Gilens asks 1,935 survey questions of national samples of the population between 1981 and 2002. The surveys are undertaken by Gallup, Harris and other reputable and independent pollsters. Each survey question asks whether respondents support or oppose some change in U.S. government policy. After compiling the answers, Giles then divides the population by income level and compares the preferences of those surveyed with the actual decisions by U.S. policy-makers.
The first part of his finding is encouraging. It confirms previous research that indicates that overwhelmingly unpopular proposals are unlikely to be adopted.
The second part is disturbing. When Gilens organized the responses by income group on policy questions in which well-off and poor Americans disagreed by 8% points or more, outcomes fairly strongly related to the preferences of the well-to-do but wholly unrelated to the preferences of the poor. Median-income Americans fare little better than the poor when their policy preferences diverge from those of the well-off.
The probability of a proposed policy change being implemented rises almost 30% as support among high-income respondents increases but only rises 6% as attitudes among median-income respondents shift from strong opposition to strong support.
Gilens concludes, " Most middle-income Americans think that public officials do not care much about the preferences of 'people like me'. Sadly, the results presented above suggest they may be right."