Back to School

MSU researchers go beyond test scores and GPAs to predict college student success

Contact: Neal Schmitt, psychology, (517) 355-9563, schmitt@msu.edu; or Geoff Koch, University Relations, (517) 432-0924, kochg@msu.edu

Just as books can’t be judged by their covers, neither can incoming college students be assessed solely by college entrance exams and high school GPAs.

A new study by MSU psychologists suggests that a combination of biographical data, judgment measures and ability can be used effectively to clump students into a small handful of categories. Such sorting, in turn, someday may improve efforts to shape important parts of the college experience and intervene with students prone to run into problems, according to the researchers.  

This new measuring stick might help smooth the playing field and reshape the idea of post-secondary achievement, according to the study authors – Michigan State University psychology professors Neal Schmitt, Fred Oswald and a team of their graduate students.

The researchers surveyed more than 2,700 freshmen at 10 colleges and universities across the United States. In addition to providing test scores and high school GPAs, the students provided a large amount of biographical data and answered questions measuring their situational judgment. In examining the responses, the researchers found the students clumped naturally into five distinct categories.     
 
Eventually, armed with scores from such biographical and judgment measures along with traditional indices such as ACT or SAT scores and high school grades, teachers and parents might take different approaches “to aid these students to adapt to college life and optimize their college experience,” write the authors in an article that will appear next year in the Journal of Applied Psychology.

For more information or to receive a prepublication version of the paper, contact Schmitt at schmitt@msu.edu. For the full-length release, including descriptions of the five categories, see newsroom.msu.edu/site/indexer/2818/content.htm