Social Norm Development Within Image Sharing Social Applications


In the last decade as the growth of cell and smart phone usage has developed into an interweaving and—more frequently than not—an overlap of social networks, people have chosen to interact with each other in a way that fosters a certain literacy with these communication platforms. When the average user is presented with an archive of past interactions between users of mobile applications, these exchanges have the potential to guide the user onto a path of commonly practiced socially expected interactions. In this research paper, the relationship between ephemeral data communication platforms and data-persistent applications will be discussed within the scope of social norms on both SnapChat and Instagram. After surveying a focus group of Cornell University undergraduate students, we concluded that there are clear trends (or lack thereof) in how people choose to interact on Instagram and Snapchat dependant on the persistence of user submitted data.


Social Norm Development Within Image Sharing Social Applications


Snapchat and Instagram are two smartphone applications that allow users to share photos with a certain audience. Whereas Instagram is an online and mobile photo/video sharing social networking application that enables users to take pictures and videos, apply filters and then share them on a variety of other social networking services as well as its own, Snapchat is a privacy-driven application that limits the amount of time a viewer can see a photo or video. On Snapchat the sender of the photo or video can choose to specify the number of seconds (up to ten) that the receiver(s) can view the content. After the allotted time period, the video or photo disappears from the receivers application and they no longer have access to it. Conversely, on Instagram, once the user uploads content to their own profile, the photo is then archived in Instagram’s database and displayed to the user’s followers without expiration.  Because of this persistent data collection, Instagram’s general user-base has grown accustomed to a specific genre and quality of images from those that they follow and that which they post. But when examining these so called “norms” that grow through continual use of these two mobile applications, due to Snapchat’s ephemeral data, no such norms should persist because of its inherent lack of cataloging its user’s content.

Among all smartphone owners within the United states, about 12% have and use Snapchat,  while 28% have and use Instagram [1]. In total, there are roughly 30 million Instagram and Snapchat users within the United States. This impressive statistic motivated our research into the relatively new trend of ephemeral data applications that attract users by allowing them to freely and authentically communicate amongst their friends [5]. Snapchat users develop unique usage patterns and norms through Snapchat’s lack of data storage, which is remarkable when compared to content imitation on more durable social media applictions like Instagram. [6] This led us to hypothesize that because of the slow social norm development on Snapchat, there will be more variance in usage and habits among its users when compared to Instagram; Snapchat users will develop similar habits to the habits of their frequent correspondents in the application because there is no other stored data to imply a common social practice; and because Instagram has an archive of previously posted images, users will be more aware and conscious of what is deemed acceptable and normal in a typical Instagram post.

Related Work



According to a paper by Esther Shein, [5] new social media applications are protecting user privacy to an extent that users are now becoming increasingly more comfortable with posting content. The new age of “ephemeral data,” ushered in by applications such as Snapchat, focuses on disappearing content after a set time period.  Ephemeral practices are giving users peace of mind, by removing possible consequences from messages sent to other users.  “People are becoming increasingly aware that some social media platforms keep a permanent record of all that they post and all they comment on or rank or like or otherwise display in their feeds and profiles, so the market is responding with new services that are more evanescent,” according to Lee Raine, director of the Pew Research Center’s Internet & American Life Project. Applications that use ephemeral data such as SnapChat are becoming increasingly popular in our world today, which allows users to have more control over their content.


In another study conducted by Xuan Zhao, the results indicated that different forms of social media are serving as personal archives.  The majority of social media records a history of past posts, which eventually develops into a comprehensive record of social interaction. According to Xuan, Z. (2014), “This ‘stickiness’ is sociotechnical. The ‘deluge of information’ on Twitter meant that tweets were perceived as expiring more quickly, and of course this is self-reinforcing. Less meaningful discussions or … more meaningless things could be posted to Twitter, whereas the higher bar for posting to Facebook meant that less content was there,  and so it expired more slowly. Tools like Snapchat, whereby expiry is built in, inevitably facilitated the generation of playful content.”  However, Xuan also discovered that social media sites sometimes make it more difficult to find personal archives, which leads to conventional social media sites to be treated as more ephemeral.  


