Women are underrepresented in the news in just about every way imaginable.

There are fewer women’s bylines on front pages, fewer women sources quoted, and fewer women as the focus of stories.

Just 24 percent of the people heard or read about in print, radio, and television are female, according to a 2010 Global Media Monitoring Project study. And male reporters dominate hard-news beats, including 67 percent of stories about politics and government, 65 percent of stories about crime, 60 percent of stories on the economy, and 56 percent of stories about science.

As a woman who quotes people for a living, I’ve often considered what I ought to do about this problem. But first, I had to figure out the extent to which I’m contributing to it.

So I turned to the wonderful Nathan Matias, a Ph.D. student at the MIT Media Lab who has spent much time studying gender representations in the media. He helped me review a full year of my reporting.

I had a hunch that the results weren’t going to be something to brag about. I expected we’d find that I quote more men than women, but I also secretly hoped the gender representation in my work would be more balanced than most. I make an effort to find women sources. I’m a feminist! Surely that would show over the course of a year.

Not exactly.

We analyzed 136 of my stories published between Aug. 13, 2012 and Aug. 13, 2013 in newspapers and websites like The Washington Post, Nieman Journalism Lab, the Denver Post, Honolulu Civil Beat, Fast Company, Medium, the San Jose Mercury News, the New Haven Register, the Lowell Sun, and elsewhere. (We skipped bloggier items, essays, and aggregations, instead combing through larger pieces that usually required me to interview multiple sources.)

Over the course of the year I covered all kinds of topics — media, technology, the 2012 election, Hurricane Sandy, the mass shooting in Newtown, Conn., the Boston marathon bombing, immigration policy, patent reform, standup comedy, Internet culture, etc., etc.

In all those stories on all those topics, I mentioned 1,566 men and 509 women.

Here’s another way to look at it: 52 of my 136 articles quote no women at all. Zero. And although 63 percent of the articles I wrote mentioned at least one woman — even those articles mentioned more men. Yikes.

The grim thing is, my numbers make me pretty average. Of 2,075 people I mentioned in my reporting over the course of a year, about 25 percent were women. Internationally, 24 percent of news subjects are female, according to the Global Media Monitoring Project. (Actually, most women reporters tend to do a slightly better job than men when it comes to quoting women; 28 percent of news subjects in stories by female reporters were female compared with 22 percent in stories by men.)

A couple of caveats before we go any further. First of all, the analysis isn’t perfect. It’s based on absolute mentions of women by name, not a list of unique women quoted. (In other words, we counted not just the number of women I quoted, but the number of times I mentioned them.)

Plus, the gender tracker that Matias used after parsing all the names from my work couldn’t determine gender in a few hundred cases. (Some of the 411 “unknown” names that didn’t count toward mentions of men or women included Voldemort, Beethoven, Hemingway, Al Pacino, Jay Z, and Beyoncé.)

Still, it’s fair to say my work reflects a pervasive problem in journalism. So what do I do?

Underrepresentation of women in media is one of those topics that’s so big and so multifaceted that a lot of people don’t really know how to begin to talk about it, let alone do anything to make a difference. Matias once hoped he’d be able to clearly identify the worst offenders in journalism, and come up with an equation that would define benchmarks for a shared goal of gender representation in media. The issue turned out to be “much more complex.”

When I mention the depressing outcome of my analysis project to journalist friends, many of them rush to my defense: But you cover politics and media and tech. It’s not your fault those industries are so sexist. You’re quoting the president and the president is a man. You can’t help it!

The larger version of their argument goes something like this: We live in male-dominated world. Women are underrepresented at major institutions, universities, top businesses, and in government. Journalists are simply reflecting the imbalance that exists all around them.

I remember New York Times and CNBC reporter John Harwood saying something to that effect during a Harvard panel discussion about presidential election coverage last year. Moderator and New Yorker writer Jill Lepore had asked three male panelists about the relative dearth of women reporters on the campaign trail.

Harwood was thoughtful in his response, and ultimately concluded: “I don’t think it’s a journalism issue.” I think what he meant was it’s not uniquely a journalism issue but his even-keeled response rankled me.

Lepore’s outrage, in contrast, better sums up the way I feel:

Here’s the essential question Matias posed to me: “Is it your job to merely reflect what’s out there, or do you have other reasons to write in a more representative fashion?”

How you answer says a lot about what you see as the critical function of journalism. If you believe that journalism has the power to affect the way people think and act — how they vote, for example — you’d probably agree that it’s important for journalism to actually represent the people it serves.

“Inequality in one place is never an excuse for inequality elsewhere,” Matias told me. “Because if everyone points at everyone else and says they have a problem, too, then no one will ever have a motivation to create change.”

