Transfer Shock in an Academic Discipline: The Relationship between Students' Majors and their Academic Performance.

The study described in this report involved associate's degree recipients who were traditional age and enrolled full-time at a two-year college before pursuing their same majors at a four-year liberal arts college. Based on their majors, students were placed into four discipline categories: fine arts and humanities, mathematics and sciences, social sciences, and professions. The authors analyzed students' pre- and post-transfer GPAs using a 2 x 4 mixed model ANOVA to determine that no significant differences existed among mean community college GPAs relative to students' majors, and no significant difference existed between two- and four-year college GPAS for the entire sample. The ANOVA did indicate, however, an interaction between location and major on students' GPAs: Tukey's HSD tests revealed a significant decline in the mathematics and sciences mean GPA from the community college to the four-year institution. The authors conclude by defining avenues for further research.

Community colleges have been criticized for inadequately preparing students who transfer to four-year colleges and universities (Susskind, 1996). For example, Diaz (1992) contends that community college transfers have been "stereotyped as risky ventures" (p. 280). Despite this negative viewpoint, community college enrollment has continued to increase, and more than 50% of first-time college students enter at the community college (Cohen & Brawer, 1996). Financial factors, stricter entrance requirements at four-year institutions, and state-initiatives to improve transfer are commonly cited as indicators that students will continue to rely on community colleges to obtain their first two years of baccalaureate education (Anglin, Davis, & Mooradian, 1995; Graham & Hughes, 1994).

Considerable research has investigated the academic performance of community college transfer students at four-year institutions. The phenomenon of transfer shock (Hills, 1965), a decline in grade point average (GPA) during the first semester at a four-year institution, has continued to be a common finding during the 1990s (Best & Gehring, 1993; Keeley & House, 1993; Preston, 1993; Soltz, 1992). Recent studies have reported that community college transfer students experienced academic dismissal or failure rates of between 18 and 22% at the conclusion of their first semester at a four-year institution (Baldwin, 1994; Graham & Hughes, 1994). Moreover, Graham and Dallam (1986) found that in comparison to continuing native students, transfer students were more likely to be placed on academic probation as a result of their first semester GPAs at the four-year institution.

A review of the literature indicates a relationship between the number of credits completed at the community college and academic performance at the four-year institution. Several studies have found that students who transferred with upper division status (defined as completion of the associate's degree or at least 60 credits) experienced a lesser degree of transfer shock than did lower division transfers (Best & Gehring, 1993; Graham & Hughes, 1994; House, 1989; Keeley & House, 1993; Miller, Janawsky, & Katz, 1977; Richardson & Doucette, 1980). Miller et al. (1977) and House (1989) indicated that community college students who transferred with upper division status achieved GPAs that were similar to native four-year college students. More important, research has shown that upper division transfers graduated at a rate that was significantly higher than lower division transfers (Best & Gehring, 1993; House, 1989). The literature also reveals a relationship between the cumulative community college GPA and academic performance at the four-year institution (Baldwin, 1994; Miller et al., 1977; Townsend, McNerny, & Arnold, 1993). The common finding of these efforts is that transfers who achieve higher GPAs at the community college are also academically successful at four-year institutions.

Although transfer shock has been the focus of numerous research studies, the common practice has been to analyze data for a heterogeneous sample or population. Only a few studies have provided information on the academic performance of community college transfers in the respective academic disciplines. Two investigations found that although community college transfers in all majors experienced transfer shock, the mean GPA declines for the respective majors varied by as much as half a grade point (Keeley & House, 1993; Richardson & Doucette, 1980). Other research (Tippin, 1982; Webb, 1985) brought to light that, in some majors, community college transfer students achieved a higher first semester GPA at the four-year institution. A final study (James Madison University, 1989) did not provide data concerning transfer shock but indicated that community college transfers who majored in biology, chemistry, math, physics, accounting, and economics realized a first semester mean GPA of 2.2, whereas transfers in all other majors achieved a first semester mean GPA of 2.73.

