Revisiting the online privacy paradigm: Towards a nuanced understanding of social media users’ attitudes to information privacy and corporate uses of personal data

(1)

Understanding social media users’ attitudes to information privacy and corporate uses of data

1

Revisiting the online privacy paradigm

Towards a nuanced understanding of social media users’ attitudes to

information privacy and corporate uses of personal data

By Denitsa Dimitrova

Master Thesis

Erasmus Mundus Master’s in Journalism, Media and Globalisation. Student ID: 12257656

Supervisor: Jakob Ohme, PhD

Wordcount: 6.956

(2)

2

Abstract

This study explores social media users’ attitudes toward the status of their information privacy on

the social media platforms they use and specifically attends to user perceptions of corporate uses of

data. The findings confirm that user attitudes are diverse and not clearly associated with frequencies

of social media usage. From a service provider standpoint, this insight problematizes the ‘one size fits all’ approach to privacy policies with regard to their capacity to meet various user privacy preferences and values. Based on the data, the discussion identifies three distinct domains that could

be approached alternatively in ways that are better attuned to specific user perceptions. Firstly, it is

suggested that privacy on social media could be conceptualized as periodically negotiable to allow

for more flexibility around user consent for how data is collected and used, including an

institutionalization of consent as revocable. Secondly, the discussion considers the implementation

of stronger feedback mechanisms to inform users about how their data is exchanged with third

parties. Lastly, it is proposed that social media providers could give users options of privacy

policies, each offering distinct terms and conditions that resonate with different user dispositions.

Index

Abstract ... 2

Introduction ... 4

Theoretical Framework ... 6

Information Privacy: Data and Communications ... 6

Corporate uses of data ... 7

Nuances in users’ attitudes and research relevance ... 9

Methodology ... 9

Operationalizing levels of social media usage ... 9

(3)

3

Internet Focus Groups ... 11

Questionnaire ... 12

Data Analysis ... 13

Results ... 13

Information privacy attitudes: the self-management paradox ... 14

Attitudes toward corporate use of personal data ... 17

Patterns across levels of usage ... 20

Discussion ... 21

Information privacy: consolidating theory and practice ... 22

Corporate uses of data: enhancing transparency ... 22

Presenting users with options: payment plans ... 23

Limitations ... 23

Conclusion ... 24

Bibliography ... 26

Table 1: Reporting themes and categories in the data... 30 Appendices ... ..

Appendix A: Google Forms Survey: What is your level of social media usage? ... .. Appendix B: Sample composition: Summary of Google Forms responses ... .. Appendix C: Recruitment Poster ... .. Appendix D: Online Privacy Research Factsheet ... .. Appendix E: Informed Consent Form ... .. Appendix F: Focus Group Facilitator Guidebook... .... Appendix G: Focus Groups’ Stimuli ... .. Appendix H: Focus Groups Transcripts ... ..

(4)

4

Introduction

The Cambridge Analytica scandal which broke out in early 2018 revealed some deep-rooted

issues with how users’ personal privacy is institutionalized on social media. While numerous users

habitually disclose a bulk of their personal data on social platforms every day, they all too often

bypass their long and extremely technical privacy policies, thus remaining in the dark about their

terms and implications (Parks, 2017; Solove, 2013). In the aftermath of the Cambridge Analytica

controversy, evaluating users’ attitudes toward their information privacy status on social media could reveal valuable insights for ways to optimize social media services to the benefit of all

involved parties.

The breadth of academic knowledge on the subject of users’ privacy attitudes is extensive: some studies focus on users’ privacy concerns in relation to other users in their networks (Johnson,

et al., 2012; Staddon, et al., 2012; Biczok & Chia, 2013; Wang, et al., 2011); while others deal with

privacy concerns regarding third-party applications and institutional risks (Pingo & Narayan, 2019;

Chi, et al., 2018; Ur, et al., 2012; Ion, et al., 2011; Patil, et al., 2012; Fahl, et al., 2012; Schlegel &

Kapadia, 2011; King, et al., 2011; Chin, et al., 2012). Indeed, much of the literature has revealed

that individuals harbour concerns about their information privacy online (e.g. Chi, et al., 2018;

Braunstein, et al., 2011; King, et al., 2011; Schlegel & Kapadia, 2011; Ur, et al., 2012). Moreover, a

widespread lack of understanding of privacy policies’ terms and conditions disrupts users’ ability to identify some implications of data disclosure, such as what contexts of data sharing are appropriate

and safe (Solove, 2013; Ion, et al., 2011; Patil, et al., 2012; Criado & Such, 2015). However, even

those with greater knowledge of data collection and its implications commonly behave at odds with

self-reported privacy concerns and continue to use social media (King, et al., 2011). This apparent

privacy paradox, referring to the mismatch between users’ privacy attitudes and behaviours, points

to some fundamental flaws in the current privacy self-management paradigm, considered in this

(5)

5 This study proposes that if social media providers were to revisit institutionalizations of

information privacy so as to cater to users’ privacy concerns and values, it is important to attain

nuanced understandings of users’ perceptions of their information privacy status on social media.

Hence, the main question driving this research is: What attitudes emerge among social media users

towards the status of their information privacy on the social media platforms they use? Better

insights into how users feel toward their personal information privacy could yield valuable feedback

for developers seeking to optimize platform services.

Additionally, this study explores how users’ specific perceptions of corporate practices of using personal data play into general attitudes to online information privacy. So far, studies have

found that users hold various attitudes to possible uses of their personal data, such as its role in

building personalized content (Chi, et al., 2018; Pingo & Narayan, 2019; Ur, et al., 2012). With a

view to generating a nuanced understanding, this paper explores corporate uses of data as a possible

influence on user attitudes by asking: What are social media users’ feelings towards practices of

corporate use of users’ personal data?

Finally, the study introduces a comparative lens on users’ perceptions by asking: Are there

any distinct patterns of attitudes emerging among users at high and low levels of social media usage? This additional layer of analysis adds to platform providers’ understanding of their

audiences by seeking to delineate any specific clusters of user concerns or beliefs about privacy that

relate to individuals’ frequency of social media use.

The paper is structured as follows: the first section lays out the theoretical framework and

elaborates on the central notions of information privacy, personal data, and corporate uses of data,

thus illustrating this study’s significance. Then, some methodological issues are discussed before

the results are presented and followed by a discussion, which in turn informs some empirical

(6)

6 specific privacy concerns. Finally, the paper addresses the research limitations and draws a

conclusion.

Theoretical Framework

Information Privacy: Data and Communications

Paving the way for the proposed research is a comprehensive body of work on information privacy.