Of course, change is the hard part. First we have to be realistic about how journalists work. When I’m thinking about whom to interview for a story, I rarely consider the person’s gender. I want the best source. Period.

That means seeking out the smartest person who says the most interesting stuff. But ultimately this person must also actually agree to talk to me, to speak on-record, and to do it before my deadline, which may or may not be later today. (Maybe now is a good time to acknowledge that Matias happens to be a man, and he’s a great source for this story.)

Time and access are among the biggest factors that determine who ends up quoted in the paper. But reporters also pick sources based on whom they’ve interviewed before, suggestions from colleagues, or — when truly desperate — sources they’ve seen quoted elsewhere.

Sometimes I’ll write about a topic that’s esoteric enough that there is really only one person to quote. (Say, for example, the one guy I found who wrote a fascinating academic analysis about the Twitter account @Horse_ebooks and how it informs our understanding of authorship.) If you’re covering a congressional delegation and both senators are men, you’re likely going to end up quoting male senators more often than their female counterparts.

There are cases in which I actively consider the gender of my sources when I’m reporting. Obviously, if I’m writing about an issue that uniquely affects women, it makes sense to quote a woman.

But I also seek out women when gender may not directly inform the person’s expertise. If I find two immigration scholars — one a man and one a woman — with roughly the same credentials, I’ll often reach out to the woman first. But if the man calls me back and the woman doesn’t, I won’t hesitate to use the material from him. If I interview both of them, I’ll use the quotes from the person who said the most interesting stuff, regardless of that person’s gender.

Would my habits be different if I had a way to track the gender diversity in my stories in real time?

Should they be?

I’m inclined to agree with the New York Times editor who told Poynter that a diversity quota could end up being a “blunt instrument that could create as many problems as it solves.” There are all kinds of nuances that influence how a journalist reports and tells a story. But it’s also worth noting that only once has an editor asked me to go back and find more women sources. Just one time in more than ten years of writing and publishing stories for news organizations.

It was last year. I was helping edit a Nieman Journalism Lab series about the evolution of journalism education, and my boss asked me and a colleague to reach out to a bunch of smart people to see if they’d write something for us.

We contacted more than a dozen men and women journalists. After a couple of weeks, we had contributions from eight men and one woman. To his credit, my editor refused to begin publishing the series until we had more women voices in the mix.

In some cases, I was blunt when I asked women to participate: “We have so many men contributing and I’d really like to have women better represented,” I pleaded. I wasn’t shy about guilting them into it. In some cases it worked. We ended up publishing five pieces by women and ten by men.

I’ve repeatedly been told that women say “no” to these sorts of requests more often than men, that women are more likely to turn down an offer to speak on a panel, appear on television, grant an interview, or run for office. There are plenty of studies that detail the myriad ways in which women speak up less than their male counterparts.

So what?

Here’s that question again: “Is it your job to merely reflect what’s out there, or do you have other reasons to write in a more representative fashion?”

Doggedness is in any good journalist’s DNA. If we aren’t finding great women sources, at least part of the problem is that we aren’t trying hard enough.

So how can we try harder? Here are some easy first steps journalists can take: Ask existing sources about smart women they know in their field, pitch more stories that focus on women, and think critically about depictions of women in your own work and the work of others.

The good news is that people like Matias are building tools to help tackle gender bias in real time. He wants to create a plug-in for WordPress, for example, that could tell you the gender breakdown for your sources as you’re writing. He’s also inviting a small group of news organizations and data scientists to come together and talk about building an open gender-tracker this fall. (He’s worked on similar tracking projects already.) Matias is part of a team working on a tool called Follow Bias that is designed to assess diversity among people you follow on Twitter.

“Anyone who looks at the overall picture and thinks that there isn’t something wrong is probably missing something,” Matias says. “Give them a chance to interpret their situation and set goals. If you’re just monitoring the front page after you’ve already made all the decisions that feed into that front page, it’s going to be less easy to create change.”

There are reasons for cautious optimism. Earlier this year, the Open Gender Tracker team found that women have written 51 percent of all Global Voices posts, for example. (Then again, much of this work is unpaid.)

Gender bias in the news is a complicated problem, but change is possible. There are so many reporters filing so much and so often — and so many people who can reach wide audiences by publishing stories to online platforms outside of traditional media — who could make a difference. But both individual journalists and major journalism institutions must be willing to put forth effort.

We need to work harder to highlight a variety of voices, not just to improve gender diversity but to make our stories better. And isn’t that always the goal? Let everyone else interview Larry Sabato and this dude for the zillionth time. You may have to work harder but you’ll also be more likely to produce something exceptional.