Unit and item nonresponse 2
|Convenor||Mr Peter Linde (Statistics Denmark )|
Although much is known about maximising postal survey response rates, survey clients and practitioners would benefit from a better understanding of how to maximise response for a fixed budget. Some methods are less costly than others. We present the results of a large experiment designed to test the impact of four design features on response rates and cost, using a fully-crossed factorial design: prenotification, postcard reminder, letter design, questionnaire length. We examine costs as well as response rates which allow us to explore the extent to which omitting one or more of these features could be a false economy.
In this presentation we explore the extent to which five response maximising features reduce the risk of non-response bias: number of mailings, pre-notification, postcard reminder, letter design, and questionnaire length. Using a large postal survey among the general population in England, we compare survey data with external data, we use a level-of-effort approach and we also have results from a randomised non-response experiment. We conclude the presentation with a discussion on whether these response maximising features are optimal for maximising data quality or whether a larger initial sample would be more cost effective.
Quality improvements frequently result in increased costs, e.g. in order to reduce nonresponse. However, nonresponse does not only depend on the number of attempts at making contact and mode, but also on the actual and experienced burden. Consequently, increased focus on reducing the experienced burden can be instrumental in increasing the achievement. The theme is discussed by examples from Statistics Denmark, where new digital solutions, better letters, prizes, reminders as to agreements, follow-up and interview training resulted in a higher degree of achievement as well as lower total costs.
One strategy of improving response rates is the use of advance letters. Advance letters can increase the trust in the survey and in the conducting institution, emphasise data protection and confidentiality issues. These aspects are particularly relevant when surveying sensitive topics like intimate partner violence.
Our analysis shows: advance letters have a positive effect on contact, cooperation and response rates.Under the advance letter condition:
- the number of refusals duet to a lack of interest was lower,
- the response rates of older respondents were higher, thereby increasing demographic accuracy,
- the reported intimate partner violence was higher.