skip to main content skip to footer

The Effects of Sample Restriction on Gender Differences

Author(s):
Lewis, Charles; Willingham, Warren W.
Publication Year:
1995
Report Number:
RR-95-13
Source:
ETS Research Report
Document Type:
Report
Page Count:
62
Subject/Key Words:
Sampling, Sex Differences, Statistical Analysis

Abstract

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 proposes 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. The 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. (62pp.)

Read More