Close

Presentation

Investigating the Key Persuasive Features for Fitness App Design and Extending the Persuasive System Design Model: A Qualitative Approach
Event Type
Poster Presentation
TimeThursday, April 152:24pm - 2:25pm EDT
LocationDigital Health
DescriptionINTRODUCTION
Physical inactivity has been recognized as one of the leading risk factors that account for many non-communicable diseases, with the World Health Organization labeling it as the fourth leading risk factor for global mortality. This has led researchers/developers to create fitness apps to support and motivate people to engage in physical activity more regularly. However, there is limited work on how collectivist and individualist users differ in terms of the persuasive features they care about in fitness apps. Having knowledge of the cultural differences, in the context of Human Factors (HF) design, will help designers/developers create better fitness apps tailored to the two main types of culture.

OBJECTIVES
The first objective of this paper is to uncover, using Hofstede’s cultural framework, the most important persuasive features users care about using Nigeria and Canada/United States (US) as a case study. The second objective is to find out whether the Persuasive System Design (PSD) model can be extended based on the uncovering of new persuasive features users care about in the fitness domain. The third objective is to uncover the cultural differences in persuasive-feature preferences between people in collectivist societies and people in individualist societies using the PSD model as a comparative framework.

OVERVIEW OF THE PSD MODEL
The PSD model is a framework for analyzing, designing and evaluating persuasive systems. Proposed by Oinas-Kukkonen and Harjumaa, the PSD model comprises four categories: Primary Task Support, Dialog Support, Credibility Support, and Social Support. Primary Task Support includes persuasive features that help and motivate users to carry out the target behavior. Dialog Support includes persuasive features that motivate users to perform the target behavior through the mechanism of feedback and system-user interaction. Social Support includes socially oriented persuasive features that motivate users to perform the target behavior. Finally, Credibility Support includes persuasive features that make the persuasive system appear credible and trustworthy. Each of these categories contain 7 persuasive features in the current PSD model as shown below:

Primary Task Support: Reduction, Tunneling, Tailoring, Personalization, Self-Monitoring, Simulation, Rehearsal.

Dialog Support: Praise, Rewards, Reminders, Suggestion, Similarity, Liking, Social Role

Credibility Support: Trustworthiness, Expertise, Surface Credibility, Real-World Feel, Authority, Third-Party Endorsements, Verifiability

Social Support: Social Learning, Social Comparison, Normative Influence, Social Facilitation, Cooperation, Competition, Social Recognition


METHOD
We used a qualitative approach to address the study’s objectives. The overarching research question is, “What are the key features of a fitness app users care about and how are they moderated by a culture?” It is broken into the following subquestions:

RQ1. What key persuasive features make users want to use a fitness app?
RQ2. Can the PSD model be extended for fitness app design based on the uncovering of new persuasive features users care about?
RQ3. Are the key persuasive features moderated by culture?

To contextualize the study, we developed a mocked-up fitness app (homepage), which we presented to the study participants and asked the question: “Please enter here [textbox provided] one key feature you would expect the app to have if you were to use it.” To analyze our data, we conducted a thematic analysis on the participants’ responses. A total of 189 validated participants from Canada/US (individualist) and 67 validated participants from Nigeria (collectivist) completed the survey.

RESULTS
In total, 17 supportive/persuasive features were teased out of the thematic analysis of participants’ comments. They include Exercise Monitor/Tracker, Reminder/Notification, Behavior Modeling, Goal-Setting, Exercise Timer, Workout Plan/Routine, Suggestion/Recommendation, Exercise Tips, List of Exercise, Reward, Social Support, Goal/Calorie/Data Input, Motivational Quotes, Exercise Summary Statistics, Tailoring/Customization, Scheduler/Planner, and Exercise History. These features were mapped to the PSD model, resulting in 9 persuasive features in total: Self-Monitoring, Goal-Setting, Reminder, Simulation (Behavior Modeling), Suggestion, Verbal Persuasion, Reward, Tailoring and Social Support. For example, Self-Monitoring in the PSD model was mapped to Exercise Monitor/Tracker, Exercise Timer, Exercise History and Exercise Summary Statistics in our thematic analysis.

