RESEARCH SUPPORT
Commissioning Statistical Analysis
Define the Terms Before the Work
Knowing when to consult a statistician comes down to timing: ideally at the design stage, and at the latest before any model is run. What a research team should settle first is the question, the data and their version, the deliverables, the review points and the roles, so that the analysis returned is the one the project needs.
No Commitment Required
Strictly Confidential
In-House PhD-Level Statisticians
Fixed Project Fee
Same-Day Response
What Should a Team Define Before Commissioning an Analysis?
Commissioning statistical analysis from outside the team means that an analyst who did not collect the data will analyze them, so everything an analyst would otherwise absorb from the project has to be written down: the research question, how the data were generated, the version of the dataset, the analyses wanted, the outputs and the role of each person. An external statistician for research data works best when the scope is explicit, and involving a statistician early lets ambiguity be resolved before any model is run and not discovered later in the results. OnlineSPSS, a specialist statistical consultancy for academic, clinical and professional research, works on this basis: a statistically trained Research Coordinator reviews the materials, the scope and one fixed project fee are confirmed, and an in-house PhD-level statistician carries out the analysis. The team still owns the question, the interpretation and every conclusion drawn from the analysis.
WHEN TO ENGAGE
Reasons a Team Looks Beyond Its Own Members
Teams bring in an outside analyst for different reasons, and the reason shapes how much of the scope concerns method and how much concerns handover.
No Statistician on the Team The group holds the data and the question but has nobody who can choose, run and check the models, so the whole analysis is passed on.
A Method Beyond the Team's Skills The team handles most analyses itself, and one component, such as a longitudinal or latent variable model, needs a specialist, one of the areas covered under Services.
Capacity at a Fixed Point A grant report, a submission or a data-lock date leaves too little time to run the analysis while the other work of the project continues.
An Independent Analyst Wanted The team prefers the analysis to be run by an analyst independent of data collection, as a separate check on what the findings show.
WHAT A TEAM SHOULD DEFINE
Terms of a Commissioned Analysis
A written scope records each of these terms, and every term settled early removes a likely surprise later in the analysis.
The Question
The primary question is stated as something to estimate, with the population, the outcome and the comparison, and secondary questions are listed separately.
The Data
Who holds the data, which version is analyzed, how it was collected and what is known to be wrong with it are written down and frozen at a stated date.
The Analyses
Each model and check is listed with the question it answers, so the team knows before work starts which results to expect, as part of Quantitative Data Analysis.
The Outputs
The tables, figures, files and written notes to be returned are named, in the formats the journal, committee or funder expects.
Roles and Contact
Who approves decisions, who answers questions about the data and how the Research Coordinator carries communication through delivery are agreed at the outset.
Review Points
The points at which the team looks at interim results are set, together with how comments from the team and later from reviewers are to be handled.
COMMON HANDOVER ISSUES
Team Oversights Found Only at Handover
Handover problems usually trace back to a point nobody wrote down at the start, and they surface when the team first tries to use the results.
Data Version Unclear The analyst works from one file while the team has updated another, and the final numbers match neither of them.
Question Drifts Midway The research question changes after the analysis begins, and models run for the first question are reported as if they answered the second.
Known Data Problems Unmentioned A change in how a variable was measured, or a site that recruited differently, is known to the team but never reaches the analyst.
Interim Results Read as Final Estimates shown at a review point are quoted in a grant report before the checks are finished, so later corrections look like changed findings.
Acknowledgment Left Late How the statistical support is to be acknowledged is decided at submission, when a journal's disclosure rules ask for it.
Two Voices, Two Instructions Two members send different requests about the same model, and with nobody named to approve decisions, the analyst cannot tell which request stands.
DELIVERABLES
Records Delivered to the Team
Files and notes are organized for the members who will use them, including anyone who joins the project after the scope was agreed.
Written Scope Record The agreed question, data version, analyses, outputs and timeline in one document, referred to throughout the project.
