Estimating unemployment levels using a mix of three interview modes
This article contains an appraisal of two recently proposed methods of inference for sequential mixed-mode surveys.
The first method is a calibration adjustment to the survey weights so as to balance the survey response to a pre-specified distribution of the respondents over the modes. The second is a prediction method that seeks to correct measurements towards a benchmark mode. The methods are applied to the Labour Force Survey and are found to agree both in terms of assumptions and results.