Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Privacy on social networking sites among Canadian teenagers Salma, Haghighat-Kashani 2019

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


24-ubc_2019_may_haghighat-kashani_salma.pdf [ 8.87MB ]
JSON: 24-1.0378336.json
JSON-LD: 24-1.0378336-ld.json
RDF/XML (Pretty): 24-1.0378336-rdf.xml
RDF/JSON: 24-1.0378336-rdf.json
Turtle: 24-1.0378336-turtle.txt
N-Triples: 24-1.0378336-rdf-ntriples.txt
Original Record: 24-1.0378336-source.json
Full Text

Full Text

Privacy on Social Networking SitesAmong Canadian TeenagersbySalma Haghighat-KashaniB.En.D., The University of British Columbia, 2013A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THEREQUIREMENTS FOR THE DEGREE OFMASTER OF APPLIED SCIENCEinThe Faculty of Graduate and Postdoctoral Studies(Electrical and Computer Engineering)THE UNIVERSITY OF BRITISH COLUMBIA(Vancouver)April 2019c©Salma Haghighat-Kashani, 2019The following individuals certify that they have read, and recommend to the Faculty ofGraduate and Postdoctoral Studies for acceptance, the thesis entitled:Privacy on Social Networking Sites Among Canadian Teenagerssubmitted by Salma Haghighat-Kashani in partial fulfillment of the requirements for thedegree of Master of Applied Science in Electrical and Computer EngineeringExamining Committee:Dr. Konstantin Beznosov, Electrical and Computer EngineeringCo-supervisorDr. Matei Ripeanu, Electrical and Computer EngineeringCo-supervisorDr. Philippe Kruchten, Electrical and Computer EngineeringExamining Committee ChairiiAbstractThe widespread popularity of social networking sites (SNSs) among teenagers continuallyraises concerns over their safety among parents, educators, and policy makers. Althougha teen’s use of such platforms plays a vital role in his or her social development, suchonline activities lead to a plethora of personal information being shared that increasesvulnerability to privacy invasion and information misuse.The employed monitoring, restriction and educational methods of privacy protection havebeen unsuccessful in encouraging teens to stay private on SNSs. While researchers haveinvestigated online practices of teens, we lack a clear understanding of the rationales behindtheir safety and confidence on SNSs. Additionally, with the rapid emergence of new socialnetworking applications each year and the ongoing evolution of educational school programson privacy, a teen’s notion of privacy and online behaviours are constantly evolving. Asa result, a thorough exploration of online interactions and thought processes of teens canhelp us better understand them and consequently communicate with them.This thesis explores the perception of online privacy by Canadian teenagers (15-17 yearolds) as well as their privacy-related concerns and behaviours on SNSs. To this end, semi-structured interviews were conducted with high school students (N = 20), and an onlinesurvey was completed by a more diverse pool of participants (N = 94).Based on our results, we grounded a theory that highlights our participants’ broad def-inition of online privacy which directly relates to their online privacy concerns. Theseconcerns shape their decision-making processes about information disclosure. Our theoryhighlights our participants’ frequently used rationales for feeling safe online, the variety ofprotective measures used to address their privacy concerns, and the factors that influencetheir choice of SNSs. Our findings can help parents and educators gain a better under-standing of a teen’s perception of online privacy and interactions on SNSs. Additionally,our findings can inform the creation of better suited policies, educational approaches, andparental supervision techniques for teens.iiiLay SummaryThis thesis explores the perception of online privacy by Canadian teenagers (15-17 yearolds) as well as their privacy-related concerns and behaviours on Social Networking Sites(SNSs). To this end, semi-structured interviews were conducted with high school students(N = 20), and an online survey was completed by a more diverse pool of participants(N = 94).Based on our results, we grounded a theory that highlights our participants’ broad def-inition of online privacy which directly relates to their online privacy concerns. Theseconcerns shape their decision-making processes about information disclosure. Our theoryhighlights our participants’ frequently used rationales for feeling safe online, the variety ofprotective measures used to address their privacy concerns, and the factors that influencetheir choice of SNSs. Our findings can help parents and educators gain a better under-standing of a teen’s perception of online privacy and interactions on SNSs. Additionally,our findings can inform the creation of better suited policies, educational approaches, andparental supervision techniques for teens.ivPrefaceThis thesis presents research performed by Salma Haghighat-Kashani.• The Introduction chapter was written by Salma Haghighat-Kashani based on sug-gestions by her supervisors, Konstantin Beznosov and Matei Ripeanu.• Chapter 2 was written by Salma Haghighat-Kashani with feedback and suggestionsby Konstantin Beznosov, Matei Ripeanu and Jennifer Shapka.• Chapter 3 was written by Salma Haghighat-Kashani with feedback and suggestionsby Konstantin Beznosov, Matei Ripeanu and Jennifer Shapka. The research wasdesigned by Salma Haghighat-Kashani based on their feedback. The study interviewswere conducted and transcribed with the help of Rachel Baitz.• Chapter 4 was written by Salma Haghighat-Kashani with feedback and suggestionsby Konstantin Beznosov, Matei Ripeanu and Jennifer Shapka. The online survey wasdesigned and the results were analyzed by Salma Haghighat-Kashani• Chapter 5 was written by Salma Haghighat-Kashani with feedback and suggestionsby Konstantin Beznosov, Matei Ripeanu and Jennifer Shapka.• Chapter 6 was written by Salma Haghighat-Kashani with feedback and suggestionsby Konstantin Beznosov, Matei Ripeanu and Jennifer Shapka.Our exploratory study was approved by UBC Behavioural Research Ethics Board (BREBnumber H15-00562), under the title of “Adolescent Understanding of Privacy in aTechnology-Filled World”.Our confirmatory study was approved by UBC Behavioural Research Ethics Board (BREBnumber H17-00785), under the title of “Understanding Influential Factors on Teenagers’Online Concerns.”vAcknowledgementsI would like to offer my special thanks to my thesis supervisors, Dr. Kostantin (Kosta) Beznosovand Dr. Matei Ripeanu, for their continuous support and mentorship throughout this work. I amsincerely grateful to Matei for taking a chance on me and helping me transition to engineering.Without his guidance and encouragement, this would have not been possible. I would also like toexpress my deepest gratitude to Kosta for steering me in the right direction, providing me withinsightful feedback, and facilitating my journey.I would like to thank Jennifer Shapka for her constructive feedback throughout this project. Also,I would like to thank my colleagues at Networked Systems Laboratory for making the years spentin the lab memorable.I am extremely grateful to my close friends for their unconditional support along the way: MoAfrasiabi for helping me with the transition, Nelou Keramati for always coming to my rescue whenI needed help, Naghmeh Kia for always listening and offering me advice, and Soheil Kianzad forchallenging my thinking. Special thanks to Kimia Nassehi for her friendship and for always beingthere through thin and thick, and to Dustin Hahn for believing in me and encouraging me to achieveanything I desire.To all my friends scattered around the world, thank you for your thoughts, well-wishes, calls, texts,editing advice, and simply being there for me.I owe a debt of gratitude to my supportive family and role models who always inspire me tobe the best I can be. I would like to thank you all for your love and devotion: Hossein, Roya,Maryam and Amir Haghighat Kashani, Banafsheh and Fatemeh Hashemi, my grandmother TayebehHashemi, and my deceased grandparents Abdullah Hashemi, Fakhrosadat Ashrafzadeh, and Gho-lamali Haghighat Kashani.Last but not least, I am forever indebted to my loving parents Fahimeh and Ahmad, and my brotherAli, for their unconditional love and support in life. To Fahimeh and Ahmad, thank you for allyour sacrifices to bring me to Canada and provide me with an incredible environment for educationand growth. To Ali, who has been my friend and mentor, for guiding me step by step through thisjourney. Thank you for paving the way and providing me with the best opportunities. This thesisis dedicated to you, as a sign of my appreciation for your love and support. This journey wouldnot have been possible without you.viContentsAbstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iiiLay Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ivPreface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vAcknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.1 Understanding Teens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21.2 Protecting Teens Against Harm . . . . . . . . . . . . . . . . . . . . . . . . . 31.3 Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31.4 Organization of the Thesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Background and Related Work . . . . . . . . . . . . . . . . . . . . . . . . . 52.1 Social Development and Parental Influence . . . . . . . . . . . . . . . . . . 52.2 Teens on SNSs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62.3 Privacy Definitions, Concerns, and Management. . . . . . . . . . . . . . . . 82.3.1 Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82.3.2 Concerns and Management . . . . . . . . . . . . . . . . . . . . . . . 93 Exploratory Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123.1 Sampling and Participants Recruitment . . . . . . . . . . . . . . . . . . . . 13viiContents viii3.2 Data Collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133.3 Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153.4.1 Perception of Privacy . . . . . . . . . . . . . . . . . . . . . . . . . . 163.4.2 Online Concerns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173.4.3 Feeling Safe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193.4.4 Channel Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263.5.1 Perception of Privacy . . . . . . . . . . . . . . . . . . . . . . . . . . 263.5.2 Online Concerns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283.5.3 Feeling Safe Rationales . . . . . . . . . . . . . . . . . . . . . . . . . 304 Confirmatory Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324.1 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324.1.1 Pilot Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 334.1.2 Participants, Recruitment, and Consent . . . . . . . . . . . . . . . . 334.1.3 Data Collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 344.1.4 Ethical Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . 354.1.5 Data Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 364.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374.2.1 RQ1: Most Important Privacy Characteristics . . . . . . . . . . . . . 374.2.2 RQ2: Privacy Levels of Online Content . . . . . . . . . . . . . . . . 384.2.3 RQ3: Justifications for the Lack of Concerns . . . . . . . . . . . . . 404.2.4 RQ4: Tactics for Privacy Protection . . . . . . . . . . . . . . . . . . 424.2.5 RQ5: Factors Linked to the Choice of SNS . . . . . . . . . . . . . . 454.3 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 475 Discussion and Implications . . . . . . . . . . . . . . . . . . . . . . . . . . . 495.1 Perception of Online Privacy . . . . . . . . . . . . . . . . . . . . . . . . . . 495.1.1 What is online privacy? . . . . . . . . . . . . . . . . . . . . . . . . . 505.1.2 What is private? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 505.2 Online Privacy Concerns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51Contents ix5.3 Decision-making Process on Information Disclosure . . . . . . . . . . . . . . 515.3.1 Determining the choice of SNS. . . . . . . . . . . . . . . . . . . . . . 525.3.2 Popular protective measures. . . . . . . . . . . . . . . . . . . . . . . 535.3.3 Justifications for the lack of concerns. . . . . . . . . . . . . . . . . . 545.4 Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 556 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57A Exploratory Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60A.1 Pre-Interview Survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60A.2 Interview Script . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62B Confirmatory Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66B.1 Instagram Recruitment Ads . . . . . . . . . . . . . . . . . . . . . . . . . . . 66B.2 Confirmatory Study Participant Recruitment . . . . . . . . . . . . . . . . . 68B.2.1 Initial Contact Message . . . . . . . . . . . . . . . . . . . . . . . . . 68B.2.2 Parental and Teen/Participant Consent Form . . . . . . . . . . . . . 68B.3 Online Survey Questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71Consent Statement . . . . . . . . . . . . . . . . . . . . . . . . 71Section 1: Your Social Networking Activities . . . . . . . . . 71Section 2: Friends & Followers . . . . . . . . . . . . . . . . . 72Section 3: Online Concerns & Private Information . . . . . . 73Section 4: Content Sharing . . . . . . . . . . . . . . . . . . . 74Section 5: About You . . . . . . . . . . . . . . . . . . . . . . 77Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79List of Figures3.1 Interview data analysis codes saturation. . . . . . . . . . . . . . . . . . . . . 153.2 Our grounded theory on teens’ online privacy. . . . . . . . . . . . . . . . . . 27B.1 Instagram ads posted on July 24 and August 1,2018. . . . . . . . . . . . . . 66B.2 Instagram ads posted on August 6 and August 14, 2018. . . . . . . . . . . 67xList of Tables3.1 Participants’ demographic information. . . . . . . . . . . . . . . . . . . . . . 134.1 Comparison of the study sample and 2017 Canadian 14-17 year olds popu-lation across provinces [1]. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 334.2 Summary statistics of privacy characteristics. . . . . . . . . . . . . . . . . . 374.3 Summary statistics of perceived privacy level. . . . . . . . . . . . . . . . . . 394.4 Pairwise comparison of participants’ privacy views on the types of onlinecontent. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 404.5 Summary statistics of lack of privacy concern justifications. . . . . . . . . . 414.6 Pairwise comparison on lack of privacy concern justifications. . . . . . . . . 424.7 Summary statistics of popular protective measures. . . . . . . . . . . . . . . 444.8 Pairwise comparison on protective measures. . . . . . . . . . . . . . . . . . 454.9 Summary statistics of factors in determining the choice of SNS. . . . . . . 464.10 Pairwise comparison on influential factors. . . . . . . . . . . . . . . . . . . . 