Page 66 - Special Topic Session (STS) - Volume 4
P. 66

STS563 Pete Jones
                      From an LMS perspective, ONS are currently in the process of assessing
                  the impact of a stratified design on labour market statistics. Compared to the
                  equal probability sampling used for the current LFS, the proposed design to
                  have unequal sampling fractions across PCS strata will result in unequal design
                  weights for labour market estimates. In addition, the proposal to select a fixed
                  number of cases within the areas selected in each strata will further reduce the
                  effective size of the LMS sample. We are currently in the process of running
                  simulations to estimate the design effects that the proposed sample design
                  will have on LMS outputs. Depending on the results we will consider ways to
                  improve  the  sample  design  if  it  is  unable  to  deliver  the  current  level  of
                  precision obtained from the LFS collection.
                      A considerable challenge will be in developing the operational capacity at
                  ONS to carry out the necessary field work to ensure response to the survey is
                  maximised. Running a continuous survey of this scale is unprecedented up
                  until now, and the current aim is to be at full scale by 2022.

                  References
                  1.  de Leeuw, E. D., & de Heer, W. (2002). Trends in household survey
                      nonresponse: A longitudinal and international comparison. In R. M. Groves,
                      A. D. Dillman, J. Eltinge, & R. Little (Eds.), Survey nonresponse (pp. 41-54).
                      New York, NY: John Wiley & Sons
                  2.  de Leeuw E., Hox J. & Luiten A. (2018), International Nonresponse
                      Trends across Countries and Years: An analysis of 36 years of Labour
                      Force Survey data. Survey Insights: Methods from the Field. Retrieved
                      from https://surveyinsights.org/?p=10452
                  3.  Census Bureau: New Technologies in Census Data Collection. Part 2:
                      Developing an Electronic Questionnaire, 2016
                  4.  AddressBase is a database of address points in Great Britain, maintained
                      by Ordnance Survey
                  5.  Methodology of Statistical Population Dataset V2.0, Office for National
                      Statistics, October 2015
                  6.  The 2011 Census and Coverage Adjustment Process, Office for National
                      Statistics, July 2012
                  7.  Tore Dalenius and Joseph L. Hodges, Jr, Minimum Variance
                      Stratification, Journal of the American Statistical Association, Vol. 54,
                      No. 285 (March 1959), pp. 88-101
                  8.  ONS output areas are the smallest level of geography for which
                      estimates of population size are produced in England and Wales (on
                      average 125 households per OA)
                  9.  Predicting patterns of household non-response in the 2011census
                      http://www.ons.gov.uk/ons/guide-method/census/2011/the-2011-
                      census/processing-the-information/statistical-methodology/predicting-
                      patterns-of-household-non-response-in-the-2011-census.pdf

                                                                      55 | I S I   W S C   2 0 1 9
   61   62   63   64   65   66   67   68   69   70   71