Content uploaded by Lukasz Piwek
Author content
All content in this area was uploaded by Lukasz Piwek on Mar 29, 2016
Content may be subject to copyright.
Full length article
“What do they snapchat about?” Patterns of use in time-limited
instant messaging service
Lukasz Piwek
*
, Adam Joinson
University of the West of England, Centre for the Study of Behaviour Change and Influence, Bristol, UK
article info
Article history:
Received 15 April 2015
Received in revised form
11 July 2015
Accepted 23 August 2015
Available online xxx
Keywords:
Instant messaging (IM)
Social network sites
Snapchat
Critical incidence technique
Social capital
abstract
The use of Snapchat e a time-limited instant messaging service e has been rapidly rising amongst ad-
olescents. However, the exact nature of Snapchat use remains difficult to examine due to the self-
destructive nature of content sent and received via this service. We report an online survey conducted
with the use of a memory sampling method to enquire about the specific details of the very last image or
video each participant sent and received via Snapchat. We found that users mainly share ‘selfies’, typi-
cally embed text and ‘doodles’ with photos they share, use it mostly at home, and primarily for
communication with close friends and family as an ‘easier and funnier’ alternative to other instant
messaging services. We also found that high intensity of Snapchat use was more associated with bonding
rather than bridging social capital. We discuss those findings in the context of existing studies on the use
of instant messaging servic es and social networking sites.
© 2015 Elsevier Ltd All rights reserved.
1. Introduction
Instant messaging (IM) has become an ubiquitous feature of
rapid communication in the ‘Global Village’ with the fast adoption
of internet-enabled mobile phones at the beginning of the 21st
century. IM is a type of online chat which offers real-time exchange
of text, images, video and voice transmission over the Internet, but
it is also used for exchanging emotions via emoticons, information
provision, behaviour change interventions and surveying (Cole-
Lewis & Kershaw, 2010; Hawn, 2009; Ramirez & Broneck, 2009;
Ogara, Koh, & Prybutok, 2014). In 2014 there were reportedly 50
billion IM sent per day e twice as many as conventional text
messages (Curtis, 2014) and it is estimated that IM apps will ac-
count for 75% of mobile trafficby2018(Juniper Research, 2014). IM
capability has been also integrated into almost every major social
networking site with smartphone app services such as Facebook
Messenger, Twitter, Googleþ or LinkedIn. There are also a large
number of popular, standalone IM mobile services such as What-
sApp, Skype,orInstagram.
In the majority of existing IM services listed above, the content
that users exchange is stored on both senders' and receivers’
devices creating a communication history, with the exception being
a real-time, streaming voice and video chat communication service
such as Skype. However, a new category of IM has recently risen to
prominence e Snapchat (http://www.snapchat.com). What makes
Snapchat stand out from other IM services is that the content users
share only persists for a limited period of time.
1.1. The overview of snapchat
The rise in Snapchat use has been one of the most rapid and
unprecedented in the history of instant messaging services and
social networking sites. Its estimated that Snapchat's base of active
users grew from 10 million in mid-2012 to over 70 million in early
2014, and 100 million in early 2015 (according to Wall Street Journal
evaluation e Snapchat doesn't reveal its numbers; Macmillan &
Rusli, 2014; Wohlsen, 2015). In December 2013 more than 400
million ‘snaps’ (the common term for video messages and photos
send via Snapchat) were received on Snapchat every day (Shontell,
2013). By comparison, it takes Facebook and Instagram combined to
match the same number of photos shared in the same period.
Snapchat
reportedly rejected an acquisition offer worth $3 billion
from Facebook (Rusli & Macmillan, 2013) and was valued to be
worth $10 billion by two independent companies in August 2014
(Rusli & Macmillan, 2014), and $19 billion in early 2015 (Wohlsen,
2015).
* Corresponding author. University of the West of England, Coldharbour Lane,
BS16 1QY, Bristol, UK.
E-mail address: lpiwek@gmail.com (L. Piwek).
Contents lists available at ScienceDirect
Computers in Human Behavior
journal homepage: www.elsevier.com/locate/comphumbeh
http://dx.doi.org/10.1016/j.chb.2015.08.026
0747-5632/© 2015 Elsevier Ltd All rights reserved.
Computers in Human Behavior 54 (2016) 358e367
The process of sharing on Snapchat works as follows: the sender
makes an image/video using the Snapchat smartphone app and
then choose how long the image/video will be viewable by the
receivers' device (between 1 and 10 s, as of April 2015). Sequences
of images/videos can also be sent. When the sender posts an image/
video to the receiver, this image/video automatically vanishes from
the senders' smartphone. The only information that persists on the
senders' device is a timestamp of when the snap was send. The
receiver now has an option to view the content but the viewing
time is limited to the specific duration chosen by the sender (i.e.
between 1 and 10 s). After the receiver views the image/video for
this particular duration, the image/video disappears from the re-
ceivers’ phone.
There are a number of additional features that make Snapchat a
unique IM service. Snapchat is exclusively a smartphone app
(available on Apple iOS and Google Android enabled devices) and
therefore it is not possible to use it with the browser (unlike Twitter
or Facebook Messenger). Any image/video is only shared with the
friend, or a group of friends, selected by the sender and those
friends have to be a Snapchat users. This way the sender always
decides who is going to receive and view the content. If the receiver
makes a screen capture of the image, the sender is notified about
this action. Additionally, the recipient must maintain tactile contact
with the device's touchscreen, thereby hindering their ability to
take a screenshot or use an external camera. However, it has been
widely reported that third-party apps such as Snapsaved allow the
receiver to make a hidden screen capture of ‘snap’ without sender
being aware of this (Cook, 2014b). Snapchat users can also embed
32 characters-long text messages, or create a finger-drawn ‘doo-
dles’, layered on the top of the photos they capture. Video chat is
also possible: users see a pulsating blue bubble that indicates
whether their friend is ‘active’ in Snapchat, and can engage in video
chat.
1.2. Snapchat use, privacy and social capital
Beside the effortless and easy-to-use interface design of Snap-
chat, the most unique features relate to the personal sharing of
content that disappears after specified period (you choose specific
person/group of people to share it, rather than share with a large
group of people or publicly by default). Informal media reports
suggest that the self-destructive nature of messages may remove
some inhibition from users who would otherwise not share such
content. It's been widely reported in social media and market
analysis that Snapchat is particularly popular amongst children and
teenagers, with half of the users aged between 13 and 17 (Statista,
2014). At the same time, there are informal reports that Snapchat
may be frequently used for
‘sexting’ (the act of text messaging
someone in the hopes of having a sexual encounter with them
later; with initially casual content transitioning into highly sug-
gestive and even sexually explicit content e UrbanDictionary,
c2008). Some market research conducted in the UK supports
these claims: half of all 18 to 30-year-old respondents reported
receiving nude pictures, while 67% had received images of “inap-
propriate poses or gestures” (Kemp, 2013). This issue has been
broadly discussed in the context of online security and privacy in
the media, especially with reference to the incident in October 2014
e a major privacy breach where 100,000 ‘snaps’ were published
online allegedly by hackers who compromised Snapsaved servers
(Buchanan, 2014). This event has been termed ‘the Snappening’.
1
It
was widely reported that a significant proportion of leaked ‘snaps’
were explicit in nature (Cook, 2014a) and due to the young Snapchat
demographics there were concerns that the stolen photos would
include indecent images of children.
