Perpetrator and Victim Gender Patterns for 21 Forms of Youth Victimization in the National Survey of Children's Exposure to Violence
Jfl. When I first saw this study and the abstract, emphasizing "gendered patterns of violence" and "significant difference" between the genders basically everywhere they could:
I just rolled my eyes at first and didn't even bother looking deeper. I thought it was just another anti-male study with p-hacked and cooked figures that wouldn't get through a rigorous and neutral replication. I thought about just forgetting about it and defaulting back on what I wrote about in this thread:
incels.is
But, then I realized that the researchers used suspiciously little, well, none at all, actual numbers in the abstract, and I just thought how hilarious it would be if all those "significant differences" they were referring to, were just those cases where, sure, the differences are mathematically notable, but mean basically nothing IRL. By that I mean differences like 2:1, 3:1, or even 4:1. Those aren't what people think of when they hear "a statistically significant difference." They think of like 10:1 differences and more, like in one Reddit comment that once got reposted that was "I was raped by a girl, but for every girl like that, there's 10 men
!"
So, I looked deeper, and wouldn't you know it, it was even better than anything I had expected:
First off, the sample size was over four and a half thousand children and teens:
.
Right off the bat, we have the gender breakdown of perpetrating each of these 21 types of victimization:
And jfl, it's exactly as I've expected when I thought over that abstract and decided to check this out more. Mathematically significant statistical differences for the vast majority of them for sure, as shown by the four stars after most of them, but ones that are irrelevant IRL. Assault with a weapon is already just a bit under 25% female, literally every fourth such attack should be female-perpetrated according to this. The only category where female perpetration is below 20% is sexual assault, and that's probably the most skewed one since girls are taught ever since childhood that they can be victims of it while boys are only ever told they can be perptetrators of it, with thist division being much bigger than for any other crime there, so that I'm sure has skewed that. The rest are all 20%+ female and a very good chunk of them even 30%+ female, which, of course, means barely any difference in IRL contexts. With those numbers, for every two or three guys who do something, you should be able to find a chick who did that as well, and most importantly, it means it's completely nonsensical to classify women as being by default incapable of doing any of this.
Right after that, we have an even better table, the best one in the study probably, which is a version of the previous one with it being broken further down by the respective percentages of how much of the total each perpetrator/victim gender combination causes.
I love this. Female-on-female violence is much, much greater than the abstract ever implies, beating out male-on-female in a lot of the categories and basically tying it in some others, while sure, female-on-male is the smallest, but still in levels similar to the general male/female differences in the previous table, and overall, it's still a third of all inter-gender victimization, just like female-perpetrated stuff is a third of the total in general:
"But, men hit harder. They are more brutal, they cause more injuries!" some people might say, so let's look at the rates of injuries, shall we.
Even with how trigger happy the researchers were with finding "significant differences" in total numbers of those categories, they couldn't find any significant differences in injuries depending on the gender of the perpetrator for most of them. If injury rates are how the seriousness of these actions is measured, then there are no sex differences at all for most of those. The authors even bemoan the injury rate being basically the same for the one type of victimization with the highest overall rates, assault with a weapon:
.
"Well, OK, but what about fear! Men are so much scarier than women!" could be the next objection, and fear in relation to perpetrator gender is the next table there:
Now, this I think has more gender differences than the injury rate alone, which is crazy when you think about it and shows how much people's perception of something is determined by their feelings and not just material facts, but even then, the "significant" differences there just fail to be anywhere near what people would think when they hear that word. Over half have no significant differences anyway according to the researchers, and a good number of those who do are only significant at the lowest level of significance there, the .05 level. Of the two who are significant at the .01 level, peer/sibling assault has just a 0.09 difference, which even on a 1-3 scale isn't much, and assault with a weapon is nerfed by it having just about no injury difference, as mentioned above. Only robbery and sexual assault have significant differences in fear on the .001 level, and as I've already talked about, the latter is the one thing women are taught to fear their entire lives and what is always and very often brought up as literally the single worst thing that could ever possibly happen to anybody and that it only ever happens to them and will certainly happen to each of them.
That I think really says it all, there's some more tables showing bivariate correlations and multiple regressions of the various variables, but nothing shocking. Men are about 50% more likely to attack strangers, women are similarly more likely to attack relatives, but again, those numbers just completely fail to live up to the day and night difference in violence people imagine for the sexes.
