For full-text data extraction, each reviewer used a separate structured extraction matrix coded in a spreadsheet. Inter-rater reliability was assessed on a random 15% subsample based on agreement ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results