"A girl phoned me the other day and said "Come on over, there's nobody home." I went over. Nobody was home." - Rodney Dangerfield
***
The BBC reports that "It is a fact that a woman born in South Africa has a greater chance of being raped, than learning how to read", and this statistic is very widely cited.
While South Africa certainly has the highest rape rate in the world, this statistic sounded sensationalistic, and too "good" to be true.
According to UNICEF, total adult literacy in South Africa from 2000-2007 was 88%, and in the same period, female literacy as a percentage of male was 98%. A simple calculation yields a female literacy rate of 86.2%.
To be even more specific, the female youth (15-24) literacy rate in the same period was 96% (which is actually a sliver higher than the male one, at 95%).
This means that a woman born in South Africa 1-2 decades ago had a 96% chance of learning how to read by 2000-2007. If we take into account the fact that some girls in the 15-24 age group will learn how to read before hitting 25 (Christmas Cake Theory!) and more will learn how to read before they die, the chance of a woman born in South Africa 1-2 decades ago learning how to read in her lifetime is even higher. And with economic development and the end of apartheid, a woman born in South Africa today will have an even higher chance of learning how to read during her lifetime.
It is implausible that 86.2% of South African women have been raped, but a figure is still preferred.
Most rape statistics don't give a woman's lifetime rape probability, but instead talk about statistics like a woman being raped every 26 seconds or 36,190 rapes in 2007-2008.
However, that gives rise to the problem of double counting - some women will be raped more than once, while others will not be raped at all, so counting the number of rapes is not helpful for determining a woman's chance of being raped.
I tried to find a source for a woman's chance of being raped in her lifetime, but such statistics were hard to get. Charlize Theron says that the number is one in three and the Treatment Action Campaign and South African Medical Research Council (SAMRC) give the same figure. The Community Intervention Centre and the Soul City Institute for Health and Development Communication give a figure of one in two. The British Medical Journal (BMJ) is more conservative, with a one-in-four estimate.
Only Soul City and the SAMRC gives a source for its statistics. The former has the cryptic phrase "Vogelman, 1991", which some digging reveals might refer to Vogelman, L. & Eagle, G. (1991). Overcoming endemic violence against women in South Africa. However, I could not find the one in two figure inside (though I did find that "In a recent rape incident in Cape Town, a rape victim managed to persuade the rapist to wear a condom before the rape"). Another possibility is a "Study by Lloyd Vogelman of the University of the Witwatersrand from the Centre for the Study of Violence and Reconciliation as cited in the Cape Times, Oct. 24, 1991" but that trail went cold. Meanwhile, the SAMRC refers to a 1994 People Opposing Woman Abuse video called "Every 83 seconds", and I suspect that even if I managed to somehow find the video, it wouldn't give me a source for the statistic.
Whatever the case, it is clear that a woman born in South Africa has a much greater chance of learning how to read than of being raped. Unfortunately, it's easier to cite a reputable source like the BBC instead of checking suspiciously high figures, so no doubt the "more likely to be raped than to learn how to read" statistic will continue being circulated and propagated.
As someone else questioning the statistics remarks:
"where activism often falls short is a cavalier attitude towards and the use of "facts" and "statistics". By overstating their case, passionate activists de-legitimize their imperative"