This paper accepts that privacy, in general, constitutes an inalienable right of individuals: Clarke

(1997) has conceptualised it as a right to sustain a personal space free from the interference of

organizations and people, while Westin (1970, as cited in Altman 1975, p.17) has seen it as a right to the self-determination of when, how, and how much personal information is communicated to

others. Information privacy, specifically, results from the conceptual marriage of the dimensions of

communications and personal data privacy and encompasses: 1) one’s right to communicate freely from the monitoring of one’s communications; 2) one’s right to have one’s personal data not made automatically available to other parties, and even where one’s data is legally possessed by other parties, 3) one’s right to have a degree of control over its usage (Clarke, 1997). Additionally, this study accepts that at the heart of conceptualisations of information privacy lie some essential

principles about offline privacy – that it is culturally conditioned, serves particular personal and

societal functions (Westin, 1970). But also, crucially, that privacy is an interpersonal regulatory

boundary that is subject to ongoing negotiation and is thus dynamic and dialectic as opposed to

fixed and static (Altman, 1975).

Personal data in this context is understood as

“

a digital trail of data or “electronic

breadcrumbs” [people leave] in their daily routines; through electronic fare cards, traffic and street cameras, internet surfing, paying for purchases with credit/cash cards, creating text messages or by

making calls, and so on”(Pingo & Narayan, 2016, p.3). On social media, personal data can be

(7)

7 geotags, family pictures, places and dates of their holidays, and so on (Dubrovitskaya, 2014).

Personal data on social media has been seen as existing on five dimensions: service data, disclosed

data, entrusted data, behavioural data, or derived data (Schneier, 2010). Rather than zooming in on

either one of these dimensions, this research has sought to create space for users’ discussions to

organically reveal their personal perceptions of data significance.

Notably, the current paradigm of information privacy online is premised on the notion of

privacy self-management, which dictates that it is users’ personal responsibility to oversee the

management of their privacy by means of certain rights granted to them (Solove, 2013; Pascalev,

2017). However, the paradigm has been scrutinized for having some deep-rooted shortfalls:

structurally, it offers privacy policies that ask people to give informed consent about disclosing their

data before users can know how technology will allow for data to be used in the future, thus actually

precluding users from having meaningful control over data (Solove, 2013; Pascalev, 2016).

Cognitively, users are treated as rational decision-makers who can evaluate the trade-offs of privacy

choices and give informed consent to disclosing their data despite evidence that users’ behaviour can be skewed by predictable systematic biases triggered by minor alluring incentives to disclose

data (Adjerid, et al., 2018; Patil, et al., 2012; Solove, 2013). These issues lay the grounds for

arguments that the nature of the exchange between users’ data and social platforms’ services is inherently unfair – a belief that has been expressed by users holding that the outcomes of this

exchange for both parties are asymmetric, with users’ data being a more valuable commodity than the free services provided by social media (Ashworth & Free, 2006).

Corporate uses of data

Exploring users’ attitudes toward corporate uses of personal data is pertinent to the study of information privacy attitudes because the notion of data is implicit in information privacy. This

study seeks to grasp users’ attitudes towards some practices in the data industry, whereby rich and comprehensive user datasets or ‘commercial profiles’ are utilized for commercial ends (Brown,

(8)

8 2014; Clarke, 2016; Esposti, 2014). Ultimately, this line of inquiry concerns the question of whether

users feel that their intrinsic information privacy rights outlined above are respected – that is, if they

feel that the social media platforms they use allow them 1) to communicate freely of monitoring or

2) to have meaningful control over their data and 3) its exchange. These questions beg the notion of

dataveillance (data surveillance), referring to the use of personal data systems in the monitoring of

individuals’ communications or actions, namely: recorded observation, identification and tracking, analytical intervention, and behavioural manipulation such as targeted advertising (Clarke, 2016;

Esposti, 2014). That is because dataveillance practices have the potential of clashing with the

aforementioned user rights, contingent upon whether users feel able to opt in or out of these

corporate practices or to regulate their privacy boundaries in the context of the latter’s occurrence

(Altman, 1975). Hence, this study seeks to explore how users feel toward the micro-level aspects of

dataveillance: namely, recorded observation, analytical intervention and behavioural manipulation,

in their concrete forms of personalized content on social media (e.g. targeted advertising) or having

their personal conversations recorded. While these practices are not in and of themselves

undesirable or harmful – they could facilitate great economic success in marketing arenas or the

realization of governmental aims (Clarke, 2016) – they entail some macro-level risks such as

amplifying societal stratification or “social sorting”, if data becomes a factor in employment,

insurance opportunities, criminal case decisions, the price availability of various products online,

and so on (Brown, 2014). While these possible ‘big picture’ implications justify the endeavour to

zoom in on users’ attitudes to corporate handling of user data, a more tangible understanding of users’ micro-level attitudes to their data could better inform corporations’ policy-making with a view to respecting those users’ right who might prefer to opt out of dataveillance practices (Irion & Helberger, 2017).

(9)

9

Nuances in users’ attitudes and research relevance

The comparative aspect of this study stems from the recognition that people are likely to

differ in terms of the value they place on various types of personal data (Patil, et al., 2012; Ur, et al.,

2012; Chi, et al., 2018; Pingo & Narayan, 2019). An attentiveness to patterns of understanding

across users’ levels of social media usage could illustrate possible conditions under which user attitudes differ. Indeed, users’ frequency of engaging on social networks has been identified as one

factor influencing privacy concerns positively (Staddon, et al., 2012: 8). While the relationship

between users’ privacy knowledge and online behaviour has been explored both quantitatively and qualitatively (King, et al., 2011; Lupton & Michael, 2017), the comparative lens between users’ sets

of concerns and attitudes across high and low levels of social media usage has not yet been applied

in qualitative terms. This study aims to contribute to the existing pool of knowledge on privacy by

filling this gap

This study’s significance is broadly rooted in its recognition of a discrepancy between the

theoretical formulation of privacy as shifting on the one hand, and the empirical rigidity of privacy

policies, on the other (Altman, 1975; Solove, 2013). The research contribution to identifying new

domains for social media providers to consider when devising more user-conscious privacy policies

underpins the paper’s social significance.

Methodology

Operationalizing levels of social media usage

Social media usage has been measured on various scales which usually belong to one of four

general methods of assessment based on one of four possible indicators: time spent online,

frequency of online engagement, reported attitudes or experiences (Rose, et al., 2013; Savci &

(10)

10 (MTUAS) template which measures frequencies in the number of uses in a particular time period, a

short survey consisting of 9 items was constructed with Google Forms (see Appendix A). The

respondents were asked to share basic demographic data like age and gender to enrich the

subsequent analysis and were subsequently asked to select the social media platforms they are

registered on from a list of options including Facebook, Instagram, Twitter, Snapchat and Other.

