Assessing the effect of social desirability on nativism attitude responses.

Attempts to measure and analyze public opinion attitudes toward racial/ethnic minorities often confront the "social desirability" problem: those who have prejudiced attitudes are rarely willing to admit them to surveyors. Instead, they may be more likely to give a socially acceptable answer rather an accurate reflection of their views. Previous research has clearly established that this effect presents a challenge for accurately measuring self-reported racial and policy attitudes that primarily affect African-Americans. It is less clear, however, how it might affect self-reported responses to attitudes dealing with Latinos and immigration. This study thus seeks to analyze the extent to which social desirability may affect survey measures of perceived levels of cultural threat (nativism). Results from two separate analyses using the Crowne-Marlowe "social desirability scale" and a survey "list experiment" demonstrate that social desirability is indeed a concern for accurately measuring nativism in the American public, but that it exerts an opposite effect from what has previously been observed: nativist attitudes tend to be over-reported in opinion surveys.