STATISTICAL SOFTWARE
Comprehensive Meta-Analysis Service
Pooled Effects With an Audit Trail
Our Comprehensive Meta-Analysis service pools your extracted studies in CMA, with effect-size type and direction harmonized before entry, heterogeneity and moderators examined, and the CMA file, plots and extraction table provided where agreed so that your team can rerun it.
No Commitment Required
Strictly Confidential
In-House PhD-Level Statisticians
Fixed Project Fee
Same-Day Response
What Is Comprehensive Meta-Analysis (CMA) Used For in Research?
Comprehensive Meta-Analysis (CMA) is a commercial program for meta-analysis from Biostat. It computes effect sizes from many forms of study data, pools them under fixed-effect or random-effects models, and produces forest plots, funnel plots, subgroup analyses and meta-regression. Review teams choose it because data entry accepts the formats studies actually report, and because a single file holds the data, the analyses and the settings. OnlineSPSS, a specialist statistical consultancy for academic, clinical and professional research, provides a Comprehensive Meta-Analysis service in which an in-house PhD-level statistician carries out the agreed synthesis in CMA from your extraction data and documents each analytic decision.
WHEN TO ENGAGE
Four Points at Which Review Teams Bring Us a CMA Project
A Comprehensive Meta-Analysis statistician is usually brought in once studies are extracted, when effects will not combine cleanly, when heterogeneity must be explained or when the team needs an auditable file.
Extracted Studies Ready to Pool The studies are extracted and the synthesis question is set, and you need effect sizes computed and pooled, with heterogeneity and its possible sources reported alongside the summary estimate.
Effects in Mixed Forms Studies report means, medians, odds ratios or correlations in different forms, and the effect measure has to be harmonized, or studies handled separately, before anything is pooled.
Heterogeneity to Explain Pooled results show substantial variation between studies, and the review needs prespecified subgroup analyses or meta-regression, not a search through moderators for a significant one.
A File the Team Can Reopen Coauthors need the file, the extraction table and the settings in a form they can reopen, so that an update with new studies does not mean redoing the work.
ANALYSES IN CMA
Syntheses Carried Out in CMA for Your Review Question
Each synthesis is matched to the type of outcome and the structure of the included studies, and its card links to the method page where the modeling choices are explained.
Meta-Analysis
CMA meta-analysis pools study effects under a fixed-effect or random-effects model chosen for the synthesis question, with the pooled estimate, its confidence interval and the between-study variance reported.
Meta-Regression
CMA meta regression relates the effect size to study-level moderators, keeps the number of covariates small for the number of studies, and reads the result as an association between studies.
Proportion / Prevalence Meta-Analysis
Single-group event rates are entered as events and sample sizes and pooled on a suitable transformed scale, with the back-transformation to a proportion stated.
Relative Risk & Odds Ratios
Binary outcomes are entered as event counts and pooled as risk ratios, odds ratios or risk differences, with the measure chosen from the baseline risk and the review question.
Correlation Analysis (Pearson & Spearman)
Correlations are pooled on the Fisher z scale and converted back for reporting, with each study's sample size entered correctly and the type of coefficient kept consistent.
Systematic Review
The pooled estimate sits inside a documented review of question, search, selection and extraction, and it is described as only as strong as the studies it combines.
COMMON CMA ISSUES
Where Meta-Analyses Go Wrong Before the Pooled Estimate
Software cannot repair inconsistent input, so each check below is made on the extracted data before a pooled effect or a forest plot is reported.
Effect Type and Direction Odds ratios, risk ratios and standardized mean differences are not interchangeable, and a scale where lower is better is reversed before pooling. Mixed definitions are harmonized study by study.
Entry Format Matched to the Paper Papers report spread as a standard deviation, a standard error or a confidence interval, and a value typed into the wrong field changes the study's weight without any warning. Each spread and event-count field is read back against the paper's table.
Counts From the Right Table Event counts and group sizes differ between a paper's tables, so the numbers come from the table that matches the analysis, and intention-to-treat figures are not mixed with completers.
Several Outcomes per Study A study with multiple outcome measures or time points contributes correlated effects. One is chosen by a rule written beforehand, or the dependence is modeled, and never entered as independent.
Few Studies and Between-Study Variance With few studies the between-study variance is estimated imprecisely, so heterogeneity statistics and random-effects intervals are described cautiously and a sensitivity analysis is shown.
Small-Study Effects Funnel plots and asymmetry tests are read only when enough studies exist, usually at least ten, and asymmetry is described as small-study effects with several possible causes, not as proof of publication bias.
DELIVERABLES
What You Receive From a CMA Analysis
The files follow an auditable meta-analysis workflow, agreed in the scope and provided where applicable, so that your review team can rerun and update the synthesis.
CMA File The CMA file with the entered data, the computed effect sizes and the analyses, so that a coauthor can reopen it and reproduce every pooled result.
