Distinct communication-behavior features significantly predict each Big-Five trait, with exclamation marks and after-class activities the strongest Neuroticism predictors¶

Claim · All claims
Evidence · 1 study (2 entries) · associational r2 · q3
Currency · newest study 2019 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q3 i? Covariance tests identify significant LASSO predictors per trait, e.g., question marks and keyboard activities for Openness, chat message counts for Conscientiousness and Extraversion. → Wen Wu 2019 q3 i? Exclamation marks in chat messages and total after-class activities are the strongest significant predictors of Neuroticism. → Wen Wu 2019 (2)

Evidence¶

Wen Wu 2019¶

Wen Wu, Li Chen, Qingchang Yang, You Li. (2019). Inferring Students' Personality from Their Communication Behavior in Web-based Learning Systems. International Journal of Artificial Intelligence in Education. https://doi.org/10.1007/s40593-018-00173-9

q3 · i? · associational · r2

Significance of predictors entering the LASSO model was tested via the covariance test at p<0.05. For Openness to Experience, predictors included number of question marks in chat messages (p=0.0002), total keyboard behavior, and tentative words in forum messages.

"Table 6 lists the significant predictors of Big-Five personality traits, where the significance (p-value) is determined via the covariance test for LASSO (Lockhart et al. 2014)."

Wen Wu 2019 (2)¶

Wen Wu, Li Chen, Qingchang Yang, You Li. (2019). Inferring Students' Personality from Their Communication Behavior in Web-based Learning Systems. International Journal of Artificial Intelligence in Education. https://doi.org/10.1007/s40593-018-00173-9

q3 · i? · associational · r2

From Table 6: exclamation marks in chat messages (covariance test statistic 16.7257, p=0.0000) and after-class activities (11.6508, p=0.0000) are significant Neuroticism predictors, with the largest test statistics in the table.

"# of exclamation marks in chat messages 16.7257 0.0000

of after-class activities 11.6508 0.0000"¶

Discussion¶

Related Claims¶

• Conscientiousness and Extraversion positively predict chat message posting, while Extraversion positively predicts forum posting — possibly the same claim (merge candidate)
• A LASSO model infers students' Big-Five personality from communication behavior with significantly better accuracy than a mean baseline — a broader claim this one bears on
• Students' Big-Five personality significantly influences their usage of communication tools in a web-based learning system, with 32 of 120 behavioral features significantly affected — a broader claim this one bears on