The full peer reviewed version of a related study by Jaeger is now available. This study only covers one institution, but it is quite large. The final sample included 14,494 students at a "a large research-extensive institution located in the
southeast" presumably NC State University, Raleigh, the authors' home institution.
They examined ten variables to see if they would predict an eleventh, student retention. The potential predicting variables were
- ethnicity
- gender
- high school GPA
- high school percentile
- rank
- SAT verbal score
- SAT math score
- total SAT score
- percent exposure to graduate student instruction
- percent exposure to part-time faculty instruction
- percent exposure to full-time faculty instruction
Of those only HS GPA, gender, number of hours attempted and exposure to part time faculty had any predictive value. Interestingly, exposure to part time faculty did not have as strong a negative effect on retention as being female.
One of their citations seems worth following up on:
Hagedorn, L., Perrakis, A., & Maxwell, W. (2002). The negative commandments: Ten
ways community colleges hinder student success. (ED 466 262).
I don't know what (ED 466 262) means, but the article seems related to this
Florida Journal of Educational Administration and Policy, Fall, 2007 by Linda Serra Hagedorn, Athena I. Perrakis, William Maxwell