STATISTICAL SOFTWARE
Statistical Software Consulting
In the Platform Your Study Requires
OnlineSPSS provides statistical software consulting services for researchers whose department, coauthors or data determine the platform: the method is chosen first, and the analysis is then carried out in that software, working from your files.
WHAT YOU WILL FIND
Analysis in 26 statistical packages
Data, syntax and output delivered with every project
SPSS is our flagship
WHAT YOU WILL FIND
Analysis in 26 statistical packages
Data, syntax and output delivered with every project
SPSS is our flagship
Statistical software consulting services carry out statistical analysis in the platform a study requires, with the method chosen first and the software second. A project may arrive in SPSS because a department standardizes on it, in Stata because the coauthors’ do-files are already written in it, or in NVivo because the data are interviews, and many quantitative questions can be answered in more than one platform. OnlineSPSS is a specialist statistical consultancy for academic, clinical and professional research, and the analysis is carried out in-house by PhD-level statisticians.
ALL STATISTICAL SOFTWARE
Statistical Software by Research Task
Statistical analysis in SPSS, R and Stata sits alongside qualitative, structural equation, planning and meta-analysis tools, grouped here by research task so you can go straight to the platform your project already uses.
IBM SPSS: Where OnlineSPSS Began
OnlineSPSS began with IBM SPSS and expanded to other platforms from there.
Survey, comparative, regression and scale analysis is carried out in SPSS by in-house PhD-level statisticians, working from your .sav and .spv files. Syntax is provided with the output where applicable, so every table can be rerun.
Quantitative Statistical Packages
Menu and syntax-driven analysis, with do-files, SAS programs, workfiles and project files.
Stata
Panel, survey and econometric models with do-files and logs
SAS
Clinical and observational analysis with programs, logs and output
EViews
Time-series, forecasting and panel econometrics in workfiles
Minitab
Design of experiments, quality statistics and regression
GraphPad Prism
Laboratory comparisons, dose-response curves and publication-ready graphs
MedCalc
ROC curves, diagnostic accuracy and method comparison
Jamovi
Open-source menu-driven analysis saved as .omv projects
JASP
Frequentist and Bayesian analysis saved as .jasp projects
Qualitative & Mixed Methods Analysis
Interview, focus group and document analysis, with project files, coded exports and a codebook where the method uses one.
NVivo
Coding, themes and framework matrices from interviews and documents
MAXQDA
Coded segments, matrices and mixed methods integration
ATLAS.ti
Quotations, memos and networks from source text to themes
Dedoose
Team coding linked to descriptor variables for mixed methods
Quirkos
Visual coding of qualitative data with a documented codebook
Delve
Browser-based coding, codebooks and thematic analysis
Programming & Spreadsheet Environments
Code-first and spreadsheet analysis, with .R, .qmd, .py, .ipynb, .m and documented .xlsx files.
Structural Equation & Latent Variable Modeling
CFA, SEM and PLS-SEM path models, with model syntax or project files, diagrams and full output files.
Study Planning & Evidence Synthesis
Power analysis before data collection and meta-analysis across studies, with every input recorded.
SOFTWARE SELECTION
How We Match the Platform to the Project
Six considerations decide where an analysis is carried out, and the method comes before the brand, so the platform changes only when the model or the team's own platform requires it.
Method and Estimator Availability Some models are better supported in one platform than another, so a latent class model, a Bayesian multilevel model or a network meta-analysis may call for a different tool.
Data Structure and Size Panel, survey-weighted, repeated and very large datasets differ in how each platform declares and stores their structure, which affects the setup and the speed of the analysis.
Reproducibility Files Required A replication request, or a coauthor who will rerun the work, needs syntax, scripts or do-files that rebuild every table from the source data, so the platform must produce them.
Your Team's Platform A department or research group that works in one platform needs files its members can open and extend, a strong reason to stay in it unless the model requires otherwise.
Reporting Outputs Journal tables, coded-data excerpts and forest plots are produced differently in each platform, so the format of the final results is settled before the work starts.
Qualitative or Mixed Data Interviews, open-text responses and documents are coded in qualitative software, and a mixed methods project may need a statistical package beside it, with the two kept linked.
COMMON SOFTWARE ISSUES
Where Software Setup Goes Wrong
These six problems recur across platforms, and each is checked before an estimate or a theme is reported, because most of them run without any error message.
Undeclared Missing Codes Codes such as 99 or 999 left undeclared as user-missing in SPSS enter means, correlations and models as real values, so the estimates are wrong without any warning.
Wrong Types on Import A column read as text, or a category code read as a number, is treated as the wrong kind of variable in R, Jamovi or JASP, so a grouping factor can enter a model as a slope.
Labels Lost in File Conversion Moving a dataset between platforms, or through a CSV file, can drop value labels, variable types and missing-value definitions, so the converted file is checked against the codebook before any model is run.
Notebooks Run Out of Order In a Python notebook, cells run out of order leave results that no longer match the code on the page, so the notebook is rerun from a clean start before anything is reported.
