Supervisor: Frank Nack
Date: 18
thof July 2014
2
ndExaminer: Daniel Buzzo
Use of Smartphone Devices to Stimulate
Social Interactions
Konstantinos Chronopoulos (10629904)
Thesis Master Information Science – Human Centered Multimedia
University of Amsterdam Faculty of Science
Use of Smartphone Devices to Stimulate Social
Interactions
Konstantinos Chronopoulos
University of Amsterdam Science Park 904 1098 XH Amsterdam, Netherlandskon.chrono@gmail.com
ABSTRACT
In this project we investigate whether face-to-face inter-actions can be stimulated by smartphone devices. We make an experiment to find if it is possible to change humans’ tendency to make less effort and dedicate less time to engage in social interactions which is claimed to be one of the results of inappropriate use of tech-nology (Brey, 2006). An exploratory questionnaire was made in order to examine the ways that students (21-31 years old) are using information and entertainment technology and how they feel about it. We designed and developed an Android application that presents to the users their friends’ availability and reminds them to check it once a day. Five subjects were used to try this application for 17 days. The data from the logfiles and the interviews, which were conducted after the tri-als, indicate that a combination of a simple application design along with the proper incentives could possibly lead to that behaviour change. The results were encour-aging; however, more extended trials and experiments should take place to draw decisive conclusions.
Keywords
behaviour change, smartphones, social interactions
1.
INTRODUCTION
Technology has changed our habits (Oulasvirta, Rat-tenbury, Ma, & Raita, 2012) and as a result we rely more and more on technology based solutions. Many believe that we are adapting to machines rather than having machines adapt to us (Pribbenow, 1999). It is claimed that there are positive aspects of technol-ogy on entertainment and communication (Brey, 2006). Access to information has become easier for many peo-ple. We are able to hear news and communicate with people from the other side of the world instantly, inex-pensively and reliably regardless of time and place. The internet facilitates the development of social rela-tionships and their maintenance (Barkhuus & Tashiro, 2010). It gives us the opportunity to meet people out-side our close vicinity and interact with new forms of leisure and entertainment (video games, sharing
con-tent like hobbies, pictures, music, etc.). Many social networking applications, that try to connect users ac-cording to their shared interests and keep them in con-tact, have been developed. People are used to them and have incorporated them in the daily routine.
It has frequently been argued that even though tech-nology has improved our lives on a great scale, it has also affected us in a negative way. The invasion of in-formation and entertainment technology in human life has made people overly dependent on it (Pribbenow, 1999). It has been claimed that people feel comfortable and safe when using technology in e.g. their home en-vironment causing ignorance of certain aspects of life. For example, some of us would prefer to spend our time watching alone a movie on television instead of going to the cinema and watch it with friends or we would choose to stay at home and play sports video games rather than go out and watch a basketball game or even play one together with friends.
The convenience of using internet-based technology has also changed our communication behaviour. It has been argued that forming relationships with people we meet in the digital world is easier but it has its costs since no face-to-face interaction takes place causing anti-social effects in the analogue world (Brey, 2006). For instance, it is claimed that today’s children prefer to spend time on social media than with their friends and families. The overload of information and the use of technol-ogy on a regular basis can cause anxiety and difficul-ties in memorising (Brey, 2006). Humans are not like machines; they need time to sort out the day’s events and relax, otherwise it is possible to suffer from tech-nostress (Pribbenow, 1999). We set high expectations from the machines. Pribbenow (1999) addresses that we count on machines to do so much that we do not know what to do when something goes wrong with our technology.
There are arguments which suggest that people need to get socialised by engaging into face-to-face interaction (more expressive and authentic than the online one) and
become physically social active. However, it is consid-ered that it is not the technology itself that causes all these problems but the kind of use that people make of it. Undoubtedly, it has simplified our lives but we have to think wisely and use it sparingly.
2.
RELATED WORK
The change of human behaviour is a field that interests both researchers and industries. There are several tech-niques, methods and domains of applications that can be used in order to accomplish behaviour change. Wiafe and Nakata (2012) studied papers published in the In-ternational Conference on Persuasive Technology from 2006 to 2010, reviewed them and collected all the tech-niques that have been used and the methods that were applied in the different domains. They address the ten most used persuasive techniques in applications, such as feedback, self-monitoring, suggestion, reward, etc. Fi-nally, they suggest that there is the need to establish a clear understanding of terms and vocabulary used in the domain because of the ad hoc definitions that designers use for techniques or methods.