47xiTo my beloved parents, Fahimeh and Ahmad,and my dear brother Ali.xiiChapter 1IntroductionThe overwhelming majority (99%) of Canadian students aged 9-17, and of American teens(97%) aged 13-17 are online and have access to a mobile device [2, 3]. Of those Canadianstudents, 57% reported Facebook as their favourite website. More than 40% indicatedsome social networking related activities, such as posting content and following friends,as their most frequent online activities [2]. Social networking sites (SNSs) are an inte-gral part of teens’ online world [3] and provide a way to satisfy their developing socialneeds [4] by allowing them to explore different social roles, form identities, and expressthemselves [5]. Most teens feel supported and more connected to their peers because ofSNSs [6]. Unfortunately, research has shown that there are risks associated with sharingpersonal information online via SNSs [7, 8]. Cyberbullying is a common outcome, andcan lead to depression, anxiety, or even suicide [9–12]. Other risks include being exposedto online predators, harassment, sexual solicitation, or exposure to financial and identityfraud [13–15].Additionally, it is well known that compared to adults, risk-taking behaviours, such asreckless driving and binge drinking are more prevalent among teens [16, 17]. Due totheir evolutionary characteristics such as being trusting and naive, and since they tendto ignore long-term outcomes, teens are most vulnerable to risks [18, 19]. It has alsofound that availability of immediate feedback, and being in circumstances that increase1Introduction 2reward system activation, such as the presence of peers, increase teens’ engagement inrisky behaviours [20]. Indeed, such conditions are met when teens engage in SNSs.1.1 Understanding TeensAt a policy level, privacy policies fail to take into account teens’ perception of their onlineprivacy. For instance, the Personal Information Protection and Electronic Documents Act(PIPEDA) assumes that transparency regarding data collection will encourage users towithhold private information [21], and SNSs tend to follow privacy models that focus onusers’ personal control over information. These models are inconsistent with teenagers’online practices, which focus on the importance of shared social norms such as trust toregulate privacy [22]. This facilitates commercial and even malicious data collection. Forinstance, Felt and Evans have reported that third parties on Facebook can verify theidentity a single user, even without having their names [23]. As a result, the potential toexploit shared information increases, which in turn puts teens at more risk.While these studies all shed some light on teens’ perception of privacy; their view ontheir information disclosure behaviour and their perception of their safety still remainsan open problem. Previous research suggested that teens’ frequent disclosure of personalinformation is due to a lack of concern for privacy [24, 25]. However, recent work indicatesthat teens are as aware of and concerned about the confidentiality of their information [26–29] and there is little variation between youth and adults’ privacy concerns [30, 31]. Thisdichotomy between information disclosure and privacy concerns is a continuing topic ofresearch. Many hypotheses have been postulated to explain this dichotomy, for example:social pressure [32, 33], a unique balance between perceived risks and rewards [34, 35], orbecause teens seek privacy differently by exercising control over other people’s access totheir information [22, 36].Introduction 31.2 Protecting Teens Against HarmTeenagers do not get much relevant guidance to engage in good privacy practices. Edu-cators at schools use ineffective techniques to influence teens’ online behaviours, such asnegatively framing the use of SNSs and using “scare tactics” [29, 37]. Additionally, parentsoften rely on top-down practices, such as placing strict limits on technology use or by usingonline safety apps (that are designed to support parental authority [38]) to control andmonitor their teens’ online activities. Although methods used in prevention campaignsand interventions lead to an increase of safety knowledge, teens’ online behaviours remainlargely unchanged [39]. Unfortunately, these parenting practices tend to undermine teens’growing need for autonomy and privacy, and often lead to teenagers engaging in high riskbehaviour as they begin to hide their online activities from their parents.In order to encourage and empower teens to be active agents in protecting themselvesonline, it is necessary to understand their perception of privacy as well as the factors thatinfluence their risk level in online interactions. Hence, it is necessary to further exploreteens’ use of SNSs and their thought processes to provide an accurate narrative from theirperspective.1.3 ContributionsIn this work, we seek a better understanding of the factors that influence teens’ on-line decision-making processes regarding privacy. To this end, we first conducted semi-structured interviews with 20 Canadian high school students, aged 15-17, with the goalof investigating the concerns and rationale behind their actions. We followed a groundedtheory methodology, analyzed the data by performing open coding, axial coding, and selec-tive coding to better understand and characterize teens’ online behaviours and concerns.To confirm our findings with a more representative sample, we then conducted an onlinesurvey with 94 Canadian teens.Introduction 4The main contribution of this research is a better understanding of teens’ notions of onlineprivacy, concerns and decision-making processes on information disclosure. Our specificcontributions are as follows: We confirm previous findings on teens’ views of online privacy, and identify additionalaspects that suggest teens have a broader definition of privacy compared to adults. We expand on the main aspects of teens’ online concerns and determine that socialstatus concerns are directly related to teens’ online decision-making processes. We identify the types of information teens perceive as more private. We provide a comprehensive list of protective tactics that teens employ and identifythe most popular tactics. We determine that the choice of SNS is often a protective tactic used by teens andwe expand the list of factors that influence this decision. We identify major rationales that influence teens’ lack of privacy concerns and providedeeper insight regarding their online behaviours.1.4 Organization of the ThesisThis thesis includes exploratory and confirmatory studies. We provide an overview of thebackground and the related work in chapter 2. In chapter 3, we present our exploratorystudy, including methodology, data analysis, results, and discussion. In chapter 4, wediscuss our confirmatory study and present the study’s methodology, data analysis, andresults. In chapter 5, we discuss our main findings, the implications, and the limitationsof our studies. Lastly, we summarize the main conclusions in chapter 6.Chapter 2Background and Related WorkTo better understand teens’ use of SNSs, this section provides an overview on social devel-opment processes that influence teens’ actions, the role of major SNSs on teens’ lives, aswell as relevant literature on teens’ privacy definitions, concerns, and their current privacymanagement techniques.2.1 Social Development and Parental InfluenceCognitive, psycho-social, and emotional developments in adolescence mark the transitionfrom childhood to adulthood. In early adolescence (ages 11-14), teens begin to define theirpersonal spaces by creating stronger physical boundaries such as closing doors and enforc-ing privacy rules such as knocking [40]. Early adolescence is also the time during which thecomplexity of teens’ conception of privacy crystallizes [41]. In later years (14-17), adoles-cents focus on self-identity and start seeking independence from parents and other familymembers [41, 42]. At this stage, teens may distance themselves from parents by not talkingto them or withholding information [43]. As a result, various parental monitoring tech-niques are used to supervise teens’ day-to-day activities. Techniques that involve “attentionto and tracking of the child’s whereabouts, activities, and adaptations” [44, p.61] are asso-ciated with reduced risky behaviour in teens [45]. Hence setting up rules and monitoring5Background and Related Work 6where teens go and with whom they hang out are common practices. In the current digitalage, more than 60% of American parents claim to have checked the websites their teenvisits along with their social networking profiles [46]. Less than half of parents have lookedat their teens’ calls and text messages. Parents have also used parental control software toblock, monitor, and filter their teens’ online activities (39%) [46, 47]. However, such tech-nical restrictions have no significant effect on protecting teens against online risks [47, 48].Although direct intervention by parents may reduce teens’ exposure to online risks, it alsoaffects their ability to interact online and effectively cope with risks [49]. At the same time,active mediation by parents such as talking to teens about their online activities has beenfound to have a positive influence on reducing problematic online behaviours [39, 49, 50].A majority of parents also prefer non-technical approaches to privacy, such as rule-makingand co-use. For instance, more than 90% of parents have talked with their teen about shar-ing and viewing (in)appropriate content online [46]. Despite these efforts, teens continueto participate in social networking activities that put them at risk.By distancing themselves from parents, adolescents seek social support from peers andbegin to disclose personal matters with friends [51]. They seek their friends’ insight, con-nection, and validation instead of parents. This results in having their social relationsbecoming paramount in their lives. During this time, adolescents become members of peergroups to explore their interests and form their identity [52]. Additionally, the develop-ment of abstract and operational thinking enables teens to think about the consequencesof their actions and understand the causes and effects related to their behaviours. Suchthinking also results in the creation of an imaginary audience (usually their peers), thatis considered to be observing and thinking about the teen [53]. Lastly, it allows them tothink about their feelings and how they are perceived by others.2.2 Teens on SNSsTeens in particular have been early adopters of SNSs. About 71% of US teens aged 13to 17 have Facebook accounts, 52% are on Instagram, and 41% are on Snapchat [54].Comparatively, 85% of Canadian teens in grades 7 to 11 have Facebook accounts, andBackground and Related Work 742% have Instagram accounts [2]. In 2016, it was estimated that 80% of teenagers withmobile devices use Snapchat [55]. SNSs such as Facebook, Instagram, and Snapchat, makecommunication and interaction effortless. Facebook enables users to stay connected andshare text, photos, and news with others. The photo sharing app, Instagram, encouragesusers to share photos and videos publicly or privately. Snapchat facilitates communicationsvia what is perceived as “self-deleting” photos and videos. Each platform supports varioustypes of activities and serves a range of communication and social needs. For instance,Facebook gratifies one’s need to have fun, and know about their friends’ lives [56]. On theother hand, instant messaging is more about maintaining and developing relationships [56].Social networking is among teens’ most popular activities online. This potentially in-creases their vulnerability to online risks. Teens use SNS to message friends, comment,post content, and follow others’ activities. They participate in social media by sharingtheir personal opinions [31]. They form their online identities and showcase their offlineexperiences [31]. SNSs allows teens to create a reputation online and manage their socialstatus through monitoring their received “likes” and comments [5, 31, 57]. During thisprocess, teens often overshare personal information in order to develop and maintain rela-tionships [34]. Although they are aware of the consequences of their online behaviours [34],inevitably, they share a considerable amount of personal information on SNSs. In 2013,Madden et al. [31] reported that an overwhelming majority (90%) of teens post their realname and personal photos online. They found that more than 50% of teens share theirschool name, the town they live in, and their email address, while 20% share their cellphone number. This should not be surprising, since Facebook’s “terms of service” askusers to provide accurate personal and contact information [58]. As noted, such disclosureof information increases teens’ vulnerability to online risks [11, 59].Some studies have argued that teens’ disclosure of personal information is an indicator fortheir lack of privacy concerns [60–63, p.51]. Sithira and Nguwi [24] have reported thatteens are not cautious while online: they are comfortable with sharing offensive views andthey feel comfortable with e-banking or downloading online materials. A study of 7,000college students revealed that although students were concerned about personal informationsuch as passwords and social security number, they considered SNSs as “private” spacesBackground and Related Work 8and were not concerned about sharing personal content [64]. Past work has also shownthat teens use a ”risk-benefit” approach to privacy. In a study with 326 high schoolstudents, Youn [65, 66] found that higher perceived benefits of information disclosure resultin more willingness to share. However, other studies with college students found no suchcorrelation [27, 33]. Tufekci [33] claimed students care about their privacy and use privacysettings and nicknames to manage access to their information. Additionally, focusing onolder teens (ages 18-19), Agosto et al. [37] found that older teens are less concerned abouttheir online safety since they believe they are capable of protecting themselves. Given thesevaried views on teens’ privacy concerns, it is important to gain more insight on teens’ useof SNSs and their notion of online privacy to better protect them against possible harm.2.3 Privacy Definitions, Concerns, and Management.2.3.1 DefinitionsPrivacy is a social construct and people’s conceptualization of privacy varies. Westin [67](1968) defined privacy as “the claim of individuals, groups, or institutions to determinefor themselves when, how, and to what extent information about them is communicated toothers.” One of the states of individual privacy identified by Westin is “reserve”, whichis the most applicable state to today’s information privacy. Taking place in public, it isthe desire to limit information disclosure to others while such desire is being recognizedand respected by them [67, 68]. Altman (1975) defined privacy as “the selective controlof access to the self” [68]. In both theories, privacy is described as culturally-specific.Hence, people’s attitudes toward privacy vary depending on their social norms, values,and practices. In a similar approach, Nissenbaum [69] explained contextual integrity asa core aspect of privacy. She identified that the subject of the data, type of information,sender, recipient, and how information is transmitted are critical parameters that controlprivacy [69].In the context of online social networking, adults’ definitions of privacy have been inves-tigated by a number of studies to demonstrate its importance in their lives. As examinedBackground and Related Work 9by boyd on privacy concerns of adult Facebook users, privacy is about the sense of controlover the information, context, and audience [70]. In another study on 19 graduate and un-dergraduate students, Dwyer highlighted the “impression management” efforts that usersput to present a good impression of themselves to the audience [71]. Dwyer findings sug-gested that users viewed their online content as public, and believed they were responsibleto control the availability of their information and the image they portray [71]. In a studyon privacy and technology, Kwasny et al. [72] reported that younger adults’ definition ofprivacy is consistent with Westin’s “reserve” state, as it involves ideas of controlling theirinformation and disclosure or non-disclosure decisions (e.g., whether to share the informa-tion or keep it to themselves). However, older adults have a narrower conceptualization ofprivacy which focuses on “something official that they are given: a legal document, healthinformation, their social security number, or a secret that a friend confides in them” [72].Studies on teens’ privacy perceptions in online context have revealed that a majority defineprivacy in terms of both their audience and access to their information [28, 29]. Boyd andMarwick [28] suggest that teens’ approaches to privacy are not simply about disclosure ornondisclosure of information. Teens want to participate in SNSs, but they only want to bevisible to certain people [28]. Hence, they seek to regulate the boundaries between privateand public by deciding on what information to share and what to keep to themselves [5, 28].Marwick and boyd [22] add that teens also perceive privacy as the ability to control theimage they portray to others. Teens place a significant