More justification for that skeptical attitude comes from the Heritage Foundation. It reports that the National Election Pool (NEP) survey
...did not ask respondents if they were married. So instead we found out  that “Men with Children” preferred the GOP 58% to 40% while “Women with Children” preferred Democrats 52% to 45%. But there is a big difference between “Women with Children” who are married and “Women with Children” who are not. For example , more than a third of single mothers are poor compared to just 6.4% of married couples with children. Single-parent families with children are almost six times more likely to be poor than married couples. How did marriage affect voters’ preferences in 2010? We don’t know because the NEP stopped asking respondents if they were married.
The survey is used by the Associated Press, ABC News, CBS News, CNN, Fox News and NBC News.
And it's just obscured a key political phenomenon.
While much is made of the "gender gap" in voting, people in the know realize this is usually a proxy for a marriage gap. Married women are more likely to vote Republican and unmarried women are more likely to vote Democrat.
But with this change to a poll, that useful bit of information will be more obscured.
Other linguistic abuses: the poll is using the wordy phrases "men with children" or "women with children" instead of the briefer term "father" or "mother." And the Heritage Foundation unwittingly uses the phrase "single parent" to hide the fact that single mothers are more prone to poverty than single fathers.
These media decisions concerning what demographics to measure shape the way in which politicians and pundits view the country. A polling environment which treats marital status as irrelevant to voting habits is not only the product of a crooked newsroom culture. It is a generator of a crooked popular culture.
(via The Brody File)