Research done by Battestini, Setlur, and Sohn [11] gave us insight into emotional ties across computer mediated communiation. In their study, they found that there was little need for the existence of an emotional connection across CMC platforms and it is not necessary for back and forth communication.  However, we feel that Snapchat is popular because of the greater emotional potential in comparison with a more conventional medium, such as text messaging, which was studied by Battestini, Setlur, and Sohn.  A study done by Shao-Kang Lo (2008),  discovered similar lack of emotion in typical text-based communcation.  Lo fount that “that when Internet users are faced with pure text without emotions, most people cannot perceive the correct emotion, attitude, and attention intents. However, when emotions are added in the same context, the receiver’s perception of the messages starts to significantly change (p. 597).” We believe that Snapchat allows more emotional range than plain-text communication, and this improved emotional comprehension contributed to the popularity of the medium.  

Another article we examined closely was  Joseph Walther’s “Selective Self Presentation in CMC: hyperpersonal dimensions of technology, language, and cognition”. This article examines the fundamental processes that occur when people are developing relationships and expressing themselves through Social Media.  This article gave us insight on how participants presented themselves through their Instagram profiles to match the social norms within each of these communities.  We hope to see this practice less often with Snapchat, because there isn’t a large quantity of durable data that indicates normal usage.  



Fifteen students participated in one of two Image Sharing Application studies.  Depending on which social media account they used more regularly, they were either assigned to use Instagram or Snapchat.  If the participant used both Instagram and Snapchat, they were assigned to the less populated group in order to balance the participation between the two applications.  The research study consisted of three different parts, which all collected information on usage of the application.


Pre-Interview: The first part of the study was an in-person interview conducted by our research team lasting approximately fifteen minutes.  The pre-interview was intended to get basic information about the participant, allowed us to assign them to one of the two groups, and then get initial information about their usage of the application.  We questioned the participants on which features they used and how often, as well as subject content of information they post, and their respective “friends” post.


Self-Reporting Surveys: The second part of the study consisted of a week long self-reporting survey, lasting approximately five minutes per day.  The self-reporting survey was sent every night at 7 PM to the participants using an email address they provided.  The self-reporting survey was created through Google Forms, which also collected their anonymized responses in a separate spreadsheet after submission.

The survey measured application usage for that particular day only, asking many of the same questions from the pre-interview, but instead of average numbers, allowed participants to give more specific responses for each given day.  The survey provided us with both qualitative and statistical data, and also prompted the participants to be more aware of their usage over the seven days.

Post-Interview: The third and final part of our study consisted of an in-person post-interview conducted again by our research team.  Since the post-interview was intended to be an open-discussion with the participants, the session lasted between fifteen and thirty minutes.

We began the session by explaining the purpose of our study and what type of information we were looking for.  We defined “social norms” in terms of our study and gave examples from other sorts of media. (e.g. “On Twitter, people use hashtags to create or join trends.” Or “On Facebook, people may use statuses to vent about their day.”) Although we did not have a specific list of questions that the participants could answer with a “yes”, “no”, or numerical value, we were interested in finding ways to prompt them to consider social norms that they noticed among particular groups of people and friends, as well as social norms that they may have adapted to.  With a given response, we then tried determine the motivations behind particular habits.



Cumalitively we collected a total of 92 self-reporting surveys from our participants, with a total of 55 responses from SnapChat users and 37 responses from Instagram users.  The majority of the data we collected from these surveys is quantitative.

We wanted to explore the amount of usage per day of SnapChat and Instagram.  On both our surveys for SnapChat and Instagram we ask them to record their amount of daily usage on the specific platform.  These are the results we received:


Number of Times Accessed per day
Instagram SnapChat
Mean 6.34375 5.798077
STD 3.525112 2.758693
Median 7 3.5
n 32 52


Additionally, we wanted to collect information about how personally engaged each participant i on both of the platforms, so we collected information about how many times people commented on their friends posted content on Instagram, and how many SnapChats people respond to per day.  Below is a graph of our results:


Figure 1



We were curious as to whether the distributions of these two variables are statistically significant.  We performed a one-tailed distribution t-test and found that the p-value (4.87E-10) to be significantly smaller than .05, therefore the difference between the number of comments contributed per day on Instagram and the number of responses to SnapChats per day is statistically significant.

