Enrollment patterns of community college students.

Abstract

The Los Angeles Community College District's population is highly saturated with Latino/Latina students. Despite high enrollment and aspirations to transfer, Latinos/Latinas transfer at rates well below national averages. Using course transcripts from the entire history of a student's enrollment, 1122 Latina students' course enrollment patterns were categorized into three separate enrollment patterns: transfer path, remedial, or occupational. Results indicate that Latinas are not enrolling in transfer-level courses at the same rate as their Caucasian peers. In addition, Latinas have lower course completion rates than Caucasians. A subsample of Latinas who are on the transfer path, however, are enrolling and completing transfer-level courses at the same rates as Caucasians.

Introduction

Community colleges across the nation are reporting high enrollment rates of underrepresented minorities, specifically Latino/Latina students. National figures indicate a disproportionate number of Latinos/Latinas concentrated in community colleges. The Latino/Latina enrollment population of 2-year colleges in the United States is 11.1% and is 6.3% at the 4-year universities (California Postsecondary Education Commission, 2000; Los Angeles Community College District, 2001). In the Los Angeles Community College District (LACCD), with a high population of Latino/Latinas in the surrounding population, 45.9% of the community population is Latino/Latina's compared to 24.5% in the California State System. Los Angeles County has a little over 9.5 million residents with 44.6% of the population reporting a Hispanic or Latino origin (Census, 2000). The LACCD resides in Los Angeles County and serves the surrounding population, which has increasing numbers of people of Hispanic or Latino heritage. It is no surprise that the LACCD has high rates of Latino/Latina enrollment. However, Latinos/Latinas are primarily entering the higher education system at community college's rather than at four-year institutions.

Women and Latinas are also disproportionately represented in community colleges. Female enrollment in the LACCD is reported at 58%, compared to 55.4% in national 4-year colleges, including state colleges and research universities (California Postsecondary Education Commission, 2000; Los Angeles Community College District, 2001). Latinas make-up approximately half of the female enrollment in the LACCD but only 32.4% of the total Los Angeles population (Census, 2002). The disproportionate representation of Latinos/Latinas in community colleges is a concern when considering the outcomes of educational attainment between ethnic groups. Latino/Latinas, despite their high enrollment rates, do not transfer to four-year institutions at the same rate as their peers. Even more perplexing is the low transfer rates despite the high aspirations of the Latino/Latina population. It is estimated that between 70-75% of Latino/Latina students aspire to a four-year degree (Nora & Rendon, 1990). Currently, estimates indicate that only 2-5% of all Latino/Latina students in the LACCD transfer to four-year institutions each year (LACCD, 2000). National figures, although difficult to estimate due to a lack of longitudinal data and a consistent definition of transfer, estimate that 19% of all students and 18.1% who enter community colleges transfer within a six-year period (Bradburn & Hurst, 2001). Community colleges are the gateway institutions for many Latino/Latina students as illustrated in the population demographics. Latinos/Latinas, however, are not transferring at a significant rate. There is little research on females and underrepresented minority transfer rates and current research focuses on transfer-as-an-outcome, ignoring the semester-by-semester course-taking process. This study takes a different view by identifying the courses in which students enroll and successfully complete on a semester-by-semester basis. Student enrollment patterns allow for an identification of student behaviors linking student aspirations to course-taking behaviors. This paper examines the course taking patterns of a group of Latino/Latina students in the LACCD by identifying the patterns of transferable level, remedial, and occupational courses and providing a complete description of how many students are on the path to transfer.

Latino/Latina Transfer Patterns in Community Colleges

Research targeting Latino/Latina community college students and their transfer rates is sparse. Only a few studies that identify factors and recommend solutions to increasing the numbers if Latinos/Latinas who transfer have been completed . Despite the few studies and the transfer mission of community colleges, the transfer rates for all students are quite low and even lower for Latinos/Latinas (Cohen & Brawer, 1996; Helm & Cohen, 2001). It is suggested that transfer rates of community college students range from 3%-20%, depending on the definition of transfer and the specific community college (Rendon & Valadez, 1993). This is especially troubling considering that community colleges are the "gateway" to higher education for many students (Nora & Rendon, 1990; Rendon & Mathews, 1989; Rendon & Valadez, 1993).

