Quantitative/Qualitative Research Study – Deception in Snapchat

ABSTRACT

The importance of communicating with others via mobile phones has increased drastically in the last few years. This has led to the introduction of new kinds of messaging applications such as Snapchat: an application in which the messages disappear once viewed. The feature-based model predicts that a non-recordable medium such as Snapchat should be more deceptive than a recordable counterpart, such as SMS. However, a limitation of this model is that it does not take photo based messages into account. This study aims to compare the deception in SMS and Snapchat via a survey. The results show that our prediction of higher deception in Snapchat than SMS based on the feature-based model did not hold true. Based on these results, there is a need to update the feature-based model for photo-based communication media.

Author Keywords

Snapchat; Deception; Computer Mediated Communication; SMS

INTRODUCTION

Recordability is an important aspect of communication media when studying deception. Studies have shown that more lies are told when they cannot be easily saved and reviewed [3]. Many popular forms of communication media including email, texting, and instant messaging leave a record and are therefore prone to less deception. According to past studies and theories including the feature-based model, media which allow for impermanent messages, should see a higher rate of deceptive content because they lack recordability [3].

Snapchat is a novel technology designed to make mediated communication more ephemeral. Given that Snapchat is designed to reduce recordability, it seems pertinent to explore how this medium fits in with other computer-mediated communication (CMC) technologies with respect to deception. Snapchats are photos that self-delete after a pre-determined time (generally a few seconds). Users can modify their pictures with drawings, filters, and a short line of text. The photo-message can be sent to an unlimited amount of recipients and will remain until the receiver views the message for the allotted amount of seconds (chosen by the sender), after which it disappears.

We have chosen to compare deception in Snapchat to that in SMS due to the main difference between them: text messages leave a trace, while Snapchats are not recorded by default. Both Snapchat and SMS are asynchronous, text-based means of communication, and typically neither participant are co-present during an exchange. However, messages sent through Snapchat are photo-based, restrict text length to about one sentence, and are not recorded.

Given these similarities and differences, we wish to explore the following research question:

RQ1: Is deception via Snapchat primarily photo-based or text-based?

The feature-based model of deception predicts that the frequency of lies told over a CMC medium will depend on the degree of recordability, synchronicity, and co-presence for the medium. Media that are more synchronous and distributed (not co-present), and less recordable, should have a higher frequency of deception [3]. Asynchronous media allow more time for a sender to craft a non-deceptive response to a receiver [3]. A communication in which parties are distributed allows for a greater range of lies, due to a lack of shared context, such as current location and actions [3]. Finally, less recordable media should lead to more lies because there is less proof of deception if the message is not saved [3]. This model predicts that the fewest lies will be told through email, then FtF and IM, with phone calls facilitating the greatest number of lies. A 2004 diary study [3] experimentally validated these findings. In terms of the feature-based model, both Snapchat and SMS are asynchronous and distributed. This is in contrast to other media such as the telephone, which allows for distributed, but synchronous communication. A main difference between the media, however, is that Snapchats have low recordability, while text messages have high recordability. Based on the feature-based model, we should see a higher degree of deception in Snapchat. However, the model does not take into account a visual aspect. Therefore, it will be interesting to see whether Snapchat’s photo-based messages fit in with the predictions of the feature-based model. We have formed the following hypothesis:

H1: People will lie more frequently using Snapchat than text messaging.

Another factor is the nature of deceptive content and whether it differs between the two media. While the format of text messages lends itself to a wide variety of communicative purposes, the photo with caption format of Snapchat, as well as its ephemeral nature, lends itself to a more narrow range of communicative motivations. A 2014 study asked participants which types of content they relay via Snapchat messages, and found that most users send more non-sensitive content than sensitive content (defined as sexual, legally questionable, mean/offensive/insulting content, and documents), and that the primary uses are to send photos of “funny things”, “people”, and “myself” [4].

In contrast, social coordination is a primary use for text messaging. This is a topic that can often lead to deception. When arranging plans with family, friends, or colleagues through text messaging, research has found that people tend to lie when coordinating future events [5]. The types of deception that are told via SMS are based on the affordances and features of the text messaging devices [3].

While social coordination of events is found to be a common form of content found in text messaging, “hyper coordination” [7] and relationship maintenance are also found to be common uses of text messaging. A content analysis done by Thurlow [8] examined over 500 different text messages of college freshmen in the United Kingdom. Relationship maintenance was found to be most prevalent form of content in text messaging at 23%.

We are interested to see whether the difference in range of topics discussed in Snapchat vs. SMS could account for differences in types of deceptive content. We hypothesize that due to the difference in common uses of Snapchat and SMS, reasons to deceive will also differ. For example, because Snapchat is often used to send “funny things,” perhaps it is best suited to non-serious lies and humorous exaggerations. In contrast, the fact that social coordination and relationship maintenance are important uses of SMS could mean that lies about whereabouts or lies to make others feel better are common. Based on this analysis, we have formed the following hypothesis:

H2: Types of deception are inherently different via Snapchat than through SMS.

