Your Unconventional Wiz has always believed that Mother Nature was fair, distributing her blessings more or less randomly among her charges. For example, some of us are attractive while others are intelligent, but very few of us win the genetic lottery and end up both brainy and good-looking.

Well, if two sociologists are right, it may be that Mother Nature does play favorites, after all. Satoshi Kanazawa of the London School of Economics and his colleague, Jody L. Kovar, assert that beautiful people also tend to be smart people -- and vice versa.

In the July issue of Intelligence, the sociologists offer a theory to explain the confluence of beauty and brains. Their argument, in a nutshell: Intelligent men achieve higher status and marry beautiful women, who pass their genes on to their disproportionately attractive and smart kids, who win mates who are good-looking or brainy, and so on. Or at least that's what they put forth in the journal article, "Why Beautiful People are More Intelligent."

Their claim, which amounts to a Grand Unification Theory of Beauty and Brains, flows from four assumptions that guided them as they examined data collected in dozens of scientific studies conducted over the past three decades around the world, which means their findings apply more or less universally. (Beauty may be in the eye of the beholder, but researchers say that there is remarkable agreement in many cultures about what beholders find appealing.)

The first two assumptions posit that intelligence and beauty are inheritable. Those are genetic "gimmes" -- there's lots of data showing that the kids of smart people are likely to be smart, just as the offspring of attractive people tend to be lookers, too.

They go on to claim that more intelligent men are more likely to attain higher social and economic status than less intelligent men. That seems undeniable, and there is lots of supporting evidence that it's true. Longitudinal studies of draftees in World Wars I and II and soldiers in the Korean conflict found a strong association between IQ and occupational prestige in later life. For example, the study of Korea vets found that the 20 percent with the highest IQs earned twice as much as those in the bottom 20 percent and were far more likely to be engineers, doctors and teachers than to toil in a lower-wage blue-collar job. A more recent study of 13,248 10th graders found that general mental ability was the single strongest predictor of occupational attainment 13 years later.

Then Kanazawa and Kovar tackle the key assumption -- that beautiful women are more likely to marry men with higher status. Again, that seems obvious, but it's also empirically true. One longitudinal study of people in Oakland, Calif., found that attractiveness among women was the "strongest determinant of their husband's occupational status." A more recent national study in the United States also found that women's physical attractiveness "has a significantly positive effect on their household income, although it has no effect on their own income." Another U.S. study found that prettier women marry men who have significantly more education, another marker for higher status, Kanazawa and Kovar reported.

QED!, Kanazawa and Kovar claim. Logically, if these four assumptions are true, then beauty attracts brains, which begets beauty and brains. Again, the authors cite evidence to show this is true. Several studies have found that smarter men and women are also more attractive than other people -- including one study showing that better-looking West Point cadets were more successful, as measured by class rank, during their years at the academy than their less comely classmates.

What about smart women and attractive men? Brains are a plus for beautiful women, but they aren't the main attraction, just as being handsome is a bonus for high-status guys but not the main draw for women, according to Kanazawa. (And if you come up way short on all fronts? Well, there's always the Internet.)

Alert Wiz readers might recall that Kanazawa was featured in this column several months ago for his clever study showing there are lots of ways that people can be smart. Clearly, professor, you're very intelligent. So it follows from your theory that you must be one fine-looking man, right?

"No, I am not at all good-looking," Kanazawa wrote in an e-mail. "Given the imperfection in assortative mating, the correlation between intelligence and beauty should be far less than perfect. Beautiful people are more likely to be intelligent, and intelligent people are more likely to be beautiful. The correlation goes both ways, but is not perfect. There are always exceptions. I am one of them."

Familiarity with the homeless breeds sympathy and not contempt, reports Penn State University sociologist Barrett Lee, with two surprising exceptions: People who were once homeless themselves weren't particularly sympathetic toward others who found themselves on the street, and neither were people whose only exposure to the homeless came from the news media.

Lee analyzed data from 1,388 adults surveyed by Columbia University to determine how contact with the homeless affected their attitudes. He found that people were more likely to feel empathy toward the homeless if they had more direct contact, and they were less likely to see homelessness as the result of some personal defect.

About one in seven adults said in the Columbia survey that they had been homeless at some point in their lives, though the experience didn't soften their views on the homeless, perhaps because they believed that others should be able to follow their example and escape life on the street, Lee found.

Respondents whose only knowledge came from the media, Lee said in a statement released with his findings, were "more likely to view homeless people as dangerous, and they are less likely to support homeless rights to public space or express a willingness to sacrifice," largely because media reports in general tend to be unsympathetic to people living on the streets.

With President Bush's approval ratings plummeting in the polls, it may be time for another late-night visit with TV host David Letterman.

It worked four years ago, when Bush's favorability numbers spiked after he traded jabs on the Letterman show three weeks before the 2000 election, reports a research team headed by communications professor Patricia Moy of the University of Washington. In the week before Bush's Oct. 19 appearance with Letterman, his favorability rating among late-night television viewers stood at 56 on a scale of 1 to 100. During the week after the show, viewers' rating of Bush rose to 59, Moy and her colleagues reported yesterday at the annual conference of the International Communication Association in New Orleans.

Coincidence? Moy and her co-researchers, doctoral students Verena Hess and Michael Xenos, think not. Al Gore, Bush's rival in '00, got a helpful bump up after his Halloween appearance on "The Tonight Show with Jay Leno."

Sadly, laughter is fleeting and so are its effects. Two weeks after the Letterman appearance, Bush's ratings had dropped back to where they were before the show. (So look for Bush and Kerry to jockey for a late-night guest appearance just before Election Day.)

The study is based on responses from the University of Pennsylvania's National Annenberg Election Survey, which interviewed a random sample of 50 to 300 adults a night between Nov. 8, 1999, and Jan. 19, 2001.