In a study that targeted a group of Latino students who intend to transfer from a community college to a four-year university, Rendon and Valadez (1993) isolated five factors influencing transfer behavior in the Latino student population: family, financial aid, knowledge regarding higher education, culture, and articulation agreements and relationships with four-year institutions (Rendon & Valadez, 1993). Latino students indicated a lack of cultural understanding with the faculty and a fear regarding cultural acceptance at four-year institutions (Rendon & Valadez, 1993). In addition, Latino students did not understand the procedures for transfer or financial resources available. Within the sample of Rendon and Valadez's (1993) study, women were more likely to cite family as factors in choosing to transfer. Supporting the research of Melendez and Petrovich (1989), females feel a strong connection to family and usually do not want to venture far from the family home. Based on their extensive research of transfer rates of Latinos in community colleges, Rendon and Taylor (1989) developed an action plan to assist Latino students transfer from two-year to four-year institutions. The ten-point plan of action calls for strong linkages with surrounding schools, both K-12 and four-year universities, involving and educating families in the higher educational process, improving student services, and increasing the quality of teaching through cultural awareness and proper assessment tools (Rendon & Taylor, 1989). Although these recommendations are helpful in developing community college support services for Latinos/Latinas, I posit that student behavior must be understood to isolate whether or not students are taking the courses necessary to transfer. Without an understanding of student course-taking behavior, student support services will not lead to successful outcomes.

Research Design

This paper describes the types of courses enrolled in by the Latina population of the Transfer and Retention of Urban Community College Students (TRUCCS) sample. The stratified random sample includes 5,000 community college students from the nine campuses of the Los Angeles Community College District. The population of this specific study will be derived from the TRUCCS sample, selecting only those students who self-identified as female and Mexican-American, Latino/Latina, or Hispanic. For comparison purposes, students who identify as Caucasian will also be selected and analyzed. Math and English courses are categorized as transferable to the University of California or California State Universities as defined by Intersegmental General Education Transfer Curriculum (IGETC) articulation agreements. The first sample of students are all the students in the TRUCCS sample, n=5,000. The second level of analysis includes those students who aspire to transfer, circumventing the students who are taking IGETC Math and English courses but who do not aspire to transfer. Course-taking patterns of Latina students in all nine LACCD campuses included in the TRUCCS sample are compared to Caucasian males and females using ANOVA's providing an extensive analysis of the differences between the groups. Frequencies and ANOVA results rely on ratios that account for the time that a student is enrolled. Many community college students "stop-out," disenroll from the community college for a few semesters and reenroll at a future date. Accounting for the number of semesters that students enroll will circumvent the large variance in student enrollment history. The ratios will also provide a metric for all comparison groups. The three ratios used in the analysis are the participation, success, and course completion ratios.

Analysis

Sample Demographics The stratified random TRUCCS sample reflects the demographics of the LACCD population, with a majority of the students being female. In 2002, the LACCD reported 60.2% of their students as female. Similarly, the TRUCCS sample is 60.9% female (n=2927) and 39.1% male (n=1883). The ethnic distribution of students in the TRUCCS sample is representative of the large urban district area. Latinos/Latinas combine for 39.3% (n=1869) of the TRUCCS sample. Females comprise the majority of the Latino population at 60% (n=1122). Within the Caucasian sample, 7.7% (n=383) of the students are female. LACCD ethnic trends are similar with 45.9% reporting as Latino.

Aspirations Degree aspirations are an integral component when analyzing course selection. Aspirations were measured on a forced choice 8 point scale. All means ranged between option 5, "At least a Bachelor's Degree, maybe more" and 6, "Master's degree." The majority of the TRUCCS students aspired to a bachelor's or master's degree. In addition, 17.8% of Caucasian students and 17.5% of Latinas indicated aspirations to the doctorate. In the sample, approximately 70-75% of Latinas and Caucasian students aspire to obtain a four-year university degree or beyond. Two-way ANOVA's were performed to test for differences in degree aspirations by gender and ethnicity. Results indicate that males (mean = 5.62; sd. = 1.44) have significantly higher aspirations than females. Ethnicity did not have a direct effect. Despite the absence of a significant interaction between gender and ethnicity, it should be noted that Caucasian females (mean = 5.46; sd. = 1.61) and Latinas (mean = 5.51; sd. = 1.52) both have lower aspirations than their male counterparts.

Course-Taking Patterns

Course Enrollment To begin understanding patterns of course enrollment and completion, it is necessary to look at representation in courses that are at the transfer-level, remedial, or occupational. The classifications of courses in the remedial and occupational categories are in accordance with the definitions provided by the LACCD. Transfer-level courses were coded using the IGETC California statewide articulation agreement between the community colleges, California State University and University of California schools. It is important to note the total number of courses enrolled and the total number of semesters of enrollment for both males and females. Females (mean = 34.35; sd. = 22.73) enroll in more courses than males (mean = 30.34; sd = 21.09). In addition, females (mean = 8.38; sd. = 5.038) have been attending college for more semesters than males (mean = 7.54; sd. = 4.633).

