The Hurricane Katrina Study
One of the focal points of our discussion this week was the feminist claim that domestic violence, which is always understood to mean violence by men against non-violent women, will increase during the COVID-19 pandemic.
Sources such as the New York Times, and many others, have reported confidently that the social distancing measures being used to slow the spread of the Coronavirus pose a particular threat to women, who are vulnerable to violence in their homes and will have even fewer resources than usual as various government offices and services shut down during the pandemic. While Paul, Tom, and I agreed that domestic violence likely will increase as a result of COVID-19, we do not agree that only (or even primarily) women are at risk, or that the demonization of men as perpetrators is a helpful approach to the problem of family violence.
Such claims about domestic violence highlight once again the serious problem of feminist research. It is increasingly common to hear from feminist sources that all natural disasters, from global warming to bushfires and now to a viral pandemic, have a disproportionate impact on women.
During the recent Australian bushfire crisis, for example, a feminist anti-violence activist named Sherele Moody provoked anger when she announced at a news conference that men returning from fighting the fires were known to abuse their intimate partners. Relying on a study by Monash University researcher Debra Parkinson, Moody asserted that advocates such as herself knew from research, and from the experience of the Black Saturday bushfires in Victoria ten years earlier, that women were subject to much harsher, much crueler domestic violence after natural catastrophes.
In fact, the study she cited, which was based entirely on anecdotal accounts from a small research sample, provided no quantitative evidence whatsoever for the alleged relationship between firefighters, or fires in general, and violence against women. Yet no news outlet researched the alleged study by Parkinson to expose the thinness of her findings.
Unfortunately, lack of evidence—or even evidence contradicting the feminist thesis–never stopped a feminist narrative or slowed our society’s willingness to “believe the women”; it is now widely established, and frequently repeated, that natural disasters lead to an increase in violence against women.
One particular study is cited over and over again in the feminist papers produced on this subject. Called “Intimate Partner Violence and Hurricane Katrina: Predictors and Associated Mental Health Outcomes,” the study was originally published in 2010 in a reputable journal called Violence and Victims (25.5). Based on data from a larger, population-based, representative study of 445 married or cohabiting persons living in the 23 southernmost counties of Mississippi at the time of the hurricane, the study claimed to have found a 98% increase in violence against women in the hurricane’s aftermath; and this sensational number has become a favorite go-to statistic for feminist advocates.
But what the study actually found—almost never accurately reported, and not prominently emphasized by the study authors themselves—is far less sensational, and of little use to feminists.
Here’s what the abstract of the study says, in part:
“Reports of physical victimization increased from 4.2% to 8.3% for women, but were unchanged for men.”
That seems straightforward, and I assume that at least some researchers don’t read past the abstract.
When one reads into the actual study, however, one discovers that this 98% increase for women is not quite as dramatic as it sounds. 11 women (4.2%) had reported physical violence before the hurricane and 17 women (8.3%) reported violence after the hurricane. With such relatively small numbers, even an increase of 2 or 3 accounts of “violence” would have translated into a significant-sounding percentage increase. Physical violence was deemed to include any of “hitting, slapping, shoving, punching [or] throwing objects,” and it is impossible to determine in any useful way from the survey data the severity or frequency of the violence recorded. Throwing a magazine is clearly not of the same magnitude as throwing a heavy glass object; just as a restrained shove does not compare to a forceful punch in the face. Without far more knowledge of the alleged incidents and their contexts, it is difficult to know what to make of the statistical increase.
Equally, or even more, interesting, however, is the abstract’s statement about “reports of physical victimization” remaining “unchanged for men.” Most people reading that statement would likely assume that the number of men reporting violence was lower than the number of women both before and after Hurricane Katrina. Wrong. The body of the study reveals that 18 men (11.7%) reported experiencing physical IPV in the 6 months before the hurricane, compared with 21 (10.0%) following the hurricane.” Being the duffer that I am, unfamiliar with social scientific research, I do not understand these numbers. I am not sure why the evident increase in the number of men reporting violence was reported in the abstract as “unchanged,” and I do not understand how a larger number of men reporting after the Hurricane can have yielded a lower percentage (I could find no explanation in the study itself, but perhaps it is there somewhere). More research into the study’s conclusions is surely in order, but suffice it to say here that the actual results of the study are not particularly helpful to the feminist cause: MORE men than women reported physical victimization both before and after Hurricane Katrina. So much for the gendered impacts of natural disasters.
Hundreds of feminist studies about violence against women are being produced every year, informing public policy and influencing public opinion. Nearly every sentence in a feminist study cites an abundance of other studies, all ostensibly proving the same things about male power and female victimization. How many of them are based on misrepresentation? How many are falsely reported, or partially reported? This is why we need armies of honest researchers to take back academia from the ideologues.