Course taking patterns of latino ESL students mobility and mainstreaming in urban community colleges in the united states

Course-Taking Patterns of Latino ESL Students: Mobility and Mainstreaming in Urban Community Colleges in the United StatesARIA RAZFAR University of Illinois Chicago, Illinois, United Sta

Course-Taking Patterns of Latino ESL Students: Mobility and Mainstreaming in Urban Community Colleges in the United States

ARIA RAZFAR

University of Illinois

Chicago, Illinois, United States

JENNY SIMON

El Camino Community College

Torrance, California, United States

In most Western countries where English is the medium of instruction,there is a substantial gap in student success between immigrant English

as a second language (ESL) students and non-ESL students In theUnited States, this situation has been observed in particular with LatinoESL students This article describes a longitudinal study of two cohorts

of Latino ESL students and compares the success of students whomainstreamed into college-level content courses and those who did not.More specifically, drawing on quantitative transcript analysis and focusgroup discussions, this study examines several factors impacting themobility of Latino ESL students in a large urban community collegedistrict in the United States The qualitative analysis focused on severalthemes including challenges to navigating the curriculum, thesignificant role of ESL in providing opportunities to use English, andthe supportive role of instructors The quantitative analysis focused onmainstreaming, enrollment patterns, and success measures, includinggrade point average (GPA) and course-completion ratio The findingssuggest that students who mainstream earlier or concurrently enroll incontent level courses are more successful in terms of course completionand GPA Implications of the study are discussed in relation toplacement, instruction, and further areas of research Although theESL programs and the linguistic-minority population of this study arelocated in the United States, the issues raised and lessons learned canenrich the broader international conversation surrounding languageminority education

doi: 10.5054/tq.2011.268060

Co-Authors listed in alphabetical order.

T he American community college serves as an important entry pointfor millions of students who would otherwise not have had thechance to go to college This is the case even more so for groups thathave not traditionally attended college, including minority groups such

as African-Americans, Latinos, immigrants, and working-class students.The modern American community college is characterized by its lowtuition, open enrollment, and multifaceted mission, includingpreparation for transfer to a university, vocational training, and basicskills instruction (Cohen & Brawer, 2003)

The state of California contains the most institutions of highereducation of any state in the United States, including the mostcommunity colleges, currently with 110 of these institutions The studydescribed here draws on transcript and demographic data from one ofthe largest urban, Latino districts in the state Given the importance ofthe community college for Latino immigrant students, this study sought

to understand the role of credit English as a second language (ESL) inthe retention, persistence, and success of Latino immigrant students.The focus on credit ESL (that is, courses for which students receiveacademic credit), rather than non-credit ESL, stems from prevailingpatterns of transfer to the university system Noncredit ESL programs,although extremely important in providing survival skills in Americansociety, are not specifically designed to provide students with the toolsnecessary to succeed academically in regular higher education pro-grams

Given the dearth of research on Latino ESL course-taking patterns incommunity college credit ESL programs, the research described in thisarticle was an exploratory, descriptive, and longitudinal study of twocohorts of Latino ESL students using student transcripts obtaineddirectly from the district; in addition, focus group discussions of self-selected participants were conducted to further enrich the findings Thisarticle examines the broader historical context for Latino ESLeducational trajectories, reviews factors impacting Latino ESL success,including the critical role of English language proficiency, providesquantitative analysis of transcript data and qualitative analysis of focusgroup discussions, and concludes with a discussion of implications forplacement, instruction, and further research The questions guiding thisstudy are the following:

1 How do the Latino ESL students begin their educational trajectory?

2 Do the Latino ESL students succeed in their goals? How do they meetthese goals (i.e., what enrollment patterns are evident)? What factorsaffect students’ progress?

