Socially-Driven Persuasive Health Intervention Design: Competition, Social Comparison, and Cooperation

Health Informatics Journal

Socially-Driven Persuasive Health Intervention Design: Competition, Social Comparison, and Cooperation Rita Orji 1, Kiemute Oyibo 2, Richard K. Lomotey 3, Fidelia A. Orji 2 1

Faculty of Computer Science, Dalhousie University, Halifax, NS, Canada Department of Computer Science, University of Saskatchewan, Saskatoon, SK, Canada 3 Department of Information Sciences and Technology (IST), Pennsylvania State University, PA, USA [email protected], [email protected], [email protected], [email protected] 2

ABSTRACT Persuasive technologies (PTs) are tools for motivating behaviour change using persuasive strategies. Socially-driven persuasive technologies employ three common socially-oriented persuasive strategies in many health domains: competition, social comparison, and cooperation. Research has shown the possibilities for socially-driven persuasive interventions to backfire by demotivating behaviour, but we lack knowledge about how the interventions could motivate or demotivate behaviours. To close this gap, we studied 1,898 participants, specifically socially-oriented strategies and their comparative effectiveness in socially-driven persuasive health interventions that motivate healthy behaviour change. The results of a thematic analysis of 278 pages of qualitative data reveal important strengths and weaknesses of the individual socially-oriented strategies that could facilitate or hinder their effectiveness at motivating behaviour change. These include their tendency to simplify behaviours and make them fun, challenge people and make them accountable, give a sense of accomplishment and their tendency to jeopardize user’s privacy and relationships, creates unnecessary tension, and reduce self-confidence and self-esteem, and provoke a health disorder and body shaming respectively. We contribute to the health informatics community by developing 15 design guidelines for operationalizing the strategies in persuasive health intervention to amplify their strengths and overcome their weaknesses. Author Keywords Persuasive technology; health; social influence; persuasive strategy; socially driven intervention; gamified design; competition; cooperation; social comparison; persuasion; strengths and weaknesses; persuasion mechanism. 1. INTRODUCTION There is growing evidence that persuasive technologies are effective tools for motivating behaviour change, influencing people, and promoting learning using various persuasive strategies 1–3. Persuasive technologies (PTs) for health—which are designed as interventions with the primary purpose of changing a user’s behaviour or attitude without using coercion or deception 4,5— have attracted the attention of researchers, physicians, and health practitioners as a novel approach for motivating desirable health behaviour using various persuasive strategies. Research has shown that persuasive technologies can be strategically designed to motivate desirable behaviour change; for example, to help people overcome addictive behaviours such as substance abuse 6–8, to promote personal wellness, to manage diseases, to engage in preventive behaviours 5,9,10, and to avoid risky behaviours 6,7,11,12. A fundamental feature of persuasive technological intervention is persuasion, an attempt to influence or reinforce behaviours, attitudes, feelings, or thoughts 4,13. Persuasion can be achieved using persuasive strategies. Socially-oriented persuasive strategies (competition, social comparison, and cooperation) are the most widely and frequently employed strategies both in persuasive health technologies and other online support interventions because of their ability to leverage social influence to motivate behaviour change. Socially-driven persuasive technologies are persuasive technologies that employ the sociallyoriented strategies, competition, social comparison, and cooperation to motivate behaviour change. Studies have shown that 1 Orji R., Oyibo K., Lomotey R., Orji F., (2018). Socially-driven persuasive health intervention design: Competition, social comparison, and cooperation. Health Informatics Journal. DOI/pdf/10.1177/1460458218766570.