A participation ratio was created to account for differences grouping the number of semesters attended by students. TRUCCS began with a cross-section of college students, thus including some students who were new to the community college and others who had attended college for a long time. The participation ratio (shown below) accounts for the rates in which students participate in courses regardless of the differences in the total number of semesters. Time or number of semesters enrolled, is a crucial issue with community college students who may take only a few courses over several years, stop-out of community colleges, or have extended enrollment periods. Success ratios, described below, will account for the completion of the enrolled courses. Participation Ratio = Number of courses enrolled / Number of semesters enrolled The participation ratio for all courses results in more female (mean = 4.12; sd. = 1.55) student participation than males (mean = 4.043; sd. = 1.49), with Latinas (mean = 3.99; sd. = 1.41) reporting a slightly higher participation rate than Caucasian females (mean = 3.99; sd. = 1.66). To test for significant difference, a One-Way ANOVA was performed between Caucasian males and females, Latinos and Latinas. The participation ratio for all courses was not significant. Delineating participation ratios by transfer-level, remedial, and occupational courses provides a comparison of enrollment in various types of courses and provides a common comparison between the categories. Males (mean = .83; sd = .37) had a higher participation ratio in transfer-level courses while females (mean = .86; sd = .75) had higher concentration rates in remedial courses. Among the male sample, Latinos (mean = .76; sd. = .31) participated in fewer transfer-level courses, with greater participation in remedial courses (mean =.73; sd. = .59) as compared to Caucasian males (mean = .63; sd. = .59). Occupational course participation favored Caucasian males (mean = 1.11; sd. = .98). The participation rates between Caucasian males, Caucasian females, Latinos, and Latinas were not significant in transfer-level courses. Participation rates were significantly different in remedial courses. Tukey post hoc tests, which provide a test of pairwise differences between tested groups, resulted in a statistically significant mean differences between Latinas and Caucasian males and Latinas and Latinos. Latinas (mean = 1.30; sd. = .62) participated in significantly more remedial courses than Caucasian males (mean = 1.18; sd. = .48). In addition, Latinas (mean = 1.30; sd = .62) participated in more remedial courses than Latinos (mean = 1.21; sd = .51). Participation rates in occupational courses were also significant. Tukey's post hoc test did not find statistically significant mean differences between groups. Despite the lack of statistical significance at the p < .05 level in Tukey's test, Latinas (mean = 1.41; sd. = .70) and Caucasian females (mean = 1.54; sd. = .90) resulted in p = .057.

Success Ratio The success ratio provides an indication of the completion rates of all courses, transfer-level, remedial, and occupational courses over time. The successful course completion ratio is a convenient way to compare course-taking successes using a sample of students who have taken courses for varying time spans. The successful course completion ratio is the number of courses completed with a "C" or better grade divided by the total number of courses enrolled. The type-specific success ratios only included students who took at least one corresponding course. For example, the remedial success ratio did not include students who did not attempt at least one remedial course. Success Ratio = Number of courses (transfer-level, remedial, or occupational) completed with a passing grade / Number of courses enrolled

Using a ratio that includes successfully completed courses over the total number of courses enrolled gives a clear indication of student success, of the time a student has been enrolled. Many students "stop-out" from community college or take a few courses each semester, skewing and misrepresenting averages. The success ratios for transfer-level courses favors Caucasian males and females (.60 and .58 respectively), while Latinas and Latinos lagged behind Caucasians with a .55 and a .54 success ratio in transfer-level courses. Remedial course success favored the female sample, with Caucasian females at a .27 remedial course success ratio and Latinas with a .35 rate. Among the occupational courses, Caucasian males and females report high success ratios at .54. To test for statistically significant differences, a One-Way ANOVA was also performed on the success ratios. Statistical significance was found in the success rate of all courses. Tukey post hoc tests found statistical pairwise differences between Latinas and Caucasian females (p < .001). Latinas' (mean = .46; sd. =. 19) success rate in all courses is less than Caucasian females (mean = .52; sd. = . 19). Statistical difference was also found between Latinas and Caucasian males. Caucasian males (mean = .50; sd. = .24) are more successful in all courses than Latinas (mean = .46; sd. = .19). In addition, the Tukey post hoc test resulted in a statistical difference between Latinos and Caucasian females and a statistical difference between Latinos and Caucasian males. Caucasian females (mean = .52; sd. = .19) and males (mean = .50; sd. = .24) were more successful than Latinos (mean = .46; sd. = .21). Success ratios were not significant for transfer-level courses. However, success ratios for remedial and occupational courses were significant. According to the Tukey post hoc test, statistical difference was found between Latinas and Caucasian males, Latinas and Latinos, Latinos and Caucasian females and Caucasian females and Caucasian males. Caucasian females (mean = .38; sd. = .32) were more successful in remedial courses than Latinas (mean = .35; sd. = .30), Latinos (mean = .29; sd. = .30) and Caucasian males (mean = .27; sd. = .33). Occupational success ratios were statistically significant. The Tukey's post hoc locate statistical differences between Latinas and Latinos and Latinos and Caucasian females. Latinas (mean = .54; sd. = .31) are more successful in occupational courses than Latinos (mean = .50; sd. = .33). In addition, Caucasian females (mean = .57; sd. = .30) are more successful that Latinos (mean = .50; sd. = .33).

