Business school professors have a knack for finding some pretty bizarre links between the personal lives of CEOs and the professional results at the companies they run.

Those who golf more than 22 times a year are linked with lower corporate performance, while those who run marathons have better results. Physical traits ranging from the width of a CEO's face to the depth of his voice have been correlated with better performance and higher pay.

Now, researchers are finding a link between the gender of a male CEO's children and how well his company behaves toward society.

The most recent issue of Harvard Business Review spotlighted new research showing that CEOs who have daughters get roughly 12 percent higher ratings for being socially responsible than those who have sons. And the median firm with a CEO who has a daughter spends an extra 13.4 percent of the firm's net income on corporate social responsibility programs, the study found.

"Having a daughter seems to make the top executives of publicly traded companies in the U.S. a bit softer, specifically in context of social responsibility," said co-author Henrik Cronqvist, a professor at University of Miami's business school, in an interview. "They seem to care more about others than just shareholders. Having a daughter seems to push the executives to care more about other stakeholders."

Cronqvist and his co-author, China Europe International Business School professor Frank Yu, used databases and Internet searches to come up with their own list of S&P 500 CEOs who have children, then looked at corporate social responsibility data from analytics firm KLD for the years 1992 to 2012. KLD rates firms on six categories-community, diversity, employee relations, environment, human rights and product-and analyzes companies on their strengths and weaknesses in such areas.

The pair found 379 CEOs for which they had both corporate social responsibility scores and information on their children's gender. They found that companies run by male executives with female children rated higher for diversity, employee relations and eco-friendliness. Of those, according to Cronqvist, the category where the impact was the strongest is diversity.

What's the reason for the link? In the paper, the researchers pull from past studies showing that women tend to attach more weight to the well-being of others. In addition, they draw on the idea that while we mostly think of parents influencing their children, the opposite is also true: Children have an impact on their parents' thinking and beliefs.

An effect known as the "female socialization hypothesis" has been shown to prompt fathers to adopt some of their daughters' values and thinking.

For instance, past research shows that congressmen who have daughters are likely to vote more liberally, particularly on reproductive rights issues. Other studies have shown that the daughter effect exists among parents generally becoming more politically liberal, as well as that of male federal judges in the United States who have a daughter and are pushed to the left on gender-related cases.

Cronqvist points to former Chief Justice William Rehnquist, who was speculated to have been influenced by his daughters in deciding to vote in favor of state governments having to abide by the Family and Medical Leave Act. British Prime Minister David Cameron mentioned his daughters when he announced his intent to force companies to publish their gender pay gap. "I want them to look back at the gender pay gap in the same way we look back at women not voting and not working - as something outdated and wrong that we overcame, together," he wrote in The Times in July.

The average age of CEOs in Cronqvist's study was 57, so many of their daughters are grown. As a result, their initial experiences in the workplace, their choice of college major or their professional aspirations may have begun to influence their fathers' thinking. Another study from 2011, for instance, found that after CEOs of Danish firms had a daughter, female employees' wages rose relative to their male peers, narrowing the wage gap.

It's unclear how having children vs. not having children, rather than whether the children were male or female, might affect a CEO's behavior. Because less than 5 percent of the CEOs had no children, Cronqvist excluded them from his study.

Another group that had a small sample in the study was female CEOs. Less than 4 percent of the executives studied were women - so while those women did have much stronger corporate social responsibility ratings than their male peers, it was such a small group that the findings are not reliable.

However, Cronqvist does use those figures to make an interesting comparison. Even male CEOs with a daughter tend to have only about one-third the social responsibility rating of CEOs who are women. Or as Cronqvist puts it: "Having the daughter seems to make male CEOs one third more female, with respect to the decisions being made."