Every year, thousands of research studies produce findings that cannot be replicated. A landmark 2015 study published in Science found that only 36% of psychology experiments yielded significant results when repeated — a finding that fueled what is now widely known as the replication crisis. Preregistration research practices have emerged as one of the most powerful responses to this crisis. By documenting your hypotheses, methods, and analysis plan before collecting data, you create a transparent, time-stamped record that separates genuine predictions from after-the-fact storytelling.
Whether you are a PhD candidate designing your first experiment, a principal investigator leading a multi-site trial, or a lab manager coordinating data collection across collaborators, this guide walks you through everything you need to know about preregistering a research study — from understanding what belongs in a preregistration to choosing the right platform and avoiding the most common pitfalls.
What is preregistration in research?
Preregistration is the practice of publicly documenting your research plan — including hypotheses, study design, data collection procedures, and analysis strategy — in a time-stamped, read-only repository before you begin collecting or analyzing data. It creates an unalterable record of what you intended to do, making it straightforward for reviewers, collaborators, and readers to distinguish between confirmatory and exploratory findings.
The concept builds on clinical trial registration, which has been mandatory for FDA-regulated trials since 2007. In the social and natural sciences, open science preregistration gained momentum after the replication crisis exposed how flexible analysis choices — sometimes called "researcher degrees of freedom" — can inflate false-positive rates. Today, the Open Science Framework (OSF) alone has surpassed 100,000 registrations, and more than 120 journals now accept registered reports, a publishing format where studies are peer-reviewed and provisionally accepted before data collection begins.
How preregistration differs from a study protocol
A study protocol is a detailed internal document describing every aspect of your methodology. A preregistration, by contrast, is a public-facing, time-stamped commitment focused specifically on the elements that could introduce bias if decided after seeing the data. You do not need to include your full literature review, theoretical framework, or exploratory plans. The goal is to make it clear which analyses were planned in advance and which emerged during data exploration.
Why preregistration matters for research credibility
Preregistration addresses several well-documented threats to scientific integrity. Understanding these threats makes the case for preregistering your work almost self-evident.
It prevents p-hacking and HARKing
P-hacking — the selective reporting of statistically significant results — and HARKing (Hypothesizing After Results are Known) are two of the most common questionable research practices in science. A study of published psychological articles from 2014 to 2017 found that only 3% had a preregistration (Hardwicke et al., 2022), leaving the vast majority of studies vulnerable to these biases. Preregistration locks in your hypotheses and analysis decisions before you see the data, making it nearly impossible to present exploratory findings as if they were predicted all along.
It increases trust in your findings
Research transparency through preregistration signals to the academic community that your work meets a higher standard of rigor. As the Center for Open Science summarizes it, preregistration "makes your science better by increasing the credibility of your results." Journals increasingly recognize this: studies with preregistrations are eligible for Open Science Badges, which signal research credibility and methodological transparency to readers and reviewers.
It improves study planning
One of the most underappreciated benefits of preregistration is the discipline it imposes on your research design. Many researchers report that the process of writing a preregistration revealed gaps in their methodology they would not have caught otherwise. As one Columbia University statistician noted, "Most of the time what I learned from the preregistration was that we didn't really adequately think about what we would do once we have the data." The act of preregistering forces you to think through your entire analysis pipeline before a single data point is collected.
It protects you from suspicion
When your results are surprising or controversial, a preregistration serves as your defense. Skeptical readers can compare your published analysis to your registered plan and verify that you followed through on what you promised. As the researchers behind the Data Colada blog put it, "the job of your preregistration is to set a skeptical reader's mind at ease" by showing that your results were not the product of flexible analysis choices.
When should you preregister a study?
The ideal time to preregister is after you have finalized your study design and analysis plan but before you begin data collection. However, preregistration is more flexible than many researchers assume. Here are the most common scenarios:
Before data collection — the gold standard for confirmatory hypothesis-testing studies
Before analyzing existing data — useful when working with archival datasets, public databases, or secondary data sources where you want to prevent your analysis choices from being influenced by preliminary results
During peer review — some journals request additional data collection or analyses after initial review, and you can preregister this new phase to maintain research transparency
Before each phase of a multi-stage project — for longitudinal or multi-site studies, you can preregister each wave separately as your design evolves
Preregistration is most valuable for hypothesis-driven, confirmatory research. For purely exploratory work — such as qualitative interviews, grounded theory research, or initial pilot studies — a full preregistration is less critical, though you can still register your data collection plan to demonstrate transparency and improve your workflow.
