Lucy Pemoni, AP
The New Family Structure Study that suggested that children of gay and lesbian parents had difficult outcomes received uneven, sometimes disappointing, coverage in the media.

As was reported by Lois Collis in detail by this newspaper earlier this month, University of Texas scholar Mark Regnerus set off something of a firestore with a study comparing the lives of children of parents involved in same-sex relationships with the children of those from other family structures.

According to the Heritage Foundation, the study found, among other things:

"Young adults whose fathers had a same-sex relationship showed significant differences from their peers in intact families on 19 of the outcomes. For example, they were significantly more likely to have contemplated suicide, to have a sexually transmitted infection, or to have been forced to have sex against their will."

I've been spending some time evaluating the media coverage of this study, the New Family Structures Study, and I find much to be frustrated about, and some things about which it is worthwhile to feel encouraged.

In essence, for those willing to engage in more than one snap article about the study, a rich, detailed portrayal of this interesting study emerged.

More specifically:

First, I was pleased that most of the conservative news outlets I observed — places like the Heritage Foundation and the National Review — were circumspect in their reporting on the data, discussing its limitations. Most social scholarship has some limitations, and acknowledging those is a hallmark of good science. Kudos to these news outlets. (I thought this article in National Review excellent.)

Also, I may not be seen as a non-biased observer here, but the treatment of the study and its aftermath by the Deseret News met all standards of objectivity as I understand them. It is fair to say that its work ranked with the best on this study. Kudos to the always terrific work of my colleague Lois Collins and to the editorial staff.

As for the type of criticism, for example, Regnerus in his sample was unable to find many long-term relationships among the male same-sex parents of those he studied. Critics loudly pointed this and other things out. Some criticized the funding of the study. Fair enough. These types of limitations were covered well in the more conservative press and in the Deseret News.

Second, however, I found little willingness to write about this study in what is sometimes called the mainstream, national press. To be sure, I didn't search for the articles every possible way, but I found no mention of the study in either the Washington Post nor the Associated Press, for example. This is of a piece of what I have observed that national news often ignores issues relating to the American family.

The New York Times, for its part, buried its one newspaper article in the science section, but the article was fair-minded and informative — excellent journalism. (Again, assuming my search parameters picked up the full extent of the newspaper's coverage.) There was a blog post on the topic, as well.

I was most disappointed in the New Yorker magazine's online blog. The New Yorker, with its dedication to craft and its history of the best writing in the business, posted a blog on the topic that was too one-sided.

My biggest quibble with the coverage in the more mainstream press was the neglect of a companion study to the New Family Structures Study.

This second study by Dr. Loren Marks at LSU observed that previous research into the children of parents involved in same-sex relationships generally used convenience samples, and those studies generally concluded that the children of gay parents turn out as well as those from more traditional family structures.

These studies are intriguing, but convenience samples are, essentially, asking your friends, and perhaps their friends, to participate in a study. All such studies call into question how reliable and how generalizable they are to broader populations.

To be sure, convenience samples are useful when no other means of research are available, but their limits are well established. In a media research class I teach to sophomore students, I usually forbid convenience samples.

Basically, many mainstream and more liberal writers said Dr. Regnerus's study had its limitations, but few of those (though certainly not all), mentioned that competing studies are also limited. It left an unbalanced impression -- even implying that the science was settled and the New Family Structures Study was an outlier.

Third, as I scanned the Internet, I found it easy to learn much about this provocative study. Despite flaws in the news coverage, especially those of omission among the national, mainstream media, this study has generated useful dialogue for those willing to engage in it. And careful readers could learn much.

In short, we really don't need mainstream media to tell us what news is any more.

One last thing has generally been mostly missed in the broader media discussion that needs saying:

Given that current studies that look at the outcomes of children of parents in same-sex relationships have significant weaknesses, given weaknesses in funding, given the nature of society, and given the evolving nature of law, it remains very difficult to craft any study today about the children of same-sex relationship parents without running into severe limitations of scope, of method or of bias.

There's much that seems unknowable, for now. At least that's my take. To fully know the outcomes for children with parents in same-sex relationships will take a few decades of children growing in relationships that include same-sex marriage.

And that's worrisome. If you start with the premise that marriage is fundamentally and first about stability for children — rather than a contract that reinforces the rights of adults — then the growing policy debate over gay marriage runs up against the current limits of science.

Perhaps I force this analogy, but when science tests medical treatments, researchers go through rigorous pre-tests and models before they ever go into clinical trial. To do otherwise is unethical.

Yet, if my reading of these family studies is correct, we embark with somewhat limited knowledge on a massive, uncontrolled experiment when or if our nation chooses to adopt gay marriage generally.

If our country over time adopts gay marriage, maybe it will all turn out well for children. Let's hope so. But if the current science is any indication, it will take years for science to sort out fully whether that hope was misplaced. By then, it will be too late to stop the experiment.

Lane Williams teaches journalism and communication at BYU-Idaho. He is a former journalist whose scholarly interests include Mormon portrayals in the media, media and religion, and religion and politics.