three years.” The three-year timeframe begins when a student declares an intent to pursue a degree or transfer to a four-year institution. Importantly, the law only applies to credit-bearing ESL sequences. To help colleges meet the compliance deadline of fall 2020, the Chancellor’s Office and the Academic Senate are developing recommendations. Memos released in July 2018 and April 2019 provide guidance on using evidence-based rules for placing students into credit ESL to maximize the probability that they will complete transfer-level English within three years (six semesters or nine quarters). The memos also offer a series of curricular recommendations. The memos also suggest that colleges use high school GPA for placement, integrate ESL skills courses, ensure that ESL sequences lead directly to transfer-level English instead of developmental English, and the like. Additional guidance is expected to inform placement within the ESL sequence, especially for students who did not graduate from US high schools and those with no high school records—groups that can comprise sizable shares of the ESL population (Perez and Stanskas 2018; Perez 2019). Our data suggest that two-thirds of degree-seeking ESL students in California do not have a US high school diploma; a similar share are noncitizens (e.g., permanent residents, students with visas, refugees/asylees, etc.) and about a third are non-traditional-age (see technical appendix Table B6). improve outcomes for students intending to complete transfer-level English, a course that is required for students intending to earn an associate degree or transfer to a four-year university. Second, it is difficult to obtain data related to the success of ESL students who learn English for other reasons. For example, learning English might improve an individual’s job prospects, but although we are fortunate to have access to longitudinal student-level data through the Chancellor’s Office, information on career outcomes is not available. We start with a detailed picture of the current landscape of ESL course offerings that highlights the variation across colleges in terms of sequence length, type of sequence, and sequence end point. Next, we identify degreeseeking ESL students and examine their demographic characteristics and progress toward transfer-level English. We then explore how ESL program features affect student achievement, including the outcomes of ESL students from historically underrepresented backgrounds. Finally, we turn to the changes currently taking place in light of AB 705, examining key issues that colleges should consider. We conclude with several policy recommendations. PPIC.ORG English as a Second Language in California’s Community Colleges 6 Current Landscape of ESL Programs ESL programs are typically designed to teach a range of language skills—listening, speaking, reading, grammar, writing, vocabulary—at a variety of proficiency levels. Traditionally, each language skill was taught discretely in a standalone course. For example, a student might enroll concurrently in a reading course and a writing course, and the two courses would have largely unrelated content. More and more, colleges are offering integrated sequences, which teach more than one skill in a single course—for example, a student with mid-range English skills might enroll in an intermediate reading and writing course. A handful of colleges have adopted new course sequences that allow students to skip levels of ESL, depending their proficiency (these are known as accordion sequences), compress multiple levels of ESL into a single course, and support ESL students enrolled in transferlevel English courses. These recent reforms attempt to shorten the pathways from ESL to transfer-level English— with the aim of reducing attrition and improving student completion of ESL sequences.5 As evidence in developmental English and math suggests, shortening the pathway to transfer-level coursework and strengthening the alignment with the transfer-level course improves the odds that students enroll and pass transfer-level courses (Cuellar Mejia, Rodriguez, and Johnson 2018; Rodriguez, Cuellar Mejia, and Johnson 2018, 2017). ESL Pathways to Transfer-Level English Roughly 27 percent of ESL course sections are prerequisites for transfer-level English, or “sequence courses.” These course offerings reflect the needs of students at a given community college, but are also shaped by alternative ESL providers in the community, such as adult schools.6 ESL sequence courses vary substantially. Most (89%) confer credit but only roughly less than in four offer credit that is degree applicable or transferable to a four-year institution (see technical appendix Table B3 for more details). Sequence courses that do not confer credit are typically many levels below transfer-level English and intended for beginning English speakers at colleges in communities with few other ESL providers. A few colleges also offer noncredit ESL sequences—which offer tuition-free courses and are not subject to the AB 705 timeframe—that mirror their credit ESL pathways. It may seem odd that some ESL courses offer transfer credits even though they are prerequisites to transfer-level English.7 However, transferable ESL courses reflect the reality that learning English is not “remedial”; it is equivalent to a native English speaker learning a foreign language. AB 705 acknowledges this reality; it states that “ESL is distinct from remediation in English” and that “students enrolled in ESL credit coursework are foreign language learners.” Faculty we spoke to agreed that transferable ESL courses make sense because their rigor is comparable to that of a foreign language course. A secondary reason for making ESL credits transferable is to allow students to make faster progress toward their degree or transfer goals while learning English. As is 5 Recent legislation aims to do something similar in the K–12 system—that is, to increase English Learners’ access to college preparatory courses. AB 2735 requires that school districts allow English Learners full access to all of a school’s standard instructional program, including a–g courses as well as AP classes. AB 2121 requires districts to permit a fifth year of high school for newcomers and students in the Migrant Education Program in order to provide more opportunities to earn required graduation credits. 6 Adult education ESL and community college-based ESL often work in tandem to meet the needs of English learners in their community. English learners, especially those with very little or no English proficiency, may start coursework outside of community college, even if they intend to seek transfer-level English and/or college credentials. If these students continue on the pathway, we only observe them once they have entered a community college ESL sequence course. In our analysis, we assume that courses at a given level below transfer-level English are comparable, so that although we cannot observe prior English training in any detail, students are at similar levels of English proficiency. Six community college districts include adult education institutions, meaning that we can observe enrollment in adult education– based ESL. In those four districts alone, it is clear that there is a great demand for ESL even beyond students who enroll in community college. For example, in 2013, 79 percent of ESL students enrolled in adult education–based ESL in the San Diego Community College district, compared to 21 percent in the colleges in that district. Unfortunately, there are many more adult education schools that are not included in the MIS data. 7 See technical appendix Table A5 for more information on the types of ESL sequences that use transferable ESL courses. PPIC.ORG English as a Second Language in California’s Community Colleges 7 discussed in a later section, there are important developments with respect to the CSU/UC transferability of ESL in response to AB 705.8 Sequence courses in ESL typically focus on reading (54%) and writing (80%) skills, or integrate the two. Relatively few of these courses teach listening, speaking, and/or grammar, but most students (53%) who enroll in sequence courses at some point in their community college career also enroll in non-sequence courses that offer these additional skills. ESL Pathway Characteristics While ESL pathways take different forms, traditionally ESL sequences to transfer-level English offer sequential, separate skills courses as shown in Figure 1. As of the 2016–17 academic year, 29 colleges offer a traditional ESL sequence. FIGURE 1 A sample of a traditional ESL sequence SOURCE: Author’s chart based on PPIC ESL program database. The most common number of levels—offered at 25 percent of colleges—is six (Figure 2). AB 705 indicates that a five-level sequence would theoretically allow students to complete transfer-level English in six terms. Almost half of colleges, 48 percent, offer a sequence that is five or fewer levels,9 but relatively few students enroll at the lowest levels: 82 percent of students begin at level five or higher. This suggests that it may not be extraordinarily difficult to shorten sequences. 8 It is also important to point out that during the timeframe we study, five colleges also offered transferable ESL courses that were equivalent to freshman composition offered in the English Department in that they meet graduation requirements. In fact, 8 percent of students in our sample completed the freshman composition requirement via the ESL version of this course. In this study we opted not to highlight these courses because they do not transfer to all colleges and universities—and even when they do transfer, they may only transfer to certain programs. In fact, we find that 58 percent of students who enrolled in an ESL freshman composition course also took it a second time in the English Department, which contributes to the problem of excess credits. 9 Keep in mind that this only includes ESL sequence courses; some students may have taken non-sequence ESL courses in college or enrolled in ESL courses at adult schools or other venues. PPIC.ORG English as a Second Language in California’s Community Colleges 8 FIGURE 2 Some ESL sequences are lengthy, but most students start ESL only a few levels below transfer-level English Percent 30% 25% 20% 15% 10% 5% 0% 6% 1% 9+ 8% 2% 14% 4% 25% 10% 17% 13% 19% 16% 21% 11% 21% 2% Colleges Degree-seeking students 8% 2% 87654321 Levels below transfer-level English SOURCE: Author calculations from Chancellor’s Office MIS data and PPIC ESL program database. NOTES: Degree-seeking students are coded according to their first sequence ESL course, for students who first enrolled between the 2009–10 and 2016–17 academic years (N=120,365). Colleges are coded according to the longest sequence offered in the 2016–17 academic year. If ESL coursework leads to developmental English, the developmental English course(s) is counted in the sequence. See technical appendix Table B4 for raw numbers. In 2016–17 (the most recent academic year for which we have data), completing the ESL sequence was not sufficient for enrolling in transfer-level English at 40 colleges.10 At these colleges, students also needed to complete one or more developmental English courses. ESL students at these colleges would enter the sequence to transfer-level English at least two levels below transfer English. Faculty told us that the developmental English requirement was intended to help ESL students acquire the language skills needed for academic English courses. It may have been partly driven by a perception that only prerequisites taught in an English department can help students acquire the proficiency needed for transfer-level English. Current Chancellor’s Office guidance on AB 705 cites ESL-to-developmental-English pathways as a focus of reform. Of the 51 colleges that had ESL sequences leading to developmental English in the 2009–10 school year, 14 had changed their sequences to lead directly to transfer-level English as of 2016–17.11 These data suggest that many colleges are already in alignment with AB 705’s requirement that ESL students be able to complete transfer-level English within three years, or six terms. But whether students actually do complete transfer-level English within three years is another matter. As we show in the remainder of this report, a relatively small percentage of degree-seeking ESL students successfully complete transfer-level English even within six years. Degree-seeking ESL Students: Demographics and Outcomes We can identify degree/transfer seeking ESL students by looking at their course-taking behavior. Our degreeseeking cohort includes any ESL student who took at least one degree-applicable course outside of ESL or transfer-level English and/or at least one developmental English/math course at some point during their college 10 See technical appendix Table A5 for more information on the types of sequences that feed into transfer-level English and developmental English. 