Gender Bias in the Measurement of Workaholism.

Past research suggests gender differences in workaholism might be due to differences in how men and women respond to the item content in workaholism measures. Using item response theory differential item functioning, we show women are less likely to report some workaholism items, leading to contamination. Specifically, women are less likely to report spending more time at work than other activities, and staying at work longer than others. We speculate that societal norms and practical restrictions on women's time results in lower endorsement rates for these items compared to men, and thus underestimates their workaholism. Results contradict past findings that men and women are similar in regard to workaholism and suggest women are in fact higher in workaholism than men. Limitations and future directions are discussed.