Pribbenow (1999) in his paper focuses on the stress which occurs when people spend too much time work-ing with machines and their minds try to think and act like machines. This is called technostress and it is usual to people who use computers and other machines on a daily basis. Technology is supposed to save time and make jobs easier. However, people feel that they cannot function properly without it (Pribbenow, 1999). Continuous and inappropriate use can cause physiolog-ical problems, computerphobia, computer stress, anxi-ety, cyberphobia, technology aversion etc. In the article even though they do not think that it is possible to elim-inate technostress they present ways of reducing or even preventing it. They conclude that the key is to identify the cause of it and then find the way to limit it. Shin and Dey (2013) detect problematic use of another technological achievement, the smartphone. They have developed a way in which they automatically take usage data from smartphones and by using machine learning algorithms they detect problematic use. Characteris-tics such as the ratio of SMSs to calls and the number of apps used per day were useful enough for the reliable detection of problematic usage. Since the accuracy of their results was 89.6% then their work is a promising step towards being able to automatically detect prob-lematic behaviour when using technology.
Oulasvirta et al. (2012) state that the use of smart-phones can form habits and more specifically the “check-ing habit” which is a brief, repetitive inspection of dy-namic content quickly accessible on the mobile device. Even though they found that in total the checking habits increased the usage of the device and that this behaviour is frequent, they found that it is experienced more as
an annoyance than an addiction.
Applications such as “AppDetox”1have been developed in order to avoid technology addiction and more specif-ically the mobile applications (L¨ochtefeld, B¨ohmer, & Ganev, 2013). All smartphones provide several services that people are using constantly, which results in chang-ing their habits and consequently they become addicted to them. The application they built gives the chance to the users to consciously create rules that will keep them from using certain apps. By monitoring the usage of the app they found that users are strict when creating the restrictive rules but they mostly suppress the use of messaging and social networking apps.
On the other hand, other research (Barkhuus & Tashiro, 2010) has shown that technological achievements, such as internet and consequently the online social networks, have a positive impact on people’s social lives. Specifi-cally, they focus on offline socialising structures around an online social network (Facebook) and on how this can ease students’ social lives. They present Facebook as a useful tool for students since they can initiate and man-age gatherings. They conclude that social networks are a powerful tool for encouraging peripheral friendships and they support that these are just another means of communication that connects with real life as a normal continuation of communication.
Many argue that one negative effect of technology and internet on people’s lives is that they place barriers between people and face-to-face communication (Brey, 2006). People take less time and effort to engage in live interactions. It is claimed that they are becoming inac-tive and prefer to stay at home and play video games instead of getting together with friends.
The fact that children are very active on the inter-net and the extensive use of online social inter-networks has raised social concerns. Abeyrathne et al. (2011) built a robotic toy for children in order to eliminate the lack of actual physical communication among them and their friends. They also developed a 3D virtual world to provide collaborative interaction and enjoyment with friends. In general, by developing this system they tried to accomplish (through the interaction) strong relation-ships and friendrelation-ships. It does not restrict online activ-ities and social networking but creates a friendly envi-ronment for the child in which their parents can partic-ipate too. They found that it encourages face to face communications, which are crucial in the appropriate development of the children in this age, and that are then extended to the online space.
3.
RESEARCH QUESTION
As we saw in the related work, poor decisions about technology use can cause problematic human behaviour
1
i.e. technostress, socialisation issues, etc. It is argued that our lives are filled with several forms of technology, such as television, laptops, internet, telephones and hu-mans are even more exposed and vulnerable to the mis-use of technology. This can camis-use problems in their so-cialisation process and prevent them from meeting their friends and participating in social activities. Through this project we try to change this behaviour by focusing on the socialisation part of the problem and the research question we want to answer is:
Can smartphone devices stimulate analog social inter-actions?
We focus on developing a simple-designed application for smartphone devices and adding the appropriate in-centives for the users.
4.
METHODOLOGY
Our research focused on university students of similar age and education level.
4.1
Exploratory Questionnaire
At first, we made an exploratory questionnaire to gather information about the kinds of information or enter-tainment technology the students are using while at home and the incentives that would make them to move away from them and participate in social activities with their friends. So, the electronic questionnaire (see ap-pendix A) was given to ten university students between the age of 21 and 31 (6 males, 4 females). Four of the questions were more general and helped us to get a clearer idea about how people of this age are using technology and how they feel about it. The remaining three were more practical, asking about the amount of time they spend on each technology (smartphone, lap-top/PC, tablet, etc.) and about the stimuli that would make them to move away.
4.2
Application development
To test our research question we developed a smart-phone application and we focused on finding ways that it could stimulate social interactions. Research was made on mobile app stores in order to find similar appli-cations and get ideas for the design and the development of our own.
During our application’s design process the general idea was to keep it as simple as possible so that the users will spend the least possible time on the application and consequently on their smartphone. We tried to use a direct and clear way that the information is pre-sented on the application, without requesting from the users useless interactions. Additionally, the application was sending daily notifications to the users, reminding them to check the application through which they would accomplish a face-to-face social interaction by meeting their friends. These were the two main ideas that would contribute in changing the users’ behaviour.
4.3
Trials
After finishing with the development of the application there was a trial period of 17 days in which the partici-pants had to use it in their daily life. The subjects were a group of 5 friends with the following properties: (i) age: 21 - 31, (ii) 2 males & 3 females, (iii) Greek na-tionality, (iv) university students and (v) living in Am-sterdam. During the trials the database on the server was constantly checked to observe how the participants are using the application.