emphasis on social norms such astrust and respect when speaking of privacy [22].2.3.2 Concerns and ManagementA majority of adult American Internet users are concerned about strangers accessing theirpersonal information (84%) and computer hackers accessing their credit card numbers(50%) [73]. A similar study by Krasnova and Kift on adult German Facebook users high-lighted that they are more concerned about the misuse of their personal information suchas name, address, phone number, and photos compared to bank data and passwords [74].Users believe that they should know all about their available information on the web-sites, and that a law should be in place to have websites delete personal information [30].Background and Related Work 10Other privacy-related concerns include having personal information shared with third par-ties, identity theft, and (prospective) employers gaining access to nonprofessional onlineactivities [75].Compared to studies on adults, it has often been suggested that teens are not concernedabout their online privacy. For instance, it has been found that compared to their parents,teens are less worried about online data collection by marketers [76]. Additionally, it hasbeen reported that teens do not utilize the privacy customization features provided by theSNS [71]. In a study on teens’ attitudes toward privacy by Lejnieks [77], teens are askedif they are concerned about online “privacy and security” issues. More than 60% of themhave claimed that they do not care [77]. Additionally, Sithira and Nguwi [24] reported thatmany teens do not secure and protect their private information on SNSs. Another study onyoung adults indicated that they are less concerned about portraying a professional imageon SNSs [78].Recent studies on teens’ online behaviour, however, have indicated that teens are concernedabout their privacy and they take steps to protect themselves [29, 59, 79].Tufekci [33] reported that instead of nondisclosure, college students’ concerns over un-wanted audience lead them to adjust their profile visibility and use nicknames. (Note thateven when their profile is set to private, teens still share personal content with up to hun-dreds friends that they know casually on SNSs [5].) Teens also use additional techniques tomanage their privacy. They delete friends from their accounts, share false information and”cloak” their messages so that only certain friends can understand them, un-tag namesfrom photos, delete friends’ comments from the accounts, and delete/edit their sharedcontent [31, 76].While researchers have investigated teens’ online practices, there is no clear understandingof the rationale behind teens’ confidence in their safety on SNSs. With this wide range ofonline activities and concerns, it is necessary to obtain a more comprehensive understandingof how teens use SNSs, what they worry about online, and how their concerns influencetheir decision-making processes on information disclosure in order to better protect themagainst online privacy violations. Given that the employed monitoring, restriction, andeducational methods of protections have not been successful in encouraging teens to stayBackground and Related Work 11safe and private on SNSs, such insights can help parents and educators to develop newstrategies to address teens’ online privacy issues.Chapter 3Exploratory StudyTo understand teen’s perception of privacy, we chose to focus on their individual experiencesusing the Grounded Theory (GT) approach. First articulated by Glaser and Strauss [80]and later modified by Corbin and Strauss [81], GT is a qualitative research method with theprimary purpose of constructing theories from data. In this thesis, we used analytical toolssuch as constant comparisons, theoretical comparisons, the “flip-flop” technique, waving ared flag technique [81],and clustering [82] to analyze our data. To this end, we conductedin-depth semi-structured interviews to explore participants’ self-reported online behavioursand their decision-making processes. Using in-person interviews allowed us to clarify givenresponses, ask follow-up questions and deviate when new topics emerged.We conducted three rounds of interviews over a year and a half. Each round was followed byan analysis process. The first round was in the Summer of 2015 from July to August, with11 participants. The second and third rounds were conducted in June to August 2016, andin January 2017, with five and four teens respectively. On average, each interview lastedabout 45 minutes. They were all audio recorded and transcribed for analysis.12Exploratory Study 13Table 3.1: Participants’ demographic information.Gender Live WithFemale 60% Both Parents 70%Male 35% Mother 20%Transgender 5% Father 10%Age Parent EducationGrade 10th 20% Post secondary 55%Grade 11th 20% Bachelor 35%Grade 12th 60% Some college 10%3.1 Sampling and Participants RecruitmentWe used criterion sampling to recruit 20 teens aged 14 to 17 (13 females; see Table 3.1). Thisage range was chosen as it represents a period when adolescents’ independent functioningfrom their parents increases [53]. Developmentally, they are focused on friendships andintimacy, hence, privacy becomes a more relevant concern. Since Facebook was at the timethe dominant SNS among teens [54], participants were targeted through Facebook ads.All participants signed a consent form and received $10 as compensation. Our study wasapproved by the university’s research ethics board (approval # H15-00562).3.2 Data CollectionBefore starting the interviews, participants were asked to complete a questionnaire thatcollected their demographic information (including age, living arrangements, e.g., livingwith one parent or both, parent’s education level), as well as general information aboutaccess to and use of SNSs (Appendix A.1). The interviews were then conducted in a flexiblemanner with the help of an interview guide which included sections on teens’ use of digitaldevices, use of SNSs, and online privacy practices. This guide was reviewed by Dr. JenniferShapka, an expert from the Department of Educational and Counselling Psychology, andExploratory Study 14Special Education at University of British Columbia to ensure that the questions addressedthe study objectively and without bias among the participants. Interviews were carried outin a conversational style. We developed the interview guide to help us carry out discussionswith teens in a conversational style. The first round of the interviews broadly covered topicssuch as teens’ ideas of privacy, their online experiences and interactions, their experiencecreating their accounts, and their advice for others. We specifically probed for recent andpersonal experiences to help them uncover the rationale behind their actions. Throughoutthe following academic year, we transcribed and performed a preliminary analysis of thedata. Based on the results of the analysis, the second round of the interviews was conductedto further focus on teens’ concerns, their privacy settings, and other safety measures theytake to stay private. Participants were always asked the follow-up “why” question toelaborate their privacy attitudes and decisions. The third round of interviews aimed todelve deeper into their use of different SNSs and their perception of privacy and safetyassociated with each platform (Appendix A.2).3.3 AnalysisThe coding process was performed after each round of interviews. After the 20th inter-view, no new codes appeared during the data analysis, indicating that saturation had beenreached (Figure 3.1). As suggested by Marshal and Rossman, our coding process includedbreaking down the data into manageable pieces, interpreting the data, and assigning con-ceptual labels (codes) to the pieces of text to best describe what we believed was indicatedby the data [82]. Memoing was done throughout the analysis process to describe the codes,our interpretations, and the relationship between codes.We started the preliminary analysis process after the first round of interviews in orderto refocus the questions based on the emerging content. We began with the initial opencoding [82], during which we looked at every sentence to identify and label the essence ofthe statements. As a result, we generated a long list of 300 codes for the first 11 interviews.During the axial coding [82], we reviewed and grouped the initial codes around commonconceptual categories. About 21 categories emerged through this process. Similarly, theExploratory Study 15open coding and axial coding processes were performed on the new interviews. When newtopics were discussed, new codes were generated. When similar topics were mentioned, theywere labeled with the existing codes to further develop and refine the existing categories.Through selective coding, we identified the core category and sub-categories. The remain-ing categories were used to expand on those. We further analyzed the transcripts by lookingfor themes that related to teens’ online concerns to develop a model that characterized theirprivacy perception. We compared the description of our model with the data to identifythe model’s accuracy and we updated the model iteratively until we felt confident that themodel correctly represented teens’ view on privacy.3.4 ResultsWe identified teens’ online concerns as the core category of our findings, and their concep-tualization of privacy, perception of safety, and protective measures as the sub-categories.In what follows, we explain each category in more detail, and provide illustrative quotesfrom interviews.Figure 3.1: Interview data analysis codes saturation.Exploratory Study 16Since the purpose of this qualitative study was exploration of a phenomenon, we avoidquantitative statistics in the reporting of the analysis results. Instead, we highlight a fullrange of answers and points of view.3.4.1 Perception of PrivacyTo confirm previous studies on teens’ definition of privacy, we asked participants to describeprivacy in their own terms and to provide examples of what they perceive as a privacyinvasion.The majority of the participants highlighted the importance of having the ability to controlwho can observe or access their shared content. Their descriptions focused on the ability tochoose either the information they would like to share, their audience, or both. For instance,P1 define privacy as “the amount of control [she has] over who sees [her] information andwho doesn’t.” P6 also stated: “Privacy is how much you choose to share, don’t chooseto share; with whom you choose to share or whom you don’t choose to share with.” Shefurther added that privacy is also about how much information other people share abouther. P9 was more specific with his description. He explained privacy as the ability “to putstuff on the internet that only the people [he has] on [his] friends list can see.”P14 focused on the importance of consent. She stated: “privacy is making sure that theinformation that you shared is not shared to other people without your consent. Like, Iguess that if I’m talking to someone, or if I share a photo, I think that the photo that Ishared on that website should stay on that website. But if it’s saved by other people andspread on other websites then it would feel to me as an invasion of my privacy.” Similarly,P19 described invasion of his privacy as someone accessing information that he had notprovided to them.From a slightly different perspective, few participants focused on traceability. For instance,P4 stated that privacy is “to do something without other people knowing and/or interfer-ing.” He gave an example and explained that it is “being able to browse different web pagesand being able to say things online without other people being able to, like, track you downExploratory Study 17or look at all your other history and stuff.” For P17, privacy was also about the ability“to do something without somebody else knowing it.”3.4.2 Online ConcernsTo shed light on teens’ concerns when using SNSs, we asked them about their privacyconcerns and their thought process when posting online content. Their responses fell intotwo general themes: social status and security concerns.Social Status ConcernsThe majority of our participants were concerned about their social status and how theirposts would be perceived by the audience. This includes concerns over the audience’sinterests and judgements. Almost all of the participants made comments such as “I thinkabout who’s gonna see” the post, or “who is my audience?”For instance, P4 stated: “So I find my biggest concern is that have people read my postor find it interesting?”. Thinking thoroughly about the audience’s judgements, P6 stated:“I like to think about how different people in my friends list would see this [post]. Likehow would I feel if my friend saw this, my best friend, my mom, my distant relative. Likewhat would they feel about me posting this?” P1 mentioned asking herself “is [this post orpicture] going to portray a good image of me to a stranger who doesn’t know me?” Shefurther explained: “Because people who don’t know you and they are more likely to judgeyou based on a post or a picture than my friend who knows me and might know why I postedthat picture, or if that wasn’t like me. But I think [to myself ]: does this show strangers whatI’m actually like?” Similarly, P8 expressed her concerns about future employers: “...if youhave an employer in a year or two, he can go on your Facebook, he can see the things youdid. If you go to a strip bar do you want to post pictures of that?” Additionally, someparticipants worried about their friends posting content about them that could damagetheir social status: “I guess my biggest concern is if other people post pictures of me. Likefor example with alcohol in my hand, that would be my biggest concern” (P1).Another common statement involved participant’s attention to not offending their audience.P7 explained: “Well, I don’t think I’m a very offensive person, I hope not at least, andExploratory Study 18I don’t say offensive things. But, as I said, without the benefit of the body language andlike intonation, sometimes things can be taken in the wrong way and so I wanna be carefulthat everything I say does not come across in any way as something that would be deemedoffensive to anybody.”To further explore teens’ concerns on SNSs, we asked our participants about their onlineregrets. We identified that they also regret actions that negatively influence their socialstatus. They regret sharing “stupid” posts, “ugly” pictures, or in general embarrassingcontent. For instance, P6 stated: “I posted like stupid things. They are not like is just stupid things that I shouldn’t have said or done.” Additionally, similar to theirconcerns about offending others, they expressed regrets about sending rude messages orcomments. For instance, P7 mentioned: “Sometimes I just say something kind of offhandif I’m in like kind of a bad mood. Not anything like offensive but it can come off kind ofrude and I’ll be like oh sorry I didn’t mean to be rude, I haven’t had any sleep and I don’tknow what I’m saying, that kind of thing.”Security ConcernsThese concerns were mainly about unauthorized access to teens’ data such as financial in-formation, passwords, and social insurance number. It also involved concerns over hackers,identity thieves, online predators, and stalkers. For instance, some participants worriedabout “people figuring out [their] passwords and leaking them to other people” (P13). P14mentioned that “... it’s just scary if they were to figure out what your password is orsomehow reset it. They have access to everything you put on there, cause there’s like alarge part of myself that goes into social media.” Similarly, P5 was concerned about onlinepredators. She reasoned, “because I know that a lot of my friends kind of talk to these guysthat say they’re however old and they have their profile picture to some young attractiveguy, but no. Like there’s more to it, I guess. You can’t really trust that.”Exploratory Study 193.4.3 Feeling SafeAlthough teens expressed having concerns when using SNSs, they felt confident that theyare safe online and they claimed to share a great deal of information online. To better un-derstand teens’ information disclosure behaviours, we asked participants to explain theirperceived safety threats, whether or not they feel safe on SNSs, and the reasons why. Par-ticipants’ generally associated safety with the absence of physical, mental, and emotionalharm. For instance, P19 defined safety as “not getting killed or hurt.” P14 explained thatsafety to her is “not being attacked and not being interfered with, in harmful ways.” P16explained that safety is “to be able to do something without having any mental or physicalor emotional harm being done to you.” He further explained that in online situations,“if somebody else sees [the information you do not wish them to see], it would have anemotional effect on you.”Information safety was mentioned by some participants. Some equated safety to not gettinghacked or having strangers obtain their personal information. For instance, P20 stated thatonline safety is “people not knowing [his] personal information like where [he] lives, what[he does], where [he] often hangs out, and stuff like that. Because if strangers know that,that would be very creepy.” Others claimed that feeling safe is about feeling comfortable,secure, and in control of their situations. P13 defined not feeling safe as the inconvenienceof dealing with problems. She explained that “it’s very unnecessary and time consumingif I were to face an online problem, for example a security one. For example, I don’t wantto not have access to my emails because I was hacked. Then I wouldn’t get the importantemails that I usually get from organizations and things.” Additionally, our participantsincluded unauthorized access to information, re-sharing content without permission, onlinepredators, and bullying into their understanding of safety threats.The majority of participants indicated feeling safe on SNSs. We noticed five major themesin their rationales for feeling safe: (1) Some reasoned that since they take the necessaryprecautions, they are safe online. The precaution methods varied (will discuss them furtherin the next section). For instance, P5 mentioned “I conceal information [online]... I havelike a million of friend requests from people who I have no idea who they are, and obviously,I’m not gonna accept that just for safety reasons.” (2) Some perceived themselves capableExploratory Study 20of dealing with the consequences of a safety breach. For instance, P9 mentioned that if astranger with a fake account messages him, he would delete the person from their socialnetworking account. In explaining why she shares her phone number online, P13 statedthat “because [she] can easily change it”. Similarly, P17 claimed that she would never shareher social insurance number online. She stated: “I don’t have a lot of knowledge on socialinsurance, how the government works with that, and how easy it is to take information fromme and steal my identity. Because I don’t know a lot about it, I don’t want to dip my toeinto it.” (3) Some other participants viewed it as unlikely to have their safety threatenedor get harmed online. For instance, P12 stated: “well, because you hear about the dangersand stuff but then it’s not really, like it could never happen to me right? It would neverhappen to me.” P17 reasoned that not a lot of things have ever happened to her online,and P18 stated he does not think anyone is going to kill him. (4) Some participantsbelieved that they have nothing to hide or no content that would create troubles for them.In particular, participants mentioned “being open” about their lives (P7, P17), and nothaving sensitive or important content online (P4, P12, P13). For instance, P4 mentioned “Idon’t think I have anything important enough worth spying or hiding I guess. As it standsright now, I don’t have a lot of reason to be really secretive.” Similarly, P7 stated “mostof the socializing I do is through the messages, so one to one. I guess in theory peoplecould hack into that, but I don’t see why they would. Nothing I say is very that interesting.There is nothing terribly secretive in my life. If I was like a spy maybe I’d be little moreconcerned, but my life is pretty much an open book.” Lastly, (5) some participants claimedto not be concerned about privacy invasions that have no direct effect on their lives. Forinstance, P16 stated that “if people have problems with Facebook seeing what you search orsomething, I think that’s just fine. That’s what they’re supposed to do. It’s their job butlike for somebody I know to look at something I do without my consent, I’d consider thatas an invasion of privacy.” He reasoned that “[accessing his information] is very unlikelyto have any direct effect on [his] life”.Only a few participants answered “no” to the question of whether or not they feel safe onSNSs, due to potential threats to their physical safety, or/and emotional safety. Focusing onher physical safety, P8 mentioned feeling unsafe while she was having a stalker a few yearsago. Highlighting the information safety, P6 believed that no one is ever safe. AlthoughExploratory Study 21she uses privacy settings, she reasoned: “but I know that it’s still possible that like when Ipost things, people can still get to them.” From slightly a different perspective, P19 statedthat she used to feel safe online before having some relatives “following” her on Instagram.She stated: “I really don’t like it. I used to feel very safe on Instagram because I could bemyself. I could be anyone I wanted to but now I just keep thinking about what [my relatives]going to see.”Protective MeasuresWe identified seven actions participants take addressing their online concerns. These ac-tions are labelled (1)..(7) below:(1) Withholding information was the major protective measure taken by the majority ofparticipants. They claimed that in order to make sure they are being private online, theyavoid sharing “too much” or “personal” information. For instance, P5 stated that “im-portant stuff like your address or anything like social security number, obviously you’renot going to want to give that out.” However, the content that is believed to be privatevaries among participants. For instance, some claimed that they would never share theiraddresses, passwords, social insurance numbers, credit card information, and phone num-bers online, while others refrained from posting status updates, song lyrics, swear words,and “stupid” or “crazy” photos of themselves. Some cited nude or inappropriate photos ofthem smoking or drinking, offensive comments, romantic crushes, and their friends’ secretsas private content that should not be shared online. A few other participants mentionedsharing personal views and stories, phone numbers and credit/debit card information.(2) Addressing security related concerns in particular, some participants acknowledged be-ing careful with their privacy settings. For instance, they set their privacy to “friendsonly”, disabling the option for others to find them through the Facebook search engine,and disabling the geolocation of their posts. (3) Some limit their contacts with strangers.They identified not adding strangers, and blocking or ignoring their messages as their maintactics. (4) Additionally, our participants claimed using more private modes of communica-tion, such as Facebook direct messaging, texting and calling, for sharing personal matters.Exploratory Study 22(5) A number of them described asking family members for help on who can see theircontent.Consistent with our participants’ concern over their social status, (6) they aimed to care-fully consider their audience and the consequences of sharing content online. For instance,P4 always asks himself: “is this something I would be okay with other people seeing?” Hefurther explained that he reviews his posts to make sure they are not offensive as he hasfriends, relatives and family members on his social media. Even those participants whohad their privacy settings adjusted to “friends only” still considered unintended audiences.For instance P18 stated: “I’m aware that if I share something, the person I intended to seemay not be the only one to see it.” Our participants also considered future universities,employers, and whether they will regret their posts in future. (7) Lastly, some mentionedhaving two different accounts on one SNS for a variety of reasons. One participant, forexample, used a public account for her work with kids at a summer camp, and a separateprivate account for sharing her personal content. Interestingly, one participant used tohave a separate Facebook account for sharing content with his friends. Then, he realizedthat having his family on Facebook is beneficial. He said: “I feel like sometimes I mightdo something stupid and it might make me think. Like I’m about to post and in the backof my mind, it’s like your parents will see this. I think it’s better.”3.4.4 Channel SelectionTo better learn about online interactions by teens, their unintentional privacy managementstrategies and other decision-making processes, we asked our participants to describe howthey used each SNSs differently.FacebookThe majority of our participants indicated that Facebook messaging app was a major toolused to directly communicate with their friends and family. Their communications tookplace in two forms: one-to-one messages and group chats. For instance, P3 and P14 usedFacebook to talk to friends who they did not see as often. P1 used Facebook messengerfor group chat, instead of texting: “... I think, if anything, the messenger app would beExploratory Study 23what I use most for Facebook.” P7 explained her use of group chat as follows: “usuallygroup messaging is easier when you’re trying to plan something with a big group of people,like ‘oh, we should all go to the park tomorrow or something. Then you message a wholebunch of people at once so you could, like, get it up there and everybody would see it.Or if it’s, like, a big group of friends from school, I feel like I just post things and saythings that I want all of them to know.” The second most frequently-mentioned activityon Facebook was following updates on their friends, important events, and news. Forinstance, P2 stated: “Facebook is for connecting, kind of seeing what everyone is up to,like, judging them, and, yeah, news. That is the big part of it.” Participants also sharedimportant news with their network. P5 stated: “I think the only things I ever post are,like, if there’s, like, big news or something like about the forest fires. I shared an articleabout that. It was an important thing to me.” We classified such content as impersonalsince they do not reveal much information about the participants. Other examples ofimpersonal content include sharing articles, cat photos, or funny videos. Participants alsoshared more personal content on Facebook, such as photos of themselves, their hobbies,believes, opinion, and political views. For instance, P8 mentioned: “sometimes I sharequotes that I like which is a lot of what people do. What else. Mainly pictures or, like,pictures with my family and family members.” P11 reported using Facebook to share herthoughts: “when I’m posting something, it’s something that irritates me. A topic that Iwant to bring up, like the last post I did, it was a really racist comment made on a website,and I had to defend it. That’s why I shared it to raise awareness like that.” Additionally,participants used Facebook for school tasks, volunteering and job related activities. P14would post on Facebook promotions for the events that she facilitates. P13 and P20 alsoused Facebook for school projects, contacting their teachers, and looking for volunteeringopportunities.The majority of our participants believed themselves to be less active on Facebook. Theymade statements that they “barely ever post,” “don’t post much” or “don’t post often” onFacebook. A reason for not actively posting on Facebook was having there a wider rangeof “friends” that included their family, close friends, school friends, and acquaintances. P8described her friends’ range as “[her] Mom who is 44, with [her] drama teacher who is60, to [her] friends who are as young as 9 years old.” She explained: “with any FacebookExploratory Study 24accounts, you have those people you probably don’t talk to, you met them once, they wantedto add you to Facebook and then you kind of drifted away and you actually don’t knowthem. You know them through other friends, but you don’t talk to them on the daily basis.”Additionally, P11 justified the lack of active posting on Facebook as: “because it documentseverything you say.”SnapchatOn the other hand, participants claimed to be more active on Snapchat and share personalcontent with friends. Such content was believed to be “daily”, “mundane”, or “stupid”photos of themselves. For instance, P1 and P9 use Snapchat to show their friends whatthey are doing. P11 also claimed that Snapchat is very personal to her. She stated: “I justsend funny stuff over. Like when I’m hanging out with friends and then like upload a photoand then like put it on the story I guess. I’d say Snapchat is more like for personal things.”Participants also mentioned sharing “random” content that we classified as impersonal.P7 explained “Snapchat is different because you can send photos and videos with it so ifyou see something cool you be, like, it reminds me of you.” I just send my friends randomstuff that I think they will find amusing.” Only a few participants with Snapchat accountsstated not using it much. They mentioned using Snapchat to mainly watch their friends’videos and pictures.Unlike Facebook, the majority of the participants claimed having only their close friendson Snapchat. P1 stated: “I would say Snapchat is mostly just for me and my close friends.I Snapchat them and don’t think about what I Snapchat.” P11 stated that he trusts hisaudience. She explained: “Even if [my friends] screenshot and save something that istotally like pretty embarrassing, I’d personally ask them to delete it. Like the next time Ihang out with them, I’ll ask them to delete it and show me. Like a lot of people that I haveon social media accounts such as Snapchat, I trust them.” Additionally, our participants’reasoning for why they post more content via Snapchat indicated that they care aboutthe more temporary and casual nature of the platform. For instance, P6 explained: “OnSnapchat, it is more likely for me to post something because it seems, like, less permanent.Like, Snapchat is more like a casual thing than Instagram.” P17 explained: “It’s a lotmore causal than Facebook. It’s always the connotation that comes with it. It’s just likeExploratory Study 25a trend that everybody is, like, subconsciously aware of it. Any day you want to send apicture, you send it through Snapchat.”InstagramOur participants engaged in a variety of activities on Instagram ranging from sharing per-sonal and impersonal content, following updates, and for job-related purposes. The ma-jority of those participants who used Instagram claimed that they use it to showcase theirlifestyle. For instance P5 explained: “with Instagram, I generally just like to post photosof what I’m doing, what I’m eating, you know, like just things I enjoy.” P15’s Instagramaccount was more personal to her than her Facebook account. She reasoned: “I guess onInstagram, people can see what I go through, see my life experiences, who I’m friends with,and what kind of stuff I do in my life.” Additionally, some participants mentioned sharingimpersonal photos such as street arts, sculptures, architecture, or sunsets.Some stated using Instagram mainly to follow their friends’ activities, along with theirfavourite actors/actresses. They stated that “Instagram is more for seeing what otherpeople are doing”, “what people are posting”, “see what everyone is up to”, and follow fanpages or popular accounts. A few participants have public Instagram pages to showcasetheir artistic abilities such as photography. A few participants claimed not using Instagramas much.Lastly, when asked about their followers on Instagram, some participants indicated thatthey are more likely to accept a stranger’s “follower” request on Instagram, comparedto a request on Snapchat and Facebook. For instance, P3 clarified: “because it’s justpictures, I don’t block people I don’t know, but my page is private on Instagram. I can seewho is following me and I can, like, say I don’t want them to follow me.” P5 explainedthat she generally accepts all the follower requests on Instagram: “Just because I feel likethere’s less of a risk. Unless someone direct messages me, there’s not really any kind ofcommunication part and it’s easier to block someone on Instagram. They just can’t seeyour profile anymore.”Exploratory Study 263.5 DiscussionOur exploratory study investigates teens’ online decision-making processes by looking attheir attitudes towards online privacy and analyzing the relationship between their concernsand online interactions. Our goal is to obtain a more comprehensive understanding of howthey use SNSs.After analysing the data obtained from our interviews, we developed a theory about teens’online privacy concerns and behaviours on SNSs. Our theory states that teens do careabout their online privacy and worry about matters on SNSs that directly relate to theirperception of online privacy (See Figure 3.2). More specifically, we theorize that teens’major online concerns that include social status and security concerns, shape their decision-making processes about their information disclosure on SNSs. For instance, when postingcontent online, teens have various justifications to address their online concerns. Thesejustifications range from believing that they have nothing to hide, to believing that theytake measures to stay private on SNSs. For instance, such measures include choosing whichSNS to use in order to communicate with their friends.The following sections explain our theory in more detail and summarize the findings thatmotivate our confirmatory study’s research questions (Section 4).3.5.1 Perception of PrivacyThe results of our interviews suggest that our participants’ notion of privacy is fairlyconsistent with the traditional views by Westin and Altman [67, 68] and findings by Kwasnyet al. [72] on young adults. Additionally, our results confirm the findings of previous studieson teens’ notion of privacy [22, 28, 29, 70]. Our study expands on their conceptualizationof privacy and also highlights their view on privacy invasions.Similar to Marwick and boyd’s findings [22], our participants perceived privacy as havingthe ability to choose which information to share with which audiences. They focusedon controlling the flow of their information by choosing their content and the channels ofExploratory Study 27Figure 3.2: Our grounded theory on teens’ online privacy.communication. This is also consistent with boyd’s assessment [70] of adult Facebook users’view on privacy that focuses on their ability to control “impressions” and social situations.Hence, privacy is about managing the boundaries to access information. Additionally, wefound that our participants’ definition of privacy was also focused on the ability to controlwhat information their audience share about them. In this respect, invasion of privacyExploratory Study 28was viewed by our participants as having their information re-shared without their consent.As a consequence, the audience was believed to be responsible for maintaining the teen’sprivacy. This is in contrast to Dwyer’s findings on young adults who generally viewed theironline content as public [71]. Lastly, our participants’ notion of privacy involved the abilityto engage in online activities without having their activity tracked back to them. In ourconfirmatory survey, we aimed to refine and validate these qualitative findings (see RQ1and Section 4.2.1).Although our participants shared a great amount of information on SNSs, they also believedthat their “private” information should never be shared online. This is in contrast to theview that teens are unconcerned about the security of their information [24]. However,what was believed to be “private” information varied. Different types of information wereassociated with different levels of privacy. Similar to Bryce and Fraser [34], we found thatdisclosure of certain type of information, such as financial or social insurance numbers, wasperceived to be too risky. In such cases, the participants saw themselves as active agents,responsible for keeping their information safe. In our confirmatory survey, we measuredthe perceived privacy level of various types of content (see RQ2 and Section 4.2.2).3.5.2 Online ConcernsThe results of our interviews highlighted the complexity of teens’ thought processes andconcerns when interacting online. We were able to provide a comprehensive, although notexhaustive, list of teens’ online privacy concerns that expands previous studies [31, 76, 83,84]. We identified social status related and security related concerns as the two majortopics of concerns among our participants.Social Status ConcernsOur participants’ major consideration about using SNSs was managing their social sta-tus and image. Their concerns were focused on the popularity of their content, and theaudience’s perception and judgment of them. This finding is similar to the study withundergraduate students reported by Christofides et al. [27].Exploratory Study 29Our participants’ online regrets further emphasized the importance of their self-image. Sim-ilar to adults [30], our participants were concerned about having photos of them uploadedwithout consent. Additionally, they regretted their earlier posts, excessive amount of shar-ing, and posting unimportant content. Such regrets were perceived as embarrassing and/orpotentially damaging to their social status. Self-image concerns were also highlighted bythe description of their thought process when sharing content online. To manage theirreputation, our participants evaluated their content carefully to consider the potential con-sequences of having it shared. This is in contrast to Wang’s finding on college students’ lackof consideration, which leads them to regret their posts later [85]. Such attention to thematters of social status on SNSs is understandable, since teens use social media to shapetheir identity, participate in the society, share their social and political views, and increasetheir social capital and popularity [4, 5, 31]. This is consistent with teens’ development ofabstract and operational thinking [53], which helps them consider an “imaginary” audienceon SNSs that is watching and judging them.Security ConcernsConsistent with previous studies, we found that our participants were generally concernedabout the security aspects of their online interactions, such as unauthorized access to theirinformation, identity theft, online predators, and stalkers. In line with their concerns,we identified that they employ different techniques to maintain the security of their in-teractions. In addition to setting their profile privacy to “friends only”, our participantswithheld information, such as addresses and financial information, used private modes ofcommunication, avoided contacts with strangers, and asked for advice to protect themselvesagainst risks. Our finding is partly consistent with the Fox’s report on American Internetusers’ concerns about strangers accessing their information [73]. However, in contrast tostudies on adults [30], we found that our participants were less concerned about access totheir information by the SNS operators and other online businesses. Some even acceptedthe SNS operator’s access to their content as normal. Additionally, our findings suggestthat, unlike adult Facebook users in Germany [74], teens are concerned by the possibilityof unauthorized access to their passwords and financial information.Exploratory Study 303.5.3 Feeling Safe RationalesWhile the majority of existing studies focused on teens’ perception of privacy and theirinteractions online, a few investigated their rationale for feeling safe on SNSs. Such ratio-nales shed light on teens’ decision-making processes, and address the dichotomy betweeninformation disclosure and concerns. Through our study, we identified five types of expla-nations for a teen’s confidence in their safety on SNSs. These are discussed below and weestimate their prevalence in our confirmatory study (RQ3 and Section 4.2.3).Three explanations—the perceived likelihood of being the target of an invasion, perceivedeffects, and the sensitivity of their content online—highlight our participants’ lack of com-plete awareness on possible privacy and security breaches and their potential consequences.The majority of our participants did not believe that they would be targets of privacybreaches. Even when they did, our participants did not consider themselves at risk unlessthe threat would result in direct emotional and physical harm. They also did not considertheir online content “interesting”, important or sensitive enough for “hackers”, governmentagencies, or others to target. Even if they were targeted, since they do not post “anythingbad”, our participants did not anticipate any serious consequences. This suggests thatteens may lack awareness on the wide range of privacy breaches that may occur other thanbeing hacked. They may also underestimate both the sensitivity of the information theyshare online, such as their lifestyle and contact information, and the potential consequencesof privacy breaches.Fourth, teens confidence in feeling safe increases when they see themselves as being capableof dealing with the consequences of privacy invasions. Comments such as “just delete theperson” or “change my number” suggest that our participants considered the consequencesof their interaction, and, if they found the consequence manageable, they felt safe enoughto continue with their actions. This is similar to the findings by Thomas [86].Lastly, our participants frequently mentioned that they feel safe online because they takethe necessary measures to ensure their safety and privacy on SNSs. In our confirmatorysurvey, we measured the popularity of various protective measures among teens (see RQ4and Section 4.2.4).Exploratory Study 31Choice of SNS as a Protective MeasureOne of the major protective measures taken by our participants was the choice of thecommunication channel for sharing their content. This is partly consistent with Dwyer’sfinding on low switching cost between SNSs that has lessen users’ privacy concerns [71].However, unlike her claim that a switch to another platform suggests that a privacy problemis already encountered, we argue that teens’ choices of the communication channel addresstheir privacy concerns in advance, rather than being a recovery measure.Based on their comments, our participants’ decisions on the selected SNS or other channelwere directly influenced by the audience of that channel. For instance, they discussedpersonal matters with a close friend face-to-face or via Facebook messaging. Similarly,depending on the perceived sensitivity of the content, our participants chose between SNSs.Their Facebook accounts hold the largest and the broadest “friend” lists. As a result, themajority of our participants claimed to “barely” being active on Facebook. They mentionedposting impersonal content such as news, quotes, and interesting articles. Additionally,they claimed sharing what is believed to be general information, such as school name,hometown, gender, age and, in some cases, phone numbers. Such information was describedas content that can be known to everyone. On the other hand, Snapchat was generallyassociated with communications with close friends. In addition to the functionality of theapp that makes the content disappear, having close friends on Snapchat resulted in ourparticipants feeling comfortable with sending “mundane” and “stupid” content to theirfriends. As some explained, they trust their friends with the content. Unlike Facebookand Snapchat, Instagram is used to display one’s lifestyle. Hence, pictures are selected fordisplay, while the “just pictures” nature of the app, makes it feel safe for teens to sharecontent. In our confirmatory survey, we refined and validated our qualitative findingsabout the factors that influence the choice of SNS as communication channels (see RQ5and Section 4.2.5).Chapter 4Confirmatory StudyIn order to refine and validate our qualitative findings, we conducted a confirmatory studyin the form of an online survey with a larger and more representative sample of teenagers.Our survey consisted of multiple-choice and rating scale questions. We received an approvalfrom the university’s research ethics board (application number H17-00785).4.1 MethodologyUpon further reviewing the literature and considering the findings of our exploratory study,we chose to investigate the following research questions through the survey:RQ1: What are the characteristics of online privacy teens care about the most?RQ2: What type of information is perceived among teens to be relatively less/more pri-vate?RQ3: What are the most frequent reasons for not being concerned about posting contenton SNSs.RQ4: What are the measures teens most frequently use to protect their privacy on SNSs?RQ5: What are the factors linked to the teens’ choice of SNSs?32Confirmatory Study 334.1.1 Pilot StudyWe initially tested the survey questions on seven participants (five adults and two teenagers)using similar methodology that is described below. Our adult and teen participants wereable to accurately comprehend the questions. A number of suggestions were made andimplemented to improve the quality of the data. For instance, a follow up question wasadded to better understand the amount of time teens spend on SNSs.4.1.2 Participants, Recruitment, and ConsentThis study involved 94 participants (52% female), ranging in age from 14-17 (14 — 12%,15 — 25%, 16 — 22%, and 17 — 32%). All participants resided in Canada. As Table 4.1illustrates, the province-wide distribution of our sample was fairly similar to the populationof the 14-17 year olds across Canada.Table 4.1: Comparison of the study sample and 2017 Canadian 14-17 year olds popula-tion across provinces [1].Province Participants 14-17 year oldsCanada 94 People 1,558,580 PeopleOntario 40% 40%British Columbia 22% 13%Alberta 17% 12%Nova Scotia 6% 3%Manitoba 4% 4%Saskatchewan 4% 4%New Brunswick 4% 2%NW Territories 2% 0.1%Quebec 2% 21%Newfoundland &Labrador0% 1.4%Prince Edward Is-land0% 0.4%Yukon 0% 0.1%Nunavut 0% 0.2%Confirmatory Study 34The results of our exploratory study indicated that Facebook, Instagram, and Snapchatwere the most popular platforms among our participants. We initially posted online adson both Facebook and Instagram. However, we found Facebook ads ineffective, as only oneparticipant contacted us via Facebook. The remaining participants were recruited throughInstagram.In the ads, we instructed the interested and qualified teens to sign-up by contacting theresearch team via sending a private message on the SNS (Appendix B.1). Once contacted,we sent an initial message to provide more details about the goals of the study, its duration,and the confidentiality of the information (Appendix B.2.1). Parents/guardians consentand participant assent forms were attached to our initial message (Appendix B.2.2).We used an opt-out consent method to obtain permission from teens’ parents/guardians.We instructed teens to read the consent form and if they agreed to participate, present theform to their parents/guardians for permission. Interested teens were asked to let us knowwithin five days whether they obtained permission. In the consent form, parents/guardianswere asked to email within five days, only if they did not wish their child to participate.Each participant received a CAD $5 Amazon electronic gift card.4.1.3 Data CollectionOur online survey (see Appendix B.3), hosted on Qualtrics survey service, consisted ofseveral sections that contained multiple-choice and rating scale questions about partici-pants’ activities on SNS, their SNS connections, their privacy concerns, content sharingpreferences, and demographics.The survey consisted of five sections. In section one, we asked participants to answermultiple-choice questions on their social networking activities such as usage frequency, de-vices used, favorite SNSs, and frequency of posting on SNSs. Section two consisted ofopen-ended