RESEARCH DISCIPLINE
Preregistered Analysis for Psychology
Confirmatory Tests Kept Distinct
For psychologists with a preregistration or a Registered Report: the planned tests carried out as specified, each departure documented, exploratory analyses labeled as such, and effects reported with intervals that show what the data can rule out.
No Commitment Required
Strictly Confidential
In-House PhD-Level Statisticians
Fixed Project Fee
Same-Day Response
What Does Preregistered Analysis Involve?
Preregistered analysis for psychology carries out, on the collected data, the tests that were specified before those data were seen, and reports every departure from them. Psychology preregistration statistics concern what the plan commits to: the hypotheses, the exclusion rules, the model for each test and the inference criteria, such as the smallest effect that would matter. A registered report analysis goes further: the plan is peer reviewed at Stage 1, before data are collected or analyzed, and the Stage 2 article reports the approved analyses, with anything unplanned set apart as exploratory. OnlineSPSS, a specialist statistical consultancy for academic, clinical and professional research, keeps the analysis in-house: PhD-level statisticians run the registered tests, and the researcher interprets each result.
WHEN TO ENGAGE
Registration Status and What It Still Allows
What can still be changed depends on whether the plan is unregistered, accepted in principle or already carried out, and each stage calls for different work.
Plan Before Registration The study is designed and the plan is not yet registered, so each test is tied to a hypothesis, with the variables, exclusions and decision rule named in full.
Registered Report at Stage 1 A Stage 1 manuscript needs a justified sample size, an analysis plan that reviewers can judge and outcome-neutral checks that show whether the results will be interpretable.
Data Collected, Plan Registered The data are in and the registered plan must now be carried out, with each departure, such as a different model after a failed check, identified and justified.
Registration Open to Two Readings The registration leaves an exclusion rule, a covariate or a transformation open to more than one reading, and the report must show how the ambiguity was resolved.
CONFIRMATORY AND EXPLORATORY
Carrying Out a Registered Plan Faithfully
Faithful execution has two halves, doing what was registered and showing clearly where the analysis went beyond it, and the work covers both.
Inference Criteria
The smallest effect size of interest and the decision rule for each hypothesis are fixed, so that an interval can be judged against a threshold set before the data.
Sample Size and Stopping
A fixed sample or a sequential design is justified within Research Planning & Study Design, with interim looks and stopping rules registered in advance.
Exclusions and Checks
Quality exclusions and attention checks are applied as registered, and the number of participants removed by each rule is reported in a flow table.
Equivalence for Null Results
When a registered test is not significant, a registered equivalence test against the smallest effect of interest separates evidence that any effect is smaller than that threshold from data that are simply inconclusive.
Deviation Reporting
Each departure is listed with its reason and with results under both approaches, and a finished analysis can be checked through Independent Statistical Review.
Experimental Designs
Models for conditions, trials and participants follow Experimental Psychology within Psychology & Behavioral Sciences.
PREREGISTRATION PITFALLS
How a Registration Stops Constraining the Analysis
A registration constrains the analysis only when it is specific enough to follow and is then followed.
Vague Plan, Flexible Analysis A registration that says only that groups will be compared leaves the test, the covariates and the exclusions to later choices, so it constrains very little.
Registered Test Replaced Quietly A planned test is swapped for another after a check fails, and the article reports only the substitute, so readers cannot see what the registered analysis showed.
Exploratory Pattern Promoted A pattern found after the data were seen is described in the language of a hypothesis test, so its p-value reads as confirmatory evidence that it cannot provide.
Nonsignificant Read as Absent A nonsignificant planned test is described as showing no effect, though the interval may include effects large enough to matter in the theory.
Registered After the Data Were Seen The registration is time-stamped after data collection began or after a first look at the data, so it cannot show that the tests were chosen without knowledge of the results, and the report should say so.
Registered Outcomes Dropped Registered secondary outcomes vanish from the article, so a reader cannot compare the report with the registration or see which results were unfavorable.
DELIVERABLES
Records That Link a Registration to a Result
Registered, departed and exploratory results are kept in separate records, so the boundary between them survives into the published article.
Plan-Linked Script A script that implements the registered tests in order, annotated with the clause of the registration that each block follows.
Deviation Log A dated list of every departure from the registration with its reason and both results, for use in the methods section of the report.
Exclusion Flow Table Counts of participants entering the analysis and removed by each registered rule, in a form suited to a participant flow description.
Confirmatory Results Table Each registered test with its estimate, 95% confidence interval and decision against the registered criterion, in registered order.
Exploratory Results Table Unplanned analyses, labeled as exploratory, with intervals and no conclusions beyond what analyses of that kind can support.
Reporting Notes for the Article Notes on describing the registered, departed and exploratory results in a Stage 2 manuscript or article, for you to check and incorporate.
ILLUSTRATIVE EXAMPLE
A Registered Persistence Experiment With a Null Result
An illustration of how a Stage 2 analysis reports an inconclusive registered test and an ambiguous exclusion rule without changing the plan.