However, a single study to date that examined privacy issues
with Snapchat use contradicts the assumption that adult Snapchat
users engage in ‘risky’ and explicit sharing. Roesner, Gill, and Kohno
(2014) surveyed 127 adult Snapchat users and found that most
users reported that they did not send sensitive content (although
25% reported they might do so experimentally). Specifically, they
found that the majority of users were not willing to send content
classified as ‘sexting’, photos of documents, messages containing
legally questionable content, or content considered insulting.
Additionally, researchers found that security was not a major
concern for the majority of respondents e most of the users un-
derstood that the messages could be recovered and that screenshot
taking was a common and expected practice Roesner et al. (2014).
The issues of privacy and online sharing lead to another
important question e what is the nature and role of Snapchat use in
facilitating social interactions and networking? One of the major
impacts of emerging social networking sites and digital commu-
nication technologies is their function as a “social lubricant” e
facilitating the building of social capital between users (Lee, Kim, &
Ahn, 2014; Steinfield, Ellison, & Lampe, 2008). Social capital has
been defined as the “connections and the associated norms of
reciprocity among people” (Putnam, 2001). Putnam distinguished
between two types of social capital: bonding and bridging. Bonding
social capital refers to strong-tie relationships such as family,
partners and close friends, where people share strong personal, or
intimate, connections and provide emotional support to each other.
On the other hand, bridging social capital refers to weak-tie re-
lationships such as previous coworkers or former classmates,
where people don't share a similar background or emotional reci-
procity. Previous research has shown that all kinds of social capital
yields positive outcomes such as self-esteem, life satisfaction, and
even health (Ellison, Steinfield, & Lampe, 2007; Kim, Subramanian,
Gortmaker, & Kawachi, 2006; Valenzuela, Park, & Kee, 2009
).
Social capital has been extensively examined in the online
context especially with the use of social networking sites (SNS),
particularly Facebook. A large number of studies on Facebook have
found a strong association between the use of Facebook and social
capital, especially for the creation and maintenance of bridging
social capital (Ellison et al., 2007, Ellison, Vitak, Gray, Lampe, 2014b;
Vitak, Ellison, & Steinfield, 2011). For instance, it has been estab-
lished that high frequency and duration of use of Facebook, high
routine access and high emotional connectedness to Facebook is
strongly associated with social capital (Ellison et al., 2007). Social
capital has been also examined in the context of SNS connection
strategies (Ellison et al., 2014b) and the frequency of features use on
Facebook (Lee et al., 2014). For instance, Lee et al. (2014) found a
strong association between the frequency of using features such as
Wall Posts, Comments, Messages and Status Updates with both
bonding and bridging social capital.
While Snapchat has rapidly risen to popularity since 2012
(Duggan, 2013) the exact nature of its use is still unknown, and its
also not clear how this use is associated with bridging and bonding
of social capital. The study by Roesner et al. (2014) mainly focused
on perceived privacy and security amongst Snapchat users:
whether users send sensitive content, how aware are they of the
security drawbacks of Snapchat, how frequently they make and
accept the making of screenshots. Utz, Muscanell, and Khalid (2015)
compared Snapchat and Facebook use in the context of romantic
jealousy, and showed that Snapchat was used more for flirting and
1
Term ‘Snappening’ comes from combination of words snap and ‘happening’ ,in
reference to an event that happened shortly before in August 2014 e ‘the Fap-
pening’ (combination of ‘fap’ e the onomatopoeic term for masturbation, and
‘happening’) where a large number of nude celebrities photos and videos leaked to
4chan.org (Kosur, 2014).
L. Piwek, A. Joinson / Computers in Human Behavior 54 (2016) 358e367 359
finding new love interests, whereas Facebook was still the main
social networking site used for keeping in touch with friends. The
only other available publications are informal blogs, online maga-
zines stories and market analysis reports. There is a paucity of
details such as specific patterns of Snapchat use, reasons for Snap-
chat use, the context in which people use it and the frequency of use
or type of content users share. No other research has so far been
conducted on Snapchat use, although there are extensive studies on
the use of other popular social networking sites and instant
messaging services. Therefore, the primary aim of the study
described in this article was to examine how people use Snapchat,
what content they tend to share, what they use it for, with whom,
how frequently, and what value it presents for them. A primary
objective was to gain a better understanding of what content
people send specifically, why they send it, and what they mainly
use Snapchat for. Our secondary objective was to examine how
Snapchat use is associated with bridging or bonding social capital,
following an approach similar to one used by (Ellison et al., 2007,
Ellison, Gray, Lampe, Fiore, 2014a) with Facebook. To this end we
conducted an exploratory survey to examine patterns of Snapchat
use by employing elements of a memory retrieval method taken
from a qualitative Critical Incidence Technique. We then conducted
a follow-up survey where we looked at the association between the
intensity of Snapchat use with a range of other factors, and bonding
and bridging of social capital.
2. Methods
2.1. Participants
We used a short invitation e-mailed to 2194 first-year students
at the University of the West of England (UWE). The e-mails were
provided to us with the permission of both Student Services and
Business Intelligence and Planning at UWE and were distributed
directly via Qualitrics Research Suite system (Qualtrics, c2013)
preserving full participants' anonymity. We used only first-year
students due to UWE regulations regarding the survey distribu-
tion and also because existing market research point to a large
proportion of Snapchat users amongst young age groups (Statista,
2014). A total of 209 participants (139 female and 70 male)
agreed to participate in an online survey. The only requirement for
participation in the survey was that the person is, or was, a Snap-
chat user. The study received ethical approval from the University
Ethics Review Board.
2.2. Measures for the exploratory survey and critical incidence
technique
The survey was constructed using the Qualtrics Research Suite
(Qualtrics, c2013) and took approximately 10 min to complete. Each
participant was presented with a brief instruction on the purpose of
the study and the approximate time needed for completion. Par-
ticipants were informed that the information they provided would
be secured under the Data Protection Act 1998. Participants were
asked to confirm that they understood the instructions by ticking a
box before they started the survey. The survey incorporated both
open-ended and closed-ended questions with mixed-type
response scales and was split into two parts: (1) general socio-
demographic questions and general information about the use of
Snapchat and other SNS/IM platforms, and (2) a memory retrieval
task for the last snap sent and received.
Demographics and the general use of Snapchat and SNS. General
sociodemographic data was collected on: gender, age, ethnicity,
marital status, sexual orientation, employment status, whether
participants were students or not, highest education level, country
of residence, and location/area of residence. General questions
about the use of Snapchat and other SNS included four items which
questioned participants about: (1) the reason they started using
Snapchat, (2) how many people they actively interacted with on
Snapchat, (3) whether they used other SNS to share photos/videos
and (4) how frequently they used various IM or chat services on
their smartphone.
Critical Incidence Technique. One of the major challenges of
conducting a study on Snapchat is the transient nature of any
content generated or received on the platform. In studies investi-
gating the ‘digital traces’ of human behaviour on social networking
sites, such as photos or ‘likes’ on Facebook (Kosinski, Stillwell, &
Graepel, 2013; Lambiotte & Kosinski, 2015), the content is present
in the recorded history of the user's profile, with visibility varying
only based on the privacy setting. However, there are no such
‘digital traces’ for Snapchat, except from a timestamp with the in-
formation on when the snap was sent/received. In such cases it is
only possible to employ introspective methods which have obvious
shortcomings: users can be biased, they may not remember clearly,
or they can omit important content. In order to maximise the
possibility of capturing a representative sample of Snapchat expe-
riences' we utilised elements of Critical Incidence Technique (CIT;
Flanagan, 1954) as a part of the online survey questionnaire. First
described by John C. Flanagan in 1954, CIT is a well-established
qualitative research tool used in the fields of health science, edu-
cation and market research. Critical incidents can be gathered in
various ways, but typically respondents are asked to tell a story
about an experience they have had. The key idea is that the ob-
servations collected from participants should be recorded as close
as possible to the time when they occurred, which improves
memory retrieval. The recollection should also be structured to
provide better contextual layout to the event e memory is
improved if the observers know in advance that they will need to
report (FitzGerald, Seale, Kerins, & McElvaney, 2008).