4.4
Interviews
The 12th and 13th day of the trials the interviews were conducted. The exact questions are listed in the ap-pendix B but in general they were asking about:
• their daily free time
• the time spent on the application
• if they used the application to arrange meetings and how they did it
• how helpful the application was • the application’s features
• the features-incentives that would like to add in the current application
The subjects were free to enrich their own answers, ex-plain and justify them or even disagree.
Three of the interviews were done in person and the other two via Skype. Their duration varied from seven to thirteen minutes and all of them were voice recorded.
4.5
Log files
Apart from the data that were necessary for the appro-priate functionality of the application, we also stored additional usage information on the server. This in-cluded usage timestamps, button taps and extra user input values.
After the interviews we let the participants to use the application for five more days and the last day we col-lected all the data from the server to analyse and com-pare them with the answers we got from the interviews.
5.
APPLICATION DESCRIPTION
To test our research question we thought of three possi-ble applications (each one of them approaches the prob-lem from a different perspective):
The first application focuses on the future planning and it aims at stimulating the people to meet up with their friends by suggesting common free days and hours, nearby places to visit and possible events that may take place that day.
The second application checks if the user is using her phone for more than x minutes playing a game or check-ing Facebook etc. and notifies her. The notification would be like a game or challenge: “Lift your head
up, look on the left and talk to the first person you see there!”. Badges and rewards are given to the users as an incentive to continue completing the challenges. This application focuses on the present and requires from the user to act now and to stop using her phone.
The final application monitors the way the user is us-ing her smartphone (how much time she spends and on which application). The user is able to see these statis-tics and get notifications about her activity. Again it makes use of rewards (badges etc.) in order to give her the incentive to use her phone as less as possible. This application focuses on the past activity on the user’s phone to present the amount of time spent and make her realise that she needs to change her behaviour. Eventually, we decided to move on with the first case and to aim on the minimum viable product (MVP) (Moogk, 2012) given the limited time and resources. According to the MVP the application would both show common free time slots between the users and their friends for the current day and send daily notifications about their availability reminding them to check the application.
5.1
Scenarios
The scenarios below present two possible ways of appli-cation usage:
Scenario 1 - Tom. Tom is a 26 years old student at the University of Amsterdam. His studies are quite demanding and he hardly finds time to meet with his friends. However, he has some free time during the day when he returns from the university and he would like to meet his friends who are also busy with their studies and jobs. What keeps him back is that most of the time his schedule does not fit with his friends’ one so he cannot be sure about their availability.
Fortunately, his friends suggested to him the MeetFriends application. So, he installs it on his smartphone and launches it. After filling his name and phone number he is asked to set his availability time and when he wants to receive notifications about his friends’ avail-ability. He finishes from the university at 5 p.m. every day so he imports his availability time from 5 p.m. to 11 p.m. Some of his friends have already seen that Tom now has the MeetFriends application on his phone and have sent him friend requests. He accepts them and he returns to the main screen where he can see now a list of his friends that will be available today during the time period he will also be. Tony is also in the list. It is more than a month since the last time they met so Tom decides to arrange a meeting with him. He clicks on his name and gets a pop-up message which says that both will be available today between 6:30 p.m. and 8:00 p.m. along with the options to either send him a text message or call him. From that point he can contact him immediately and arrange meetings.
Figure 1: Use Case Diagram
Scenario 2 - Tina. Tina is 30 years old, lives and works in Amsterdam and she is also a part-time mas-ter student. Consequently, she rarely has free time and when she does it is not more than two to three hours daily. That is why she and some of her friends have installed the MeetFriends application on their smart-phones.
Her schedule is very tight so it is very difficult for her and her friends to find common free time slots and most of the times she is too busy to check their availability herself. After finishing her work she is too exhausted to initiate a meeting; however, she misses her friends and she wants to meet them. So, she has set to get notifi-cations from the application in the morning in order to remind her to check her friends availability and arrange a meeting that would fit in her tight schedule.
Today is Tuesday and Tina is heading to her office when she receives a notification from MeetFriends telling her that Tim, Tracey and Tatiana will be available today. Nevertheless, she just receives an email from her mas-ter’s thesis supervisor telling her to meet for one hour in the afternoon to talk about her thesis. So, she opens the Google Calendar application and updates her sched-ule by adding this meeting. Then, she returns back to the MeetFriends, but due to the changes in her schedule Tatiana is now the only available friend. She decides to call her anyway and arrange a meeting.
5.2
Use cases
The use case diagram in figure 1 presents the interaction between the user and the system.
On the first launch of the application the user is able to set her personal contact information that is her name and her phone number. Additionally, she has to set the time slot that she will be available along with the
spe-cific time she wants to receive notifications about her friends availability. She is able to change these prefer-ences whenever she wants to.
