Skip to main content

Table 4 Author and systematic review characteristics

From: The prevalence of and factors associated with inclusion of non-English language studies in Campbell systematic reviews: a survey and meta-epidemiological study

Variables LOE-restricted reviews (n = 15) LOE-open reviews (n = 84) LOE-inclusive reviews (n = 17)
Author characteristics  
 Number of authors  
  Mean (SD) 4.2 (1.2) 4.6 (2.6) 4.9 (2.1)
  Median 4 4 5
 Number of institutions represented by authorsa  
  Mean (SD) 2.3 (1.0) 2.4 (1.6) 2.8 (1.9)
  Median 2 2 3
 Number of different working countries represented by authorsa  
  Mean (SD) 1.3 (0.6) 1.3 (0.7) 1.8 (1.3)
  Median 1 1 1
Questionnaire variables (n = 6) (n=30b) (n=9b)
 Authors’ languages  
  Mean (SD) 1.3 (0.5) 3.2 (2.2) 4.2 (2.5)
  Median 1 2 4
 Authors’ nationalities  
  Mean (SD) 2.0 (0.9) 2.1 (2.2) 2.7 (1.2)
  Median 2 2 2
Methodological inclusion criteria  
 RCTs (%) 8 (53.3) 31 (36.9) 4 (23.5)
 Quasi-experiments (%) 3 (20.0) 48 (57.1) 12 (70.6)
 Non-experiments (%) 3 (20.0) 5 (6.0) 1 (5.9)
 Unclear (%) 1 (6.7) 0 (0.0) 0 (0.0)
Scope of search strategy  
 Languages of search terms  
  Mean (SD) 1 (0.0) 1.3 (1.1) 1.4 (1.5)
  Median 1 1 1
 Databases, registers and journals searched in  
  Mean (SD) 34.6 (36.8) 25.3 (23.0) 34.4 (22.0)
  Median 26.0 21.5 26.0
 Experts contacted  
  Yes (%) 9 (60.0) 63 (75.0) 14 (82.4)
  No (%) 6 (40.0) 21 (25.0) 3 (17.6)
  1. The seven LOE-undefined reviews are not included in the regression analyses. The total number of reviews included in the analysis is thus 116
  2. aIn 35 reviews, one or more author institutions were not reported accounting for 18.4% of the total 570 authors
  3. bTwo respondents did not give complete answers about the nationalities and languages of their review team; thus, there was one missing value for LOE-open and one for LOE-restricted
  4. N/A Not Applicable, i.e. no methodological inclusion criteria were unclear