Open Conference Systems, ITC 2016 Conference

Font Size: 
POSTER: Core Affect: Two- or Three-Dimensional?
Ana Carla Crispim, Anna Brown

Building: Pinnacle
Room: 2F-Harbourside Ballroom
Date: 2016-07-03 03:30 PM – 05:00 PM
Last modified: 2016-05-22

Abstract


Introduction

Core affect is a neurophysiological state that relates to subjective feelings that are consciously accessible (Russell, 2003). Affective constructs, like core affect, have extensive empirical and theoretical evidences; however, disagreements are still encountered regarding the number of dimensions needed to explain the construct. Most researchers acknowledge the presence of valence and activation dimensions, but disagree about a third dimension, which can be dominance or approach/avoidance.

Objectives

The aim of our research was to verify dimensionality of core affect, using methodology appropriate for categorical data.

Methodology

We developed a measure with 68 items (adjectives describing emotional states) rated on a 3-points scale. The adjectives were hypothesised to be influenced by three dimensions: valence, activation, and approach/avoidance. A sample of N=422 native English speakers took the survey via Qualtrics online. In order to explore dimensionality, we performed an exploratory factor analysis (EFA) with categorical outcomes, extracting 2, 3, and 4 factors, using the Diagonally Weighted Least Squares estimator. The oblique Geomin rotation was used to interpret the factor solutions initially, and then the target rotation was used to approximate previously published 2-factor circumplex solutions for core affect.

Results

The three-factor model was identified as the most parsimonious one (RMSEA=0.03, CFI=0.95, SRMR=0.07), while the two-dimensional model clearly was not sufficient to describe the data complexity. Our final model had a good model fit (RMSEA=0.03, CFI=0.95) and most of the items with targets had factor loadings near to 0 and acceptable standard errors. The valence dimension has 63 items with significant loadings, the activation dimension has 44 items with significant loadings, and the approach/avoidance dimensions has 64 items with significant loadings.

Conclusions

The factor analysis provided evidence for a three-dimensional model of core affect. As a result of this study, a shorter three-dimensional measure of core affect was developed.


An account with this site is required in order to view papers. Click here to create an account.