Zombie Feminist Statistics: Catcalling

The Washington Post via The Straits Times:

"The Girl Scouts of America notes that one in 10 girls is catcalled before her 11th birthday. Surveys conducted by Cornell University's School of Industrial and Labor Relations and the non-profit organisation Hollaback! reveal that nearly 85 per cent of American female respondents report facing street harassment before age 17 and nearly 70 per cent before age 14. Almost 80 per cent of women report they were followed by harassers at some point.

Amanda Burgess-Proctor, associate professor of criminal justice at Oakland University and an expert in gender-based violence, sexual assault and crime, points out that girls and women of colour are more likely to experience street harassment and at a greater risk of being victimised physically or sexually."


The "one in 10 girls is catcalled before her 11th birthday" claim seemed fishy, so I decided to investigate it.

The Girl Scouts didn't provide more information on this claim, simply blandly stating "Two years ago, a study showed that one in ten American girls had been catcalled before her 11th birthday" (One in Ten Girls is Catcalled Before Her 11th Birthday. Here Are 6 Things Parents Can Do About It).

This article isn't even explicitly dated, but I inferred that it was referring to a 2015 report, Catcalling happens to most women between the ages of 11 and 17, which mentioned the anti-street harassment group Hollaback! (which, as a "wife, mother, sister, daughter" noted, "sends the message that if my sons make eye contact with, or say “hello” to, a woman they don’t know, they are a predator, or at the very least, a “creepy douchebag.”")

In any event, the study is presumably the one conducted with the help of Cornell, grandly titled Cornell International Survey on Street Harassment.

While the headlines are sexy, the methodology is anything but; while the full methodology is not available, we can see already that there are several problems.

In the US survey alone, right on the first page we are told that:

"Survey was not randomly distributed to a random sample of participants, and thus cannot be generalized in the same was [sic] as, say, a Gallup survey"

To wit, we can see that there are serious self-selection issues as the survey was given out in a very non-random way:

"Site leaders... Could send the links out however they wishes"

Even more damning,

"Sample: Relatively highly educated, moderately economically secure, and engaged with street harassment (63.1% have visited Hollaback! online)"

In other words, this is like going to a feminist convention, finding that most of the women there are feminists and then concluding that most women are feminists.

This is confirmed with their analysis of the difference between those who have visited Hollaback! online and those who haven't:

"We found significant differences on ALL types of harassment, age at harassment, and comfort talking about harassment. We cannot tell if these experiences led them to Hollaback!, or if being active in Hollaback! made them more sensitive to observing such experiences"

The fine print, as often, was left out in the rush to report sexy (and sexed up) headlines, and it is probable that this "one in 10 girls is catcalled before her 11th birthday" statistic is destined to become another zombie feminist statistic like 2% of rape claims being false, the gender wage gap and one in five college women being raped.

I also cannot find out how they define "verbal" harassment or "nonverbal" harassment (but it is safe from Hollaback!'s past record to assume that "hello" and "good morning" count). It is notable that nonverbal harassment does not mean exposing, groping/fondling or following - which are in their own categories. So respondents reported being "nonverbally" "harassed" in other ways (maybe being stared or even looked at)