THE PRESENT STUDY

When doing initial research on this topic, we found that very few studies have been done on Snapchat, and we found no previous research on deception through Snapchat. Due to this lack of prior research, we decided to test our hypotheses and address our research question with an exploratory survey. The purpose of the survey was to get a sense of how people use this new technology, as well as how usage might differ from text messaging. We designed the survey to get a broad sense of how often and in what ways deception occurs via Snapchat. To put this information in context with previous deception and CMC research, our survey allowed people to report both their beliefs about deception in Snapchat and SMS, as well as the nature of their actual deceptive practices via these media.

The survey touched upon many aspects of Snapchat use; for comparison purposes, participants were also asked to report SMS use. The survey consisted of multiple choice, Likert-style, and short answer questions. Participants were asked general questions about their Snapchat and SMS use including frequency of use, common topics discussed, and types of people they communicate with via each media. They were also asked questions specific to deception, focusing on both their actions (who they deceive and what they deceive about), as well as their beliefs about the deception of others (how often they believe others are deceitful about certain topics). For this study, we defined deception as “A successful or unsuccessful deliberate attempt, without forewarning, to create in another a belief which the communicator considers to be untrue” [1]. We included this definition next to each question about deception so that participants would have the same understanding of its meaning. We categorized types of deception for some questions based on a 1996 study by Depaulo, et al., which divides deceptive content into the following categories: Feelings, Achievements, Explanations/Reasons, Knowledge, and Actions/Plans/Whereabouts [2]. The categories for other questions, such as “What do you usually discuss” and “With whom do you usually communicate with” were determined based on our own knowledge of typical usage of the media, and by getting qualitative feedback from peers who also use Snapchat and SMS.

One challenge with collecting Snapchat data is that messages are not recorded, and therefore it is difficult to analyze actual Snapchats. Our study relies on self-reported data based on participants’ memories of past Snapchats. We included examples of deceptive texts for many of the topic categories with the hope that this would help participants recall similar Snapchats they have sent or received. For example, for the Explanations/Reasons category, we added the example, “Saying ‘I’m sick’ to get out of a commitment”, something we could imagine people communicating via both Snapchat and SMS.

METHOD

Subjects. Respondents were college-aged students (18-22) who participated in the study for class credit or a $5 Amazon gift card. Participation in the study was restricted to respondents who self-identified as users of both Snapchat and SMS.  We received 86 total responses, one of which who did not follow directions, leaving us with 85 valid responses.

Procedure. Our data was collected strictly through an online survey that was advertised through a variety of online systems. The study was posted through a web-based recruitment system and publicized through email; those who participated were eligible to receive class credit or a $5 gift card upon completion of the survey.

Measures of Behavior. Participants were asked to report their Snapchat and text messaging behaviors. Questions required respondents to recount how many Snapchats and text messages they sent in a day as well as who the participants communicate most often with, how often they deceive others via these two communication mediums as well as how often they felt that they were being deceived through these two communication mediums. Finally, participants were asked a few open ended questions regarding their potential motives and thought processes behind deception.

RESULTS

The open ended questions in the survey (Q21-22) were coded by three different people and were used to answer our research question and second hypothesis. Regarding the research question; Is deception via Snapchat primarily photo-based or text-based?, we found the answer to be photo-based. Many responses included answers similar to: “Snapchat [deception] would usually be more of an exaggeration, because it is photo proof of whatever you are lying about.” Therefore, it makes sense that deception in Snapchat is primarily photo based.

Our findings also proved our second hypothesis correct; Types of deception are inherently different via Snapchat than through SMS. 25% of participants said they find it harder to lie in Snapchat because of the warranting value of photos, 9% said that Snapchat affords more “self-preservation ” lies, and 15% claimed they lie differently because of the variance in affordances between the two media. Our hypothesis is correct because, from our findings, participants use Snapchat and SMS very differently.  As one survey respondent said, “Snapchat is used when I’m trying to demonstrate some sort of feeling or visually show some sort of image or video that would be useful to the other person involved. SMS would be used for more serious information, tangible information, and information I would need to have stored somewhere that I can access again.” 45% of participants echoed this sentiment by answering that SMS is used for more serious and personal conversations or that Snapchat is used for sending humorous and entertaining messages.

