This paper demonstrates the peculiar dependency of the correlation (fDX) between initial scores (X) and change scores (D) on the variance of final scores (SY2). Parameters of the primary distributions are manipulated to do so. The paper provides further support for the contentions of redundancy and spuriousness of rDX without relying on the claim of additiviry of errors.
References
1.
BereiterC.Some persisting dilemmas in measurement of change. In HarrisC. W. (Ed.), Problems in measuring change.Madison: Univer. of Wisconsin Press, 1963. Pp. 3–20.
2.
FleissJ. L.Comment on Overall and Woodward's asserted paradox concerning the measurement of change. Psychological Bulletin, 1976, 83, 774–775.
3.
GarsideR. F.The regression of gains upon initial scores. Psychometrika, 1956, 21, 67–77.
4.
LaceyJ. I.The evaluation of autonomic responses: Toward a general solution. Annals of the New York Academy of Sciences, 1956, 67, 123–164.
5.
LeebS.WeinbergS.Unreliability of difference scores: A clarification of the issues. Psychological Reports, 1977, 40, 931–935.
6.
OverallJ. E.WoodwardJ. A.Unreliability of difference scores: A paradox for measurement of change. Psychological Bulletin, 1975, 82, 85–86.
7.
OverallJ. E.WoodwardJ. A.Reassertion of the paradoxical power of tests of significance based on unreliable difference scores. Psychological Bulletin, 1976, 83, 776–777.
