Should You Preregister Participant Inclusion and Exclusion Criteria?

Should Participant Eligibility Be Decided Before Recruitment Begins?

You plan to investigate the effectiveness of a digital learning intervention among university students. You know the population you want to study, but several questions remain unresolved. Should students with prior experience using the intervention be eligible? Should participants who miss several sessions be excluded? What happens if someone completes the baseline assessment but not the posttest?

These decisions can influence the composition of the sample and, potentially, the study's conclusions. When researchers establish or modify eligibility rules after observing participant characteristics or outcomes, it may become difficult to distinguish legitimate methodological decisions from selective exclusion.

Preregistration offers a way to document the intended rules before relevant results are available. The challenge is determining which criteria belong in the original plan, how precisely they should be defined, and what flexibility remains when unforeseen circumstances arise.

Yes, Prespecify Eligibility and Consequential Exclusion Rules

How to Preregister Inclusion and Exclusion Criteria Properly

What Is the Difference Between Inclusion and Exclusion Criteria?

Inclusion criteria define the characteristics required for a person or observational unit to participate in a study. Exclusion criteria identify conditions that disqualify an otherwise potentially eligible person or unit.

For example, a study of first-year university students might require current enrollment in a specified course. It might exclude students who previously completed that course if prior exposure would compromise the intended comparison.

Both types of criteria contribute to defining the target study population. Their purpose is not simply to produce a convenient sample. They should reflect the research question, design, participant protection requirements, and relevant practical constraints.

The terminology is sometimes used differently across disciplines. What matters most is that the operational rules are unambiguous and their application can be reconstructed.

Why Should Eligibility Criteria Be Prespecified?

Eligibility criteria influence who enters the study, which characteristics are represented, and the population to which findings may reasonably apply.

For example, excluding participants with limited prior technology experience may make an educational intervention easier to administer. However, it may also restrict the relevance of the findings to students who are already comfortable with digital tools.

Prespecification makes these restrictions visible before the results are known. It also helps researchers avoid changing the sample composition in response to observed findings.

There are several methodological benefits:

• Recruitment personnel can apply consistent screening procedures.
• Eligibility decisions can be checked against documented criteria.
• Researchers can explain how the final sample relates to the intended population.
• Readers can distinguish planned restrictions from later exclusions.
• Analysts can assess whether departures may have introduced selection bias.

Prespecification does not guarantee representativeness. A consistently applied eligibility rule can still produce an unnecessarily restricted or systematically unrepresentative sample.

What Information Should Each Criterion Include?

A useful criterion should identify the relevant characteristic, the operational definition, the method of assessment, and the stage at which it will be applied.

Consider the statement, "Participants must have sufficient digital literacy." It leaves unanswered what counts as sufficient, how literacy will be assessed, and who will make that judgment.

A more reproducible specification might require completion of a particular digital-skills screening assessment using a predefined threshold, provided the threshold is justified and the instrument is suitable for the population.

Not every criterion requires a complicated verification procedure. Self-report may be appropriate for some characteristics. The chosen method should be proportionate to the consequences of an incorrect eligibility decision.

Should Inclusion Criteria Be as Broad as Possible?

Not necessarily. Broad eligibility may improve the relevance of findings to a wider population, but some restrictions are essential for scientific validity, feasibility, or participant safety.

A study of a specialized instructional intervention may reasonably require participants to have completed a prerequisite course. A clinical intervention may require particular diagnostic or safety-related conditions.

The central issue is justification. Each restriction should have a defensible relationship to the research objective or study procedures.

Unnecessary restrictions can reduce external validity and limit the diversity of participants represented in the evidence. Conversely, removing scientifically necessary restrictions may introduce substantial heterogeneity or compromise the intended comparison.

Researchers should therefore evaluate the trade-off between a clearly defined study population and the broader applicability of the findings.

Are Eligibility Exclusions the Same as Data Exclusions?

No. This distinction is particularly important when preparing a preregistration.

Eligibility criteria determine whether a person or unit qualifies for participation. Data exclusion rules determine whether information from an enrolled participant or collected observation will contribute to a particular analysis.

These situations should not automatically receive the same treatment. In particular, a protocol deviation or missing outcome does not necessarily justify removing a participant from every analysis.

The separate question of preregistering data exclusion rules concerns how collected observations will be retained, corrected, or excluded during analysis.

Should Participants Be Excluded for Noncompliance or Incomplete Participation?

That depends on the study design and the estimand, meaning the precise effect or quantity the analysis is intended to estimate.

