Data Extraction (or Data Abstraction)
Once the eligible studies are identified, the next step, known as “Data Extraction” (or “Data Abstraction”), involves gathering the “required information” about each study from the journal articles and other reports of studies identified from the Data Sources.
This “required information” includes information about the characteristics of the study, such as where the study was done, how many participants were included, and what interventions were tested. It also includes information about the reported outcomes and the results for each outcome.
Notice how the ARAS study authors describe both the process for extracting this data and a list of the data items that were extracted:
The ARAS study authors also describe how they used extracted data to assess the potential risk of bias for each research study:
After researchers extract the necessary data, they next combine the data from the individual studies that were selected for the systematic review. This is known as “Data Synthesis.” When researchers use statistical tools to combine data, this is known as a “meta-analysis.” It’s important to name the statistical model and software program that were used to do this work.
In the case of the ARAS study, meta-analysis wasn’t possible, so here is how the study authors described their approach to data synthesis approach this way: