Factor Analysis of Law School Grades LSAC
- Author(s):
- Boldt, Robert F.
- Publication Year:
- 1973
- Report Number:
- RB-73-42
- Source:
- ETS Research Bulletin
- Document Type:
- Report
- Subject/Key Words:
- Law School Admission Council (LSAC), Bias, Correlation, Factor Analysis, Grades (Scholastic), Law School Admission Test, Legal Education
Abstract
Grades of 116 law school students were obtained from transcripts, along with LSAT, Writing Ability and Background scores. The grades were factored, and an analysis of the goodness of fit of one- through four-factor systems indicated that the system of grades was essentially one factor in nature. A plot of factor loadings for the various courses made for the two- and three-factor systems indicated that a few courses were somewhat different than the rest, but the reason was not clearly identifiable. The two-factor system was decomposed into two uncorrelated components, one which correlated .96 with the single factor scores and the other which correlated zero with the single factor scores. The latter component had a multiple correlation of .03 with LSAT, Writing Ability and Background scores. These analyses produced no persuasive evidence for more than one factor. The results agree with those of a similar study on the business school context. The residuals produced were used to study the possible existence of instructor bias. Two instructors were found with residuals significantly different from zero. One was found to be a slightly easy grader, and the other was found to be a hard grader. (Author) (24pp.)
Read More
- Request Copy (specify title and report number, if any)
- http://dx.doi.org/10.1002/j.2333-8504.1973.tb00653.x