In the American Council on Education's policy statement on the transfer function, Palmer and Eaton (1991) called for a change in the evaluation of post-transfer academic performance. The authors stressed the need to examine the academic performance of community college transfer students in specific programs, maintaining that such studies will "either confirm transfer success or detect weaknesses that need attention"(Palmer & Eaton, 1991, p. 39).

The limited research on academic performance in specific disciplines has examined community college students who transferred to public universities or state colleges. No such investigations have been conducted from the institutional perspective of the private, liberal arts college. In addition, results have not been subjected to tests of significance. The primary purpose of this study, therefore, was twofold: (a) to examine the transfer shock phenomenon from a discipline-based approach using the perspective of a private, liberal arts college as the senior institution, and (b) to determine whether statistical significance existed in the academic performance of community college transfer students in these different disciplines.

Limitations

Based upon previous research, the researchers assumed that the findings of this study could be influenced by a number of factors. Specifically, we expected that community college students who transferred with upper division status and higher GPAs would perform better than students who transferred with lower division status. Based upon the findings of Tinto (1987) and Pincus and DeCamp (1989), we also expected that transfer students who matched the demographic characteristics of the liberal arts college student, predominately traditional age (under age 25) and enrolled on a full-time basis, would perform better than would nontraditional age students (age 25 and over) enrolled on a part-time basis. As a result of these assumptions and other considerations, significant limitations were placed on the subjects included in the study.

Transfer students included in this study attended community colleges in a state-wide system. The system had designated the associate of arts (AA) degree, with a common 45-hour general education curriculum, as a "transfer" degree. This designation indicated that the student had completed between 60 and 64 credit hours at the community college, including the general education curriculum, and had earned a minimum grade of C in each course. By mandate, public four-year institutions accepted general education courses as "equivalent" to their requirements and accepted all hours completed in the AA degree. Virtually all private four-year institutions in the state adopted a similar understanding with the community college system. A final aspect of the AA degree was a declared transfer "major," such as associate of arts in business.

Limiting the subjects to those who had completed the AA degree satisfied the performance expectations related to upper division status and GPA. Demographic assumptions further limited the study to those students who were traditional age and enrolled on a full-time basis. As a final limitation, the study included only those subjects who pursued the same major at both the community college and the liberal arts college. This measure eliminated the possibility that results could be influenced by changes in major.

It is important to note that these limitations result in a sample that differs from the majority of community college students (nontraditional age, enrolled part-time, transfers at various points prior to completing a degree). Recent trends, however, support the limitations incorporated in this study. Examples from two states illustrate curricular and degree efforts similar to those used in this investigation. Facing enrollment limitations at public four-year institutions, the state of Florida defined the AA degree as the "transfer degree" and, subsequently, guaranteed admission to the state university system to all students that completed the degree (Harden, 1991). Florida has since enacted legislation that mandated standard requirements for baccalaureate programs among the 9 public universities and 28 community colleges (LeMon & Pitter, 1996). After developing a common general education core curriculum for transfer programs, the state of Illinois launched an initiative to articulate lower-division courses required in 17 majors to ensure transfer between any and all community colleges and four-year institutions (Illinois Community College Board, 1996).

Although national statistics are not collected, institutional and state data suggest that more traditional age students are enrolling in community colleges, and that many of these students attend on a full-time basis. Piedmont Virginia Community College attributed the vast majority of a 19% rise in full-time enrollment to increased attendance by recent high school graduates (Head, 1988). Dramatic enrollment increases of traditional age students were reported by a number of community colleges in the late 1980s. For example, Delgado Community College in Louisiana reported the number of 18-to-24-year olds attending increased 75% (from 2,292 to 4,019) between 1987 and 1990 (Collison, 1991). A study of college choices found that 21% of the 1994 graduates of Oregon secondary schools enrolled in one of the state community colleges the following fall (Oregon State Department of Education, 1996). Recent enrollment statistics from the state of Pennsylvania reveal that 46% of the students attending community colleges are traditional age and that more than half (53%) of these students are freshmen, enrolled on a full-time basis (Pennsylvania Association of Colleges and Universities, 1997).