Furthermore, friends are needed in order this applica-tion to have a meaning. So, it gives the possibility to the user to add friends from her contact list so that she can interact with them. It will only show contacts that already use the application and the user can send invitations to the ones that she wants. She can also re-ceive friend requests from her friends and she can either accept or decline them. Eventually, the list of all her current friends is presented to her.
The user is also shown the available friends of the day on the first screen whenever she launches the application. Then, she can choose a friend and receive the time slot that both will be available today. In case that she would like to meet him, she is given the options to contact him either by sending a text message or by calling him. Both options prompt the user to the respective pre-installed applications on the user’s smartphone.
Finally, the user receives a notification from the appli-cation once per day at the time that she has set. This notification not only shows the friends that will be avail-able today but also works as a reminder for the user to check her friends availability.
5.3
Application Requirements
From both the scenarios and the use cases we can ex-tract several requirements for our application:
Smartphone device: The users must own a smart-phone device running Android OS, version 4.0 (Ice Cream Sandwich) and above.
Google account: They must also own a Google ac-count. This is required anyway in order to use an An-droid smartphone.
Google calendar: The application takes also Google Calendar (only events that the user is shown as busy) into account in order to calculate the free time. How-ever, it is not mandatory for the user to use Google Calendar for the application to work properly.
Internet connection: The application needs to com-municate with the server in order to get the user’s friends, the invitations and the availability times. Consequently, it requires to have internet access. Without this the ap-plication gives no result and it prompts the user to open the smartphone’s Wifi.
Database: A database is needed to save user and anal-ysis information.
6.
APPLICATION DESIGN AND
DEVELOP-MENT
For the development of the application the Android SDK was used along with the Java programming lan-guage.
(a) User information (b) Availability (c) Notification
Figure 2: Application’s first launch screens and the notification
6.1
Application Interfaces
Research was made on the Google Play store in order to find similar applications and get ideas for the de-sign. We tried to keep it as simple as possible so that the users would be able to find what they need easily without spending too much time on the application and consequently on their smartphone.
To begin with, when the application is launched for the first time it asks from the users to enter their name (as they want to appear on their friends) and their phone number (figure 2a). Then, they have to add their avail-ability by setting the time slot that they are available during the day. This means that the application will only suggest meetings that fit in this slot. In addition, they have to set the time they want to get notified by the application about their friends availability (figure 2b). This notification is shown in figure 2c. It is received daily at the specific time that the user has set it, it en-lists the available friends of this day and prompts the user to touch it for more details. By tapping the no-tification the MeetFriends application opens and more information is presented to the users.
After setting the first launch preferences the users are prompt to the first main screen and they are ready to start adding friends and use the MeetFriends applica-tion. The three main screens are the “Meet ’em”, the “Friends” and the “Preferences” screen.
Pressing the “Friends” option another sub-screen ap-pears with three tabs: “Friends”, “Add” and “Invita-tions”. The first one (figure 3a) enlists all the user’s friends; the user can also delete people from her friends list. In the second tab (figure 3b) the user is able to see which of the contacts on her smartphone are using the MeetFriends application too and send them invitation to become friends. A contact has to be a “friend” in order to interact with the user and to be able to see her availability. Finally, the invitations from other users appear in the third tab. From here the user can either accept or decline the friend request (see figure 3c).
(a) Friends screen (b) “Add” screen (c) Invitations screen
Figure 3: Friends tabs
(a) Meet ’em screen (b) “Meet” dialog (c) Contact options
Figure 4: First Main screen - Arranging a meeting
The user can change the time slot she is available and the notification time anytime by visiting the “Prefer-ences” screen (figure 2b).
The application’s first main and most important screen is the “Meet ’em” screen (figure 4a). Here the user can see all the friends that will be available today and that have common free time. By tapping on the names a popup dialog appears (figure 4b) informing the user about the exact time slot that she and her friend will be available. For example, “Tom and you will be free today between 19:00 and 21:30.”. Additionally, she is given the options to either text or call him. In case she wants to meet him then she can press one of the two buttons and get the respective menus. In figure 4c the user has pressed the “call” button and she has the options to make a call via skype, viber, or just place an ordinary phone call. If she had more calling applications, then these would also appear in the options. If the user wants to check others friends’ availability then she just presses “Cancel” to return to the “Meet ’em” screen and choose another friend. She can also update the list by pressing the refresh button at the top right corner.
Finally, there is feedback from the application for every action that the user performs. If any error or event oc-curs then the respective popup message appears
inform-ing the user about it. “No active internet connection”, “Invitation sent”, “Preferences updated” are some of these messages. Furthermore, if the smartphone is not connected on the internet then the user still gets notifi-cations from the application which prompt her to open the Wifi in order to receive her friends’ availability.