What to include in your preregistration
A strong preregistration is specific enough to constrain your analysis decisions but concise enough to be readable. The key is to include everything a skeptical reader would need to verify that your findings were not the product of flexible analysis — and nothing more.
Essential elements every preregistration needs
Research questions and hypotheses — state exactly what you predict and in which direction
Study design — describe the type of study (experimental, observational, survey-based), conditions, groups, and any randomization procedures
Sample size and rationale — specify your target sample, how you determined it (e.g., power analysis), and your stopping rule for data collection
Variables — define your primary outcome variable(s), independent variables, and any covariates or control variables with precise operational definitions
Exclusion criteria — state in advance which data points or participants you will exclude and the specific rules that determine exclusion (e.g., "participants who incorrectly answered at least 2 out of 3 comprehension checks")
Analysis plan — describe the specific statistical tests you will use, how you will handle missing data, what corrections you will apply for multiple comparisons, and what constitutes a significant result
Exploratory analyses — briefly note any additional analyses you plan to run that are not part of your confirmatory tests, clearly labeled as exploratory
What to leave out
Your preregistration does not need a full literature review, a detailed theoretical background, or procedural minutiae that do not affect analysis decisions. A preregistration "buried in theoretical background and procedural detail becomes unreadable and defeats its own purpose," as the Data Colada researchers note. Keep it focused on the decisions that could introduce bias if made after seeing the data.
Where to preregister: comparing the top study preregistration platforms
Several platforms support study preregistration, each with different strengths. The right choice depends on your discipline, study type, and how much flexibility you need.
Open Science Framework (OSF)
OSF is the most widely used preregistration platform in the social, behavioral, and life sciences. It offers multiple registration templates — including the standard OSF Preregistration template, discipline-specific options, and the AsPredicted-style template — and allows you to embargo your registration for up to four years before it becomes public. OSF also integrates with your broader research project, so you can link your preregistration to data, materials, and manuscripts in one place.
Best for: most academic research, multi-component projects, researchers who want a full open-science ecosystem.
AsPredicted
Created by researchers at the Wharton School (University of Pennsylvania), AsPredicted is designed for speed and simplicity. It asks nine standardized questions that guide you through the essential elements of a preregistration without overwhelming detail. Unlike OSF, AsPredicted registrations can remain private permanently — a feature some researchers prefer when working on competitive or sensitive topics.
Best for: researchers who want a quick, streamlined process, especially for straightforward experimental studies.
ClinicalTrials.gov
The standard registry for clinical trials involving human participants, ClinicalTrials.gov is maintained by the U.S. National Library of Medicine. Registration is legally required for many clinical trials under the FDA Amendments Act, and results reporting is mandatory for applicable trials. It is the most recognized registry in biomedical research.
Best for: clinical and biomedical research involving human participants, especially drug trials and intervention studies.
PROSPERO
PROSPERO is the international prospective register of systematic reviews. If you are conducting a systematic review or meta-analysis, PROSPERO is the standard platform, used by researchers worldwide to reduce duplication and promote research transparency in evidence synthesis.
Best for: systematic reviews, rapid reviews, and umbrella reviews across all disciplines.
How to preregister a study: a step-by-step workflow
Follow these seven steps to move from a finalized study design to a published preregistration.
Step 1: finalize your research design
Before you can preregister, you need a clear, complete research design. This includes your research questions, hypotheses, variables, sample plan, and analysis strategy. If any of these elements are still in flux, resolve them first. A vague preregistration offers little protection against bias and may actually create a false sense of rigor.
Step 2: choose your platform and template
Select the platform that best fits your discipline and study type (see the comparison above). On OSF, browse the available templates to find one that matches your research design. The standard OSF Preregistration template works for most studies, but there are also templates specifically designed for qualitative research, secondary data analysis, and replication studies.
Step 3: draft your preregistration document
Work through each section of your chosen template systematically. Be specific about your hypotheses, define your variables precisely, state your sample size and stopping rules, describe your analysis plan in enough detail that another researcher could execute it independently, and list your exclusion criteria with concrete thresholds. Write as though a skeptical reviewer will compare every line to your final manuscript — because they likely will.