The questions inquired about respondents’ frequency of checking their accounts for each of the

available options; and the frequencies of communicating with other users on each platform. The

response options for frequencies of usage exist on a 10-point scale with 1 being ‘never’ and 10

being ‘all the time’, thus allowing for a median score calculation.

Drawing a precise line between high and low levels of usage is challenging and a decision

was been made to do a median split, thus creating a binary by splitting the entire usage range in half

and clumping users on either of its respective ends in the same category (Iacobucci, et al., 2015).

This was decided with a view to minimizing the risk of ending up with unequal group sizes, which

might have compromised the viability of collecting sufficient data. To classify participants as low-

or high-level users, the survey question about users’ frequency of engaging (posting, sharing, liking,

commenting) on social media served as a metric to calculate a median score, as it was deemed to

reflect the most tangible example of disclosing personal data. The median score amounted to 3,62,

thus splitting participants in groups of 10 low-level users and 7 high-level users. Notably, while

there are nuances within each either category (i.e. participants with a median of 4 likely engage in

significantly different patterns than those with an average of 10), the scores are taken as an

orientation mark to inform group formation rather than as a rigorous reflection of users’ true

ranking.

Recruitment and Group Composition

Recruitment occurred on social media (see Appendix C) and followed a purposive sampling

(11)

11 on a level of social media usage (Bryman, 2012: 422; Braun, 2013: 59). Out of the 27 individuals

who completed the Google form, 17 responded to the focus group invitations. The final sample was

comprised of 15 nationalities aged between 24 and 30, likely representing an educated cosmopolitan

stratum of society (see Appendix B). Following the snowball sampling technique, some individuals

recommended prospective participants from their networks, which might have resulted in some

participants’ acquaintance although care was taken to group them separately where possible (Bryman, 2012: 424; Harrell & Bradley, 2009: 80). Given disagreements on the criterion of

participants’ unfamiliarity, it was not deemed this would undermine the data’s credibility (Harrell &

Bradley, 2009: 80).

Four groups consisting of four to five participants were formed. Although the original

calculation of individuals’ usage median score was affected following some participants dropping out, the groups’ formation remained strategic insofar as users’ usage level is an accounted for factor

in the discussions. One group was comprised of two low-level and two high-level users; one was

formed with three high-level and one low-level users; one consisted of two high-level and three

low-level users; and one was made up of four low-level users. As some major themes continued

recurring throughout the four discussions, it was deemed that optimal levels of theoretical saturation

were achieved (Bryman 2012: 421, 426).

Internet Focus Groups

Focus groups lent themselves as the most suitable method for the study, given 1) its emphasis on

how information privacy is understood by users with similar patterns of social media activity, and 2)

that focus groups cultivate an environment where dynamic group interaction lends itself as the

source of data (Harrell & Bradley, 2009: 6). This method is appropriate for exploring attitudes and

understandings, which are the foci of the study’s research questions (Braun, 2013: 45). Focus

groups yield in-depth data which allows for exploring constructions of meaning, and are suitable for

(12)

12 (Harrell & Bradley, 2009: 10). While the findings of focus groups research are not generalizable

beyond the groups conducted, the groups’ composition was strategic insofar as it was anticipated

that their distinct conglomerations of participants on varying usage levels would promote unique

conversations (see Appendix I), thus laying out the framework for subsequent analysis (Harrel &

Bradley, 2009: 10, 83). Hence, the research design aimed for heterogeneity across groups (Braun,

2013: 56). Given that research was conducted amid the global Covid 19 pandemic, the groups were

conducted as an hour-long chat-based conversation on a Google call, with questions posed by the

moderator and responses typed in by participants (Krueger & Casey, 2000: 189). Although this

method modification might have limited the moderator’s ability to get a deeper sense of the group

dynamics, some audiences have been found to respond well to this process (Krueger & Casey,

2000: 190). A decision was made to avoid a speech-based internet discussion due to the risk that

data collection could have been compromised in instances of technical errors such as low network

speed, overlap of participants’ responses and muffled microphone sounds.

Questionnaire

The discussions proceeded by obtaining participants’ consent, familiarizing them with their rights,

and asking 8 main questions, some followed by context-appropriate prompts (see Appendix F).

Probing and occasional member checks served as quality control techniques during the discussions

(Carey, 2015: 277). The questionnaire reflected the research questions, for example by asking

participants about their feelings toward personalized content and targeted advertising on social

media. The second part of the discussions included a short interactive segment to encourage

participants’ reflections: through screen-sharing, they were shown some data-related privacy features on Facebook and Instagram – the most used platforms as demonstrated by the Google

Forms survey (see Appendices B and G). These features aimed to explore participants’ levels of

familiarity with data-related privacy settings and their perception of their transparency and

(13)

13

Data Analysis

The data was analysed by means of a thematic analysis approach that is suitable for the

chosen method (Braun, 2013: 50). The group discussions were coded manually and inductively

following three stages of coding: open, axial, and selective (Corbin & Strauss, 1990: 12). In the

open coding phase, the data was coded for various emerging issues; during axial coding, the

first-cycle codes was revisited and merged into categories with the research questions in mind, where

irrelevant categories were dropped out; in the selective coding stage, second-cycle categories were

clustered into overarching themes that speak directly to the research questions (Corbin & Strauss,

1990). Each sign in the codes of participants’ names reflects the following order: the sequence of

the discussion, the user’s position in the discussion, and their level of usage marked as H for high and L for low (e.g. 1P2H indicated the second participant from the first focus group on a high-level

of usage).