ILLUSTRATIVE EXAMPLE · MINDSET INDUCTION AND TASK PERSISTENCE RESEARCH · UNITED STATES
A doctoral researcher in social psychology had an accepted Stage 1 Registered Report testing whether a brief mindset induction increased persistence on an unsolvable puzzle task, compared with a neutral induction.
The registered primary test and an equivalence test against the smallest effect of interest were carried out as specified. The registered primary test was not significant, and the equivalence test, together with the interval, showed that the data were inconclusive and not evidence of no effect. A second issue was an exclusion rule for failed attention checks that did not say whether one failure or two counted. The report stated the reading used, gave results under the alternative and logged the item as a deviation. An unplanned age difference was shown in a separate exploratory table, with no conclusion drawn from it.
Outcome: The researcher received the confirmatory results with 95% confidence intervals, the equivalence result, the deviation log, the exclusion flow table, the exploratory table and notes for the Stage 2 manuscript.
This example does not describe a real client or project. We never share details of actual client work, in line with the confidentiality we promise every researcher. This illustrative example is included to show the type, scope, and standard of work involved in this kind of analysis.
SOFTWARE
Matching Each Registered Test to Its Code
A registered analysis should be rerunnable from the registration, so the program is chosen for how clearly each test can be matched to its clause.
HOW WE ASSESS YOUR PROJECT
We Review Your Actual Research
Your assessment is based on a review of your actual research materials and analytical requirements, not a generic project category.
You Submit Your Project Details
Share whatever is available at the current stage, such as the research aim, protocol or proposal, dataset, variable list, manuscript or statistical requirements.
We Review Your Research and Define the Scope
Your statistically trained Research Coordinator reviews the materials to understand the study design, research questions, data structure, analytical requirements and expected deliverables. If clarification is needed, additional documents or information can be requested before the assessment is finalized.
Your Analysis Is Completed and Delivered
Once the scope and fixed project fee are agreed, the in-house statistician completes the work and the Research Coordinator remains the communication point through delivery, with the analysis, deliverables and timeline as agreed.
RESEARCH RESPONSIBILITY
A Public Commitment Made in the Researcher's Name
A registration is the researcher's public commitment, so the hypotheses, the content of the plan and the conclusions drawn from the tests are theirs.
Agreed Statistical Support OnlineSPSS provides the agreed methodological, analytical and statistical reporting support, defined in a scope that is confirmed before any work begins.
No Intellectual Contribution OnlineSPSS does not claim any intellectual contribution to your research. Authorship of the registration, the Registered Report and the article remains yours.
Your Research Decisions The hypotheses, the registered decisions, the interpretation of the results and the conclusions remain yours, with overall responsibility.
Confidential Handling Research data and project details are handled confidentially, project information is not published, and an NDA is available where required.
FREQUENTLY ASKED QUESTIONS
Registered Reports, Bayes Factors and Unplanned Analyses
How a Registered Report differs from a registration, where Bayes factors fit and what may be reported beyond the plan.
What does preregistered analysis for psychology cover?
Preregistered analysis for psychology covers carrying out the registered tests as specified, documenting each departure, labeling exploratory analyses and reporting estimates with intervals against the registered criteria. Both a standard registration and a Registered Report at either stage are covered, with the work bounded by what the registration commits to.
How does a Registered Report differ from a preregistration?
In a Registered Report the plan is peer reviewed at Stage 1, and in-principle acceptance does not depend on the results. The analysis plan must therefore be complete and defensible, and Stage 2 review usually checks that it was followed or that departures were approved.
Can I use a Bayes factor in a registered analysis?
Yes, if the plan registers the prior, the model and the evidence threshold before the data are seen. A Bayes factor and an equivalence test answer different questions, and the sensitivity of the Bayes factor to the prior is reported so that a reader can see how far the conclusion depends on it.
Am I allowed to run analyses I did not preregister?
Yes. The registered analyses are reported first and in full, and the additional ones appear under an exploratory heading with intervals and without threshold language. They generate hypotheses for a later study, and they do not confirm anything. A finding that matters is best tested in a new, registered study.
How do you protect my research data and confidentiality?
Your data and project details are handled confidentially, and project information is not published. Researcher names and institutions are not used publicly without permission. An NDA is available where required and can be arranged before sensitive materials are shared. Read more about confidentiality and data security.
What happens when I request an assessment, and how long does it take?
You share your research question, design and relevant materials, such as the preregistration or Stage 1 manuscript and the dataset, and a statistically trained Research Coordinator reviews the request and responds the same day. The analysis, deliverables, timeline and one fixed project fee are then confirmed before any work begins, so the timeline follows from the agreed scope.
Still have a question? Find more answers in our full FAQ.
NEXT STEP
A Registered Plan
Carried Out in Full
Will a reader be able to match each reported result to its clause in the registration? That standard governs the analysis from the first script to the Stage 2 manuscript.
No Commitment Required
Strictly Confidential
Fixed Project Fee
In-House PhD-Level Statisticians
Same-Day Response
Serving Researchers Since 2014