RCT design checklist

Use for design and preparation. Check applicable ethics/registration and distinguish completed, planned and non-applicable matters.

When to use this worksheet and what completion involves Fictional completed example

EvalCompass Worksheet

RCT design checklist

Method: Randomized Controlled Trial

Use for design and preparation. Check applicable ethics/registration and distinguish completed, planned and non-applicable matters.

Basic information

Enter evaluation basics

Completed example for this section (fictional)

This fictional example is for explanation. Adapt it to your own program. Your entries will not change.

Program name: Consultation-information trial
Evaluation organization: Evaluation team (fictional)
Lead evaluator: Lead evaluator A (fictional)
Evaluation period: start to planned end: April–July 2026 (fictional)
Creation date: April 1, 2026 (fictional)

Reason for this entry: Specify period and responsibility scope.

Program name

Evaluation organization

Lead evaluator

Evaluation period: start to planned end

Creation date

Writing tips and explanation for this section

About this section

Enter evaluation basics

Specify period and responsibility scope.

View the method explanation

Writing tips

Follow each field’s heading to record the program name, focus, period, and other details. Leave undecided matters clearly marked as unresolved.

Read the full fictional example

Questions and primary outcome

Specify for whom, which intervention, compared with what and what is measured

Completed example for this section (fictional)

This fictional example is for explanation. Adapt it to your own program. Your entries will not change.

Scroll horizontally to view the table

Item	Content
Targets	People receiving ordinary information
Intervention and comparison	Additional individual / ordinary information
Primary outcome	Applied by deadline, with all assigned targets as denominator

Reason for this entry: Do not exclude recipients after assignment.

Scroll horizontally to view the table

Item	Content

Writing tips and explanation for this section

About this section

Specify for whom, which intervention, compared with what and what is measured

Do not exclude recipients after assignment.

View the method explanation

Writing tips

Keep the information in each row aligned. Scroll wide tables horizontally to see every column.

Read the full fictional example

Assignment design

Enter assignment methods and units

Completed example for this section (fictional)

This fictional example is for explanation. Adapt it to your own program. Your entries will not change.

Unit: individual / household / school / area, etc.: Individuals
Simple / stratified / cluster randomization: 1:1 assignment using pre-generated random numbers
Intervention group definition: Ordinary plus individual information
Control definition: pure control / waiting list, etc.: Ordinary information

Reason for this entry: Both groups receive standard services; ethical suitability requires separate review.

Unit: individual / household / school / area, etc.

Simple / stratified / cluster randomization

Intervention group definition

Control definition: pure control / waiting list, etc.

Writing tips and explanation for this section

About this section

Enter assignment methods and units

Both groups receive standard services; ethical suitability requires separate review.

View the method explanation

Writing tips

Follow each field’s heading to record the program name, focus, period, and other details. Leave undecided matters clearly marked as unresolved.

Read the full fictional example

Sample size calculation

Enter power results and rationale

Completed example for this section (fictional)

This fictional example is for explanation. Adapt it to your own program. Your entries will not change.

Expected effect size and evidence: Not calculated; 15 points is illustrative, not a minimum detectable effect
Significance level (α): Choose at design and justify
Power (1-β): Choose at design and justify
Required intervention-group sample: Not calculated; example uses 100
Required control-group sample: Not calculated; example uses 100
ICC for cluster RCTs: Not applicable here: individual assignment

Reason for this entry: Do not treat fictional 100 as a power-calculation result.

Expected effect size and evidence

Significance level (α)

Power (1-β)

Required intervention-group sample

Required control-group sample

ICC for cluster RCTs

Writing tips and explanation for this section

About this section

Enter power results and rationale

Do not treat fictional 100 as a power-calculation result.

View the method explanation

Writing tips

Follow each field’s heading to record the program name, focus, period, and other details. Leave undecided matters clearly marked as unresolved.

Read the full fictional example

Ethics and feasibility

Record completed / in progress / not addressed and plans for unaddressed items

Completed example for this section (fictional)

This fictional example is for explanation. Adapt it to your own program. Your entries will not change.

Scroll horizontally to view the table

Check item	Status	Response plan
Consent for participation, assignment and data use	Not yet checked	Check organizational review and participant explanations
Preregistration	Not yet planned	Choose a registry suited to the field; UMIN is not universally mandatory

Reason for this entry: Keep outstanding pre-trial matters explicit.