3 How do these enrollment patterns affect the students’ level of success?

HISTORICAL CONTEXT: LATINOS AND THE

EDUCATIONAL PIPELINE

The California postsecondary system, otherwise known as theCalifornia Master Plan for Higher Education, is a three-tier system —University of California (UC), California State University (CSU), andcommunity colleges—and is amongst the most robust and coherentsystems of higher education in the world intended to provide equalaccess to all.1 The UC serves as the primary research institution for thestate and selects the top eighth of the graduating high-school seniors,and the CSU selects the top third, whereas the community colleges areopen access for any student who can benefit from them In other words,the community colleges operate as open admissions institutions Latinosare the fastest growing demographic represented in the precollege (i.e.,grades K–12) setting, constituted nearly 50% of the K–12 enrollment in

2005, and are projected to surpass that mark in the next decade(California Department of Finance, 2000) However, they are signifi-cantly underrepresented at all levels of higher education (Ornelas &Solorzano, 2004) With most Latinos concentrated in the communitycollege (California Community Colleges Chancellor’s Office, 2007), it isimportant to examine how this group of students proceeds and advancesthrough the community college as it relates to the transfer function ofthe community colleges, because these colleges serve as the principalgateway to success in higher education at both the undergraduate andpostgraduate levels Although a large percentage of students aspire totransfer, only a handful actually does (Rivas, Pe´rez, A´ lvarez, & Solo´rzano,2007) According to California Postsecondary Education Commission(CPEC) data (2004), only 7 out of 100 first-time Latino college studentstransfer to either a CSU (6) or a UC (1) Although every stage of thepipeline (i.e., the way students proceed through the curriculum) iscritical to examining questions of success and achievement, and thefactors impacting persistence are multiple and complex, the focus of thisanalysis is on where most Latino students begin their higher education:

at the community college

THE FIRST STOP IN THE COLLEGE PIPELINE: THE ROLE

OF CREDIT ESL AT THE COMMUNITY COLLEGE

As the first stop in the pipeline, which for only a minority of studentsends with graduation from a university, community college credit ESLcourses play the unenviable role of gatekeeper to the rest of the

1 See University of California Office of the President (2009) for more information on the California Master Plan for higher education.

curriculum It is assumed that English language learners need the skillsand knowledge attained in credit ESL courses in order to succeed in therest of the curriculum and overcome all the hurdles in the pipeline.Thus, students take these courses to prepare for regular college courses.After credit ESL, students must pass English 101 (or first yearcomposition), college-level math, as well as other lower-division contentcourses in order to prepare for the next hurdle: transferring to auniversity Thus, credit ESL plays a major role in leading students towardsuccess in college Given the importance of these courses, there is atremendous need to better understand the role of credit ESL inpreparing students for success in higher education and beyond(Crandall & Sheppard, 2004; Kuo, 2000) This issue is particularlyproblematic because of the diversity of students who enroll in ESLcourses ESL may play differing roles in a student’s educationaltrajectory, depending on the educational background of the studentwho enrolls in the course If a student’s educational backgroundincludes graduation from high school and even some college, then ESLoperates primarily as a foreign language course—the student may havereading and writing skills in his or her own language but just need helpwith carrying these skills over to English; however, if a student’seducational background does not include graduation from high school(if he or she is an adult), or the student’s education was interrupted toimmigrate to the United States (generation 1.5), then ESL might playthe dual role of a foreign language course and a developmental course—not only strengthening the student’s English, but also his or heracademic skills, mostly in the areas of reading and writing (Ignash, 1996;Mellissinos, 1993) Thus, ESL instructors have the particularly difficultjob of figuring out how to meet the needs of all their students, whenstudents of such diverse backgrounds may be contained in one class.This goes against the common assumption that ESL students are amonolithic population; instead, ESL students are a large, hugely diversepopulation in terms of educational background and economic andsocial status, among other factors (Echevarria, Short, & Vogt, 2004;Gerardi, 1996).