Extraction and Effect-Size Table A reproducible table of each study's extracted values, the effect size and its variance, and the reason for any value that was derived or converted.
Forest and Funnel Plots A CMA forest plot for each analysis showing every study's weight, and funnel plots only where at least ten studies make them readable, labeled with the effect measure and its direction.
Subgroup and Moderator Results Prespecified subgroup and meta-regression results with the heterogeneity measures, and a note of which analyses were planned and which were exploratory.
Tables and Methods Text Publication-ready tables and a methods paragraph stating the effect measure, model and software settings, in the form your target journal expects.
Statistical Interpretation Each pooled result is explained within what the included studies support, with the certainty of the evidence and its limitations stated plainly.
OTHER PLATFORMS
When Another Platform May Be More Appropriate
CMA suits many standard syntheses, but the software follows the review question, and some reviews fit another tool better.
ILLUSTRATIVE EXAMPLE
How a CMA Analysis May Look in Practice
An illustration of how studies reporting the same outcome in different forms are harmonized before they are pooled.
ILLUSTRATIVE EXAMPLE · NUTRITION INTERVENTION EVIDENCE REVIEW · UNITED STATES
A university public health group wanted to pool randomized trials of school-based nutrition programs on a measure of dietary quality, with the trials spread across several countries and instruments.
The review question and the effect measure, a standardized mean difference, were fixed in the protocol. During extraction the team found that some trials used scales where a higher score meant a better diet and others used scales where a higher score meant a worse one, and that one trial reported a standard error where the entry form expected a standard deviation. The direction of every scale was aligned and recorded, the field mismatch was corrected against the paper, and the entries were rechecked by a second person before pooling. A random-effects model was used, heterogeneity was described with its uncertainty, and the planned subgroup by age group was reported as prespecified, while no funnel plot was drawn because there were too few trials to read one, and this was stated in the methods paragraph.
Outcome: The group received the CMA file, the extraction and effect-size table with every conversion noted, the forest plots, the subgroup results and a methods paragraph describing the model and settings.
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.
RESEARCH DISCIPLINES
Research Fields Where CMA Is Common
CMA is used in medicine, psychology, education and the social sciences, and the same questions about effect measures and heterogeneity arise in every field that synthesizes quantitative studies.
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.
RESEARCHER VOICE
What Researchers Say
Feedback from an evidence synthesis researcher on consistent effect-size direction and type, and on outputs the project team could audit and rerun.
“Keeping the analysis in Comprehensive Meta-Analysis (CMA) mattered because the rest of the project team needed to rerun it later. They dealt with making sure effect-size direction and type were consistent before pooling, documented the analytical steps clearly, and gave us outputs that were much easier to audit and report.”
Statistical Support Within Your Research OnlineSPSS provides the agreed methodological, analytical and statistical reporting support, including the CMA files described in the scope. The researcher or research team retains the scientific question, substantive interpretation, conclusions and overall responsibility for the research.
FREQUENTLY ASKED QUESTIONS
Questions About CMA Analysis for Research
Short answers to what review teams ask before sharing a CMA project, from the analyses we carry out and existing files to mixed data formats, heterogeneity and confidentiality.
Which analyses do you carry out in CMA?
Our work in CMA covers pooled meta-analysis under fixed-effect and random-effects models, subgroup analysis, meta-regression, pooled proportions, correlations and binary-outcome effects, with forest and funnel plots. The choice follows your review question and the data the studies report, and it is confirmed during the assessment.
Can you work from my extraction sheet and an existing CMA file?
We work from the extraction sheet and the CMA file themselves, and not from the pooled figures alone. Existing entries are checked against the source papers before more studies are added, and a completed synthesis can be examined through Independent Statistical Review, which traces each pooled result back to the values entered and the source papers.
Can studies that report different statistics go into one CMA analysis?
Yes, when they estimate the same effect. CMA lets each study be entered in the form it reports, such as means with standard deviations, events and totals, or a test statistic with sample sizes, and computes one effect size index from each. The format used for every study is recorded in the extraction table, and a study whose result cannot be expressed on that index is analyzed separately or described without pooling.
What do I do about high heterogeneity?
First check the data for entry errors and mixed effect definitions, then examine prespecified moderators. We report the between-study variance and its uncertainty, and describe heterogeneity as a feature of the evidence, not as a problem to remove. Post hoc subgroup findings are labeled exploratory.
How do you protect my research data and confidentiality?
Your extraction sheets, CMA files and review protocol 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, such as the review protocol and extraction sheet, 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
Discuss Your
CMA Analysis With Us
Share your review question, your extraction data or CMA file and the effect measure you plan to use. We check how the studies were entered and confirm the analysis before any work begins.
No Commitment Required
Strictly Confidential
Fixed Project Fee
In-House PhD-Level Statisticians
Same-Day Response
Serving Researchers Since 2014