11 Some colleges introduce ESL programs/sequences after the 2009–10 academic year that are included for in the 40 colleges we observe as still offering ESL sequences that feed into developmental English coursework as of 2016–17, but are not part of the original 51 that offered this type of sequence in 2009–10. We derive 14 by asking the original 51 colleges how many changed their ESL sequences to feed directly into transfer-level English. PPIC.ORG English as a Second Language in California’s Community Colleges 9 journey.12 One-third of the 356,209 students who took at least one ESL course between 2009–10 and 2014–15 also took at least one ESL course that is a prerequisite for transfer-level English (i.e., a sequence class).13 As Figure 3 shows, the majority of ESL students (53%) only took ESL courses that are not prerequisites for transferlevel English courses (i.e., a non-sequence class).14 And the remaining 13 percent took at least one ESL sequence course but did not take any other courses consistent with degree/transfer-seeking.15 FIGURE 3 A third of ESL students are seeking degrees 53% 47% 34% non-sequence sequence, non-degree sequence, degree-seeking 13% SOURCE: Authors’ calculations based on MIS data and college catalog scan. Because our research focus is on ESL students seeking degrees or transfers, we omit the two-thirds of community college students who enrolled in ESL but did not demonstrate course-taking consistent with an intent to pursue a degree or transfer. In technical appendix Table B9, we compare demographic characteristics of degree-seeking ESL students and other ESL students. Overall, we find that degree-seeking ESL students are more likely to be Asian, to have a foreign high school diploma, to be a permanent resident, to be under the age of 25 (traditional college age), and to be a recent high school graduate than non-degree-seeking ESL students. We find that ESL students seeking degrees typically enter an ESL sequence three or four levels below transferlevel English, and 62 percent are traditionally college-aged. The most common prior educational attainment is a foreign secondary school diploma (41%), while 33 percent hold a US high school diploma.16 The most common immigration status is legal permanent resident (38%), followed by student visa holder (25%). US citizens make up 23 percent of degree-seeking ESL students. Nearly half of degree-seeking ESL students are Asian (47%); 23 percent are Latino, 19 percent are white, and 2 percent are black. There are some important educational and immigration-related differences across racial/ethnic groups that may be linked to students’ ESL experiences. Latino students are much more likely to have been US educated than other 12 Our definition of degree and/or transfer-seeking student relies on course taking behavior. CCCCO data administrators cautioned us against using the educational goal data element collected during our sample timeframe because it is considered an “uninformed” educational goal which is collected during the application process. Starting summer 2014, this data element is considered an “informed” educational goal because it is updated after a student has reviewed assessment results and received orientation and other student services. This educational goals variable is now being used by the CCCCO Student Success Metrics to define the degree-seeking population. See Technical Appendix A for more information. 13 Enrolled for the first time in an ESL course within this timeframe. We limit our sample to this timeframe in order to have at least a three-year follow-up for outcomes. 14 Non-prerequisite courses are advisory or elective courses that commonly teach listening/speaking, grammar, and some reading skills. See technical appendix Table B3 for more detail on the characteristics of these courses. 15 Some of these ESL students not seeking a college degree may have pursued or obtained a certificate. 16 It is possible to graduate from a US high school and still be considered an English Learner. High school graduation rates are lower for current English Learners than for students who have been reclassified or were never English Learners. PPIC.ORG English as a Second Language in California’s Community Colleges 10 students (Figure 4); 62 percent have either a high school diploma or GED. Only one-third of Asian students, 39 percent of black students, and 29 percent of white students have US diplomas or GEDs. The majority of white students (55%) hold a foreign secondary school diploma, as do 44 percent of Asian students and 42 percent of black students, compared to only 18 percent of Latino students. Black and white ESL students start four levels below transfer-level English, while Latino and Asian students start three levels below, on average. FIGURE 4 Latino degree-seeking ESL students are much more likely to be US-educated 100 90 14 10 16 11 11 80 18 Unknown Percent 70 41 60 50 40 5 30 20 33 10 06 Overall 9 53 9 Latino 44 5 29 5 Asian 42 9 31 7 Black 55 4 25 3 White Foreign secondary school diploma GDEDE// HS equivalency/ CA HS proficiency certificate High school diploma Adult school Less than HS SOURCES: Authors’ calculation based on MIS data. NOTE: Sample includes students first enrolled in ESL in the CCC system between the 2009–10 and 2014–15 academic years. See technical appendix Tables B1 and B2 for more details. While more than half of white ESL students are permanent residents, as are approximately 40 percent of black and Asian ESL students, only 29 percent of Latino ESL students are legal permanent residents (Figure 5). More than one-third of Asian ESL students are visa holders, compared to only 4 percent of Latino ESL students. Among ESL students, we find that half of Latinos are US citizens, but citizenship rates are much lower for other racial/ethnic groups. FIGURE 5 Half of Latino degree-seeking ESL students are US citizens Percent 100 7 90 80 25 70 6 60 50 40 38 30 20 10 23 0 Overall 14 4 29 50 Latino 5 34 3 41 15 Asian 2 12 9 40 31 Black 3 9 21 52 14 White Other/Unknown Student Visa holder (F1 or M1) Refugee/Asylee Temporal resident Permanent resident U.S. Citizen SOURCES: Authors’ calculation based on MIS data. NOTE: Sample includes students first enrolled in ESL in the CCC system between the 2009–10 and 2014–15 academic years. See technical appendix Tables B1 and B2 for more details. PPIC.ORG English as a Second Language in California’s Community Colleges 11 ESL students with either a GED or US high school diploma start three levels below transfer-level English, on average, whereas all other educational groups start four levels below (see technical appendix Table B2 for more detail). More than three-quarters of high school diploma holders are traditionally college-age; they make up the youngest group of ESL students. These varying characteristics can help us understand how different groups of students enter and progress through ESL sequences. ESL Progression to Transfer-level English Completing transfer-level English is one of the foundational general education requirements at California’s community colleges. For students intending to obtain an associate degree or transfer to a four-year university this requirement is a key milestone. Yet many CCC students fail to achieve this outcome, and ESL students are even less likely to do so. We estimate that only 36 percent of ESL degree-seeking students who first enrolled during the 2011–12 academic year successfully completed transfer-level English within six years (Table 1) compared to 48 percent among those who did not receive ESL instruction (CCCCO 2018a). While this number has improved over time, Figure 6 shows it is apparent that starting at a lower level of ESL coursework is correlated with lower completion rates of transfer-level English. TABLE 1 Successful completion of transfer-level English by year first enrolled in CCC system (%) Overall 2009–10 2010–11 2011–12 2012–13 2013–14 2014–15 3-year 26 22 23 25 27 28 30 6-year 34 33 34 36 n/a n/a n/a SOURCE: Author calculations from Chancellor’s Office MIS data and PPIC ESL program database. NOTES: Students who first enrolled between 2009–10 and 2014–15 are tracked for three years (N=120,365); Students who first enrolled between 2009–10 and 2011–12 are tracked for six years (N=64,527). See technical appendix Table A3 for cohort sample sizes. Figure 6 charts student progress through an ESL sequence depending on the level of entry. One thing is clear: as is the case in traditional developmental math and English sequences, the more levels students must traverse, the more exit points there are, and the less likely students are to complete a transfer-level course (Bailey, Jeong, and Cho 2010; Cuellar-Mejia et al. 2016). To illustrate this, we highlight two ESL sequence starting points: (1) two levels below transfer English, from which theoretically a student could complete transfer-level English in three terms or 1.5 years, and (2) five levels below transfer English, from which a student could theoretically complete a transfer-level course in the AB 705 timeframe of six terms or three years. We see that among the 13,212 students in our sample who started in ESL two levels below transfer-level English between the 2009–10 and 2011–12 academic years, 86 percent successfully completed this level, 62 percent subsequently took the next course in the sequence (one level below) and 56 percent successfully completed the ESL sequence. Ultimately, 56 percent of students starting in ESL two levels below enrolled in a transfer-level English course within six years and 51 percent successfully completed the course. Among the 8,509 students who started five levels below transfer we find that only 20 percent completed it within six years and only 9 percent achieved this outcome within the AB 705-mandated three-year timeframe.17 17 This statistic is presented for comparison purposes to align with the three-year AB 705 timeline. However, given length of ESL sequences our tables and figures focus on six-year completion of transfer-level English. Table with three-year completion of transfer-level English by ESL starting level is available upon request. PPIC.ORG English as a Second Language in California’s Community Colleges 12 FIGURE 6 Attrition is high in ESL sequences ESL Cohort (2010–2012): Six Year Progression in ESL Sequence Percent 100 100 100 90 85 80 86 70 69 60 60 50 Six Levels Below Transfer (6,981) 54 46 62 56 56 51 40 Five Levels Below Transfer (8,509) 30 Four Levels Below Transfer (12,702) 41 35 32 28 20 Three Levels Below (12,793) 23 20 Two Levels Below (13,212) 10 One-Level Below (5,122) 0 SOURCE: Author calculations from Chancellor’s Office MIS data and PPIC ESL program database. NOTES: Includes students who first enrolled during the 2009–10 and 2011–12 academic years in one through six levels below transfer (LBT) English. Numbers in parentheses indicate number of students starting at each level. See technical appendix Table B5 for underlying data for these and other levels below transfer. Figures for attrition by ESL sequence type (feeding directly into transfer-level English or developmental English) and characteristics (traditional, integrated, mixed, and accordion sequences) are available on request. There are a few points to highlight. First, the high first-course pass rates (above 80%) indicate that a student who enrolls in an ESL course is likely to successfully complete it. Second, many students who successfully complete a first course do not enroll in the next course in the sequence. This may not always be a negative outcome. As we track and interpret student progression though the ESL sequence, it is important to note that some colleges have designed their sequences or instituted departmental policies that enable students to skip levels based on achievement. For example, colleges might have ESL sequence structures intended to accelerate student progress, first-day-of-class writing assessments, or departmental rules that allow students to bypass developmental English.18 We can see that 12 percent of students who began two levels below transfer English and successfully completed that course actually skipped that level and moved directly to transfer-level English. However, a larger share (17%) of those who successfully completed ESL two levels below transfer failed to show up again in an ESL course or in transfer-level English. Our faculty interviews suggest that high attrition rates at upper levels may partly be driven by ESL sequence structures that require some ESL students to complete one or more levels of developmental English prior to enrolling in transfer-level English. Given the link between successful completion of transfer-level English and the level at which a degree-seeking ESL student starts, it seems important to look at differences across races/ethnicities and levels of prior education. 