Posts Tagged figures

Surveys On Rape And The Need For Clean Stats

Recently we asked you what really pushes your buttons and makes you angry. You may have answered, you may not – I hadn’t, and didn’t intend to… but bugger it, my spleen needs venting. So here goes – I have a couple of thing that particularly piss me off: psychics are definitely one of them. Sexuality discrimination (in either direction) is very much a second. And another biggie? Bad stats, where it matters.

Now, I appreciate it might seem like a bit of a nothingness, after all. So some numbers get inflated to make it look like men are shitty to their girlfriends, or that knife crime is on the rise, or that more than half of teenage girls are pregnant – these kind of issues might seem relatively minor, if slightly sexist, sensationalist or downright stupid. Nobody’s getting hurt here, you might think, and after all more than 33% of statistics are made up, and over half of the remaining two thirds are meaningless cliche anyway. However, consider the following headline, from Tuesday’s Metro:

One in four women has been raped, a shocking new survey reveals

I think it’s fair to say the statement that 25% of women have been raped is a shocking statement. Truly. If it were, in fact, true. But is it? Well, it’s right there in the headline, and surely nobody running those figures could do so without being 110% sure of their accuracy, and at the very least they’d make sure they were about 4/3rds positive of the interpretation? Well, a little digging around and I was able to locate a summary of the survey this stat was taken from – it was an online survey of 1061 people in London, broken down into 349 men and 712 women. There’s no indication as to how that sample of 1061 people was put together, so any discussion of the stats has to be with the caveat that any potential bias is undisclosed. Interestingly, when looked at in terms of self-defined sexuality there were only 71 homosexual, 52 bisexual and 16 asexual respondents – yet the summary merrily extrapolates the data of around four dozen bisexual respondents into statements of comparative risk Read the rest of this entry »

, , , , , ,

5 Comments