The effect of computer-mediated communication on the expression of opinions

Abstract

The aim of this study was to determine if computer-mediated communication (CMC) has a significant effect on the strength with which people express their opinions. Written CMC data and spoken face-to-face (F2F) data were generated through an online discourse production task (DPT) and an interview in which participants provided their opinion on a topic. This data was quantitatively analysed by scoring appraisal words for sentiment strength using AFINN, a sentiment lexicon. Hypothesized was that computer-mediated opinionated discourse would show a stronger sentiment than F2F discourse, largely because of anonymity and lack of social cues in CMC. A t-test was performed and showed a significant difference in sentiment strength between the computer-mediated and F2F opinionated discourse. A subsequent qualitative analysis was carried out to uncover underlying patterns and was done according to Appraisal theory (Martin & White, 2003). It was expected that computer-mediated discourse would show higher sentiment strength than F2F discourse when the same topic is addressed. Secondly, more modality tools are expected to be employed in F2F discourse. Results indeed showed a stronger sentiment in CMC discourse and a greater use of modality tools in F2F discourse. Most opinionated discourse research in CMC has been done on microblogs for marketing purposes and aimed to identify the orientation of microblog posts. This research is not interested in orientation, but in sentiment strength.