Application of CIT was ideal in our study because we didn't have
direct access to content that participants shared and we were not
certain what factors were important in the use of Snapchat. At the
same time it was easy to isolate a single ‘incidence’ of Snapchat use
as a recollection of the last snap that was sent and received. This
approach was especially relevant because the only ‘digital trace’
that is left on a user's smartphone is a timestamp of their recent
communication. Therefore, we used a set of questions to enquire
about participants' memories of the last snap image/video they
sent and received. This set of questions was designed following the
guidelines from Flanagan (1954) and FitzGerald et al. (2008) to
facilitate a detailed memory retrieval of the last snap incidence.
The
Snapchat memory retrieval task started with the following
question: ‘do you have access to your Snapchat app at the moment?’
If participants replied ‘no’ to this question, they were asked to give a
rough estimate of the date when they sent the last snap (day,
month and year). If they replied ‘yes’, they were asked to open the
Snapchat app, look at the interaction history, and record the exact
date (day, month, year) and time (hour, minute) when the last snap
was sent. This task therefore also aimed to create a better ‘anchor’
for memory retrieval for the last snap sent (Flanagan, 1954;
Gremler, 2004).
After participants recorded a timestamp (or estimate of when
they sent the last snap), they were asked a number of questions
about the last Snapchat they've send. Questions related to five
different description categories: time of sending the snap; the
content of the snap (whether it was photo or video, what was on
the snap, whether they ‘doodled’ on it, whether it was a reply to
another snap, and whether they made a screenshot of the received
snap); the reason for sending the snap; participants' location when
they sent the snap; and socially-related factors (whether they sent
L. Piwek, A. Joinson / Computers in Human Behavior 54 (2016) 358e367360
it to single person or a group of people, who this were specifically,
what their mood was when they sent it, and whether they had been
drinking alcohol when they sent it). The open-ended responses
obtained with CIT were classified by two separate judges and
Cohen's
k
was used to calculate agreement amongst the judges.
Cohen's
k
result for inter-rater reliability are provided with each
figure that corresponds to the relevant items.
2.3. Measures for the follow up survey on social capital
After obtaining the initial results, we decided to examine how
our participants used Snapchat to build bridging and bonding social
capital. We contacted 209 participants who completed the first part
of the survey and asked them to participate in a short, 5 min follow-
up survey. A total of 96 participants completed the follow up survey
(67 female, 30 male). The follow up survey included following three
scales: (1) Snapchat use Intensity, (2) Bridging (3) and Bonding
social capital on Snapchat. We also asked participants how much
time on average they spend every day using the internet and
Snapchat.
The Snapchat Use Intensity scale was adopted from Ellison et al.
(2007) Facebook Use Intensity Scale to obtain a better measure-
ment of Snapchat use than frequency or duration indices. The
Ellison et al. (2007) Facebook Use Intensity Scale has a broad scope
of questions that are easily generalised to any other social network
e which is why we decided to apply it with Snapchat. This measure
included two self-reported assessments of Snapchat behaviour,
designed to measure the extent to which the participant was
actively engaged in Snapchat activities: (1) the number of Snapchat
friends and (2) the amount of time spent on Snapchat in a typical
day. In addition, the measure also included a series of Likert-scale
attitudinal questions designed to examine the extent to which
the participant was emotionally connected to Snapchat and the
extent to which Snapchat was integrated into their daily activities.
Those remaining questions included the following items: (3)
Snapchat is part of my everyday activity; (4) I am proud to tell
people I'm on Snapchat; (5) Snapchat has become part of my daily
routine; (6) I feel out of touch when I havent logged onto Snapchat
for a while; (7) I feel I am part of the Snapchat community; and (8) I
would be sorry if Snapchat shut down. The eight-item index was
found to be reliable (Cronbach
a
¼ 0.89) and descriptive results for
this scale are summarised in Supplementary Table 4.
Bonding and Bridging social capital with Snapchat was measured
using a 10-item scales adapted from Williams (2006). We chose this
particular scale because of its broad and generic nature. Addition-
ally, a large number of studies adapted Williams (2006) scale to
examine social capital on Facebook (e.g. Brooks, Hogan, Ellison,
Lampe, & Vitak, 2014; Ellison et al., 2014a), WhatsApp (Aharony,
2015), and Twitter (Hofer & Aubert, 2013).
Items for Bonding social capital with Snapchat asked respondents
to rate, on a five-point Likert scale, the extent to which they agreed
or disagreed with the following statements: (1) There are several
people on Snapchat I trust to help solve my problems; (2) There is
someone on Snapchat I can turn to for advice about making very
important decisions; (3) There is no one on Snapchat that I feel
comfortable talking to about intimate personal problems; (4) When
I feel lonely, there are several people on Snapchat I can talk to; (5) If
I needed an emergency loan of £50 0, I know someone on Snapchat I
can turn to; (6) The people I interact with on Snapchat would put
their reputation on the line for me; (7) The people I interact with on
Snapchat would be good job references for me; (8) The people I
interact with on Snapchat would share their last dollar with me; (9)
I do not know people on Snapchat well enough to get them to do
anything important; and (10) The people I interact with on Snap-
chat would help me fight an injustice. The ten-item index was
found to be reliable (Cronbach
a
¼ 0.86) and descriptive results for
this scale are summarised in Supplementary Table 4.
Items for Bridging social capital with Snapchat asked respondents
to rate, on a five-point Likert scale, the extent to which they agreed
or disagreed with the following statements: (1) Interacting with
people on Snapchat makes me interested in things that happen
outside of my town; (2) Interacting with people on Snapchat makes
me want to try new things; (3) Interacting with people on Snapchat
makes me interested in what people unlike me are thinking; (4)
Talking with people on Snapchat makes me curious about other
places in the world; (5) Interacting with people on Snapchat makes
me feel like part of a larger community; (6) Interacting with people
on Snapchat makes me feel connected to the bigger picture; (7)
Interacting with people on Snapchat reminds me that everyone in
the world is connected; (8) I am willing to spend time to support
general Snapchat community activities; (9) Interacting with people
on Snapchat gives me new people to talk to; and (10) On Snapchat,I
come into contact with new people all the time. The ten-item index
was found to be reliable (Cronbach
a
¼ 0.91) and descriptive results
for this scale are summarised in Supplementary Table 4.
3. Results
3.1. Exploratory survey
The majority of participants (89%) were aged between 16 and 25
years, of white ethnic origin (89%), not in a relationship (72%),
heterosexual (88%), and living in an urban or suburban area (83%).
The summary of demographic details for participants is shown in
Table 1.
Over 47% of participants reported that they started using
Snapchat because their friends were using it, and because it's fun to
use (17%), with other reasons being: easy and free (8%), curiosity
(6%), communication (5%), and privacy (2%). Almost 80% of users
reported that they use Snapchat to interact with no more than 12
people on a regular basis (Fig. 1). The majority of participants also
use Facebook (96%), Instagram (59%) and Twitter (58%) to share
photos or videos using a smartphone, with less frequently used SNS
being Tumblr (13%), Pinterest (8%), Flickr (4%) and WhatsApp (1%).