Extending the PSD Model with New Persuasive Features
Our thematic analysis reveals two new persuasive features (Goal-Setting and Verbal Persuasion) can be added to the Primary Task and Dialog Support categories, respectively, to extend the PSD model.

Goal-Setting: In the Primary Task Support category, we added Goal-Setting composed of Goal-Setting itself, Scheduler/Planner and Workout Plan/Routine. Goal-Setting from the perspective of Goal-Setting Theory is the development of an action plan to motivate and guide a person towards reaching a goal. Hence, Goal-Setting as a Primary Task Support feature in fitness apps entails the target goal itself (e.g., 1000 calories a week) and doing scheduling to achieve the action plan. Overall, 11.64% of the individualist participants and 4.48% of the collectivist participants requested the Goal-Setting feature. For example, P195 (a collectivist participant) requested the app provide “Tailored goal setting based on physical index input.” Similarly, P122 (an individualist participant) wanted the app to provide, “A large variety of goals you can track.”

Verbal Persuasion: In the Dialog Support category, we added Verbal Persuasion (comprising Exercise Tips and Motivational Quotes), which the persuasive system provides as feedback from time to time to motivate and encourage users. Proposed by Bandura, Verbal Persuasion increases users’ self-efficacy by making them believe that they can perform a behavior that seems difficult or they failed doing in the past. Regarding Exercise Tips, 3% of participants in each culture requested this feature. For example, P210 (a collectivist participant) requested “Tips on healthy living.” Similarly, P53 (an individualist participant) requested “Exercise tips for specific exercises.” Regarding Motivational Quotes, 3% and 2.1% of the collectivist and individualist participants, respectively, requested this feature. For example, P222 (a collectivist participant) requested “Words of encouragement for progress made.” Similarly, P154 (an individualist participant) requested “Motivational phrases.”

Based on the findings from the thematic analysis, the PSD model can be extended for fitness app design from 28 to 30 features as follows:

Primary Task Support: Reduction, Tunneling, Tailoring, Personalization, Self-Monitoring, Simulation, Rehearsal, Goal-Setting [8 features]

Dialog Support: Praise, Rewards, Reminders, Suggestion, Similarity, Liking, Social Role, Verbal Persuasion [8 features]

Credibility Support: Trustworthiness, Expertise, Surface Credibility, Real-World Feel, Authority, Third-Party Endorsements, Verifiability [7 features]

Social Support: Social Learning, Social Comparison, Normative Influence, Social Facilitation, Cooperation, Competition, Social Recognition [7 features]


Five Most Important Persuasive Features in Each Culture
The five most important features in the individualist culture include the following:
1. Self-Monitoring (55.56%)
2. Goal-Setting (11.64%)
3. Behavior Modeling (8.47%)
4. Suggestion/Recommendation (5.29%)
5. Verbal Persuasion (5.29%)

Moreover, the five most important features in the collectivist culture include the following:
1. Self-Monitoring (29.85%)
2. Reminder/Notification (28.36%)
3. Behavior Modeling (8.96%)
4. Suggestion/Recommendation (8.95%)
5. Verbal Persuasion (5.98%)

The results show that, regardless of culture, users care about Self-Monitoring the most. Other persuasive features they care about include Behavior Modeling (Simulation), Suggestion and Verbal Persuasion.

Culture Difference in Persuasive Feature Preference
Overall, Primary Task Support was the most requested set of persuasive features (76.20% - individualist and 47.77% - collectivist), followed by Dialog Support (16.40% - individualist and 47.77% - collectivist) and Social Support (2.12% - individualist and 2.99% - collectivist). It turns out that Credibility Support features were not requested. One reason for this is that Credibility Support features such as Trustworthiness, Surface Credibility, Real-World feel, etc., are abstract and perceived, unlike Primary Task and Dialog Support features which are concrete and visual. The main difference between the two cultures is that the individualist culture prefers Primary Task Support [e.g., Self-Monitoring (55.56%) and Goal-Setting (11.64%)] more than the collectivist culture [Self-Monitoring (29.85%) and Goal-Setting (4.48%)]. However, the collectivist culture prefers Dialog Support [e.g., Reminder/Notification (28.4%) and Suggestion/Recommendation (8.95%)] more than the individualist culture [Reminder/Notification (3.70%) and Suggestion/Recommendation (5.29%)].