Take the 2-minute tour ×
Cognitive Sciences Stack Exchange is a question and answer site for practitioners, researchers, and students in cognitive science, psychology, neuroscience, and psychiatry. It's 100% free, no registration required.

I'm in the process of performing a meta-analysis that includes Grade Point Average (GPA) as one of the variables. For some purposes, it might be useful to have an estimate of the reliability of measurement for GPA.

  • What has empirical research estimated the reliability of GPA to be?
  • What factors influence the reliability of GPA measurement?

Additional points

  • I'm looking for specific coefficient values that I could use in a meta-analytic context. I am not interested in personal opinions about the value of GPA.
  • I can also see that reliability would vary based on how you define reliability.
  • I imagine that GPA would be more reliable and valid when it is obtained from a group of students at a single university. I can also see that GPAs based on more units of study would be more reliable.
  • I can also see that in some contexts GPA is an end itself, and therefore it would not be appropriate to adjust for reliability.
share|improve this question

1 Answer 1

up vote 2 down vote accepted

Bacon and Bean (2006) discuss the issue of the reliability and validity of GPA. They report the results of two studies that looked at the reliability of GPA in samples of business students at the same university.

  • Average intercourse grade correlations were typically between $\bar{r} =.18$ and $\bar{r} =.38$.
  • Estimates of reliability for GPA ranged from .67 to .94.
  • There is a simplex structure to GPA whereby correlations between GPA for a given year correlate more highly with GPAs of other years that are closer in time (e.g., correlations between consecutive year GPAS from years 1 to 2, 2 to 3, and 3 to 4 where .74, .79, and .74 respectively. In contrast the correlation between GPA year 1 with year 4 was .60.
  • Average intercourse correlations within disciplines tend to be higher than across course correlations.
  • In general, reliability increased as more courses/subjects were included.

The above estimates are based on correlations that use "+/-" system of grading. Universities that use straight letter grades should have lower reliability and those that use a 100 point numeric grading system should be more reliable.

References

  • Bacon, D.R. & Bean, B. (2006). GPA in research studies: An invaluable but neglected opportunity. Journal of Marketing Education, 28, 35-42.
share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.