U.S. flag

An official website of the United States government

Skip Header

Measuring marriage rates over time: Implications for empirical analyses

Written by:
Working Paper Number SEHSD-WP2016-13


Marriage behavior has been the subject of extensive research across many academic disciplines.  However, the marriage data used in this research have often been less than ideal.  This research fills an important void assessing the use of alternate sources of marriage data for empirical analyses utilizing new questions on marital history in the American Community Survey.  This work considers how data on marriage stock data (information on marital status) and marriage flow data (from information on marital history) might yield different results.  We examine differences in descriptive estimates as well as exploring the relationship between aggregate unemployment rates and marriage rates using an ordinary least squares regression approach and considering the implications for a difference-in-differences approach.  Results show marital status and marital history data yield divergent results in terms of sign and significance.  Differential effects by cohort and differential migration patterns by marital status may contribute to this discrepancy.  These results suggest using caution when conducting analyses using marital status data as they may produce false negative results or spurious false positive results.

Page Last Revised - October 8, 2021
Is this page helpful?
Thumbs Up Image Yes Thumbs Down Image No
255 characters maximum 255 characters maximum reached
Thank you for your feedback.
Comments or suggestions?


Back to Header