The Anatomy of an Incorrect Citation

This is about an article on asexual media representation by Sarah E. S. Sinwell which was featured in the Asexualities: Feminist and Queer Perspectives book. You can see my summary of the chapter here, and my other reactions to it are coming in a day or so.

This is a post about the importance of proper and accurate citation of sources, and how it’s possible for citations to go wrong in so many ways. This particular example was only a single sentence, but it’s a good example of a larger problem in the growing field of “asexuality studies”.

The story starts in Sinwell’s recent essay, where she starts off with what seems a fairly simple claim: “According to Anthony Bogaert, between 1% and 6% of the American population describe themselves as Asexual”. Seemed straightforward, which is why at first, I was confused – I was sure I’d read all of Bogaert’s work, how could I have missed that!

Of course, it turns out I hadn’t missed it, it was just flat out wrong – it’s an incredibly confused misinterpretation at best, or deliberate deception at worst. Even then, it took a little  bit of detective work to figure out where she was pulling that claim from, since it’s most likely from several different bits of data that have been mashed together.

The first number – the 1% part of the 1-6% estimate – seems to most likely be linked to the most famous of Anthony Bogaert’s papers (published in 2004), which found that 1% of respondents to a British national health survey stated that they did not experience sexual attraction to any gender, and could perhaps be considered asexual. But this sample comes from Britain, and had nothing to do with self identity, unlike Sinwell’s claim.

The other likely source for the 6% figure – the other half of the supposed 1-6% range –  appears to come from an informal internet poll attached to a CNN article about asexuality, in which 6% of respondents (of unknown nationality, though most were probably american) chose the “Asexual” option. This poll is mentioned in Bogaert’s 2006 paper, but it’s not actually from any study found by him. In fact, Bogaert expressly states that that number is most likely incorrect.* Casual internet polls are already an unreliable source of information in general, and the fact that the poll was linked to an article about asexuality means it likely had a pretty significant responder bias in favor of asexual people; polls like that are also completely unrandomized and absolutely cannot be used to make generalizations about the general population in the way Sinwell’s statistic is claiming to.

Then, of course, there’s the Poston & Baumle study (2010), which is listed in the bibliography but never mentioned in the actual paper, which actually did find a range of figures – from I think something like .7-6.1%  – which actually was from a national American probability sample. Unfortunately, it also defined asexuality as something like “anyone who chose “not sure” as their sexual orientation, haven’t had sex yet, or identified as anything other than heterosexual, homosexual, and bisexual”, which is kind of meaningless. Despite being the closest to Sinwell’s claimed percentage range, this study had no connection to Anthony Bogaert and didn’t really look at self-identity.

The mysterious transformation then happens somewhere in Sinwell’s head, so that a British 1% figure from a national population study and a 6% figure from an American convenience sample somehow got combined into a 1-6% percentage range from a supposed American national population sample. How does that even happen? This isn’t advanced math, it’s basic statistics and a little reading comprehension. She even directly cites the Bogaert study even though, if you actually read it, it completely undermines her false statistic.

Author’s Update from 2018: 

I was re-reading some discussion of another article recently, and now I have a pretty good guess where Sinwell got those incorrect numbers from! It looks like Sinwell just repeated the wording and citations made in a seperate article by Breanne Fahs, which has been notably been called out for incorrectly citing these same statistics before. This is why it’s important to always double check the original sources instead of relying on indirect references!

/end of update

This was just one sentence in one paper, but it’s representative of a much larger problem in the field of asexual research. This kind of casual misinterpretation and misuse of statistical “findings” to make whatever claims an author wants regardless of the actual original findings is a rampant problem in current asexual research, as is the process of just making up random claims and then citing semi-related works in an attempt to pretend they are validly sourced (for example, things like citing AVEN as the source of your definition of asexual even when your definition is completely different from AVEN’s clearly stated positions, as other articles have done). Misrepresentation of research results and poor citation problems happen in all fields, but I feel like they are even more rampant in the field of asexual research, and the problem seems especially noticeable in self-identified queer and feminist approaches to ‘asexual’ research.

Part of the problem is likely the small size and relative novelty of the field of asexual research – a lot of poor quality research that would be ignored in other fields gets published and gets a lot of attention simply because there isn’t anything better out there yet. Another part of the problem is the fact that most of these articles are probably getting peer reviewed by people who know nothing about asexuality or asexual research, so things that are obviously and blatantly incorrect to someone familiar with the field may seem reasonable to an unfamiliar reviewer who doesn’t have the time/effort to spare to actually factcheck things. And in the case of some queer theory approaches to asexuality, they are often being written or reviewed by people who don’t have much knowledge of statistics either, which I suspect may make these kinds of errors even more prevalent.

I have no idea if this error or ones like it are the result of simple carelessness or outright intentional deception. But either way, it’s a problem that queer and feminist asexual research really needs to solve if it wants to be taken seriously.


*For reference, the original statement from Bogaert was:

when the interest of the popular press surrounding the issue of asexuality reached its height in late 2004, CNN conducted an Internet poll asking people to self-identify their sexual orientation. A sizable proportion (6%) of the nearly 110,000 respondents reported that they identify as asexual (“Study,” 2004). The point of presenting this result is not that this percentage accurately reflects the true proportion of asexuals in the population—it likely does not— but rather that a sizable minority are choosing to identify with a term that is not part of the traditional academic and clinical discourse on sexuality and sexual identity.”

5 thoughts on “The Anatomy of an Incorrect Citation

  1. Misuse of statistics is a rampant issue throughout the humanities (and even the social sciences). When I was taking my overview of Japanese modern historiography class, there were A LOT of dodgy statistical uses–either pulling data from multiple places and pretending that all the data sets were measuring the same thing (mixing governmental numbers vs. religious movement numbers from different time periods, for example), stripping numbers from their context (like the infamous 60% of Japanese men are asexual thing, which is actually that somebody did a small-scale survey of men between the ages of 16 and 23-ish in an urban area and asked them if they identified as herbivore men), or just straight up fabricating numbers. So, yeah, it’s an issue, and I wish people were better at it.

  2. Thank you for noticing this problem and pointing it out, Sennkestra. It’s a pretty big problem and it’s important that you noted it. The percentage of people who are asexual in a population (such as in the USA) is something we unfortunately do not have good statistics on, lots of different studies with different figures, etc, and I hope the state of research on asexuality improves soon, with more conducted, so that this particular detail, the number of people who are asexual, a statistic everyone is striving to cite, can be one cited with more… credibility, I suppose. That we can avoid people thinking up to 6% of people are asexual thanks to just one lame internet poll, or thanks to one researcher having cited said poll… :P

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s