Please enjoy another guest blog from our AmeriCorps Member, Evan Abramsky – we only have a month and a half left of his term, so please enjoy his thoughts while you can! Thanks for your service, Evan!
This past month, I was referred to several sets of research on intimate partner violence, all of which came to some very surprising and questionable findings. The studies all concluded that bidirectional violence (between both partners) was overwhelmingly the most common form of IPV reported by large samples of couples. Among couples that reported unidirectional violence, women were twice as likely as men to be perpetrators. Immediately, I greeted the findings with skepticism, and upon further review, discovered the researchers’ key flaw across their studies: a failure to validly measure IPV.
In constructing social science research, one of the trickiest projects is creating measures that are both reliable and valid. While the two terms may sound like synonyms, they are actually quite distinct. A reliable measure will record the same responses from an individual each time that individual is measured, without meaningless fluctuations. A valid measure, on the other hand, must actually measure the concept we are attempting to study. For example, if I’m attempting to measure gender and I decide to do so using an individual’s height, I would be using a measure this technically reliable (their height won’t change between measurements), but is completely invalid.
There are many reliable tests and surveys designed to study intimate partner violence, however, they are often invalid measures of the true weight of the violence and this poses serious problems to creating effective IPV policy. For many researchers, it is convenient and reliable to administer a survey of the types of violence employed by abusers in an attempt to understand IPV, which is the basic methodology of surveys like the Conflict Tactics Scale (CTS), and then use the results to draw conclusions on the frequency of domestic violence in a population. However, such a methodology is not valid, making the results of such a study questionable at best. This can be dangerous for IPV response policy, as it may overstate the relative occurrence of violence among women, while fueling arguments that are counterproductive to our understanding of IPV.
The results of such measures are invalid because they fail to fully grasp the nature of intimate partner violence. IPV is not the simple use of violence against an intimate partner; it is a pattern-based perpetration of coercive control that must be properly contextualized to understand. An isolated incident of a wife throwing a plate in anger, is not intimate partner violence. A husband repeatedly coming home and bombarding his wife with dishes, is. However, a survey like CTS would say that both are so.
It is also extremely important that measures account for more subtle and complicated forms of violence, such as sexual violence and psychological violence. These are often parts of a perpetrator’s pattern of control, depriving the victim of any decent treatment or sense of self-worth. Surveys that neglect to include sexual or psychological violence miss a major aspect of the IPV mechanism of control.
For those working in the intimate partner violence field or directly researching IPV, it is important that we remain vigilant of these distinctions in scientific methodology. It is also important that we apply new measures of violence that attempt to contextualize the frequency, severity, and initiation of IPV. These measures should take into account the violence’s situational aspects, woman’s history and perception, cultural aspects, and more comprehensive set of demographic factors.
Coercive control patterns are often hidden beneath the surface of seemingly non-violent actions—a glance, the cracking of knuckles, a sharp exhale—and those studying and working with IPV survivors and perpetrators must recognize these for the non-verbal controlling threats they are.
We must always ask the question: is this the whole story?