After this finding we wanted to also explore the different between the quantity that people post to their profile on Instagram versus the quantity people post to their My Story on SnapChat:


Figure 2


Figure 2 shows the amount of daily posts to Instagram’s MyWall vs SnapChats posted to MyStory collected from our surveys




From the quantitative and qualitative data we collected, we can identify some of the trends that differentiate the typical usage and social norm development on Snapchat from other forms of social media.  

Self-Reporting Data

Both Instagram and Snapchat participants were asked to record how many times they accessed the application per day, and the mean of the results indicated that Instagram users access the application more frequently.  However, a t-test run on the same data indicates that the difference is not statistically significant.  From participant responses during pre- and post interviews, we could infer that Instagram users are more likely to browse Instagram newsfeeds casually, while Snapchat users are more likely to open the application when they have a notification.  


The contrast between the number of comments on Instagram posts and responses to Snapchats is shown in Figure 1.   This difference implies that Snapchat is more individually engaging  as a medium.  Each user needs to choose exactly who will be receiving their Snapchat, while Instagram uploads are automatically sent to hundreds of followers.  The more individualized and personal communication that takes place in Snapchat results in more replies.  We encountered similar sentiments in our pre- and post interviews.  A significant portion of participants felt that Snapchat was more conversational, Snapchat is more appropriate to reach out to a smaller group of people more directly, and the more conversational nature leads to more Snapchat replies.  One participant reported that he sent upwards of thirty snapchats each day, and received more than fifty.  He also stated that he responded to every Snapchat he received.  We feel that according to the norms he experienced, each Snapchat requires a response, because a Snapchat is an individual message meant for a particular recipient, similar to a text message.  The trend in number of Snapchat responses vs Instagram also fits in with the responses we received pertaining to the number of friends on each medium.  Most Instagram participants followed hundreds of people, while the number of Snapchat friends was often lower.  Some participants also stated that out of their official Snapchat friends, they regularly communicated with an even smaller percentage.  This implies that Snapchat is meant to reach out to a small group of close friends, while Instagram is meant to broadcast to a larger group of more distant friends and acquaintances.  


In Figure Two, the difference between quantities of Instagram posts and My Story posts is shown.  The My Story feature of Snapchat has very similar affordances to normal Instagram posts, because Snapchat friends can see My Story posts as many times as desired for an extended period of time, and the My Story posts are broadcasted to all contacts.  A significant number of Snapchat participants reported days in which they did not use the My Story function, and a large number of Instagram participants also reported that they did not upload a post to their profile for a certain number of days.  However, the difference between the two is not necessarily relevant, because there were a higher number of Snapchat participants than Instagram participants.    The more interesting trend is the number of Snapchat participants that reported using the Snapchat story multiple times per day.  Even though there is small number of reported days (less than 5) for the higher number of My Story posts, the non-zero result tells us that some Snapchat participants use My Story frequently.  However, judging by the number of reported days,  this is a relatively small group.  Since most Snapchat participants had some alternate form of more durable social media, there is not much motivation to use the My Story feature of Snapchat.  In pre and post interviews, must participants still reported that a portion of their Snapchat contacts frequently used the My Story feature.  My Story seems to be an additional feature separate from the core Snapchat mechanic, while picture uploads are a fundamental feature of Instagram, so it would be more likely that more people on Snapchat would not use My Story than people not posting to Instagram.   


Pre- and Post Interviews

In addition to the quantitative data provided by the self-reporting phase, we were able to identify certain trends in usage and social norm development using the qualitative data provided by our pre and post interviews.  


We encouraged participants to examine their usage of Snapchat in comparison with other forms of social media, and the responses gave us an idea of how Snapchat was used.  Since most participants had multiple social media platforms at their disposal, there must be some type of norm that determines what information and communication is allocated to what platform.  An Instagram participant mentioned that “things put on Facebook and Instagram are different,” and another mentioned that “[Instagram] is more casual than Facebook.” Yet another stated that Instagram is “for documenting something that you’re not sure you want to put on Facebook.” So in some cases, formality or importance determines what information is allocated to what platform.  Most participants felt that Snapchat was used for communication topics and information that was not as important as the information on other forms of social media.  Some participants even mentioned being upset when something important was told to them via Snapchat.  One participant received a Snapchat of a friend’s job acceptance letter, and that participant was upset because they did not get any details and they were not able to examine the photo closely because it was already removed by the application.


