I am working on a paper about social disorganization and domestic violence. My fixed-effect panel regression result shows that higher racial heterogeneity can significantly contribute to lower domestic violence occurrence. It is not only a significant result but also very robust. The classical hypothesis of social disorganization theory is that higher racial heterogeneity is related to weaker social bonds, and thus higher violence occurrence. How would you interpret it?
Hi Xinquan Xiang. To help you, it is probably best if you clarify what you mean by domestic violence (violence within the family or intimate partner violence?) and how you measure racial heterogeneity (is it a social context variable, or is it related to the races of the member of a couple?).
Thank you for the reply, Marta! In terms of the definition of domestic violence, it has something to do with my data source. I collected my DV data from annual reports of domestic violence organizations regarding crisis call numbers and shelter receivers. They define domestic violence primarily as violence within the family. The racial heterogeneity in my thesis is in a social context, denoted by the Shannon diversity index in each geographical unit.
A little more context of my research. I am researching social disorganization and domestic violence at the county level in the US. The county-level data on economic and social conditions are obtained from the American Community Survey, and the DV data is collected from annual reports of domestic violence organizations and merged into the county-level data.
I am not sure if this is of any help. Still, I would try to understand if there are any differences in those results concerning types of violence within the family (children-to-parent violence is a very different phenomenon from intimate partner violence, for example) or levels of this diversity index. This could give you different and more easily interpreted results. I would also check each parameter incorporated in the diversity index and see if these could explain the differences between your results and the literature. I think it is noteworthy to account for the size of this contribution. Albeit significant, is it large or small? Could sample size explain it? I hope it gives you some ideas. Good luck!
Thank you Marta! I really appreciate your ideas. I will look into the difference between the parameters I use and those used by literature.
Hi Marta, I suggest controlling for other variables I guess that high social heterogeneity may be linked to third factors such as rural/urban density, Instruction levels and other well known factors associated to IPV incidence. Alternatively, your data may grasp a changing dynamic between variables with respect to previous studies that may be to dated to allow a pertinent comparison.
- Badges
- Science topic
- Similar topics
- Filing