RESEARCH SUPPORT
Online Stata support
when results do not reproduce
Support starts from the dataset, do-files and logs you share. The cause of a difference between runs is found in those files, and the do-files, logs and tables that come back are arranged so that a rerun on your own installation of Stata reproduces them.
No Commitment Required
Strictly Confidential
In-House PhD-Level Statisticians
Fixed Project Fee
Same-Day Response
How Does Online Stata Support Work?
Online Stata support at OnlineSPSS is help with a result that does not reproduce: a do-file gives different numbers on another machine, another release or a second run, and the cause has to be found in the files. It is research-analysis support, not StataCorp support, licensing or installation help. A do-file records the commands in order and a log records what Stata returned, so a Stata reproducibility check reads the do-file for paths, version, seeds and installed commands, and compares the log with the dataset to confirm which sample and options produced a table. The same routine serves Stata analysis help for a thesis chapter whose table cannot be regenerated. It is carried out by in-house PhD-level statisticians at OnlineSPSS, a specialist statistical consultancy for academic, clinical and professional research, and the study and its conclusions are yours.
WHEN TO ENGAGE
Cases Where a Remote Stata Project Fits
Distance is rarely an obstacle for Stata work, since everything the analysis needs already exists in files that can be exchanged.
Results Differ Between Two Machines The same do-file and dataset give different coefficients on your computer and a coauthor's, and the difference has to be traced to the release, an installed command or the data before either version is reported.
Do-File That Will Not Rerun An inherited do-file stops with an error, depends on a folder path or a community-contributed command, or gives numbers that differ from the table in the draft.
Log Without a Do-File Only a log or a set of screenshots survives from earlier analyses, and the commands behind the tables have to be reconstructed and checked against the data.
Sensitive Data, Controlled Sharing The dataset contains confidential information, and the arrangement for sharing it, including an NDA where required, is settled before any file changes hands.
REMOTE WORKING PARTS
How a Remote Stata Project Is Organized
Working at a distance puts weight on the files themselves, so each card concerns what is exchanged and what it must contain.
Shared Files
A remote Stata consultation starts with the dataset, any do-files and logs, and a description of the design and variables, and further files are requested if the first set leaves a gap.
Log Against Do-File
Each log is read against its do-file and dataset to confirm the estimation sample, the options and the version that produced a table in your draft.
Written Queries on the Files
Points the files leave open, such as a variable's coding or a sample restriction, are put to you in writing, so each answer becomes part of the record of the analysis.
Models Fitted From a Do-File
The agreed models are fitted by do-file within Quantitative Data Analysis, each run against the dataset version named in the query and decision record, so later changes to the data are visible.
Check of Earlier Work
Analyses already run elsewhere can be rerun and compared with the originals through Independent Statistical Review, with any discrepancy explained.
Return of Results
Output is returned as files that run on your own installation of Stata, arranged so that a coauthor can repeat the run and obtain the same tables.
BREAKS BETWEEN MACHINES
Why Stata Results Fail to Reproduce
A result that differs between two machines nearly always has an identifiable cause in the files, and most causes are ordinary.
Folder Paths Tied to One Machine A do-file loads data from a folder that exists only on the original computer, so it stops on any other machine until the paths are made portable.
Interactive Edits The dataset was changed in the data editor or by commands typed into the window and never saved to the do-file, so rerunning the file does not recreate the data.
Random Elements Unseeded Imputation or simulation was run without a seed, so each run gives slightly different numbers and the published table cannot be regenerated exactly.
Version Differences A do-file prepared for an older release is run on a newer one, or the reverse, and a changed default or output label alters what the table shows.
Package Not Installed A community-contributed command used in the original analysis is missing on the second machine, so the do-file stops at that line and the original estimates cannot be regenerated.
Ties Broken by Chance in a Sort A step that keeps the first record per group follows a sort with tied values, and Stata orders ties arbitrarily unless the stable option is used, so reruns keep different records.
