| Article ID | Journal | Published Year | Pages | File Type |
|---|---|---|---|---|
| 956514 | Social Science Research | 2006 | 28 Pages |
We illustrate the difficulty of measuring gender relations in surveys by comparing couple responses to survey items on the wife’s autonomy in various domains using data from 23 communities in India, Pakistan, Malaysia, Philippines, and Thailand. We employ an item response model to show that the level of women’s autonomy depends on whether wives or husbands are respondents and that the response categories do not have the same cognitive or semantic meanings to men and women. The disagreement between men and women varies across communities for reasons that are not easy to explain. The items also contain random measurement error that attenuates the correlation between spousal reports. We conclude that these survey questions are of limited utility for understanding differences in gender stratification across context.
