Table 1.
Dataset characteristics.
Table 2.
Editors with at least 100 comments by status and gender.
Table 3.
Description of LIWC measures (as per http://www.liwc.net).
Table 4.
Example messages with their corresponding LIWC Positive and Negative scores; SentiStrength Positive (P+) and Negative scores (N-); and ANEW Valence scores.
Table 5.
Emotions and Status: Administrators promote a generally neutral tone on article talk pages.
Table 6.
Emotions and Status: Admins are more emotional in personal talk pages compared to the article talk pages - they express more positive emotion compared to regular editors, but also more anxiety and sadness.
Table 7.
Dialogue and Status: Administrators are more impersonal in article talk pages. Regular editors are more concerned with others.
Table 8.
Dialogue and Status: Regular editors send messages in a socio-emotional speech style. Administrators are more socially detached.
Table 9.
Emotions and Gender: Female Wikipedia editors express more positive emotion than male editors.
Table 10.
Emotions and Gender: Female editors express more positive emotion than males, while the expression of negative emotion is generally similar for men and women.
Table 11.
Emotions, Gender and Status: Wikipedia female administrators express more positive emotion than male administrators in article talk pages, but are similar in the expression of negative emotion.
Table 12.
Emotions, Gender and Status: Wikipedia female administrators are more emotional than male administrators in personal talk pages.
Table 13.
Emotions, Gender and Status: Wikipedia male regular editors differ significantly from male admins.
Table 14.
Dialogue and Gender: Female editors use a relationship-oriented speech style.
Table 15.
Dialogue and Gender: Female editors write and receive more relationship-oriented speech on personal talk pages.
Table 16.
Dialogue, Status and Gender: Male admins are the least relationship-oriented, female regular editors are the most relationship-focused.
Table 17.
Emotional congruence: Surplus of positive vs. negative differences between metrics of replies and messages to which they reply.
Table 18.
Emotional assortativity: large positive Z-scores indicate homophily in the “reply” network, according to emotions in the messages written by each editor.
Figure 1.
Assortativity in the reply network according to the expression of anger.
The color of each node depends on the proportion of words expressing anger in the comments written by the corresponding editor, from blue (low) to red (high). Two editors are connected if they exchanged at least 10 replies in article talk pages. Node size is proportional to the number of connections.
Table 19.
Language assortativity: large positive Z-scores indicate homophily in the “reply” network, according to the communication style of the messages written by each editor.