Comparing Cooperative vs Competitive Gamification for Daily Walks

Master Thesis

Project Overview

Walks have many health benefits such as lowering the chances for a stroke, type-2-diabetes, and vitamin D deficiency...but it is not a very engaging activity to do on a daily basis. For my bachelor thesis, I (with the help of my supervisors) investigated how gamified competitive and cooperative feedback influences step count while controlling for individual differences in social value orientation. Although we did not find any significant effects due to low sample sizes, we gained qualitative insights from participant feedback and I learnt how to individually conduct an entire research cycle using an experience sampling method.

My Contribution

This was the first research project that I coordinated by myself and my supervisors were very helpful throughout the process. We had recurrent meetings where they provided me with feedback on the design and implementation. They also provided me with valuable feedback on my written work.

A mobile preview of different gamification elements used for daily walks

Concepts apps for the competition (left), cooperation (middle), and neutral (right) conditions

IDEATE

Background

The concept of gamification is defined as the use of game design elements in non-game contexts. Many gamification designs contain competitive elements such as rewards, high scores, and rankings. But with the resurgence of games that also require cooperation (e.g. League of Legends), it is interesting to examine how competitive gamification elements compare with cooperative elements in promoting physical activity. This relationship between cooperation vs competition elements on physical activity could also be influenced by individual differences in social value orientation. For example, pro-social people might be more likely to enjoy cooperative elements.

Process

Although we used the university database to recruit participants, we did not achieve our target sample size (based on a power analysis). Therefore, we had to hand out flyers around the campus and recruit participants on the spot. The initial idea was to create a mobile application that uses geo-fencing to automatically recognize if a person captures a given location and counts the number of steps taken. Due to time constraints however, we were unable to complete a mobile application and had to switch to a paper prototype where participants received a map along with capture points. Additionally, participants downloaded a commercial mobile application to monitor their daily step count but this was often not accurate. Therefore, every evening we had to manually analyze the data from each participant, calculate the total step count (using google maps) and email participants gamification feedback for the day. Although we did not find a significant effect, throughout the experiment we received feedback from participants that allowed us to gather qualitative data on their user experience.

Research Question & Hypotheses

RQ: The current study aims to compare the effectiveness of using competitive versus cooperative gamification in promoting moderate physical activity and how individual social value orientation moderates the effect.
‍
H1: When participants are placed in the cooperative condition they will walk more on average than when placed in the competitive condition, resulting in a relative increase in moderate physical activity.
H2: Participants who scored either low (competitive) or high (prosocial orientation) in the social value orientation survey will be likely to walk more steps on average in the competitive or cooperative conditions respectively (condition by SVO interaction effect).

DESIGN & ANALYZE

Method

The gamification prototype was developed by setting various capture
points within the TU/e campus and providing each participant with a manual that contained instructions and a map with the capture locations. To measure social value orientation, participants were asked to perform a theoretical resource allocation task before the start of the study. The study lasted for 9 days; 4 days of control and 5 days of intervention. During the study, participants had to capture the locations on the map by taking pictures and sending these locations via email. In the intervention, participants were randomly assigned to receive competitive or cooperative feedback while playing the capture game at the end of each day.

Main Result

The results indicated that although participants generally found the experience enjoyable, there were no statistical differences between the competition and cooperation conditions and there was no statistically significant interaction effect between social value orientation and participants’ assigned condition. Reasons for lack of effect could be because of daily variations in weather, participants combining their walks in groups and the restriction for performing the daily walks on campus and the nature of the social value orientation task.

The participant timeline for my gamfication study

The experimental timeline

Map used to located the various capture points in the tu/e

The capture locations on campus.

IMPACT

The results did not provide any statistical evidence for the initial hypotheses however from participant feedback, there was valuable information regarding how the app can be improved such as including a group goal for the cooperative condition, emphasizing the exploration aspect of walks, and the idea to promote walking in groups through gamification. The current research provides the important insight that properly designing a prototype or mobile application by performing an iterative design cycle that includes user experience tests is not only important for application developers, but also for researchers who want to quantitatively examine gamification effects under controlled conditions.