Scroll horizontally to view the table

Check item	Status	Response plan

Writing tips and explanation for this section

About this section

Record completed / in progress / not addressed and plans for unaddressed items

Keep outstanding pre-trial matters explicit.

View the method explanation

Writing tips

Keep the information in each row aligned. Scroll wide tables horizontally to see every column.

Read the full fictional example

Design checklist

Completed example for this section (fictional)

This fictional example is for explanation. Adapt it to your own program. Your entries will not change.

☑ Checked: Predefined primary outcome and multiple-indicator analysis if relevant
☐ Not checked: Completed sample-size calculation beforehand
☐ Not checked: Checked applicable ethics and recorded application/approval status or plans
☐ Not checked: Checked applicable trial-registration/analysis-publication requirements and recorded status/plans
☐ Not checked: Can ethically explain control services and intervention differences
☐ Not checked: Predefined expected attrition, follow-up, missingness responses and analytical limits
☑ Checked: Confirmed ITT as the primary approach

Reason for this entry: Primary indicator and ITT defined in the example; actual power, review, registration and attrition responses incomplete.

Predefined primary outcome and multiple-indicator analysis if relevant
Completed sample-size calculation beforehand
Checked applicable ethics and recorded application/approval status or plans
Checked applicable trial-registration/analysis-publication requirements and recorded status/plans
Can ethically explain control services and intervention differences
Predefined expected attrition, follow-up, missingness responses and analytical limits
Confirmed ITT as the primary approach

Writing tips and explanation for this section

About this section

Primary indicator and ITT defined in the example; actual power, review, registration and attrition responses incomplete.

View the method explanation

Writing tips

Read each item and check those you have confirmed. Do not use the fictional example’s checkbox states as findings about your own program.

Read the full fictional example

Before writing, check when to use the worksheet and what completion involves

RCT design checklist is a worksheet for applying Randomized Controlled Trial in meetings or individual work. For the overall duration of the method, use “Illustrative planning durations, not common standards: design 2–6 months; implementation depends on intervention, from several months to multiple years” as a reference, adjusting for scale and resources. Fill in what you know provisionally and leave unresolved matters as questions to check next.

When this worksheet is useful

Plan investigation of new social interventions or policies before implementation
Design assignment and comparisons beforehand to examine causal effects
Sufficient participants and technically/ethically feasible randomization

Participants and preparation

Lead evaluator, statistician, field coordinator and ethics committee. Specialist organizations such as J-PAL may be practical at scale.

Bring the materials used for decisions, existing data, and differing stakeholder views to help complete the worksheet.

Suggested order

1. Basic informationEnter evaluation basics
2. Questions and primary outcomeSpecify for whom, which intervention, compared with what and what is measured
3. Assignment designEnter assignment methods and units
4. Sample size calculationEnter power results and rationale
5. Ethics and feasibilityRecord completed / in progress / not addressed and plans for unaddressed items
6. Design checklist

Signs the worksheet is ready for review

Report with ITT, effect sizes and confidence intervals
Baseline and endline datasets
Preregistered trial plan

Cautions for review

Differential attrition distorts comparisons if lost participants differ. Always compare rates and characteristics in both groups.
Trial awareness may change behavior: Hawthorne effects. Take care in social experiments where blinding is difficult.
Do not equate significance with meaningful effect sizes. Large samples detect small effects; always report effects and confidence intervals.

Guides to read before starting

Read a completed fictional example (your entries will not change)

Fictional example; not actual results

A fictional calculation: randomly assign 100 each to ordinary or additional individual information and establish all application status. 100 is not a recommended sample size.

Basic information

Program name: Consultation-information trial
Evaluation organization: Evaluation team (fictional)
Lead evaluator: Lead evaluator A (fictional)
Evaluation period: start to planned end: April–July 2026 (fictional)
Creation date: April 1, 2026 (fictional)

Reason for this entry: Specify period and responsibility scope.

Questions and primary outcome

Scroll horizontally to view the table

Item	Content
Targets	People receiving ordinary information
Intervention and comparison	Additional individual / ordinary information
Primary outcome	Applied by deadline, with all assigned targets as denominator

Reason for this entry: Do not exclude recipients after assignment.