Participants reported SMS, Facebook Messenger and Snapchat as the
most frequently used instant messaging services, as seen in Fig. 2 .
When required to recall the context and description of the last
snap participants sent, almost all users reported their they last snap
was a photo (95%) with a ‘doodle’ embedded (74%) which was
mostly in the form of a text message (94%) and rarely a drawing
(6%). Half of the participants reported that the last snap they sent
was was a ‘selfie’,
2
while the remaining participants sent a broad
range of content such as screenshots (7%), food images (7%), or
various other objects (6%) e see Fig. 3a for more details. Almost 55%
of participants reported that the snap they sent was a reply to one
they received. Similar to the ‘sent’ content, a selfie was the most
frequently reported snap received (63%; Fig. 3a) and almost all
(96%) participants reported that they did not screenshot the ‘snap‘
they received. The result of inter-rater reliability Cohen's
k
showed
a substantial degree of agreement between raters for judging both
‘send’ (
k
¼ 0.69, z ¼ 26.1) and ‘ received’ content (
k
¼ 0.54, z ¼ 14.5).
The majority of participants reported that communication (48%)
and desire to share funny, personal or emotional content (40%)
were the main reasons for sending the snap, with other reasons
being boredom (5%); 7% of participants did not remember why they
sent it. Interestingly, most participants reported being in various
2
Selfie is a photograph that one has taken of oneself, typically one taken with a
smartphone or webcam and shared via social media (Ef tekhar et al., 2014).
L. Piwek, A. Joinson / Computers in Human Behavior 54 (2016) 358e367 361
locations at their home (75%), such as in their room or bed, while
sending the last snap (Fig. 3d). The majority of ‘snaps’ were sent in
the late morning/early afternoon hours between 10 am and 2 pm
(27%), and in the evening between 7 pm and 11 pm (32%), as seen
on Fig. 3e.
The majority of participants reported that the recipient of the
snap was a single person (73%) with close friend (55%) and partner
(18%) being the most common recipients (Fig. 3b). Participant's
who sent their snaps to a group of people (27%) sent it mainly to
close friends (62%), although a mix of random people was also
highly reported as a recipient (29%; Fig. 3c). The majority of par-
ticipants reported being in a good or very good mood (76%) when
they sent the snap (18% reported neither good nor bad mood, and
5% reported being in bad mood). The majority of participants had
not have been drinking when they sent the snap (92%) and those
that did (8%) had an equivalent of four pints of lager on average.
3.2. Follow-up social capital survey
In the follow up survey we examined whether Snapchat is used
for bridging and bonding social capital and we collected more de-
tails on the intensity of Snapchat use. As we described in the
Methods section 2.1, we recruited a subset of participants (n ¼ 97)
who took the main exploratory survey and this subgroup had very
similar sociodemographic characteristics (see Table 1 for detailed
comparison).
In order to explore the relationship between Snapchat use and
the various forms of social capital, we conducted a number of
regression analyses. In each regression, we controlled for socio-
demographic, Internet and IM use factors, and intensity of Snap-
chat use, in order to see if the use of Snapchat accounted for vari-
ance in social capital over and above these other independent
variables. A descriptive summary of the results for intensity of
Snapchat use, bonding, and bridging can be found in the
Supplementary Table 4.
Due to relatively small sample size included in the regression
analysis we first conducted regression diagnostics to establish
whether any of the four assumptions of linear regression were
violated. To this end we applied a global test procedure
c
G
2
4
devel-
oped by Pe
~
na and Slate (2006). The test can be viewed as a Neyman
smooth test and it only relies on the standardised residual vector
(Rayner & Best, 1990). If the global procedure indicates a violation
of at least one of the assumptions, the components of the global test
statistic can be utilised to gain insights into which assumptions
have been violated (Pe
~
na & Slate, 2006). The advantage of such
procedure is that it reduces the oftentimes subjective assessment of
the validity of model assumptions when using existing graphical
techniques (Pe
~
na & Slate, 2006 ). We found that none of the as-
sumptions were violated in our linear model for bonding (Skew-
ness (
c
G
2
4
¼ 0.71, p ¼ 0.4); Kurtosis (
c
G
2
4
¼ 0.03, p ¼ 0.85; Link
Function (
c
G
2
4
¼ 0.24, p ¼ 0.62, Heteroscedasticity (
c
G
2
4
¼ 0.05,
p ¼ 0.83) and bridging social capital (Skewness (
c
G
2
4
¼ 1.44
p ¼ 0.23); Kurtosis (
c
G
2
4
¼ 2.3, p ¼ 0.13); Link Function (
c
G
2
4
¼ 0.72,
p ¼ 0.4, Heteroscedasticity (
c
G
2
4
¼ 0.58 p ¼ 0.45).
3
Table 1
Sociodemographic characteristics (given in % and N size) for participants who
completed main exploratory survey (n ¼ 209), and follow-up social capital survey
(n ¼ 96).
Main Follow-up
%(N)%(N)
Gender
Female 67 (139) 69 (67)
Male 33 (70) 31 (30)
Age
16e20 62 (130) 58 (56)
21e25 27 (57) 27 (26)
26e30 6 (13) 10 (10)
31e35 2 (5) 3 (3)
36e40 1 (2) 2 (1)
41 or more 1 (2) 2 (1)
Ethnicity
White 89 (187) 88 (85)
Asian/Asian British 4 (9) 2 (2)
Mixed/multiple ethnic groups 2 (5) 3 (3)
Black/African/Caribbean 2 (4) 4 (4)
Other ethnic group 1 (2) 2 (2)
I prefer not to say 1 (2) 1 (1)
Marital status
Single 72 (150) 68 (66)
Relationship e not co-habiting 15 (26) 15 (15)
Relationship e co-habiting 12 (31) 14 (14)
Divorced 0.5 (1) 1 (1)
I prefer not to say 0.5 (1) 1 (1)
Working status
Full Time 11 (22) 10 (10)
Part Time 46 (97) 44 (43)
Not Employed 42 (87) 42 (41)
I prefer not to say 1 (3) 1 (3)
Area of living
Urban 54 (112) 52 (50)
Suburban 29 (60) 32 (31)
Rural 18 (37) 16 (16)
Sexual orientation
Heterosexual 88 (183) 84 (81)
Gay 8 (16) 10 (10)
Bisexual 3 (9) 5 (5)
I prefer not to say 1 (1) 1 (1)
Fig. 1. The number of people participants regularly interact with on Snapchat
(n ¼ 209).
Fig. 2. The average frequency of use for various instant messaging services (n ¼ 209).
3
It's important to note that although the assumptions in linear model were not
violated, we didn't introduce interactions due to small sample size (n ¼ 97) used in
the follow-up survey. As Leon and Heo (2009) showed in a relevant simulations, we
would require a sample size of at least
*
n
*
¼ 208 for theoretical statistical power of
80% to detect the interaction in a mixed-effects linear regression model.
L. Piwek, A. Joinson / Computers in Human Behavior 54 (2016) 358e367362
We first investigated the extent to which socio-demographic
factors, and basic Internet/IM use, predicted the amount of
bonding social capital reported by participants. In the regression
analysis predicting bonding social capital (Table 2), those control
variables accounted for 12% of the variance with significant effects
of age (scaled
b
¼0.37, p < 0.05), although this effect disappeared
after adding Snapchat use intensity into the model. R
2
increased to
36% with the addition of Snapchat use intensity variable (scaled
b
¼
0.51, p < 0.01). This indicates that intensity of the Snapchat use is
positively associated with bonding social capital. When Snapchat
use intensity variable was added to the model, we also found a
significant effect for gender (scaled
b
¼0.41, p < 0.05). This in-
dicates that female participants reported greater bonding social
capital than their male counterparts.