Results

The data illuminates some recurring themes in users’ information privacy attitudes synthesised by

theme and category in Table 1. The following sections are structured topically around the driving

research questions, namely: What attitudes emerge among social media users toward the status of

their information privacy on the social media platforms they use? What are social media users’

feelings towards practices of corporate use of users’ personal data? Are there any distinct patterns of attitudes emerging among users at high and low levels of social media usage? Notably, during the

discussions, participants most often organically referred to Facebook as the platform for their

(14)

14

Information privacy attitudes: the self-management paradox

In line with classic conceptualisations of privacy, participants alluded to the shifting nature of their

privacy needs: some participants explicitly reflected on this, and virtually all expressed having

adjusted their privacy settings at least once in the past (Altman, 1975). Occasionally, participants

voiced their perception that default privacy settings do not grant optimal privacy protection:

2P3L: …the default [settings are] way more public than what id see as safe

While some participants did not mind adjusting their privacy settings, many found this to be an

annoying or time-consuming task:

1P4H: … To be more private you have to spend a lot of time in settings, which I find some times very

annoying

When specifically questioned about their perceptions of their communications privacy status on

social media, most participants expressed feeling their conversations are not private, in addition to

frequent mentions of feeling they had no control over this:

1P4H: I dont feel [that my conversations] are private at all. Although for instance whatsapp has

encrypted chats, the fact that it is a Facebook company, make me feel that they still have the

information from the private communications, and probably it is anlaysed by machines for marketing purposes, also, maybe political

This sense was expressed alongside frequent comments about feeling watched, observed or listed to

by platform providers and third-party applications on platforms, which was characterised negatively

as invasive, disturbing and creepy:

3P4L: It's terrible, I always see ads for things I was discussing with a friend or something earlier

(especially on Facebook) … it is a source of distress for most users, to feel like you're being listened to at all times

(15)

15 When asked about their personal data privacy, users again painted an overwhelming picture of a

perceived lack of control:

2P4H: [My data] isn't [private]. Almost all the income of social media platforms are generated by

extracting information about me.

However, similarly to previous findings, differential attitudes emerged towards the perceived lack

of privacy: some participants were unsettled, while others seemed unconcerned insofar as the data

they disclosed remains unidentifiable (Patil, et al., 2012; Ur, et al., 2012):

1P3L: I like to think … that the data they are collecting on me is just another point in a cloud and

that kind of anonymity makes me feel safer… i dont care if the details of my network and activity are used to develop algorithms, but i would be mad if someone used my picture on an ad or something… there is personal data that is about me as an internet user and then theres personal data that could be used to pick me out of a crowd of other humans

Besides the differential attitudes to identifiable and non-identifiable data, another theme which

participants reflected on differently was that of the very paradigm of online privacy (Solove, 2013).

Specifically, two distinct attitudes emerged around the notion of responsibility to protect users’

information privacy. Some participants, conforming to the privacy self-management perspective,

internalized personal agency over their data footprint (Solove, 2013; Pascalev, 2017):

2P1H

:

…I still try to keep in mind that it is my choice, whether or not to publish some information,

upload a photo or share an article. Even to put my like under a post is a personal desicion and in the moment I do it it kinda accept what comes with it as far as I cannot avoid the conceque[n]ces.

By contrast, others regarded the expectation of users to self-regulate their privacy to be inadequate

for a few reasons. For example, some participants referred to their inability to make informed

decisions about disclosing their data, given their overall lack of understanding of what data is

(16)

16 2P3L: Supposedly i have a choice to agree to share [my data] or not, but i have absolutely no idea

what happens to them afterwards.

2P2H: As I don't know how and what is used of my data, I don't really know what would be most

important to protect

Additionally, some participants illustrated scenarios where they found themselves cognitively

unable to make informed privacy self-management decisions due to the unpredictability of how user

data could be handled in the future as a result from technological innovation (Tucker, 2019; Solove,

2013). This additionally muddled the notion of consent and induced a sense of unease in

participants (Solove, 2013):

3P3L: I don't like the fact that one day they can use my information for any reason, just because

when downloading the applications, I had to agree to this clause.

Users perceived more issues pertaining to platforms’ structural setups, such as websites’ leniency

toward weak default foundations for privacy protection:

4P1L: I disagree with 4P4L … the default privacy setting makes your information more prone to

economic/political etc. exploitation… while I agree that over-/undersharing is up to us, the default setting nevertheless creates a greater foundation for misuse of data

An additional privacy risk discussed by users relates to what has been conceptualised as privacy

interdependence, whereby individuals’ personal privacy is dependent on others’ online behaviour

(Biczok & Chia, 2013). In this light, some participants felt that nothing could guarantee that social

media companies would not end up having their personal data:

3P1L: Oh and also: if you live in a big city it is unavoidable you end up on the social feed of soem

tourist (accidently) making a picture of you.

To sum up, a range of attitudes emerged towards users’ information privacy status in the forms of

(17)

17 communications as something not entirely private and often beyond their control. While some users

did not mind this as long as their data remained personally unidentifiable, others voiced stronger

concern. Notably, disagreements surfaced regarding the notion of responsibility to protect users’

information privacy: one position upheld the personal responsibility trope, supported by references

to user consent and informed decision-making; while the other alluded to corporate responsibility

by pointing to some inadequacies in the self-management paradigm, which prevent users from

having meaningful control over managing their personal privacy (Solove, 2013; Pascalev, 2017).

Attitudes toward corporate uses of personal data

Participants’ attitudes towards corporate practices of recorded observation, analytical intervention

and behavioural manipulation such as targeted advertising also varied considerably (Esposti, 2014).

Many characterised targeted advertisements as ‘creepy’, alongside reflecting on feeling ‘observed’,

but a few users admitted to enjoying the convenience of personalized content:

3P4H: [Targeted advertising] feels intrusive

3P4L: It's terrible, I always see ads for things I was discussing with a friend or something earlier (especially on Facebook)

4P4L: …targeted ads… are helpful to be honest, and I get many whitepapers and event

recommendations based on my interests in marketing even if I haven't done relevant Google search.

There was a widespread perception of a lack of transparency on the end of media corporations

regarding what types of data are collected and how they are used afterwards:

1P3L: … the part that I have feelings about is the data they collect that you cant really do anyting

about

2P4H: …there is no way I can really know [how my data is used]. I don't know what info about me is

bought and sold. There is no way to find out.

(18)

18 3P1L: I don't really know what my data is for them. What is it that they want to know about me? 3P4H: I get the sense that a lot of it is just incorporated into some invisible database for unknowable

ends and that it could be retrieved at any point down the line.

The perceived obscurity around corporations’ business tended to persist even after walking

participants through the data-related privacy features on Facebook and Instagram (Appendix G).

While a handful of users considered these features to be effective, many remained unconvinced that

they offer users any meaningful control over their data:

1P1L: It's nice to see you get at least some control over facebook.

3P4L: It always annoys me however, that Facebook and Instagram seem to (at least based on what

we saw when Denitsa was screen sharing) give you options with privacy and Instagram even shows you a list of your interests, despite this I don't really see any actual transparency on the matter of these website using our information

3P3L: The fact that I can turn off some adds doesn't mean that I can control the usage of my data

beyond the adds

Despite being specifically designated to provide transparency and user control, the Facebook

settings were often considered to only give superficial control and limited transparency by allowing

users to customize their personalized content and allowing them to see what ‘interests’ they are

assigned based on online activity. Meanwhile, Instagram’s features were seen as not allowing any

user control.

