PISA 2022 country notes
Nine short independent analyses of public PISA 2022 microdata. Each note leads with the question that is peculiar to its own country, and each one tests whether the comparison it makes is measurement-supported before making it.
Every note in this series carries the same three guarantees. Standard errors pooled over ten plausible values and computed from 80 replicate weights, because omitting either understates uncertainty by a factor of two or more. A reproduction of figures the OECD has already published for that country, before anything new is presented. And a measurement-comparability verdict on any questionnaire index the note compares across groups. Two of the nine report that the comparison they tested is not supported, and withhold the figures.
Montenegro
What Montenegro’s PISA 2022 data says about the north, and what it cannot say
Uzbekistan
Uzbekistan has the flattest socio-economic gradient in PISA, and a language gap larger than it
Viet Nam
The one country whose PISA comparability the OECD has publicly qualified is the one most often held up as a model
Brazil
Brazil’s regional gap is not the poverty gap in disguise, and it is not the rural gap either
Canada
Canada’s immigrant advantage is real, and it is not selection into rich provinces
United Arab Emirates
The United Arab Emirates does not have one school system, and its PISA mean is an average over seven
Why these countries
One per region, chosen so that the series covers the range of structures a national analysis has to deal with: language of instruction, tracking, regional inequality, parallel curriculum systems, and grade retention. No South Asian system participated in PISA 2022, so Uzbekistan is published in that slot and the substitution is stated on the note itself.
The method, in one paragraph
All nine are produced by the same country-parameterised analysis code from the same public files. Achievement is pooled over ten plausible values by Rubin’s rules; sampling variance comes from 80 Fay-adjusted balanced repeated replication weights; adjusted gaps come from weighted least squares run through the same machinery so their standard errors carry both error components. Questionnaire constructs are tested with a staged configural, metric, scalar and strict invariance cascade in lavaan, with full-information estimation for the rotated 2022 questionnaire design and a categorical sensitivity re-run. The code carries golden tests whose expected answers come from theory rather than from a previous run.
Commissioning the same analysis for your country
These notes are published free. The equivalent analysis for any participating country, at a fixed scope and a fixed price, is consulting work delivered by AdriaMont Consulting DOO. Write to milos@adriamont.me.
About
Dr Milos Kankaras is a psychometrician and policy analyst with more than twenty years of international large-scale assessment work, including with the OECD, UNESCO and Eurofound. His published specialism is measurement equivalence and cross-cultural comparability. He holds a PhD in social sciences from Tilburg University.
This hub indexes the nine notes in the series, all produced from the same analysis code. Published by the Center for Psychology, Podgorica. The Center for Psychology and the AdriaMont Institute are both operated by AdriaMont Consulting DOO, Cetinjski put 36, 81100 Podgorica, Montenegro. Commissioning enquiries go to milos@adriamont.me.
Dr Milos Kankaras | milos@centerforpsychology.me | miloskankaras.com | centerforpsychology.me | ORCID | LinkedIn