We used exactly the same independent variables as predictors
in bridging social capital (Table 3). In the regression analysis pre-
dicting bridging social capital, those control variables also
accounted for 12% of the variance but with no significant effects.
The Snapchat use intensity variable was again positively associated
with bridging social capital (scaled
b
¼ 0.43, p < 0.01) accounting for
28% of the variance.
Fig. 3. Percentage of participants who (a) sent (n ¼ 209) received (n ¼ 114) specific snap, which (b) single person (73%; n ¼ 153) or (c) a group of people (27%; n ¼ 56) they sent it to,
(d) location (n ¼ 209), and (e) time of the day when they sent it (n ¼ 203).
L. Piwek, A. Joinson / Computers in Human Behavior 54 (2016) 358e367 363
4. Discussion
We report the findings of the first study on the patterns of
Snapchat use by means of a detailed survey and analysis of the last
snap sent and received. We also looked at more general aspects of
Snapchat use, as well as the association between intensity of
Snapchat use and social capital.
To start with, Snapchat was reported to be amongst the top three
IM services that respondents used most frequently e on a par with
Facebook Communicator and conventional text messaging (SMS).
Almost all participants also shared their photos on Facebook.Those
results were not surprising e as Quan-Haase and Young (2010)
pointed out, users tend to employ a broad range of digital commu-
nication tools that become integrated into a bundle of media use.
Snapchat was mainly used to communicate with a single person
rather than a group of people, and this person mainly includes close
friends, partners and family members. The overall number of
contacts people interacted with using Snapchat was relatively small
in comparison to Facebook. These results are in line with Roesner
et al. (2014) and Utz et al. (2015) who also found that users have
small and close social networks on Snapchat. Small networks
typically consist of people who are in our closest social circles
(Sutcliffe, Dunbar, Binder, & Arrow, 2012). Dunbar (1992, 1998)
hypothesized that small networks are easier to manage due to
inherent cognitive limitations in the number of people with whom
one can maintain stable social relationships. The importance of
small-network size was also highlighted in studies looking at the
differences in gratification from use of IMs and SNS. With IM, users
can engage in more intimate and private conversations, allowing
them to share their problems with communication partners more
easily, and allowing for better intimacy and a sense of connection
(Hu, Wood, Smith, & Westbrook, 2004). In contrast, SNS more
closely resemble a mix of e-mail and an online forum, where
messages are visible to the entire community. Quan-Haase and
Young (2010) argue that this is a key distinction in the use of SNS
and IM e that IM platforms such as Snapchat allow communication
partners to engage in deeper exchanges with affection, whereas
SNS such as Facebook tend to support the exchange of short mes-
sages via a public wall. Although messages can be exchanged pri-
vately via tools such as Facebook Communicator, this feature is
similar to e-mail and hence does not really support emotional
closeness.
The argument about the more intimate use of Snapchat was
further supported by our findings on the relationship between
Snapchat use and social capital. Similarly to Ellison et al. (2007),who
examined social capital on
Facebook, we found a positive association
between intensity of Snapchat use and social capital. However,
Snapchat appeared to be more useful for bonding rather than
bridging of social capital, which is opposite to what Ellison et al.
(2007) found for Facebook. Facebook, used with large social net-
works, serves to accelerate the intensity of relationships and lowers
barriers for participation in social groups, but appears to be less
useful in creating the close kind of relationships associated with
bonding capital (Vitak et al., 2011). Snapchat, used to forge small
networks of close relationships, is more a facilitator for the bonding
of social capital. As highlighted above in the context of IM vs SNS
differences, Snapchat offers a more intimate, private and ‘conver-
sation-like’ mode of communication, and therefore its intensity of
use is associated more with bonding rather than bridging of social
capital. Bridging social capital is typically associated with the
informational benefits of a diversified network of so called “weak
ties”.Those“weak ties” are loose connections between individuals
who may provide useful information or new perspectives for one
another but not necessarily emotional support (Granovetter, 1983;
Steinfieldetal.,2008). Bridging has been highly associated with
the intensity of Facebook use, where users typically communicate
with a large and diverse network of people, contrasting with small
networks in Snapchat. To summarise, a stronger association with
intensity of Snapchat use and bonding, rather than bridging, may
well stem from the fact that people use Snapchat mainly to enhance
a “strong” emotional ties with friends, partners and family, rather
than cultivate a large and “weak” networks, like in case of Facebook.
We also found that general demographics, Internet and IM use
were not significant predictors of bonding social capital, suggesting
that only certain kinds of uses of the Internet support the genera-
tion and maintenance of bonding and bridging of social capital
(Ellison et al., 2007). However, we found that bonding social capital
was predicted not only by Snapchat intensity of use, but also by
being female. This was surprising because majority of existing
studies have found no relationship between gender and bridging or
bonding social capital on SNS (e.g. Ellison et al., 2007; Lee et al.,
2014). However, women and men tend to have different styles in
valuing and sustaining relationships (Duck
& Wright, 1993; Eagly &
Steffen, 1984), so one expects there might be differences in the way
they use the Internet for interpersonal communication. Indeed,
previous research have suggested that males and females use the
same ICT for different purposes, with females reported using email
to maintain relationships while males used it to organise meetings
offline (Boneva, Kraut, & Frohlich, 2001). It may be that females are
more adept at using the affordances of new ICT, including Snapchat,
in order to build bonding social capital, although more research is
required to better understand the exact nature of this effect with
time-limited IM services.
Table 2
Relation between Snapchat use and bonding social capital.
Model 1: Controls Model 2: Controls þ
Snapchat intensity
Gender 0.342(0.204) 0.406
**
(0.175)
Age 0.369
**
(0.182) 0.136 (0.162)
Sexual orientation 0.112 (0.239) 0.162 (0.206)
Employment 0.105 (0.171) 0.134 (0.147)
Relationship 0.114 (0.185) 0.084 (0.163)
Hours of internet/day 0.022 (0.034) 0.028 (0.029)
Frequency of IM use 0.055 (0.136) 0.147 (0.123)
Snapchat Intensity 0.510
***
(0.092)
Constant 3.778
***
(0.662) 2.891
***
(0.590)
Observations 92 92
R
2
0.122 0.361
Adjusted R
2
0.049 0.299
F Statistic 1.668 (df ¼ 7; 84) 5.858
***
(df ¼ 8; 83)
Note: Gender was coded as 0 ¼ female, 1 ¼ male
**
p < 0.05;
***
p < 0.01.
Table 3
Relation between Snapchat use and bridging social capital.
Model 1: Controls Model 2: Controls þ
Snapchat intensity
Gender 0.189 (0.208) 0.136 (0.189)
Age 0.329 (0.186) 0.133 (0.175)
Sexual orientation 0.251 (0.244) 0.294 (0.222)
Employment 0.055 (0.175) 0.080 (0.159)
Relationship 0.247 (0.189) 0.081 (0.176)
Hours of internet/day 0.007 (0.035) 0.013 (0.031)
Frequency of IM use 0.228 (0.139) 0.058 (0.132)
Snapchat Intensity 0.427
***
(0.099)
Constant 1.552
**
(0.675) 0.810 (0.637)
Observations 92 92
R
2
0.120 0.281
Adjusted R
2
0.046 0.212
F Statistic 1.630 (df ¼ 7; 84) 4.061
***
(df ¼ 8; 83)
Note: Gender was coded as 0 ¼ female, 1 ¼ male
**
p < 0.05;
***
p < 0.01.
