Education and knowledge.

Via Ezra, interesting results from a new Pew Poll on knowledge of current events among the American public:

Distinct patterns emerge when these results are analyzed by key demographic groups. Education proves to be the single best predictor of knowledge. Holding all other factors equal, levels of knowledge rise with each additional year of formal schooling. At the extremes, these educational differences are dramatic: People with postgraduate degrees answer, on average, about 17 of the 23 questions correctly, while those who did not finish high school average only about eight correct answers.


But an increasingly educated American population isn't translating into a more informed public, because the level current-events knowledge is falling across all education groups:

But a deeper analysis of the five identical questions asked in both 1989 and 2007 reveals a surprising pattern: Americans didn't do as well in 2007 compared with how similarly-educated Americans performed in 1989. Across the board, scores declined significantly among college graduates, those with some college as well as for those with a high school education or less.

...What keeps the overall knowledge scores from declining is that college grads still know more than less well-educated Americans - even if they know less in absolute terms than college grads in the past - and there are proportionally more of them now than there were 18 years ago. Currently about 27% of the public are college graduates, compared with 17% in 1989. At the same time, there are fewer people who have only a high school education (50% now compared with 60% in 1989). Education still leads to increased knowledge about prominent people and events in the news - but it does not confer as much of an advantage now as it did in 1989.




Granted, it's a comparison across only five questions, which is problematic given that the distribution across types of question has been changing. (People seem to know more about issues and are less good at name recognition than in the prior survey.) Still, it raises pretty interesting questions. Despite Pew's language, the relative increase in knowledge across the the three education categories appears to have increased slightly between 1989 and 2006. Individuals with a college degree are 44% instead of 39% more likely than somebody with high school or less to answer four or five questions correctly. A simple explanation would be that as Americans get more education on average, more marginal people, in terms of being informed about and involved in society, are entering each education category, pulling down the category averages while the individual effect of education on knowledge of the world remains unchanged. Or maybe there's something more complicated at work.