Back to History

Fields Changed

Registration

Field Before After
Trial End Date December 31, 2024 December 31, 2029
Last Published May 21, 2024 11:34 AM May 11, 2026 05:22 AM
Primary Outcomes (End Points) Education-related variables such as performance, attendance, and approval. Labor market variables: employment, formal employment, earnings, wages, and field of work (occupation and sector). Education-related variables such as performance, attendance, and approval, measured through bimonthly SEDUC administrative records and through scores from a standardized exam administered by the São Paulo State Secretariat of Education. The standardized assessment was originally pre-registered as SARESP (Sistema de Avaliação de Rendimento Escolar do Estado de São Paulo, Portuguese Language, Mathematics, and Natural Sciences); following the substitution of SARESP by Provão Paulista at the state level in September 2023, the standardized assessment outcome is operationalized through Provão Paulista (Portuguese Language and Mathematics only; Natural Sciences not covered by the substituted instrument). Labor market variables: employment, formal employment, earnings, wages, and field of work (occupation and sector).
Primary Outcomes (Explanation) Standardized assessment scores are standardized within cohort × edition using the control group mean and standard deviation. Available editions: Cohort 1 (2021 entry) has the 2023 Provão edition (3rd year); Cohort 2 (2023 entry) has both 2023 and 2024 editions (2nd and 3rd years respectively). The originally pre-registered SARESP 2021 wave is unavailable because SARESP was discontinued before the cohort reached the relevant grade, and Provão was not yet operational. Differential attrition in Cohort 2 Provão coverage (≈+7 percentage points, p < 0.0001) is addressed through Lee bounds (Lee 2009) as a robustness check alongside naive ITT estimates.
Secondary Outcomes (Explanation) Socio-emotional and aspirational outcomes are measured through a phone-based survey scheduled for 2026, covering both cohorts (treatment and control), with approximately 3,800 interviews (≈1,900 per cohort). Sample replenishment respects original stratum assignment; contact is considered lost after 5 unsuccessful attempts or explicit refusal.
Back to top