RESEARCH SUPPORT
Stata for Academic Research
Analysis Prepared for Submission and Replication
Stata regression tables for publication, built by code, for faculty and doctoral researchers preparing an article: models suited to the design, tables in the journal's format and a replication package that a reviewer or data editor can run.
No Commitment Required
Strictly Confidential
In-House PhD-Level Statisticians
Fixed Project Fee
Same-Day Response
What Does Stata for Academic Research Involve?
Stata for academic research is the use of Stata to produce results that can be published and verified: models suited to the design, tables in the target journal’s format and code that reproduces every reported number from the source data. Many economics journals, and a growing number in other fields, require or request replication materials, and referees may ask for additional specifications, so the analysis is built to withstand both. OnlineSPSS, a specialist statistical consultancy for academic, clinical and professional research, provides this support to faculty and doctoral researchers: an in-house PhD-level statistician carries out the agreed analysis in Stata and documents it so that the authors can stand behind it. Authorship of the article and every claim in it remain with you.
WHEN TO ENGAGE
Stages of a Manuscript at Which Stata Work Is Needed
Requests come at different points in the life of a manuscript, from the first draft of the empirical section to a resubmission.
Empirical Section to Draft The data and the design are settled, and the models, tables and figures for the empirical section of the manuscript have to be produced and interpreted.
Data and Code Requirements The target journal asks for a data and code package with the submission, and the analysis must be organized so that it can meet the request.
Revise and Resubmit Referees ask for additional specifications or subgroup results, and the new analyses have to fit with the ones already reported in the paper.
Inherited Code, New Paper A paper extends a dissertation or an earlier project whose code was written for other aims, and it has to be checked before it is reused.
WHAT THE SUPPORT COVERS
From Models to a Replication Package
A journal article has two readers to satisfy, the referee who judges the models and the data editor who reruns them.
Specification for the Design
Models are chosen from the design and the question and described so that their assumptions can be stated in the methods section, following the conventions of the field, as in Economics.
Main and Additional Results
The main specification and any prespecified alternatives are kept apart from exploratory ones, so that the paper reports what was planned and what was added afterwards.
Journal-Format Tables
Exporting Stata tables to Word or LaTeX is done from stored estimates with esttab or etable, so no number is typed by hand, and the tables follow the journal's style through Results Reporting & Interpretation, with intervals and the sample for each column.
Replication Files
The package holds the code, a readme and the data or instructions for obtaining them, arranged so that a data editor can run them in order and reach the published tables.
Restricted Data
Where the data cannot be shared, the package documents how to obtain them and the code runs once access is granted, within the data provider's terms.
Review Before Submission
A completed analysis, including one written by a coauthor, can be examined through Independent Statistical Review before it goes to the journal.
COMMON PUBLICATION ISSUES
How Empirical Articles Lose Credibility at Review
Referees and data editors tend to find the same problems, and each of the six below is something the analysis is built to withstand.
A Table That Cannot Be Reproduced A number in the paper cannot be regenerated because the code was edited after the table was made, so tables are produced by code from a clean run.
Specification Search Many models are tried and the one with a significant coefficient is reported, so the specifications are planned and the number tried is disclosed.
Random Steps Without a Seed Bootstrap, simulation and random sampling steps give slightly different numbers on each run unless a seed is set, so a rerun does not match the paper.
Packages Not Recorded A user-written command is updated and can change its output, so the packages used and their versions are listed with the code.
Columns With Different Samples An added control with missing values shrinks the sample, so the columns of a table no longer share one sample, and the number of observations in each column is stated.
Data Statement Beyond the Package The data availability statement promises more than the package holds, so what is shared, what is restricted and how to request it are stated accurately.
DELIVERABLES
What an Article-Ready Stata Analysis Contains
What goes into the package is set by the target journal's data and code policy, and these items cover what such policies commonly ask for.
Analysis Code Set Do-files that run in order from the source data to every table and figure in the manuscript.
Readme for Reviewers A readme describing the files, the run order, the software and its release, the data sources and which output belongs to which table or figure.
Manuscript Tables Regression tables and figures exported to Word, LaTeX or the journal's format, with estimates, confidence intervals and the number of observations in each column.
Data Availability Notes A description of the data, what can be shared and what is restricted, to support the journal's data statement.
Statistical Methods Wording Statistical wording for the methods, matched to the output and to what was done, for you to check and incorporate.
Referee Response Analyses Where a revision is under way, the additional analyses that referees requested, produced in the same code and reported consistently with the original tables.
ILLUSTRATIVE EXAMPLE
A Referee Request That Changed the Sample
An illustration of how a revision is handled within an analysis that has to be replicable.
ILLUSTRATIVE EXAMPLE · ENERGY ECONOMICS RESEARCH · UNITED STATES
A faculty member in economics had a manuscript on how household electricity use changed after a revision of the pricing schedule, using several years of monthly billing data, and the journal required a replication package.
The model was a fixed-effects specification with errors clustered by household, described as estimating the change associated with the new schedule, with the limits of the design for a causal reading stated. At revision, a referee asked for an additional control. The new variable was missing for part of the sample, so the added column of the table no longer rested on the same households as the others. The columns were rerun on one common sample, the original estimation sample was shown beside it, and the number of observations was reported in every column. A seed was set for the one resampling step, the user-written commands and their versions were recorded, and the readme matched each output to its table.
Outcome: The author received the do-files that run from the billing data to every table, a readme for the data editor, the revised tables with 95% confidence intervals, the referee-response analyses and the statistical methods wording.
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
Choosing the Platform for the Article
The platform follows the journal's field, the coauthors and the model, and the replication package is written in whichever one produced the reported results.
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
Statistical Support Within Your Own Article
OnlineSPSS provides the statistical analysis behind the manuscript, and the paper, with its claims, stays the work of its authors.
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 article. Authorship remains the researcher's, from the first draft to the published paper.
Your Research Decisions The research question, the theoretical framing, the interpretation, the discussion 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
Questions From Authors Using Stata
Replication packages, confidential data and referee requests are where authors most often need a clear answer before submission.
What does Stata analysis for an academic article include?
OnlineSPSS covers the models, the regression tables in your target journal’s format and the code that produces and reproduces them, with a replication package prepared where the journal asks for one. The scope is confirmed in the assessment and follows the manuscript and the journal’s requirements.
What does a journal's replication package usually contain?
Requirements differ by journal, but the usual elements are the code, a readme, and either the data or instructions for obtaining them, together with a data availability statement. Some journals check the package before acceptance. The author guidelines of your journal decide what is required, and the package is built to them.
Can I share code when the data are confidential?
Often yes. The code can be shared with a description of the variables and the route to obtain the data, so that it runs once access is granted. Whether any data or derived files can be released depends on the provider’s terms and your ethics approval, and journals set their own policies for restricted data, so the journal’s guidance is checked first.
How should additional specifications requested by referees be reported?
As additional results, labeled as such, with the sample, the estimates and their intervals shown, and without replacing the original specification unannounced. When the new analyses are exploratory, say so, and keep a record of how many specifications were tried so that readers can judge the evidence.
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 manuscript draft, the dataset, any existing do-files and the journal’s guidelines, 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
Ready Your Article
for Review and Replication
Name the journal, describe the data and share the draft or code you have, and the review will identify what the referees and the data editor are likely to ask of the analysis.
No Commitment Required
Strictly Confidential
Fixed Project Fee
In-House PhD-Level Statisticians
Same-Day Response
Serving Researchers Since 2014