Jfl. When I first saw this study and the abstract, emphasizing "gendered patterns of violence" and "significant difference" between the genders basically everywhere they could:
I just rolled my eyes at first and didn't even bother looking deeper. I thought it was just another anti-male study with p-hacked and cooked figures that wouldn't get through a rigorous and neutral replication. I thought about just forgetting about it and defaulting back on what I wrote about in this thread:
Women consistently commit 20-30% of serious assaults across Western countries at the least
I remember seeing an infographic a year or two ago that said that women are behind 25% of aggravated assaults (which, of course, is substantially more than what the average person seems to think) and ever since that, the topic had quietly been on my mind, so now I've decided to find some more...
incels.is
But, then I realized that the researchers used suspiciously little, well, none at all, actual numbers in the abstract, and I just thought how hilarious it would be if all those "significant differences" they were referring to, were just those cases where, sure, the differences are mathematically notable, but mean basically nothing IRL. By that I mean differences like 2:1, 3:1, or even 4:1. Those aren't what people think of when they hear "a statistically significant difference." They think of like 10:1 differences and more, like in one Reddit comment that once got reposted that was "I was raped by a girl, but for every girl like that, there's 10 men
!"So, I looked deeper, and wouldn't you know it, it was even better than anything I had expected
:First off, the sample size was over four and a half thousand children and teens:
Which I hope is a large enough sample size for anyone who would doubt this, because we have some really nice data here
.Right off the bat, we have the gender breakdown of perpetrating each of these 21 types of victimization:
And jfl
, it's exactly as I've expected when I thought over that abstract and decided to check this out more. Mathematically significant statistical differences for the vast majority of them for sure, as shown by the four stars after most of them, but ones that are irrelevant IRL. Assault with a weapon is already just a bit under 25% female, literally every fourth such attack should be female-perpetrated according to this. The only category where female perpetration is below 20% is sexual assault, and that's probably the most skewed one since girls are taught ever since childhood that they can be victims of it while boys are only ever told they can be perptetrators of it, with thist division being much bigger than for any other crime there, so that I'm sure has skewed that. The rest are all 20%+ female and a very good chunk of them even 30%+ female, which, of course, means barely any difference in IRL contexts. With those numbers, for every two or three guys who do something, you should be able to find a chick who did that as well, and most importantly, it means it's completely nonsensical to classify women as being by default incapable of doing any of this.Right after that, we have an even better table, the best one in the study probably, which is a version of the previous one with it being broken further down by the respective percentages of how much of the total each perpetrator/victim gender combination causes.
I love this
. Female-on-female violence is much, much greater than the abstract ever implies, beating out male-on-female in a lot of the categories and basically tying it in some others, while sure, female-on-male is the smallest, but still in levels similar to the general male/female differences in the previous table, and overall, it's still a third of all inter-gender victimization, just like female-perpetrated stuff is a third of the total in general:"But, men hit harder. They are more brutal, they cause more injuries
!" some people might say, so let's look at the rates of injuries, shall we.Even with how trigger happy the researchers were with finding "significant differences" in total numbers of those categories, they couldn't find any significant differences in injuries depending on the gender of the perpetrator for most of them. If injury rates are how the seriousness of these actions is measured, then there are no sex differences at all for most of those. The authors even bemoan the injury rate being basically the same for the one type of victimization with the highest overall rates, assault with a weapon
:You can just tell they were hoping for this most extreme category to have a big gender difference
."Well, OK, but what about fear! Men are so much scarier than women
!" could be the next objection, and fear in relation to perpetrator gender is the next table there:Now, this I think has more gender differences than the injury rate alone, which is crazy when you think about it and shows how much people's perception of something is determined by their feelings and not just material facts, but even then, the "significant" differences there just fail to be anywhere near what people would think when they hear that word. Over half have no significant differences anyway according to the researchers, and a good number of those who do are only significant at the lowest level of significance there, the .05 level. Of the two who are significant at the .01 level, peer/sibling assault has just a 0.09 difference, which even on a 1-3 scale isn't much, and assault with a weapon is nerfed by it having just about no injury difference, as mentioned above. Only robbery and sexual assault have significant differences in fear on the .001 level, and as I've already talked about, the latter is the one thing women are taught to fear their entire lives and what is always and very often brought up as literally the single worst thing that could ever possibly happen to anybody and that it only ever happens to them and will certainly happen to each of them.
That I think really says it all, there's some more tables showing bivariate correlations and multiple regressions of the various variables, but nothing shocking. Men are about 50% more likely to attack strangers, women are similarly more likely to attack relatives, but again, those numbers just completely fail to live up to the day and night difference in violence people imagine for the sexes.
