ERIC Number: ED387523
Record Type: Non-Journal
Publication Date: 1995-Mar
Reference Count: N/A
The Effects of Sample Restriction on Gender Differences.
Lewis, Charles; Willingham, Warren W.
As strongly suggested by recent work, patterns of gender difference can change because of changes in the selectivity of the sample itself. This is a statistical influence connected with the distributions of female and male scores, rather than a substantive influence related to demographic characteristics of the sample such as age or ethnicity. It is, nonetheless, an important influence because gender differences that are partly statistical in origin can easily confuse possible implications regarding education and assessment. This report proposed a general model to account for the effects of sample restriction on gender differences. Simulations showed the model to be quite accurate in reproducing standard mean differences and other statistics in a restricted sample. Three primary contributing factors were identified: the range-restricting effects of sample selection, differential variability of female and male scores in the original sample, and the representation of females and males in the restricted sample. A test with actual data showed reasonably good consistency between trends predicted by the model and trends in gender differences that have been widely observed in advanced tests administered to select samples. (Contains 29 references, 9 tables, and 10 figures.) (Author)
Publication Type: Reports - Research
Education Level: N/A
Authoring Institution: Educational Testing Service, Princeton, NJ.