The impact of a program for the disadvantaged on student retention

College Student Journal, March, 2008 by Andrew W. Braunstein, Mary H. Lesser, Donn R. Pescatrice

This study compares the freshman to sophomore retention rate for all freshmen at a medium-sized college to the retention rate for freshmen who participated in a program for disadvantaged students. The retention rate (over a three-year period) was nearly identical for the two groups. That finding is an indicator of the benefits of the special program, given that the students who participated in the program had lower SAT scores, lower high school grade point averages, and lower family incomes compared to all freshmen at the college. Based on previous research in the area of student retention, one would clearly expect a significantly higher freshman to sophomore retention rate for the larger group. Furthermore, a comparison of logistic regression models shows that the retention of the student population in general was influenced to a greater degree by demographic, academic, and financial factors than was the retention of the smaller group.

**********

Many programs aimed at achieving equal opportunity in education had their beginnings in the Higher Education Act of 1965. One program created was Student Support Services (SSS), formerly known as Special Services for Disadvantaged Students. As described by McCants (2002), the program is aimed at helping low-income and first generation college students and individuals with disabilities graduate from college. Services provided include assistance with obtaining financial aid, special instruction, and academic and non-academic counseling.

A medium-sized, private, liberal arts college in the Northeast provided these free services for twenty years to eligible students with the help of funding from the U.S. Department of Education. To participate in the program, a student needed to meet one or more of the following criteria: neither parent graduated from a four-year college; the family's taxable income was below a level mandated by the federal government; the student had a physical or learning disability. Among the specific services provided to students participating in the program were personal, academic, and financial aid counseling, help with study skills, tutoring, career planning, peer mentoring, and exposure to cultural enrichment activities.

In this study, we examine data for a recent three year period. Of specific interest is a comparison of the freshman to sophomore retention rate for the group of students participating in the special program to that for all students enrolled in the college during that same time period.

Results and Discussion

Table I below presents some important data characteristics for the two sets of students:

From Table I, we see that the overall retention rate was nearly identical for the two sets of students, but that outcome speaks very highly for the program for the disadvantaged students for several reasons:

   Both the average SAT score and the average high school GPA for
   students participating in the program were significantly lower than
   the same measures for the student body as a whole. There is
   consistent evidence in the literature that student grades,
   achievement, or academic performance have an overwhelming positive
   correlation with persistence (St. John, 1989; St. John, 1990; St.
   John, Kirshstein, & Noel, 1991; Mallette & Cabrera, 1991; Cabrera,
   Nora, and Castaneda, 1992; Perna, 1997; Tinto, 1997; Murtaugh,
   Burns, & Schuster, 1999; St. John, Hu, Simmons & Musoba, 2001). The
   overwhelming majority of students participating in the special
   program were in the two lowest income categories under
   consideration, whereas there was a fairly even distribution among
   income categories when looking at all freshmen. It has been often
   revealed that students from families with larger incomes tend to
   persist more than students from families with lower incomes (St.
   John, 1989; Cabrera, Stampen, & Hanson, 1990; St. John, 1990; St.
   John et al., 1991). Students participating in the program received
   loans at a much greater rate than the student body as a whole. Some
   research, including that done at the institution in question
   (Braunstein, Lesser, and Pescatrice, 2006) has shown that the
   receipt of loans (either alone or in a package with grant money)
   has a negative impact on retention.

Logistic regression techniques were also employed to analyze the data. One of the goals of using logistic regression is to attempt to calculate the change in the probability of retention associated with a change in one of the demographic, academic, or financial factors. For instance, one might try to find out how much an extra $1,000 of grant money increases the probability of a student being retained. Or one might calculate the change in the probability of retention associated with being a commuter as opposed to being a resident student (holding the values of all other demographic, academic, and financial variables constant). Table II below compares the results of one particular logistic regression model estimated for all freshmen to estimated results for that same model for those participating in the program. A series of logistic regression models estimated for all freshmen at the institution (Braunstein et al., 2006) found that ethnicity/race, high school grades, family income, and financial aid (for instance, grants have a positive effect; loans have a negative effect) are all statistically significant factors in determining whether or not a freshman student is retained. For the model whose results are displayed in Table II, we see that, for both groups of students, money received in the form of grants has a marginally significant positive impact on retention. Whereas race is a highly significant factor in the retention of all freshmen at the institution (Whites are more likely to return), it is only marginally significant for the program participants. Income is seen to be a significant factor in the retention of the entire group of freshmen, but is not statistically significant for the smaller group. Finally, high school grades are not a significant factor in the freshman to sophomore retention of those in the special program, but it is a highly significant factor for all freshmen (those with grades in the bottom 30% are less likely to return than those in the middle 40%, and those with grades in the top 30% are more likely to return than those in the middle 40%). It seems clear that the programs and services provided to the disadvantaged students in some sense "leveled the playing field", in that the retention of those students did not seem to be influenced (or, in the case of ethnicity, was influenced to a lesser degree) by some of the factors that seemed to influence the retention of the college's freshmen in general.