Emotions in Digital Media: Subjective ratings of emotions and behavioral tendency concerning user interface (UI) interventions supporting the emotion regulation of online news commenters (vignette study) – Dataset by Aleksi Syrjämäki, Mirja Ilves, Joel Kiskola, Anna Rantasila, Poika Isokoski, Thomas Olsson, and Veikko Surakka. The data consist of subjective ratings of emotions and behavioral tendency. An online experiment investigated the perceived effects of a user interface (UI) intervention aiming to support online news commenters’emotion regulation. In the study, participants were presented with a vignette, describing a user reading online news comments, feeling agitated in response, and writing a highly uncivil comment. Then, the experimental manipulation was administered, and participants were randomly allocated to three experimental groups: 1) labeling, 2) analysis-control, and 3) guidelines-control. The groups differed in how the UI reacted to the comment. In the labeling group, the UI showed a text that the platform had analyzed the content of the user’s comment, and it indicated that the user had expressed strong emotions. In the analysis-control group, the text was similarly described as having appeared as a result of the analysis of the comment, but instead of describing the tone of the comment, it referred to posting guidelines. In the guidelines-control group, the text similarly referred to posting guidelines, but was described as having been visible below the comment field the whole time. Next, the participants assessed the commenter’s probable responses to the intervention. In four items, the participants were to rate whether the user’s emotional state would have changed to a more positive, negative, aroused and unaroused direction, when compared to how they felt while writing the uncivil comment. The rating scale was a seven-point scale varying from 1 (completely disagree) to 7 (completely agree). After that, the participants were asked in a multiple-choice item whether they believed the user would (1) send the comment without changes, (2) make the tone of the comment more positive and then send the comment, (3) make the tone of the comment more negative and then send the comment or (4) not send the comment. In addition, the study included one attention check item and two manipulation check questions. The participants who failed the attention check or either of the manipulation checks were excluded from all analyses. The file included contains the followng data: the participants' id numbers, ages and genders, the ratings of emotional state, responses to multiple-choice question, and the responses to attention check and manipulation check questions.

2023