Working Paper: NBER ID: w12065
Authors: Ronald G. Ehrenberg; George H. Jakubson; Jeffrey Groen; Eric So; Joseph Price
Abstract: The Andrew W. Mellon Foundation's Graduate Education Initiative (GEI) provided over $80 million to 51 treatment departments in the humanities and related social sciences during the 1990s to improve their PhD programs. Using survey data collected from students who entered the treatment and 50 control departments during a 15 year period that spanned the start of the GEI, we use factor analysis to group multiple aspects of PhD programs into a smaller number of characteristics and then estimate which aspects of PhD programs the GEI influenced and how these different aspects influenced attrition and graduation probabilities. From these analyses, we identify the routes via which the GEI influenced attrition and graduation rates and also indicate which aspects of PhD programs departments should concentrate on if they want to improve their programs' performance.
Keywords: Doctoral Education; Attrition; Graduation Rates; PhD Programs; Graduate Education Initiative
JEL Codes: I2
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| GEI (D50) | attrition rates (J63) |
| GEI (D50) | completion probabilities (C25) |
| GEI (D50) | mean time-to-degree (C41) |
| clarity factors (C38) | attrition rates (J63) |
| financial factor (G32) | cumulative attrition probabilities (C41) |
| better advising (G53) | attrition rates (J63) |
| program requirements (C88) | attrition rates (J63) |
| importance of finishing dissertations quickly (Y40) | graduation probabilities (C29) |