Another large influence on typical Snapchat content is humor, and we discovered that multiple participants mentioned some form of humor while discussing their Snapchat habits.  One participant used Snapchat primarily for “silly stuff you wouldn’t want to show people for an extended period of time,”  and multiple participants mentioned sending frequent Snapchats of funny faces.   Most participants mentioned that Snapchat is more appropriate for humorous content than more durable forms of social media, but more generally, participants mentioned that Snachat is more emotional than other conversational forms of communication such as text messaging.  Compared to other forms of social media, this range of emotion is enhanced because of the lack of data durability.  Because Snapchats are not saved permanently, we feel that users are more comfortable showing their genuine emotions in Snapchat compared to a more durable form of social media such as Instagram or Facebook.


The lack of range in emotion in Instagram seems to be caused by perfectionist attitudes we encountered in our Instagram participants.  Because Instagram photos are stored for an extended period of time and broadcasted to hundreds of people, participants felt that they had to construct the best possible impression of themselves on Instagram.  Other forms of durable social media, such as Facebook, lead to similar attitudes.  One participant even stated that “who you are on social media is who you are,” and therefore her goal was to “present the best version of myself.”  Another participant mentioned that she “deleted a[n Instagram] photo because it was under-liked.”  Yet another participant said she won’t post a photo on Instagram “unless I have something spectacular that is going to happen to me so I can post high quality images on there.”  A fourth participant stated that “I almost always ask my friends which filter I should use before I actually post anything.”  All of these participants are demonstrating a similar attitude, which is  that Instagram posts should be carefully crafted to curate an idealized online personality.  We believe this trend is partially caused by data durability, because participants often displayed  much more raw emotion and unrestricted communication on Snapchat.


Our Snapchat participants felt much more comfortable showing a range of emotion or unflattering versions of themselves, often for the sake of entertainment or humor.  This also caused an interesting social practice, which was the adjustment of the Snapchat time limit based on content. If a Snapchat was more embarrassing or more private, participants stated that they were more likely to decrease the time limit of the Snapchat to prevent screenshots.  


From the quantitative self-reporting data, we inferred that Instagram was often used to broadcast to large numbers of distant friends or acquaintances, while Snapchat is used to communicate directly and personally with a smaller circle of close friends.  From the more qualitative pre and post interviews, we were also able to determine that our participants had even smaller groups of friends within their Snapchat contacts.  The members of these groups frequently send Snapchats to each other, and often demonstrate the same social norms.  Most of our participants were college students, so a common grouping of friends was based on geography.   Participants often communicated with a group of friends from their hometown, as well as a group of friends from college.  Some participants also had additional groups of friends that went to a different school, worked in a certain city, or were all members of the same family.  Among these smaller groups of friends, participants often reported experiencing different social norms.  For example, one friend thought the group of friends from his hometown sent positive Snapchats, while people from his college were more negative because they complained about the weather or amount of academic work.  Another participant mentioned that there is one group of friends that frequently uses the draw feature.   A third participant mentioned that one group of friends exclusively used My Story.  Another interesting social norm that varied between groups of friends was the commonality of responses.  One participant reported that one group of friends almost always sent conversational Snapchats, while other groups rarely responded to Snapchats.  Some participants even mentioned friends responding with a plain picture, just to send a text caption.  For example, one participant mentioned that they received a Snapchat that they felt required a response, so they took a picture of a tree and responded in text with a caption.  At this point, Snapchat required a picture to be sent in order to send any response text, so if a user wanted to respond with just text, they would have to send some sort of image or switch to a different communication platform.  Around half way though our study, Snapchat released an update that allowed plain text chat without decaying data.  

