The value of traditional cognitive variables for predicting performance in graduate speech-language pathology programs.

Background: Student performance on the Praxis examination and in clinical settings is considered indicative of university graduate program quality. Utilization of admission variables most predictive of graduate school performance in speech-language pathology is therefore critical. This study evaluated the relationship between cognitive variables (GRE scores, undergraduate GPA, and course-specific grades) and performance on the Praxis and between the cognitive variables and first-year clinical performance.

Methods: Admissions data for 230 students from four graduate programs in two states were analyzed. Participants were assigned to three groups based on Praxis scores (high performance, moderate performance, low performance/fail) and two groups based on clinical performance (low need supervision, high need supervision).

Results: Statistically significant relationships were found between all independent variables and the dependent variables (Praxis scores and first-year clinical performance). Participants' pass-fail status on the Praxis was predicted with a high degree of accuracy based on speech-hearing science grades, physical science grades, and the GRE-Total (GRE-T) scores alone. The low need supervision group for clinical performance demonstrated statistically higher group means for GRE-T and GRE-Quantitative (GRE-Q) scores compared to the high need supervision group.

Conclusions: The use of the GRE and American Speech-Language-Hearing Association-required science course grades in graduate admissions is warranted.