Skip to content
Social Research

Assessment of the utility of new interviewer observation variables for non-response weighting on the National Survey for Wales

Bookmark and Share
  • Release date: 19 June 2013
The objective of this small research project was to evaluate whether a more strategic approach to the selection of interviewer observation variables might result in more powerful weighting adjustments.

Given the substantial cost of household surveys and their use in understanding people’s attitudes and behaviours for the purpose of policy-making and evaluation, it is essential that every effort is made to ensure that the estimates produced from them are accurate.

While most interviewer observations are selected to improve the prediction of response propensity, the objective here was to also improve the prediction of specific survey outcomes. If weighting variables are correlated with both response propensity and with survey outcomes, they become more effective in reducing bias in survey estimates.


Tel: 0300 025 5996

Tel: 0300 025 8099

Keep up to date @StatisticsWales@StatisticsWales Sign up to our newsletterSign up to our newsletter