These days, ESL is starting to be recognized for its special role inserving such a diverse population of students and is considered a specialtype of developmental program In the past, ESL has been included withforeign language courses (Ignash, 1994; Mellissinos, 1993); however,ESL has slowly come to be recognized for the very large role it plays notonly in teaching English, but also in improving the academic skills ofimmigrant students For example, California’s Basic Skills Initiative(Center for Student Success and the RP Group, 2007), a new initiative toinform California community colleges of best practices in teaching andprovide support and funding for community colleges to adopt these best

practices on their campuses, includes ESL under basic skills, with thiscaveat:

The inclusion of English as a Second Language in this definition recognizesthat all ESL is not, by definition, subsumed under basic skills To the extentthat a student is unable to succeed in college-level coursework due to inability

to speak, read, write or comprehend English, ESL skills may be considered asfoundation skills in accordance with the definition (p 4 and see footnote 2)

Thus, this perspective acknowledges that whether ESL is a ‘‘basic skill’’depends on the ability and background of individual students Thisreport goes on to recognize the dual role of ESL, in that instructors notonly ‘‘[assist] in English language acquisition’’ but also ‘‘[teach] basicliteracy skills’’ (p 47) Other authors have also recognized the role ofESL in teaching students how to effectively navigate the academicenvironment (Gerardi, 1996; Kuo, 2000), thus adding an additionallayer to the complexity of ESL instruction It stands to reason, then, thatESL students with weak educational backgrounds, as many Latinostudents have, will struggle more than those with stronger educationalbackgrounds.2 Thus, as most community college programs are set upnow, student success in ESL is necessary for success in other parts of thecurriculum

MAIN FACTORS IMPACTING SUCCESS OF LATINO

STUDENTS THROUGHOUT THE PIPELINE

Researchers have discussed many factors impacting Latino studentsuccess in higher education, including: immigration status, poverty,work, first-generation college status, and lack of academic preparation

Of particular relevance to the arguments being made in this article arethe issues of language background, curricular tracking, types of coursestaken, and the kinds of learning opportunities that exist in theclassrooms (although not specifically addressed by the analysis in thisarticle) One study concerned with these issues, the Transfer, Retention

in Urban Community Colleges study (Hagedorn, 2006) surveyed over2,400 Latino community college students across a large urban districtand provides a useful framework for understanding the broader context

2 For the purpose of this article, we define weak and strong educational backgrounds in relation to those measures related to achievement in schools or standardized testing either

in the United States or abroad Thus, ESL students with strong academic backgrounds and preparation have a demonstrated record of achievement (e.g., already have a diploma, Bachelors, or graduate degree in their native language) In contrast, weak students have minimal record of achievement in formal schooling, usually place in the lower tier of standardized measures, and may have a history of dropping out or never attending school.

of Latino course-taking behavior That study, although not focusingspecifically on Latinos starting their education in credit ESL as this studydid, drew their sample from the same population It found that whereasLatinos had the same aspirations for academic success as other groups, theywere less likely to have taken college Algebra, trigonometry, precalculus,physics, or chemistry in high school or while in college; in addition, theywere less likely to enroll in college level English courses Thus, thetraditional gatekeeper courses of Math and English, which are the bestpredictors of persistence and transfer, need to be examined closely inrelation to the curriculum objectives and student course-taking practices.

In examining this issue in relation to the questions posed by thisstudy, the most salient factors impacting persistence and retentionthrough the education pipeline are the following: non-English speakingbackground, (i.e., Spanish; Datnow, Borman, Stringfield, Overman, &Castellano, 2003; Ovando & McLaren, 2000; Post, 1990); curriculartracking (Oakes & Guiton, 1995; Padilla & Gonzalez, 2001;Yonezawa,Wells, & Serna, 2002); academic English language proficiency (Razfar,2003; Rendon & Hope, 1996); the types of courses that Latinos take(remedial versus college prep; vocational versus academic; Oliva, 1999);and, finally, the type of pedagogy and learning opportunities available inthose courses (Gutie´rrez, 2002; Losey, 1995) These studies show that theplight of Latinos in the educational pipeline has been well documented.According to Ga´ndara & Contreras (2009):

