Method Data
The source of data for this analysis is the restricted version of the National
Postsecondary Aid Study (NPSAS:96), a nationally comprehensive sample of
students enrolled in postsecondary education in 1995-96. Included in NPSAS:96
is a sample of first-time postsecondary students (of all ages) who make up the
Beginning Postsecondary Student (BPS) longitudinal study cohort.
Sample
The total number in the sample for this study is 6,351: 1,814 students who chose
a two-year college and 4,537 students who chose a four-year college. Caucasian
students comprise 68 percent of the two-year college sample and 73.4 percent of
the four-year sample. African American students comprise 14.2 percent of the
two-year college sample and 10.4 percent of the four-year college sample, while
Latino students make up 12.4 percent of the two-year sample and 8.4 percent of
the four-year college sample. All other ethnic groups comprise 5.5 percent of the
two-year sample and 7.4 percent of the four-year sample.
In terms of gender, 45.8 percent of the two-year college sample is male and
45.4 percent of the four-year college sample is male. Thus, females comprise a
majority of both the two-year and four-year sample.
Most of the students are under the age of 22, comprising 74.8 percent of the
two-year sample and 96.7 percent of the four-year sample. Those over 22
comprise 23.1 percent of the two-year sample and 3.3 percent of the four-year
sample.
Model
The model for this study examines the nexus between student background
characteristics (17 variables), student aspirations (2 variables), high school
experiences (9 variables), college experience (8 variables), price and subsidies (8
variables), debt load (4 variables), and BPS/choice questions (5 variables) (Table
1).
Statistical Method
The statistical method consisted of two steps. First, an ANOVA was performed
on the 88 BPS/college choice variables to determine which variables were a best
fit for the model. Five variables were significant (.05) for both two- and four-year
students. The second step in the statistical method was to perform a logistic
regression analysis with the complete model (Table 2). The outcome variable
was whether or not a student enrolled in a two-year college. Because of the
large sample size, we set our significance level of p = .001 (see Thomas & Heck,
2001 for further consideration of working with large databases).
Whether a student chooses a two-year or a four-year college, the outcome is
dichotomous: either yes or no (coded as 1 or 0). The resulting graph of the
relationship is not a straight line, but a curved line bounded by 0 and 1.
Regardless of the values of the constants βI or the variables XI, this equation still
results in values between 0 and 1 because of the properties of the natural
logarithm. The value P can also be thought of as a probability measure that the
outcome variable will be 1 (yes). This is precisely what a dichotomous model
requires (Cabrera, 1994; Menard, 1995).
6 Journal of Student Financial Aid Volume 39 • Number 1 • 2009
Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46  |  Page 47  |  Page 48