Influence of Personality and Social Context
Damien Dupre, University College Dublin
Gary McKeown and Nicole Andelic, Queen's University Belfast
Gawain Morrison, Sensum Ltd.
People are sharing two types of content:
[…] I think that would qualify as not smart, but genius….and a very stable genius at that!
— Donald J. Trump (@realDonaldTrump) 06. Jan. 2018
[…]NEVER, EVER THREATEN THE UNITED STATES AGAIN OR YOU WILL SUFFER CONSEQUENCES THE LIKES OF WHICH FEW THROUGHOUT HISTORY HAVE EVER SUFFERED BEFORE.[…]
— Donald J. Trump (@realDonaldTrump) 23. Jul. 2018
An emotion is the result of internal and external modifications triggered by an “event” (Tcherkassof, 2008).
Seven characteristics distinguish Emotions to other cognitive states:
People can actively express emotions in social media by sharing texts, emojis, and medias (images, sounds or videos).
However with the wild spread of front facing cameras it is now possible to share emotions by analysing people's facial expressions.
Personality traits play an important role not only in the way people are expressing emotion (Wache et al., 2015) but also in the way people share them (Dhir et al., 2016).
According to the Big 5 model, personalities can be distingused by 5 strable traits:
Motivation to use Social Medias can also influence people's willingness to share emotions.
Ujhelyi and Szabo (2014) have identified four types of motivations that underpin sharing behaviours:
Participants were asked if they were willing to share their self-reported emotion as well as their facial expression on social media with Yes/No answers.
197 participants were recruited via online survey platforms (93 males, 97 females, age M = 44.9, SD = 14.7).
After each video, a feedback was given to the participants to increase their motivation to share their emotion:
Necessity to fit data in a Generalized Linear Mixed Models (GLMM) which incorporates both fixed-effects parameters and random effects in a linear predictor.
MASS::glmmPQL(data = df_raw,
fixed = share_sr ~
sessid * Extraversion +
sessid * Agreeableness +
sessid * Conscientiousness +
sessid * Emotional_Stability +
sessid * Openness,
random = ~ 1 | task,
family = binomial)
MASS::glmmPQL(data = df_raw,
fixed = share_fe ~
sessid * Extraversion +
sessid * Agreeableness +
sessid * Conscientiousness +
sessid * Emotional_Stability +
sessid * Openness,
random = ~ 1 | task,
family = binomial)
AUC for Facial Expression model is 0.72 whereas AUC for Self-Report model is 0.63.
Prediction more accurate in a GLMM with random effect and taking into account not only contextual and dispositional variables but also their interactions as fixed effects.
(too?) many limits of the procedure:
But a very interesting approach mixing dispositional factors (personality) and contextual factors (motivation) to evaluate social media users activity.
Willingness to Share Emotion Information on Social Media: Influence of Personality and Social Context
Damien Dupre, University College Dublin
Gary McKeown and Nicole Andelic, Queen's University Belfast
Gawain Morrison, Sensum Ltd.