18 Several faculty we spoke to indicated that it is not uncommon for ESL faculty to give students an in-class writing assessment during the first weeks of class, which may result in students moving up levels. Additionally, one college shared that although their sequence officially leads to developmental English, up to two-thirds of its students are cleared move directly to transfer-level English based on demonstrated achievement. PPIC.ORG English as a Second Language in California’s Community Colleges 13 At almost every level of ESL entry, Asian degree-seeking students are the most likely and Latino students the least likely to complete a transfer-level course (Figure 7). Recall that black and white students start four levels below transfer-level English on average, while Asian and Latino students start three levels below. We find that white students starting four levels below are more likely than Latino students starting three levels below to complete transfer-level English (30% versus 24%). Focusing on racial/ethnic differences among students starting two levels below (the largest category), we find even more dramatic variation in completion rates at the extremes: 62 percent for Asian ESL students and 33 percent for Latino ESL students. FIGURE 7 Equity gaps persist in completion for students who enter ESL sequences at all levels 70 62 60 % completing transfer-level English from each starting level 50 40 33 30 20 14 10 45 42 36 27 17 15 15 1 LBT 2 LBT 3 LBT 4 LBT 5 LBT 6 LBT 0 Latino Asian Black Two or More White SOURCE: Author calculations from Chancellor’s Office MIS data and PPIC ESL program database. NOTES: Sample includes degree-seeking students who first enrolled in ESL at a California Community College between the 2009–10 and 2011–12 academic years and are tracked for six years, through the 2016–17 academic year. LBT refers to level below transfer English. In sum, black students, a relatively small percentage of all degree-seeking ESL students, start at low levels and have lower throughput rates at every level. Latino students start at high levels (a greater proportion have US high school diplomas) but have the lowest completion rates. Analyses of degree-seeking ESL students by prior education suggest that foreign high school graduates have the highest completion rates no matter where they start in the sequence, with one notable exception: adult school students who start one level below transfer-level English have the highest completion rates.19 Students who report a US high school diploma or a GED have roughly equal transfer-level English completion rates across starting levels. These findings are described in more detail in technical appendix Figure B2. This descriptive examination of student progression makes it clear that the overwhelming majority of degreeintending ESL students are not reaching a key milestone needed to achieve their educational goals, even after six years. It also reveals important differences across races/ethnicities and prior education levels. Ultimately the high attrition in ESL sequences highlights the importance of redesigning sequences to meet AB 705 goals. Early Reform Efforts in ESL Programs Given system-wide information on relatively low transition rates of ESL students to transfer-level English, as well as the evolution of curricular models for teaching ESL, a number of ESL programs have been adapting their course models in recent years (CCCCO 2018b; Bunch 2008; Henson and Hern 2018; Hodara 2015; Park 2018; 19 It should be noted that this group comprises a small sample (N=61). PPIC.ORG English as a Second Language in California’s Community Colleges 14 Rodriguez 2013). Some colleges have shortened or compressed their ESL pathways to transfer-level English (technical appendix Figure B1). However, most ESL reforms have focused on how English skills are taught. Since 2010, ESL sequence courses that teach integrated language skills have overtaken those that teach skills in the traditional, standalone manner. In academic year 2009–10, roughly 31,000 students enrolled in traditional sequence courses;20 by 2016–17, that number had fallen to around 22,000 (Figure 8). Over that same period, 81 colleges included traditional courses in their ESL sequences and 91 offered integrated courses. ESL sequences leading to transfer-level English often include a mix of traditional and integrated courses; in recent years, students have increasingly been choosing integrated courses over stand-alone offerings. FIGURE 8 Overall ESL enrollment is holding steady, but traditional ESL enrollment is declining 70,000 Students enrolled in ESL sequence courses 60,000 50,000 40,000 30,000 20,000 TrRaAdiDtiIoTnIaOlNAL InNtTeEgrGatReAdTED ToOtaTlAL 10,000 0 2010 2011 2012 2013 2014 2015 2016 2017 SOURCE: Author calculations from Chancellor’s Office MIS data and PPIC ESL program database. NOTES: Enrollment for students who enrolled in ESL sequence courses over time where 2010 refers to 2009–10, and so on. Students are counted once in a given academic year, according to the type of ESL course(s) he or she enrolled in. If the student enrolled in both traditional and integrated coursework in a given academic year, they are counted in both lines. Otherwise students are unduplicated within college but may be duplicated across colleges if students enrolled in multiple colleges at any point in their college career. See Table B7 in Appendix B for raw numbers. ESL courses have traditionally taken a discrete skills approach for a variety of reasons. One faculty member told us that discrete skills were taught at her college partly because faculty were accustomed to discrete skill structures or because they specialized in particular skills. Some faculty believed that students needed to learn skills separately to be successful in academic English settings. For instance, one faculty member said that having separate reading and writing courses is important because strategies that reading instructors feel are critical for success might not be covered thoroughly in an integrated reading and writing course. These comments illustrate the challenges of integrating ESL skills and reinforce the notion that true integration requires more than simply combining discrete skills into a single, longer course (Bickerstaff and Raufman 2017). They also underscore the importance of professional development to help ESL faculty move toward an integrated approach. However, faculty who transitioned from discrete to integrated skills courses believed that some degree of integration had already been happening. Given that reading and grammar is necessary for academic writing, it 20 It is important to note that even when a college does offer a traditional ESL sequence with separate writing, reading, grammar and listening/speaking courses, it may require students to take only the writing sequence to advance to college English—other skill areas are often offered as electives. PPIC.ORG English as a Second Language in California’s Community Colleges 15 was natural to embed grammar exercises into a writing course—for example, by identifying error patterns and providing “just-in-time” support. Similarly, students in integrated courses read a variety of material—including novels, short stories, and articles sometimes based on a theme—to support their writing and develop their vocabulary. Integrated courses were often described as being designed with the transfer-level English course objectives in mind. Some faculty also indicated that they used the California Acceleration Project’s (CAP’s) principles of design in integrated courses—which includes “just in time” interventions, low-stakes collaborative practice, group work, and an emphasis on addressing students’ affective needs. One faculty member noted that the traditional approach made it difficult to evaluate student preparedness in English because skills were taught in separate courses and not all skills were required. Integration helps address this problem. On the logistical front, faculty indicated that enrollment difficulties and faculty availability were motivations for integrating courses—it was not uncommon for courses in some skill areas to be cancelled due to low enrollment, leaving students stranded and faculty without courses to teach. Integrated ESL sequences are now the dominant approach across California community colleges (Figure 9). In addition, a handful of colleges have implemented course sequences that allow students to make faster progress toward transfer-level English. Only 3 percent of colleges have adopted reforms that shorten sequences by compressing courses and only 4 percent allow students who make significant progress to skip ahead. Both approaches eliminate exit points in the sequence and allow students to progress at a faster rate. However, because only a few colleges and a few students have adopted these reforms, it is not possible to assess their impact on student outcomes.21 FIGURE 9 The integrated sequence has become the most common type 100 90 80 Percent of colleges 70 60 50 44 40 29 30 20 21 10 4 3 0 Traditional Integrated Mixed ESL sequence type Accordion Compressed SOURCE: PPIC ESL program database. NOTES: Sequence structures of colleges offering ESL sequences as of the 2016–2017 academic year. Our study includes a total of 105 colleges. The nine colleges not included in our study did not have readily available or comprehensible information about their respective ESL sequences, and/or could not be reached for more information. In the timeframe of our analytical sample, the total number of colleges in each category fluctuates, as some colleges established ESL programs in different years, and colleges have changed their structure and sequence types between 2010 and 2017. See Technical Appendix A Glossary for more details. 