I have made a selection of studies that are focused on a specific research question in a meta-analysis for a master thesis. However, it is my first meta-analysis, and it is unclear to me how to do data abstraction, test homogeneity of the studies, and test publication bias. I am using the software RevMen 5 and SPSS statistics. What are the steps to proceed after the selection of the studies? Thank you in advance.
https://training.cochrane.org/handbook/current/chapter-10
If you are using Revman, the above will help you guide what you will need to do. Otherwise, you could search in the link or another search such as PubMed.
Systematic review ----> Construct the PRISMA diagram ----> Data extraction and construct Table 1 for describing the included studies ----> Evaluate study quality -----> Meta-analysis -----> Get Forest plot and Funnel plot from RevMan (heterogeneity and publication bias are derived from those plots)
@Vaska Ristova, I think you already have enough inputs as @Shin Murakami and @Nattawut Leelakanok have provided very good resources and suggestions. It is good to have a protocol for your SR.
Writing a protocol is as previously mentioned a key step of the systematic review preparation before starting it. You could define with your team the inclusion and exclusion criteria (based on the question you are trying to answer, the protocol may differ), the number of person involved in each step of the inclusion/exclusion criteria as well as for data extraction. This would also be very helpful to a priori state on the data you want to extract from the studies, the type of ROB tool you need to use as well as in specifying the analyses that are planned.
The Cochrane collaboration site is a very good ressource as indicated. It can be complemented and adapted to your specific research area if you also look for what has been previously reported on your specific field.
Statistical analysis should also be planed a priori (with all possible covariates that may impact the results) avoiding analyses done after seeing the data (associated with a high risk of false discovery rate).
other interesting website for conducting SRMA: https://www.ccace.ed.ac.uk/research/software-resources/systematic-reviews-and-meta-analyses
share about Nattawut Leelakanok; may help example:
https://www.hsj.gr/medicine/methods-of-ergonomic-evaluation-in-upper-limb-a-review-in-agricultural-manual-work.php?aid=26935
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Meta-Analysis