Publications & Resources
Net-Shift Analysis for Comparing Distributions of Test Scores
Erick L. Lindman
This paper suggests a technique for analyzing distributions of test scores. The technique is intended for comparing distributions of scores made by groups of pupils on standard tests with distributions made by other groups of students upon the same tests. Briefly, it does this by identifying the percents of student scores which must be shifted to an adjacent cell (interval) to make the two distributions exactly the same.
The technique is intended to reveal changes in score distributions which may occur when different teaching methods are used. For example, a new remedial program might cause a shift of low scores toward the mean without altering the distribution of scores above the mean. The technique also provides a more complete comparison between the distribution of scores made by a selected group of pupils and a norm group.
Lindman, E. L. (1968). Net-shift analysis for comparing distributions of test scores (CSE Report 45). Los Angeles: University of California, Los Angeles, Center for the Study of Evaluation.