Vignettes

Vignettes were adopted due to their effectiveness in collecting qualitative data from younger
children (Barter & Renold, 2000). Vignettes addressed notions including over-disclosure, social
capital, self-presentation, and cyberbullying. A vignette about co-use was also to open a dialogue
about parents’ mediation behaviours. These were broken down into sub-notions to ensure that
nuances within these notions would not skew the data (Table 3). Children were asked to provide
advice for an imaginary child and outline whether they would model this behaviour, providing
explanations for their reasoning (‘Would you do the same? Why/why not?’; Table 3). Names of
imaginary children were consistent across all interviews.

Table 3. Vignettes and their related theoretical notions and sub-notions used in the child interviews

Theoretical notions Sub-notions Vignette

Over-disclosure Public Claire has a Facebook account. On her public profile she has her date of
birth, school, and the name of the town she lives in

Private Sam sends Sarah direct messages on Instagram telling her about his
secrets

Social capital Bridging David made a new friend on Facebook

Bonding Adam uses Instagram to keep in touch with his old friends from primary
school

Self- Azeem worries about posting photos on Instagram in case he does not
presentation get any likes

Cyberbullying Victimization Rachael read a status on Facebook that was about her and it made her
feel upset

Perpetration Craig posted a photo of Rebecca on his SnapChat story to make his
friends laugh

Co-use Sameer shares his SnapChat account with his mum

Procedure
The procedure of this study was conducted in accordance with the COREQ guidelines (Tong,
Sainsbury, & Craig, 2007). Prior to data collection, this study was submitted for a full ethical
review to the first author’s university research ethics committee. Ethical approval was granted
following this. This study also complied with the ethical guidelines of the British Psychological
Society. The lead researcher had a full Disclosure and Barring Service (DBS) check and
completed all of the interviews with children, parents, and teachers.

All interviews took place between May and July 2018 and were conducted by the lead
researcher: a female PhD student. As an ex-teacher, the lead researcher had experience in
safeguarding and child protection protocol as well as experience communicating with children,
parents, and teachers within a school setting. The lead researcher had prior experience of
qualitative methodologies and analysis; they also completed advanced training in preparation for
the data collection of this study. The lead researcher had a pre-existing relationship with three of
the schools recruited and therefore some children, parents, and teachers were familiar with them.
To mitigate biases or assumptions towards the lead researcher, first names were used throughout
the interview; this was utilized in particular to ensure that children did not feel as though they
were communicating with a teacher. All participants were reminded of the lead researcher’s role
as a PhD student and their interest in exploring children’s SNS use.

Most interviews took place during the school day within the school premises, two parents and
three children (two families) were interviewed in separate rooms in their homes on request.
Interviews were designed to take approximately 20 min in length to avoid difficulties fitting into
the school day. Interviews averaged at 19 min in length for parents and teachers, and 16 min in
length for children. Each interview was recorded using a digital recording device that was placed
on a table between the participant and the lead researcher. Participant consent for the interviews
to be recorded was obtained verbally prior to turning on the device. All recordings were
immediately transferred for transcription. All participants were assigned a unique numerical code
alongside their category (e.g., Child 1). The corresponding participant’s unique ID code and their
demographic information were stored within a password-protected file to later be added to the
transcription. All participants received a written and verbal brief and consent form prior to
commencing the interview, and a verbal and written debrief following completion.

Data analysis
All recordings were transcribed verbatim by the lead researcher (to ensure accuracy and depth of
familiarization with the data) into Microsoft Word documents, subsequently imported into
NVivo software. Inductive thematic analysis was used, in accordance with Braun and Clarke’s
(2006, 2013) framework, to elicit and interpret semantic patterns within relevant context. Within
NVivo, codes were constructed independently within the context of each individual transcription
to ensure that themes and sub-themes were not formulated prematurely (Braun & Clarke, 2013),
these were then semantically compared. Initial codes were compared contextually to identify
potential emerging sub-themes. Finally, these codes were compared across all participant groups
to identify larger themes (Braun & Clarke, 2013). These themes were combined to form broader
themes and sub-themes via thematic maps. These themes were then further analysed and refined
both via the repetition of the above process to ensure consistency and homogeneity (Braun &
Clarke, 2013) and through discussion with co-authors.