Last week, I complained about a Wall Street Journal op-ed by Prof. Arthur C. Brooks of Syracuse University (subscription required) that framed social and economic mobility as a “he said”/”she said” issue rather than considering any empirical evidence. Here’s the key passage from his piece:

Consider the evidence. While 92% of conservatives believe that hard work and perseverance can help a person overcome disadvantage, only 65% of liberals think so…

Naturally, well-to-do liberals must be amazed at the gullibility of the millions of poorer conservatives who still cling to the idea of America’s promise of a better future through hard work and perseverance. Sunny conservatives of all economic classes may very well prefer to see things their way about America. Are conservatives naïve, or are liberals unjustifiably dour? Reasonable people disagree on this question.

But as Joel Wiles, a former Duke student, pointed out to me, the poll question says something different. Brooks makes the position of 35% of liberals sound unreasonable — of course “hard work and perseverance can help a person overcome disadvantage” (my italics). However, the actual question used in the survey, which was conducted by the public affairs school at Syracuse where Brooks teaches, adds a crucial qualifier — the word usually:

Would you agree or disagree with the following: While people may begin with different opportunities, hard work and perseverance can usually overcome those disadvantages.

The word “usually” shifts the meaning of the question from the possibility of overcoming disadvantage to the probability of doing so. And as I said before, the probability of moving upward in the United States today is far too low.

(For those who are interested, the data are here. I replicated the 65%/92% figures using the 2005 data, but the file also includes data from 2004 and 2006.)