Assessing methods for within-household selection in self-administered push-to-web surveys: an experimental comparison

For self-administered postal push-to-web surveys in countries such as the UK, where only address frames are available, it is challenging to devise a method for within-household respondent selection that has both desirable statistical properties and high compliance levels. 

Several methods are currently used, with no consensus about their relative merits. The aim of this research is to provide much-needed evidence about the relative effectiveness of alternative methods of within-household selection in the absence of an interviewer. 

Two common approaches, the any two method (non-random with potential cost advantages) and the next-birthday method (quasi-random), are compared via an experimental design. Each is believed to be the most practical of its kind. A factorial experiment was incorporated into the experimental design comparing two common styles of invitation letters, which could have an impact on the implementation of both within-household selection methods. 

Research questions address differences between the methods in population representation, precision of estimation, and costs. 

Overall, accounting for design effects, the any two method is cheaper to a small extent. Little variation in precision of estimates or the distribution of answers to attitudinal questions is observed. 

There is some indication that the proportion of the achieved sample in the next-birthday group has a higher proportion of younger respondents. 

The average educational attainment is consistent across within household selection approaches, though this appears to be moderated by the type of letter used.