Designing and evaluating user interface mechanisms for affect labeling to enhance online discussion environments
Carvalho, Mariana (2020)
Carvalho, Mariana
2020
Ihmisen ja teknologian vuorovaikutuksen maisteriohjelma, FM (engl) - Master's Degree Programme in Human-Technology Interaction
Informaatioteknologian ja viestinnän tiedekunta - Faculty of Information Technology and Communication Sciences
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2020-05-19
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tuni-202005034880
https://urn.fi/URN:NBN:fi:tuni-202005034880
Tiivistelmä
The development of technology and diversification of means of communication have in-creased the range of social interactions that people have in their daily lives. As an adverse side effect, it is not uncommon to see heated and uncivil discussions happening online, creating a hostile environment that highly affects both the users and the reputation of the services. This is a particularly troublesome problem in online news sites, which different studies have shown to suffer from different negative consequences caused by hostile interactions on their online platforms.
While several theories have tried to explain why communication decays in digital medias, scholars have arrived at conflicting conclusions regarding the responsible factors. One possible approach is to better understand the processes of emotion regulation in order to ad-dress the hostile behaviors frequently seen online. Recent research suggests that affect labeling, that is, explicitly naming one’s emotional reactions, could be an option for emotion regulation online. Therefore, this thesis proposes to design and evaluate different options of affect labeling mechanisms to help individual emotion regulation, aiming to enhance the quality of online discussions.
This thesis studies design factors which might lead to higher acceptability in an affect labeling enabling mechanism. In order to do so, this work was composed by three stages. First, different news websites from around the world were benchmarked to identify mechanisms being used to ensure user engagement and quality, which were then categorized. These categories served as base for the creation of fourteen user interface designs that would enable an affect labeling process. Next, a user study was conducted, where eighteen participants were interviewed regarding their online engagement habits and perceptions on affect labeling and then evaluated six of the designs regarding their acceptability.
The study confirms some of the negative impacts of online hostility shown by previous studies, such as a general lack of motivation from all the participants to participate in online discussions. It also shows that participants’ perception on affect labelling change depend-ing on the approach used for the process to happen: obvious options, such as open text, were less preferred, while more subtle variations (e.g. reactions, emojis) were well accepted, perceived as a safer form of expression. The evaluations show that the participants preferred simplicity and a small number of steps to use a tool. At the same time, they also valued the tool providing a feeling of focus on the user, as an individual, and allowing certain level or personalization on the users’ inputs.
Based on the results, guidelines for future affect labelling designs were compiled. It would be interesting to perform new evaluations with participants with more active online engagement behavior in order to compare if the results are valid and gather new insights.
While several theories have tried to explain why communication decays in digital medias, scholars have arrived at conflicting conclusions regarding the responsible factors. One possible approach is to better understand the processes of emotion regulation in order to ad-dress the hostile behaviors frequently seen online. Recent research suggests that affect labeling, that is, explicitly naming one’s emotional reactions, could be an option for emotion regulation online. Therefore, this thesis proposes to design and evaluate different options of affect labeling mechanisms to help individual emotion regulation, aiming to enhance the quality of online discussions.
This thesis studies design factors which might lead to higher acceptability in an affect labeling enabling mechanism. In order to do so, this work was composed by three stages. First, different news websites from around the world were benchmarked to identify mechanisms being used to ensure user engagement and quality, which were then categorized. These categories served as base for the creation of fourteen user interface designs that would enable an affect labeling process. Next, a user study was conducted, where eighteen participants were interviewed regarding their online engagement habits and perceptions on affect labeling and then evaluated six of the designs regarding their acceptability.
The study confirms some of the negative impacts of online hostility shown by previous studies, such as a general lack of motivation from all the participants to participate in online discussions. It also shows that participants’ perception on affect labelling change depend-ing on the approach used for the process to happen: obvious options, such as open text, were less preferred, while more subtle variations (e.g. reactions, emojis) were well accepted, perceived as a safer form of expression. The evaluations show that the participants preferred simplicity and a small number of steps to use a tool. At the same time, they also valued the tool providing a feeling of focus on the user, as an individual, and allowing certain level or personalization on the users’ inputs.
Based on the results, guidelines for future affect labelling designs were compiled. It would be interesting to perform new evaluations with participants with more active online engagement behavior in order to compare if the results are valid and gather new insights.