Fit-Chasing With Modification Indices Adding paths or correlated errors in AMOS or LISREL only because a modification index suggests them capitalizes on chance, so each change needs a theoretical reason and is reported as exploratory.
Overgrown Code Lists Hundreds of overlapping codes in NVivo or ATLAS.ti with no written definitions make the analysis hard to trace, so codes are reviewed, merged and documented before themes are developed from them.
ANALYSIS SERVICES
Analysis Services Delivered in Your Platform
Analysis support in statistical software at OnlineSPSS is the analysis itself, carried out in the platform your project uses, not instruction in how to use the program. These six services describe what that analysis can cover.
Research Planning & Study Design
The study design, sample size and analysis plan are settled before data collection, so the platform and the files the analysis will need are known from the start.
Quantitative Data Analysis
Data preparation, models and diagnostics are carried out in SPSS, R, Stata, SAS or Python, with syntax or scripts provided where applicable so the work can be rerun.
Qualitative & Mixed Methods Analysis
Coding and thematic, framework or content analysis are carried out in NVivo, MAXQDA or ATLAS.ti, with the project file, and the codebook where the method uses one, returned for your own use.
Results Reporting & Interpretation
Results from any platform are turned into interpreted tables, figures and methods text, prepared to the reporting standard your target journal expects and to what the design supports.
Meta-Analysis & Evidence Synthesis
Effect sizes are pooled and heterogeneity examined in tools such as R, RevMan or Comprehensive Meta-Analysis, with forest plots and the extraction data kept.
Independent Statistical Review
An analysis already run in any platform can be reviewed from its syntax, scripts and output, with a clear statement of what stands and what should change.
STATISTICAL METHODS
The Method Comes First, the Platform Second
Whichever package a project uses, the method follows the research question, the study design and the data structure. These are six of the method families we carry out across the platforms above.
Comparative & Hypothesis Testing
Group, condition and time-point comparisons
Regression & Predictive Modeling
Associations, adjusted effects and prediction
Multivariate, Scale & Reliability
Scales, factors, reliability and agreement
Structural & Latent Variable Modeling
Latent constructs, paths and measurement models
Longitudinal & Multilevel Analysis
Repeated, clustered and nested observations
Clinical, Survival & Epidemiological
Time-to-event, diagnostic and observational designs
RESEARCH DISCIPLINES
Software Used Across Research Fields
Fields tend to favor particular platforms, such as SPSS in psychology and education, and the analysis is carried out in the one your field, institution and coauthors expect.
RESEARCHER VOICE
What Researchers Say
Feedback from two research managers, on keeping the analysis in the required platform and on projects that arrive as datasets and project files without syntax.
“Our projects arrive in different software depending on the university, client, or research team, so being able to keep the analysis in the required platform matters. The team could work from existing files, preserve reproducibility, and move between statistical and qualitative packages without forcing everything into one tool.”
“I needed the work kept in IBM SPSS because that is what our team uses internally. The main problem was having .sav and .spv files but no syntax to reproduce what had already been run. OnlineSPSS corrected the setup and returned the project files with clear reporting, which made later changes much easier.”
FREQUENTLY ASKED QUESTIONS
Questions About Statistical Software Consulting
Short answers to what researchers ask before sharing a project, from choosing a platform and keeping your team's files to journals, qualitative tools and confidentiality.
Which statistical software should I use for my research?
Choose the method first: the research question, design and data structure decide which models are needed, and the platform follows from whether it supports them and what your team, institution and coauthors use. Choosing research statistics software is therefore a decision about files and reproducibility as much as features. If you are unsure, describe the project and we confirm the platform during the assessment.
Can the analysis stay in the software my team or university uses?
In most cases, yes. The platform is agreed for the project, and it is usually the one your team, department or coauthors already use, with the work carried out from the datasets, syntax, scripts and project files you already have. A different platform is suggested only when the requested model cannot be carried out properly in yours, and the reason is explained first.
Do journals expect code or a replication package?
It depends on the journal and the field. The American Economic Association’s data and code policy expects code that reproduces the published results from the source data, while many other journals only encourage sharing, so the author instructions of your target journal decide. Syntax, scripts or do-files can be provided where applicable, organized so the results can be rerun.
Do you work with qualitative data analysis software such as NVivo, MAXQDA and ATLAS.ti?
Yes. Interviews, focus groups and open-text data are coded in qualitative software with a documented coding approach (and a codebook where the analytic method calls for one), and the project file is returned with the findings. Where a study combines qualitative and quantitative strands, the two are kept linked so that the findings can be integrated.
How do you protect my research data and confidentiality?
Research data and project details are handled confidentially, and project information is not published. 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, including the platform your files are in, and a statistically trained Research Coordinator reviews the request and responds the same day. The analytical work, 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
Discuss Your
Analysis and Software Requirements
Share your research question, your data and any existing syntax, scripts or project files, and tell us which platform your team requires. We review them and confirm the approach before any work begins.
No Commitment Required
Strictly Confidential
Fixed Project Fee
In-House PhD-Level Statisticians
Same-Day Response
Serving Researchers Since 2014