and multiple-choice questions that focused on participants’ online friends andfollowers. We obtained estimations on the number of friends and followers participantshave on the popular SNSs such as Facebook, Instagram, Snapchat and Twitter. We wereConfirmatory Study 35specifically interested to learn about the percentage of the friends and followers they com-municate with regularly or have met in person. In section three, we focused on teens’ onlineconcerns and private information. Participants were asked to indicate what informationthey share on their profiles. They were also asked to rate how different statements wouldapply to them. Statements focused on teens’ concerns on unauthorized access to theirshared information and their perceived sensitivity level of various types of information.In section four, we focused on teens’ content sharing behaviours. Participants were againasked to rate how each statement reflected their online behaviours. These statements fo-cused on the influence of different factors on teens’ choice of SNSs when communicatingwith their friends, the privacy protection methods they use to stay private on SNSs, andtheir thought process when sharing content online. Lastly, in section five we obtaineddata on participants’ age, gender, their parents’ education, city they live in, and theirrelationship with their peers.Based on our pilot study, we found the survey took between 10 and 30 minutes to complete.A 10-minute threshold was identified as the absolute minimum amount of time needed toread and answer the survey questions. As such, this was used to exclude responses that werecompleted in shorter times. The average completion time was approximately 22 minutes.Four participants completed the survey in less than 10 minutes, hence, their responses wereeliminated in order to ensure that the integrity of the data was statistically sound. Theremaining 90 participants (52% Female) were included in further analysis.Additionally, to identify careless respondents, we included three attention check questionsin the survey (e.g., option d of Question 10 in Appendix B.3).4.1.4 Ethical ConsiderationsThere were no known risks involved for participating in the study. For the analysis pur-poses, the identities of the participants were anonymized. All data was kept confidential.The anonymity of participants is protected in this report.Other posts to the Facebook and Instagram pages such as likes or follows were not analyzed.Only teens that sent private messages through the Instagram page were considered for theConfirmatory Study 36study and were provided with additional information on the content of the study. Theresearch team had access to the Instagram page and the messages. All communicationswith the participants took place through the same platform.The university’s survey tool (Qualtrics) was used to host the surveys. All data was storedand backed up in Canada. In addition to obtaining initial parent consent, participants wereasked to provide consent electronically by reading the consent statement and selecting“agree” before starting the survey questions. Since we did not ask for any identifiableinformation, participants were informed that they were not able to withdraw their responsesafter submitting the survey. However, if they withdrew before submitting, the incompletesurveys were to be deleted from the server. We did not have any incomplete surveys orparticipants asking to withdraw from the study.4.1.5 Data AnalysisInferential statistics such as the Friedman test [87]1 and pairwise sign test were conductedto check for statistically significant differences among groups to infer behaviours. Once asignificant difference among groups was detected using the Friedman test, we performedthe sign test, a non-parametric paired-sample test on each pair, to identify which oneswere significantly different. All results of the sign tests reported in this thesis are for p-value < .05. To protect against Type I error, the False Discovery Rate (FDR) methodwas used on all comparisons to adjust the p-values for multiple comparisons. Additionally,Kendall’s W (a.k.a Kendall’s coefficient of concordance) was used as an effect size statisticto assess the agreement among participants. The value of W ranges from 0, indicating noagreement, to 1, complete agreement.1Friedman test is a non-parametric statistical test that identifies significant differences in treatmentsamong participants.Confirmatory Study 374.2 Results4.2.1 RQ1: Most Important Privacy CharacteristicsParticipants were asked to rate how much they care about various privacy characteristics(see Question 14 in Appendix B.3) on the scale from 0 (“Don’t care at all”) to 4 (“Extremelycare about”). Privacy characteristics were put into the following statements on teens’ onlinebehaviours:Content: “The ability to choose what information I share online and what I keep tomyself.”Audience: “Once I posted my information online, the ability to control who can accessthat information.”Tagged: “The ability to control what information about me is shared online by others.”Traceable: “The ability to browse web pages and say things online without other peoplebeing able to track me down or look at my activity history.”Self-image: “The ability to control how I am perceived.”The summary statistics of privacy characteristics are presented in Table 4.2. There wasno significant difference among the five characteristic groups based on the Friedman’s test(chi-squared χ˜2 = 3.53, df = 4, and p-value = 0.47). As such, we could not determine thatthe participants cared about some privacy characteristics more than others.Table 4.2: Summary statistics of privacy characteristics.Factor Mean SD MedianContent 3.20 0.80 3Audience 3.08 0.86 3Tagged 3.03 0.94 3Traceable 2.94 1.10 3Self-image 2.84 1.10 3Confirmatory Study 384.2.2 RQ2: Privacy Levels of Online ContentParticipants were provided with a list of content types (see Question 11 in Appendix B.3)and were asked to rate how private they find each type of content (e.g., “Romantic Issues”)on the following scale: (5) “I’d NEVER share this with anyone”, (4) “I’d share ONLY withmy close friends”, (3) “I’d share with all my friends”, (2) “I’d share with all people on myfriends list or followers”, (1) “I’d share with friends of friends”, and (0) “I’d share this withEVERYONE”. The summary statistics of the results are provided in Table 4.3.Results of Friedman’s test (χ˜2 = 1056.5, df = 18, p-value < 2.2e−16) indicated that one ormore types of information are more important for participants to keep as private. Kendall’sW was greater than 0 (Wt = 0.65), indicating some agreement among participants.Pairwise sign tests revealed that the mean ratings of each item in the following list issignificantly higher than the next items on the list and the remaining 11 items: FinancialInfo, Nude Photos, Account PSW, Identifiable Info, Phone PSW, Romantic Issues, andPhone Number. See Table 4.4 for the complete summary.Confirmatory Study 39Table 4.3: Summary statistics of perceived privacy level.Content Mean SD MedianFinancial Info 4.98 0.15 5Nude Photos 4.86 0.59 5Account PSW 4.77 0.43 5Identifiable Info 4.51 0.56 5Phone PSW 4.367 0.53 4Romantic Issues 3.88 0.81 4Phone # 3.44 1.01 4Friends 3.28 1.01 3Personal Views 2.84 1.54 3Email 2.82 1.46 3Hang Out Places 2.62 1.20 3School Name 2.46 1.48 3Personal Photos 2.44 1.22 3City 2.22 1.27 2Activities 2.10 1.28 2Age 1.91 1.36 2Full Name 1.87 1.46 2Sexual Orientation 1.70 1.67 1Favorite Shows 1.26 1.35 1SD stands for Standard Deviation.Confirmatory Study 40Table 4.4: Pairwise comparison of participants’ privacy views on the types of onlinecontent.Mean Content a b c d e f g h i j k l m n o p q r s4.98 a. Financial Info4.86 b. Nude Photos4.77 c. Account PSW4.51 d. Identifiable Info4.37 e. Phone PSW3.88 f. Romantic Issues3.44 g. Phone #3.28 h. Friends2.84 i. Personal Views2.82 j. Email2.62 k. Hang Out Places2.46 l. School Name2.44 m. Personal Photos2.22 n. City2.10 o. Activities1.91 p. Age1.87 q. Full Name1.70 r. Sexual Orientation1.26 s. Favorite ShowsHow to read: Each content type is assigned an alphabetic letter and is listed in the 2nd column(e.g., a. Financial Info). Each cell with the corresponding letter represents that content. Eachcolumn represents a pair-wise comparison between one content (the cell that is highlighted inblue) and the rest. The gray cells indicate contents that are found significantly different from theblue content. If gray contents are above the blue content, they are found significantly higher.4.2.3 RQ3: Justifications for the Lack of ConcernsTo learn about the popular rationale that teens use for not being concerned when postingcontent on SNSs, we asked participants to rate how well the following justifications applyConfirmatory Study 41to their behaviours (Question 15 in Appendix B.3). The rating scale ranged from 0 (“nevertrue for me”) to 3 (“always true for me”).Nothing to Hide: “I don’t believe I have anything to hide.”Unlikely to Happen: “I believe I am unlikely to have my privacy invaded online.”Able to Deal: “If privacy invasions happen online, I am confident that I can deal withthe consequences and stay away from harm.”Take Precautions: “I feel like when I am interacting online, I take the necessary precau-tions to stay safe.”Unconcerned with Authorities: “I do not worry about having my privacy being in-vaded online by authorities and strangers that I don’t know.”Unconcerned with Parents: “I do not worry about having my privacy being invadedonline by my parents and other adults.”Unconcerned with Friends: “I do not worry about having my privacy being invadedonline by my friends/classmates/other peers.”No Direct Influence: “If there is no direct influence on my personal life, I do not mindother people accessing my online content.”The summary statistics are provided in Table 4.5. There was some agreement among ourparticipants (Kendall’s Wt = 0.06) that some of the 14 rationale applied to their onlinebehaviors more than others (Friedman’s test χ˜2 = 35.85, df = 14, p-value < 7.745e−06).Table 4.5: Summary statistics of lack of privacy concern justifications.Rationale Mean SD MedianTake Precautions 2.11 0.69 2Able to Deal 2.06 0.81 2No Direct Influence 1.83 0.96 2Nothing to Hide 1.79 0.88 2Unconcerned with Friends 1.78 0.92 2Unlikely to Happen 1.68 0.85 2Unconcerned with Authorities 1.63 0.98 2Unconcerned with Parents 1.50 1.06 2Confirmatory Study 42The pairwise sign tests revealed that although not significantly different from each other,the mean rating of Take Precautions and Able to Deal were found significantly higher thanthe three rationales of Unlikely to Happen, Unconcerned with Authorities, and Unconcernedwith Parents. See Figure 4.6 for the complete summary.Table 4.6: Pairwise comparison on lack of privacy concern justifications.Mean Rationale a b c d e f g h2.11 a. Take Precautions2.06 b. Able to Deal1.83 c. No Direct Influence1.79 d. Nothing to Hide1.78 e. Unconcerned with Friends1.68 f. Unlikely to Happen1.63 g. Unconcerned with Authorities1.63 h. Unconcerned with ParentsHow to read: Each rationale is assigned an alphabetic letter and is listed in the 2nd column(e.g., a. Take Precautions). Each cell with the corresponding letter represents that rationale. Eachcolumn represents a pair-wise comparison between one rationale (the cell that is highlighted inblue) and the rest. The gray cells indicate rationales that are found significantly different from theblue rationale. If gray rationales are above the blue rationale, they are found significantly higher.4.2.4 RQ4: Tactics for Privacy ProtectionTo learn about popular protective measures that teens use to stay private on SNSs, weasked participants to rate from the scale of 0 (“never true for me”) to 3 (“always true forme”), how much each of the following protective measures described their online behaviours(Question 13 in Appendix B.3):Private Communication: “To share personal matters with friends, I use private modesof communications (such as Facebook Messenger, texting, and calling), rather thansharing the content on my social networking sites.”Confirmatory Study 43No Regrets: “I don’t share something online that I would regret later.”Privacy Settings “I use privacy settings to protect myself online.”Consequences: “I carefully consider who I share my online content with and the conse-quences of sharing it.”No Personal Info: “I don’t share too much about myself online.”No Strangers: “I don’t accept requests from online strangers or respond to their mes-sages.”Not Posting Much: “I don’t post much information online.”Multiple Accounts: “For some social networking sites, I have multiple accounts that Iuse for different purposes (e.g., 2 Instagram accounts).”Delete Unpopular: “I delete content that I’ve posted online, if it doesn’t gather enoughattention.”Delete Received: “I delete some comments and messages that I receive online after Iread them.”Hide Meaning: “When I share things online, I hide the meaning so that only my friendscan understand what I am saying.”Adults Help: “My family and friends help with configuring my privacy settings on socialnetworking sites.”Fake Info: “I provide fake or inaccurate personal information online, such as a fake name,address, phone number, and email address.”Delete Posted: “I delete some comments and other content that I leave on others’ pagessome time after I post them.”Deactivate: “I keep my social networking accounts (e.g., Facebook, Instagram) deacti-vated when I am not using them, to stay private.”Table 4.7 reports summary statistics on these protective measures. There was an agree-ment among participants (Kendall’s Wt = 0.44) that they used some of the 15 protectivemeasures more than others (Friedman’s test χ˜2 = 559.99, df = 14, p-value < 2.2e−16).Confirmatory Study 44Table 4.7: Summary statistics of popular protective measures.Protective measure Mean SD MedianPrivate Communication 2.68 0.58 3No Regrets 2.46 0.80 3Privacy Settings 2.17 0.90 2Consider Consequences 2.08 0.85 2No Personal Info 1.88 0.87 2No Strangers 1.84 0.99 2Not Posting Much 1.83 0.92 2Multiple Accounts 1.54 1.18 2Delete Unpopular 0.96 0.98 1Delete Received 0.94 0.798 1Hide Meaning 0.90 0.67 1Adults Help 0.74 0.96 0Fake Info 0.72 0.81 1Delete Posted 0.52 0.64 0Deactivate 0.51 1.00 0The privacy measures in the above list are shown in descending order of their mean ratings.As such, below we explain only the statistically significant differences between each mea-sure, and the measures below it. The pairwise sign tests revealed that the mean rating ofPrivate Communication was significantly higher than all the remaining 14 measures. Nexton the list, No Regrets was found significantly higher than all the remaining measures ex-cept one (Privacy Settings). Next, Privacy Settings measure was found significantly higherthan 10 out of 12 remaining measures; no significant difference was found with ConsiderConsequences, and No Personal Info. Lastly, the mean ratings of Deactivate and DeletePosted were found significantly lower than all the other protective measures. However, theywere not found significantly different from each other. See Figure 4.8 for a full summary.Confirmatory Study 45Table 4.8: Pairwise comparison on protective measures.Mean Protective measure a b c d e f g h i j k l m n o2.68 a. Private Communication2.46 b. No Regrets2.17 c. Privacy Settings2.08 d. Consider Consequences1.89 e. No Personal Info1.84 f. No Strangers1.83 g. Not Posting Much1.54 h. Multiple Accounts0.96 i. Delete Unpopular0.94 j. Delete Received0.90 k. Hide Meaning0.74 l. Adults Help0.72 m. Fake Info0.52 n. Delete Posted0.51 o. DeactivateHow to read: Each protective measure is assigned an alphabetic letter and is listed in the 2ndcolumn (e.g., a. Private Communication). Each cell with the corresponding letter represents thatmeasure. Each column represents a pair-wise comparison between one measure (the cell that ishighlighted in blue), and the rest. The gray cells indicate measures that are found significantlydifferent from the blue measure. If gray measures are above the blue measure, they are foundsignificantly higher.4.2.5 RQ5: Factors Linked to the Choice of SNSWe asked participants to rate the importance of the following factors in determining