Latinos for the most part are now stalled at the level of high schoolcompletion, with dropout rates remaining very high across generations Onlyone in ten Latinos has a college degree, compared to more than one in fourwhite Americans and more than one in three Asians (p 5)

Research has shown that Latinos are in the lowest quartile of academicachievement in reading and math This gap clearly puts Latinos in avulnerable position as they pursue higher education and precludes themfrom entering the top-tier four-year institutions Although most starttheir higher education pursuits in the community college, the over-whelming majority never transfers, and many Latino students findthemselves increasingly segregated from others, in part due to limitedEnglish use (Fry, 2008; Fry & Gonza´lez, 2008)

It is important to note that deficit views (negative views of bilingualism,ethnic minorities, etc.) and a history of marginalization within schools ingeneral have seriously impacted Latino success across the pipeline.Although all these factors are deeply interdependent and difficult toisolate, for ESL planners, language proficiency, especially the ability toparticipate in academic discourse, is the most salient domain (Razfar,2003) From a curricular point of view, this is represented in practical

terms by enrollment and success in early college level English courses(such as the mandatory first year composition course), which serve as aprerequisite for almost all other academic coursework However, giventhe importance of language in all fields, ESL programs are not onlypreparing students for transfer level English, per se, but for all contentareas that are dependent on academic language proficiency Thus,language skills cannot be reduced to literary and grammatical functionsbut should be viewed holistically as the primary meaning making toolthat gives students access to higher-order learning opportunities(Darder, Torres, & Gutie´rrez, 1997; Garcı´a, 2001; Ovando & McLaren,2000; Valencia, 2002; Valverde, 2007).

As previously stated, this research was guided by three majorquestions In addition to these questions, there are more specificsubquestions related to the data listed below:

1 How do the students begin their educational trajectory?

a What are their educational goals?

b What is their initial ESL level?

c What is their educational background?

2 Do the students succeed in their goals? How do they meet these goals(i.e., what enrollment patterns are evident)? What factors affectstudents’ progress?

a Are students enrolled long enough to succeed in their goals?

b At what rate did students finish the credit ESL sequence?

c For those students whose goals require courses beyond creditESL courses, do these students mainstream?

d For those students whose goals require courses beyond creditESL courses, do they concurrently enroll in these courses whilestill taking ESL or after completing ESL?

e Did students’ initial ESL level or educational background affecttheir enrollment patterns?

f Did these students enroll in and pass freshman composition andcollege-level math (i.e., the most important courses required fortransfer)?

3 How do these enrollment patterns affect the students’ level of success?

a Was there a difference in grade point average (GPA), completion ratio, or success in ESL courses based on students’enrollment patterns (i.e., nonmainstreamers, concurrent main-streamers, or linear mainstreamers)?

course-METHODOLOGY: CONTEXT AND DESCRIPTION

Context: The District and the Latino ESL Students

The Latino ESL students in this study were enrolled in the largesturban community college district in the state of California, which coversnearly half of a major California county Drawing on Census 2000 data,

of the nearly ten million Latinos who lived in the county, nearly half(4,729,661) lived within a 10-mile radius of the district’s communitycolleges, with the total adult population (ages 18–64 years) reported to

be 2,953,365 (62%) Nearly a quarter of the adult population wide was reported to not speak English well or not at all (664,306, or22.5%) The majority (about 58%) of this population was concentrated

district-in the downtown area served by three of the district’s communitycolleges With respect to the ESL cohorts of this study, Latinos made up40.9% of the district population They tended to be older than otherethnic groups in the district, with an average age of 32 years, and alsowere largely female (66.7% female; 33.3% male)