21 Given the newness of accordion and compression models, we are unable to track student outcomes for six years. However, emerging evidence from colleges suggests that these approaches can work in ESL. Faculty at Cuyamaca College shared that after compressing the top levels and incorporating an accordion model the ESL sequence was reduced from seven levels to five; this resulted in all students “going faster”—about one-third of students completed transfer-level English in four terms compared to fewer than 10 percent under the old model. Future research should assess the efficacy of these models. PPIC.ORG English as a Second Language in California’s Community Colleges 16 How Do ESL Reforms Affect Student Achievement? Our analysis of student course-taking data and academic outcomes covers the period prior to the passage and implementation of AB 705. However, as we have seen, a number of colleges were either attempting reforms or already had structures similar to those recommended in the AB 705 guidance (including integrated courses, direct pathways to transfer English, and transferable ESL credits). In this section, we use statistical analyses to assess the relationship between some of the reformed ESL course features and academic outcomes for degree-seeking ESL students. Our analysis focuses on four primary outcomes: completion of transfer-level English, transferable credits awarded, community college degree completion, and transfer to a four-year college. This part of the analysis focuses on students whose outcomes of interest we could track for three or six years. To align with the AB 705 timeframe, we track completing transfer-level English (C or better) within three and six years; given the lengthy ESL sequences, however, we limit tracking subsequent outcomes—accruing transfer credits and obtaining a degree or transferring—to six years. Assessing Individual ESL Sequence Features To help inform AB 705 reforms, we now examine the role of different ESL sequence features in promoting improved student outcomes. We focus on three features: (1) integrated sequences, which teach two or more language skills per class; (2) sequences that lead directly to transfer-level English; and (3) transferable ESL courses, which award credits that transfer to a CSU and/or UC school.22 In this phase of the research, we used statistical models to control for differences in student, college and cohort characteristics that may also influence student outcomes (see Technical Appendix A for more detail). After controlling for student characteristics and college and cohort effects, including the level at which a student enters the ESL sequence, we find evidence that each of these ESL sequence features helps students progress, and that some help more than others. Holding all else constant, we find that taking a transferable ESL course has the biggest impact on a student’s likelihood of completing transfer-level English within three and six years: between 16 and 20 percentage points (Figure 10). Students taking integrated courses are 11 to 12 percentage points more likely to complete transfer-level English within three and six years, respectively. A sequence that leads directly to transfer-level English is also associated with a higher likelihood of completion of that course within three and six years (between 7 and 12 percentage points). While our research finds evidence to support what we saw in the ESL sequence progression charts (Figure 6)— that the lower the level at which students enter, the less likely they are to complete transfer-level English—we also find that students starting at the lowest levels of ESL benefit less from integration and from sequences that lead directly to transfer-level English compared to those who start at the top two levels. Interestingly, students starting at levels three or four appear to benefit more from transferable ESL in a six-year timeframe, compared to their peers who begin in the top two levels (see technical appendix Tables B11–12). These findings suggest that students who start at different levels have varying experiences with sequence features and that more attention should be given to their differing needs. Our research also found some evidence that ESL sequence structures support longer-term student success—that is, the proportion of transferable credits obtained, degree completion, and transfer to a four-year college within six 22 See technical appendix Table A5 for more information on the how these different sequence features overlap (i.e., 55% of integrated sequences feed directly into transfer-level English). PPIC.ORG English as a Second Language in California’s Community Colleges 17 years. All else being equal, as would be expected, we find that taking a transferable ESL course increases the proportion of transfer FIGURE 10 ESL sequence features increase the likelihood of completing transfer-level English Transferable ESL Integration 11.2 11.8 16.1 20.4 6 Years 3 Years Direct to TLE 11.5 7.4 0 5 10 15 20 25 Change in likelihood of completing transfer-level English (percentage point) SOURCE: Author calculations from Chancellor’s Office MIS data and PPIC ESL program database. Full results available in technical appendix Tables B11 and B12. NOTES: Results from individual-level regression models. Our sample includes degree-seeking first-time college students entering one of California’s community colleges between 2009–10 and 2014–15 who are tracked through the 2016–17 academic year. All results are statistically significant at the 90% level or above. credits earned by 9 percentage points, and that students entering three to four levels below the sequence benefit more from taking a transferable ESL course than their peers who enter in the top two levels. We find that the only ESL sequence feature that is associated with obtaining a community college degree is a transferable ESL course— which improves the likelihood of earning a degree by almost 13 percentage points. Integration was also found to be beneficial, but it was only associated with the share of transferable credits awarded (4 percentage points more than traditional ESL courses). Finally, we find little evidence that ESL sequence features improve a student’s likelihood of transferring after six years; taking a transferable course improves this likelihood marginally (by 3 percentage points). Across the board, we find that students entering at the lowest levels of ESL are much less likely to have positive academic outcomes over the six-year timeframe (see technical appendix Tables B13–B15). Our ESL faculty interviews shed light on our findings. They indicate that transferable ESL courses consistently emerge as a strong correlate of student outcomes because these courses “hold students to high a standard.” Faculty we spoke to confirmed that offering transferable ESL courses as part of the sequence served to emphasize their rigor as university-level foreign language work. The opportunity to earn transferable credits while they are learning English allows students to advance toward their degree or transfer goals—and this, in turn, probably boosts aspirations, especially among those who begin in lower levels. We also learned that integration supports student success partly due to the fact that in an integrated course, students are more likely to receive “just in time” help with skills that are necessary to progress through the sequence (e.g., grammar and reading comprehension associated with a given writing assignment). We also learned that integrated sequences that are designed with transfer-level English course objectives in mind and taught with a theme-based approach that aligns reading and writing assignments can be especially helpful in facilitating student success. These design features mimic the type of reading and writing that students do in transfer-level English and other content courses outside of ESL. By contrast, in a traditional sequence, all of these skills would be addressed separately, so that the material in a reading course does not always align with what students are learning in the writing course. PPIC.ORG English as a Second Language in California’s Community Colleges 18 Do ESL Sequence Reforms Help All Student Groups? Our descriptive analyses have already demonstrated differences in outcomes across groups of degree-seeking ESL students. In particular, we see lower sequence completion rates for Latino and black students. Here, we examine whether there are statistically significant differences in the impact of ESL sequence features on the likelihood of degree-seeking students across racial/ethnic groups completing sequences, controlling for student characteristics (including a student’s level of entry), college characteristics, and cohort effects. Taking transferable ESL courses is associated with higher rates of throughput within three and six years for white, Latino, and Asian students (Figure 11). Such course-taking is associated with 13 to 19 percentage point increases in three-year throughput and 17 to 25 percentage point increases in six-year throughput. Taking a transferable ESL course does not appear to be associated with higher throughput for black students. FIGURE 11 ESL sequence features increase the likelihood of completing transfer-level English, but varies by race/ethnicity Three-Year Throughput Transferable ESL Feed Direct TLE Integration 0 12.7 18.8 16.3 7.1 15.1 11.9 7.3 12.5 5 10 15 20.8 20 25 Black White Latino Asian Six-Year Throughput 32 30 35 Transferable ESL Feed Direct TLE 11.5 16.8 20.5 18.4 24.9 36.1 Integration 0 5.9 15.2 5 10 15 20 25 30 35 40 Black White Latino Asian SOURCE: Author calculations from Chancellor's Office MIS data and PPIC ESL program database. Full results available in technical appendix Tables B16 and B17. NOTES: Results from individual-level regression models, estimated separately by race. Our sample includes degree-seeking first-time college students entering one of California’s community colleges between 2009–10 and 2014–15 who are tracked through the 2016–17 academic year. Only statistically significant results displayed. While our sample size for black students is small, our coefficients are estimated with sufficient precision to be certain of differences between black ESL and other ESL students on many measures. Only statistically significant results at the 90% level or above are shown. We also test for statistically significant differences between race/ethnic groups. In three-year throughput, all differences between black and other racial/ethnic groups reported are statistically significant, with the exception of the black/white difference in the association between integration and throughput. For six-year throughput, all differences between black and other racial/ethnic groups and between Asian and other racial/ethnic groups were statistically significant for feeding directly into transfer-level English. Racial/ethnic group differences were not statistically significant in six-year throughput for transferable ESL or integration. See technical appendix Table B20 for a summary of the tests for between group statistical significance. PPIC.ORG English as a Second Language in California’s Community Colleges 19 We find that Asian students appear to benefit from integrated ESL courses: they are 13 percentage points more likely to complete transfer-level English within three years and 15 percentages points more likely to do so within six years if they have ever enrolled in an integrated course. Latino ESL students also appear to benefit from integrated course structures, but the effect is smaller (7 percentage points in the three-year timeframe and 6 percentage points in the six-year time horizon). Integrated ESL course structure also improves successful completion within three years for white and black students, increasing throughput by 12 and 21 percentage points, respectively. Integration is the only ESL feature that appears to increase throughput for all racial/ethnic groups. Being at a college where the ESL pathway feeds directly into transfer-level English appears to be particularly beneficial to black students, who are 32 and 36 percentage points more likely to reach transfer English within three and six years, respectively, than their peers in colleges where ESL feeds into developmental English. We also find a statistically significant association for white students—an 18 percentage point increase in six-year completion. Black and white students start sequences at lower levels, on average, than do Asian or Latino students. While Latino students typically start at higher levels than black and white students, they also have higher throughput rates at colleges where ESL feeds into transfer-level English (a 12 percentage point increase in six-year completion and a 15 percentage point increase in three- year completion). It is also important to consider how a number of successful longer-term academic outcomes across racial/ethnic groups are related to ESL course structure, controlling for all else. In particular, we examine the association between the percentages of attempted transfer credits earned, transfers to a four-year college, and degree attainment within six years and our three ESL design features: integrated courses, sequences that feed directly into transfer-level English, and transferable courses. We find that ESL sequences ending in transfer-level English are most beneficial to Latino students when it comes