PIAAC Seminar: Enriching PIAAC: Longitudinal follow-ups and linkage with other data, OECD Headquarters, Paris, France, 21 October 2022

Large-scale assessments like PIAAC are normally conducted as cross-sectional surveys: they offer a snapshot of the distribution of proficiency of the population at one point in time. They are therefore precious tools to understand how different countries compare to each other, but they are less well suited for investigating the dynamics of skills formation and how individual trajectories are shaped by skills. To answer these questions, a longitudinal dimension is often necessary. This longitudinal dimension can be obtained in two ways: (1) by conducting follow-up studies, where participating adults are re-interviewed and, possibly, their skills are re-assessed, or (2) by linking cross-sectional survey data to administrative records (for example, social security or school archives), or to other surveys containing a more complete history of education or labour market outcomes. These two strategies are not mutually exclusive. Some countries have enriched their data from the 1st cycle of PIAAC in either or both ways. Others have linked other datasets to PIAAC using Education and Skills Online. The purpose of this workshop is to review these different experiences, illustrate the potential they offer for research and policy evaluation, and reflect on how these initiatives can involve a wider number of countries. (Agenda; Presentations)

PIAAC Methodological Seminar: "The use of test scores in secondary analysis", OECD, Boulogne-Billancourt, France, 14 June 2019

The increasing availability of microdata from large-scale assessments of students and adults have spurred a great deal of research in labour economics and economics of education. Test scores from these assessments are often seen as better and more precise proxies of cognitive skills or human capital than years of schooling. Researchers using these data are however often unaware of how assessments are designed, and of how test scores are estimated on the basis of the answers provided by respondents sitting an assessment. Large-scale assessments are primarily designed to describe the distribution of proficiency in the population of interest, and are not intended to provide unbiased measures of the individual proficiency of any single respondent. This has important implications for their use in secondary, individual-level analysis. The purpose of this Seminar is to raise awareness about the methodology used to estimate test scores in large-scale assessments and to stimulate a reflection on how methodological choices made at the production stage affect the use of data in secondary analysis. The Seminar intends to bring together users and producers of this type of data, in order to stimulate an inter-disciplinary dialogue and reflect on how the production and the dissemination of the data could be improved in order to better serve the multiplicity of demands that these data are called to satisfy. (Agenda; Concept Note; List of speakers; Presentations; videos)

OECD Methodological conference on cross-country comparability of questionnaire scales, OECD Headquarters, Paris, France, 8-9 November 2018

This conference aimed at exploring novel approaches in measurement equivalence research and at highlighting promising areas to advance the field of invariance testing. It also aimed at delineating recommendations and guidelines for dealing with the issue of measurement equivalence, both in the preparation of public-use datasets and in the analytical work based on the data collected in these surveys. The target audience for this conference included researchers and analysts working with international datasets as well as the key suppliers in the industry of international large-scale surveys (Conference webpage; Agenda). 

OECD-GESIS Seminar on: 'Translating and Adapting Instruments in Large-Scale Assessments', OECD Headquarters, Paris, France, 7-8 June 2018  

The aim of the methodological seminars is to encourage discussion and cross-fertilisation of ideas between the range of actors and communities involved in the design, management and analysis of large-scale assessments. The topic of translation and adaptation of instruments has been chosen for 2018 as it is a key to achieving the goals of cross-linguistic and cross cultural comparability in international studies. (Agenda; List of speakers; Presentations; meeting report).

OECD Seminar on: Managing the Quality of Data Collection in Large Scale Assessments, OECD Headquarters, Paris, France, 11-12 May 2017 

The purpose of this seminar was to convene a group of experts in the area of large scale surveys and assessments to discuss the management of the quality of data collection and the detection, mitigation and treatment of errors arising in the course of data collection (Agenda; Speakers; Presentations)

For information on forthcoming PIAAC events, please contact: edu.piaac@oecd.org