In a randomized trial, excluding participants after assignment because they did not follow the intervention can undermine the comparability created by randomization. Such exclusions may introduce bias when adherence is related to participant characteristics or outcomes.

An intention-to-treat approach generally seeks to analyze participants according to their assigned groups, although its precise implementation and the handling of missing outcomes require further specification.

Per-protocol or adherence-based analyses may answer different questions, but they require careful definition and appropriate methods. They should not automatically replace the primary randomized comparison.

In other research designs, minimum participation or exposure requirements may be appropriate. Researchers should explain why those thresholds are scientifically meaningful rather than choosing them after observing the outcomes.

How Should You Handle Participants Who Withdraw or Have Missing Data?

Withdrawal, loss to follow-up, and missing observations should not be treated as interchangeable with initial ineligibility.

A participant may have satisfied every inclusion criterion but later withdraw from the intervention. Another may remain enrolled yet fail to complete a particular assessment.

The preregistration should distinguish continued participation, permission to use previously collected information, and the statistical treatment of missing measurements. Applicable consent terms, ethics approvals, and data-protection requirements may constrain what information can be retained or analyzed.

Statistical handling should be planned according to the design and plausible missing-data mechanisms. Complete-case analysis, multiple imputation, and likelihood-based methods involve different assumptions and are not universally interchangeable.

Where possible, specify how missingness will be described and which sensitivity analyses will assess the robustness of conclusions.

Can Eligibility Criteria Be Changed After Recruitment Begins?

Yes, when justified, but changes deserve careful evaluation because they may alter the recruited population or introduce differences between earlier and later participants.

Suppose a recruitment criterion proves unnecessarily restrictive. Researchers may consider broadening it to improve recruitment or external validity. Before doing so, they should assess whether the change affects the research question, study design, or comparability of participants.

Document the original criterion, revised rule, reason for the change, date of implementation, and number or characteristics of participants potentially affected, where appropriate and permissible.

Changes involving participant protection or approved recruitment procedures may require ethics or governance approval before implementation.

Importantly, the principles for amending preregistered decisions require preservation of the original record rather than retrospective replacement of the original eligibility rules.

How Do Eligibility Criteria Affect Generalizability?

Eligibility rules determine the boundaries of the population directly represented by the study. Their implications for generalizability depend on how participants are selected within those boundaries and how the sample compares with the population of interest.

For example, findings from students who have reliable home internet access may not readily generalize to students experiencing unstable connectivity.

This limitation does not automatically invalidate the study. It does mean that conclusions should be framed consistently with the recruited population and the assumptions required for broader application.

When planning criteria, consider whether excluding particular groups is scientifically necessary and whether the resulting population remains aligned with the intended use of the findings.

What About Secondary Data and Existing Datasets?

Preregistration can also be useful when researchers analyze previously collected data. In such studies, eligibility criteria may determine which records, institutions, time periods, or observations enter the analytic sample.

For example, a researcher using institutional learning analytics data might specify that records must correspond to students enrolled during a defined academic period and contain a valid course identifier.

However, researchers should disclose what they already know about the dataset. If they have inspected outcome patterns before defining inclusion criteria, the resulting plan cannot be treated as fully prospective with respect to those outcomes.

Prespecification is still valuable for documenting subsequent decisions, but its evidential interpretation must reflect prior data access.

Defining Eligibility for a Study of Online Learning Support

The resulting protocol allows recruitment personnel and analysts to apply consistent rules. It also prevents the researchers from redefining eligibility simply because some participants show unexpectedly low achievement or limited engagement.

The example does not imply that previous course completion must always be an exclusion criterion. Its appropriateness would depend on the intervention, intended population, and scientific rationale.

Common Mistakes When Defining Participant Eligibility

How to Choose and Apply Defensible Eligibility Criteria

Begin by defining the population needed to answer the research question. Then evaluate each proposed criterion according to its scientific purpose, operational clarity, and potential effect on representativeness.

A useful criterion should not exist merely because it makes recruitment or analysis more convenient. It should help define the intended population, protect participants, or address a genuine design requirement.

Eligibility criteria should also align with the study's primary outcome and intended analysis population. For example, requiring participants to possess a particular level of baseline proficiency may substantially change the interpretation of an intervention's effect.

Finally, coordinate eligibility planning with sample-size and recruitment stopping rules. Restrictive criteria can affect recruitment rates, the number of eligible participants, and the precision of the resulting estimates.

Before Preregistering Participant Eligibility Rules

Frequently Asked Questions About Preregistering Eligibility Criteria

Define Eligibility Before Participant Selection Can Influence the Results

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.