Health Informatics Journal these socially-driven persuasive technologies can be effective for motivating health behaviour change. As a result, researchers have applied them for motivating behaviour change across several health domains, including binge-drinking prevention 6–8, drug-and-substance-use prevention, HIV and STDs prevention and treatment 14, smoking cessation 15,16, being physically active 17,18 and healthy eating 19–21. They have also been applied in the area of disease management, including in teaching skills for self-management of type 1 diabetes 22 and asthma 23. Despite the wide application of the socially-driven persuasive technologies across many persuasive health domains as an intervention, there have been mixed findings regarding their effectiveness at motivating the desired behaviour change. Although there are many stories of successful socially-driven persuasive interventions with respect to their effectiveness at motivating healthy behaviour change 19,20,24–26, research has also shown that using socially-driven persuasive intervention as a tool to motivate behaviour change could be counterproductive 20,27–30. For example, Palinkas et al. 31 in their evaluation of a sociallydriven intervention for primary and secondary prevention of drug and alcohol use among multiethnic female adolescents found that the intervention was largely ineffective and mainly counterproductive as the prevalence of alcohol and drug use increased significantly over the intervention period. There is still a gap in knowledge on how and why the socially-oriented strategies (competition, social comparison, and cooperation which are the driving force of the socially-driven persuasive interventions) could influence behaviour either positively or negatively and the mechanisms through which they motivate or demotivate behaviours. This is essential for effective operationalization of the strategies in intervention design and the success of socially-oriented persuasive health intervention in general. To advance research in this area, we conducted two large-scale studies of 1108 and 790 (a total of 1898) participants to investigate the strengths and weaknesses of the three socially-oriented strategies (competition, comparison, and cooperation) that are widely used in persuasive health interventions design. We investigated these strategies in the context of PTs for promoting healthy eating behaviour (study one) and PTs for motivating change in risky health behaviours such as binge drinking (study two). Investigating two different health domains allows us to uncover a wide range of strengths and weaknesses that could be generalized across other domains. As a secondary objective, we validated the persuasiveness of the strategies and showed their comparative persuasiveness with respect to their ability to motivate healthy behaviour change. We used prototype persuasive implementation of the individual strategies that has been validated in other studies 13,32. The results of a thematic analysis of 278 pages of qualitative data reveal important strengths and weaknesses of individual strategies that could facilitate or hinder their effectiveness at motivating desirable behaviour changes. Specifically, we found that, because each of the socially-oriented strategies harnesses the principle and the power of social influence to motivate behaviour change, they share some common strengths (such as their tendency to make behaviours fun, exciting, interesting, and appear easier to do than usual) and weaknesses (such as their tendency to be invasive and jeopardize an individual’s privacy and relationships). However, because each of the strategies differs in their operationalization of the social influence principle, some strengths and weaknesses are unique to the individual strategies or subsets of the strategies. Some of the strengths include that: Competition gives a sense of accomplishment; Comparison allows for subtle and empowering peer pressure; and Cooperation provides opportunities for mutual support, group encouragement, and reinforcement and offers opportunities to collaborate, make and interact with friends. On the other hand, some of their weaknesses include that: Competition could trivialize the importance of healthy behaviour and annoy and discourage people; Comparison and Competition could encourage body shaming and health disorder (e.g., eating disorder) and reduce self-esteem; Cooperation and Competition could cause unnecessary stress, tension, pressure and make people anxious. This knowledge of the strengths and weaknesses of the strategies will not only inform the choice of the appropriate strategies to employ in a particular PT design but will also inform the manner in which the strategies will be operationalized in PT designs. The operationalization of the strategies can amplify their strengths or their weaknesses. Our findings also reveal that the three socially-oriented strategies differ significantly with respect to their overall persuasiveness, with the social comparison being the most persuasive of the strategies. 2 Orji R., Oyibo K., Lomotey R., Orji F., (2018). Socially-driven persuasive health intervention design: Competition, social comparison, and cooperation. Health Informatics Journal. DOI/pdf/10.1177/1460458218766570.