Transfer The TRUCCS sample is generally representative of the LACCD as well as other large urban community college districts. Despite their ethnic and gender categories, LACCD students have high aspirations for baccalaureate degrees, thus requiring the completion of transfer-level courses. However, as illustrated in the introduction, the transfer rates of all students are less than optimal. Past studies of transfer rates continue to view transfer as an outcome providing a set of predictor variables, allowing an institution to target a specific population or tailor services to facilitate greater transfer rates. Despite research and institutional changes transfer rates have not drastically improved. Taking a more continuous approach, the following analysis begins to analyze transfer-as-a-course-taking process that defines the transfer path as successful completion of two or more transfer-level Math and English courses. The IGETC articulation agreement between the California community colleges and the California State Universities and the University of California schools consists of six areas. Each area is a specific subject matter such as English, Math, Social Sciences, and Science. Students are encouraged to enroll and complete the first two areas, English and Math, before enrolling in subsequent areas. The English and Math courses provide the basic skills required to successfully complete other areas of the IGETC. Using English and Math in this analysis captured students are as they enter the transfer path. The entire enrollment history of students was included in the analysis with the number of semesters enrolled, ranging from 1 to 68 semesters.

Transfer Path The transfer path analysis was conducted over the entire enrollment history of the students. Students, however, have wide ranges of the number of semester enrolled. Within the TRUCCS sample, the first year of enrollment ranged from 1974 to 2001. The course completion ratios accurately represent the success in transfer-level courses compared to the number of semesters enrolled. Course Completion Ratio = Total number of two or more transfer-level courses completed with a passing grade / Total number of two or more transfer-level courses in which the student enrolled It is important to note that the students in this sample are only those students who reported aspiration to transfer. Although their aspirations may appear to be high, the actual process of course-taking describes a small amount of students who are actually on the transfer path. An analysis of students on the transfer path or students who completed more than two transfer-level English and Math courses reveals that 60% of the students attempted two or more transfer-level Math and English courses but only 50% completed the transfer-level courses. The course completion ratio for all male and female students was .49 and .48 respectively. To identify statistical differences between the gender and ethnic groups on the transfer path, a One-Way ANOVA was performed on the course completion ratio for students on the transfer path. The ANOVA identified statistical differences and a Tukey Post Hoe test identified the specific pairwise differences between groups. Results from the One-Way ANOVA are not significant. Caucasian males and females, Latinos, and Latinas are completing transfer-level courses at similar rates.

Summary of Findings

In addition to the information provided by the participation ratio, the success ratio results in a similar metric for all groups, comparing the number of courses completed with a passing grade by the number of semester enrolled. The success ratio, similar to the participation ratio, gives a clear idea of the successful course completion patterns of Caucasian males and females, Latinos, and Latinas. ANOVA results indicate that in all courses, Latinas' success rate is lower than all other groups. Course completion in transfer-level courses indicates that Latinas are competitive with Caucasian males and females and Latinos. In remedial courses, Latinas lag behind Caucasian females but report greater success than Latinos and Caucasian males. Latinas were also less successful in occupational courses than Caucasians but more successful than Latino males. Overall, Latinas course-taking patterns indicate that they are less successful than any other group; however, the patterns change when looking only at those students who aspire to achieve a bachelor's degree or more. In this subsample, Latinas are enrolling and completing transfer path courses at similar rates. The transfer path is defined as enrollment and completion in two or more transfer-level Math and English courses. In terms of course completion for transfer path courses, One-Way ANOVA results do not indicate statistical difference. Those Latinas who aspire to transfer, therefore, are enrolling and completing transfer-path courses at the same rate as their peers. All community college students, however, are reporting small numbers who enroll and complete transfer-level courses. Out of the total sample of students who aspire to transfer, only approximately 60% are enrolling in the transfer path, and only 50% complete transfer-level courses. Overall course completion rates for students on the transfer path results is 50%. Although Latinas are competing with their peers, the number of students who are on the transfer path is less than optimal. In summary, Latinas participate in more non-transferable courses than transfer-level courses as compared to Caucasian males and females. Success rates indicate that Latinas are successful in remedial courses at greater rates than other groups with one exception, Caucasian females. Dividing the Latina sample by those who only aspire to achieve a bachelor's degree or master's degree shows different results. Latinas with high aspirations enroll in and complete two or more transfer-level Math and English courses at the same rate as their peers. Latinas may be competitive with their peers but the overall rates for all students are quite low. Although the results of this study do not indicate the factors associated with low enrollment and completion in transfer-level courses; a pattern has emerged that requires attention. As the population of Los Angeles and the LACCD will continue to have more and more Latinos/Latinas, it is important to examine the course-taking behaviors of students to push Latinos/Latinas successfully through the educational pipeline.