Snapchat Updates

One of the difficult aspects of studying Snapchat, or any form of social media, is that the platform can be updated and changed without notice.  In our case, a Snapchat update was released that allowed users to send text messages to one another within the Snapchat application.  This new affordance could cause quite a few changes in Snapchat use, and it would change the implications of our study as data durability and different response affordances are added to the application.  However, since this change was made before the end of the study (and many participant’s post interviews) we were surprised that no participants mentioned the update.  This demonstrates one of the difficulties of designing applications for touch screen and gesture devices, which is that affordances are often invisible to the user.  Even though there was an update to the application that added additional functionality, it may not have been clear to our participants that the new functions were added.  In order to access the chat feature, a user would have to swipe left on a received snapchat.   There was no additional buttons or visible indicators to show that this action was possible, and many users may not have realized that the functionality was present.  We encountered similar invisible functionality problems with other Snapchat features during our study.  In the pre-interview process, we ask what features each participant frequently uses, and in some cases, a participant responded that they weren’t familiar with the feature.  Most of the time, it was the filter feature (accessed by swiping left or right once a picture was taken) and in one case, it was the video feature (accessed by holding down the capture button instead of pressing once).  Both of these features do not have any visible indicators, and for a user unfamiliar with Snapchat, it could be difficult to determine how to access these advanced features.  One participant learned about the filter feature when a friend sent them a filtered picture, and then asked them directly how to add the filter via text messaging.  So even though a new update changed the implications of our study, it also indicates that touchscreen and gesture based applications need to include clear instructions and even demonstrations for each feature.  Otherwise, the only way to hear about these features is seeing it done by a friend or through word of mouth, which can result in completely different trends in usage.  An example would be a user that did not hear about the new text message update, and instead continues to send blank or plain pictures with a caption.  Even though the new update added new functionality to Snapchat, the core mechanic, which is sending ephemeral images, is still in place and is still the main attraction for users of Snapchat.  As a result, many of the norms we discovered throughout our study will still exist, and identifying these norms can be helpful for the designers of future ephemeral communication applications.  


For this study we set out to examine and compare the two social platforms Snapchat and Instagram.  Both of these image sharing applications provide users with similar abilities, but how they have been designed has lead to different usage, and significantly different social norms.


We were able to observe how Instagram is used as more of social platform, whereas SnapChat provides the ability of more of a conversational application.  Additionally, we were able to see that SnapChat users often communicate within smaller social circles, whereas Instagram users tend to follow a larger group of people.  Another social norm we observed was the difference in content posted on Snapchat versus Instagram, which is likely caused by the data durability on Instagram and lack thereof on SnapChat.


We thought that this explorative study helped to reveal a lot about these social platforms.  Due to the scope of the project and the timespan allocated, we were only able to collect limited data.  For future work, it would be interesting to perform a similar study with a larger sample of people with more diverse backgrounds than just the Cornell community.   Additionally, the framework for this study could be expanded to include other image sharing application beyond just SnapChat and Instagram.






[1] Duggan, M. (2013, October 28). Photo and Video Sharing Grow Online. Pew Research

Center’s Internet American Life Project RSS. Retrieved May 16, 2014, from


[2]  Joseph B. Walther & Lisa C. Tidwell (1995) Nonverbal cues in computer‐mediated

communication, and the effect of chronemics on relational communication, Journal of Organizational Computing, 5:4, 355-378, DOI: 10.1080/10919399509540258


[3] Laineste, L. (2013). Funny or Aggressive? Failed Humour in Internet Comments.

Folklore: Electronic Journal of Folklore, (53), 29-46.


[4] Michaeli, N. (2013). Interpersonal competence among users of computer-mediated

communication. (Order No. 3602961, Alliant International University). ProQuest Dissertations and Theses, , 93. Retrieved from (1468954584).


[5] Shein, E. (2013). Ephemeral data. Communications of the ACM, 56(9), 20-22.


[6] Xuan, Z. (2014). Curation Through Use: Understanding The Personal Value of Social

Media. Proceedings of the SIGCHI Conference on Human Facts in Computing Systems, 2431-2440.


[7] Computer-Mediated COmmunication : A reveiw.  Vol 24. no 3. 766-785.


[8] Hancock, T. Jeffery, & Dunham, J. Phillips. (2001). Impression Formation in Computer Mediated

Communication Revisited:An analysis of the Breadth and Intensity of Impressions. Communication Research. vol. 28 no. 3 325-347. doi: 10.1177/009365001028003004.


[9} Lo, Shao-Kang. (2008). A Pragmatic Multimodal analysis of emoticons and gender in social

networks. Accessed on 5/1/2010



[10] “Internet Use Over Time”. Pew Research Center, Washington, D.C. (2014), accessed on 5/1/2014


[11] Battestini, A., Setlur, V., Sohn, T. A Large Scale Study of Text Messaging Use, Mobile Human

Computer Interaction (MobileHCI) 2010.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Create a free website or blog at

Up ↑

%d bloggers like this: