Skip to main content
SpringerLink
Log in

Comparing survey data with an official administrative population: assessing sample-selectivity in the IAB Establishment Panel

  • Published:
Quality & Quantity Aims and scope Submit manuscript

Abstract

The IAB Establishment Panel is a large annual establishment survey and serves as a data source for many empirical analyzes, e.g., on labor demand, wages and industrial relations but also for influential policy evaluations. We analyze the selectivity of the IAB Establishment Panel compared with the administrative population of all establishments in Germany to test whether the IAB Establishment Panel is biased towards “good employers”. By design the survey over-samples large, thus high paying and stable establishments. After flexibly controlling for elements of the sampling design, we do not detect any meaningful difference between establishments participating in the survey and the full population regarding key economic performance indicators. If anything, we observe a slight overrepresentation of stable and slow growing establishments. For applied research, the results highlight that “typical” econometric specifications may not be sufficient to control for selectivity.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1

Data source: Establishment History Panel, 2010

Similar content being viewed by others

Notes

  1. Bollen et al. (2016) argue similarly and highlight that most research in social sciences follows a ‘tradition’ of generally not using weights (even when using complex survey data) when modeling relationships in regression models. While it is clear that “not weighting” must not be confused with “not correcting for selectivity”, it seems that a lot of applied econometric work does not reflect selectivity issues of survey designs in a sufficient way.

  2. Detailed information about the IAB Establishment Panel can be retrieved from Ellguth et al. (2014) or Fischer et al. (2009).

  3. Approximately 70% of the interviews are conducted in face-to-face mode (see Ellguth et al. 2014).

  4. In practice, we cannot rule out that administrative and survey information refer sometimes to different entities of the establishment (or even the company). Therefore we conducted additional robustness tests for a subsample of establishments which are less likely to differ in their definition of the establishment. This check did not change our conclusions.

  5. We refer to the survey for 2010. There were changes in the classification of industries in other years. Modifications became necessary due to revisions of the underlying official classifications. We adjust our categories for some analysis accordingly.

  6. A detailed description of the BHP is presented in Gruhl et al. (2012). Our version differs from the publicly available data set in that it is the 100% sample.

  7. We checked whether this restriction is selective with regard to the probability of participation in the IAB Establishment Panel and found no correlation (see Table 4).

  8. But IAB Establishment Panel aims nevertheless at being representative for a given year's population of establishments as most of its questions refer to a reference date which is 30 June of the respective year.

  9. As empirically assessed by Janik and Kohaut (2012) there is selectivity in panel continuation, which can be explained by operative variables. E.g. when interviewers change, the likelihood of panel continuation shrinks by about 12% points. But the question in this regard is whether or not this is also related with the “quality” of the establishment or any other economic relevant concepts.

  10. The results are qualitatively similar for other years of the data.

  11. We do not present the constant in our tables as it is not very informative. Instead, we present baseline probabilities in Table 1.

  12. We show implications of such strategies in Sect. 6.1.

  13. In the administrative data about 15% of the individual wages are top-coded. Using the establishment-level median wage should be robust towards this data limitation. By contrast, average wages are severely biased.

  14. The spike at a daily median wage slightly above €10 represents establishments at which the median person is in a marginal employment contract (Mini-Job). In 2010, these are jobs with a monthly salary below € 400.

  15. Remember that we have a substantial amount of missing wage information, see Table 1. This is mainly due to the fact that the median wage in the BHP is computed for full time employees only.

  16. It is known that closures may reflect also other events like take-overs, retirement decision and not necessarily economic failure (Müller and Stegmaier 2015), but for practical reasons and due to data constraints we stick to closures.

  17. These definitions are based on employment flows and require a sufficient number of employees. Hence, the smallest size category is omitted and the number of observations decreases (see Table 1) when we analyze closures according to this concept.

  18. The results are qualitatively similar for all three definitions of closure. For simplicity we present the results only for the first definition.

References

  • Addison, J.T., Teixeira, P., Evers, K., Bellmann, L.: Is the erosion thesis overblown? Alignment from without in Germany. Ind. Relat. 55(3), 415–443 (2016)

    Article  Google Scholar 

  • Bartolucci, C.: Gender wage gaps reconsidered. J. Hum. Resour. 48(4), 998–1034 (2013)

    Google Scholar 

  • Beckmann, M., Kräkel, M.: Internal rent seeking, works councils, and optimal establishment size. Eur. Econ. Rev. 56(4), 711–726 (2012)

    Article  Google Scholar 

  • Biemer, P.P.: Total survey error: design, implementation, and evaluation. Publ. Opin. Q. 74(5), 817–848 (2010)

    Article  Google Scholar 

  • Bollen, K.A., Biemer, P.P., Karr, A.F., Tueller, S., Berzofsky, M.E.: Are survey weights needed? A review of diagnostic tests in regression analysis. Ann. Rev. Stat. Appl. 3, 375–392 (2016)

    Article  Google Scholar 

  • Bossler, M., Gerner, H.-D.: Employment effects of the new German minimum wage. Evidence from establishment-level micro data. IAB-Discussion Paper, 10/2016, Nuremberg (2016)

  • Bossler, M., Upward, R.: Employee turnover and the expansion and contraction of employers. In: Saridakis, G., Cooper, C.L. (eds.) Research handbook on employee turnover, pp. 305–346. Elgar, Cheltenham (2016)

