Can the GRE predict valued outcomes? Dropout and writing skill.

Area of Science:

• Educational Measurement
• Graduate Admissions

Background:

• Graduate programs are reconsidering the GRE due to bias concerns and questionable prediction of student success.
• This paper investigates the GRE's predictive validity for graduate student outcomes.

Purpose of the Study:

• To assess the GRE's ability to predict graduate student dropout.
• To evaluate the GRE Analytical Writing scores' relationship with graduate coursework performance.

Main Methods:

• Two studies were conducted using data from chemistry and computer engineering graduate programs at two universities.
• Statistical analyses examined the predictive power of GRE Quantitative (GRE-Q), Verbal, and Analytical Writing scores for dropout and coursework quality.

Main Results:

• GRE-Q scores significantly predicted dropout in chemistry programs but not overall in computer engineering programs (significant only for domestic students).
• GRE Analytical Writing scores showed a significant positive relationship with the quality of graduate students' written coursework.

Conclusions:

• The GRE, particularly GRE-Q, shows some predictive validity for graduate student outcomes like dropout and writing performance.
• Findings suggest that the GRE may retain value in specific graduate program admissions contexts, warranting nuanced consideration rather than outright dismissal.