ERIC Number: ED262081
Record Type: Non-Journal
Publication Date: 1985-Mar
Reference Count: N/A
Robustness of Regression Estimates for Ordered Dichotomous Variables.
Wolfle, Lee M.; Ethington, Corinna A.
The purpose of this paper is to examine the validity of regression estimates when skewed dichotomous scales are used as independent variables. When Pearson product-moment correlations are used to measure zero-order associations involving dichotomous variables, the resulting coefficients underestimate the true associations. As a result, using product-moment correlations involving dichotomous variables in regression equations apparently yields biased partial regression estimates. The analysis reported here was based on fifty sets of simulated data with 500 cases and four independent variables in each set. We found that tetrachoric and polyserial correlations for associations involving dichotomized variables yield more accurate estimates of the regression coefficients based on the underlying continuous data than did product-moment correlations for the same dichotomous variables. (Author)
Publication Type: Speeches/Meeting Papers; Reports - Research
Education Level: N/A
Authoring Institution: N/A