Propensity score weighting balanced the two non-intervention control groups on all outcomes at T2 and T3, supporting internal validity¶

Claim · All claims
Evidence · 1 study · causal r2 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? Placebo tests comparing Control Group 1 and Control Group 2 found no significant differences on any outcome after propensity weighting, indicating no residual detected confounding between controls. → Jiang Su 2026

Evidence¶

Jiang Su 2026¶

Jiang Su, Woodcock Anna, Haddad Kamel, Hernandez Paul R. (2026). Engaging Latinx or Hispanic-Identifying Families in Science Education: A Longitudinal Intervention Study. Journal of Educational Psychology. https://doi.org/10.1037/edu0001047

q2 · i? · causal · r2

Diagnostic placebo analyses compared Control Group 1 and Control Group 2 on all outcome variables at T2 and T3 using propensity score-weighted post hoc tests with a Scheffé adjustment. Only descriptive statistics suggested a group difference in family obligations at the six-month follow-up before latent modeling.

"no significant differences emerged between the control groups on any outcome after weighting, suggesting that the propensity score adjustment successfully balanced the groups on observed characteristics and that subsequent treatment effects are unlikely to reflect preexisting differences between control groups"

Discussion¶

Related Claims¶

• The treatment group reported higher state family obligations than Control Group 2 at six months, with group differences dissipating by 12 months — related
• The intervention group reported higher state parental support than the unreachable/non-interested control group at the 12-month follow-up — related
• After two years, RSF Cohort 1 students generally performed at the same levels as propensity-score-matched comparison students, with few statistically significant differences — related