ABSTRACT

Consumers regularly access user-generated content when making purchase decisions, and a growing proportion of this content is generated on smartphones rather than personal computers. The authors argue that because the use of smartphones (vs. PCs) leads consumers to write more succinctly and focus on the overall gist of the experiences they describe, consumers privilege the inclusion of emotional content on the device. Across six studies, including two field studies and four controlled experiments, the authors use both natural language processing tools and assessments made by human judges to analyze the linguistic characteristics of user-generated content. The findings show that the use of smartphones results in the creation of content that evinces more positive and, to a lesser extent, negative emotions than content generated by the use of PCs. Supportive evidence is provided for the hypothesis that the greater emotionality of smartphone-generated content is driven by the relative succinctness of the content rather than alternate mechanisms such as the timing of writing, differences in topical content, and user self-selection. Implications for research and practice are discussed.

Keywords: Mobile Marketing; User-Generated Content; Word of Mouth; Social Media; Emotion; Natural Language Processing; Computational Linguistics