Williams and Zimmerman in 1998, when considering the reliability and validity of gain scores (g), proposed that combinations of pretest and posttest parameters favoring high reliability of gains are more likely in practice than are unfavorable combinations. It is argued here that this is an empirical question and that no support was given for their position. Further, those authors made a potentially misleading statement about the validity of the gain score. A final point is offered, namely, that the higher the pretest-posttest reliability, the lower the reliability of g.
Get full access to this article
View all access options for this article.
References
1.
MayK.HittnerJ. B. (1997) Reliability and validity of gain scores in three possible scenarios. Perceptual and Motor Skills, 85, 179–184.
2.
WilliamsR. H.ZimmermanD. W. (1996) Are simple gain scores obsolete?Applied Psychological Measurement, 20, 59–69.
3.
WilliamsR. H.ZimmermanD. W. (1998) A note on May and Hittner's three scenarios for the reliability and validity of gain scores. Perceptual and Motor Skills, 86, 664–666.