Conclusion

This study presents evidence that Latinas are enrolling in fewer transfer-level and more remedial courses than their female Caucasian peers. This evidence suggests that Latinas are entering community colleges less prepared and/or are not receiving appropriate academic counseling. Questions remain as to the necessity of remedial courses. Do Latinas need remedial courses because of a lack of academic preparation or are they entering remedial courses due to ethnic tracking or lack of academic counseling? This study suggests that an achievement gap exists, implying that Latinas are not being appropriately served by K-12 schools and/or community colleges. Latinas' success rates are also not competitive with Caucasians except in transfer-level courses. In remedial and occupational courses, Latinas are not completing courses at the same rate as Caucasians. Latinas may not be academically prepared, may not be receiving appropriate academic services, or are not enrolling in appropriate courses for their level of academic preparedness. Again, community colleges may need to research the academic needs of the Latina population and review counseling and academic services as well as course offerings. The area in which Latinas find compatible success is in the transfer path definition that only includes those students who aspire to achieve a bachelor's degree. These results call into question the assumption that Latinas are not academically prepared. When Latinas have high aspirations, they are successful. Latinas may need support services that seek to bolster their aspirations. For details relating to the statistical analysis, please contact the author.

References

Census, U. S. (2000). State and county quickfacts. Retrieved March, 2003, from http://www.census.gov/main/www/cen2000.html

Census, U. S. (2002). American factfinder. Retrieved January, 2002, from http://www.census.gov

Cohen, A. M., & Brawer, F. B. (1996). Policies and programs that affect transfer. Washington D.C.: American Council on Education.

California Postsecondary Education Commission. (2000). California postsecondary education commission data and reports. Retrieved November 25, 2002, from http://www.cpec.ca.gov/SeeondPages/DataReports.asp

Helm, P. K., & Cohen, A. M. (2001). Leadership perspectives on preparing transfer students. New Directions for Community Colleges, 30(2), 99-103.

Los Angeles Community College District. (2000). Los Angeles community college institutional research report, from http://research.laccd.edu/research/

Los Angeles Community College District. (2001). Los Angeles community college district enrollment and student characteristics. Retrieved November 15, 2002, from http://research.laccd.edu/research/

Melendez, S. E., & Petrovich, J. (1989). Hispanic women students in higher education: Meeting the challenge of diversity. In J. G. Touchton (Ed.), Educating the Majority: Women challenge tradition in higher education. New York: Macmillan.

Nora, A., & Rendon, L. I. (1990). Determinants of predisposition to transfer among community college students: A structural model. Research in Higher Education, 31(3), 235-255.

Rendon, L. & Mathews, T. (1989). Success of community college students: Current issues. Education and Urban Society, 21(3), 312-327.

Rendon, L. & Taylor, M. (1989). Hispanic students: Action for access. Community, Technical, and Junior College Journal, 60(3), 18-23.

Rendon, L. & Valadez, J. (1993). Qualitative indicators of hispanic student transfer. Community College Review, 20(4), 27-37.

Rendon, L., & Mathews, T. (1993). Success of community college students: Current issues. In B. K. Townsend (Ed.), Women in Higher Education: A Feminist Perspective. Needham Heights, MA: Ginn Press.

Jaime Lester, University of Southern California 

Linda Serra Hagedorn, University of Southern California

Lester is a doctoral student at the Rossier School of Education. Dr. Hagedorn is an Associate Professor at the Rossier School of Education.