Health Informatics Journal Our findings advance existing knowledge in the area of persuasive health intervention design by offering six main contributions: First, we provide insights into the mechanism through which socially-driven persuasive health interventions could motivate or demotivate health behaviours by revealing the strengths and weaknesses of the individual strategies that should be taken into account by PT for health designers when employing each of the strategies. Second, we provide a comparative analysis of the mechanism through which the three strategies could influence behaviour positively and negatively by comparing and mapping the strengths and weaknesses of individual strategies. Third, based on our findings, we offer 15 design guidelines on how to operationalize/implement the strategies in persuasive health intervention design to amplify their strengths and overcome their weaknesses. Fifth, we validate and compare the perceived effectiveness of individual socially-oriented strategies and show that they differ significantly in their overall persuasiveness via a large-scale study. Finally, we provide an extensive review of existing socially-driven persuasive health interventions (classifying them into interventions for health promotion and prevention and interventions for disease management). We also deconstruct how existing interventions operationalized the three socially-oriented strategies. These contributions build on and extend our previous work 33 where we only reported a few of the strengths and weaknesses and six design recommendations based on a preliminary analysis. Besides, the current study is based on a larger dataset of 1898 participants and an in-depth analysis of 278 pages of qualitative data which allows us to uncover more details, provide more insights, 15 design guidelines, and extensive discussions of design implication for persuasive health intervention by situating them within the context of existing knowledge. 2. RELATED WORK In this section, we present a brief overview of socially-oriented persuasive strategies and socially-driven persuasive health interventions employing the strategies. 2.1 Socially-oriented Persuasive Strategies A fundamental feature of persuasive technological intervention is persuasion, an attempt to influence or reinforce behaviours, attitudes, feelings, or thoughts 4,13. Persuasion is often achieved using various persuasive strategies. Persuasive strategies are approaches that can be used in persuasive technological intervention design to promote desirable behaviours or attitudes 32. Over the past few decades, a number of persuasive strategies have been developed. For example, Fogg 4 developed seven persuasive strategies, and Oinas-Kukkonen and Harjumaa 34 built on Fogg’s strategies to develop 28 persuasive system design strategies. Among all the strategies, socially-oriented persuasive strategies (competition, social comparison, and cooperation adopted from Oinas-Kukkonen and Fogg 4,34), have been widely employed in persuasive technological intervention for health designs and other online health support systems 3,32,35. The major distinction between socially-oriented strategies and other persuasive strategies is their ability to leverage the power of social influence to motivate behaviour change 34. Social influence is a term used to explain the effect that other people have on us; our behaviours, beliefs, and attitudes 36. Social influence occurs when an individual’s behaviour or attitude is affected by others. Almost all our behaviours can be shaped by the power of social influence. According to Berger 37, “without our realizing it, other people’s behaviour has a huge influence on everything we do at every moment of our lives, from the mundane to the momentous occasion. Even strangers have a startling impact on our judgments and decisions.” As a result, research has applied the principle of social influence to influence behaviours in many domains including marketing 38, education 39, transportation 40, environmental sustainability 41, energy and water conservation 42,43, and of course health 17,18. For a typical example of an application of social influence principle in the marketing domain, see Amazon.com 38. Amazon.com employed the power of social influence in their eCommerce website to motivate customers to purchase products and hence increase sales using the customers who bought this also bought that concept and the product rating tactics. Similarly, in the area of energy conservation, Gustafsson et al. 42 described a mobile application called Power Explorer which allows teenagers to compete with their peers on who used the less energy as a way of saving energy. With respect to education, Christy and Fox 39 investigated the impact of social influence on students’ educational performances using a leaderboard which allows the students to view and compare their performance with that of others. In the 3 Orji R., Oyibo K., Lomotey R., Orji F., (2018). Socially-driven persuasive health intervention design: Competition, social comparison, and cooperation. Health Informatics Journal. DOI/pdf/10.1177/1460458218766570.

Health Informatics Journal health domain, Tosco et al. 17 developed a mobile health application that leverages the power of social influence to motivate teenage girls to exercise by comparing their activity levels. This is in contrast with the Fish ‘n’ Step 18 which employed the self-monitoring strategy to motivate physical activity by allowing users to track their own behaviours. Considering this increasing application of social influence principle to affect individual’s behaviours in various domains, recent research efforts have focused on developing systematic approaches for operationalizing the principle in various application domains. In the field of persuasive technologies, Oinas-kukkonen 44 proposed three socially-oriented strategies—competition, social comparison, and cooperation. - Competition strategy provides opportunities for users to compete with one another while performing the behaviour as a way of motivating the desired behaviour. - Social Comparison strategy offers users the opportunity to view and compare their behaviour performance data with that of other user(s). However, comparison strategy does not involve winning or losing, and it does not involve an obvious and direct interaction with others (peer groups), differentiating it from the competition.

