Remi Piatek and Pia Pinger, "Maintaining (Locus of) Control? Data 
Combination for the Identification and Inference of Factor Structure 
Models", Journal of Applied Econometrics, Vol. 31, No. 4, 2016,
pp. 734-755.

The zip archive pp-data.zip contains the Stata do-files used to
produce the main data set from the raw data provided by the DIW, as
well as ASCII files (in DOS format) providing the regional and federal
unemployment rates and the price index used in our application. More
details about the files contained in the archive are below. Unix/Linux
users should use "unzip -a".

A web appendix called Piatek_Pinger_Appendix.pdf containing supplementary
material is also available from the JAE Data Archive.

1. Data sources

1.1 German Socio-Economic Panel (SOEP)

The primary data source is the SOEP, waves 2004-2011 of the main household 
sample and waves 2000-2011 of the youth sample. We used release number 28.
Access to the SOEP data must be obtained from the DIW-SOEP webpage (Berlin), 
see http://www.diw.de/de/diw_02.c.222836.de/datenzugang.html

1.2 Price index data

We used the German consumer price index to adjust wages to 2009 levels 
(Source: OECD). The data can be retrieved from 
http://research.stlouisfed.org/fred2/series/DEUCPIALLAINMEI#

1.3 Regional and federal unemployment rates

The data were obtained from the German Federal Employment Agency
(Bundesagentur fuer Arbeit) in December 2013 at
https://statistik.arbeitsagentur.de/nn_217688/Statischer-Content/Statistik-nach-Themen/Arbeitslose-gemeldete-Arbeitsstellen/Arbeitslose/Arbeitslose.html
Please note that these data are regularly updated retrospectively, so
that it is not possible to download the original data we used.
However, these data are provided in the ASCII files
unemployment_regional.txt and unemployment_federal.txt, as well as in
the Stata do-file youthdata.do used to process the data.

For the regional unemployment rate, we used the unemployment rate as a
share of the total civilian workforce in each German state for the
years 2004-2011 (Arbeitslosenquote bezogen auf alle zivilen
Erwerbspersonen - in Prozent).  For the unemployment rate at the time
of the education decision, we used the federal unemployment rate as a
share of the total civilian workforce for the years 1950-2011.

2. Stata Do-files

MAIN.do (executes sequentially the other do-files)
adultdata.do
youthdata.do

Before running the do-files, change the global paths in MAIN.do (lines
12 to 20) as indicated. Then run the do-file either interactively in
Stata, or in batch mode using "stata -b do MAIN.do" (Unix/Linux
users).

3. List of raw SOEP-data files required to run the Stata Do-files

SOEP (wide format data):

  bakind.dta
  bap.dta
  bapage17.dta
  bioage17.dta
  biobrthm.dta
  biobirth.dta
  bioparen.dta
  biosoc.dta
  cogdj.dta
  pkind.dta
  ppfad.dta
  qkind.dta
  rkind.dta
  skind.dta
  tkind.dta
  ukind.dta
  vkind.dta
  wkind.dta
  xkind.dta
  ykind.dta
  zkind.dta
  vp.dta
  wpage17.dta
  xpage17.dta
  ypage17.dta

SOEP (long format data):

  hgen.dta
  hl.dta
  pl.dta
  pgen.dta

Please address any questions to:

  Pia Pinger
  
  Department of Economics
  University of Bonn
  
  pia [DOT] pinger [AT] gmail [DOT] com