L. Piwek, A. Joinson / Computers in Human Behavior 54 (2016) 358e367364
When examining the specific incidence of the last snap that
participants shared via Snapchat, we found that users typically send
(and receive) a selfie e a self-portrait photograph. The practice of
taking and sending selfies has progressively developed with the
proliferation of digital media and now represents a recognised
element in culture, particularly amongst young people. Luders,
Proitz, and Rasmussen (2010) pointed out that smartphones have
become a common medium that contribute to this disciplined, yet
playful, visual self-authoring. While research on selfies has been
limited, especially in the psychological domain, existing studies
show that photos shared on social networking sites are a practical
and informative means of representing self-image, interpersonal
impressions, and identity management (Eftekhar, Fullwood, &
Morris, 2014; Saslow, Muise, Impett, & Dubin, 2012; Tosun, 2012;
Van Der Heide, D'Angelo, & Schumaker, 2012). One study argues
that making and editing a large number of selfies is associated with
Dark Triad traits, especially narcissism (Fox & Rooney, 2015). Fox
and Rooney (2015) argue that narcissists are prone to social com-
parison (Krizan & Bushman, 2011), and may present these edited
and optimised selfies on SNSs as a strategy to convey their
perceived superiority to others (Jonason, Lyons, Baughman, &
Vernon, 2014). However, all the studies mentioned above exam-
ined selfie posting strategies on Facebook, and there is a distinctive
lack of studies on making selfies with IM services. Snapchat is used
with much closer contact groups than Facebook or other social
networking sites where photos are typically shared with larger
groups of people. Hypothetically, the high intimacy level on Snap-
chat discussed above may have a different impact on the strategies
and motivation for sharing selfies on SNS such as Facebook. More
research is needed to better understand differences in users’ mo-
tivations and strategies for sharing their selfies on various social
media sites.
One of the intuitive and attractive features of Snapchat design is
that it affords the quick and effortless making of selfies, while the
ability to add short text comments and doodles makes it even more
playful. In fact, the playfulness of Snapchat is refl
ected in a number
of other results we obtained including: reason why participants
started using it (25% reported in different way that they use
Snapchat because it's enjoyable activity); positive mood reported
while sending their last snap (76% positive); as well as the playful
and funny type of content users reported sending. In previous
studies measures of playfulness have been used to establish the
degree to which user experiences fun when using the technology
(Davis, Bagozzi, & Warshaw, 1992; Moon & Kim, 2001; Van der
Heijden, 2003). For instance, Sledgianowski and Kulviwat (2009)
examined user adoption on Facebook, Friendster, and MySpace.
Playfulness has been identified as one of the most critical factor in
using those SNS, next to other factors such as critical mass of other
users, and trust and perceived ease of use. It's possible that the
narrative, conversation-like, and intimate nature of Snapchat, with
an interface that affords the easy exchange of short impressions,
becomes a preferred medium to playfully socialise in a more private
setting than public SNS such as Facebook. Snapchat is immersive to
use because you have to hold your finger on the screen to see the
content, and you only have one chance to view the received content
before it disappears therefore you need to stay focused when
receiving message. Arguably, the combination of self-destructing
images with an immersive interface that restricts the scope of
user interaction with the content makes Snapchat an instant
narrative vehicle that is similar to verbal story exchange.
We found that just three participants shared and received sen-
sitive content, specifically a naked or semi-naked photo of them-
selves, and they reported sending it to their partners. We also found
that security and privacy were not a user concern and that only 2%
of participants reported it as one of the reasons they use Snapchat.
This supports Roesner et al. (2014) finding that Snapchat's success is
not due to its security properties but because users find Snapchat to
be fun. This also goes against common misconceptions in the media
that self-destructing content lowers inhibitions and therefore in-
creases the chances of sexting.
One limitation of our study is that we only sampled a single snap
and therefore it's difficult to say how representative this single
incidence is in the broader use of Snapchat. While we get a clear
pattern of activity that is common across users (such as sharing
selfies), some behaviours might be underrepresented. However,
because content is unavailable to examine amongst Snapchat users,
the memory sampling method seemed an appropriate way for
obtaining accurate picture of Snapchat use. A larger sample size
could further refine these results. There are still a number of
unanswered questions that could be further explored by employing
more direct, qualitative methods of engaging Snapchat users, such
as focus groups. Do people ‘dose
’ themselves with snaps and how
they do it? Is the content that people share on Snapchat more
spontaneous than content shared on other IM services or SNS?
Perhaps content that people share on Snapchat is less self-censored
than content shared via other SNS? Or maybe Snapchat is a part of
media disposal culture e younger generations may lack a prefer-
ence for physical media and perceive instant media as more
desirable mode of communication. While young people seem to be
a key Snapchat user group, it would be interesting to compare the
differences between younger and older users. With Facebook now
releasing similar tools, and Instagram offering disposable content,
we may be facing a new chapter in how content is generated,
shared and stored e one that moves from default public sharing, to
the default removal of shared content.
5. Conclusions
Snapchat's rapidly increasing popularity among young age
groups rises a number of questions about how users utilise IM
services with time-limited and self-destructing content, and how
this relates to the use of other popular SNS such as Facebook. The
current study is amongst the first that investigate a detail patterns
of Snapchat use by surveying the very last incidence of snap that
participants send and receive. The study also examines the rela-
tionship between intensity of Snapchat use and social capital. Re-
sults indicate that Snapchat is mainly used as a playful mobile IM
service to rapidly communicate and share content, especially self-
ies, with a small group of close friends, partners and family. Such
“strong ties” oriented use is further reflected by a strong association
between Snapchat intensity of use and bonding, rather than
bridging, of social capital. It seems that popularity and patterns of
Snapchat use highlighted in our study might be a sign of a new form
of digital narrative rising amongst younger population of social
media users e a narrative that is achieved by seamless and playful
use of smartphones to capture and share content-rich moments
that cease to exist a second later.
Our study highlights how Snapchat become effortlessly
embedded within its users daily communication practises and is
currently the most popular form of IM in par with SMS and Face-
book Communicator. Although our study shows that privacy risk for
Snapchat users are less profound than indicated in the popular
media, parents and educational institutions should be aware of risk
associated with such services. Due to selfie-oriented use of Snap-
chat that pose a potential risk of unintended disclosure of sensitive
personal content, parents of the youngest users should be espe-
cially aware of Snapchat use. However, the fact that Snapchat offers
such playful form of communication could be also utilised by
educational institutions as a new mean of engagement. In addition
to helping young students populations, the use of Snapchat could
L. Piwek, A. Joinson / Computers in Human Behavior 54 (2016) 358e367 365
support variety of other populations, including community mem-
bers, and others who benefit from maintained ties. However, more
research is needed to fully understand how ubiquitous and
disruptive such use of self-destructing messaging is in the cultural
and socio-psychological context of the digital media use.
Acknowledgements
The authors would like to thank Dr Yvette Morey for her valu-
able comments and discussions regarding the results presented in
this manuscript.
Appendix A . Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.chb.2015.08.026.
References
Aharony, N. (2015). Whats App: a social capital perspective. Online Information
Review, 39,26e42. http://dx.doi.org/10.1108/OIR-08-2014-0177.
Boneva, B., Kraut, R., & Frohlich, D. (2001). Using e-mail for personal relationships:
the difference gender makes. American Behavioral Scientist, 45, 530e549. http://
dx.doi.org/10.1177/00027640121957204.
