Item		Point in time (where applicable)	Rating
1a	Study design (Potential confounders taken into account)	End of intervention	High confidence: RCT, RDD, ITT, IV
			Medium confidence: DiD with matching, PSM
			Low confidence: other matching
1b	Study design (Potential confounders taken into account)	Longest follow up (if applicable)	Study design may change at post endline follow up, usually loss of RCT as control becomes treated. Same codes as 1a
2	Masking or blinding (RCTs only)		High confidence: any blinding or any mention of blinding
			Medium confidence: no blinding
			Low confidence is not used for this item
3	Power calculations are reported		High confidence: any mention of power calculations as basis for sample size
			Medium confidence: no mention of power calculations
			Low confidence is not used for this item
4a	Losses to follow up are presented and acceptable*	End of intervention	High: attrition within IES bounds
			Medium: attrition close to IES bounds
			Low: attrition not reported or attrition outside IES bounds
			N/A for ex post studies
4b	Losses to follow up are presented and acceptable*	Longest follow up (if applicable)	High: attrition within IES bounds
			Medium: attrition close to IES bounds
			Low: attrition not reported or attrition outside IES bounds
			N/A for ex post studies
5	Intervention if clearly defined		High confidence: intervention clearly and fully described
			Medium confidence: brief description of intervention
			Low confidence: intervention named but not described, or not named
6	Outcome measures are clearly defined and reliable		High confidence: outcome measure clearly and fully described, preferably with reference to validation
			Medium confidence: brief description of outcome
			Low confidence: outcome named but not described
7	Baseline balance (N.A. for before vs. after)		High confidence: RCT or baseline balance report and satisfactory (imbalance on 5% or <5%)
			Medium confidence: Imbalance between 5% and 10%
			Low confidence: Baseline balance not reported, or reported and lack of balance on 10% or >10%
	Overall confidence in study findings	End of intervention	Lowest rating across items 1a, 4a, 6 and 7
	Overall confidence in study findings	Longest follow up (if applicable)	Lowest rating across items 1b, 4b, 6 and 7 (N/A if 1b and 4b N/A)

*Maximum acceptable rate of differential attrition for each overall attrition rate (Source: Deke, Sama‐Miller & Hershey, 2015).

DiD, difference in difference; IES, Institute of Education Studies; ITT, intention to treat; IV, instrumental variable; PSM, Propensity Score Matching; RCT, randomised controlled trial; RDD, regression disconyinuity design.