Bad Data, Worse Consequences

Who lies for you will lie against you.
Bosnian Proverb

Science (including methodology and statistics) is an interest (almost a love) of mine. And I get pretty angry when science is misused to serve ideologies. Unfortunately, this is something that happens frequently in the media. The reasons might be benevolent, but the consequences are not. Even if it really is an issue, using bad science obscures the issue. Even if biased studies and fabricated statistical data draws the “necessary” attention to an important topic, it creates resentment when people find out the data is manipulated. Personally, I don’t like to be lied to — no matter how beneficial the cause.

So I’m always happy to read articles that deal with the misuse of science. put out an interesting opinion piece by Christina Hoff Sommers (not sure why it is put under “opinion”) about some frequent misuses of statistics: “5 Feminist Myths That Will Not Die“. Well worth a read.

As she beautifully starts:

“If we’re genuinely committed to improving the circumstances of women, we need to get the facts straight”
Christina Hoff Sommers

I would go a bit further and say: “If we’re genuinely committed to improving the circumstances of women and men, we need to get the facts straight”.

Thing is, biased statistics and misuse of science paints a pretty ugly picture of the world — and frequently men are held responsible. Not the tiny minority of men who commit crimes, but all men (likewise, women who commit crimes are frequently ignored). To put it differently: Bad data have even worse consequences. Not only for the victims who will not get the help they actually need, but also for others. The only ones who profit — for a while — are ideologists who use the data to further their cause. Their cause, not the cause of the people they claim to work for.

And this is also a point the article makes multiple times, e.g.,:

Bad data not only undermine credibility, they obstruct progress by making it impossible to measure change.
Christina Hoff Sommers


A few hundred children is still a few hundred too many, but they will not be helped by thousand-fold inflation of their numbers.
Christina Hoff Sommers

and also beautifully near the end:

“But killer stats […] also promote bigotry. The idea that American men are annually enslaving more than 100,000 girls, sending millions of women to emergency rooms, sustaining a rape culture and cheating women out of their rightful salary creates rancor in true believers and disdain in those who would otherwise be sympathetic allies.”
Christina Hoff Sommers

I would further stress that these studies are used to constrain the view to men as perpetrators and women as victims — for issues that affect men and women. And you can’t have equality if you do not see the whole issue. Even if the discussion is limited to female victims and male perpetrators (why, if the problem affects men and women?), given that the “solutions” usually affect men as well, this is something that has to be discussed openly by women and men.

But nevertheless, really well-written article and highly recommended.