In 2015, in the Design of Interactive Computational Media course, my team and I worked on a project involving audience engagement at sports events. Through literature reviews, covert observation, questionnaire, and interviews, we gained a better understanding of this problem space. After a brainstorming session, we designed and built a prototype that affords initiation of group activities such as cheers or the wave. Then we performed various usability tests and modified the design of our mobile application accordingly. 

Problem Space

The spectator experience at sports events goes beyond the passive consumption of the sport but is an experience shaped by the active participation of individuals and other fellow spectators [1]. Audience members at a sports event enjoy active, collective crowd activities, such as creating the wave and singing the national anthem [2]. In other words the spectator experience is co-created by the spectator themselves [1]. However, audience members have to passively wait for an opportunity to interact with the crowd. The lack of an established system to facilitate interaction among audience members at a sports event, prevents fans from having the best possible spectator experience. 

Literature review

Designing for Crowds [2]: This paper illustrates the challenges of designing for a crowd. The primary research setting is a football game and the author tackles the problem of treating the crowd as a whole rather than a group of individuals. Currently, the closest interaction mechanism in the market is the big screen and camera interactions between audience and the sport centre. This research aims to understand the behaviour and characteristics of a crowd. Part of the problem at a football game is that the crowd has to deal with long periods of less intense gameplay while waiting for half time to end or while commuting to the stadium. One of the key observations is that audience members enjoy engaging in group activities, such as singing the national anthem together. Moreover, some activities are initiated by an individual and other members of the crowd join the activity, resulting in a group collaboration. Another important observation is that fans will interact with the supporter fans of the opposite team, despite the existence of the "fan supporting team segmentation".

Designing Technology for Active Spectator Experiences at Sporting Events [3]: This paper argues that the spectator experience goes beyond the passive consumption of the sport but one where spectators are active members that collectively shape the experience. Bringing to light the lack of technology designed for spectators of sports despite the active nature of the spectator experience. The researchers created large digital display banners which were placed on the peripherals of the football fields, in the view of the audience. The digital banners (deemed BannerBattle) were used to facilitate active audience engagement such as cheering and flag displays, provided live feedback to the audience and served to facilitate cheering competitions between the home and away-team. The researchers found audience interaction with the BannerBattle displays increased dramatically during less intense times of gameplay, and subsequently decreased during more intense gameplay. The BannerBattle was also shown to actively promote social interaction among spectators. BannerBattle allowed for increased active audience participation at times, by addressing the social and the collective co-creational aspects of the spectator experience.

A Look at Spectator Technology [4]: This paper studies the potential of mobile technology targeting sports spectators. The survey results from 83 college sports fans offer insight as to the expectations and concerns of these fans. The study showed that less than half of respondents (43%) use a location-based checking app mainly for gamification elements such as points and badges, as well as seeking nearby friends. 40% of these respondents said they would be especially likely to check-in if a reward or incentive was offered. In contrast to the incentive approach, the idea of a competition that ranked "fan status" against fellow attendees was viewed negatively. A large number of participants (29%) said that they are unlikely to check-in because of disinterest in using smartphones at a game or resistance to sharing their location. These participants reported that no reward or incentive could overcome their concerns. This suggests that while check-ins can be an impactful tool, publishing the location information should be up to the security preferences of the user.

covert observation

The covert observation was done at the U of T versus Western men’s hockey game, on October 23rd 2015, at Varsity Centre. Two different kinds of covert observations were performed during the same game: event observation and spectator observation.

The event observation consisted of recording every observed collective social activity, such as noise made by audience members after a goal. For each observed event the following data was collected: the event description, time of occurrence, event trigger, estimated duration, number of spectators involved (less than 5, 5-20 or 20+), and the team involved in the event (home, away, or both). In total 17 events were observed during the game. During the spectator covert observation, 20 spectators were observed. For each spectator the following data was collected: dominant colour of outer bottom layer of clothing, presence of supporting sports merchandise (eg. UofT hoodie to show support), technologies used during the game (more specifically cell phone usage), and any additional notes on communication. Supporters of both teams were present in the Varsity Centre, and they belonged to various age groups. In addition, audience members were given noise-makers by the Varsity Club to support cheering. Note that throughout the observation no spectator noticed the observers.


The questionnaire was completed by 8 sports fans who had attended live games in the past. The questionnaire consisted of the following four short multiple choice questions, to ensure that it can be quickly completed by the participants:

1. On average, how many times do you attend a sports event at a stadium or arena in a year?

2. Do you usually use your mobile device during the game?

3. Have you ever posted on social media about the sports event during the game?

4. Would you be willing to provide your seat number to participate in a crowd activity?


For the semi-structured interviews, two volunteers were recruited. Each of the interviews took less than 20 minutes. The two volunteers were sport enthusiasts that followed at least one sport and enjoyed attending games in a stadium. Six main questions were asked during the semi-structured interview:

1. Which sports do you follow?

2. When was the last time you went to a sports game?

3. How often do you attend sports games?

4. During the game do you engage in any group activities? What activities? What motivates you to follow others in a group activity?

5. Do you usually initiate group activities like cheers or the wave? If yes, how?

6. What do you do during less intense times of gameplay?

During the interview process, we encouraged the interviewees to “say more about that” in order to get a more elaborate answer for each question. Depending on the answers they provided us, we also asked additional questions that were not predetermined.

results and discussion

By conducting research we were able to learn more about what problems audience members face, what actions they are willing to take to solve those problems, and what motivates them to take those actions. These problems can be organized in the following categories: disengagement and boredom during less intense times of gameplay, cell phone usage, security concerns, and communication gap between audience members.