The ESL Pipeline in the District

The ESL pipeline in the district is relatively long compared to mostdistricts Based on an analysis of the curriculum across the campuses ofthe district, there are seven levels of ESL throughout the district (with onlytwo campuses of the nine offering all seven levels, most campuses offeringfive or six levels, and one offering only four levels at the highest part of theprogression) The ESL programs in the district all offer a requiredsequence of writing courses as well as recommended sequences oflistening or speaking courses and reading courses Some offer optionalcontent-based courses as well, such as English for Business After finishingESL classes, the students then have to take at least one developmentalEnglish class before enrolling in the first college-level English course.The ESL programs at each college in the district differ in terms ofboth their numbering system as well as the prerequisite(s) that the lastclasses in the ESL series satisfy At some colleges, the last class in theseries satisfies the prerequisite for the developmental class that is onelevel below transfer; at others, the last class in the series satisfies theprerequisite for the developmental class that is two levels below transfer.When coding the data for this study, the ESL levels were determinedbased on their status as a prerequisite for these developmental classes.This status was determined by looking at both the college catalogs andthe transcript data Thus, for ESL students it was mandatory that theytake at least one precollegiate developmental class before enrolling inthe mainstream freshman (first year) composition course required of all

students In contrast, for non-ESL students in the district, an Englishplacement assessment was used to determine whether they have to takethe precollegiate class before taking the freshman composition course.Enrollment in a precollegiate course was not considered mainstreamingfor the purpose of this study.

Description of the Data and the Participants

The data used in this study were drawn from the enrollment anddemographic database maintained by the district’s office of institutionalresearch Upon collaboration with the district’s institutional researchoffice, a student was classified as an ESL student if he or she enrolledand received a grade or a W (withdrawal from course) in at least onecredit ESL course during his or her academic career in the district.Students were then assigned to fall and spring cohorts based on theirfirst credit enrollments in either the fall or spring terms Based on thisprocedure, there were 80,923, ESL students district-wide between Fall

1992 and Spring 2004, who were assigned to either fall or spring cohorts.Students entering in fall 1999 or fall 2000 were selected for inclusion inthis study This group was further narrowed by selecting only the Latinostudents, and only those students who had enrolled in a credit ESL class atany time during their enrollment in the district The data were selected inFall 2005, and at that point the study participants had been in the districtlong enough to either complete their education or, in some other way,stop their enrollment in the district The reason that pre-1999 cohortswere not selected was because of concerns that we would not be able tofind any students for our qualitative follow-up study who were also part ofthe quantitative transcript study As a follow-up to the transcript study, asmall group of students was selected to take part in focus groups.Therefore, it was possible that some of the students who took part in thefocus groups were also a part of our transcript study, or had at least gonethrough the same curriculum that we were examining in our study

Procedures

Although the principal method used for this study is quantitative, wehave also incorporated qualitative findings as a way to raise questions forfuture research and to contextualize the limitations of relying only ontranscript data Given the ex-post facto nature of the data analysis, wecould not incorporate a more robust qualitative design (i.e., classroomobservations, interviews with more participants, etc.) Mixed-methodsapproaches are based on the assumption that quantitative andqualitative methods are compatible, and the use of both enhances

validation of the findings More important, a mixed-methods approachprovides nuanced detail for an understudied phenomenon, hence raisingissues for further exploration Mixed methodologies also allow for greaterbreadth of analysis and triangulation of findings using multiple sources ofdata (Brannen, 2005; Creswell, 2003; Greene, Caracelli, Valerie, &Graham, 1989; Johnson & Onwuegbuzie, 2004) In this study, we clearlygave more prominence to the quantitative findings, but the qualitativefindings that emerged from focus groups allowed us to name issues forfuture research Thus, the quantitative methodology was more ‘‘domi-nant’’ and the qualitative findings are ‘‘less dominant,’’ leading to a