Hypothesis

With the limitations in mind, we established the following null hypothesis to apply tests of significance to the data collected in this study: There is no significant difference in the mean GPA change of community college transfers among the various academic disciplines of student major.

Methodology

The subjects included transfer students from a public community college system in a midwestern state who enrolled in a private, liberal arts college in the same state during a seven-year time period (fall 1989 to fall 1996). A population of 434 students transferred from the state community college system to the liberal arts college during the years included in this study. Applying the limitations to this population identified a sample of 250 students, 57.6% of the population.

An examination of the respective majors of the sample revealed instances of single transfer students in each of the majors of economics, journalism, physics, religious studies, and theater. As the liberal arts institution was a college, and did not have schools within the college, we placed the students in one of the four following categories of disciplines:

Fine arts and humanities--majors in art, music, theater, English, foreign language, history, religious studies;

Mathematics and sciences--majors in mathematics, biology, chemistry, physics;

Social sciences--majors in economics, government, psychology, sociology;

Professions--majors in business administration, elementary education, journalism.

This placement resulted in the following homogeneous discipline sizes: fine arts and humanities (N=51), mathematics and sciences (N=54), social sciences (N=66), professions (N=79).

To test for significance we computed a 2 x 4 mixed model ANOVA and conducted post hoc tests. The ANOVA determined differences between the means for the four following comparisons:

(a) differences between the community college GPA means of the respective disciplines;

(b) differences between the liberal arts college GPA means of the respective disciplines;

(c) differences between the pre- and post-transfer GPA means of the total sample;

(d) differences between the pre- and post-transfer GPA means of the respective disciplines.

Post hoc Tukey's honestly significant difference (HSD) tests were conducted to determine which differences were statistically significant.

Results

Mean pre- and post-transfer GPAs of the respective disciplines are summarized in Table 1. Although a wide range between the mean community college GPAs for the respective disciplines is not evident, a substantial amount of difference exists among individual subjects in the study. The lowest community college GPA was 2.418 (fine arts and humanities); the highest community college GPA was 3.750 (professions). Data from the first semester at the liberal arts college indicates an increase in the range of both mean and individual GPAs when compared to the community college GPAs. The mean range for the respective disciplines increased 52% (.185 to .352) with 2.04 (mathematics and sciences) as the lowest GPA and 3.83 (social sciences) as the highest GPA.

Table 1
Pre- and Post-transfer GPA by Discipline

                                      Mean GPA

Discipline            N     Pre-transfer    Post-transfer    Change

Fine Arts &
 Humanities           51   [2.963.sub.a]    [3.121.sub.a]    +.158
Mathematics &
 Sciences             54   [3.142.sub.a]    [2.870.sub.b]    -.272
Social Sciences       66   [3.125.sub.a]    [3.222.sub.a]    +.097
Professions           79   [3.148.sub.a]    [2.999.sub.a]    -.148

TOTAL                250   [3.103.sub.a]    [3.055.sub.a]    -.048

Note. The same subscript indicates the means of the respective discipline or total are not statistically significant at p <.05 in the Tukey honestly significant difference comparison.

As shown in Table 1, a minimal amount of transfer shock (.048) was experienced by the entire sample included in this study. The data for the homogeneous disciplines, however, indicates both decreased and increased GPAs at the four-year institution. Students in the disciplines of mathematics and sciences and professions realized transfer shock, with declines of .272 and .1489, respectfully. Students in the disciplines of fine arts and humanities and social sciences experienced mean GPA increases of .158 and .097, respectfully. Nickens (1972)coined the term transfer ecstasy to represent such post-transfer GPA increases.

To test differences in means for the four comparisons a 2 (location: community college vs. liberal arts college) x 4 (discipline: fine arts and humanities vs. mathematics and sciences vs. social sciences vs. professions) mixed model ANOVA was conducted. The dependent measure in this test was the GPA at both the community college and liberal arts college. The results of the ANOVA indicated no significant difference for two comparisons. First, no significant difference existed between the mean community college GPAs of the respective disciplines, F (3,246) = 1.23, p [is greater than] .05, MSe = .35. In other words, students' majors did not significantly influence GPA at the community college. Second, no significant difference existed in the location (community college vs. liberal arts college), F (1,246) = 1.61, p [is greater than] .05, MSe = .13. In other words, the GPA decline (.048) experienced by the entire sample was not statistically significant.