6.2
Database
A PHP web server and a MySQL Database is used to store online user information and usage data. The database consists of five tables; the four tables contain information about the users, the invitations, the users’ friends and the users’ free time. The last table contains information regarding the usage of the application from the users, which is used later in the analysis to draw conclusions and answer our research question.
A timestamp and the respective user’s identification number are stored when the application requests her friends’ availability. An extra value is also stored for each of these requests, that states whether it was initi-ated: (i) by the user (she pressed the refresh button), (ii) by the notification (the application requests infor-mation from the server in order to notify the user with the appropriate message), (iii) automatically by the ap-plication (whenever the user visits the apap-plication’s first main screen). Finally, the total number of “sms” and “call” buttons pressings is also stored.
6.3
Algorithm
All the users have to set the time slot that they are daily available. These time slots along with the saved events on the users’ Google Calendar are used in order to calculate their daily free time and consequently the common free time slots with their friends. The algo-rithm is divided into two phases:
Users’ availability. To begin with, we need to find the time slots that the user will be available today. So, first of all we check her Google Calendar. There, we seek for events that the user is shown as “busy” and that take place during her availability time slot. When-ever we find one then we re-calculate her availability. For example, if her availability time slot is from 18:00 to 21:00 and the algorithm finds an event from 19:00 to 20:00 then the user now will be available only on 18:00-19:00 and 20:00-21:00. If there is no event then the slot remains as it is. This information is saved on the “freeslots” table. A sample of how these slots are estimated can been seen in figure 5. The green bars represent the time slots that the user is available. Common free time slots. The next step is to take the time slots mentioned before and compare them in order to find common free time with her friends. We follow again a similar process as above. For the scope of this project the algorithm only takes the time slots that last more than 30 minutes and in cases where a
Figure 5: User’s free time estimation
user has more than one slot then the first one is chosen.
7.
RESULTS & ANALYSIS
The results we acquired from the exploratory question-naire revealed to us that many of the participants re-ferred on the scheduling issues (their schedule does not fit with their friends’ one or they maybe do not know their availability) as the one of the reasons that keeps them at home. Furthermore, among the answers, the two of the most popular incentives that would make them to move away from home-technology were (a) the need for meeting their friends and to have face-to-face interactions and (b) the easier coordination with friends when the schedule is tight.
During the trials the participants had maximum four hours free time daily, with the 40% of the interviewees having maximum two hours (see figure 6). All of them were students and the trials were conducted during ex-ams period which means more studying, projects, mas-ter’s theses, etc. Two of the interviewees mentioned that during this time period they were indeed too busy studying and their free time was limited.
As regards the time the subjects spent on the applica-tion the 40% of the users identified in figure 7 stated that used the application for 1 to 5 minutes while 20% for less than a minute. Additionally, the rest 40% of the users stated that they used the application for ap-proximately 5 to 10 minutes daily (figure 7). However, the logfiles showed that the daily average time usage for every user was as presented in figure 8. So, from the moment the application was launched by the subjects, it was used for 1 (or less) to approximately 3 minutes. This means that indeed our application was direct, easy to use and did not require useless actions from the users. They were getting the information they wanted without spending valuable time on their smartphones.
Additionally, figure 9 presents the amount of times they used the application each day. The first two days the users had to invite and accept each other as a “friend”. This, plus the fact that they were exploring the
appli-Figure 6: Daily free time Figure 7: Daily usage time (interviews)
Figure 8: Daily usage time (logfiles)
cation led to such big values these days. The maximum application launches per user were 8 times in a day pe-riod and the minimum was 0. On average the users used the application 10 out of the 17 days that the whole ex-periment lasted and they were launching it on average once a day.
In figure 9 we also see fluctuations that demonstrate activity in the application. Although there were days that some of the subjects did not use the application, we observe a consistent usage during the 17 days by all the users until the last day of the trials. This shows us that users cared to see their friends’ availability and had the need for more social interactions.
The 40% of the interviewees answered that they were passive as a user, which means that the users preferred to wait for the notification to remind them to check their friends’ availability (figure 10). Actually, this is also confirmed from the logfiles not only because these users
Figure 10: Kinds of users Figure 11: Meeting’s possibil-ity
spent less time on the application but also because they pressed the “refresh button” the least times compared to the rest of the users.
On the contrary, the remaining 60% of the interviewees stated that they were more active and they not only waited for the notification but they were also checking their friends availability by themselves. Nevertheless, 2 out of 3 people stated that they did both of them mostly at the beginning but afterwards they just waited for the notifications; as one answered, at first the application was still “new” and she was exploring it.
These application could work for both passive and more active users, since both groups arranged meetings. The majority was the one who did both, including checking the availability of its friends by itself, which means that these users wanted to find some friends and meet them, which was easy via the application.