‘‘more dominant, less dominant design.’’ (Creswell, 1994, pp 177-178).Considering ESL students were previously invisible in the districtdatabase, the focus group themes provide a human face to thequantitative analysis

Quantitative Analysis

In this section, the definitions of key variables and constructs for thequantitative analysis are provided, followed by a description of thequalitative methods used

The most salient variable for the purposes of this study wasMainstreaming A student had mainstreamed if he or she had enteredthe regular college curriculum—that is, the student had taken non-ESLand nondevelopmental courses These courses could be eithervocational or academic courses Based on the tenets of content-basedinstruction and content-based college ESL instruction (Kasper, 2000;Spurling, Seymour, & Chisman, 2008), mainstreaming was disaggregatedfurther, based on two possible patterns of mainstreaming in relation tonon-ESL and core subject areas: linear and concurrent mainstreamers.Thus, the following three categories were examined longitudinally:

N Linear mainstreamers: students who enrolled in a college-levelcourse for the first time after completing ESL courses

N Concurrent mainstreamers: students who enrolled in a college-levelcourse for the first time while concurrently enrolled in an ESL course

N Nonmainstreamer: a student who never took courses outside of ESL

or developmental tracks

In addition to the Mainstreaming variable, the following variableswere also used in the study: Initial Educational Goal, EducationalBackground, Taking and Passing of Courses by Course Type, SuccessRate, Initial ESL Level, ESL Levels Progressed, Mainstreaming, OverallGPA, and Overall Course Completion Rate (definitions below):

N Initial Educational Goal: The goal declared by the student uponentering the district The categories were: career-related, certificate/

Associate of Arts (AA), transfer, basic skills/General EquivalencyDiploma (GED), personal development, and unknown/undecided(see Table 1).

N Educational Background: The highest level of education the studentcompleted upon entering the district We examined the followinglevels: did not complete high school; graduated from high school;additional education past high school

N Taking of Courses (by course type): Whether a student took a course of

a particular type The course types were: college-level English man English), ESL, transfer-level math, science, social science, andvocational This variable is dichotomous: Either the student took acourse of that type or did not

(fresh-N Passing of Courses (by course type): Whether a student passed a course

of a particular type (course types are listed above) This variable isdichotomous and determined by whether a student had passed onecourse of that type This calculation was only performed for studentswho had actually taken that course type In our analysis, this variablewas used only for college-level English, because a student could havetaken college-level English several times and passed it once For othercourse types, the next variable was used It is a more accurate way ofdetermining success in courses, because it takes into account the factthat a student might have taken many courses of that type

N Success Rate (by course type): The number of times a student passedcourses of a certain type divided by the number of times a studenttook courses of a certain type

N Initial ESL Level: The first ESL level that a student enrolled in

N ESL progress: The number of levels that a student progressed inESL, not including the first level So, if a student took and passed alevel, but did not go to the next level, ESL progress equaled zero

N Overall GPA: total number of grade points divided by the totalnumber of units completed

N Overall Course Completion Ratio: Total number of units completeddivided by the total number of units attempted Units attempted isdefined by having earned either a W (withdraw) or a grade in thecourse

All variables were calculated based on a student’s cumulativeacademic record This method allowed for longitudinal analysis ofLatino ESL educational trajectories, which few studies have conducted.Some studies have examined the course-taking behavior of ESL anddevelopmental students in the community college (e.g., Illich, Hagen, &McCallister, 2004; Weissman, Bulakowski, & Jumisko, 1997; Weissman,Silk, & Bulakowski, 1997) However, many of these studies were limited

to course-taking patterns within a single semester Although they are

valuable in providing a snapshot of student goals and course-takingbehaviors, longitudinal studies can help improve understanding of goalattainment over time Longitudinal studies have added value, in thatthey track the same students over several semesters, rather than differentstudents in different semesters.