Thinking about the large-scale implications of corporate handling of user data, some

participants expressed strong concern over how data use could amplify the phenomena of filter

bubbles or echo chambers:

2P4H: I am worried that people are getting locked in their echo chambers. The more they gather, the

(19)

19

worry about political ads. There have been prior scandals where Pro-Publica found that facebook allowed advertisers to specifically target anti-semites

4P1L: I would say owning this information on users has vast negative economic and political

implications. Next to all the democratizing positive ones… [For example] Facebook, cambridge analytica, political targeting/manipulation, consumer capitalist effiency increase, throwing out a couple of big ones.

Some participants explicitly voiced concern over the effects of corporate behavioural manipulation

such as targeted advertisements (Esposti, 2014):

1P4H: What we see on the internet shapes our opinion on many things, and if most of the things we

do see on the internet are designed for us personally, it means that our opinion is indeed shaped in a way that suits companies/corporations/governments

Participants’ acknowledgement of such broader societal effects of corporate uses of data indicated

their understanding of data as a valuable commodity. For some, this insight coupled with a

perception of no transparency on the end of corporations, pointed to an unfairness in the exchange

between users and platforms:

3P1L: And [corporations] also do not tell you what they want to do with your data: normally in any transaction you know what you pay for and you know exactly how much you pay for it. Byt social media is not transparant in that aspect at all

Specifically, corporate obscurity was perceived regardinghow much and what data is traded

between service providers and third parties, as well how this data informs corporate analytical

intervention such as the labels assigned to users (Esposti, 2014):

2P4H: … I don't feel like there has ever been a very sincere attempt at transparency on what is being

gathered. For example, you can download an archive of your facebook data, but they also buy info on everyone from third party brokers and you cannot know what they have gathered on you that way.

(20)

20

So info about me is bought and sold pretty much freely with not much awareness on what exactly it is on my part.

One participant was the exception to this sentiment and expressed agreement with the fairness of the

user-platform exchange by holding that personal data is a fair price to pay for the range of

communication and entertainment services provided on platforms.

These contrasting views were generally reflected in the two camps which formed around the

question of paying for stronger privacy protection on social media. Many participants expressed

willingness to enter paid contracts with providers, specifically listing the following as aspects worth

paying for: encryption of messages, the guarantee that providers do not use more data than the basic

data needed to provide the service, breaking up the monopoly of big social media corporations over

user data, and an advertisement-free experience. These participants generally had a wish for an

alternative social media experience:

3P4H: I think [contemporary social media] makes for a dystopian world where data is mined as an

asset that profits/advertising revenues for a few corporations at the user's expense. So I just wish some alternatives would arise soon!

Simultaneously, others expressed unwillingness to pay for stronger privacy for different reasons:

some held that privacy is an intrinsic right that should be respected without the need to pay, others

distrusted that providers would respect these contracts, and yet others simply did not deem

information privacy to be worth paying for.

To conclude, a range of attitudes emerged toward social media corporations’ practices of

handling user data– some users found these helpful and others felt their privacy was violated. A re-emerging theme was users’ perception of obscurity around how user data is handled by

corporations.

Patterns across levels of usage

(21)

21 between users’ level of social media usage and their self-reported perceptions. The lack of clear patterns of understanding emerging from users’ usage level applies to both lines of inquiry in this study: usage frequency did not seem to dictate patterns in participants’ attitudes to their information

privacy, nor perceptions of corporate uses of data. On the contrary, diverse privacy attitudes

characterised both ends of the usage spectrum, with some frequent users expressing strong concern

for their information privacy thus confirming the privacy paradox (Solove, 2013):

3P2H: It's kind of scary. I don't like to talk about a subject and then see an advertisement about

exactly the same thing on my social media! I feel "violated", without any privacy.

At the same time, some low-level users expressed general indifference or even willingness to

support social media providers by sharing their digital footprint:

1P3L: the suggested content that I enjoy seeing has to be built on data from somehwere

In summary, participants’ attitudes to both their information privacy status on social media and to corporate uses of personal data were diverse, with both negative and positive sentiments

associated with each, although usage level was not a predictive force of distinct attitudes in the

context of the discussions. Some notable themes to emerge from the discussions were the

disagreement over who bears responsibility for protecting users’ information privacy, the perceived

lack of a corporate transparency about the collection and handling of user data, and some

differential user attitudes to identifiable and non-identifiable data.

Discussion

The focus groups outlined a widespread feeling that users’ personal data was beyond their scope of control on social media – a finding which diverges from both 1) Clarke’s (1997) conceptualization

that individuals should have a right to sustain a personal space free from the interference of

organizations, and 2) Westin’s (1970) theorization of privacy as a right to the self-determination of

(22)

22 interrogated from the lens of the data’s significance to social media providers seeking to attune

institutionalizations of privacy to users’ distinct values and perceptions. Three recommendations are

made based on the data.

Information privacy: consolidating theory and practice

The shortfalls of the privacy self-management paradigm as identified by participants lay grounds for

the scrutiny of privacy institutionalization on social media. Specifically, the configuration of user

consent as static obstructs users’ ability to exercise their privacy self-management, especially

long-term as user data begins to be used in novel ways, yet user consent is not requested anew (Tucker,

2019; Solove, 2013). In the reported experience of many participants, their theoretical privacy rights

– to communicate freely from corporate monitoring, to not have their data automatically available to third parties, and to have control over their data’s usage – were not respected in the context of social

media (Clarke, 1997). To circumvent this issue, social media providers could begin to consider new

theoretical and practical approaches to online privacy, ones premised on the idea of privacy as

dynamic and subject to periodical negotiation, that would allow users more flexibility around

consent to data collection and use, as well as its withdrawal and negotiation (Altman, 1975).

Corporate uses of data: enhancing transparency

The conversation about the need for social media mechanisms that demonstrate more concretely and

personally how privacy settings influence users’ information privacy is not new (King, et al., 2011;

Schlegel & Kapadia, 2011). Nonetheless, following the discussions’ revelation of participants’

considerable dissatisfactions with the effectiveness of Facebook’s and Instagram’s data

transparency features, it is proposed that social media providers might benefit from revisiting these

structures. The recommendation for developing more thorough, continuous feedback loops to

inform users about their data exposure to third party applications, seems to hold value in light of the

present study (Schlegel & Kapadia, 2011). Additionally, it is hereby suggested that users’

(23)

23 implementation of archives of data exchanges. For example, the availability of lists of third parties

like data brokers whom a user’s data has been sold to or purchased by, alongside the option to

permit or forbid certain parties from trading a user’s dataset, could be a starting point to giving users more knowledge and control over their information privacy.