In addition, the 66% of the users who were more active found the application “very helpful”. The rest 33% of them along with all the passive users described the ap-plication as “helpful”. There were two keypoints that convinced and helped the subjects to arrange meetings with their friends: (a) the fact that through this ap-plication they had all their friends’ availability on their hands and (b) both the daily notifications and the fact that through this application they had all their friends’ availability on their hands. The 50% of the users who found the application “very helpful” answered that the keypoint was “a” the easy access they had on their friends’ availability while the 66% of the users who stated that the application was “helpful” answered “b” both. The results indicate that having these two incentives-features in the application was a proper choice since both had impact on the users.
In figure 12 it becomes clear how the users were reacting on the notifications during the trials. We can be sure that on average on 6 out of 17 days the notifications succeeded and convinced the user to open the applica-tion. However, this does not mean that the rest of the days the notification was not successful. In figure 12, the orange bars (“not sure”) declare that the user did receive a notification but she was not online so that we could “catch” this activity. In each of these days there is also activity in the application; nevertheless, we cannot be sure if the notification convinced the user to launch
Figure 12: Notifications’ status
Figure 13: Meetings’ fre-quency
Figure 14: Arranging meet-ings’ process
the application or not; some users told us that they indeed were offline when they received the notification and they went online afterwards to check friends’ avail-ability. Except for the “failed”, in every other case the user used the application (“failed” means that the user received a notification but he/she ignored it.)
The 60% (66% were active users and 33% were pas-sive) answered that during the trials it met its friends as often as before the application while the 40% (50% active and 50% passive) met them more often than be-fore (figure 13). Another 60%, out of which the 66% were passive, stated that arranging meetings with their friends now seems easier (figure 14).
Taking also the piechart in figure 11 into considera-tion, we conclude that the application worked both as a provocation and as an enabler. It enabled the ex-isting intend of the users to meet their friends or even provoked them to arrange new interactions.
Data that came from both the logfiles and the inter-views showed that the users preferred to contact via text and specifically they used other messaging/calling apps (Viber, WhatsApp, Skype, etc.) to contact each other. However, we cannot be sure if the users completed the contact but we know that they at least initiated it by pressing either the sms or call button.
The “automatic notifications”, “setting the time slot that the user is available”, “ the communication op-tions” and “showing the free common time slots” were the most important features while the “setting the no-tification time” feature did not take so much votes. As regards the “Google Calendar” feature only the 40%
Figure 15: Application features
Figure 16: Application new features-incentives
of the participants actually used it. 20% do not use “Google Calendar” in their daily life and another 40% did not happen to use it during the trials. However, they stated that this feature would be important be-cause they would not have to update their availability in the application every time their schedule changes. Generally, all the features were found at least impor-tant by at least one user (see figure 15), which indicates that having these in the MVP was a correct decision.
8.
DISCUSSION & FUTURE WORK
After analysing the results and the experiment obser-vations we have discovered limitations and possible al-ternatives to our methods.
The trials lasted for 17 days and although we got some encouraging results, this amount of time is considered not enough in order to draw decisive conclusions as re-gards the behaviour change. One participant also com-mented in the interview that they felt they needed more time to adjust to the kind of spontaneity promoted by the application.
Additionally, the time period that the trials were con-ducted was not optimal since it was during the exams period, and master’s thesis preparation resulting in lim-ited free time. One of the subjects stated during the interview that she would not go out often even without the use of the application. She believes though, that in any other time period she would make better use of the application and she would give different answers. It would have been advantageous if a more extended re-search on the participants’ past activity and on the fea-tures that they would like to see in this application had taken place. Some of the features presented in figure 16 could have been used in the test application making it
more appealing to the users. We observe that the first three incentives are close to each other and by far the most desirable compared to the rest. Consequently, in a future update these features would provide additional value to the application.
Having information about the subjects’ activity before the beginning of the trials would also be useful. It would be beneficial if we had another time period, under the same circumstances, to track the frequency that the users arrange meetings and the way they do that. Our user group for this project was limited. Expand-ing our research in other user groups with different age range, jobs or even more participants would require dif-ferent approaches and incentives. Probably, the results would be different in that case but our research would be more complete and generalised.
Although by using this application the users agree to give information about their schedule, privacy issues may occur since some of them may do not want to share their free time publicly.
There are also some comments regarding the applica-tion and its features. It would be positive to know with certainty if the user arranged a meeting or not. By tak-ing the number of button presses we know that the user initiated the contact process but we cannot be sure if she completed it.
Pribbenow (1999) mentions in his paper that human dependency on the technology is growing as people tend to use it more and more. Technology is supposed to save us time and this is what we tried to accomplish through our application by making it direct. However, we could make it even simpler so that the users would spend even less time on it. We could omit the “invitation” process that gives the users the possibility to interact with their friends since we observed from the logfiles that takes some time. The 60% of the users (66% of which had described the application as “very helpful”) found this feature very important, because they have too many contacts and they would prefer to choose the ones they want to interact with. However, instead of waiting for friend requests, the users would be able to just choose from their contacts’ list the people that they would like to have access to their schedule and the ones that would like to share their availability.
9.
