---
title: "PISA 2022: Synthetic Survey Fit Across Five… | Minds"
canonical_url: "https://getminds.ai/research/pisa-2022-synthetic-survey-validation"
last_updated: "2026-09-08T16:20:31.840Z"
meta:
  description: "A five-country PISA comparison evaluates 600 students and ten held-out questions, with weighted distribution errors and explicit country and language limits."
  "og:description": "A five-country PISA comparison evaluates 600 students and ten held-out questions, with weighted distribution errors and explicit country and language limits."
  "og:title": "PISA 2022: Synthetic Survey Fit Across Five… | Minds"
  "twitter:description": "A five-country PISA comparison evaluates 600 students and ten held-out questions, with weighted distribution errors and explicit country and language limits."
  "twitter:title": "PISA 2022: Synthetic Survey Fit Across Five… | Minds"
---

Minds

September 5, 2026·Validation·Minds Team # **PISA 2022: Synthetic Survey Fit Across Five Countries** A five-country PISA comparison evaluates 600 students and ten held-out questions, with weighted distribution errors and explicit country and language limits. The PISA 2022 comparison reached 93.80% aggregate approximation using compact-profile probability answers. Weighted mean distribution error was 6.20 percentage points across ten held-out items from 600 students in five countries. The repeatable positive finding concerned response format. Generic probabilities reduced weighted error by 14.86 points relative to generic forced choices, with positive effects across every evaluated country and question family. ## The international comparison The sample included 120 students from each of the United Kingdom, Germany, Japan, Mexico and the United States. Ten questions covered five question families. Conditions compared generic forced choice, generic probability, country-plus-demographic probability, and compact-profile probability. The test used English master wording across all countries. It evaluated country context, not the performance of five localized-language questionnaires. Human evaluation outcomes were held out from runtime inputs. ## Weighted distribution results_### **Weighted mean absolute error** Reported results for this study. The table and discussion retain the study scope, baselines and uncertainty._- Generic forced choice**22.29**