go backGo Back
sub-category Research

Basics of when and how to perform a Meta-analysis

  • Charlesworth Author Services

Basics of when and how to perform a Meta-analysis

A meta-analysis is a quantitative (statistical) means by which to determine whether a particular effect reported in literature is real. In essence, a large number of studies is combined, and their data extracted and evaluated in a formulaic (statistical), unbiased manner, to determine validity and to summarise the results.

When to perform a meta-analysis

Essentially, it is best suited to quantitative data, as the use of the statistical methods in a meta-analysis are powerful tools to account for differences in sample sizes and variabilities in study approaches, and therefore findings. Meta-analyses also include tools to determine whether the protocol used is sensitive to differences in study selection. 

Common use of meta-analysis

A very common setting where meta-analysis is used is the pharmaceutical industry. In order to determine whether a drug has a bona fide effect, it is very common practice to look at the existing body of data and perform a meta-analysis. This will usually inform whether a pharmaceutical company takes on a project or not, and can highlight areas where more work is needed. 

Meta-analysis vs. systematic review

In a systematic review, a specific research question is answered by collecting empirical evidence according to inclusion and exclusion criteria. In fact, a systematic review may be a concomitant requirement of a meta-analysis.

How to perform a meta-analysis 

Here, broadly, are the stages in which a meta-analysis is conducted.

  1. Define a research question. This will often be a binary question, where an effect is either shown or not shown across a number of studies. A common question posed for a meta-analysis is the efficacy of a drug intervention in clinical studies, where potentially multiple studies are conducted across groups, and an overall answer for whether the intervention is efficacious or not is required.
  2. Perform a systematic review. A systematic review is often undertaken to supplement the findings of the meta-analysis, and to identify sources of data to analyse.
  3. Extract the data, including sample sizes, methods and specific measures used to signify data variability. This is likely to be a huge wealth of data, which now needs to be standardised, with some studies being weighted differently to others, depending on sample size and data variability.
  4. Estimate the overall effect using statistical modelling. The goal here is to compute a quantitative outcome measure, representative of the data being analysed, and taking into account variation caused by both random observations and methodological differences. Numerous approaches exist, going from simple linear modelling methods all the way to more complex, higher order approaches. Needless to say, performing a meta-analysis requires a highly specific skillset, which can be gained using online courses and other literature.

Conclusion

Taking on a meta-analysis is no small task, and can take several months to complete. As a single researcher, you may find this an overwhelming undertaking, so the best way forward is not to plunge yourself in at the deep end, but to rather take it slowly. There are many free resources and online courses available to get support in performing meta-analyses. So, take advantage of those, and you’ll be on your way to writing your first meta-analysis.

 

Charlesworth Author Services, a trusted brand supporting the world’s leading academic publishers, institutions and authors since 1928. 

To know more about our services, visit: Our Services

Visit our new Researcher Education Portal that offers articles and webinars covering all aspects of your research to publication journey! And sign up for our newsletter on the Portal to stay updated on all essential researcher knowledge and information!

Register now: Researcher Education Portal

Maximise your publication success with Charlesworth Author Services.

Share with your colleagues

cwg logo

Scientific Editing Services