The ANOVA did, however, indicate an interaction between location and discipline on students' GPAs, F (3,246) = 9.00, p [is less than] .01, MSe = .13. Post hoc Tukey's HSD tests were computed to determine which of the eight means were significantly different. As shown in Table 1, the Tukey's procedure revealed a significant decline only in the mathematics and sciences mean GPA from the community college to the liberal arts college. In other words, although all disciplines experienced decreases or increases in post-transfer mean GPAs, only the mathematics and sciences discipline experienced a change that was statistically significant.

As shown in Table 2, the Tukey's HSD indicated significant difference between three of the liberal arts GPA means. The mean liberal arts GPA of the mathematics and sciences majors was significantly lower than the mean liberal arts GPA of two disciplines--fine arts and humanities and social sciences. The mean liberal arts GPA of the professions majors was significantly lower than the mean liberal arts GPA of the social sciences majors. Based on these findings, we rejected the null hypothesis and concluded that there is a significant difference in the mean GPA change of community college transfers between the various academic disciplines of student major.

Table 2
Matrix of Liberal Arts GPAs by Discipline

Discipline                    Mathematics/
                 Fine Arts/     Social        Social
                 Humanities    Sciences      Sciences   Professions

Fine Arts &
 Humanities         --
Mathematics &
 Sciences          .251(*)       --
Social
 Sciences          .101         .352(*)         --
Professions        .122         .129           .223(*)       --

Note. Critical Value Tukey a < .05 = .1964. (*) p < .05.

Discussion

This study supports the call for investigations of the academic performance of community college transfer students in specific programs (Palmer & Eaton, 1991). Previous research has reported that community college transfer students commonly experienced mean GPA declines of .50 or greater. Considering the small GPA decline experienced by the total sample in this study, the faculty and administration might erroneously assume that all community college transfer students would perform at essentially the same GPA level at both the two- and four-year institution. Thus, attention would not be drawn to the difficulties encountered by transfers in mathematics and sciences.

This finding also questions the definitions of transfer shock (Hills, 1965) and transfer ecstasy (Nickens, 1972), which need further clarification. Although students in both the mathematics and sciences and professions disciplines experienced GPA declines, only the mathematics and sciences decline was statistically significant. Students in both the fine arts and humanities and social sciences disciplines experienced GPA increases, yet neither increase was statistically significant. Shock and ecstasy would better describe instances of decrease or increase that are statistically significant, yet these terms have been used to describe all instances of decrease or increase.

Results of the study support previous findings that the academic performance of community college transfer students at the four-year institution varies across the respective majors (Keeley & House, 1993; Richardson & Doucette, 1980; Tippin, 1982; Webb, 1985). Statistical analysis revealed a significant difference when comparing the post-transfer means of the respective disciplines. Because there was not significant difference between the pre-transfer GPA means or between the pre- and post-transfer GPA mean of the total sample, there is strong evidence to suggest a relationship between the students' major and their academic performance at the four-year college.

As this study is an institutional one, focusing on a particular liberal arts college, it suggests a number of areas for future research. Additional research incorporating, and perhaps comparing, the differing types of four-year institutions would contribute greater insight to the relationship between transfer students' majors and their performance at four-year institutions. Similar studies could be conducted with the sample limited to nontraditional aged students enrolled on a part-time basis. This would add to the existing research on the influence of demographic characteristics on the academic performance of community college transfer students. Moreover, repeating the methodology of this study when there is a sufficient sample size in specific majors would provide valuable information for both community colleges and four-year institutions. A number of repons have identified the major(s) pursued by significant populations of community college students (Grosset & Estrada, 1994; Lucas & Meltesen, 1994). Information concerning the academic performance of transfer students in these majors, however, has yet to be explored.

