PEDAR: Research Methodology College Persistence on the Rise? Changes in 5-Year Degree Completion and Postsecondary Persistence Rates Between 1994 and 2000
Beginning Postsecondary Students Longitudinal Study
1989-90 Beginning Postsecondary Students Longitudinal Study
1995-96 Beginning Postsecondary Students Longitudinal Study
Bias Analysis
Accuracy of Estimates
Data Analysis System
Statistical Procedures
Differences Between Means
Interaction Effects (Changes in Completion Gap)
Executive Summary
Full Report (PDF)
Executive Summary (PDF)
 Beginning Postsecondary Students Longitudinal Study

The Beginning Postsecondary Students Longitudinal Study (BPS) is based on a sample of students who enrolled in postsecondary education in institutions located in the 50 states, the District of Columbia, and Puerto Rico for the first time in a specific academic year. Two BPS surveys have been conducted thus far; one followed students who first began their postsecondary education in 1989–90 (BPS:90/94) and a second followed students who began in 1995–96 (BPS:96/01). Unlike other NCES longitudinal surveys (such as the National Education Longitudinal Study of 1988), which follow age-specific cohorts of secondary school students, the BPS sample includes nontraditional students who have delayed their postsecondary education due to financial need or family responsibilities, or other reasons. Students who began their postsecondary studies before the base year of the study, or who stopped out, and then returned to their studies in the base year were not included, nor were students who were still enrolled in high school. Eligible students for the BPS samples were identified from participants in the two corresponding National Postsecondary Student Aid Studies (NPSAS:90 and NPSAS:96).

The BPS survey samples, while representative and statistically accurate, are not simple random samples. Instead, the samples are selected using a more complex three-step procedure with stratified samples and differential probabilities of selection at each level. First, postsecondary institutions are selected within geographic strata. Once institutions are organized by zip code and state, they are further stratified by institution control (i.e., public; private not-for-profit; or private for-profit) and highest degree offering (less-than-2-year; 2-year to 3-year; 4-year non-doctorate-granting; and 4-year doctorate-granting).

A computer-assisted telephone interview (CATI) was conducted with BPS students twice after the Base Year survey. The final follow-up survey for BPS:90/94 took place about 5 years after college entry (in 1994), while the final follow-up survey for BPS:96/01 took place 6 years after (in 2001). The CATI collected information concerning enrollment, program completion, education financing, employment, and family formation; graduate school access and enrollment; and civic participation.

The BPS survey data underwent several data quality evaluations, which included both online data editing procedures and post-data collection editing. The online data editing ensured that the data collected fell within legitimate ranges and where feasible, items were cross-checked against other related items. After data collection, the data were cleaned and edited using several steps including verification of one-way frequencies for each item, cross-tabulations of related items, standard variable recoding and formatting (such as dates), the determination of outlier values, and logical imputations. After the CATI data were cleaned and edited, composite variables for specific data analyses were created, which were subjected to similar cleaning and checking procedures.

next section