    Google Scholar 

  • Davis, W.R., Pihama, N.: Survey response as organisational behaviour: an analysis of the annual enterprise survey, 2003–2007. Paper presented at New Zealand Association of Economists Conference, Wellington (2009)

  • Dustmann, C., Ludsteck, J., Schönberg, U.: Revisiting the German wage structure. Q. J. Econ. 124(2), 809–842 (2009)

    Article  Google Scholar 

  • Ellguth, P., Kohaut, S.: Tarifbindung und betriebliche Interessenvertretung. Ergebnisse aus dem IAB-Betriebspanel 2015. WSI-Mitteilungen 69(4), 283–291 (2016)

  • Ellguth, P., Kohaut, S., Möller, I.: The IAB Establishment Panel—methodological essentials and data quality. J. Labour Mark. Res. 47(1–2), 27–41 (2014)

    Article  Google Scholar 

  • Fackler, D., Schank, T., Schnabel, C.: Does the plant size–wage differential increase with tenure? Affirming evidence from German panel data. Econ. Lett. 135, 9–11 (2015)

    Article  Google Scholar 

  • Fischer, G., Janik, F., Müller, D., Schmucker, A.: The IAB Establishment Panel. Things users should know. Schmollers Jahrb. J. Context. Econ. 129(1), 133–148 (2009)

    Article  Google Scholar 

  • Fisher, S., Bosley, J., Goldenberg, K., Mockovak, W., Tucker, C.: A qualitative study of nonresponse factors affecting BLS establishment surveys: results. In 163rd Annual Joint Statistical Meetings (2003)

  • Göbel, C., Zwick, T.: Are personnel measures effective in increasing productivity of old workers? Labour Econ. 22, 80–93 (2013)

    Article  Google Scholar 

  • Gruhl, A., Schmucker, A., Seth, S.: The establishment history panel 1975–2010. Handbook version 2.2.1. FDZ-Datenreport 04/2012 (en), Nuremberg (2012)

  • Hethey-Maier, T., Schmieder, J.F.: Does the use of worker flows improve the analysis of establishment turnover? Evidence from German administrative data. Schmollers Jahrb. J. Context. Econ. 133(4), 477–510 (2013)

    Article  Google Scholar 

  • Hirsch, B., Mueller, S.: The productivity effect of temporary agency work: evidence from German panel data. Econ. J. 122(562), F216–F235 (2012)

    Article  Google Scholar 

  • Janik, F., Kohaut, S.: Why don’t they answer? Unit non-response in the IAB Establishment Panel. Qual. Quant. Int. J. Methodol. 46(3), 917–934 (2012)

    Article  Google Scholar 

  • Kohaut, S., Möller, I.: Führungspositionen in der Privatwirtschaft: Im Osten sind Frauen öfter an der Spitze. IAB-Kurzbericht, 02/2016, Nuremberg (2016)

  • Müller, S., Stegmaier, J.: Economic failure and the role of plant age and size. Small Bus. Econ. 44(3), 621–638 (2015)

    Article  Google Scholar 

  • Müller, S., Stegmaier, J.: The dynamic effects of works councils on labor productivity. Br. J. Ind. Relat, First evidence from panel data (2016). doi:10.1111/bjir.12200

    Google Scholar 

  • Schmidt, C., Zimmermann, K.: Work characteristics, firm size and wages. Rev. Econ. Stat. 73(4), 705–710 (1991)

    Article  Google Scholar 

  • Snijkers, G., Haraldsen, G., Jones, J., Willimack, D.: Designing and conducting business surveys. Wiley, Hoboken (2013)

    Book  Google Scholar 

  • Taris, T.W.: On selectivity of nonresponse in discrete-time multi-wave panel studies. Qual. Quant. 31(1), 79–93 (1997)

    Article  Google Scholar 

  • Tomaskosvic-Devey, D., Leiter, J., Thompson, S.: Organizational survey non-response. Adm. Sci. Q. 39, 439–457 (1994)

    Article  Google Scholar 

  • Trax, M., Brunow, S., Suedekum, J.: Cultural diversity and plant level productivity. Reg. Sci. Urban Econ. 53, 85–96 (2015)

    Article  Google Scholar 

  • Winship, C., Radbill, L.: Sampling weights and regression analysis. Sociol. Methods Res. 23(2), 230–257 (1994)

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Institute for Employment Research (IAB), Regensburger Str. 104, 90478, Nuremberg, Germany

    Mario Bossler & Jens Stegmaier

  2. University of Erlangen-Nuremberg, Nuremberg, Germany

    Gregor Geis

  3. Halle Institute for Economic Research (IWH), Halle, Germany

    Jens Stegmaier

Authors
  1. Mario Bossler
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Gregor Geis
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Jens Stegmaier
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Jens Stegmaier.

Additional information

We thank the research department “Establishments and Employment” at the IAB, Daniel Fackler and Mark Trappmann for helpful comments and discussions.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Bossler, M., Geis, G. & Stegmaier, J. Comparing survey data with an official administrative population: assessing sample-selectivity in the IAB Establishment Panel. Qual Quant 52, 899–920 (2018). https://doi.org/10.1007/s11135-017-0495-6

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11135-017-0495-6

Keywords

JEL Classification

Search

Navigation