Presenting users with options: payment plans

The sheer diversity of user attitudes toward information privacy online hints that a ‘one size fits all’

approach to privacy policies is likely ineffective in accommodating the full variety of user

preferences. Some participants’ wish for alternative forms of privacy protection points to the

uncharted territory of different privacy policies presented to users to pick from, each offering

distinct terms, conditions, and guarantees. For example, introducing a monetary value to the

user-platform exchange, where the former would pay for stronger privacy and the latter would adjust the

service accordingly, might benefit both parties: users would find their privacy needs met, while

providers would continue making revenue directly from users while bypassing third parties that

monetize data. Additionally, paid social media plans have the potential of appeasing users who

believe that privacy policies make for an unfair exchange where the values of the inputs of both

parties, e.g. user data and free services respectively, are unequal. Paid plans could entail some

services which the participants expressed willingness to pay for, such as an advertisement-free

experience and the guarantee than only minimal user data is collected. Finally, social media

providers who are sceptical toward real user interest in this possibility could turn to those users who

expressed unwillingness to enter these contracts as an assurance.

Limitations

The research limitations exist along three main lines: sample limitations, limitations of internet

focus groups, and some methodological issues. Future research exploring user attitudes to

(24)

24 background and specifically education level, which might have been an unexamined force in the

present research. It is possible that educational background might influence individuals’ general

understanding of online information privacy and their preferences. Secondly, the internet-based

nature of the discussions is not ideal and might have facilitated a less immersive and lively

discussion than a face-to-face context would permit. Thirdly, users’ levels of social media usage

were not determined too rigorously in this study, but only give a sense of interpretive orientation in

the analysis. Future studies seeking to explore the relationship between usage frequency and

attitudes could deploy more methodologically rigorous ways for estimating usage levels in order to

draw more meaningful comparative conclusions. Additionally, the discussion is mostly based on

user reflections on Facebook but scarcely any other media, since the focus groups were not strictly

controlled regarding that aspect. To gain more specific understanding of user attitudes to distinct

platforms, future research should attend to these nuances.

Finally, the recommendations made based on themes in the data begin to identify some

possible directions which social media providers could follow in order to attune their service to

users’ privacy preferences and values. Future research could deepen scholarly insight into the technological plausibility and specifics of the recommendations or could improve our understanding

of users’ perception of the suggested platform innovations.

Conclusion

This study has found that user attitudes to the status of their information privacy on social media

and their perceptions of corporate uses of data – specifically, data collection, recorded observation

of personal communications, and behavioural manipulation like targeted advertising - are diverse

and not predictable by users’ frequency of engaging on social media. Overwhelmingly, participants reported feeling that, contrary to classic conceptualizations of privacy, they were not in control over

safeguarding the confidentiality of their communications and personal data, and the latter’s

(25)

25 to self-manage their information privacy online, on the one hand (Solove, 2013; Pascalev, 2017),

and users’ self-perceived ability to do so in terms of the tools and information that platforms offer them, on the other hand. The qualitative value of this study reveals that from a service provider

standpoint, a ‘one size fits all’ approach to privacy policies is likely to be ineffective in

accommodating the myriad of user privacy values and concerns. Hence, the discussion identifies

three distinct directions which social media providers could consider, were they to pursue an

alternative institutionalization of online privacy that is better attuned to users’ reported concerns.

Firstly, it is suggested that privacy on social media could be conceptualized as periodically

negotiable to allow for more flexibility around user consent for how data is collected and used,

including an institutionalization of consent as revocable. Secondly, the discussion considers the

implementation of better corporate feedback to users regarding how their data is exchanged with

third parties. Lastly, it is proposed that social media providers could offer users options of privacy

policies, each entailing distinct terms, conditions, and guarantees. For example, paid social media

plans could benefit both parties in the contract. While these recommendations are a preliminary

sketch and might not be the ideal solutions to improving users’ privacy experience, they discern

some considerable issues. Specifically, if the privacy self-management paradigm is to persist, then it

is reasonable to aim for developing social media architectures that render it feasible by offering

users tools and options to enable individual privacy-self management. Alternatively, if the paradigm

is not configured as technologically viable on social media platforms, scholars might want to

continue the conversation about different configurations of privacy, such as the viability of user

ownership over personal data, and how alternative approaches could offer solutions to the

(26)

26

Bibliography

:

Adjerid, I., Peer, E., & Acquisti, A. (2018). Beyond the Privacy Paradox: Objective Versus Relative

Risk in Privacy Decision Making. MIS Quarterly 42(2), 465-488. DOI:

10.25300/MISQ/2018/14316

Altman, I. (1975). The Environment and Social Behaviour: Privacy, Personal Space, Territory,

Crowding. In Wadsworth Publishing Company, Inc. Belmont, California

Ashworth, L., & Free, C. (2006). Marketing Dataveillance and Digital Privacy: Using Theories of

Justice to Understand Consumers' Online Privacy Concerns. Journal of Business Ethics 67,

107-123. DOI 10.1007/s10551-006-9007-7

Biczok, G. & Chia, P. H. (2013). Interdependent Privacy: Let Me Share Your Data. In A.-R.

Sadeghi (Eds.), Financial Cryptography and Data Security (338-353). Springer-Verlag.

Braun, V., & Clarke, V. (2013). Successful qualitative research: A practical guide for beginners.

Thousand Oaks: Sage

Braunstein, A., Granka, L., Staddon, J. (2011). Indirect Content Privacy Surveys: Measuring

Privacy Without Asking About It. Symposium on Usable Privacy and Security (SOUPS),

1-14

Brown, I. (2014). Social Media Surveillance. In R. Mansell and P. H. Ang (Eds.) The International

Encyclopaedia of Digital Communication and Society (1st_{, 1-7). Oxford, United Kingdom:}

John Wiley & Sons, Inc. DOI: 10.1002/9781118290743.wbiedcs122

Bryman, A. (2012). The nature of qualitative research. In Social research methods (4th, 380-413).