DELIVERABLES
Files That Make the Work Repeatable
Each file is meant to be run, not only read, and together they let you or a reviewer reproduce the numbers on your own installation.
Single Run-All Do-File A do-file that runs the whole analysis in order from the dataset, with comments marking each stage and a header giving the Stata version.
Final Log File The log of the complete run, kept as the record of what Stata returned for every model and table.
Log Review Findings Where earlier logs were supplied, a list of matches and discrepancies between the logs, the do-files and the reported tables.
Query and Decision Record Each question raised about the files, the answer received and the analytic decision that followed, dated, so a coauthor can see why the analysis took its shape.
Exported Model Tables Tables with estimates, confidence intervals and the observations used, exported in a form ready to place in your document.
Plain-Language Output Notes Notes explaining the main blocks of output and which numbers your draft should cite, prepared for readers who do not use Stata.
ILLUSTRATIVE EXAMPLE
A Do-File That Would Not Rerun
An illustration of how a remote review of a do-file and its log resolved a table that would not reproduce.
ILLUSTRATIVE EXAMPLE · MARITIME PORT EFFICIENCY RESEARCH · UNITED STATES
A transport economics researcher held a do-file from a coauthor on another continent and could not reproduce the regression table in a working paper on container dwell times at ports.
The log showed that the original model had run on fewer observations than the shared dataset held. An earlier line, typed interactively and never saved to the do-file, had dropped ports with missing fuel price data, so the do-file as shared produced a larger sample and different coefficients. The step was reconstructed from the log, added to the do-file with a comment, and the model was rerun. The Stata version and one community-contributed command were recorded, the original clustering by port was kept for the repeated yearly observations, and the dwell-time estimates remained worded as associations.
Outcome: The researcher received the run-all do-file, the final log, the log review findings, the query and decision record and the exported tables with 95% confidence intervals.
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
File Formats Between Stata and Other Programs
Remote projects often involve a coauthor who works elsewhere in another program, so the formats and conventions that cross between programs matter.
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
Remote Collaboration, Research Authorship
Working from shared files changes where the work happens, and the ownership of the research question and its conclusions stays where it was.
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 paper or to the analysis files returned to you. Authorship remains the researcher's.
Your Research Decisions The research question, the interpretation of the estimates and the conclusions stay with you, as does overall responsibility for the study.
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
Transfers, Releases and Inherited Do-Files
Data transfer, software versions and the status of an inherited do-file are the practical matters that come before a remote Stata project.
What is a Stata reproducibility check?
A Stata reproducibility check at OnlineSPSS reruns your analysis from its do-file and log on the data you share, finds why a result differs between runs, machines or releases, and returns a do-file and log that you can rerun on your own installation. A statistically trained Research Coordinator is your point of contact, and an in-house statistician carries out the check.
Do I need to send a do-file, or is a dataset enough?
A dataset and a description of the research question are sufficient for a new analysis. A do-file and log are useful where they exist, because they show what was done before and let earlier results be compared with new ones, but nothing requires them. Where they are missing, the work begins with the dataset and the questions.
Will the do-file run on my copy of Stata?
It is prepared for the Stata release you report, and the version, any seeds and the community-contributed commands are recorded. Differences between releases can change some defaults or output labels, so a result that differs on a much older release is investigated before the difference is attributed to the data.
Can a log file show that a published table was produced correctly?
A log shows what Stata returned at the time, so read with the do-file and the dataset it can confirm the sample, options and estimates behind a table. It cannot show that the data were collected or coded correctly, or that the model suits the design, which are judged separately.
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 dataset, any do-files and logs and notes from your coauthors, 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 Stata Project Run
From Your Own Files
Distance changes little in a Stata project when the dataset, the do-files and the logs travel together, since anyone holding all three can follow how each table was made.
No Commitment Required
Strictly Confidential
Fixed Project Fee
In-House PhD-Level Statisticians
Same-Day Response
Serving Researchers Since 2014