ABSTRACT
In chapter 2, I have argued that the important variation found in the answers to consensual income questions undercuts the validity of the resulting income thresholds. Two reasons for the variation were that respondents had to make estimates for household situations that were not very well defined and with which they were unfamiliar in most cases. An obvious alternative, which avoids these problems, is to ask questions about the respondent’s own situation, instead of a hypothetical household type. For example, the Gallup question might be reformulated to: “How much income would you need under your conditions for your household to get along?”
