7 Steps of Meta-Analysis.
This page contains instructions on how to conduct a meta-analyis from its beginnings through its completion. For explanations of each step in greater detail, click the corresponding link.
01
Understand What Meta-Anlaysis is.
Meta-analysis is a powerful tool used to review the existing research in a particular field. However, unlike a literature review, which produces as qualitative representation of the field, meta-analyses QUANTITATIVELY summarize an area.
02
Establish a Question.
Before you can go any further, you must establish your question. This question will determine not only the search terms you use to collect studies, but also what your inclusion criteria will be, as well as the moderators you eventually decide to analyze. Choose wisely!
03
Collect Studies.
The study collection process begins the initial phase of your data collection. It is important that you choose your search terms carefully, in order to obtain all the relevant articles. Be sure to keep a detailed record of your collection process as well.
04
Create a Coding Scheme.
The coding framework that you design will determine the data that you collect from each study. There are certain pieces of information that every coding scheme should include, but the majority of the columns should be specific to your variable of interes, including potential outside factors that could moderate (affect) the overall results each study reports.
05
Code Articles.
As the coding process begins, carefully read each article and enter its data into the corresponding columns in your data sheet. Make sure you understand how to derive multiple effects from a single study, as well as whether or not your question requires a control group comparison.
06
Analyze.
There are 7 steps to the analysis process. Once you have completed the coding process, you will need to decide if your data is standardized or unstandardized, as well as which model best represents the effect size you are trying to calculate. Then, you will begin analyzing your data using SPSS macros.
07
Draw Conclusions.
As you draw conclusions from your findings, there are a number of specific ways in which you must display your data. In addition, keep in mind that null findings for your various moderators are just as important to discuss as significant findings.