This section of the review should include details of the types of data extracted for inclusion in the review. Data extraction begins with recording the type of text. Once data extraction of the background details is complete, the extraction becomes highly specific to the nature of the data of interest and the question being asked in the review. In JBI SUMARI, elements of data extraction are undertaken through the text and opinion analytical module, and the data extracted is automatically transferred to the exported report.
Extracting data from text and opinion
As detailed in the protocol section, this section of the review should include details of the types of data extracted for inclusion in the review. An extraction in JBI SUMARI includes fields relating to the type of text, its authors and participants, and the content of the paper. Textual data (conclusions) are extracted from papers included in the review using the standardized data extraction tool for text and opinion reviews. The data extracted will include specific details about the phenomena of interest, populations, and any outcomes of significance to the review question and specific objectives.
It is recommended that double textual data extraction is performed independently by two reviewers.
The specific fields and types of text to extract are as follows: (see Appendix III)
1. Types of text
The type of opinion that is being extracted, for example, an expert opinion, a consensus guideline, conference proceedings, policy reports or reports accessed from web pages of professional organizations.
2. Population represented
To whom the paper refers or relates.
3. Setting / Context (may be clinical, cultural or geographical)
Setting is the specific location where the opinion was written, for example, a nursing home, a hospital or a dementia specific ward in a sub-acute hospital. Some papers will have no setting at all.
The geographical context is the location of the author(s) – be as specific as possible, for example Poland, Austria, or rural New Zealand.
The cultural context is the cultural features in the publication setting, such as, but not limited to, time period (16th Century); ethnic groupings (indigenous Australians); age groupings (e.g. older people living in the community); or socio-economic groups (e.g. working class). When entering information it is important to be as specific as possible. This data should identify cultural features such as time period, employment, lifestyle, ethnicity, age, gender, and socio-economic class or context.
4. Stated allegiance/position
A short statement from the expert voice summarizing the main thrust of the publication.
5. Conclusion (with illustration from text and page number)
Use this field to describe the main finding/s of the publication. This includes an assessment of the clarity of the argument’s presentation and logic. Is other evidence provided to support assumptions and conclusions? Is it based on clinical or life experience?
Levels of credibility (Unequivocal/Credible/Not Supported) can be assigned in this section (see further detail in data synthesis section)
6. Reviewer’s conclusion
Use this field to summarize the strengths and weaknesses of the paper.
7. Notes
This section of the report should include any other notes the reviewer wants to make. It may also include techniques that have been used to analyze the data, e.g. named software program.
Has the NOTARI data extraction tool been appended to the review? Have all of the extracted findings been discussed and assigned levels of credibility in the review?