Skip to main content
. Author manuscript; available in PMC: 2018 Jan 31.
Published in final edited form as: CSCW Conf Comput Support Coop Work. 2017 Feb-Mar;2017:1217–1230. doi: 10.1145/2998181.2998213

Table 1.

The proportion of user-written posts that were labeled as trolling (and proportion of words with negative affect) was lowest in the (PosMood, PosContext) condition, and highest, and almost double, in the (NegMood, NegContext) condition (highlighted in bold).

Proportion of Troll Posts Negative Affect (LIWC)
PosMood NegMood PosMood NegMood
PosContext 35% 49% 1.1% 1.4%
NegContext 47% 68% 2.3% 2.9%