Protocol vs. Preregistration: What’s the Difference?

If You Already Have a Research Protocol, Do You Still Need Preregistration?

Before beginning a study, you may already have a detailed protocol describing the research question, participants, procedures, measures, data collection, and analysis. Perhaps the protocol was prepared for an ethics review, research team, funder, thesis committee, or clinical trial registry.

Then someone asks whether the study was preregistered.

It is tempting to answer yes because the research plan already existed before the study began. Sometimes that may be functionally close to preregistration, particularly when the protocol itself was prospectively registered. But a protocol and preregistration describe different aspects of research planning.

The distinction becomes clearer when you separate the content of a plan from the way that plan is recorded in time.

A Protocol Is the Plan; Preregistration Establishes When the Plan Was Recorded

How Protocols and Preregistration Overlap Without Meaning the Same Thing

A Research Protocol Describes How the Study Will Be Conducted

A research protocol is the prospective plan for a study. Its exact content depends on the discipline and study design, but it may describe the rationale, objectives, research questions or hypotheses, design, participants, interventions or procedures, outcomes, data collection, analysis, ethics, and other operational details.

Protocols can serve several purposes. They may coordinate a research team, support ethics or regulatory review, establish standardized procedures across sites, document a planned methodology, or provide enough detail for others to understand how the research is intended to proceed.

Some research traditions have formal protocol requirements. Clinical trials, systematic reviews, animal studies, and other fields may have discipline-specific expectations for protocol preparation or registration. Researchers should therefore follow the requirements that apply to their particular design rather than assuming that “protocol” always refers to one standardized document.

Preregistration Is Defined by Timing and Preservation of the Plan

Preregistration addresses a related but different question: what had researchers committed to before the relevant evidence was known?

The Open Science Framework currently defines preregistration as posting a time-stamped, read-only version of a study plan before beginning the relevant data collection or analysis. The resulting registration establishes a record of research intentions that can later be compared with what researchers actually did.

This is why preregistration is more than simply writing a plan early. The plan must be preserved in a way that establishes its earlier state.

The Same Document Can Sometimes Serve Both Purposes

The distinction does not mean that researchers need two entirely separate documents.

Imagine that you prepare a detailed protocol containing your hypotheses, sampling plan, primary and secondary outcomes, exclusion criteria, procedures, and statistical analyses. Before collecting the relevant data, you submit that protocol to an appropriate registry that creates a time-stamped, preserved record.

The document remains your protocol because it describes how the study will be conducted. Its prospective registration can also provide the preregistration function because the earlier plan is now preserved before the relevant outcomes are known.

Indeed, the Open Science Framework currently provides a Registered Report Protocol Preregistration template for protocols that have received in-principle acceptance through a Registered Report journal. Other research traditions similarly speak of protocol registration.

So the concepts are not mutually exclusive. One describes the study plan itself; the other concerns the prospective registration of relevant decisions.

A Protocol Sitting on Your Computer Is Not Automatically a Preregistration

Suppose you wrote a protocol before collecting data but kept it as an editable document in your research folder. Months later, after analyzing the study, you upload that document somewhere and state that it was the original protocol.

The document may genuinely have been written beforehand. But readers have limited independent evidence that the version they are seeing is exactly the version that existed before the results were known.

A prospective registration addresses this problem by creating a dated record whose earlier state is preserved.

This distinction may sound procedural, but it is central to what preregistration contributes. The issue is not merely whether researchers say they had a plan. It is whether there is a record that can establish what that plan contained at the relevant point in the research process.

A Preregistration May Be Narrower Than a Full Protocol

A detailed research protocol may contain information that is essential for running the study but less consequential for the inferential purpose of preregistration.

For example, a protocol may contain administrative responsibilities, staff procedures, equipment instructions, data-management logistics, safety procedures, contact information, or detailed operational schedules.

A preregistration, by contrast, may concentrate on research questions, hypotheses, sampling and stopping decisions, variables, exclusions, outcome definitions, and analysis plans. These are among the decisions whose timing can be particularly important when readers evaluate the eventual claims.

Exactly what should be preregistered therefore depends on what advance specification needs to accomplish in the particular study.

A Protocol Can Also Be Less Specific Than a Useful Preregistration

The reverse is possible too.

A protocol might state that “appropriate statistical tests will be used” or that “outliers will be excluded where necessary.” Those statements may be sufficient for some administrative purpose, but they leave consequential analytical choices unresolved.

If the objective of preregistration is to clarify what analyses and decision rules were established before seeing the data, greater specificity may be needed.

The label attached to the document therefore tells you less than its content. A document called a “protocol” can be highly specific. A document called a “preregistration” can be vague. What matters is whether it records the decisions relevant to the claims researchers intend to make.

Protocol Registration Is Where the Concepts Most Obviously Meet

Many fields explicitly register protocols.

The ARRIVE guidelines for animal research, for example, ask authors to state whether a protocol containing the research question, key design features, and analysis plan was prepared before the study and whether and where it was registered. Their examples include preregistered protocols and Registered Reports.

Systematic reviews and clinical research also have established traditions of prospective protocol registration, although the registries, required information, and governing standards differ.

In these settings, asking whether something is “a protocol or a preregistration” can create a false choice. A protocol can be prospectively registered, and that registration may perform some or all of the transparency functions associated with preregistration.

Neither Term Tells You Whether the Plan Was Good

A detailed protocol can contain weak methods. A preregistration can preserve a poor analysis plan with impeccable timing.

Neither label should substitute for methodological evaluation.

When assessing a protocol, ask whether the planned design is capable of addressing the research question. When assessing a preregistration, also ask whether the relevant decisions were registered at an appropriate time and with enough specificity to make later comparisons informative.

Both Can Change After the Study Begins

Neither a protocol nor a preregistration makes the original plan infallible.

Procedures may need to change. Recruitment may fail to reach the expected target. A measure may become unavailable. An analysis may prove inappropriate. Ethical or practical considerations may require adaptation.

What preregistration adds is a visible reference point against which those changes can be understood. Researchers can document consequential deviations from the registered plan rather than rewriting the history of the study.

The protocol, meanwhile, may need amendments according to the procedures governing the particular study, institution, ethics body, registry, or regulatory environment. Those requirements are context-specific and should be verified directly.

When One Research Plan Becomes Both a Protocol and a Preregistration

Common Confusions About Protocols and Preregistration

Decide What You Need the Research Plan to Accomplish

If you already have a protocol, do not duplicate work automatically. First examine whether it contains the information needed for meaningful preregistration and whether your intended registry accepts an appropriate format or template.

Terminology varies across disciplines, so focus on function rather than labels. Ask what the document contains, when it was finalized, whether it was prospectively registered, how its earlier state is preserved, and what requirements apply to your particular research context.

That will usually tell you more than the filename.

Before Deciding Whether Your Protocol Also Functions as a Preregistration

Frequently Asked Questions About Protocols and Preregistration

A Protocol Describes the Plan; Preregistration Preserves Its Timing

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.