Women Don't Know No Politics

It's a fairly standard, and often undiscussed for obvious reasons, finding in political knowledge research -- women do poorly as compared to men in tests of what people know.  I've blogged about this before.

I bravely return to this topic.

Why?  Because there's a study published in Journal of Politics that attempts to explain why gender differences occur.  It uses a strategy I've blogged about before, one previous studies have used, to help explain why women tend to do less well than men.  A number of theories have been posited to explain this gap.  Men guess more than women on such questions and the way we construct a political knowledge index tends to score "no answer" and an incorrect answer the same.  In other words, a of their few guesses will be right, thus giving men the advantage versus women who if they don't know, will simply say they don't know.  But a more plausible explanation is the kinds of questions we ask.

Or, as Kathleen Dolan writes in her JOP piece:

When we include political knowledge measures that ask for information on the present state of women in American politics, we see women’s traditional gender disadvantage wiped out. Women in the survey reported here hold as much information about women’s place in politics as do men. This offers support for the notion that knowledge levels are, in part, a reflection of the content of the items we employ.

To put it simply, methodology matters. Unfortunately there are no media variables in the study and I'd be curious to know if exposure to the news plays any role in creating differences, or evening the results, depending on the kinds of questions one asks to men and women.  It's an interesting question, one worth exploring.

What Women Know

Men know more about politics than women.

At least that's what you'd think based on a zillion years of political knowledge research.  The sex variable is a consistent, significant predictor of political knowledge in its various traditional formats -- and almost always in such a way that men score higher than women.

But, as you may know, I love digging up the exceptions.  I've found a number of studies that attempt to explain this effect, and I've blogged about them.  Here's another (abstract here).  Quite simply, the gender gap disappeared or even reversed when the measure of political knowledge turned to the more "practical" aspects of the concept, such as benefits and services, rather than the mere recall of names.

Previous studies I've pointed to have shown women do better when they're asked about politicians or public figures who happen to be female rather than all male.  Also men tend to guess more than women, so they often get it right while women are more likely to say "don't know."  Depending on how you code your political knowledge index, this can make a difference.

But this latest study is interesting in that it looks at the kinds of knowledge.  And the authors also explore the consequences in terms of voting to the political left or right.  Neat stuff both from a methodological perspective and, I suppose, if you're a feminist scholar.

Women in Social Media

New survey out about social media and women.  PDF is here.  Busy Saturday, so may try to read it later and comment.

Women, Knowledge, and Blogging

Women generally perform less well than men on tests of current events or political knowledge.

Methodology Alert

(Why? In part men are more likely to guess compared to women who say they "don't know." Both are coded the same way, with a "0" if wrong or don't know, a "1" if correct. If guys guess, there's a 50-50 shot they'll get lucky, thus increasing their scores. Some of it is the way girls and boys are raised, a socio-cultural explanation. Some of it has to do with a political culture skewed heavily toward men in gray suits, which improves male scores. Find a state with a woman senator and women tend to do better. Ask political questions that involve women in roles of authority, women do better. And so on.)

End Methodology Alert

So I was reading this excellent piece on female bloggers by Sheila Gibbons published in The Villager. It strikes me that as girls and women blog more, with shifting media diets toward online, we should see improvements in the political knowledge scores of females. Women will write about topics that matter to them, other women will read them, and hopefully the gender knowledge gap will disappear.

That of course presumes we ask the right sorts of questions that do not increase the gender gap, sometimes artifically. I'm not a PC guy. If the president and chief justice are men and knowledge tests include those legitimate questions about policy or office holding, women have no excuse in getting them wrong as compared to men. But I do believe the online world will flatten some of the traditional differences we've often found in the data. At least I hope so.

The Gender Knowledge Gap

Men know more than women. That's a fairly universal finding in tests of political knowledge, a gender gap most scholars prefer to leave undiscussed.

There are a lot of potential explanations for the gender gap in what people know about public affairs. A study that attempts to answer at least some of the reasons for the gap makes for interesting reading. To quote the authors, "the gender disparity on political knowledge is large, and the gap persists in multivariate analyses." Translation: even if you statistically control for lots of other stuff like education, it's still there.

Okay, but why?

Rather than delve into the sociological waters of how boys and girls are raised and trained and brought into political life, the authors instead look at something more simple -- measurement.
Men are more reluctant to say they don't know when asked a survey question. Pluck out the "don't knows" and half of the gender knowledge gap disappears. Poof. Gone. And because men will try and answer you get a "guessing effect" in the results. There's a lot more here in the study, so check it out yourself.

Basically, we men don't ask directions and we don't say we "don't know" if asked a knowledge question on a survey. In both cases, we're willing to guess.