Figure 2
The flowchart is titled “Identification of publications via databases,” and comprises three stages labeled on the left from top to bottom as follows: “Identification,” “Screening,” and “Included.” In the “Identification” stage, a text box on the left is labeled “Records identified from: Business Source Elite (n equals 784) and A B I or I N F O R M (n equals 586).” A right-pointing arrow leads to a text box labeled “Records removed before screening: Duplicate records removed within each source – Business Source Elite (n equals 13) and A B I or I N F O R M (n equals 6); Duplicate records removed across the two sources (n equals 517).” A downward arrow leads from the first box to “Records screened for titles or abstract asterisk (n equals 834),” with a right-pointing arrow leading to “Records excluded (n equals 718)” in the “Screening” stage. A downward arrow leads to “Records screened for full text asterisk (n equals 116),” with a right-pointing arrow leading to “Records excluded (n equals 11).” A downward arrow connects to “Final sample (n equals 105).” A downward arrow leads to the last text box labeled “Publications included in review (n equals 105): Non-Empirical (45) and Empirical (60)” in the “Included” stage.PRISMA flow diagram. * Against inclusion and exclusion criteria. Source: Author’s own work