ERIC Number: ED448209
Record Type: Non-Journal
Publication Date: 2000-Nov-16
Reference Count: N/A
Alternative Approaches for Interpreting Alpha with Homogeneous Subsamples.
Roberts, J. Kyle; Onwuegbuzie, Anthony J.
Much of the current research concerning reliability emphatically suggests that researchers should gather their own reliability estimates when administering an instrument. It has also been recommended that data with low reliability be discarded. While some data obtained from instruments that originally yielded reliable results may be unreliable, it does not necessarily follow that the data are not useful to researchers. This paper contends that although data that are homogeneous might yield less reliable results than an inducted sample, these data should not be discarded until further examination of the data is conducted. Methods for determining data homogeneity are discussed in detail. The examination of the mean item variances, the variance of the mean item variances, and the squared standard error of measurement is encouraged, but, in fact, no specific guidelines have been developed for determining what is and is not an acceptable threshold for data homogeneity. Contains 3 tables and 11 references.) (Author/SLD)
Publication Type: Opinion Papers; Speeches/Meeting Papers
Education Level: N/A
Authoring Institution: N/A