Table 3 Proportion of overlap between the social, collaboration, and professional support networks.

From: The impact of gender diversity on scientific research teams: a need to broaden and accelerate future research

Proportion

Men average

Women average

p-value

Effect size

(Cohen’s D)

Proportion collaborationa given socialb

0.77

0.71

0.34

0.19

Proportion collaboration given professional supportc

0.78

0.76

0.76

0.06

Proportion social given collaboration

0.60

0.58

0.77

0.06

Proportion social given professional support

0.66

0.67

0.96

0.01

Proportion professional support given collaboration

0.59

0.55

0.43

0.15

Proportion professional support given social

0.66

0.58

0.22

0.24

  1. For each pair of networks X and Y, we compute the proportion of individuals that have relation type X that also report a relation of type Y, averaged across all individuals. For example, the first row shows the proportion of collaboration relations that were also listed as social relations, averaged across all individuals. p-values are shown for t-tests comparing the average proportions between men and women. Bold numbers indicate the higher proportion in each row.
  2. aCollaboration = social network questions: grants/publications/new research.
  3. bSocial = social network questions: friend/fun (and with Team 5 it is just social). cProfessional support = social network questions: advice/mentor/student committees.
Back to article page