- Cooperation strategy offers an opportunity for people to collaborate (work together) to achieve a shared behaviour objective and applaud them for achieving their goals together. For a detailed discussion of these strategies, see 34. Among all the domains of applications, the use of socially-driven persuasive technologies (which employ these strategies) in the health domain have received special attention 45,46, likely due to the importance of maintaining good health and wellness. Therefore, this study will focus on socially-driven persuasive technologies for health with special emphasis on socially-driven persuasive technologies for promoting healthy eating and socially-driven persuasive technologies for motivating change in risky health behaviour (risky alcohol behaviour change). 2.2 Socially-driven Persuasive Health Interventions. Persuasive technologies aim to bring about desirable change in attitude and/or behaviour without using coercion or deception 4 . Socially-driven persuasive technologies are persuasive technologies that employ the power of social influence (via the three socially-oriented strategies: competition, comparison, and cooperation) to motivate behaviour change. Studies have shown that these socially-driven persuasive technologies can be effective for motivating desirable behaviour change 15–21,47. As a result, researchers have employed these strategies in persuasive health technologies for motivating behaviour change in many health domains, including binge drinking prevention 6–8, drug and substance use prevention 6,31,48, HIV and STD prevention and treatment 14, smoking cessation 15,16, being physically active 17,18 and healthy eating 19–21,49. In general, socially-driven persuasive health technologies can broadly be categorized into two main areas: socially-driven persuasive health technologies for health promotion and prevention and socially-driven persuasive health technologies for disease management. 2.2.1 Socially-driven Persuasive Health Interventions for Health Promotion and Prevention Preventive health behaviours are behaviours that are undertaken to prevent sicknesses, detect early signs of sickness, and maintain general health and wellbeing 50. Examples include smoking cessation 15,16, being physically active 17,18, healthy eating 19–21 , binge drinking prevention 6–8. Several socially-driven persuasive technologies have been developed for health promotion and prevention. For example, National Mindless Eating Challenge (NMEC) is a mobile phone-based persuasive health intervention aimed at promoting healthy eating behaviour 21. NMEC employs the social comparison strategy to motivate behaviour change. NMEC users are required to care for virtual pets or plants and that entails them following some healthy eating recommendations. At the beginning of the application, users are assigned tasks that are relevant to their healthy eating goals and are given the 4 Orji R., Oyibo K., Lomotey R., Orji F., (2018). Socially-driven persuasive health intervention design: Competition, social comparison, and cooperation. Health Informatics Journal. DOI/pdf/10.1177/1460458218766570.

Health Informatics Journal opportunity to view and compare their performance with the performance of others (social comparison). Similarly, LunchTime is a persuasive application for motivating healthy eating 19. LunchTime employs the competition, and comparison strategies to motivate behaviour change. Users assume the roles of restaurant customers, and their goals are to choose the healthiest food from a list of food choices. Each user is allowed to compare his/her performance with that of other users (competition and comparison). Finally, RightWay Café is a game-based persuasive health intervention that employs competition to promote healthy eating and physical activity 51. At the beginning of the game, the users create a personified avatar using their own personal characteristics, such as height, gender, body frame, name, physical activity, weight, and age. Users are tasked with managing the daily activity of their avatar (such as calorie consumption and physical activity) to enable them to reach a healthy weight. The user who is most successful at managing his/her avatar’s daily diet and physical activity in a healthy way wins (competition). Typical examples of socially-driven persuasive intervention in the area of physical activity are Neat-o-Games 52, Phone Row 53 , Fish ‘n’ Step 54, and Chick Clique 17. Neat-o-Games 52 is a persuasive game-based intervention that employed competition and comparison to promote physical activity. The virtual racing game requires users to race with other users in their mobile network. The user’s physical activity (monitored using wearable accelerometers) is used as input to control the speed of the his/her avatar in the race. At the end of every day, players’ activity points are compared and winners are announced (competition and comparison). Similarly, Neat-o-Games, Phone Row 53 employed the competition and comparison to motivate moderate intensity physical activity by requiring users to control the movement of a smartphone racing boat using their body movement. Users compete and compare their performance with the performance of other users (competition and comparison). Fish ‘n’ Step 54, employs competition and cooperation to promote physical activity. A user’s daily step count is associated with the growth and flourishing of a virtual fish in a tank. A user’s fish tank includes other users’ fish, thereby fostering both cooperation and competition. Users could compete with one another (competition) as an individual or as part of a team (cooperation) and are provided feedback regarding their calories burned, personal progress, and ranking. Chick Clique 17 is a persuasive mobile health application that leverages competition and comparison strategies to motivate teenage girls to exercise. Chick Clique allows up to four friends to engage in a friendly competition where the group’s walking statistics are tracked, ranked (competition), and compared with the other members (comparison). Several persuasive technologies have also employed the socially-oriented strategies highlighted above to motivate risky health behaviour change. For example, N-Squad Web Adventure 8 is a socially-driven persuasive technology which employed the cooperation to impart knowledge about the consequences of alcohol consumption and hence discourage risky drinking behaviours. The application simulated the impact of unhealthy alcohol intake on an individual’s body including the circulatory, digestive, and nervous systems. It requires that users work as part of a group to resolve some challenges about risky drinking behaviours embedded in the application (cooperation). Similarly, Thinking Not Drinking: A SODAS City Adventure 48 is a game-based persuasive intervention for preventing unhealthy alcohol use. Thinking Not Drinking employs the social comparison to motivate behaviour change. Each game session begins with skill-specific goals that a user must accomplish. The game trains players on how to overcome peer pressure (social comparison) and avoid alcohol abuse. For a review of applications of persuasive strategies to motivate changes in alcohol and other risky health behaviour, see Lehto and OinasKukkonen 35. 2.2.2 Socially-driven Persuasive Health Interventions for Disease management Persuasive Technologies have also been employed to impart disease-specific knowledge and self-management skills on patients. These include educating patients on how to manage certain illnesses, helping them conform to treatment directives by providing relevant health information, providing timely reminders to the patients, simulating health behaviour, and providing opportunities for users to rehearse health behaviours that are related to their specific health illness 32,55. Persuasive interventions for disease management are mainly targeted at people who identified themselves ill and the overarching objective is to help them manage their illness or get well with the aid of the PTs. For example, Packy and Marlon is a persuasive health intervention that helps children and teenagers self-manage their type 1 diabetes. Users are tasked with keeping their virtual characters’ diabetes under control by monitoring the avatar’s vital body 5 Orji R., Oyibo K., Lomotey R., Orji F., (2018). Socially-driven persuasive health intervention design: Competition, social comparison, and cooperation. Health Informatics Journal. DOI/pdf/10.1177/1460458218766570.

