THE EVAPORATING INDEPENDENTS: REMOVING THE “INDEPENDENT” OPTION FROM THE NES PARTY IDENTIFICATION QUESTION

Abstract

In this brief paper we contend that the number and composition of individuals identifying with no political party may change substantially when the “Independent” option is omitted from the classic National Election Study (NES) party identification question. Data from two surveys of Vermont residents reveal that when the independent choice is not presented the percentage of self-identified independents drops significantly. Moreover, the relationship between age and partisanship varied depending on whether or not the nonpartisan choice was given. When the independent option was announced, a much higher percentage of younger than older respondents chose the option, but with the nonpartisan choice omitted, more older individuals actually opted to identify as independents. A similar analysis using nationwide survey data from the Roper Organization and NES supports our Vermont conclusions.