GESIS Leibniz Institute for the Social Sciences: Go to homepage
Programme for the International
Assessment of Adult Competencies

Network modules of the PIAAC Leibniz Network

Module 1 "Research" aims at evaluating PIAAC competence measurements in the context of other measures of comptence and general cognitive ability. Alternative measures, e.g., of skill mismatch or adult further education, will be developed and evaluated. Further more, aquisition and use of competencies will be analyzed in more detail.

Module 2 "Knowledge Transfer" aims at distributing PIAAC data and PIAAC research results into the scientific community, into politics and into practice.

Module Overview

Goal and Research Questions

The goal of this sub-project is to investigate whether and to what extent competence tests, for example those implemented in PIAAC, measure something different than only general cognitive skills.

Competence tests are intended to model, beyond general cognitive skills, the outcomes of learning processes typically taking place in the context of schools. One example of such a learning outcome is reading comprehension. Tests for general cognitive ability, in contrast, are expected to measure more general skills that remain stable over time and are not learned in the school context, for example reasoning.


To investigate the research question, the project will draw on data from PIAAC competence assessments of Literacy and Numeracy and assessments of general cognitive competencies from PIAAC-L. Within this sub-project, a computer-based numerical series test, which serves as a central measure of general cognitive ability, will be presented to 800 persons in PIAAC-L.

To answer the research question whether competence tests do not only measure general cognitive skills but also model the outcomes of educational processes, the project will investigate

(1) to what extent competence measures can be distinguished from of general cognitive skills (with regard to dimensionality). For instance, whether success in competence tasks and general cognitive ability depends on a common skill, or whether different skills are involved.

(2)  whether the relationship between processing parameters such as processing time and success is comparable across general cognitive ability tests and competence tests.

(3) whether the relationship to background variables, e.g. age, duration and type of general and vocational education, is comparable across general cognitive ability tests and competence tests.

Participating Researchers

At DIPF: Prof. Dr. Frank Goldhammer (Project Management), Lena Engelhardt

Co-operation: Prof. Dr. Jürgen Baumert (Max-Planck Institute for Human Development); Prof. Dr. Oliver Lüdtke, Prof. Dr. Olaf Köller (IPN); Prof. Dr. Jürgen Schupp (DIW); Prof. Dr. Claus H. Carstensen (University of Bamberg)

To the project website at DIPF

The tasks of Module 1.2 from the PIAAC Leibniz Network deal with the PIAAC items on adult education and training. Part of the work focuses on further developing adult education items in order to measure distinct aspects of adult education and reasons for non-participation. The aim is to better understand participation in adult education and to study individual barriers to participation in adult education. Furthermore, new questions on informal learning were developed in order to additionally answer questions related to non-formal learning.

Selected questionnaire items were tested in terms of methodological and questionnaire-related aspects in a cognitive pretest by the GESIS cognitive lab. The results were used to further develop the questionnaire items.

The items on participation in adult education and on barriers were tested as part of the PIAAC-L questionnaire (wave 2016), the questions on informal learning were included in the GESIS Panel (wave 12/2016). This will enable the project to gain knowledge on which individuals take part in adult education, which groups are excluded from training activities and the reasons associated with it. These results will inform policy recommendations with the aim to enable all groups of the population to take part in training activities.

Participating Researchers

Natascha Massing: GESIS – Leibniz Institute for the Social Sciences (GESIS)

Britta Gauly: GESIS – Leibniz Institute for the Social Sciences (GESIS)

Prof. Dr. Harm Kuper: Freie Universität Berlin (FU)

Prof. Dr. Josef Schrader: German Institute for Adult Education - Leibniz Centre for Lifelong Learning (DIE)

Andreas Martin: German Institute for Adult Education - Leibniz Centre for Lifelong Learning (DIE)

Education and skill mismatch can be seen as major problems of modern labor markets. Both phenomena represent a discrepancy between workers’ education and job requirements. While educational mismatch, or qualification mismatch, refers to formal qualifications, skill mismatch concerns the actual skills possessed by the worker (and required by the workplace).