to percentage of attempted credits completed (technical appendix Table B18).23 For Latino students, these sequences are associated with obtaining community college degrees (technical appendix Table B19).24 These sequences do not appear to be associated with transfers to four-year colleges for any racial/ethnic group (technical appendix Table B20). As with successful completion of an ESL sequences, there is no positive association for Asian ESL students between sequences that feed into transfer-level English and longer-term outcomes. In general we do not find strong evidence that integrated sequences are particularly helpful when it comes to successful longer-term outcomes, with just a couple of exceptions: Latino and black students at colleges with integrated sequences are more likely to obtain a community college degree (4 and 10 percentage points, respectively).25 It is not surprising that transferable ESL is associated with the percentage of attempted transfer credits earned. However, this feature seems to be especially helpful for black students, who experience a 17 percentage point increase in credits earned, compared to the other racial/ethnic groups (whose increases range from 8 to 12 percentage points), although differences between black and other ESL students are not statistically significant. And while we did not find that transferable ESL is linked to black ESL students completing transfer-level English, it is associated with helping black students earn more college credits. In addition, taking transferable ESL is positively associated with completing a community college degree for all race/ethnic groups, but particularly for Asian and white ESL students.26 Transferable ESL credits appear associated with modest increases in transferring to four- year colleges for Latino and white students but not for Asian or black students. 23 Differences between Latinos and whites and Latinos and Asians are statistically significant. 24 Differences between Latinos and Asians are statistically significant. 25 Differences between black and white ESL students are statistically significant. 26 Differences between Asian and Latino and Asian and white students are statistically significant. PPIC.ORG English as a Second Language in California’s Community Colleges 20 In short, our analysis suggests that integration increases the odds of all groups of students completing transferlevel English and improves several longer-term outcomes. As they did with our overall findings, faculty interviews suggest that these effects are likely driven by the key principals of integration itself—in particular, content designed to lead students toward the goal of transfer-level English; the ability to provide “just in time” support on skills that students need to progress through the sequence (e.g., helping students with grammar, in in the context of a writing assignment); and a thematic approach that is relevant to students’ lives (e.g., social justice). Interestingly, we find that a direct transition from ESL to transfer-level English has differential impacts across student groups: Latino and black students benefit more from going directly into transfer-level English, while Asian students at colleges that offered an ESL sequence that led directly to transfer English earned a smaller share of attempted credits. Faculty we spoke to indicated that an ESL sequence leading to transfer-level English provides language development support throughout a student’s pathway to transfer English. Still, it appears that Asian students may benefit from transitioning to the English department sooner via an ESL sequence that leads to developmental English. Finally, transferable ESL credit was important for a variety of student subgroups. This suggests that the opportunity to take more rigorous ESL courses and be considered foreign language learners may be raising educational aspirations among traditionally underserved groups. ESL Reforms in Response to AB 705 We have shown that few degree-seeking students in traditional ESL programs achieve key educational milestones, but that integrated ESL courses, direct pathways from ESL to transfer-level English, and transferable ESL coursework are associated with an improved likelihood of achieving key academic outcomes. Many colleges are already implementing these structures, but many others are not—and, until recently, few questioned the assessment and placement system that assigned students to an ESL sequence. AB 705 aims to boost outcomes for ESL students by requiring all colleges to rethink their assessment and placement systems and traditional ESL course structures. Interviews with ESL department chairs, faculty, and administrators at 13 colleges across the CCC yielded insights into the changes colleges are making in response to AB 705. In this section, we highlight the most salient themes. Changes to ESL Assessment and Placement In earlier research on assessment and placement in California’s community colleges, we found that all colleges reported using standardized placement tests in ESL during the 2015–16 academic year. We also found that the use of other measures was very limited. The most common multiple measure, used at 24 percent of colleges, was an instructor/counselor recommendation (Rodriguez et al. 2016). A recent AB 705 implementation survey confirmed that the majority of colleges continue to use standardized tests to assess and place students in ESL, even those with four years of high school data (Multiple Measures Assessment Project 2019).27 The inclusion of ESL programs in AB 705 has completely upended this status quo—indeed, one ESL faculty member described the law as “the biggest change that has happened at the community colleges.” 27 This is partly due to the fact that a high share of students in ESL have not graduated from a US high school. In our sample of degree-seeking students, only one-third have a US high school diploma. This small share is not surprising: a recent estimate using Cal-PASS plus data finds that 87 percent of ESL students who graduated from a California high school and enrolled in community college did not take ESL; instead, they enrolled in English for native speakers (Willett 2017). PPIC.ORG English as a Second Language in California’s Community Colleges 21 Starting in fall 2020, colleges must use “evidence-based multiple measures” for placement into ESL coursework (California Legislative Information 2017; CCCCO 2018c, 2018d).28 High school records are the default evidencebased measure. If high school records are unavailable, the law considers guided self-placement (GSP) to be a viable method. The AB 705 ESL workgroup recently provided colleges with guidance on evidence-based rules for placing students into credit ESL (Perez and Stanskas 2018). To begin, the workgroup outlined default placement rules for ESL students with four years of high school records (Table 2). The default placement guidelines mirror the recommendations for English placement—i.e., that all students can be placed in transfer-level English, and that students with lower GPAs should receive concurrent support. Importantly, the workgroup clearly states that this does not mean that all recent high school graduates will be ready for transfer-level English—some students may still need to enroll in ESL to acquire academic language proficiency. Additional guidance is expected on placement within the ESL sequence, especially for students with fewer than four years of high school records (e.g., older immigrants, refugees, and foreign students). Still, several faculty voiced concern about the default placement rules making it difficult for students to take ESL—as one faculty member noted, choosing ESL over a co-requisite could require extra time and money. TABLE 2 Chancellor’s Office AB 705 default placement rules for ESL High school performance metric Recommended AB 705 placement Throughput for students enrolling directly in transfer-level English GPA >=2.6 Transfer-level English composition No additional academic and concurrent support required 78.6% GPA 1.9to 2.6 Transfer-level English composition Additional academic and concurrent support recommended GPA <1.9 Transfer-level English composition Additional academic and concurrent support strongly recommended SOURCE: Perez and Stanskas 2018. 57.7% 42.6% As of fall 2018, 10 of the 13 colleges we spoke to had begun to plan for or had already implemented changes in response to AB 705. Changes to assessment and placement were partly driven by the elimination of standardized placement tests in English as a result of AB 705—some colleges use the English assessment for ESL, and others have been using ESL placement tests that they now think they will need to eliminate. New guidance issued in April 2019 clarified that AB 705 does not require that colleges eliminate placement tests for ESL at this time— noting that some tests can still be used at least through July 2021 (Perez 2019).The colleges that did not report making changes indicated that they were waiting for more guidance from the Chancellor’s Office. It is anticipated that as new guidance and information is released, an increasing number of colleges will be implementing changes in response to AB 705. A recent AB 705 guidance memo indicated that all colleges will be asked to submit AB 705 adoption plans for ESL by July 2020 (Perez 2019). Below we summarize the most commonly described changes to ESL assessment and placement policies in response to AB 705. 28 In December 2018, the CCCCO and the Academic Senate published a frequently asked questions document in which they clarify that the three-year timeline for AB 705 “begins at the point that a student enrolled in a credit ESL sequence declares intent for degree or transfer as determined by their educational goal, education plan, and/or major declaration” (CCCCO, 2018c). In math and English, the one-year timeline begins when a student first enrolls in a math or English course—the assumption is that all students who enroll in math or English courses intend to pursue a degree or transfer. PPIC.ORG English as a Second Language in California’s Community Colleges 22 Guided Self-placement In our interviews with ESL faculty we frequently heard that colleges would be adopting a guided self-placement (GSP) approach in addition to or instead of a standardized placement test.29 As part of a GSP process, a college provides students with course information that is intended to help them evaluate their readiness for a given course level. This information includes course descriptions, sample assignments, projects, and exams. The GSP process is meant to help students select courses that align with their educational goals and abilities (CCC Academic Senate 2018). GSP is an attractive placement option for a large group of ESL students (e.g., older adults, refugees, or those who went to foreign high schools) who may not have high school records from a US high school. ESL faculty described a typical GSP process as providing students with reading and/or writing samples for different levels of ESL and allowing them to choose the example that aligns most closely with their reading/writing abilities. This in turn helps students select the most appropriate ESL placement level. Some examples of the guiding questions in the GSP process are “Which of these writings is most like the one you would do? What level do you think you should be in?” or “Do I understand the reading? Can I write this well?” Changes to assessment and placement at Cuyamaca College Prior to fall 2018, Cuyamaca College primarily relied on standardized placement tests and writing samples. For instance, applicants would be asked a few questions to determine whether they need to be referred to ESL (e.g., is English used in the home, is English your first language, etc.), then they would take the Combined English Language Skills Assessment (CELSA) and respond to a writing prompt that was graded by faculty. Because this approach had low predictive value, the college completely transformed the way it assessed and placed students into ESL. The college has eliminated the placement test and instead uses the “best of” three measures: Self-reported high school grade point average—which use the MMAP rule set for placement into the different levels