|
Original sample in wave 1: Two-stage random sample including two partial populations
1st stage: selection of 300 postal code areas serving as primary sampling unit (PSU) for both partial samples. The sampling probability of the postal code areas depended on the particular area size in terms of number of residents (probability proportional to size/pps).
2nd stage, sample I: Benefit Units were drawn from register data of the Federal Employment Agency (BA) with July 2006 as refrence date. The size of the gross sample per PSU was dependent on the PSU size in terms of the relative proportion of benefit recipients within each postal code area (probability proportional to size/pps). The gross sample comprised an average number of N=100 per postal code area.
2nd stage, sample II: at first, a residential buildings sample was drawn from a commercial database stratified according to a stratification index provided with the database at a ratio of 4:2:1, thus identifying households with a low, medium or high status. Interviewers of the surveying institute visited the selected buildings. Where the particular building accommodated several households, one was selected by the institute, which was later interviewed. The gross sample comprised N=100 households per postal code area.
BA Refreshment in wave 2 (sample III):
The samples drawn from wave 1 were continued in wave 2. Additionally, a refreshment sample for sample I was drawn from register data of the Federal Employment Agency. Benefit Units were selected if they were in receipt of unemployment Benefit II in July 2007 but not in July 2006 – therefore representing new transitions into receipt of Unemployment Benefit II. These new households were drawn using the same procedure and PSUs as in wave 1.
BA Refreshment in wave 3 (sample IV):
Again a refreshment sample for sample I was drawn from register data of the Federal Employment Agency in wave 3. Benefit Units were selected if they were in receipt of Unemployment Benefit II in July 2008 but not in July 2006, or July 2007 – therefore representing new transitions into receipt of Unemployment Benefit II. These new households were drawn using the same procedure and PSUs as in wave 1.
BA Refreshment in wave 4 (sample V):
Again a refreshment sample for sample I was drawn from register data of the Federal Employment Agency in wave 4. Benefit Units were selected if they were in receipt of Unemployment Benefit II in July 2009 but not in July 2006, July 2007, or July 2008 – therefore representing new transitions into receipt of Unemployment Benefit II. These new households were drawn using the same procedure and PSUs as in wave 1.
In wave 5 the sample was repopulated with two additional samples using a two-stage random sample including two partial populations.
1st stage: selection of 100 postal code areas serving as primary sampling unit (PSU) for both partial samples. The sampling probability of the postal code areas depended on the particular area size in terms of number of residents (probability proportional to size/pps).
2nd stage, sample VI: For sample VI the sample was drawn from the registers of the local residents regstrations offices. The 100 postal codes where matched to 96 municipalities. Individual addresses where drawn from the sample frame within the municipalities using a systematic random sampling procedure (interval sampling). Addresses where drawn for individuals born in 1992 or before. 144 addresses where drawn in each PSU.
2nd stage, sample VII: Benefit Units were drawn from register data of the Federal Employment Agency (BA) with July 2010 as refrence date. The size of the gross sample per PSU was given as the product of the fixed sample size (individuals per PSU) in the general population sample and the quotient of the benefit recipient rate per point and the national benefit recipient rate.
BA Refreshment in wave 4 (sample V):
In wave 5 another refreshment sample for sample I was drawn from register data of the Federal Employment Agency. Benefit Units were selected if they were in receipt of Unemployment Benefit II in July 2010 but not in July 2006, July 2007, July 2008, or July 2009 – therefore representing new transitions into receipt of unemployment Benefit II. These new households were drawn using the same procedure and PSUs as in wave 1.
|