Most importantly this investigation adds to the existing body of research, but it is limited in scope and should not be used for generalized conclusions. Not all states have adopted transfer policies similar to the state involved in this study. Community college students transfer prior to earning the AA degree as well as after earning degrees other than the AA. The majority of community college enrollment continues to be composed of nontraditional age and part-time students. In addition, inherent differences exist between and among community colleges and four-year institutions. The limitations of this study and organizational differences emphasize the need for additional research focused on the relationship between students' majors and their academic performance as community college transfers in pursuit of the baccalaureate degree.

References

Anglin, L.W., Davis, J.W., & Mooradian, P.W. (1995). Do transfer students graduate? A comparative study of transfer students and native university students. Community College Journal of Research and Practice, 19, 321-330.

Baldwin, A. (1994). Indicators of the university success of associate degree recipients in the fields of business, computer science, and engineering. Journal of Applied Research in the Community College, 1, 113-128.

Best, G. A., & Gehring, D. D. (1993). The academic performance of community college transfer students at a major state university in Kentucky. Community College Review, 21(2), 32-41.

Cohen, A. M., & Brawer, F. B. (1996). The American community college (3rd ed.). San Francisco: Jossey-Bass.

Collison, M. (1991). More traditional-age students consider community colleges. The Chronicle of Higher Education, 37(30), Al, 29-30.

Connecticut Community-Technical Colleges. (1992). The community and technical colleges of Connecticut student enrollment statistics, 1991-92. Hartford: Connecticut Community-Technical Colleges Board of Trustees. (ERIC Document Reproduction Service No. ED 351 084)

Diaz, P. E. (1982). Effects of transfer on academic performance of community college students at the four-year institution. Community/Junior College Quarterly of Research and Practice, 16(3), 279-291.

Georgia State Department of Education. (1994). How to make transfer work: Recommendations from the literature. Georgia Department of Technical and Adult Education Briefing Paper, No. 4. (ERIC Document Reproduction Service No. ED 391 557)

Graham, S. W., & Dallam, J. (1986). Academic probation as a measure of performance: Contrasting transfer students to native students. Community/Junior College Quarterly of Research and Practice, 10, 23-34.

Graham, S. W., & Hughes, J. C. (1994). Moving down the road: Community college students' academic performance at the university. Community College Journal of Research and Practice, 18, 449-464.

Grosset, J., & Estrada, D. (1994). A profile of CCP (community college of Philadelphia) transfer students enrolled at the state system of higher education universities during the fall 1993 semester. Institutional Research Report No. 79. Philadelphia: Community College of Philadelphia Institutional Research Office. (ERIC Document Reproduction Service No. ED 375 884)

Harden, H. D. (1991, March). Challenges of articulation between sectors. Paper presented at Florida State University's National Conference "Celebrating Leadership for the 21st Century," Tallahassee, FL. (ERIC Document Reproduction Service No. ED 331 559)

Head, R. B. (1988). PVCC Full-time students: Fail quarter 1987. Research Report No. 5-88. Charlottesville, VA: Piedmont Virginia Community College Office of Institutional Research and Planning. (ERIC Document Reproduction Service No. ED 297 804)

Hills, J. R. (1965). Transfer shock: The academic performance of the junior college transfer. Journal of Experimental Education, 33, 201-215.

House, J. D. (1989). The effect of time of transfer on academic performance of community college transfer students. Journal of College Student Development, 30, 144-147.

Illinois Community College Board. (1996). Status report on the Illinois articulation initative. Springfield: Author. (ERIC Document Reproduction Service No. ED 390 503)

James Madison University. (1989). The academic performance of transfers from Virginia's public two-year institutions at James Madison University. Harrisonburg, VA: James Madison University Office of Planning and Analysis. (ERIC Document Reproduction Service No. ED 321 778)

Keeley, E. J., & House, J. D. (1993, May). Transfer shock revisited: A longitudinal study of transfer academic performance. Paper presented at the Annual Forum of the Association for Institutional Research, Chicago, IL. (ERIC Document Reproduction Service No. ED 357 774)