of ESL (MMAP 2017). The ESL placement survey developed by Irvine Valley College. Guided self-placement: students are given writing samples from different levels of ESL and asked to pick the level that best matches the type of writing they would do. The department chair believes the new assessment and placement approach has been working well. Students are good at placing themselves and appear to be more satisfied with this approach. It is important to highlight that in addition to the changes described above, Cuyamaca College has also transformed their ESL sequence over the past several years by incorporating acceleration strategies. In making these changes to the course structures the college reduced the amount of time students spend in ESL from 3.5 to 2 years. After the changes to the ESL sequence and assessment and placement, the college believes it is ahead of the game and in compliance with the law. In fact, the ESL department chair believes they are “grinning about AB 705” – noting that the policy “caught up to us” not the other way around. Multiple Measures Faculty also reported that their colleges are moving toward multiple measures in ESL. Several reported using a multiple measures ESL survey that was developed by Irvine Valley College (IVC). The IVC survey gathers 29 A recent AB 705 implementation survey conducted by the Multiple Measures Assessment Project (MMAP) finds that of 104 colleges that responded, 13 are currently using GSP and 44 plan to use it in the future (Multiple Measures Assessment Project 2019). PPIC.ORG English as a Second Language in California’s Community Colleges 23 information about a student’s academic attainment, self-evaluated English proficiency, and use of English in everyday life (e.g., phone language settings). Researchers at IVC have begun to assess survey data from other colleges to assess the survey’s predictive utility. A couple of colleges reported using the Multiple Measures Assessment Project rules to determine ESL placements. For example, Cuyamaca College has used self-reported high school GPA as part of a disjunctive, or “best of,” placement process that also included the IVC survey and a guided self-placement tool (see text box). Changes to ESL Course Structure Ten of the thirteen colleges we spoke to were in the process or had already made changes to their ESL course structures.30 Changes to course structures are primarily driven by AB 705’s mandate that colleges maximize the probability that ESL students complete transfer English within three years. The guidance provided by the Chancellor’s Office also provided a series of curricular recommendations for colleges as they prepare for the fall 2020 compliance deadline. Colleges are advised to:  integrate the teaching of language skills  develop direct pathways from ESL to transfer-level English  design ESL courses from the vantage point of transfer-level English  allow credit ESL faculty to teach transfer-level English or create an ESL equivalent  increase professional development opportunities for credit and transfer-level English faculty  seek CSU humanities transfer credit for advanced transfer-level ESL courses Several colleges we spoke to said that their college already had acted on transferable ESL, designing sequences from the vantage point of transfer-level English, sequences that were five levels or less, and direct pathways to transfer-level English. During our interviews we heard that colleges were planning to change their ESL course structures in other ways. Below we summarize what we learned. Non-credit ESL Non-credit ESL, which was not included in the Chancellor’s Office recommendations, was being considered or implemented at six colleges. Colleges described using non-credit ESL in two ways: (1) to replace the lowest levels of credit ESL and (2) to mirror the credit ESL sequence. Both approaches focus on students who need more than three years to reach transfer-level English. As one faculty member noted, the three-year timeframe could drive some students toward the non-credit pathway—especially those at lower levels. Several faculty at colleges with lengthy sequences spoke about turning the lowest levels into non-credit ESL. Others spoke about developing non-credit courses that mirror the credit ESL sequence—so that all students could benefit from the same ESL course whether it offers credit or not. Faculty shared that in addition to being outside a three-year timeframe, noncredit courses are free and repeatable. Several faculty also noted that a non-credit option would be helpful for students who do not intend to obtain a degree or transfer to a four-year school. ESL Sequences that Lead Directly to Transfer-level English Faculty at colleges with ESL sequences that feed into developmental English emphasized the urgent need for ESL sequences that lead directly to transfer-level English.31 This need is driven by AB 705, which is prompting colleges to replace developmental English with co-requisite remediation, and which has made a clear distinction 30 The MMAP survey of AB 705 implementation finds that 13 campuses are currently piloting curricular changes to credit ESL in 2018–19; 27 plan to pilot curricular changes in 2019–20. 31 The MMAP (2019) survey found that eight colleges are developing a credit pathway from ESL to transfer-level English or an ESL version of this course. PPIC.ORG English as a Second Language in California’s Community Colleges 24 between ESL—a program that involves advanced foreign-language learning—and developmental English. ESL faculty spoke about developing additional ESL courses that would parallel developmental English. Our research suggests this will reduce attrition among ESL students. One faculty member noted that it was not uncommon for students to do well ESL but fail to progress in developmental English. One possible explanation is that because developmental English is taught from “an aspect of teaching English speakers,” ESL errors are penalized and there is no language support. Accelerated ESL Completion Several colleges indicated that they were planning to accelerate ESL sequences by compressing, integrating, using an accordion model, or having faculty certified in both ESL and freshman composition teach transfer-level English and a linked co-requisite course.32 Some of these colleges began their acceleration efforts prior to AB 705; at other colleges, the policy created a heightened sense of urgency. The primary motivation for accelerating ESL pathways was data showing that few students were completing transfer-level English. At two colleges that had been offering seven-level sequences, fewer than 10 percent of students were making it to transfer-level English. ESL faculty members stressed that these outcomes were “horrible” and “not good enough.” Irvine Valley College has or is in the process of implementing several reforms (see text box). Interestingly, two colleges implemented acceleration strategies that did not work for them. One college introduced an accordion model and then realized that most students transitioned from “A to A,” so there was little need for a “B” level. Moreover, the accordion model had “always been confusing for data collection purposes” because the system considers it to be twice as long. Starting fall 2019, this college decided to allow students who need additional support—which would have otherwise been provided in a “B” level course—to take a non-credit version of the credit ESL course—the hope is that adopting the mirrored sequence will result in faculty electing to drop the “B” level course. Another college introduced a co-requisite but quickly learned that having English faculty teach the transfer-level English course and ESL faculty teach the co-requisite was not going to work. Having different instructors was “logistically impossible” because students could go through five sections of freshman composition with five different sets of readings and assignments. It is important to highlight that some colleges are not considering reducing the number of ESL levels even though their sequences include six or more levels. Faculty cited the need to provide longer sequences because they serve student populations with very low English skills—and because they believe that many are only there to learn English, not to pursue a degree or transfer. They noted that AB 705 does not outlaw sequences with more than five levels; it prohibits the college from placing students any lower than level five, but students could place themselves at level six or lower. Other Innovative Changes in ESL Colleges we spoke to are exploring ways to support the progression of ESL students more broadly in college and beyond. 32 On acceleration front, the MMAP (2019) survey found that eight colleges are working on creating an ESL program that lasts three semesters. A similar number of colleges identified plans to pilot concurrent support models in ESL. PPIC.ORG English as a Second Language in California’s Community Colleges 25 Changes to ESL course structures at Irvine Valley College The ESL transformation at Irvine Valley College (IVC) exemplifies several of the approaches described above: IVC implemented an accordion sequence, created a direct pathway to transfer-level English, designed a noncredit adult ESL program, and expects to soon offer a co-requisite paired with transfer-level English taught by ESL instructors certified in composition. IVC launched its reform effort in 2012 in response to low throughput percentages, particularly at the transition point from ESL to developmental English. In 2014, IVC redesigned their sequence so that it would lead directly to transfer-level English. In doing so it designed the top two levels of ESL writing/reading courses to mirror the developmental writing sequence, and brought the curriculum in line with California Acceleration Project (CAP) design principles. However, these changes did not improve throughput. IVC staff believed that one of the reasons for this was that many ESL students did not want to pursue a degree or transfer—after surveying their students they found only about half were pursuing these goals. In 2016 they implemented two pathways: academic ESL (credit) and adult ESL (non-credit). The credit ESL path also includes an accordion model, which has been in place for many years. With the implementation of AB 705, IVC recognized that it can no longer offer six or seven levels of credit ESL courses. It plans to offer only the top four levels in credit ESL; lower-level courses will be part of the non-credit pathway, which also includes intermediate and advanced courses to meet the needs of the noncredit student population. Adult ESL students who intend to transfer or pursue a degree will be able to transition to the credit ESL sequence by enrolling in the first level of academic writing or higher, depending on their ability and needs. Additionally, starting in fall 2019, the college plans to offer a co-requisite linked to transfer-level English that will be taught by an ESL instructor certified to teach freshman composition. The co-requisite will provide the language support needed to achieve the objectives of transfer-level English. In addition to changing its ESL course structures, the college has been transforming its ESL assessment and placement system. In fall 2019, the college will replace the standardized placement test with a new in-house survey, while continuing to use a writing sample. In addition, credit ESL students will be able to use the English guided self-placement tool. ESL Certificates Recognizing the benefits of aligning its student success efforts with two major statewide reforms—AB 705 and Guided Pathways—Cypress College has created a Guided Pathways ESL Milestone certificate of achievement. The certificate recognizes that by successfully completing the highest levels of ESL—which currently transfer to a CSU and UC as elective units and fulfill Category C2 Humanities requirements for the local degree—and completing an introductory or core course in a meta major pathway, a student has demonstrated academic English reading and writing skills at an advanced, postsecondary level and achieved success in a pathway toward a degree, certificate, or general education transfer. The college has developed the