An experimental analysis of the impact of survey design on measures and models of subjective wellbeing

We analyse the results of experiments on aspects of the design of questionnaire and interview mode in the 2009 wave of the new UK Understanding Society panel survey. The randomised experiments relate to job- and life-satisfaction questions and vary the labeling of response scales, the mode of interviewing and the location of questions within the interview. We find a highly significant impact of these design features on the distributions of reported satisfaction in various life domains and some important impacts on the findings from conventional cross-section models of satisfaction.