13Nonresponse weights Sample-based nonresponse methods Split set sample into weighting classesEstimate weighted response rates in each classNew weight is 1/RR
14Nonresponse weights Response rates different in each weighting class Means for major survey variables must differ between each classMeans for major survey variables must be same for R and NR within each class
15Nonresponse weights GHS and EFS Based on Census-link studies 1991 Target nonresponse in specific demographic groupsSampling frame informationInterviewer observations
24When to use weights Always (whenever you can) Problems with presentation /interpretationestimates / sample sizes / SEsNR & probability weights tend to increase variances
25When to use weightsStat packages (e.g. SPSS) don’t always deal with weighting correctlyScale weights to average 1Stata/SAS survey estimation proceduresCalibration tends to reduce variances
26Conclusions Weights combination of probability, NR, calibration Required for unbiased estimationMay require specialist software for correct hypothesis testing
27Current Issues Use of 2001 Census data Census-linked NR studies Change in Pop. Controls (back-weighting)Integrated survey (CPS)LFS:Attrition weightingLocal LFSNumber of controls
28References Weighting for non-response, Elliot, D. NM17 Grossing Up - when and how, Butcher, B. SMB 14The presentation of weighted data in survey report tables, Elliot, D. SMB 38Using weights in regression analysis: A comparison between SPSS and STATA packages, Insalaco, F. SMB 45Developing a weighting and grossing system for the GHS, Barton, J. SMB 49Evaluation nonresponse on household surveys, Foster, K. GSS Methodology Series 8.Report of the Task Force on Weighting and Estimation, Elliot, D. GSS Methodology Series 16.