Measurement error in job evaluation and the gender wage gap

Job evaluation is used to establish pay for many workers in the United Stat es and has been used to measure the extent of pay discrimination. However j ob evaluations are subject to measurement error that can bias these estimat es. Using computed reliability ratios to adjust for measurement error in a study of Iowa state government jobs, we find that measurement errors exagge rate the implied extent of discrimination against predominantly female jobs by 34%-44%. Measurement errors also exaggerate the number of independent j ob factors which affect pay. A practical procedure far making these correct ions is illustrated.