Brooks, B., Hogan, B., Ellison, N., Lampe, C., & Vitak, J. (2014). Assessing structural
correlates to social capital in Facebook ego networks.38,1e15. http://
dx.doi.org/10.1016/j.socnet.2014.01.002.
Buchanan, R. T. (2014). The snappening: Thousands of photos and videos released
through third party Snapchat app (13 Oct 2014). The Independent Online URL
http://www.independent.co.uk/life-style/gadgets-and-tech/the-snappening-
thousands-of-teenagers-photos-and-videos-released-through-third-party-app-
9790298.html.
Cole-Lewis, H., & Kershaw, T. (2010). Text messaging as a tool for behavior change in
disease prevention and management. Epidemiologic Reviews, 32,56e69. http://
dx.doi.org/10.1093/epirev/mxq004.
Cook, J. (2014a). Hackers access at least 100,000 snapchat photos and prepare to leak
them, including underage nude pictures (10 Oct 2014). Business Insider URL
http://uk.businessinsider.com/snapchat-hacked-the-snappening-2014-10.
Cook, J. (2014b). Snapsaved admits it was the source of 100,000 leaked snapchat
photos and videos (13 Oct 2014). Business Insider URL http://uk.businessinsider.
com/snapsaved-admits-it-was-source-of-leaked-snapchat-photos-2014-10.
Curtis, S. (2014). Instant messaging overtakes texting in the UK (13 Jan 2014). The
Telegraph Online URL http://www.telegraph.co.uk/technology/news/10568395/
Instant-messaging-overtakes-texting-in-the-UK.html.
Davis, F. D., Bagozzi, R. P., & Warshaw, P. R. (1992). Extrinsic and intrinsic motivation
to use computers in the workplace. Journal of Applied Social Psychology, 22,
1111e1132. http://dx.doi.org/10.1111/j.1559-1816.1992.tb00945.x.
Duck, S., & Wright, P. H. (1993). Reexamining gender differences in same-gender
friendships: a close look at two kinds of data. Sex Roles, 28, 709e727. http://
dx.doi.org/10.1007/BF00289989.
Duggan, M. (2013). Photo and video sharing grow online (28 Oct 2013). Paw Research
Center URL http://www.pewinternet.org/2013/10/28/photo-and-video-sharing-
grow-online/.
Dunbar, R. (1992). Neocortex size as a constraint on group size in primates. Journal
of Human Evolution, 22, 469e493. http://dx.doi.org/10.1016/0047-2484(92)
90081-J.
Dunbar, R. I. M. (1998). The social brain hypothesis. Evolutionary Anthropology, 6,
178
e190. http://dx.doi.org/10.1002/(SICI)1520-6505(1998)6:5<178::AID-
EVAN5>3.3.CO;2-P.
Eagly, A. H., & Steffen, V. J. (1984). Gender stereotypes stem from the distribution of
women and men into social roles. Journal of Personality and Social Psychology,
46,735e754. http://dx.doi.org/10.1037/0022-3514.46.4.735.
Eftekhar, A., Fullwood, C., & Morris, N. (2014). Capturing personality from Facebook
photos and photo-related activities: how much exposure do you need? Com-
puters in Human Behavior, 37,162e170 . http://dx.doi.org/10.1016/
j.chb.2014.04.048.
Ellison, N. B., Gray, R., Lampe, C., & Fiore, A. T. (2014a). Social capital and resource
requests on Facebook. New Media & Society, 13,873e892. http://dx.doi.org/
10.1177/1461444814543998.
Ellison, N. B., Steinfield, C., & Lampe, C. (2007). The benefits of facebook “friends:”
social capital and college students' use of online social network sites. Journal of
Computer-Mediated Communication, 12,1143e1168. http://dx.doi.org/10.1111/
j.1083-6101.2007.00367.x.
Ellison, N. B., Vitak, J., Gray, R., & Lampe, C. (2014b). Cultivating social resources on
social network sites: facebook relationship maintenance behaviors and their
role in social capital processes. Journal of Computer-Mediated Communication,
19, 855e870. http://dx.doi.org/10.1111/jcc4.12078.
FitzGerald, K., Seale, N. S., Kerins, C. A., & McElvaney, R. (2008). The critical incident
technique: a useful tool for conducting qualitative research. Journal of dental
education, 72, 299e304.
Flanagan, J. R. (1954). The critical incident technique. Psychological bulletin, 51,
327e358. http://dx.doi.org/10.1037/h0061470.
Fox, J., & Rooney, M. C. (2015). The Dark triad and trait self-objectification as pre-
dictors of mens use and self-presentation behaviors on social networking sites.
Personality and Individual Differences, 76,161e165. http://dx.doi.org/10.1016/
j.paid.2014.12.017.
Granovetter, M. (1983). The strength of weak ties: a network theory revisited. So-
ciological Theory, 1,201e233.
Gremler, D. D. (2004). The critical incident technique in service research.
Journal of
Service Research, 7,65e89. http://dx.doi.org/10.1177/1094670504266138.
Hawn, C. (2009). Report from the field: take two aspirin and tweet me in the
morning: how twitter, facebook, and other social media are reshaping health
care. Health Affairs, 28,361e368. http://dx.doi.org/10.1377/hlthaff.28.2.361.
Hofer, M., & Aubert, V. (2013). Perceived bridging and bonding social capital on
Twitter: differentiating between followers and followees. Computers in Human
Behavior, 29,2134e2142. http://dx.doi.org/10.1016/j.chb.2013.04.038.
Hu, Y., Wood, J. F., Smith, V., & Westbrook, N. (2004). Friendships through IM:
examining the relationship between instant messaging and intimacy. Journal of
Computer-Mediated Communication, 10(00). http://dx.doi.org/10.1111/j.1083-
6101.2004.tb00231.x.
Jonason, P. K., Lyons, M., Baughman, H. M., & Vernon, P. A. (2014). What a tangled
web we weave: the Dark Triad traits and deception. Personality and Individual
Differences, 70,117e119 . http://dx.doi.org/10.1016/j.paid.2014.06.038.
Juniper Research. (2014). Mobile messaging markets. Technical Report. URL http://
www.juniperresearch.com/press-release/mobile-messaging-pr1.
Kemp, N. (2013). What marketers should know about Snapchat? (13 Jun 2013).
Marketing Magazine Online URL http://www.marketingmagazine.co.uk/article/
1186152/marketers-know-snapchat.
Kim, D., Subramanian, S. V., Gortmaker, S. L., & Kawachi, I. (2006). US state- and
county-level social capital in relation to obesity and physical inactivity: a
multilevel, multivariable analysis. Social Science and Medicine, 63, 1045e1059.
http://dx.doi.org/10.1016/j.socscimed.2006.02.017.
Kosinski, M., Stillwell, D., & Graepel, T. (2013). Private traits and attributes are
predictable from digital records of human behavior. Proceedings of the National
Academy of Sciences of the United States of America, 110, 5802e5805. http://
dx.doi.org/10.1073/pnas.1218772110.
Kosur, J. (2014). What is the fappening? its a dirty moment captured in time (1 Sep
2014). Business2Community URL http://www.business2community.com/social-
buzz/fappening-diving-deep-scandal-0995227.
Krizan, Z., & Bushman, B. J. (2011). Better than my loved ones: social comparison
tendencies among narcissists. Personality and Individual Differences, 50,
212e216. http://dx.doi.org/10.1016/j.paid.2010.09.031.
Lambiotte, B. R., & Kosinski, M. (2015). Tracking the digital Footprints of personality.