Data Handling Log What was received, which version was analyzed, the exclusions applied and the reason for each, so the final sample can be reconciled.
Scripts, Syntax and Logs The scripts or syntax and logs for each agreed analysis, in the software the team uses, rerunnable on the frozen dataset.
Results in the Agreed Format Tables and figures laid out as the scope specified, with estimates, 95% confidence intervals and sample sizes.
Findings Note A note reading the results against the question, stating what the design supports and what it does not.
Team Briefing Note A short note for members who were not in the discussions, covering what was done, where the files are and how to rerun them.
ILLUSTRATIVE EXAMPLE
A Dataset That Kept Changing Under the Analysis
An illustration of how an unstable data file and two readings of the question were settled in a written scope.
ILLUSTRATIVE EXAMPLE · COASTAL FISHERIES MONITORING RESEARCH · UNITED STATES
A research institute team that monitors coastal fish catches wanted an outside analysis of how catch per unit of fishing effort had changed across several years.
The scoping review showed two difficulties. The team held two versions of the catch file, one updated after a late data-entry correction, and two members read the primary question differently: the trend over the whole period, or the change after a management rule. The scope fixed one dataset version with a stated freeze date, named the trend as the primary question and the change after the rule as a secondary one, and listed a count model with an effort offset as the agreed analysis. The team lead reviewed interim results at a point set in the scope. The findings were worded as associations over time, without attributing the change to the rule.
Outcome: The team received the written scope record, the data handling log, the analysis files, the trend estimates with 95% confidence intervals and a briefing note for members who had not joined the discussions.
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
Software the Team Already Uses
The scope names the programs in which files are returned, so that the team can open, rerun and extend them without conversion.
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
What Stays With the Team When Analysis Goes Outside
Handing the analysis to OnlineSPSS transfers the statistical work, while the research, its authorship and its conclusions stay with the team.
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 study, report or publication. Authorship remains the researcher's.
Your Research Decisions The research question, the substantive interpretation, the discussion and the conclusions remain with the team, which keeps 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
Questions From Teams Commissioning an Analysis
Before an outside analysis starts, a team usually wants to know what data to hand over, how a changing question is handled and how the work is credited.
When should a team consult a statistician, and what is involved?
A team should ideally consult a statistician at the design stage, before data are collected, and at the latest once the data exist and before any model is chosen; at OnlineSPSS commissioning means sharing your question and materials, having them reviewed by a statistically trained Research Coordinator, and agreeing the analyses, deliverables, timeline and one fixed project fee before any work begins. Design questions such as sample size belong to Research Planning & Study Design. An in-house PhD-level statistician then carries out the agreed analysis on the dataset version frozen in the scope.
What data does an outside statistician need from my team?
The dataset in the form it exists, a codebook or variable list, a description of how the data were collected and the protocol or proposal where there is one. Direct identifiers are rarely needed for the analysis itself, so the team can remove them before sharing, within its own ethics approval and data-sharing rules. What is known to be wrong with the data is more useful to the analyst than a polished file.
What if our research question changes during the analysis?
A change is possible and should be written down. Models already run for the earlier question are kept as such, and results are not presented as if the new question had been planned from the start. Where a change alters the agreed analyses or outputs, it should be confirmed with the Research Coordinator before further work is done.
How is the analysis credited in the paper?
Credit follows the rules of the journal and the policy of your team. OnlineSPSS does not claim any intellectual contribution to the research, and authorship remains the researcher’s. Where a journal asks for statistical support to be acknowledged or disclosed, the wording is for the team to decide.
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 protocol, the current data version and a list of known data issues, 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
Set Out the Terms
You Already Know
Unsettled points are normal before an outside analysis, and a scope written while they are still open needs less rework than one rewritten after models have run.
No Commitment Required
Strictly Confidential
Fixed Project Fee
In-House PhD-Level Statisticians
Same-Day Response
Serving Researchers Since 2014