Health Informatics Journal signs, administering insulin, and other drugs as may be necessary 22. Packy and Marlon simulates various real-life diabetes challenges. To win, each one of the two users—Packy and Marlon must effectively manage their characters’ diabetes and insulin intake; therefore, they must support each other (cooperation). Similarly, Bronki the Bronchiasaurus is a persuasive health intervention aimed at imparting asthmatic children with asthma self-management skills 23. The persuasive application simulates good and bad real-world asthma self-management skill to impart self-monitoring skills to kids with asthma. The application presents two virtual characters (Bronkie and Trakie), and users are required to help the characters control their asthma. To achieve this, users have to avoid triggers such as dust and smoke that have the tendency of impacting their asthma negatively while they go on their mission. They also need to measure and monitor breath strength, take medications and use the inhaler as required. The user’s health decisions are determined by the character’s health outcome and a good health outcome is required to win (competition). This review is by no means exhaustive; however, it is a good representation of common practice in the area of socially-driven persuasive health technologies design. This review shows that the socially-oriented strategies have been widely employed in PTs to motivate behaviour change across various health domains. The evaluated persuasive health interventions reported varying degrees of success at achieving the intended health objectives and mixed findings with respect to the effectiveness of the socially-oriented strategies 18,21,30,32. However, there is a lack of research into the mechanism through which the strategies could motivate or demotivate behaviours; why they may work in one context and fail in another. Revealing the strengths and weaknesses of the individual strategies would shed light on the mechanism through which these socially-oriented strategies could motivate or demotivate behaviours and suggest ways of operationalizing them to reduce their weaknesses and amplify their strengths. It will also shed light on why many persuasive interventions record varying degrees of success, mixed findings, and even failures 28. Again, because persuasive socially-oriented strategies are often employed en masse, there is little knowledge on their comparative efficacy. 3. STUDY DESIGN AND METHOD In this section, we focus on the purpose of our study, the study instruments and data collection method which were presented in a previous study 33. 3.1 Purpose In our study, we set out to address two important issues in persuasive health intervention research, which include: (1) investigate the perceived persuasiveness of three socially-oriented persuasive strategies (competition, comparison and cooperation) with respect to their ability to motivate changes in healthy behaviours; and (2) investigate the mechanism through which these three strategies could motivate or demotivate the adoption of healthy behaviours by examining their strengths and weaknesses. Specifically, we focus on two common application domains of persuasive health interventions to ensure uniformity and generalizability. They include: (1) persuasive technologies for encouraging healthy eating behaviour; and (2) persuasive technologies for motivating change of risky healthy behaviour (risky alcohol behaviour change). 3.2 Measurement Instrument To achieve the aims of our study, we conducted two separate empirical studies. The first study focuses on persuasive technologies for motivating healthy eating behaviours, while the second study focuses on persuasive technologies for motivating a change of risky alcohol behaviours. To gather the required data for both empirical studies, we used prototype persuasive technology implementation of each of the socially-oriented strategies. The study instruments have been a validated in previous studies 13,32. Specifically, in each study, we represented each of the three socially-oriented strategies in different storyboards. In study one, the storyboards focus on PTs for encouraging healthy eating, while, in study two, the storyboards focus on PTs for promoting change of risky alcohol behaviours. All of the storyboards were designed by an artist following the recommended storyboard design guidelines proposed by Truong et al. 56. Storyboards, in general, provide a common visual language, which individuals from different socio-cultural backgrounds can easily read and understand 57. Moreover, in previous research 32,58,59, they have been found to be an effective way of evaluating and depicting persuasive strategies to users of a persuasive intervention to elicit the right responses, which will eventually inform the intervention design. In our studies, the storyboards portray a character, which represents the user (or a study participant), and his/her interactions with simulated PTs 6 Orji R., Oyibo K., Lomotey R., Orji F., (2018). Socially-driven persuasive health intervention design: Competition, social comparison, and cooperation. Health Informatics Journal. DOI/pdf/10.1177/1460458218766570.

