|
 |
THE ACCOUNTABILITY PROBLEM AND TRANSFER “RATES” |
For many years, the academic research community has struggled to identify a uniform
measure of transfer activity that could be used to understand the health of the transfer
function within community colleges. But because there are many different types of
transfer, no single rate can capture all the transfer activity. For students who begin at two-
year colleges, for example, upward vertical transfer to a baccalaureate degree institution
is just one option. Researchers also disagree about which students to count in the transfer
“base” (all entering students, or only those in degree-granting programs, or only those
indicating a desire to transfer), and when in students’ academic careers to count them (all
first-time students, after 12 units, etc.). The two studies described below are particularly
important in understanding the metrics of 2/4 transfer rates.
Transfer Assembly Project
The Transfer Assembly Project, based at the Center for the Study of Community Colleges
at the University of California at Los Angeles and headed by Arthur Cohen, is the
longest-standing study focusing on statewide measures of community college–
baccalaureate transfer. Since 1989, the project has collected data on transfer rates initially
for 18 and now for 24 states, using the following measure: the transfer rate is the
percentage of all first-time community college students who complete at least 12 units at
that college and who take at least one class from a public in-state university within four
years of leaving the community college.* Data are collected from individual institutions
within a state, sometimes through the statewide agency, and are aggregated into a
statewide rate, which is subsequently reaggregated into a national transfer rate. Because
of confidentiality agreements, data are not published for individual institutions or for the
states. Analysis of the changes in the rates, however, indicates that there are larger
disparities in transfer rates between institutions within states than there are between
states.
The Transfer Assembly Project’s base for calculating transfer rates is a subset of all first-
time community college students (it excludes students who are not first-time students as
well as those who fail to complete at least 12 units). But the base includes students in
“vocational” as well as academic courses—not just those students who have indicated
that they plan to transfer—because of the view that many students change their minds
about plans after enrolling in college, and because many “vocational” courses are offered
for credit that sometimes is transferable.
Although the Transfer Assembly Project has retained the same definition of “transfer
rate,” the methodology for collecting data and the number of institutions in the database
have changed over the years. Thus the data are not entirely comparable from year to year.
The most recent study, published in 2001, tracks transfer rates for students who first
enrolled in 1995. The trend data show a dip in transfer rates in the 1980s and a rise in the
1990s (see table 1), changes that the authors attribute to overall economic conditions and
the emphasis on academic (in contrast to vocational) education within community
colleges.
NCES Study of Alternative Transfer Rates
Under the auspices of the National Center of Educational Statistics, Ellen Bradburn and
David Hurst (2001) explored the consequences of using different populations of potential
transfer students in calculating transfer rates for first-time students enrolling between July
1, 1989, and June 30, 1990. “Transfer” was defined as initial enrollment in a community
college followed by subsequent enrollment at any four-year institution (public or private,
in any state) within the five-year period. The initial pool of “potential transfer” students
included all students eligible for transfer, and the alternative definitions were increasingly
restrictive. Their analysis showed that the transfer “rate” increases as the pool of students
narrows (see table 2). In addition, student socioeconomic characteristics vary among the
different pools, with the least restrictive pools containing the most diverse group of
students, including the largest proportion of students of color from low-income families.
* Under the "12 units or more" standard, identical entry pools that produce the same number of transfer students can have vastly different transfer rates. For example, state A's two-year colleges enroll 100,000 first-time students; 2,500 of them complete 12 or more units; 800 of those students transfer to a four-year college; and so the transfer rate is 32% (800 of 2,500). State B's two-year colleges also enroll 100,000 first-time students, but 10,000 of them complete 12 or more units, and 800 of those students transfer to a four-year college. Here, the transfer rate is only 8% (800 of 10,000).
|