The results from the covert observation suggest that spectators feel bored and disengaged due to the lack of stimulation during times of less intense gameplay, and especially during halftime. We observed that spectators use their cell phone applications such as camera and messaging applications, for entertainment during these periods. It was also observed that the spectators were poorly informed of halftime activities organized by the event organizers. For instance, during the puck throwing halftime activity, many spectators were heard asking where other spectators got the pucks, when in fact to get a puck one had to go through a special ticket registration process prior to the game. This shows a void in communication between the spectators and the event organizers. Not only are the spectators disengaged during halftime they also face a communication barrier which prevents them from participating in the existing activities. 

Cell Phone Usage

Through the research process, we learned more about the usage of cell phones at sports events. In particular, the interviews gave more insight into the use of cellphones in the stadium, as one interviewee said:

“I never look at my phone when they are playing ... unless it’s a commercial break ... then I might have taken a photo before and put it on Facebook or something. If it’s during intermission I might ... to see if someone has messaged me”.

This suggests that audience members who do not use their mobile devices during the game, are likely to use them during the commercial breaks or halftime. Despite the fact that the interview results suggest that some spectators may be reluctant to use their cell phone during the game, other research instruments suggest that most audience members are willing to use their cell phones. From the questionnaires, we can conclude that most audience members use their mobile devices during the game, as 75% of the participants said yes to this question. The covert observation also confirms this finding, as 17/20 observed members of the audience used their mobile devices at one point during the game. These spectators mainly used their cell phone for messaging or taking pictures. Cell phone usage was observed during the entirety of the game; however, it was more dominant during less intense times of gameplay. 


One potential problem that fans deal with is privacy; one paper from literature reviews mentions that audience members are reluctant to provide their location for privacy reasons. However, from the questionnaire results it can be concluded that at this point in time, most audience members are willing to provide their seat number, as 100% of the participants confirmed this statement. However, this information should be carefully handled and should not be shared to avoid user privacy intrusion.


We observed that large group activities were triggered by events during the game, such as a good save or a goal. Small group activities took place at all times during the game, where the spectators communicated with friends in their immediate vicinity. From the interview questions, it can be concluded that audience members are interested in crowd activities and are likely to collaborate with other audience members during the game. One interviewee gave more insight into the needs of fans during the game, as he said:

“It would benice to know what people think… it would be nice to get a sense of the sentiment people have toward how the game is going. … I definitely did this during the baseball world series, I was on Twitter almost the entire time watching what people thought of what was happening in the game.”

This statement suggests that the audience members at the stadium are interested in sharing their opinion about the game with one another.


Through literature reviews, covert observation, questionnaire, and interviews we have come to the conclusion that there are many potential points of intervention in this problem space. However, we chose to focus on the issue of lack of engagement, by facilitating crowd activities such as cheers, or the wave. 

Mobile devices provide means for all audience members to be able to collaborate with one another simultaneously. Therefore the problem space can be address using a mobile application which can serve as platform for simultaneous crowd interaction, and the target audience can be narrowed down to ‘sports fans who are interested in engaging in group activities, own a smartphone, and are willing to use their device during a sports event at a stadium or arena’. The stakeholders are all audience members, viewers that are not present at the stadium, players, governing bodies, media, and anyone who provides a service at the stadium such as food vendors. These stakeholders will be affected by the designed solution.


We conducted a 30 minute brainstorming session, where each member sketched around 8 ideas. Then, we shared our ideas with the group and combined them into the following sketch: 



Our final prototype ROAR, is a mobile application that facilitates cheers, songs, movements, and pixel posters at sports events. Users can enter the application by scanning their ticket and selecting the team that they support. The colour scheme of the app will then change to match the colour of the supported team. Users will be asked to follow the instructions for these activities, roughly every 5 minutes and only during less intense periods of the game. We also display each activity on the Jumbotron so that all audience members, even those who do not use the app can participate. After a feedback session, we decided to removed the settings option for notifications, as they can be found in the built-in settings of the user's cell phone. The username selection page was also removed to ensure anonymity.


1. Cheer

Users can write their own slogan, chant, or one-liner for everyone to say simultaneously using this feature. The following video demonstrates the sequence of interactions required for initiating a Cheer and uses a cursor to represent touch:

2. Song

Users can choose a song for everyone to sing simultaneously using this feature. They can also include the lyrics of the song. The following video demonstrates the sequence of interactions required for initiating a Song and uses a cursor to represent touch:

3. Pixel

Users can choose a text with limited number of characters. Then each person’s phone screen will act as a pixel to create a collective poster. The following video demonstrates the sequence of interactions required for initiating a Pixel and uses a cursor to represent touch:

4. Movement

Users can define their action in order to create a simultaneous motion like clapping or an in order motion like the wave. The following video demonstrates the sequence of interactions required for initiating a Movement and uses a cursor to represent touch:

usability testing

Three usability studies were conducted and modifications were made to the final prototype accordingly. These studies involved having each volunteer initiate activities and execute actions that our prototype supported. These actions included: 

1. Voting on a post previously created on the dashboard

2. Initiating a cheer  

3. Creating a pixel poster

The results showed that additional explanations could be used to avoid confusion. As a result, various texts in the app have been modified to make each instruction more clear. In addition, an information icon was added to the "Initiate" page that creates an overlay explaining each activity. For the Pixel activity in particular, we added a new feature that shows the user which pixel of the poster their phone screen corresponds to. 

abstract and references

The abstract in the CHI Extended Abstracts format and the references are attached. 

my team

  • Sky Zhou
  • Elaine Wei
  • Samuel Chien
  • Linda Zhang
  • Parastoo Abtahi