CONCLUSIONS
In this project we tried to give an answer to whether using smartphone devices appropriately can stimulate social interactions. At first, we made an exploratory questionnaire to decide on what we should focus more, then developed a smartphone application, had user tri-als and finally conducted interviews on the participants.
The time we had to test our research question was lim-ited and we could not draw decisive conclusions regard-ing the human behaviour change. However, the results from the interviews and the logfiles were encouraging and we were able to come up with promising findings. In this project the application was developed deliber-ately easy and simple. One of our goals was that the users will spend as less time as possible on the applica-tion, in contrast to what other smartphone applications are considered to do and that it will avoid anything that could cause problematic usage as Shin and Dey (2013) mention. We believe that the simplicity of the design played an important role. The directness of the pre-sented information along with the daily notifications provoked the users to check the application, without though requiring from them to spent too much time on it. In addition, we believe that with our design we are able to fight against people’s tendency to make less ef-fort and dedicate less time to participate in face-to-face interactions (Brey, 2006) since arranging meetings via the application is easy and does not require great effort. Among the ten ways which Pribbenow (1999) enlists and limit the techonology addiction are the human in-teractions, the frequent breaks away from technology and the timeouts when people will not be online; these are features that our method could provide. We believe that an application like ours (if used for more time) could change the users’ behaviour even more dramati-cally by making them realise how much they like meet-ing their friends. It may lead people to think about these social interactions even when not using the appli-cation and eventually prefer to have a meeting instead of spending their time e.g. at home on the internet alone. However, more extended trials should be conducted in order to answer it.
From the interviewees’ answers we observe that the ap-plication could either work as an enabler (it enabled the existing intend of the users to meet their friends) or as a provocation (it provoked them to arrange new face-to-face interactions). On average, the participants were constantly visiting the application during the trials in-dicating their need to meet their friends. Additionally, the incentives used in this application could work as memory aid, reminding them to meet friends or even as a balance against other applications which require the users’ constant attention and make them addictive to them (L¨ochtefeld et al., 2013). We believe that if an application like this was used for a longer time pe-riod it could make people more mindful as regards the smartphone use and the face-to-face social interactions. In that case, we may not take any usage data from our logfiles meaning that the users do not need the applica-tion to “guide” them any longer.
10.
ACKNOWLEDGEMENTS
I am particularly grateful for the assistance given by Daniel Buzzo and Frank Nack during my master’s thesis and for all the friends that volunteered to participate in our experiments.
References
Abeyrathne, D., Edirisinghe, C., Ranasinghe, N., Karunanayaka, K., Zhu, K., Peiris, R. L., . . . Morisawa, Y. (2011, July). Connected online and offline safe social networking for children. Comput. Entertain., 9 (2), 9:1–9:8. Retrieved from http://doi.acm.org/10.1145/1998376.1998380 doi: 10.1145/1998376.1998380
Barkhuus, L., & Tashiro, J. (2010). Student socialization in the age of facebook. In Pro-ceedings of the sigchi conference on human factors in computing systems (pp. 133–142). New York, NY, USA: ACM. Retrieved from http://doi.acm.org/10.1145/1753326.1753347 doi: 10.1145/1753326.1753347
Brey, P. (2006, September). Evaluating the social and cultural implications of the internet. SIG-CAS Comput. Soc., 36 (3), 41–48. Retrieved from http://doi.acm.org/10.1145/1195716.1195721 doi: 10.1145/1195716.1195721
L¨ochtefeld, M., B¨ohmer, M., & Ganev, L. (2013). Appdetox: Helping users with mobile app addiction. In Proceedings of the 12th in-ternational conference on mobile and ubiq-uitous multimedia (pp. 43:1–43:2). New York, NY, USA: ACM. Retrieved from http://doi.acm.org/10.1145/2541831.2541870 doi: 10.1145/2541831.2541870
Moogk, D. R. (2012). Minimum viable product and the importance of experimentation in technol-ogy startups. Technoltechnol-ogy Innovation Management Review (March 2012: Technology Entrepreneur-ship).
Oulasvirta, A., Rattenbury, T., Ma, L., & Raita, E. (2012). Habits make smartphone use more perva-sive. Personal and Ubiquitous Computing , 16 (1), 105–114.
Pribbenow, K. (1999). Maintaining balance: Mile-high expectations vs. technostress. In Proceedings of the 27th annual acm siguccs conference on user services: Mile high expectations (pp. 180–184). New York, NY, USA: ACM. Retrieved from http://doi.acm.org/10.1145/337043.337138 doi: 10.1145/337043.337138
Shin, C., & Dey, A. K. (2013). Automatically detect-ing problematic use of smartphones. In Proceed-ings of the 2013 acm international joint confer-ence on pervasive and ubiquitous computing (pp. 335–344).
Wiafe, I., & Nakata, K. (2012). Bibliographic analysis of persuasive systems: Techniques, methods and domains of application. Persuasive Technology, 61.