whichSNS they use (Question 12 in Appendix B.3). The rating scale ranged from 0 (’‘Not at allimportant”) to 4 (“Extremely important”).Confirmatory Study 46Friends’ Access: “My expectations of how my friends access and use various social net-working sites (e.g., constantly using Snapchat, rarely using Facebook, and don’t haveTwitter).”Content Type: “Type of the content that I am planning to share with my friends (e.g.,text, images, videos, articles, media news, and opinions).”Content Sensitivity: “Sensitivity of the content that I am planning to share with myfriends (e.g., mundane content on Snapchat, lifestyle photos on Instagram, medianews on Facebook, and my romantic troubles via Facebook Messenger).”Adults’ Access: “Whether I want my parents, relatives, and/or other adults to see whatI post online (e.g., whether my parents are my Facebook friends).”Relationship Level: “How close I am with the people that I intend to share my contentwith (e.g., close friends on Snapchat and acquaintances on Facebook).”SNS Constraints: “What the social networking site allows me to do online (e.g., groupchatting, direct messaging, and video calls).”Contact Urgency: “The urgency of the interaction (e.g., calls for asking for a ride andFacebook Messenger for daily updates).”Table 4.9 presents summary statistics of these factors. Results of Friedman’s test (χ˜2 =25.318, df = 6, p-value < 0.0003) revealed that there was an agreement among participants(Kendall’s Wt = 0.05) that one or more factors influenced their choice of SNS more thanothers.Table 4.9: Summary statistics of factors in determining the choice of SNS.Factor Mean SD MedianAdults’ Access 2.57 1.19 2Content Type 2.52 0.85 3SNS Constraints 2.51 1.08 3Relationship Level 2.33 0.95 2Contact Urgency 2.31 1.01 2Content Sensitivity 2.22 .17 2Friends’ Access 2.02 1.06 2Confirmatory Study 47The pairwise sign tests revealed that the ratings of Adults’ Access was significantly higherthan the ratings of Content Sensitivity, and Friends’ Access. Additionally, the ratings ofFriends’ Access was found significantly lower than the three factors of Content Type, SNSConstraints, and Contact Urgency. Complete results of the pairwise comparison test areprovided in Table 4.10.Table 4.10: Pairwise comparison on influential factors.Mean Factor a b c d e f g2.57 a. Adults’ Access2.52 b. Content Type2.51 c. SNS Constraints2.33 d. Relationship Level2.31 e. Contact Urgency2.22 f. Content Sensitivity2.02 g. Friends’ AccessHow to read: Each factor is assigned an alphabetic letter and is listed in the 2nd column (e.g.,a. Adults’ Access). Each cell with the corresponding letter represents that factor. Each columnrepresents a pair-wise comparison between one factor (the cell that is highlighted in blue) and therest. The gray cells indicate factors that are found significantly different from the blue factor. Ifgray factors are above the blue factor, they are found significantly higher.4.3 DiscussionThrough our confirmatory study, we aimed to answer five research questions on teens’interactions on SNSs. In this section, we summarize our findings. In Chapter 5, we discussthe findings of our exploratory and confirmatory studies in more detail and compare themwith previous findings.Confirmatory Study 48RQ1: Most Important Privacy CharacteristicsDue to the lack of statistically significant differences in the data, we are unable to identifywhich characteristics teens care about most with respect to their online privacy.RQ2: Privacy Levels of Online ContentThrough our analysis, we found that teens perceive financial information, nude photos,account passwords, identifiable information (e.g., home address and SIN), phone passwords,and romantic or family issues as types of private information that they would only sharewith their close friends or never share with anyone at all.RQ3: Justifications for the Lack of ConcernsOur results highlight that the presence of adults in a SNS is perceived as more influentialin determining teens’ choice of SNS, compared to the sensitivity of the content teens intendto share or the degree of their friends access to the SNS.RQ4: Tactics for Privacy ProtectionOur results indicate that using private modes of communication, not sharing regrettablecontent online, and using privacy settings are among the most popular strategies teensutilize to protect themselves on SNSs.RQ5: Factors Linked to the Choice of SNSLastly, we found that the two justifications of taking necessary precautions online and beingcapable of dealing with consequences of privacy invasions are used often among teens. Thisis compared to the beliefs that privacy invasions are unlikely to happen to teens, or thatthey do not worry about having their privacy invaded by authorities, strangers, parents,or other adults.Chapter 5Discussion and ImplicationsIn this thesis, we sought to expand our current understanding of Canadian teens’ notionof privacy and online interactions in order to better communicate with them and empowerthem to exercise best safety practices. To this end, we extend past findings of three differentaspects of online privacy with regards to SNS usage: (1) teens’ general perception of onlineprivacy, (2) their online concerns, and (3) decision-making processes about informationdisclosure. In the following sections, we discuss our findings, compared with other relatedresearch on adults and teens, and highlight implications.5.1 Perception of Online PrivacyTo shed more light on teens’ view on online privacy, we first asked our participants todescribe it in their own terms and provide us with examples of privacy invasions. Next,in our confirmatory study, we aimed to identify online privacy characteristics that ourparticipants care about the most along with the information that they perceive as more/lessprivate.49Discussion and Implications 505.1.1 What is online privacy?Understanding the core aspects of teens’ conceptualization of privacy is key to refiningprivacy theories. It is also necessary for designing better options to help teens to be activeagents in protecting themselves online. Previous studies found that teens view their onlineprivacy as the ability to control (1) what information they share on SNSs, (2) with whomthey share that information, and (3) how they are perceived online [22, 70]. Similar to youngadult’s conceptualization of privacy [72], our findings suggest that teens have a broaderdefinition of privacy compared to the adults. We found that their definition includes theability to control (4) what information other people share about them and the ability to(5) engage in online activities without having their activity traced back to them. The lackof statistically significant differences in the data for RQ1 from our confirmatory study(Section 4.2.1) warrants further investigation of whether these five abilities constitute thecore privacy characteristics that form teens’ notion of online privacy.5.1.2 What is private?Previous research has found that perceived sensitivity of the personal information hasa great impact on the perception of online privacy [63]. In 1993, Weible has defined“information sensitivity” as “the level of privacy concern an individual feels for a type ofdata in a specific situation” [88, p. 30]. Information sensitivity among adults varies fromperson to person as well as by situation [89]. Additionally, some types of data are moresensitive than others [89]. For instance, financial and medical information are consideredhighly sensitive [90], whereas, information about product purchases and media habits isperceived as less sensitive [63, 89].Similar to findings on adults and on young people by Bryce and Fraser [34], our study con-firmed that teens perceive some types of information as more private than others. Throughour confirmatory study, we expanded and elucidated this list: information linked to therisk of physical threats and real-life consequences (e.g., addresses, financial information,passwords, and phone numbers), content with unfamiliar invasion consequences (e.g., so-cial security numbers), “stupid” or “crazy” photos, nude or inappropriate photos (e.g.,Discussion and Implications 51while smoking or drinking), offensive comments, romantic crushes, and relationship sta-tuses. Among these items, our findings suggest that teens find financial information, nudephotos, account passwords, identifiable information (e.g., social insurance numbers), andmedical information to be the most private and they are less likely to share them withanyone. They agreed that other information, such as phone passwords and romantic orfamily issues (crushes, breakups, family dramas), is less private. Hence, they would bewilling to share such information but only with close friends.Such findings can help parents and educators better understand the extend to which varioustypes of information are shared online. Although it is safe to say that teens’ privacyintentions are to keep perceived risky personal information to themselves, future studiesshould examine cases where teens’ privacy intentions and actual behaviours differ.5.2 Online Privacy ConcernsThe results of our interviews provided a more comprehensive list of teens’ privacy concernson SNSs (Chapter 4, section 3.5.2). Although most online safety educational curriculatarget security concerns (e.g., stranger danger messaging) [29], teens’ concerns about theirsocial status play a major role in influencing their online interactions.Our findings can help parents and educators design privacy protection programs that betteraddress teens’ online concerns to improve their privacy on SNSs. For instance, targetingteens’ concerns about social status and explaining the potential invasion consequences canhave a stronger influence in encouraging them to stay safe on SNSs.5.3 Decision-making Process on Information DisclosureIn this thesis, we sought a better understanding of teens’ online decision-making processes.This included diving deeper into their online privacy protection measures, the factors thatinfluence their choice of SNSs, and identifying major rationales they have to justify theirlack of privacy concerns.Discussion and Implications 525.3.1 Determining the choice of SNS.Studies on adults have shown that users have distinct attitudes toward different SNSs [91]and each SNS has its unique purposes [92]. For instance, Dwyer et al. [91] reported thatcompared to MySpace users, Facebook users trust Facebook and its members significantlymore. They are also more willing to share personally identifiable information on Facebook.Other factors, such as the gained gratifications on a platform, influence users’ choice ofSNS [93]. For instance, Twitter users gain social capital on weak and distant ties witha diverse network [94]. On the other hand, Snapchat users gain gratification on strongrelationships and emotional support from their interactions [93, 95].The results of our study suggest that teens’ privacy concern is another factor that influencestheir choice of SNSs. We found that to protect their privacy, teens make decisions aboutwhich SNS to use. Our findings expand the list of the previously reported factors andidentified those factors that are more influential. In addition to considering aspects suchas the audience’s consumption of SNSs, technical affordance, timing, and disruption (e.g.,calls for urgent, text for non-urgent matters), their friends availability on SNSs, costs [70],and the presence of parents and other adults on SNSs [96], teens consider other factorssuch as their relationship with the intended audience (e.g., close friends on Snapchat,acquaintances on Facebook), types of activity, content type, and the perceived level ofsensitivity of the shared information (e.g., mundane photos on Snapchat, lifestyle photoson Instagram, and news articles on Facebook). Further, we found that adults’ presenceon a SNS (i.e., whether teens want their parents, relatives, and/or other adults to seewhat they post online) is a more important factor than the sensitivity of the content theyintend to share or their expectations of how their friends consume SNSs. Surprisingly,this last factor about how their friends use SNSs was found to be the least important indetermining teens’ decisions, compared to the type of the content they intend to share,technical affordance, how close they are with their audience, or the timing and disruptionof the interaction.Our findings contribute to the knowledge base on teens’ online decision-making processes,which can help future researchers better understand how they use SNSs. For instance,future research should consider the importance of teens’ concerns about their social statusDiscussion and Implications 53when investigating their use of various social networking platforms. Lastly, teens’ choice ofSNSs and their activities on each platform should be considered when providing them withrelevant privacy information. For example, talking about privacy breaches on Facebookmay not resonate with teens, if they mainly use Snapchat to communicate with theirfriends.5.3.2 Popular protective measures.In a study about disclosure of personal information on SNSs, Young and Quan-Haasereport that adults’ privacy protection measures include exclusion of personal information,using privacy settings, and use of private modes of communications such as emails [97].Additionally, findings by Tufekci [98] on college students highlight that students also useprivacy settings to alter the visibility of their online profile and use nicknames, instead ofreal names, to protect their privacy.The results of our research extend past findings and provide a more comprehensive list ofprotective measures that teens employ to address their privacy concerns. Previous studieson teens [22, 32, 70] report such measures as using privacy settings, encoding the content(referred to as “steganography”), manipulating the technical affordance (e.g., deactivatingtheir account during the day or deleting all communications after), concealing or obscuringinformation, providing inaccurate identifiable (e.g., name, age, and location) and otherpersonal information, refusing to provide information, maintaining multiple profiles, andflaming.1 Our findings suggest that teens’ protective measures also include the avoidanceof sharing “too much” or “personal” information online, using privacy settings and SNSsfeatures such as deleting content or accounts, limiting contacts with strangers, using moreprivate modes of communications (e.g., Facebook direct messaging, texting, and calling) forsharing personal matters, asking family members for help on who can see teens’ content,and carefully considering the consequences of sharing their content online. We furtheridentified private modes of communication as the most popular protective measure used byteens, followed by avoiding sharing something online that they would regret later. After1Flaming is the act of sending highly negative messages to those sending unsolicited emails.Discussion and Implications 54these two measures, the use of privacy settings was found more popular than many (but notall) other measures, such as not adding strangers, claiming not to post much, or deletingcontent. As expected, temporary deactivation of accounts and deleting content that theyhad posted online were among the least popular measures. Surprisingly, providing fake orinaccurate personal information was also among the least used protective measures.These findings can help parents and educators better communicate with teens on theirapproaches to privacy to help them navigate and manage the online risks. Additionally,future research is needed to measure the effectiveness of these measures in order to clarifyand address potential shortcomings.5.3.3 Justifications for the lack of concerns.Our study revealed major rationales for teens’ lack of concern, when they share contentonline. Teens justify by claiming that they take the necessary precautions to stay safe. Theybelieve that they can deal with the consequences of privacy invasions, security breachesare unlikely to happen to them, and that they have nothing to hide or nothing important.This is similar to research findings on adults and older teens [37, 74]. Krasnova et al. [74]report that perceived