Health Informatics Journal for motivating the respective health behaviour change – illustrating the individual socially-oriented strategies. For example, in Figure 1, the user is the character (in blue dress and red hair) interacting with the computer. In designing the study instruments for both of our studies, a number of steps were followed. First, we evaluated and iteratively refined the storyboards by taking into consideration recommendations from domain experts, with whom we interacted. Upon completing the expert evaluations, we proceeded to carry out the initial online user evaluations with the sole purpose of ensuring that our storyboards accurately depicted the respective strategies. Figure 1 shows an example of a storyboard which illustrates the competition strategy in the context of PT for motivating healthy eating, while Figure 2 shows competition in the context of risky alcohol behaviour change. Second, following each storyboard, we presented to participants a validated scale for assessing the perceived persuasiveness of each socially-oriented persuasive strategy to elicit quantitative feedback on its effectiveness. The scale was adapted from Orji et al. 32 and has been used previously by a number of studies 13,60,61. Specifically, we asked participants the following questions: “Imagine that you are using the system presented in the storyboard above to track your daily eating (or alcohol intake in study 2), on a scale of 1 to 7 (1-Strongly disagree to 7-Strongly agree), to what extent do you agree with the following statements: a) The system would influence me. b) The system would be convincing. c) The system would be personally relevant for me. d) The system would make me reconsider my eating (or alcohol drinking) habits.” Third, each strategy was followed with an open-ended question. The open-ended questions allow participants to provide qualitative feedback (comments) on the strategy represented in each storyboard and how they would use the persuasive system. Furthermore, the comments allowed them to justify their ratings of each strategy with respect to its effectiveness (i.e., strengths and weaknesses). Prior to assessing the persuasiveness of each strategy, we ensured that the participants understood the strategy illustrated in the storyboard. We achieved this by asking them two comprehension questions. In the first question, we asked the participants to identify the illustrated strategy on the storyboard from a list of strategies (“What strategy does this storyboard represent”). In the second question, we asked them to describe what is happening in the storyboard in their own words (“In your own words, please describe what is happening in this storyboard”). Responses from participants who answered both comprehension questions incorrectly were discarded. Together with the responses from participants who gave correct answers to the two comprehension questions, we also retained responses from participants who answered one of the comprehension questions correctly. Moreover, we asked them questions about their eating and drinking behaviours as well as demographic information. 3.3 Data Collection Study one was approved by the University of Saskatchewan’s ethics board while study two was approved by the University of Waterloo. Afterwards, we recruited participants for the studies using Amazon Mechanical Turk (AMT): an online recruitment platform, which allows access to a global audience at a relatively low cost, and ensures efficient survey distribution, and reliable results 62,63. We used SurveyMonkey to design our questionnaires. However, to eliminate possible bias due to a fixed ordering of the storyboards, we used the page randomization functionality provided by SurveyMonkey. This randomized the positions of the storyboards (and their associated questions) by varying their ordering for each participant. Prior to conducting the main study, we carried out two pilot studies. The first pilot study (which focused on healthy eating) consisted of 30 participants (15 were recruited from AMT and 15 from a university in Canada). The second pilot study (which focused on binge drinking behaviour change) comprised 20 participants (10 were recruited from AMT and 10 from a university in Canada). The preliminary evaluation of the pilot studies confirmed the suitability and understandability of our study instruments.

