This chapter outlines the methodological elements of the PIAAC survey that are focused on how skills are measured and reported in this release.
In addition to assessing proficiency, PIAAC collected detailed information on how frequently adults used various skills in their work and daily lives. To support meaningful analysis, responses to these behavioural questions were combined into standardised indices using Item Response Theory (IRT). Each index aggregated multiple items that reflected the use of a particular skill (e.g. reading at work), resulting in standardised indices developed through IRT. Each index combined responses to multiple survey items into standardised scores, the resulting scales were continuous variables, standardised to have a mean of 2 and a standard deviation of 1 across the pooled sample of respondents in all participating countries and economies. Higher scores indicated more frequent use of the skill.
Indices were available for a range of skill areas, including:
Each index was based on responses to a specific set of questions (listed in the Background Notes) and allowed for comparative analysis by age, education, gender, and economic sector.
Proficiency in literacy, numeracy, and adaptive problem solving (APS) were assessed directly through performance tasks completed by survey respondents. These responses were analysed using IRT methodology, which adjusts for differences in test item difficulty and provides each respondent with a scale score between 0 and 500. These scores represented the individual's estimated proficiency, even if they did not complete all test items.
The results are presented in two main forms:
The literacy and numeracy domains are divided into five levels, while the APS domain is divided into four. These levels are used to describe both the overall distribution of skills and differences across demographic groups.
While the core assessment frameworks and many survey questions remained consistent between PIAAC Round 1 (2012) and Round 2 (2022/2023), some updates were introduced to reflect changes in the labour market and digital technology. As such, direct comparisons across cycles should be made with caution.
Learn about our data and confidentiality safeguards, and the steps we take to produce statistics that can be trusted by all.