Assignment design

Unit: individual / household / school / area, etc.: Individuals
Simple / stratified / cluster randomization: 1:1 assignment using pre-generated random numbers
Intervention group definition: Ordinary plus individual information
Control definition: pure control / waiting list, etc.: Ordinary information

Reason for this entry: Both groups receive standard services; ethical suitability requires separate review.

Sample size calculation

Expected effect size and evidence: Not calculated; 15 points is illustrative, not a minimum detectable effect
Significance level (α): Choose at design and justify
Power (1-β): Choose at design and justify
Required intervention-group sample: Not calculated; example uses 100
Required control-group sample: Not calculated; example uses 100
ICC for cluster RCTs: Not applicable here: individual assignment

Reason for this entry: Do not treat fictional 100 as a power-calculation result.

Ethics and feasibility

Scroll horizontally to view the table

Check item	Status	Response plan
Consent for participation, assignment and data use	Not yet checked	Check organizational review and participant explanations
Preregistration	Not yet planned	Choose a registry suited to the field; UMIN is not universally mandatory

Reason for this entry: Keep outstanding pre-trial matters explicit.

Design checklist

☑ Checked: Predefined primary outcome and multiple-indicator analysis if relevant
☐ Not checked: Completed sample-size calculation beforehand
☐ Not checked: Checked applicable ethics and recorded application/approval status or plans
☐ Not checked: Checked applicable trial-registration/analysis-publication requirements and recorded status/plans
☐ Not checked: Can ethically explain control services and intervention differences
☐ Not checked: Predefined expected attrition, follow-up, missingness responses and analytical limits
☑ Checked: Confirmed ITT as the primary approach

Reason for this entry: Primary indicator and ITT defined in the example; actual power, review, registration and attrition responses incomplete.

Common errors in judgment

Compare only readers of the information

How to revise: Predefine ITT by assigned groups and separately show actual use

Reason: Selecting users loses randomized comparability.

Assume RCTs have no bias after dropping missing outcomes

How to revise: Examine group-specific missingness/reasons and conduct sensitivity analysis

Reason: Randomization does not guarantee equal follow-up probability.

Editorial and checking information

Content updated: 2026-10-07Source checked: 2026-10-07 (includes AI cross-checking)No documented human reviewIncludes fictional examplesEditorial policy and scope of checking Cite this page Report a correction

References and checked sections

The elements of a randomized evaluation
Michael Gibson・Anja Sautmann／J-PAL
Section consulted
3. Designing the evaluation / 4. Randomization / 5. Data analysis
How it relates to the text
After randomization, attrition, non-compliance and spillovers threaten internal validity; generalization requires further examination.
Source checked: 2026-09-13 (includes AI cross-checking)
Magenta Book Annex A: Analytical methods for use within an evaluation
HM Treasury / Evaluation Task Force · 2026
Section consulted
A2.1 pp.20–21, How is it used? / Pros and cons: design before implementation, statistical power, control group exposure, ethics
How it relates to the text
Check pre-implementation randomization feasibility, sufficient power, no influence on comparison groups and ethical risks.
Source checked: 2026-10-07 (includes AI cross-checking)

How to cite this page

EvalCompass Editorial Team. “RCT design checklist”. EvalCompass, updated 2026-10-07. https://evalcompass.jun-nakatani.com/en/worksheets/rct/

Add the date you accessed the page. Text created by this site is available under CC BY 4.0. Quotations, external materials, and bundled fonts have their own terms of use. When reusing content, credit the source, link to the license, and indicate any changes.

Suggest a correction or improvement to this page

Change history

2026-09-13 — Checked sources for definitions and conditions and added fictional examples, judgment rationales and completed entries for every field. Revised unsupported historical or uptake claims. Reviewed procedures and cautions, adjusting fixed counts, thresholds and institutional generalizations. Preserved field positions, types and counts while revising fixed conditions and thresholds.
2026-10-07 — Added Magenta Book Annex A (May 2026) A2.1 solely for checks of pre-intervention randomization, power, comparison-group influences and ethics. Retained definitions, finite-sample/missingness cautions, J-PAL sources, worksheet and examples.