Step 4: get feedback from collaborators
Share your draft preregistration with co-investigators, advisors, or methodological consultants before submitting. Fresh eyes often catch ambiguities, unstated assumptions, or gaps in the analysis plan. This review step is especially important for early-career researchers, teams working on complex multi-site studies, or anyone preregistering for the first time.
Step 5: submit and time-stamp your registration
Once your team has finalized the preregistration, submit it on your chosen platform. The platform will generate a time-stamped, read-only version of your document. On OSF, you can choose to make it public immediately or set an embargo period of up to four years. On AsPredicted, you decide when — or whether — to make it public.
Step 6: conduct your study as planned
Proceed with data collection and analysis according to your preregistered plan. If you need to deviate from the plan — which happens in real research — document the deviation clearly and explain why it was necessary. Deviations are not failures. Undisclosed deviations are.
Step 7: report transparently in your manuscript
When writing your manuscript, reference your preregistration URL and clearly distinguish between confirmatory and exploratory results. Include a link to your time-stamped registration so reviewers and readers can verify your claims. Note any deviations from the preregistered plan in your methods section or supplementary materials. Many journals now award Open Science Badges for preregistered studies, which can strengthen your paper's reception.
Common preregistration mistakes and how to avoid them
Even well-intentioned preregistrations can fall short. Here are the most frequent pitfalls researchers encounter and how to sidestep them.
Being too vague. Saying "we will measure well-being" leaves room for dozens of operationalizations. Instead, specify the exact instrument, scale, and scoring method. A skeptical reader should never be able to wonder whether you chose your measure after seeing the data.
Including too much. A preregistration overloaded with theoretical background, detailed procedural descriptions, and lengthy exploratory plans becomes unreadable. Focus only on the decisions that could introduce bias, and save the full narrative for your manuscript.
Forgetting to specify exclusion criteria. If you do not define your exclusion rules upfront, any post-hoc exclusions will look suspicious — even if they were perfectly reasonable. Define concrete thresholds (e.g., response time cutoffs, attention check failures) in advance.
Not planning for deviations. Research rarely goes exactly as planned. Build in language for how you will handle common complications: what happens if your sample is smaller than expected? What if a measurement instrument fails? What if you discover a coding error? Having a contingency plan in your preregistration shows that you thought carefully about your design.
Treating preregistration as a one-time event. For multi-phase or longitudinal projects, consider registering each phase separately. Your study evolves over months or years, and your preregistration should reflect the specific stage you are committing to — not a speculative plan for the entire project.
How ScholarDock helps you manage the preregistration process
Managing a preregistration involves coordinating multiple documents — your protocol draft, analysis plan, feedback from collaborators, platform-specific templates, and links to your registered document. For research teams working across multiple studies, keeping all of this organized is a challenge in itself.
ScholarDock, a research project and reference management platform, brings your entire preregistration workflow into one connected workspace. You can draft your study protocol and analysis plan directly within your project space, link relevant references from your source library to support your hypotheses, and track the status of each preregistration alongside your broader project timeline. When collaborators need to review a draft preregistration, they can comment, suggest edits, and track revisions — all within the same environment where your research materials, tasks, and notes already live.
For teams managing multiple studies simultaneously, ScholarDock makes it easy to see which projects have been preregistered, which are still in the drafting stage, and which need collaborator review. Instead of scattering your preregistration documents across OSF, Google Docs, email threads, and shared drives, you keep everything connected — from your initial literature search through to your final registered plan. ScholarDock's AI tools can also help you organize references, extract key findings from supporting literature, and keep your protocol drafts linked to the sources that informed them.
Preregistration is becoming a professional expectation
The momentum behind preregistration research practices is undeniable. With over 100,000 registrations on OSF, more than 120 journals accepting registered reports, and growing mandates from funding agencies like the NIH and European Research Council, preregistration is rapidly shifting from a "nice to have" to a professional standard in many disciplines.
The good news is that getting started is straightforward. Choose a platform, answer the key questions about your study design, and submit your time-stamped plan. The process takes a few hours for a simple study and pays dividends in research credibility, rigor, and peace of mind for years to come.
If your research team is ready to bring more structure and transparency to every stage of the research process — from organizing references and drafting protocols to tracking project milestones and collaborating in real time — ScholarDock gives you a single connected workspace where preregistration planning fits naturally into your broader research workflow. Start organizing your next study today.