Examining an Extension of Johnson’s Hypothesis: Is Male Perpetrated Intimate Partner Violence More Underreported than Female Violence? Item Non-Response in the Project on Human Development in Chicago Neighborhoods

摘要

This paper examines two hypotheses about under-reporting in intimate partner violence data. The first hypothesis holds that significant amounts of under-reporting of intimate partner violence occur due to stigma. The second examines the empirical evidence behind Johnson’s (Journal of Marriage and the Family 57:238–294, 1995) contention that controversial findings of equal rates of intimate partner violence perpetration among men and women occur through a combination of heterogeneity in type of intimate partner violence and missing data. E.M. and Data Augmentation are used to correct for item non-response in the Project on Human Development in Chicago Neighborhoods. Strong support is found for general under-reporting; weak support is found for greater under-reporting of male violence. Keywords Intimate partner violence - Domestic violence - Missing data - Multiple imputation