Drawing on Kirk (1995), it was determined that independent sample tests were most appropriate for comparing the two groups in the study;furthermore, because it was decided to further break down mainstreamersinto concurrent mainstreamers (n 5 628) and linear mainstreamers (n 5 75) and

t-to conduct multiple comparisons, Bonferroni corrections were applied inorder to make adjustments to the Type I error rate In addition, analysis ofvariance (ANOVA) analysis was conducted to further validate the results.Because the difference in the variance was not significant, there was no illeffect due to the difference in the sample size of each group

Qualitative Data Collection and Analysis

Focus groups discussions were conducted with students who hadprimarily enrolled at the largest community college within the district.Given our focus on Latino students, this college also had the largestpercentage of Latino students (over 65%) The ESL coordinator andinstructors were asked to solicit volunteers to participate in semistruc-tured, focus-group discussions about ESL and broader communitycollege issues Based on the number of volunteers, three focus groupsessions were conducted with up to five participants in each Thesesessions took place in reserved library study rooms on the campus wherethe students were enrolled They were an hour long, audio-recorded,and conducted by the authors of this article These sessions weresemistructured with guiding questions (see Appendix A) designed toprovide rich narratives about course taking, persistence, and success inthe community college Although we had guiding questions, the sessionswere purposefully open-ended and designed to optimize the collection

of narratives Of the 10 participants, 3 were Latinos (all 3 were female)and 7 were Asian (6 females, 1 male) The three female participantsMaria, Rosa, and Lupe (pseudonyms) were the first generation ofcollege students in their respective families Maria and Rosa were both

18 years old and immigrated to the United States while in middle school.Lupe was a middle-aged adult who immigrated at the age of 18 years withtwo adult children and decided to attend college after her children hadgrown up She completed high school in her native country of Mexico

In conducting and later analyzing the focus group transcripts, agrounded, qualitative approach was used With the use of the constantcomparative method (Denzin & Lincoln, 2000; Glaser & Strauss, 1967), thefocus group transcripts were analyzed and coded for the purpose ofdeveloping emergent themes relevant to the study’s focus on Latino ESL

course-taking practices The goal was to further elucidate the results onLatino ESL educational trajectories with students who had actuallyparticipated in the ESL pipeline at one of the community colleges in thestudy The selection of students for the focus groups came entirely from themainstream group, because they were the more accessible population.There were many institutional and logistical constraints to tracking downnonmainstreamers Despite the constraints, efforts were made to contactnonmainstreamers through e-mail and telephone calls; however, theseefforts were unsuccessful Drawing on NVivo 8.0 qualitative software(produced by QSR International www.qsrinternational.com), we calculatedthe length of the narratives (lines and word counts) and determined apercentage in relation to the total amount of focus group text Bothquantitative and qualitative results are discussed in the next section.

RESULTS

Quantitative Findings: Transcript Analysis

Drawing on the student demographic data and enrollment records,the first set of the quantitative findings corresponds to the first researchquestion: How do Latino ESL students begin their educational trajectory? Datainclude initial educational goal (Table 1), educational background(Table 2), and initial ESL level (Table 3) The second set of thequantitative findings responds to these related research questions: Do thestudents succeed in their goals, how do they meet their goals, and what are thefactors that impede their progress? Data include length of enrollment in thedistrict (Table 4), rate of completion of ESL course sequence (described

in the text), mainstreaming categories (Table 5), mainstreamingcategories by initial educational goal (Table 6), enrollment patterns bymainstreaming category (Table 7), and enrollment in and passing offreshman composition and college-level math (described in the text)

TABLE 1

Educational Goals of Latino ESL Students

Note ESL 5 English as a second language; BA 5 Bachelor of Arts; AA 5 Associate of Arts; GED

5 General Education Diploma; NA 5 not available.

Finally, the third set of the quantitative findings answers the thirdresearch question: How do these enrollment patterns affect students’ level ofsuccess? This section examines academic success as measured by GPA,course completion ratio, and ESL success rate (Table 8).