Oxford, UK: Oxford University Press

Carey, M. A. (2015). Focus Groups. Elsevier Ltd, 274-279

Chi, Y., Jeng, W., Acker, A., & Bowler, L. (2018). Affective, Behavioural, and Cognitive Aspects

(27)

27 G. Chowdhury, J. McLeod, V. Gillet, P. Willett (Eds.). Transforming Digital Worlds: 13th

International Conference (442-452), Springer

Chin, E., Felt, A. P., Sekar, V., & Wagner, D. (2012). Measuring User Confidence in Smartphone

Security and Privacy. Symposium in Usable Privacy and Security, 1-16

Clarke, R. A. (1988). Information Technology and Dataveillance. Communications of the ACM,

31(5), 498-512

Clarke. R. (2016). [1997]. Introduction to Dataveillance and Information Privacy, and Definitions of

Terms. Xamax Consultancy Pty Ltd. http://www.rogerclarke.com/DV/Intro.html Clarke, R. (2016). The Digital Persona and its Application to Data Surveillance. Available at:

http://www.rogerclarke.com/DV/DigPersona.html

Corbin, J. & Strauss, A. (1990). Grounded theory research: Procedures, canons and evaluative

criteria. Zeitschrift für Soziologie, 19(6), 418-427

Criado, N. & Such, J. M. (2015). Implicit Contextual Integrity in Online Social Networks.

Information Sciences 325: 48-6

Dubrovitskaya, M. (2014). Introduction. In Cryptographic Protocols for Privacy-Preserving Access

Control in Databases (pp.1-11). ETH-Zürich.

Esposti, S. D. (2014). When big data meets dataveillance: The hidden side of analytics. Surveillance

& Society 12(2), 209-225. ISSN: 1477-7487

Fahl, S., Harbach, M., Muders, T., Smith, M., & Sander, U. (2012). Helping Johnny 2.0 to Encrypt

His Facebook Conversations. Symposium on Usable Privacy and Security (SOUPS), 1-17

Harrell, M. C., & Bradley, M. A. (2009). Data Collection Methods: Semi-Structured Interviews and

Focus Groups. Rand Corporation

Iacobucci, D., Posavac, S. S., Kardes, F. R., Matther, J. S., & Popovich, D. L. (2015). Toward a

more nuanced understanding of the statistical properties of a median split. Journal of

(28)

28 Ion, I., Sachdeva, N., Kumaraguru, P., & Capkun, S. (2011). Home is Safer than the Cloud! Privacy

Concerns for Consumer Cloud Storage. Symposium on Usable Privacy and Security

(SOUPS), 1-20

Irion, K. & Helberger, N. (2017). Smart TV and the online media sector: User privacy in view of

changing market realities. Telecommunications Policy, 41(3), 170-184.

https://doi.org/10.1016/j.telpol.2016.12.013

Johnson, M., Egelman, S., & Bellovin, S. (2012). Facebook and Privacy: It's Complicated.

Symposium on Usable Privacy and Security (SOUPS), 1-15

King, J., Lampinen, A., & Smolen, A. (2011). Privacy: Is There An App for That? Symposium on

Usable Privacy and Security, 1-20

Krueger, R. A. & Casey, M. A. (2000). Modifications of Focus Groups. In Focus Groups: A

Practical Guide for Applied Research (187-192). Sage Publications, Inc.

Lau, W. W. F. (2016). Effects of social media usage and social media multitasking on the academic

performance of university students. Computers in Human Behaviour, 68, 286-291

Lupton, D., & Michael, M. (2017). 'Depends on Who's Got the Data': Understandings of Personal

Digital Dataveillance. Surveillance & Society 15(2), 254-268

Parks, P. J. (2017). Online Privacy. San Diego, CA: Reference Point Press

Pascalev, M. (2017). Privacy exchanges: restoring consent in privacy self-management. Ethics

Information Technology 19, 39-48. DOI 10.1007/s10676-016-9410-4

Patil, S., Norcie, G., Kapadia, A., & Lee, A. J. (2012). Reasons, Rewards, Regrets: Privacy

Considerations in Location Sharing as an Interactive Practice. Symposium on Usable Privacy

and Security (SOUPS), 1-15

Pingo. Z. & Narayan, B. (2016). When Personal Data Becomes Open Data: An Exploration of

(29)

29 Liew (Eds.), Digital Libraries: Knowledge, Information, and Data in Open Access Society

(pp.3-9). Cham: Springer

Pingo. Z. & Narayan, B. (2019). Privacy Literacy and the Everyday Use of Social Technologies. In

S. Kurbanoğlu, S. Ünal, J. Huotari, E. Mizrachi, L. Roy (Eds.), Information Literacy in

Everyday Life (pp.33-49). Springer

Rose, L.D., Whaling, K., Carrier, L. M., Cheever, N. A., & Rokkum, J. (2013). The Media and

Technology Usage and Attitudes Scale: An empirical investigation. Computers in Human

Behaviour 29, 2501-2511

Savci, M. & Aysan, F. (2016).The Relationship between Impulsivity, Social Media Usage and

Loneliness". Educational Process: International Journal (EDUPIJ), 2, 106-115

Schlegel, R., & Kapadia, A. (2011). Eyeing your Exposure: Quantifying and Controlling

Information Sharing for Improved Privacy. Symposium on Usable Privacy and Security

(SOUPS), 1-14

Schneier, B. (2010). A Taxonomy of Social Networking Data. IEEE Computer and Reliability

Societies, 88

Solove, D. J. (2013). Introduction: Privacy Self-Management and the Constant Dilemma. Harvard

Law Review, 126(7), 1880-1903

Staddon, J., Huffaler, D., & Brown, L. (2012). Are privacy concerns a turn-off? Engagement and

privacy in social networks. Symposium on Usable Privacy and Security (SOUPS), 1-13

Tucker, C. (2019). Privacy, Algorithms, and Artiﬁcial Intelligence. In A. Agrawal, J. Gans, & A. Goldfarb (Eds.), The Economics of Artificial Intelligence: An Agenda (pp. 423–437).

University of Chicago Press

Ur, B., Leon, P. G., Cranor, L. F., Shay, R., & Wang, Y. (2012). Smart, Useful, Scary, Creepy:

Perceptions of Online Behavioural Advertising. Symposium on Usable Privacy and Security

(30)

30 Wang, Y., Komanduri, S., Norcie, G., Acquisti, A., Leon, P. G., Cranor, L. F. (2011). "I regretted

the minute I pressed share": A Qualitative Study of Regrets on Facebook. Symposium on

Usable Privacy and Security (SOUPS), 1-16

Westin, A. F. (1970). Privacy and Freedom. Lowe & Brydone Ltd

Table 1: Reporting themes and categories in the data

Table 1. Reporting themes and categories in the data

Research question 1

User attitudes to information privacy on social media

Research question 2

User attitudes to corporate uses of data

Category Illustrative example in

data

Illustrative example in

data

Positive Positive would not pay for stronger privacy protection

1P3L: generally no, I dont think I would pay for it. I am making money for these companies by looking at their ads

Enjoys targeted advertising

4P4L: … targeted ads (…) are

helpful to be honest, and I get many whitepapers and event recommendations based on my interests in marketing even if I haven't done relevant Google search. Yes they can be ridiculous sometimes but they're useful 95% of the time with me

(31)

Understanding social media users’ attitudes to information privacy and corporate uses of data 31 designated privacy/ transparenc y features grant some transparenc y / control over user data

1P1L: It's nice to see you get

at least some control over facebook.