certificates in various programs and plans to begin awarding them in fall 2019 (see text box). Humanities Credits Several faculty we spoke to were considering or had already submitted advanced ESL courses that offer not just elective credit but humanities credit. The goal is to have advanced ESL coursework fulfill CSU C2/IGETC 3B Humanities requirements. One faculty member emphasized that this is an equity issue: ESL students should PPIC.ORG English as a Second Language in California’s Community Colleges 26 receive credit for being foreign language learners able to engage in rich cultural content inherent to ESL courses. Several faculty indicated that the language competencies in top-level ESL courses are at a higher level than those in Spanish 1. If this change were to happen, faculty argued that students could potentially save up to 15 units of coursework. This would align with the AB 705 goal for that students to make quicker progress toward degree and transfer without excess credits. Our findings suggest this change could further boost the impact of transferable ESL courses on student outcomes, and could be especially helpful for students entering ESL sequences at lower levels. Guided Pathways ESL Milestone Certificates at Cypress College Starting in fall 2019, Cypress College will award certificates to students who master advanced transfer-level academic English skills and complete at least one course in a Guided Pathways meta major. The idea is that these students have achieved academic English reading and writing skills at an advanced postsecondary level and taken prerequisite or introductory coursework in a degree, certificate, or general education pathway. Most certificate programs recommend but do not require that students take an educational planning elective. To determine which meta majors to prioritize for the 2018–19 credit ESL Milestone certificates, institutional researchers gathered data on the top awards and/or certificates earned by ESL students at the college. This rich collaboration between research and faculty produced a list of pathways:  Business and Computer Information Systems (CIS)  Career Technical Education/Aviation and Travel Careers  Dental Hygiene/Nursing/Psychiatric Technology  Diagnostic Medical Sonography, Health Information Technology, Radiologic Technology  Kinesiology  Language Arts/Oral Communication  Language Arts/Written Communication  Science Engineering and Math (SEM)  Social Sciences For example, to earn the GP ESL Milestone certificate for completing the Science Engineering and Math (SEM) pathway (the most commonly completed pathway), students would need 10 units of the top two levels of ESL (ESL 185 and 186) 4 units of College Algebra; and 4–5 units of Preparation for General Chemistry, General Chemistry I, or Elementary Physics. Students would also be encouraged to take an educational planning course (COUN 140) for 1 unit. As of April 2019, the certificate program had been approved by the college, district, and Chancellor’s office. Conclusion and Recommendations One of the most striking findings in this report is the simple fact that a large share of degree-intending students drop off the ESL pathway to transfer-level English: only one-third of all degree-seeking ESL students who first enrolled between 2010 and 2012 successfully completed transfer-level English within six years. Moreover, this share includes a high of 56 percent of those who started out one level below transfer-level English to just 9 percent of those who begin eight levels below. And while nearly half of colleges offer no more than five levels of ESL, which PPIC.ORG English as a Second Language in California’s Community Colleges 27 would theoretically allow students to complete transfer-level ESL in six terms, only 20 percent of students starting at this level complete transfer-level English within six years and even fewer did so within three years. Many community colleges have already begun structuring their ESL sequences in ways that are aligned with the AB 705 implementation guidelines provided by the Chancellor’s Office and Academic Senate (Perez and Stanskas 2018). In particular, colleges have made great strides in integrating the teaching of multiple language skills, structuring sequences to lead directly to transfer-level English, incorporating transferable ESL coursework into the sequence, and adopting other strategies that have the potential to reduce the number of pre-requisite courses ESL students need to take. Still, there is room for improvement at many colleges that continue to offer lengthy sequences using a traditional approach. Drawing on our research findings, we offer the following recommendations, several of which support the recommendations provided by the AB 705 ESL subcommittee. Colleges should consider shortening ESL sequences. Given our finding that students who enter the sequence at lower levels are less likely to complete transfer-level English within six years, we recommend that colleges shorten their sequences to help eliminate exit points and boost student outcomes. AB 705 has led some colleges to shorten their sequences already, either by incorporating accordion features or by compressing levels. While these acceleration strategies are relatively new and not to scale in many instances, we were unable to assess whether students do indeed progress to transfer-level English at a faster pace. Future research will need to examine whether acceleration strategies help students and whether all students benefit equally. Colleges should consider offering transferable ESL courses. Our study found that all ESL students benefit from the transferability of ESL coursework, and those who begin at lower levels benefit to a greater extent. While we cannot attribute causality to the findings, our analysis suggests that colleges with fewer (or no) transferable ESL courses have the potential to improve student success with this reform. Offering transferability for ESL allows students to make quicker progress toward their degree or transfer goals while learning English; in addition, transferability for ESL affirms the academic rigor and value of the coursework. In response to AB 705, colleges are also seeking humanities credit (not just elective credit) for ESL coursework; this will further boost the impact of transferable ESL coursework. Colleges should adopt an integrated approach to teaching English skills. Our analysis suggests that colleges that currently offer courses that focus on one skill (teaching writing and reading in separate courses) have the potential to improve student success by transitioning to an integrated approach (teaching reading and writing in the same course). Faculty indicated that in an integrated course, students are more likely to receive “just in-time” support on skills that are necessary to progress through the sequence (e.g., grammar support associated with a writing assignment). Integrated courses are also often are designed with the transfer-level English course objectives in mind and assign readings and associated writing exercises that are linked to a given theme. As a result, ESL students are exposed to the type of reading and writing they will need to do in transfer-level English and other courses—possibly gaining higher levels of comfort and confidence vis-à-vis course materials outside of ESL. However, simply offering an integrated course is not sufficient to promote improved outcomes: professional development that supports faculty in teaching integrated skills is critical. Many students can benefit from a direct pathway from ESL to transfer-level English. As colleges search for strategies to support the success of traditionally underrepresented students, our research found that a direct pathway from ESL to transfer-level English is especially beneficial for black and Latino students. Faculty we spoke to indicated that this kind of direct pathway provides language development support throughout a student’s journey to transfer English. However, Asian students may benefit from a faster transition to the English department via an ESL sequence that leads to developmental English. Therefore, each college should take account of its ESL student population when implementing this change. PPIC.ORG English as a Second Language in California’s Community Colleges 28 Non-credit offerings should seamlessly connect with credit ESL. Our research finds that colleges are offering non-credit ESL coursework at the lowest levels of ESL (some in partnership with adult schools) or parallel/mirroring credit ESL sequences. If this becomes an important strategy for shortening sequences, it will be imperative that colleges connect non-credit with credit ESL sequences. Our finding that Latinos are much less likely to be degree-seeking ESL students suggests that they may be more likely to enroll in these non-credit pathways. Therefore, strengthening connections between non-credit and credit ESL pathways can also help promote more equitable outcomes. New assessment and placement policies should be monitored to ensure accuracy, effectiveness, and equity. Many colleges indicated that they were making changes to assessment and placement in response to AB 705. Guided self-placement (GSP) figured prominently in these discussions. While the movement away from a reliance on standardized test scores is supported by research, it is unclear if the move toward GSP will promote improved outcomes. While GSP has been called the most socially just form of placement (Toth 2018) because it assumes that students are autonomous and can honestly determine their skill levels, it may also contribute to greater inequality due to its heavy reliance on information that is often strongly linked to social and cultural capital. It appears that some colleges may be using GSP to allow students to place themselves at lower ESL levels. One faculty member we spoke to voiced hesitation about using GSP in ESL because she questioned whether students low skill levels would result in poor accuracy and have an impact on equity. It will be critical to monitor student placements under GSP policies to ensure they are effectively and equitably placing students. Non-sequence students should be encouraged to pursue a degree or transfer. Given that Latino students are underrepresented among associate degree earners and transfer students, the high number of Latinos taking nonsequence courses presents an opportunity for the system. However, it is also possible that a growth in noncredit offerings could pull these students further off the degree-seeking ESL sequence track. It will be important to ensure that students are not unfairly tracked into these pathways. It will also be important to help students learn about and enroll in sequence courses and other degree applicable courses, which can potentially raise aspirations and lead to college degrees. Stronger student supports, including advising, placement, and information about degree and certificates available at the college might help more underrepresented students advance toward college degrees. The Guided Pathways ESL Milestone certificates at Cypress College is a great opportunity on this front. If non-sequence students learn about the prospect of earning a credential while learning English, they may be motivated to pursue further education. AB 705 is motivating colleges across the system to reform ESL programs and has initiated conversations about evidence-based practices that could improve long-term outcomes for ESL students. Careful implementation of assessment and placement and curricular reforms will be critical to ensuring that the goals of AB 705 are realized. To ensure equity in access and outcomes, it will also be important to include ESL students in broader college and system-wide initiatives such as the Vision for Success and Guided Pathways. PPIC.ORG English as a Second Language in California’s Community Colleges 29 REFERENCES Almon, C. 2012. “Retention of English Language Learners at a Community College.” In Linguistic Minority Students Go to College: Preparation, Access, and Persistence, ed. Y. Kanno and L. Harklau (New York: Routledge). Bailey, T., D. W Jeong, and S. W. Cho. 2010. “Referral, Enrollment, and Completion in Developmental Education Sequences in Community Colleges.” Economics of Education Review 29 (2): 255–70. Bailey, Thomas and E. Weininger. 