Test
Description
The study deals with the relationship between personality and emotions in situations of decision making uncertainty. It was part of an academic research with two goals: - Psychological study: The study deals with the relationship between personality and emotions in situations of decision making uncertainty - Software engineering study: The study deals with a new data classification algorithm for generating a set of effective models We expect to identify how people with diverse personalities express different emotion-action relationships. As a key component, personality traits are expected to clarify and improve the understanding of the scale of emotions and their influence in situations of decision-making under uncertainty. We found that, the effect of the relationship between emotions and decisions differs for people of different genders and different degrees of locus of control. The novelty of this approach is that we do not aim to predict decisions or emotional uprisals during paradoxical situations or in times of uncertainty. Instead, we identified the complex patterns underlying such decision-making. We differentiated the patterns of emotional decision-making for different groups with diverse personality traits. Data: The data was collected via an online experiment (See 'Steps to reproduce'). The raw experiment data is attached hereby as 'raw_data_ellsberg.csv'. Our data preparation Python code to standardise the data and prepare it for further learning can be found here: https://github.com/dtrugman/edt/blob/main/data_prepare_ellsberg.ipynb
Files
Steps to reproduce
We conducted an experiment using Qualtrics with 200 participants (100 males and 100 females), whom we recruited using Prolific.co, an online recruitment platform. Each participant was identified using their unique Prolific ID and paid £9.26/h for their time (about 7 minutes). The participants completed the emotion elicitation questionnaire, the Ten-Item Personality Inventory (TIPI), and Rotter's locus of control questionnaire. See attached 'experiment_ellsberg.docx' for the full experiment document. Part 1 - Emotion elicitation questionnaire: The Ellsberg paradox demonstrates that people prefer to bet on known rather than unknown probabilities (Ellsberg, 1961). The lottery games were presented as a random draw of a single ball out of an urn containing 90 balls, of which 30 were red. The other 60 were either black or yellow, but of unknown numbers. The participants were informed about the two lotteries that comprised the same payoffs for drawing different balls and were asked to choose between the two. Following the decision regarding their preferred option, participants were asked to state the certainty they felt toward their choice on a 5-point scale, ranging from 1 (very uncertain) to 5 (very certain). Participants were then asked to choose the expected emotional experience if they were to participate in any of the two options described in the games (the one they chose and the other they did not). They selected their expected emotional experience across six axes of emotions, which held different opposite emotions on a 4 point scale: Anxiety-Confidence, Boredom-Fascination, Frustration-Euphoria, Dispirited-Encouraged, Terror-Enchantment, Humiliation-Pride. (An adjustment from the original, 6-point scale (Kort, Reilly & Picard, 2001), where we cut out the extreme margins from each axis that participants are less likely to experience). Part 2 - Two psychological questionnaires: 1. TIPI Scale: Brief Big Five personality questionnaire Participants were asked to choose the answer that describes them most accurately, using a brief, 10-item measure of the Big Five personality dimensions (TIPI; Gosling, Rentfrow, & Swann, 2003). The five dimensions are openness to experience, conscientiousness, extraversion, agreeableness, and emotional stability. The TIPI items were rated on a 7-point scale, ranging from 1 (strongly disagree) to 7 (strongly agree). 2. Rotter's locus of control questionnaire To expand the understanding of participants’ personalities, we examined their perceptions regarding their control of reinforcement. We used a 29-item measure of internal versus external reinforcement control. Each item of the Internal-External Locus of Control Scale presents a forced-choice pair of statements with one internally and another externally oriented. The participants were asked to select one statement of each pair that they strongly related to. The scale includes six filler items intended to make the test’s purpose more ambiguous.