Studies that look at treatment effects or other effect sizes

Dichotomous (number of events)

  • Unmatched groups, prospective (e.g., controlled trials, cohort studies)
    • Events and sample size in each group
    • Non-events and sample size in each group
    • Events and non-events in each group
    • Event rate and sample size in each group
    • Chi-squared and total sample size
  • Matched groups, prospective (e.g., crossover trials or pre-post designs)
    • Matched 2x2 table
    • Events and sample size in each treatment, and external correlation
    • Non-events and sample size in each treatment, and external correlation
    • Events and non-events in each treatment, and external correlation
    • Event rate and sample size in each treatment, and external correlation
  • Unmatched groups, retrospective (e.g., case control studies)
    • Exposed and unexposed for cases and controls
    • Exposed and total for cases and controls
    • Proportion exposed and total for cases and controls
    • Matched 2x2 table for case-control study
  • Computed effect sizes
    • Odds ratio and confidence limits
    • Log odds ratio and standard error
    • Log odds ratio and variance
    • Peto's (O-E) and V
    • Risk ratio and confidence limits
    • Log risk ratio and standard error
    • Log risk ratio and variance
    • Risk difference and confidence limits
    • Risk difference and standard error
    • Risk difference and variance

Continuous (means)

  • Unmatched groups, post data only
    • Mean, SD and sample size in each group
    • Difference in means, common SD, and sample size
    • Cohen's d (standardized by pooled within-groups SD) and sample size
    • Means, sample size, and t-value
    • Difference in means, sample size, and t-value
    • Sample size and t-value
    • Means, sample size, and p-value
    • Difference in means, sample size, and p-value
    • Sample size and p-value
  • Unmatched groups, pre and post data
    • Means, SD pre and post, N, in each group, Pre/Post Corr
    • Means, SD difference, N, in each group, Pre/Post Corr
    • Means pre and post in each group, t within groups, N
    • Means pre and post in each group, p within groups, N
    • Means pre and post in each group, F for difference between changes, N
    • Mean change, SD pre and post, N, in each group, Pre/Post Corr
    • Mean change, SD difference, N, in each group, Pre/Post Corr
    • Mean change in each group, t within groups, N
    • Mean change in each group, p within groups, N
    • F for difference between changes, N
  • One group (pre-post) and matched groups
    • Mean difference, SD of difference, and sample size
    • Means, SD Pre, SD Post, Pre/Post correlation, and sample size
    • Means, sample size, and paired t-value
    • Means, sample size, and paired p-value
    • Mean difference, sample size, and t-value
    • Mean difference, sample size, and p-value
    • Sample size and t-value from paired t-test
    • Sample size and p-value from paired t-test
  • Computed effect sizes
    • Raw difference in means and confidence limits (independent groups)
    • Raw difference in means and standard error (independent groups)
    • Raw difference in means and variance (independent groups)
    • Cohen's d (standardized by pooled within-groups SD) and confidence limits
    • Cohen's d (standardized by pooled within-groups SD) and standard error
    • Cohen's d (standardized by pooled within-groups SD) and variance
    • Hedges's g (standardized by pooled within-groups SD) and confidence limits
    • Hedges's g (standardized by pooled within-groups SD) and standard error
    • Hedges's g (standardized by pooled within-groups SD) and variance
    • Raw mean difference and confidence limits (paired study)
    • Raw mean difference and standard error (paired study)
    • Raw mean difference and variance (paired study)
    • Cohen's d (standardized by SD of difference scores) and confidence limits
    • Cohen's d (standardized by SD of difference scores) and standard error

Correlation

  • Computed effect sizes
    • Correlation and sample size
    • Correlation and standard error
    • Correlation and variance
    • Fisher's Z and sample size
    • Fisher's Z and standard error
    • Fisher's Z and variance
    • Correlation and t-value
    • t-value and sample size for correlation
    • p-value and sample size for correlation

Rates (events by person years)

  • Unmatched groups
    • Events and person years in each group
    • Event rate and person years in each group
  • Computed effect sizes
    • Rate ratio and confidence limits
    • Log rate ratio and standard error
    • Log rate ratio and variance
    • Rate difference and confidence limits
    • Rate difference and standard error
    • Rate difference and variance

Survival (time to event)

  • Computed effect sizes
    • Hazard ratio and confidence limits
    • Log hazard ratio and standard error
    • Log hazard ratio and variance


Studies that look at events or means for one group

Dichotomous (number of events)

  • Raw data
    • Events and sample size
    • Non-events and sample size
    • Events and non-events
    • Event rate and sample size

Continuous (means)

  • Raw data
    • Mean, SD and sample size
    • Computed effect sizes
    • Mean and confidence limits
    • Mean and standard error
    • Mean and variance

Rates (events by person years)

  • Raw data
    • Events and person years
    • Event rate and person years


Generic effects analyzed in raw scale units

Data analyzed in raw scale

  • Computed effect sizes
    • Point estimate and confidence limits
    • Point estimate and standard error in raw units
    • Point estimate and variance in raw units


Generic effects analyzed in log units

Data analyzed in log scale

  • Computed effect sizes
    • Point estimate and confidence limits in raw units
    • Point estimate and standard error in log units
    • Point estimate and variance in log units
people working
Comprehensive Meta-Analysis

Download 10 Day Trial


"CMA 2.0 is perhaps the most user‐friendly statistical software I have used to date. It makes the complicated process of conducting a meta‐analysis much easier. It can be used from start to finish ‐ from direct data entry to publication‐quality graphics. I highly recommend this program!"

Nathan Thoma, M.A. - Psychology Intern, Payne Whitney Clinic, Weill Cornell Medical Center, New York ‐ Presbyterian Hospital


"I found the Comprehensive Meta‐Analysis software program to be extremely user friendly, providing instant computational data from the simplest to the most complex statistical problems, a versatile database to help organize and restructure large volumes of multifaceted data, and parallel visuals that help better understand your data. I also found the support staff very helpful in providing responses to both my technical questions about the program itself as well as helpful addressing more general theoretical questions about meta‐analyses. I would highly recommend this program to a novice or more seasoned researcher interested in conducting meta‐analytic statistics."

Talin Babikian, PhD, MPH - UCLA

Join our E-mail Newsletter

Valid e-mail is required.
Learn about new meta-analysis workshops & software features, and receive special tools to help with your research.

Comprehensive Meta-Analysis

software image

Comprehensive Meta-Analysis (CMA) is a powerful computer program for meta-analysis. The program combines ease of use with a wide array of computational options and sophisticated graphics.