The first step of the analysis is based on PIAAC 2011/12 data and examines the individual and institutional determinants of overeducation in Europe, in the light of the recent economic crisis. The project investigates cross-country differences in the incidence of substantial overeducation – that is, having a higher qualification than one’s job requires – by explicitly taking into account the role of the unstable labor market situations in the emergence and the persistence of the phenomenon in some countries. While previous research on overeducation has mainly focused on graduates and labor market entrants, we consider both medium and highly educated individuals and focus on prime-age workers. The second step takes on a longitudinal perspective and analyzes the socio-economic conditions individuals in Germany face related to the experience of educational mismatches in their career by making use of both PIAAC 2011/12 and the PIAAC-L study.

With PIAAC 2012, a first attempt was made to measure skill mismatch in addition to educational mismatch. However, the skill mismatch measure implemented has methodological issues. It is based on a subjective measure that classifies only 3 % of the German workers as well-matched, i.e., possessing the skills required by their workplace. So far, there is no well-established measure of skill mismatch. The aim of this module is, therefore, also to develop an improved skill mismatch measure. A new version of a subjective measure is tested in the PIAAC-L study.

Participating researchers

Dr. Camilla Borgna (WZB), Dr. Anja Perry (GESIS), Dr. Paula Protsch (WZB), Prof. Dr. Heike Solga (WZB)

This module aims to improve the understanding of causal relationships between education and labour market outcomes by applying modern microeconometric methods and linking PIAAC data to external datasets. The causal effect of skills is identified using compulsory schooling reforms as instrument for the skills assessed in PIAAC. The resulting variation does not suffer from reverse causality or omitted variable bias. For this purpose, PIAAC data is linked to information on the timing of compulsory schooling reforms, especially in Germany (e.g. Piopiunik, 2013), and survey participants of PIAAC-L are asked about the actual number of years they have spent in school. The insights gained from this analysis may help to improve the international PIAAC background questionnaire. In addition, information about compulsory years of schooling will be provided in a scientific use file.

Participating Researchers

Prof. Dr. Ludger Wößmann, ifo Institute Munich

Dr. Simon Wiederhold, ifo Institute Munich

Franziska Hampf, ifo Institute Munich

To the project website at ifo

A central determinant of labour market success – especially in Germany – is the type of educational program taken (vocational or general). Using data from IALS in a differences-in-differences approach, Hanushek et al. (forthcoming) show that any initial advantage of vocational relative to general education at labour market entry decreases over time and eventually flips to a lower relative employment probability at older ages. Having information on tested skills is important for this analysis because it allows for a comparison of individuals who are similar with respect to their skill level but differ by educational program.

Hence, improved skill measures in PIAAC render possible an updated and detailed analysis of the importance of vocational vs. general education in the labor market. Moreover, linking German PIAAC-L data with respondents’ employment histories from the Federal Employment Agency addresses the problem of relying only on cross-sectional information of an individual’s employment status. The linkage is complemented by a retrospective acquisition of information on initial occupation and field of study as well as on specific aspects of the employment history.

Besides a better understanding of the causal relationship between education and labour market outcomes, this module has the potential to improve future PIAAC surveys with respect to the possibility of linking the PIAAC data to external sources and to advance the background questionnaire with respect to the acquisition of education information and work histories.

Participating Researchers

Prof. Dr. Ludger Wößmann, ifo Institute Munich

Dr. Simon Wiederhold, ifo Institute Munich

Franziska Hampf, ifo Institute Munich

To the project website at ifo

This module aims at distributing the analytic potential of PIAAC data into different knowledge  communities. It includes the following tasks:

  • Available PIAAC Data will be extended by linking them to administrative data of the German Federal Employment Agency (Module 2.1).
  • Workshops on the analysis of the international PIAAC data will be held (Module 2.2).
  • An international conference will be organized on the evidence of PIAAC and PIAAC longitudinal data (Module 2.3).

Results of the research modules will be distributed into the research community by scientific publications (Module 2.4).