7 Orji R., Oyibo K., Lomotey R., Orji F., (2018). Socially-driven persuasive health intervention design: Competition, social comparison, and cooperation. Health Informatics Journal. DOI/pdf/10.1177/1460458218766570.

Health Informatics Journal

Figure 1: Storyboard illustrating the Competition persuasive strategies in the context of persuasive technology for motivating healthy eating behaviour – adapted from 32

Make me slim, Emily’s BAC Eat Less! is 0.02% today

If you drink less than Emily, you’ll win 400 bonus points!

Your BAC today is 0.01. Congrats! You won 400 bonus points! Keep it up!!

Figure 2: Storyboard illustrating the Competition persuasive strategies in the context of persuasive technology for motivating risky alcohol behaviour change – adapted from 13

Table 1: Participants’ demographic information Total Participants = 1,898 Gender

Females (48%), Males (52%).

Age

18-25 (31%), 26-35 (39%), 36-45 (18%), Over 45 (12%).

Education

Less than high school (1%), High school (30%), College diploma (13%), Bachelor’s degree (38%), Master’s degree (16%), Doctorate degree (1%), Others (1%). 8

Orji R., Oyibo K., Lomotey R., Orji F., (2018). Socially-driven persuasive health intervention design: Competition, social comparison, and cooperation. Health Informatics Journal. DOI/pdf/10.1177/1460458218766570.

Health Informatics Journal 3.4 Participants’ Demographic Information A total of 1898 valid responses were included in our analysis after filtering out incomplete responses and incorrect responses to comprehension and attention-determining questions. 1108 of these responses came from study one and 790 responses from study two. In appreciation of their time, participants received a $2 USD compensation. Moreover, our participants were at least 18 years of age as at the time of data collection; they also read and understood English well. Finally, specifically for study two, participants were those who consumed alcohol often or had consumed alcohol at least once. In general, we have a relatively diverse population in terms of gender, age, education level attained (see Table 1). Our sample in each study was relatively diverse, as there were participants from different continents and countries, such as the United States of America, India, Canada, Mexico, and other countries of the world. 4. DATA ANALYSIS The two main aims of this paper are: One, to examine the persuasiveness of the three socially-oriented strategies that are commonly employed in persuasive health intervention design. Two, to investigate the mechanism through which these strategies could motivate or demotivate healthy behaviour by investigating the strengths and weaknesses of the individual strategies. To achieve this, we used several well-known analytical tools and procedures. We summarize the various steps taken to analyze our data in this section. 1. We validated that our storyboards accurately depicted the intended socially-oriented strategy by running the chi-squared test 59. 2. We ascertained the appropriateness of our data for analysis using the Kaiser-Meyer-Olkin (KMO) sampling adequacies and the Bartlett Test of Sphericity 64. 3. To evaluate and compare the persuasiveness of the socially-oriented strategies, we calculated the average score for each strategy and conducted a Repeated-Measure ANOVA (RM-ANOVA) followed by pairwise comparison. The analysis was conducted after validating for the ANOVA assumptions. 4. We also used notched boxplot to visualize the persuasiveness of the strategies (see Figure 3). 5. To tease out the strengths and weaknesses of the individual socially-oriented persuasive strategies, we conducted a thematic analysis 65 of 278 pages of qualitative participants’ comments from our participants about the individual strategies. 4.1 Storyboard Validation To ensure that participants understood the socially-oriented strategy illustrated in each of the storyboards, we ran chi-squared tests on the participants’ responses to the multiple-choice questions that asked them to identify the strategy represented in each of the storyboards. The results for all of the three socially-oriented strategies were significant at p