I’ve been hemming and hawing over whether to post this in here or not for some time. Firstly because I’ve pretty much said my piece – and apparently pretty well – about Atheism+. I was going to let it go at that but then I had an ‘altercation’ on Twitter with an A+ supporter neck deep in feminist theory and saw an article in the New Statesman on A+ that said the following:
Less than a week old in its current form, Atheism+ is the brainchild of Jen McCreight, a Seattle-based biology postgrad and blogger at the secularist Freethought network. She has called for a “new wave” of atheism on that “cares about how religion affects everyone and that applies skepticism to everything, including social issues like sexism, racism, politics, poverty, and crime.”
And well, reading that following the altercation stuck in my craw because from what I’ve seen thus far, it’s simply not true. Scepticism is not being applied to sacred cows or presumptions arrived at within the echo chamber of their existing movements.
Put on the spot to come up with an example I went to the 1/4 rape stat that gets bandied about. Rather than try and convince me – as one would hope a rational sceptic who has arrived at a decision would do – I was instead subjected to a barrage of insults, accusations of being delusional and of warping the stats.
My reasons for being sceptical of the 1/4 ‘statistic’ are expanded on below for sake of completeness, but aren’t really relevant. The point was that they were being completely non-sceptical of their own claims and abusive rather than trying to back it up or explain how those stats were arrived at.
Scepticism towards everything?
Female population of England and Wales: 27,503,500 – 2001 census data. 
Reported cases of Rape (British legal definition) British Crime Survey: Worst recorded year 2010/11 – 14,624 recorded incidents.
UK female life expectancy (CIA World factbook): 82.25. 
Rounding down population to: 27,500,000 
Rounding up rape incidents: to 15,000
Rounding up life expectancy: to 83
As a percentage of female population 15k rapes is a yearly incidence rate of 0.05454545454%. 
Rounding up yearly incidence rate to 0.06%
Multiply percentage by life expectancy: 4.98%. 
Round up to 5%.
To get to 1/6 two out of every three would need to be unreported.
To get to 1/5 three out of every four would need to be unreported.
To get to 1/4 four out of every five would need to be unreported.
This does not seem likely as efforts to destigmatise and otherwise make it safe to make a rape report have had no particular discernible effect on reports.
RAINN suggests that around 50% of rapes go unreported. That would take us to 10%, which is ghastly enough, but far from the 1/4 that is often bandie about. Despite giving the claim every extra opportunity it doesn’t seem to measure up.
The BCS self-reported section, which has the advantage of not having anything to hold people back from reporting, but the disadvantage of no hard data, reports an incidence of 0.4 for the year 2010/2011 between ages of 16 and 59 (43 year span). Even if you presume that and multiply it up (as above) that’s still ‘only’ 17.2%. That’s enough to meet the 1/6 but not the larger extents and is a probably a gross overestimate.
Rape is clearly a terrible crime and far too prevalent. 1% would be far too prelevent.
It is simply not necessary to mangle the stats to make it seem worse than it is and this may, in fact, be deleterious to the cause through the effect of making people suspicious.
Now, it’s possible that the US (origin of these claims) is much more rapey than the UK or has lower reporting rates, but it seems unlikely the difference would be that marked.
 – UK population has been overestimated for some time but this appears to be about right for 2011/12
 – These are reported/recorded incidents. Female only. It does not take into account false accusations.
 – This is still going up. IIRC a recent BBC story said it was now 83.
 – This is the only case in which I have rounded down, rather than up. Everything else has been rounded to favour the high-incidence rate hypothesis.
 – Yearly incidence, not lifetime chance/number.
 – This isn’t realistic. Odds of being raped are not evenly spread throughout age. Again, it’s done to favour the high incidence hypothesis.