2002. “Performance, Graduation, and Transfer Rates of Immigrants and Natives at City University of New York Community Colleges.” Educational Evaluation and Policy Analysis 24 (4) 359–77. Betts, Julian, Sam M. Young, Andrew C. Zau, and Karen Volz Bachofer. 2016. College Prep for All: Will San Diego Students Meet Challenging New Graduation Requirements? Public Policy Institute of California. Bickerstaff, Susan, and Julia Raufman. 2017. From "Additive" to "Integrative": Experiences of Faculty Teaching Developmental Integrated Reading and Writing Courses. Columbia University Teachers College, Community College Research Center. Bleakley, H. and Chin. A. 2004. “Language Skills and Earnings: Evidence from Childhood Immigrants.” Review of Economics and Statistics 86 (2): 481-496. Bunch, George, C. 2008. “Language Minority Students in California Community Colleges: Current Issues and Future Directions.” C4 eJournal, Issue 1. UC Riverside: California Community College Collaborative (C4). Bunch, George C., Ann Endris, Dora Panayotova, Michelle Romero, and Lorena Llosa. 2011. Mapping the Terrain: Language Testing and Placement for US-Educated Language Minority Students in California’s Community Colleges. Report prepared for the William and Flora Hewlett Foundation. Bunch, George and Ann Endris. 2012. Navigating “Open Access” Community Colleges. In Linguistic Minority Students Go to College: Preparation, Access, and Persistence, ed. Y. Kanno and L. Harklau (New York: Routledge). California Community Colleges Academic Senate. 2018. An Introduction to Guided Self-Placement (GSP). California Community College Chancellor’s Office. California Community Colleges Chancellor’s Office. 2018a. Statewide Student Success Scorecard, 2018. California Community College Chancellor’s Office, accessed January 15, 2019. California Community Colleges Chancellor’s Office. 2018b. Chancellor’s Office MIS Basic Skills Cohort Tracker. California Community College Chancellor’s Office. California Community Colleges Chancellor’s Office. 2018c. “Frequently Asked Questions about AB 705 for Credit English as a Second Language (ESL).” California Community College Chancellor’s Office. California Community Colleges Chancellor’s Office. 2018d. “AB 705 Credit ESL Implementation Timeline.” California Community College Chancellor’s Office. California Legislative Information. 2017. “AB-705 Seymour-Campbell Student Success Act of 2012.” California Legislative Information. Chiswick, B. R., and P. Miller. 1995. “The Endogeneity between Language and Earnings: International Analyses.” Journal of Labor Economics 13 (2): 246–88. Cuellar Mejia, Marisol, Olga Rodriguez, Hans Johnson, and Bonnie Brooks. 2018. Reforming English Pathways in California’s Community Colleges. Public Policy Institute of California. Gonzalez, A. 2000. “The Acquisition and Labor Market Value of Four English Skills: New Evidence from NALS.” Contemporary Economic Policy 18 (3): 259-269. Henson, Leslie and Katie Hern. 2018. A Seat at the Table: Supporting Student and Teacher Capacity in Corequisite English Remediation and Accelerated ESL. The California Acceleration Project. Hern, Katie. 2016. Sample Accelerated, Integrated Reading and Writing Course. Chabot College. Hodara, Michelle. 2015. “The Effects of English as a Second Language Courses on Language Minority Community Colleges Students.” Educational Evaluation and Policy Analysis, 37 (2): 1-24. Johnson, Hans, Marisol Cuellar Mejia, and Sarah Bohn. 2018. Higher Education as a Driver of Economic Mobility. Public Policy Institute of California. Kuehner, Alison V. and Jennifer Hurley. Forthcoming. “How Integrating Reading and Writing Supports Student Success.” Working Paper. Fremont, CA: Ohlone College. PPIC.ORG English as a Second Language in California’s Community Colleges 30 Merriam, Sharan. 1998. Qualitative Research and Case Study Applications in Education. (San Francisco: Jossey-Bass). Multiple Measures Assessment Project. 2017. Multiple Measures High School Variables Model Summary, Phase 2—Revised February 2017. Research and Planning Group for California Community Colleges. Multiple Measures Assessment Project. 2018. AB 705 Implementation Survey: ESL Focus. Research and Planning Group for California Community Colleges. Park, Elizabeth. Forthcoming. “Examining Community College Students’ Progression through the English as a Second Language Sequence.” University of Southern California. Perez, Alice. 2019. “Assembly Bill 705 and 1805 Spring 2019 Guidance Language for Credit English as a Second Language (ESL).” California Community College Chancellor’s Office. Perez, Alice, and John Stanskas. 2018. “Assembly Bill 705 Initial Guidance Language for Credit English as a Second Language.” California Community College Chancellor’s Office. Rivera-Batiz, F. L. 1990. “English Language Proficiency and the Economic Progress of Immigrants.” Economics Letters 34(3): 295–300. Rodriguez, Olga. 2013. “Examining the Effects of Academic English as a Second Language Pathways at the Community College: A Mixed Methods Study.” Doctoral Dissertation, Columbia University. Rodriguez, Olga, Marisol Cuellar Mejia, and Hans Johnson. 2016. Determining College Readiness in California’s Community Colleges: A Survey of Assessment and Placement Policies. Public Policy Institute of California. Rodriguez, Olga, Marisol Cuellar Mejia, and Hans Johnson. 2018. Remedial Education Reforms at Community Colleges: Early Evidence on Placement and Curricular Reforms. Public Policy Institute of California. Rodriguez, Olga, Hans Johnson, Marisol Cuellar Mejia, and Bonnie Brooks. 2017. Reforming Math Pathways at California’s Community Colleges. Public Policy Institute of California. Suárez-Orozco, Carola, Marcelo Suárez-Orozco, and Irina Todorova. 2008. Learning a New land: Immigrant Students in American Society. (Boston: Harvard University Press). Toth, C. 2018. “Directed Self-Placement at ‘Democracy’s Open Door’: Writing Placement and Social Justice in Community Colleges.” In Writing Assessment, Social Justice, and the Advancement of Opportunity, ed. Mya Poe, Asao B. Inoue, and Norbert Elliot. (Fort Collins, CO: University Press of Colorado). Willett, Terrence. 2017. English as a Second Language Multiple Measure Development. Research and Planning Group for California Community Colleges. PPIC.ORG English as a Second Language in California’s Community Colleges 31 ABOUT THE AUTHORS Olga Rodriguez is a research fellow at the PPIC Higher Education Center. She conducts research on the impact of programs and policies on student outcomes, with a particular focus on college access and success among underserved students. Her recent research focuses on statewide developmental education reform, assessment and placement systems, and place-based efforts to help students get into and through college. Before joining PPIC, she was a postdoctoral research associate at the Community College Research Center at Teachers College, Columbia University. She holds a PhD in economics and education from Columbia University. Laura Hill is a policy director and senior fellow at the Public Policy Institute of California. Her areas of expertise are K–12 education and immigration. She is currently researching English Learners in California schools, special education, and ESL students in community colleges. Her recent publications examine student achievement on the state’s K–12 assessments and the link between language reclassification policies and student success. Prior to joining PPIC, she was a research associate at the SPHERE Institute and a National Institute of Aging postdoctoral fellow. She holds a PhD in demography and an MA in economics from the University of California, Berkeley. Sarah Bohn is director of research and a senior fellow at the Public Policy Institute of California, where she holds the John and Louise Bryson Chair in Policy Research. She is also a member of the PPIC Higher Education Center. As director of research, she works with PPIC staff to bring high-quality, nonpartisan research to important policy issues in California. Her own research focuses on the role of social safety net policy and education policy in alleviating poverty and enhancing economic mobility. Her other areas of expertise include immigration policy, the workforce skills gap, and California’s economy. Her work has been published in major academic journals, including the American Economic Review, Demography, American Economic Journal: Economic Policy, and The Review of Economics and Statistics. She holds a PhD in economics from the University of Maryland, College Park Bonnie Brooks is a research associate at the PPIC Higher Education Center. Her work focuses on developmental education and career technical education at community colleges. Before joining PPIC, she was a data and research associate at the Los Angeles Promise Zone Initiative, where she measured the initiative’s impact on areas like public safety and education in local neighborhoods. She also worked at Grinnell College’s Data Analysis and Social Inquiry Lab, where she helped students and faculty integrate data into research and classroom work. She holds a BA in economics with a concentration in global development studies from Grinnell College. ACKNOWLEDGMENTS The authors would like to thank the community college ESL faculty who generously shared their time and valuable insights with us. We also thank faculty leaders from CATESOL and the Academic Senate who helped us better understand the complexity of ESL sequences and course data. We are especially thankful to Kathy Wada, Nikki Edgecombe, Niu Gao, Marisol Cuellar Mejia, and Lynette Ubois who provided valuable comments and reviews of earlier drafts. We also appreciate the critical editorial support provided by Mary Severance and government affairs support provided by Kellie Longo Flores and Deborah Gonzalez. This research would not have been possible without the data provided by the California Community Colleges Chancellor’s Office. Any errors are our own. PPIC.ORG English as a Second Language in California’s Community Colleges 32 PUBLIC POLICY INSTITUTE OF CALIFORNIA Board of Directors Steven A. Merksamer, Chair Senior Partner Nielsen Merksamer Parrinello Gross & Leoni LLP Mark Baldassare President and CEO Public Policy Institute of California Ruben Barrales Senior Vice President, External Relations Wells Fargo Bank María Blanco Executive Director University of California Immigrant Legal Services Center Louise Henry Bryson Chair Emerita, Board of Trustees J. Paul Getty Trust A. Marisa Chun Partner Crowell & Moring LLP Chet Hewitt President and CEO Sierra Health Foundation Phil Isenberg Former Chair Delta Stewardship Council Donna Lucas Chief Executive Officer Lucas Public Affairs Mas Masumoto Author and Farmer Leon E. Panetta Chairman The Panetta Institute for Public Policy Gerald L. Parsky Chairman Aurora Capital Group Kim Polese Chairman ClearStreet, Inc. Gaddi H. Vasquez Retired Senior Vice President, Government Affairs Edison International Southern California Edison PPIC.ORG English as a Second Language in California’s Community Colleges 33 The Public Policy Institute of California is dedicated to informing and improving public policy in California through independent, objective, nonpartisan research. Public Policy Institute of California 500 Washington Street, Suite 600 San Francisco, CA 94111 T: 415.291.4400 F: 415.291.4401 PPIC.ORG PPIC Sacramento Center Senator Office Building 1121 L Street, Suite 801 Sacramento, CA 95814 T: 916.440.1120 F: 916.440.1121" ["_edit_lock"]=> string(12) "1557248182:1" ["post_date_gmt"]=> string(19) "2019-04-29 17:56:27" ["comment_status"]=> string(6) "closed" ["ping_status"]=> string(6) "closed" ["post_password"]=> string(0) "" ["post_name"]=> string(64) "english-as-a-second-language-in-californias-community-colleges-2" ["to_ping"]=> string(0) "" ["pinged"]=> string(0) "" ["post_modified"]=> string(19) "2019-04-29 10:56:49" ["post_modified_gmt"]=> string(19) "2019-04-29 17:56:49" ["post_content_filtered"]=> string(0) "" ["guid"]=> string(106) "https://www.ppic.org/wp-content/uploads/english-as-a-second-language-in-californias-community-colleges.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) }