Our results did not match the predictions of the feature-based model that would have been expected with regards to deception in Snapchat. Our first hypothesis, People will lie more frequently using Snapchat than text messaging, was proven incorrect.  The survey data shows that there is more total deception as well as a larger variety of deception sent via SMS than Snapchat. We found that on a scale from 1-5 (1 representing never deceiving and 5 for always deceiving), the majority of respondents reported deceiving an average of 2.6 for SMS and 1.8 for Snapchat. Based on our coded responses, one reason for this result is that photos in Snapchat act as a warrant preventing deceptive snapchats.
This finding also reveals that the feature-based model is outdated, as it does not take into account photo-based messages. As communication technology continues to evolve and integrate different types of media, the feature-based model must be modified to account for more factors that may influence deception.

Snapchat-Chart

DISCUSSION

This work has implications toward our understanding of how deception differs among communication media. One important finding is that photo-based media such as Snapchat often facilitates deception by photo exaggeration rather than outright lies.

Visual cues such as photos make it harder for people to deceive others, regardless of other features of the media including recordability and synchronicity. This is due to the warranting value of photos, which is not present with text-based media such as SMS. Our results’ disagreement with the prediction of the feature-based model implies a need to update the model to incorporate visual warrants such as photos.

Furthermore, there was a significant difference in the type of deception occurring via Snapchat when compared to SMS. Snapchat was found to be used mostly between friends as a space for non-personal and humorous conversation. In contrast, SMS was found to be a more serious space used to communicate more urgent, complex, or important matters.

In addition to the above findings, it is interesting to note that people felt they were deceived more in SMS than in Snapchat. The reason could be that the users believe a photographic proof  to be more truthful than a text message as is evident from the quote: “I don’t really lie a lot, but snapchat kind of forces you to be more truthful because they can see you.”

In photo-based messaging applications such as Snapchat, the amount of deception is lower even when messages are not recordable. The feature-based model predicts that the rate of deception would be higher, but it seems that the warranting value of photos is more powerful than the lack of recordability. One implication of this research is that the feature-based model must be updated to incorporate photo-based media. As communication technology advances, existing models must be updated to accommodate new trends.

Snapchat-Pie

Limitations and Future Work

While this study provides valuable insight into deception through Snapchat and drawbacks of the Feature-Based Model, there are important limitations to consider. Due to the ephemeral nature of the medium, our survey relied on the accuracy of participants’ memory of their Snapchat use. While participants could review their text messages while answering the survey questions, this was not the case with Snapchats. In the future, more accurate data may be acquired if users could be prompted to answer a few questions upon sending a Snapchat. For example, they might be asked, “Was that Snapchat deceptive?” with some follow-up questions if the user answered yes.

Another limitation is the homogeneity of our participants. We chose to focus on students at a particular university for this study; however, this made for a sample that is not representative of the general population of Snapchat users. Future work should focus on acquiring data from a more diverse population.

We also realize that there is a significant difference in the number of Snapchat and SMS messages sent since the use of SMS is more widespread in our participants. As a result, there is a proportional difference between the lies in SMS vs Snapchats that were sent. Future work should focus on recruiting participants who use both media equally.

There are features of Snapchat that we did not focus on in our study. It may be interesting to study the types of deception that may exist with Snapchat videos, story timelines, or text messages. These features of the medium are also ephemeral, but differ from traditional Snapchat messages, and thus would be interesting to look at in the future.

REFERENCES

  • Vrij, A. (2008). Detecting lies and deceit: Pitfalls and opportunities (2nd ed.). Chichester: John Wiley.
  • Depaulo, Bella M., Deborah A. Kashy, Susan E. Kirkendol, Melissa M. Wyer, and Jennifer A. Epstein. “Lying in Everyday Life.” Journal of Personality and Social Psychology 70.5 (1996): 979-95. Web. 6 Oct. 2014.
  • Hancock, J.T., Thom-Santelli, J., & Ritchie, T. (2004). Deception and design: The impact of communication technologies on lying behavior. Proceedings, Conference on Computer Human Interaction, 6, 130-136. New York, ACM.
  • Roesner, Franziska, Brian T. Gill, and Tadoyoshi Kohno. “Sex, Lies, or Kittens? Investigating the Use of Snapchat’s Self-Destructing Messages.” (2014): n. pag. Web. 6 Oct. 2014.
  • Porath, S. (2011). Text Messaging and Teenagers: A Review of the Literature. Journal Of The Research Center For Educational Technology, 7(2). Retrieved November 11, 2014.
  • Grinter, R. E., & Eldridge, M. A. (2003). Wan2tlk?: Everyday text messaging. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (pp. 441-448). New York: ACM Press.

 

  • Ling, R., & Yttri, B. (2002). Hyper-coordination via mobile phones in Norway. In J. Katz & M. Aakhus (Eds.),Perpetual contact: Mobile communication, private talk, public performance (pp. 139-169).Cambridge: Cambridge University Press.

 

  • Thurlow, C. (2003). Generation Txt? The sociolinguistics of young people’s text-messaging.Discourse Analysis Online, 1(1)

 

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 WordPress.com.

Up ↑

%d bloggers like this: