Do UBC Undergrads Think Atheists Are Less Trustworthy Than Rapists?

Yesterday a blog post claiming to be about a study that showed that UBC undergrads think atheists are less trustworthy than rapists was spread widely around Twitter, and had many people up in arms about discriminatory views about atheists.  The post in question claims that “even in Vancouver, and at UBC, the graph shows that respondents are more prejudiced against atheists than against rapists.”  This seemed to me like it was, at the very least, a very unusual result, and so I decided to look into the study in question and see what exactly it said.

Firstly, the study in question doesn’t actually make the claim that students “are more prejudiced against atheists than against rapists.” What it actually says is “people did not significantly
differentiate atheists from rapists.”   But even that claim I thought was very unusual, and I think that if we look at the construction of the study in question in a bit more detail, we’ll see that there are some significant methodological problems that make even the paper’s milder claim questionable.

What exactly did the authors of this study do?  They selected 105 undergraduate psychology students who had signed up to be part of the Psychology Human Subject Pool for extra course credit.  As described in the example above, the students were then presented the following scenario:

Richard is 31 years old. On his way to work one day, he accidentally backed his car into a parked van. Because pedestrians were watching, he got out of his car. He pretended to write down his insurance information. He then tucked the blank note into the van’s window before getting back into his car and driving away.  Later the same day, Richard found a wallet on the sidewalk. Nobody was looking, so he took all of the money out of the wallet. He then threw the wallet in a trash can.

Then they were asked if it was more likely that Richard was a) a teacher; or b) a teacher and X.  X was either “a Christian”, “a Muslim”, “a rapist”, or “an atheist(someone who does not believe in God)”.  The correct answer is always a).  So now we know how the study was organised, but what are the problems with it?

First of all, everyone participating was a psychology student; no students from other disciplines were included.  All of the students who were involved were participating for extra credit.  I don’t know if that factor would change the results in any meaningful way, but it certainly seems like a potential complicating factor.  The sample size was quite small, at just 105.  71% of the participants were female.  I don’t know if males or females would answer this question any differently, but it’s definitely something as a researcher that I’d want to try to minimise as a potential conflict.  Further, if I’m reading the paper correctly, each student was presented just one of the four potential options.  If the authors split up the options evenly, that means just 26 students saw each potential option (except for one option, which was seen by 27 students).  Of the 26 who saw the “a teacher and an atheist” option, some chose correctly (though, unfortunately, the paper doesn’t reveal the individual breakdown), meaning the number who incorrectly chose “atheist” is even smaller than 26.

Because such a small sample saw each potential answer, and the potential answers were each shown to different students, there is absolutely no basis on which to conclude that students were, on the whole, more likely to distrust atheists than rapists.  Not a single individual chose atheist over rapist, making it impossible to compare their answers.  The students who chose “atheist” or “rapist” chose it as the only option, not in comparison to the other.  So we have absolutely no information that would allow us to conclude that the answers are comparable.

Even if we put all that aside, there are still problems with the interpretation.  Perhaps, for example, respondents believe that teachers are unlikely to be rapists.  They may also believe that educators are more likely than the rest of the population to be atheists (at the university level this is definitely true, I’m not familiar with the figures for high school or elementary school teachers).  Therefore, an answer of “atheist” may indicate that respondents are zeroing in on the teacher part of the question, rather than the untrustworthiness of various groups of individuals. Maybe, as one commenter on the article linked above suggested, people simply believe that rapists have a fairly low incidence in the population in general, and thus are less likely to choose them on that factor alone.  If we spent much time thinking about it, I don’t think it would be difficult to come up with other similar possible explanations.

So the original headline, “UBC Undergrads Think Atheists Are Less Trustworthy Than Rapists” is hugely misleading.  It may be possible that such a statement is true, but the study in question doesn’t provide us with evidence that would lead to that conclusion.

* * *

Full disclosure of my own religious background – I’m agnostic, but I was a pretty devout Christian throughout high school.  Interpret my statements above in light of that information, if you think it’s relevant.