APPENDIX
A.
QUESTIONNAIRE
1. What kind of technology (information or enter-tainment technology) do you use while being at home?
2. How much time do you spend on each kind of tech-nology while being at home?
3. Can you imagine your life with spending less time using this technology? (Yes, No) - Please specify 4. Do you think that you can replace some activities
you use technology for with others that do not require technology? (Yes, No) - Please specify 5. Do you find yourself often exhausted from all this
use of technology or having the need to move away from it and do something else? (Yes, No) - Please specify
6. What is that keeps you at home? (e.g. why watch a movie at home alone instead of going out and watch it with friends?)
7. What kind of incentives do you think could make you to move away from home-technology and go out to do a similar activity with friends?
B.
INTERVIEW QUESTIONS
1. How much was your daily free time during the tri-als? (less than 1 hour, 1-2 hours, 3-4 hours, more than 4 hours)
2. How much time did you spend daily on the applica-tion when you were using it? (less than a minute, 1-5 minutes, 5-10 minutes, 10-15 minutes, more than 15 minutes)
3. Were you an active user or a more passive one? (Did you wait for the application to notify you about your friends’ availability, Did you check yours friends availability yourself, both, other-specify) 4. Did you have any meeting with your friends that
had been arranged via the application? (Yes, No) - If No then skip 6, 10, 11
5. How much helpful was the application in arranging meetings with your friends? (very helpful, helpful, OK, could be better, not helpful at all)
• If “very helpful, “helpful, “OK then, which were the keypoints that convinced you to ar-range a meeting your friends? (the daily noti-fications, the fact that through this application you have all your friends’ availability on your hands, both, other-specify)
• If could be better, not helpful at all then, please say why.
6. Would you meet these friends that day anyway without the help of the application? (definitely, maybe yes, maybe no, no)
7. How did you find the methods, provided by the ap-plication, to communicate with your friends and arrange meetings and which of them you used? (sms, call, other messaging/calling apps(viber, what-sapp, skype, etc.), I did not communicate through the application, other - specify)
8. Below are listed features of the application. For each one of them please specify how much impor-tant these features are for you. (very imporimpor-tant, important, neither important or unimportant, unim-portant, very unimportant)
• auto notifications about your friends availabil-ity
• setting the time slot that you are available • setting the time you want to get notified • the communication options
• the invite friends feature • the Google Calendar feature
• showing the specific time slot that you and your friends are available
9. Below are listed other features-incentives that could be in this application. Please put them in order from the most important to the least important according to what you would like to see in this application:
(a) suggest events that take place that day (b) suggest nearby places (bars, cafe, etc.) (c) show friends availability for more than one
day (week, month)
(d) complete achievements and earn badges when you arrange meetings
(e) notify you about your friends availability when-ever you use your smartphone for more than x hours
(f ) showing common free time slots with group of friends (not just one)
(g) chance to meet new friends by suggesting peo-ple around you that you could go out with (h) showing the weather conditions
(i) other - specify
10. Compared to the time before using the application, how often did you meet with your friends the time period you used the application? (more often, the same, less often, I did not meet my friends at all) 11. After 10 days of usage would you say that arrang-ing meetarrang-ings with your friends even without the help of the application seems easier, the same or harder than before?
Figure 17: Application’s contribution
Figure 18: Daily usage (in minutes)
C.
EXTRA RESULTS
The interviewees’ answers regarding the application’s contribution in arranging meetings are presented in fig-ure 17.
Similar to the graph in figure 9 we also observe in fig-ure 18 that the first two days the users were more active since they had to invite/accept each other as friends. This, plus the fact that they were exploring the appli-cation led to such big values the first two days.
Figure 19 presents the total amount of days that each participant launched and used the application.
Figure 12 informs us about the status of the notifica-tions per user while in figure 20 we can also see the notifications’ statuses per day.
As presented in figure 21 the users preferred to contact mostly via text. From this graph is also verified what one of the users (user 4) stated during the interview that she/he received invitations for meetings but she never initiated one herself/himself; that is why none of the buttons is pressed.
The results in figure 22 agree with what the users stated in the interviews. Users 2 and 3 were waiting for the
Figure 19: Amount of days each subject used the application
Figure 20: Notifications’ status per day
Figure 22: Number of refresh button taps
Figure 23: Number of availability changes per user
notification; they were more passive users compared to the other three who were more eager and pressed re-fresh button more times in order to see their friends availability.
As concerns the daily free time slots we observed that:
1. events from Google Calendar appeared 3 times, 2. the users were mostly available in the afternoon
and evening. The earliest was 16:00 and latest 23:59 and
3. the users’ availability slots ranged from 1 hour to 7:30 hours.
Figure 23 presents the amount of times each user changed his/her availability.
Finally, during the trials the notification times were be-ing changed periodically by the users. However, the latest notification times were all set in the afternoon. The earliest was at 12:00 pm and the latest at 17:00 pm.
