PIAAC-Longitudinal (PIAAC-L), Germany

In 2011/2012, key adult competencies were assessed in 24 countries (including Germany) as a part of the OECD Programme for the International Assessment of Adult Competencies (PIAAC). The German PIAAC Longitudinal Project (PIAAC-L) follows up the German PIAAC sample with three additional waves of data collection (in 2014, 2015, and 2016), each with a somewhat different focus. The PIAAC-L questionnaires are largely based on core instruments from the German Socio-Economic Panel (SOEP), but sometimes include parts of the PIAAC background questionnaire and various additional questions and modules on the respondent’s background adopted from other surveys as well as a number of new questions. In addition, assessment instruments from PIAAC and NEPS (National Educational Panel Study) measuring key competencies are implemented. The objective of the PIAAC-L project is to significantly expand the German PIAAC database by adding a longitudinal dimension and by enriching the depth and breadth of information available on the German PIAAC respondents. This approach extends the analytical potential of the German PIAAC data and allows a myriad of additional research questions to be addressed. PIAAC-L is a collaborative effort undertaken by GESIS – Leibniz Institute for the Social Sciences (lead) together with the German Institute for Economic Research (DIW Berlin) and the Leibniz Institute for Educational Trajectories (LIfBi).

The German PIAAC 2012 respondents that had given their consent to being re-contacted for a potential follow-up survey (anchor persons) form the starting point for PIAAC-L. In order to obtain further information on the context of these anchor persons, the PIAAC-L approach additionally includes household members ages 18+ as well as the administration of a comprehensive household questionnaire (in waves 1 and 3).

The 2014 data collection (wave 1) implemented core SOEP questionnaires (for persons and households). For the 2015 data collection (wave 2), we developed an extensive background questionnaire (including items from PIAAC, NEPS, SOEP, as well as a number of other surveys). This questionnaire, followed by an assessment of literacy and numeracy using PIAAC and NEPS measurement instruments, was administered to anchor persons and their partners – if the latter lived in the same household as the anchor. As in wave 1, the 2016 data collection once again collected interviews from all adults living in the anchor person’s household using core SOEP questionnaires; the SOEP person questionnaire was extended to include new questions and modules (for example on adult education). In wave 3, respondents were also administered the SOEP short cognitive performance tasks. In addition, cognitive items from a Number Series Study carried out by the German Institute for International Educational Research (DIPF) were included as an add-on module.

Data Collection 2014
Core SOEP questionnaires were implemented in the 2014 data collection. The household questionnaire includes the following topics:
- Living situation, conditions and costs
- Household income and benefits, wealth
- Children and other household members

The person questionnaire includes questions on:
- Background information, family, and childhood
- Biographical calendar
- Formal education (general and vocational education), continuing professional education
- Work status, situation and history
- Income and benefits
- Health, attitudes, personality, opinions, satisfaction
- Time use and leisure activities

Data Collection 2015
In 2015 respondents were administered first a CAPI questionnaire and then a cognitive assessment using PIAAC instruments and NEPS instruments. The person questionnaire encompasses questions from several surveys (e.g., PIAAC 2012, SOEP, NEPS, AES, or PISA) and includes the following topics:
- General, vocational and professional education
- Current status and employment, income
- Skills used at work
- Computer skill use
- Mother tongue(s) and knowledge of foreign languages
- Self-assessment of numeracy and literacy
- Health, leisure, friends
- Family and relationships
- Background information (e.g., parents, citizenship)
- Satisfaction

The cognitive assessment implemented the following instruments:
- PIAAC Literacy
- PIAAC Numeracy
- NEPS Reading Speed
- NEPS Reading
- NEPS Mathematics
As in PIAAC 2012, the PIAAC assessment items were administered per default computer-based, but a paper-based version was available for respondents who could or would not work on the items on the computer. The NEPS items were paper-based only, as is usually the case in the NEPS.

The design of the cognitive assessment is different for anchor persons and their partners. For the anchor persons, there were eight different assessment conditions: Two with only NEPS items, two with only PIAAC items, and four with some items from PIAAC and some from the NEPS. The partners of the anchor persons worked exclusively on NEPS items (two assessment conditions).

Data Collection 2016
Core SOEP questionnaires were implemented in the 2016 data collection, with various extensions. In addition, a number of cognitive tasks were administered.

The household questionnaire includes the following topics:
- Living situation, conditions, and costs
- Household income and benefits, wealth
- Household possessions and customs
- Children

The person questionnaire includes questions on:
- Background information, family, childhood, and cultural identity
- Biographical calendar
- Formal education (general and vocational education)
- Adult education and training (vocational and general further training) [new module]
- Work status, situation and history, skill mismatch
- Income and benefits
- Health, satisfaction, attitudes
- Time use and leisure activities
- Life events

The following cognitive competencies short scales (from SOEP) were administered:
- Animal naming task
- Multiple-choice vocabulary intelligence test
- Symbol-digit test

Add-on Module:
A subset of anchor persons participated in the Number Series Study (German Institute for International Educational Research, DIPF) which was carried out as a cooperative undertaking of the PIAAC-L project with the PIAAC Leibniz Network. The Number Series Study conducted by the DIPF was a part of research project within the PIAAC Leibniz Network (PIAAC-LN) and was funded by the Leibniz Community (Grant number SAW-2015-GESIS-2) and the Centre for International Student Assessment (ZIB; Grant Number ZIB2016). This study adds an important measure to the PIAAC-L data, namely a measure of reasoning.

Various data sets are available under this study number:

• Persons (2014)
Units: All PIAAC-L 2014 respondents (anchor persons and household members 18+, with participation in data collection 2014)
Content: Data from person questionnaire, including derived variables

• Household (2014)
Units: All PIAAC-L 2014 households
Content: Data from household questionnaire, including derived variables

• Weights (2014)
Units: Anchor persons 2014
Content: Weighting factors

• Persons (2015)
Units: All PIAAC-L 2015 respondents (anchor persons and partners living in the same household, with participation in data collection 2015)
Content: Questionnaire data, derived variables, cognitive assessment data, proficiency measures (plausible values for PIAAC literacy and PIAAC numeracy; weighted likelihood estimates (WLE) for NEPS reading and NEPS mathematics; WLEs for PIAAC literacy and PIAAC numeracy)

• Weights (2015)
Units: Anchor persons 2015
Contents: Weighting factors

• Persons (2016)
Units: All PIAAC-L 2016 respondents (anchor persons and household members 18+ with participation in data collection 2016)
Content: Data from person questionnaire, including derived variables

• Cognit (2015)
Units: All PIAAC-L 2016 respondents (anchor persons and household members 18+ with participation in data collection 2016)
Content: Data from three short tests of cognitive ability

• NumberSeries (2015)
Units: Pre-selected anchor persons in 2016
Content: Data from Number Series Study

• Household (2016)
Units: All PIAAC-L 2016 households
Content: Data from household questionnaire, including derived variables

• Weights (2016)
Units: Anchor persons 2016
Content: Weighting factors

• Calendar
Units: All PIAAC-L 2014 and 2016 respondents (anchor persons and household members 18+ with participation in data collection 2014 and/or 2016)
Content: Data from biographical calendar, spell data; this data set is incremental

• Registry
Units: All persons ever registered in PIAAC-L
Content: Basic information on participation in the different waves of data collection; this data set is incremental and was updated each wave

In order to comply with data protection legislation and to protect the identity of our respondents, a number of data confidentiality edits were implemented (suppression of variables or coarsening of information).