Keywords: gender, PSID, survey response
Among married and cohabiting couples, the percentage of female respondents has increased substantially in the PSID (Panel Study of Income Dynamics) from 10% in 1970 to 57% in 2013. This shift in gender composition has taken place despite a formal policy that designates male head of households as respondents. We look at the relationship between this shift and changing family composition, labor force participation, income, education, and additional demographics. Our findings suggest that while home ownership and geography contribute to an initial uptick in female respondents, it is influenced in more recent years by increasing incomes and educational opportunities for women. We also model how likely a respondent is to be female, given certain individual and household characteristics. Finally, we explain how the measurement of aggregate variables, such as unemployment and average household income, may have changed if the gender composition of survey respondents had been kept constant over time. As gender roles continue to evolve, our research provides important insights into the quality of survey data and the changing role of women in households.