الفهرس 
Measuring Domestic Violence and its Cost Using a Latent Variable Model
AbstractOnly 14 pages are availabe for public view
 |  | from | 87 |  |  |
| | | from | 87 | | |
Abstract
Violence against women, especially Spousal Violence, is a threatening worldwide problem, that has its serious effects not only on women, but on the entire humanity. Almost 1 in 3 women, in Egypt and globally, has ever faced physical or/and sexual violence by their intimate partner. The consequences of domestic violence against women can be dramatic, as it negatively affects women’s general wellbeing including physical, mental, emotional and sexual health, also the economic burden of domestic violence cannot be ignored. Children are not an exception either, as they can be highly affected by domestic violence.
Both domestic violence and its cost are multidimensional constructs that can only be measured via a number of observed variables. In this study, we use latent variable models to measure both domestic violence and its consequent cost, and also to assess the relationship between them, in addition to assessing the main determinants of spousal violence within the structural model. The proposed model is fitted to data from the 2015 Egypt Economic Cost of Gender-Based Violence Survey (ECGBVS) using Mplus software.
A latent trait model is fitted through applying the item response theory approach. Domestic violence is measured onto two levels. At the first level four dimensions of domestic violence are measured; namely economic, physical, psychological and sexual violence; are measured via observed items. The second level employs these latent dimensions in having an overall measure of domestic violence. Cost is also measured onto two levels, whereas the first level measures three types of cost: cost on woman’s quality of life, cost on children, and economic cost, while the three types are combined in the second level.
The overall latent trait model also estimates the relationship between the two main latent variables of interest; domestic violence and its consequent cost. The effect of a number of socioeconomic covariates on domestic violence is studied within the model.