(Cliff Owen/AP) (Cliff Owen/Associated Press)

In Sunday’s New York Times, Maureen Dowd dispensed with all the speculation about whether Hillary Rodham Clinton would be running for president in 2016: “It’s such a silly question. Of course Hillary is running,” wrote Dowd.

Part of Dowd’s argument invoked a trait of Clinton’s every bit as consequential as her health, her ideology and her relationship with big Democratic Party donors. Her locks, that is:

Hillary jokes that people regard her hair as totemic, and just so, her new haircut sends a signal of shimmering intention: she has ditched the skinned-back bun that gave her the air of a K.G.B. villainess in a Bond movie and has a sleek new layered cut that looks modern and glamorous.

In a hot pink jacket and black slacks, she leaned in for a 2016 manifesto, telling the blissed-out crowd of women that America cannot truly lead in the world until women here at home are full partners with equal pay and benefits, careers in math and science, and “no limit” on how big girls can dream.

That, on balance, is a pretty positive review of Clinton’s appearance. What could possibly be wrong with that?

Something, concludes a study conducted for Name It. Change It., a non-partisan project of She Should Run, Women’s Media Center and Political Parity. The findings stem from an online survey of 1,500 likely voters across the country conducted in March by Lake Research Partners (a Democratic polling outfit) and Chesapeake Bay Consulting (a Republican polling outfit).

The goal of the survey was to figure out the impact of media references to the appearance of female candidates on their electoral prospects. Toward that end, the survey constructed a hypothetical congressional race between one Jane Smith and one Dan Jones. Participants read a profile of each and “heard” news stories about them. An elaborate experiment revolved around coverage of Smith:

In the stories about the female candidate, in addition to focusing on the education bill, we included various descriptions of her appearance. A quarter of the voters, a control group, heard a description of Jane Smith that did not reference her appearance in any way. A quarter of covers heard a neutral description; a quarter heard a positive description; and a quarter heard a negative description.

Celinda Lake, president of Lake Research Partners, says the brains behind the study didn’t sit down and riff on those descriptions. Instead, they pulled the descriptions from coverage in newspapers or prominent blogs. Here’s the positive description of Smith that was put before survey respondents:

In person, Smith is fit and attractive and looks even younger than her age. At the press conference, smartly turned out in a ruffled jacket, pencil skirt, and fashionable high heels.

Now for the negative:

Smith unfortunately sported a heavy layer of foundation and powder that had settled into her forehead lines, creating an unflattering look for an otherwise pretty woman, along with her famous fake, tacky nails.

That description derives from three stories in the mainstream media: This piece, from Huffington Post’s style section, which laments the “fake nails” that Michele Bachmann wore during a Republican presidential debate; this piece, also from Huffington Post’s style offerings, which addresses foundational issues; and this piece, from the Washington Post’s Style section, which takes on Bachmann’s tacky-nails question.

Whatever the provenance of the negative description, it scuppers Smith’s chances with voters. After taking in a news story that doesn’t make any reference to physical appearances, respondents rate Jane Smith and Dan Jones about even. Then, after they’re exposed to a story that contains negative information about Smith’s appearance, she tanks, trailing by double digits. Reports with positive and neutral descriptions also tilt respondents against the female candidate.

Lake says that the study is the “first of its kind that simulates the campaign setting — and the first of its kind in separating the negative impact of neutral, positive and negative comments on appearance.” But can we really glean anything substantive from a hypothetical race and descriptions pieced together from previous media stories? Yes, responds Lake. She notes that her people were watching the “dial” reactions of survey participants — that is, they were registering their reactions to various data points about Jane Smith. And when they heard something about appearance, they pushed the needle downward.

“When appearance started to be discussed, even positive comments on appearance, they started to dial down immediately,” says Lake, noting that the swings exceeded the study’s margin of error. “What we found is that the coverage was diminishing of woman being taken seriously. It greatly diminishes the seriousness of the woman’s candidacy.”

Just who were these folks delivering this verdict on the coverage of female appearance? Lake says that the respondents were recruited “randomly” online. The country’s polling establishment has long maintained that online polling has severe shortcomings when it comes to generating findings representative of the American public. However, Scott Keeter, director of survey research at the Pew Research Center, notes:

Generally speaking, well-designed experiments of this sort with large and diverse samples can be used to discover cause-and-effect relationships reasonably well. But they can also fail for reasons other than the representativeness of the sample. For example, because the impact of the experimental treatment (the characterization of a politician’s appearance) may be stronger in an online setting where people are focusing on the presentation more closely than they would be in, say, reading your column as they consume their daily diet of news. If that were true, the experiment would overstate the impact. On the other hand, if the effect is “dose related,” then repeated references to a candidate’s appearance during the campaign might make the effect greater than it is in a one-time experiment.

One of the study’s insta-critics is Robin Givhan, the longtime Washington Post reporter (and current contributor) who has specialized in analyzing the fashion choices of both female and male politicians and celebrities. Based on a quick look at the study, Givhan said the respondents didn’t come off as a “particularly neutral group. It sort of reads like they went into it to prove their theory, so they were looking for a particular outcome,” she says.

A more balanced approach, says Givhan, would have looked at how media comments on appearance affect male candidates, a set of scenarios that the study did not examine. “I don’t think you can make a judgment on how it’s damaging to women when you’re not comparing it to whether it’s damaging to men,” says Givhan. When asked why the survey didn’t approach men in the same way, Julie Burton, president of the Women’s Media Center, responded, “men do get appearance coverage, but it is in isolation. Women get it in quantity and it has an impact in how the voters view them.” On that same question, Lake responded, “Men aren’t covered in terms of appearance. There’s almost no coverage.”

To which Givhan responds, in effect: What about New Jersey Gov. Chris Christie?!

So Givhan favors not necessarily dialing back female-related appearance coverage, but doing extensive amounts for both genders: “When they’re standing in front of an audience in a public forum — both male and female — their appearance should be considered,” she says, noting that guy politicians may have been getting off “scot-free” for some time.

As the Women’s Media Center has criticized news outlets over the years for obsessing over the appearance of female politicians, says Burton, they’ve often received the explanation that newspapers’ style sections are logical repositories for such lighter stuff. “The real coverage is the politics section,” notes Burton, summarizing this strain of thought. The Huffington Post offered this very defense when pressed on the two stories on Michele Bachmann’s makeup and nails. A statement from Huffington Post:

No one takes this topic more seriously than we do. But seriously, come on, look at us in a reasonable context. We were founded and are run by a prominent woman. Our reporters on women’s rights issues have won numerous awards. We take women leaders and newsmakers with gender-neutral seriousness, and have from the moment of our founding. The stories cited here were from the Style section, not Politics.

The Style-Politics distinction is impressive to editors, publishers and the 2,000 Americans who are obsessed with news categorization and nomenclature. Outside of those folks, it’s meaningless. Notes Burton: “[R]eaders (and voters) often don’t see distinctions in media channels and sections. Many news aggregation and delivery platforms, including Google News and Huffington Post’s front page, lump together headlines from all channels and types of media, rendering moot any one outlet’s organizational chart.”

A better defense for the Huffington Post is that it publishes something on everything, including, for example, the fashion choices of a certain male politician.

Whether you believe in the Politics-Style divide; whether you believe this study has methodological or conceptual limitations; whether you want more appearance coverage of men, less of women or more of both; whether you believe that the range of clothing options available to women makes commentary on their choices inevitable, perhaps we can all agree that Clinton’s hair has had its day in the sun. As Burton is fond of noting, that topic is certain to fetch millions of Google hits. It also has its own topic page on the Huffington Post.