Proceedings of the Institute of Electrical and Electronics Engineers (IEEE), 102,
1934e1939. http://dx.doi.org/10.1109/JPROC.2014.2359054.
Lee, E., Kim, Y. J., & Ahn, J. (2014). How do people use Facebook features to manage
social capital? Computers in Human Behavior, 36,440e445. http://dx.doi.org/
10.1016/j.chb.2014.04.007.
Leon, A. C., & Heo, M. (20 09). Sample sizes required to detect interactions between
two binary fixed-effects in a mixed-effects linear regression model. Computa-
tional Statistics & Data Analysis, 53, 603e608. http://dx.doi.org/10.1016/
j.csda.2008.06.010.
Luders, M., Proitz, L., & Rasmussen, T. (2010). Emerging personal media genres. New
Media & Society, 12,947e963. http://dx.doi.org/10.1177/1461444809352203.
Macmillan, D., & Rusli, E. (2014). Snapchat is said to have more than 100 million
monthly active users (26 Aug 2014). Wall Street Journal Blogs URL http://blogs.
wsj.com/digits/2014/08/26/snapchat-said-to-have-more-than-100-million-
monthly-active-users/.
Moon, J. W., & Kim, Y. G. (2001). Extending the TAM for a world-wide-web context.
Information and Management, 38,217e230. http://dx.doi.org/10.1016/S0378-
7206(00)00061-6.
Ogara, S. O., Koh, C. E., & Prybutok, V. R. (2014). Investigating factors affecting social
presence and user satisfaction with mobile instant messaging. Computers in
Human Behavior, 36, 453e459. http://dx.doi.org/10.1016/j.chb.2014.03.064.
Pe
~
na, E. A., & Slate, E. H. (20 06). Global validation of linear model assumptions.
Journal of the American Statistical Association, 101,341e354. http://dx.doi.org/
10.1198/016214505000000637.
Putnam, B. Y. R. (2001). Social capital measurement and consequences. Canadian
Journal of Policy Research, 2,41e51. http://dx.doi.org/10.1 017/S1474746403001 0 52.
Qualtrics. (2013). Qualtrics. URL http://www.qualtrics.com.
Quan-Haase, a., & Young, a. L. (2010). Uses and gratifications of social media: a
comparison of Facebook and instant messaging. Bulletin of Science, Technology &
Society, 30, 350e361. http://dx.doi.org/10.1177/0270467610380009.
Ramirez, a., & Broneck, K. (2009). ‘IM me’: instant messaging as relational main-
tenance and everyday communication. Journal of Social and Personal Relation-
ships, 26,291
e314. http://dx.doi.org/10.1177/0265407509106719.
Rayner, J. C. W., & Best, D. J. (1990). Smooth tests of goodness of fit: an overview.
International Statistical Review, 58,9.http://dx.doi.org/10.2307/1403470.
Roesner, F., Gill, B., & Kohno, T. (2014). Sex, lies, or kittens? Investigating the use of
snapchats self-destructing messages. In N. Christin, & R. Safavi-Naini (Eds.),
Financial Cryptography and data security (pp. 64e76). Springer Berlin Heidel-
berg. volume 8437 of Lecture Notes in Computer Science.
L. Piwek, A. Joinson / Computers in Human Behavior 54 (2016) 358e367366
Rusli, E., & Macmillan, D. (2013). Snapchat spurned $3 billion acquisition of fer from
facebook (13 Nov 2013). Wall Street Journal Online URL http://blogs.wsj.com/
digits/2013/11/13/snapchat-spurned-3-billion-acquisition-offer-from-facebook/
.
Rusli, E., & Macmillan, D. (2014). Snapchat fetches $10 billion valuation (26 Aug 2014).
Wall Street Journal Blogs URL http://www.wsj.com/articles/snapchat-fetches-
10-billion-valuation-1409088794.
Saslow, L. R., Muise, A., Impett, E. A., & Dubin, M. (2012). Can you see how happy we
are? Facebook images and relationship satisfaction. Social Psychological and
Personality Science, 4,411e418. http://dx.doi.org/10.1177/1948550612460059.
Shontell, A. (2013). The Truth about Snapchat's active users (The numbers the company
doesn't want you to See; 9 Dec 2013). Business Insider URL http://www.
businessinsider.com/snapchat-active-users-exceed-30-million-2013-12?IR¼T.
Sledgianowski, D., & Kulviwat, S. (2009). Using social network sites: the effects of
playfulness, critical mass and trust in a Hedonic context. Journal of Computer
Information Systems, 49,74e 83.
Statista. (2014). Age distribution of Snapchat users worldwide as of July 2014. Tech-
nical Report. URL http://www.statista.com/statistics/315398/snapchat-user-
age-distribution/.
Steinfield, C., Ellison, N. B., & Lampe, C. (2008). Social capital, self-esteem, and use of
online social network sites: a longitudinal analysis. Journal of Applied Develop-
mental Psychology, 29, 434e 445. http://dx.doi.org/10.1016/
j.appdev.2008.07.002.
Sutcliffe, A., Dunbar, R. I., Binder, J., & Arrow, H. (2012). Relationships and the social
brain: integrating psychological and evolutionary perspectives. British Journal of
Psychology, 103,149e168. http://dx.doi.org/10.1111/j.2044-8295.2011.02061.x.
Tosun, L. P. (2012). Motives for Facebook use and expressing “true self” on the
internet. Computers in Human Behavior, 28, 1510e1517. http://dx.doi.org/
10.1016/j.chb.2012.03.018.
UrbanDictionary. (2008c). Definition of ‘sexting’. URL http://www.urbandictionary.
com/define.php?term¼sexting.
Utz, S., Muscanell, N., & Khalid, C. (2015). Snapchat Elicits more jealousy than
facebook: a comparison of snapchat and facebook use. Cyberpsychology,
Behavior, and Social Networking, 18,141
e146. http://dx.doi.org/10.1089/
cyber.2014.0479.
Valenzuela, S., Park, N., & Kee, K. F. (2009). Is There social capital in a social network
site?: facebook use and college student's life satisfaction, trust, and participa-
tion. Journal of Computer-Mediated Communication, 14, 875e901. http://
dx.doi.org/10.1111/j.1083-6101.2009.01474.x.
Van Der Heide, B., D'Angelo, J. D., & Schumaker, E. M. (2012). The effects of verbal
versus photographic self-presentation on impression formation in facebook.
Journal of Communication, 62,98e116. http://dx.doi.org/10.1111/j.1460-
2466.2011.01617.x.
Van der Heijden, H. (2003). Factors influencing the usage of websites: the case of a
generic portal in The Netherlands. Information and Management, 40,541e549.
http://dx.doi.org/10.1016/S0378-7206(02)00079-4.
Vitak, J., Ellison, N. B., & Steinfield, C. (2011). The ties that bond: re-examining the
relationship between Facebook use and bonding social capital. In Proceedings of
the 2011 44th Hawaii International Conference on system sciences (pp. 1e10).
Washington, DC, USA: IEEE Computer Society. http://dx.doi.org/10.1109/
HICSS.2011.435.
Williams, D. (20 06). On and off the' Net: scales for social capital in an online Era.
Journal of Computer-Mediated Communication, 11, 593e628. http://dx.doi.org/
10.1111/j.1083-6101.2006.00029.x.
Wohlsen, M. (2015). Im too old for snapchat, which is exactly why it should be worth
$19B (23 Feb 2015). Wired Online URL http://www.wired.com/2015/02/im-old-
get-snapchat-exactly-worth-19b/.
L. Piwek, A. Joinson / Computers in Human Behavior 54 (2016) 358e367 367