Supports social media providers’ services

4P2L: …But my point was that

[when companies utilize user data it] makes the services better for users. Benefits of corporations’ services outweigh risks

4P4L: I think yeah there are more

advnatages as long as users fulfill their duties of sharing

responsibly... social media companies shouldn't receive all the blame or responsibility

personal data and communicat ions are somewhat private 1P2H: i think as long as I share my personal data online it won't be purely in my possession any more

Negative

Speculation about corporate intentions

3P1L: I don't really know what

my data is for them. What is it that they want to know about me?

3P4L: Feel like they're trying to

make people feel secure, like give everyone a false sense of it

it is users’ responsibilit y to self-regulate their privacy

4P4L: I think yeah there are

more advnatages as long as users fulfill their duties of sharing responsibly... social media companies shouldn't receive all the blame or responsibility Does not enjoy targeted advertising (e.g. creepy, intrusive, irrelevant, etc.)

3P1L: That is the thing, I don't

want to feel like a product to them. Like, I sometimes watch these family vlogs. And before I installed adblocker and ghostery and stuff I would get all these ads for baby stuff and pregnancy tests, and I am really not in the market for them so it just became annoying

Negative The

obscurity of the data industry

2P4H: …For example, you can

download an archive of your facebook data, but they also buy info on everyone from third party brokers and you cannot know what they have gathered on you

would pay for stronger privacy protection

3P1L: …There are smaller alternatives to whatsapp, that I would happily for if that means they can further develop the

(32)

32

app without reliying on investors and add companies

that way. So info about me is bought and sold pretty much freely with not much awareness on what exactly it is on my part.

Feels observed / monitored / listened to / watched

1P3L: someitmes its creepy, i was

talking about robotic arm technology… and then opened facebook on my phone and htere was an ad for robotic arms. the likelihood of that being a coincidence given my other searches and internet footprint is very low

personal data are not in users’ control / private

1P2H: I don't think they are

private and confidential at all. That's why there are other chatting products made especially for "privacy" like Signal

Unethical corporate practices

3P4L: I think, a lot of the

information that these companies use does lead to greater

innovation (…) through a research method that basically equates to espionage. However I doubt the benefit outweighs the wrong here as I personally don't feel overjoyed at the prospect of Facebook using some deep conversation I had to sell information to a company that can gain something from

(33)

Understanding social media users’ attitudes to information privacy and corporate uses of data 33 designated privacy/ transparenc y features on platforms grant weak protection 3P4L: It always annoys me

however, that Facebook and Instagram seem to (at least based on what we saw when Denitsa was screen sharing) give you options with privacy and Instagram even shows you a list of your interests, despite this I don't really see any actual transparency on the matter of these website using our information Corporation s’ monopoly over user data | Do not want to support social media corporations 3P4H: I would be interested in

paying for such a service only if it would mean breaking up the monopolistic control of just a handful of companies over your data and more broadly too. Facebook especially;

3P3L: I don't think I would pay them for not using my info. I would rather delete my account

privacy is an intrinsic right that social media should protect

3P4L: …a lot of people (including myself) would argue that everyone using social media accounts has the right to privacy, I dont think Facebook needs more money to improve user environment online Negative societal effects of corporate handling of user data

1P1L: I agree that [targeted

advertising] is helpful sometimes, but I also think it can create a tunnel vision and beats the purpose of what internet was designed for. If I search for something I'd like to feel like I am getting objective information, not something based on my likes or previous searches etc;

Neutral Neutral

Indifference 1P3L: I dont feel like my data is mine, but I also dont see how it would be valuable to someone else in a way that would make me feel

uncomfortable? if that makes sense

Both

negative and positive feelings

4P1L: I would say owning this

information on users has vast negative economic and political implications. Next to all the democratizing positive ones. I wouldn't want to leave that unacknowledged

Indifference toward non-identifiable data

1P3L: i dont care if the details of my network and activity are used to develop algorithms, but i would be mad if someone used my picture on an ad or

(34)

34

something, but I dont think that anyone would want that

(35)

1/4

Appendix A: Google Forms Survey: What is your level of social media

usage?

What is your level of social media usage?

Please answer the following questions. The survey will take approximately 2 minutes.

*Required

1. Please write your name. *Your personal data will be stored safely and your anonymity is guaranteed. *

2. Please write an email address that you can be contacted at. *

3. What is your age? *

4. What is your gender? *

Mark only one oval. Male

Female Other

Prefer not to say

5. Do you have any social media accounts? If the answer is "no", skip to the end of the survey. If the answer is "yes", answer and continue to the next question. *

Mark only one oval. Yes

(36)

2/4 6. Which social media platforms are you registered on?

Tick all that apply.

Facebook Instagram Twitter Snapchat Other

7. How often do you check your: ( *If not applicable, select option "Does not apply") Mark only one oval per row.

Once a Several Once a Several Once a Once Several All

Never

month times a

month week

times a

week day an hour

times an hour ti Facebook account? Instagram account? Twitter account? Snapchat account? Other social media account(s)?

(37)

3/4 8. How often do you post or engage (share, like, comment) on your: ( *If not applicable,

select option "Does not apply") Mark only one oval per row.

Once a Several Once a Several Once a Once Several All

Never

month times a

month week

times a

week day an hour

times

an hour ti

9. How often do you communicate with other users on: (*If not applicable, select option "Does not apply")

Mark only one oval per row.

Once a Several Once a Several Once a Once Several _Al

Never

month times a

month week

times a

week day an hour

times t an hour Facebook (or Messenger)? Instagram? Twitter? Snapchat? Other social media? Facebook account? Instagram account? Twitter account? Snapchat account? Other social media account(s)?

(38)

4/4 Thank you completing this survey!

The researcher will contact your within the next two weeks to arrange a time for holding a focus group.

This content is neither created nor endorsed by Google.

Forms

(39)