LeMon, R. E., & Pitter, G. W. (1996). Standardizing across institutions: Now that we all look alike, what do we look like? Tallahassee, FL: Florida State Board of Regents. (ERIC Document Reproduction Service No. ED 399 858)

Lucas, J., & Meltesen, C. (1994). Follow-up study of 1991 Harper College transfer alumni. Palatine, IL: William Rainey Harper College Office of Planning and Research. (ERIC Document Reproduction Service No. ED 373 827)

Miller, H. F., Janawsky, R., & Katz, A. (1977). The academic achievement of two-year college graduates in New Jersey four-year colleges. Research Report No. 77-2. Trenton: New Jersey State Department of Higher Education. (ERIC Document Reproduction Service No. ED 143 397)

Nickens, J. M. (1972, April 6). Transfer shock or transfer ecstasy? Paper presented at the Annual meeting of the American Educational Research Association, Chicago, IL. (ERIC Document Reproduction Service No. ED 116 721)

Oregon State Department of Education. (1996). College choices: 1994. Part A. By college attended. Salem, OR: Office of Educational Policy and Planning. (ERIC Document Reproduction Service No. ED 399 984)

Palmer, J. C., & Eaton, J. S. (1991). Building the national agenda for transfer: A background paper. In American Council on Education, Setting the national agenda: Academic achievement and transfer (pp. 17-46). Washington, DC: American Council on Education.

Pennsylvania Association of Colleges and Universities. (1997). Statistical profile of higher education in Pennsylvania. Harrisburg, PA: Author.

Pincus, F. L., & DeCamp, S. (1989). Who transfers to four-year colleges? A study of a matched sample of B.A. recipients. Community/Junior College Quarterly of Research and Practice, 13, 191-219.

Preston, D. L. (1993, May). Interfacing two-year and four-year transcripts for transfer students. Paper presented at the Annual Forum of the Association for Institutional Research, Chicago, IL. (ERIC Document Reproduction Service No. ED 360 017)

Richardson, R. C., Jr., & Doucette, D. S. (1980). Persistence, performance and degree achievement of Arizona's community college transfers in Arizona's public universities. Tempe, AZ: Arizona State University Department of Higher and Adult Education. (ERIC Document Reproduction Service No. ED 197 785)

Soltz, D. F. (1992). JCCC transfer students: Their destinations and achievements. Overland Park, KS: Johnson County Community College Office of Institutional Research. (ERIC Document Reproduction Service No. ED 354 022)

Susskind, T. Y. (1996, August). Opportunities and challenges: Bridging the two-year four-year college gap. Paper presented at the symposium "Articulation from 2- to 4-Year Colleges: Is it an Allowed or Forbidden Transition?" Auburn Hills, MI. (ERIC Document Reproduction Service No. ED 399 990)

Tinto, V. (1987). Leaving college: Rethinking the causes and cures of student attrition. Chicago: University of Chicago Press.

Tippin, K. A. (1982, March). The academic performance of Missouri community college transfer students at the University of Missouri-Columbia. Paper presented at the Annual Community College Transition Conference, Columbia, MO. (ERIC Document Reproduction Service No. ED 216 746)

Townsend, B. K., McNemy, N., & Arnold, A. (1993). Will this community college transfer student succeed? Factors affecting transfer student performance. Community College Journal of Research and Practice, 17, 433-443.

Webb, E. (1985). Followup study of transfer students from C.O.S. to California State University, Fresno, & California Poly-Technic State University, San Luis Obispo, fail, 1984. Visalia, CA: College of the Sequoias Office of Institutional Research. (ERIC Document Reproduction Service No. ED 269 076)

Brent O. Cejda is an assistant professor of higher education at Texas Tech University in Lubbock, Texas ([e-mail]).

Alice J. Kaylor serves as associate academic dean at Saint Vincent College in Latrobe, Pennsylvania ([e-mail]).

Kirsten L. Rewey is an assistant professor of psychology at Saint Vincent College in Latrobe, Pennsylvania ([e-mail]).

Authors' Note: An earlier version of this paper was presented at the Annual Conference of the New England Transfer Association, Newport, Rhode Island, April 28-30, 1997.