Initial Educational Goals, Background, and ESL Level

Starting with the first research question, asking how students begintheir educational trajectory, their self-reported educational goal uponentry is examined Upon entry into the community college, students areasked to complete a questionnaire whereby they state their educationalgoal in terms of the following categories: 1) transfer to a bachelor’sgranting institution, 2) completion of an AA degree or certificate, 3)career-related, 4) basic skills or high-school equivalency diploma (GED),5) personal development, or 6) undecided Based on the data, the vastmajority of the Latino ESL students who started in fall 1999 and 2000indicated initial educational goals that were either vocational, basicskills, or related to personal development (more than 70%) Hence, veryfew indicated that they planned to transfer to a 4-year university (8.3%)

or pursue an AA degree/certificate3 (6.8%; Table 1)

TABLE 2

Educational Background of Latino ESL Students (n 5 1,420)

Additional education past

high school

TABLE 3

Initial ESL Level of Latino ESL Students

The two most commonly selected options—career-related and basicskills/GED—constitute nearly 67% of the total, suggesting that most ofthese students had a fundamentally practical orientation or outlook ontheir college level study.

In terms of educational background, the highest level of education ofmost of the students was a high school diploma (59.9%), and very few(4.0%) had completed any formal education beyond high school(Table 2)

In terms of initial ESL level, most students started in the intermediatelevels 3, 4, or 5 (60.5%), and few students started off at either the highest

or the lowest levels, as shown in Table 3

Student Success: Persistence and Mainstreaming

The second research question examined the relationship between thestudents’ initial stated goals and actual outcomes in terms of length ofenrollment, persistence, and mainstreaming The majority of thestudents were enrolled in the district for no more than two semesters(62.1%; Table 4) After three semesters, almost three-fourths of thestudents had stopped enrolling, and by five semesters, over 85% hadstopped It is possible that many could have achieved their stated goalswithin this amount of time However, in most cases, it takes much longerthan 2 years to complete a certificate, degree, or any significant

TABLE 4

Number of Semesters Enrolled (n 5 1,479)

Number of semesters Number of students Percent

educational goal Although very few stated the 2-year degree, certificate,

or transfer as their educational goal (n 5 224, 15.1%), in this respect,many students did not achieve their goals

In terms of persistence in the ESL program, the majority of students(62.5%) did not advance one level beyond where they started Even ifstudents who started at the highest levels, from which it is not possible toprogress within the ESL program, are excluded, the proportion ofstudents not progressing was between 55% and 60% In other words,ESL success was limited If Latino ESL students are dropping out afteronly one or two semesters of ESL, and after one or two semesters ingeneral, then the likelihood of retention and persistence overall isgreatly reduced Furthermore, the lower the initial level of the student,the less likely the student was to make it through the whole ESL series

In terms of the mainstreaming patterns that were exhibited, moststudents (57.5%) in the sample were nonmainstreamers (n 5 851), withmainstreamers consisting of 42.5% (n 5 628; see Table 5) As previouslystated, the mainstreamers were divided into two trajectories: concurrentmainstreamers, who constituted 88.1% of all mainstreamers (i.e., 37.4%

of the entire sample), and linear mainstreamers, who were 11.9% ofmainstreamers (i.e., 5.1% of the entire sample) These results indicate thatmost students did not mainstream Conceivably, students who declared theireducational goal to be career-related, basic skills, personal development, orunknown/undecided (84.9% of the sample), could have met their goals bytaking only ESL classes However, as we see later, large numbers of studentswhose goals could only be attained by taking courses outside of ESL (i.e.,transfer and certificate/AA students) were also not mainstreaming

In investigating the educational trajectories of students with variouseducational goals, a cross-tabulation of initial educational goal by

TABEL 6

Mainstreaming Categories by Initial Educational Goal

Educational

goal

Nonmainstreamers Mainstreamers Total

Number Percent Number Percent Number Percent