likelihood of a privacy violation is a primary determinant of adults’self-disclosure on SNSs. Additionally, it was found that older teens who believe that theyare capable of protecting themselves on SNSs have less privacy concerns.Our findings highlight that teens also believe that access to their information by SNSs’operators or government has no direct effect on their lives. Our findings further identifythat the two justifications—taking the necessary precautions and being able to deal withthe consequences—are more common among online teens. This is compared to the beliefsthat privacy invasions are unlikely to happen to teens, or that they do not worry abouthaving their privacy invaded by authorities, strangers, parents, or other adults.Our findings provide a deeper insight on teens’ online behaviours, which can help with thedevelopment of models to explain online information disclosure and privacy paradox.Discussion and Implications 555.4 LimitationsOur work has several limitations. Due to the self-reported nature of the data, a commonissue with individual interviews and an online survey is the possibility of inconsistencies be-tween participants’ descriptions and their actual behaviours. Participants’ desire to portraya positive self-image or memory limitations may influence their responses. However, usingself-reported data collection method in exploratory studies is essential for an unconstrainedexploration of the subject matter [99].Our results are constrained by the data we obtained from a sample of participants recruitedthrough one SNS (Facebook for the exploratory study and Instagram for the confirmatorystudy). Additionally, our participants were teenagers who responded to online ads. Hence,since our sample is not representative of all teenage Canadian SNS users, we are unable togeneralize the findings of our confirmatory study. For instance, teens who use other SNSsand never respond to online ads may potentially display different online behaviours. Furtherresearch with a larger and a more representative sample of the population is necessary toestablish the external validity of our findings.Furthermore, we did not factor in gender and individual differences in analyzing the resultsof our studies. Previous research has indicated the importance of examining such factorsin online behaviors [100]. A study on adults Facebook users found significant differencesbetween genders in SNSs usage purposes [101]. For instance, males favored making newcontacts, whereas females favored maintaining existing relationships. Such differences canhelp us better understand teens’ online decision-making processes and online perceptions.Lastly, in qualitative studies such as our exploratory study, the researchers’ life experi-ences, personal views and cultural background can influence the insight obtained from thedata [81]. For instance, during the data collection and coding process, the researcher’sthoughts can affect the design of the interview guide and the interpretation of the collecteddata. Inevitably, that will influence the categories that emerge from the data [102]. Sinceit is impossible to completely void our minds, we took measures to reduce the researcher’sbias through triangulation [82]. Investigator triangulation [103] was done by involving anDiscussion and Implications 56expert in designing the interview guide and conducting the interviews. The emergent the-ory was also discussed with multiple researchers to ensure that our analysis was grounded inthe data. Methodological triangulation [103] was done by obtaining data through multiplemethods including interviews and an online survey,Chapter 6ConclusionOur work provides a detailed look at a diverse group of Canadian teens’ privacy-relatedconcerns, online decision-making processes, and rationales when using SNSs. We conductedqualitative (N = 20) and quantitative (N = 94) studies with Canadian teens recruited fromFacebook and Instagram (hereafter, teens).Exploratory StudyGrounded on the data we collected, we developed a theory that teens do care about theironline privacy and their online concerns are directly linked to their conceptualization ofonline privacy. We theorized that teens’ concerns shape how they share content on SNSs.For instance, teens have various justifications to address their privacy concerns and feelsafe on SNSs. One important justification is taking measures to protect their privacy andsafety which includes choosing which SNS to use.Through our exploratory study, we confirmed previous findings that privacy among teensincludes the ability to control their content, audience, and the image they portray on SNSs.However, we found that teens have a broader notion of privacy. They also view privacy asthe ability to control what information is shared about them by others and the ability toengage in online activities without other people tracing the activities back to the teenagersor interfering with those activities.57Conclusion 58We found that teens describe information linked to physical threats or real-life consequencessuch as addresses, financial information, passwords, and phone numbers as private. Theyconsider contents (e.g., social security numbers) with unfamiliar privacy invasion conse-quences as private. Additionally, “stupid”, “crazy”, nude or inappropriate photos of themsmoking or drinking, offensive comments, and their romantic crushes were also perceivedas private.Furthermore, We identified social status and security concerns as the two major categoriesof teens’ online privacy concerns. We found that the majority of teens’ information disclo-sure behaviours are influenced by the former.Despite these concerns, our results highlighted that teens have various justifications tofeel safe on SNSs. (1) Teens take measures to protect their privacy online. (2) They feelthat they are capable of dealing with the consequences of a privacy invasion and (3) thatsecurity breaches are unlikely to happen to them. (4) Teens believe that they have nothingto hide and (5) they are not concerned if a privacy invasion has no direct effect on theirlives.Further analysis of teens’ protective measures confirmed previous findings that teens takemeasures such as (1) using privacy settings, (2) manipulating the technical affordance(e.g., deleting comments and messages), and (3) maintaining two different accounts on oneplatform in order to protect their privacy on SNSs. We also found additional measuresthat teens take to ensure their privacy and safety online. (4) Teens avoid sharing “toomuch” or “personal” information online. (5) They limit their contact with strangers, (6)use private modes of communication such as Facebook direct messaging, texting, or callingfor sharing personal matters, (6) ask family members or their friends for help on who canview their content, (7) carefully consider the consequences of sharing their content online,and (8) decide which SNS to use in order to communicate with their audience.Lastly, the results of our analysis highlighted that teens consider various factors in decidingwhich SNSs to use. Confirming previously existing findings, (1) we found that teens accountfor their intended audience (2) how they consume different SNSs, (3) the technical affor-dance (what the platform allows them to do), and (4) the cost (e.g., free WiFi) in choosingConclusion 59which SNSs to use. (5) We added that the content type (e.g., photos, text or videos) and(6) the perceived level of content sensitivity (e.g., mundane photos on Snapchat, lifestylephotos on Instagram, News articles on Facebook) also influence teens’ decisions.Confirmatory StudyIn our confirmatory study, (RQ1) we failed to identify which privacy characteristics teenscare about the most. (RQ2) We confirmed that teens perceive certain types of contentas more private than others and provided a more detailed ranking of their response. Ourfindings revealed financial information, nude photos, account passwords, identifiable in-formation (e.g., home address and SIN), phone passwords, and romantic or family issuesas types of private information that teens would only share with close friends or nevershare with anyone at all. (RQ3) We highlighted the major influential rationales in teens’lack of privacy concerns. We found that teens justify they are safe online because theytake the necessary precautions and that they are capable of dealing with consequences ofprivacy invasions. (RQ4) We confirmed that similar to adults, participants in our samplecare about their privacy and take measures to stay safe online. Our findings highlight avariety of protective tactics that teens employ to manage their privacy. In particular, weobserved that using private modes of communication, not sharing regrettable content, andusing privacy settings are among the most popular tactics.(RQ3) Lastly, the results of ourstudy show that teens’ choice of SNS is yet another protective measure. Focusing on thefactors that influence this choice, we identified the presence of adults in a SNS as a keyfactor that is perceived as more important than the sensitivity of the content teens intendto share or their friends access to the SNS.Our findings have implications for parents and educators. Through considering teens’point of view, their online privacy concerns, and how they interact on SNSs, parents andeducators can more effectively communicate with teens about the consequences of privacyinvasions. Additionally, through further research, we should investigate the effectivenessof teens’ protective tactics and highlight potential shortcomings in order to empower teensto better protect themselves from adverse consequences.Appendix AExploratory StudyA.1 Pre-Interview Survey1. How old are you?——-2. What grade are you in? (Mark only one.)© 8th, © 9th, © 10th, © 11th, © 12th3. What do you identify your gender as?© Female, © Male, © Transgender, © Prefer not to answer4. With whom do you live? (Mark only one.)© Both parents, © Mother, © Father, © Siblings, © Roommates, © Other:——-5. What is the highest level of education that your parents/guardian have? (Mark onlyone.)© Less than high school, © High school, © Some college, © Bachelor, © PostGraduate, © Not sure, © Prefer not to answer, © Other:——-6. How often do you use internet on your phone, computer, tablet or other mobiledevice(s)? (Mark only one.)© Daily, © Almost constantly, © Several times a day, © About once a day, ©Weekly, © Several times a week, © Once a week, © Less often60Appendix A: Exploratory Study 617. Please select the devices that you have or have access to. (Check all that apply.) A smartphone,  A cellphone that is not a smartphone,  A desktop computer, A laptop,  A tablet (e.g., iPad, Samsung Galaxy),  Other:——-8. Which of the following social media services do you use? (Check all that apply.) Facebook,  Twitter,  Snapchat,  Tinder,  Instagram,  Google+,  Vine, Tumblr,  Other:——-9. Which of these social media services do you use MOST often? (Check all that apply.) Facebook,  Twitter,  Snapchat,  Tinder,  Instagram,  Google+,  Vine, Tumblr,  Other:——-10. How many friends do you have on Facebook? (Mark only one.)© I don’t use Facebook, © 0 - 100, © 101-200, © 201-300, © 301-500, © 501-1,000,© More than 1,000 friends, © Not sure, © Prefer not to answer11. How many friends do you have on Twitter? (if applicable.)© Mark only one oval., © I don’t use Twitter, © 0 - 100, © 101-200, © 201-300,© 301-500, © 501-1,000, © More than 1,000 friends, © Not sure, © Prefer not toanswer12. How many friends do you have on Instagram? (if applicable.)© Mark only one oval., © I don’t use Twitter, © 0 - 100, © 101-200, © 201-300,© 301-500, © 501-1,000, © More than 1,000 friends, © Not sure, © Prefer not toanswer13. What information do you share online? (Check all that apply.) Full Name,  Age,  Phone Number,  Email Address,  Home Address,  SchoolName,  Medical Information,  Credit Card,  Favorite TV Show,  FavoriteSnack,  Other:Appendix A: Exploratory Study 62A.2 Interview ScriptIntroductionHello, good morning/afternoon. My name is ** and my colleague’s name is **. We willbe moderating your interview today. Can we get you a glass of water or anything else todrink?To begin, we would like you to review this assent form. It contains important informationabout today’s interview. If you assent to the terms and would like to participate in thestudy, please sign the form and hand it back to us.You and your classmates are one of the first groups of people in the world to grow up usingcomputers and Internet. We want to learn more about what it is like for you to grow upusing computers and the Internet, and what we can do to make these things more safe andfun. Our interview will take approximately one hour. You are free to choose not to answerany questions, or to stop the interview at any point if you feel uncomfortable. We greatlyvalue your honest and candid responses.We would like to make an audio recording of this session. This recording will only be usedfor the purposes of this study and will only be accessible to the researchers. Do you consentto having this session audio recorded?Interview QuestionsPrivacy[To learn about online privacy, private information, and invasion of privacy.]• How would you explain privacy to someone? What does (online) privacy mean toyou? How would you describe privacy?• What is private information to you? Examples?• What are some content that you won’t share with anyone?• What is the most personal content that you have shared online (posted or throughFacebook messenger)?Appendix A: Exploratory Study 63• What would be an example of something that you will never share online? Why isthat?• How would you feel if your friends share photos of you? Or tag you in posts?• What do you consider as an invasion of your privacy?• How would you feel if someone accesses the information that you never share onlinewithout your consent? (eg., on your phone or laptop)• What would be some consequences of that?• How likely do you think this is to happen?• How would you feel if someone gain access to the information that you have sharedon social media without your consent?• What would be some consequences of that?• How likely do you think this is to happen?• How would you feel if websites or businesses access the information that you shareon social media?• What would be some consequences of that?• How likely do you think this is to happen?• How would you feel if your friends re-share your content without your consent?• What would be some consequences of that?• How likely do you think this is to happen?Safety• How do you explain online safety to someone?• Do you feel safe online?• Why do you think you need to be safe?• What are some online safety threats? Can you give me some examples?• What experiences have you had that made you more sensitive to online privacy?Experiences that now make you be more cautious? (For example cyberbullying orsomeone stalking you)• Have you had any negative experiences online? Any situation that has made youuncomfortable?Appendix A: Exploratory Study 64Concerns[To learn about their online concerns on each platform, and what info they see as private.]• What are some of your privacy concerns?• What was the most recent content that you shared online?• What do you think about right before sharing something online?• How about when commenting? Or liking something online?• How is your thought process different when sharing something on Facebook, vs.Instagram, vs. Snapchat, vs. Twitter?• Do you think more about staying safe online, or presenting a good image?• If you want to share something very private, which SNS do you choose?• Which platform do you feel you have the most privacy?• Which platform do you feel more comfortable to share content on?• Do you feel safer on one platform versus another? Why?• Have you ever posted something and regretted it later?• If you could start over with your social networking accounts, what would you dodifferently?Protective Measures[To learn about what platforms they use and for what purposes. Delving deeper into theirprivacy settings for each SNSs and how that helps them to stay safe.]• What do you do to stay private?• What safety measures do you take?• Can you tell me about how you use different SNSs?• What makes you use them differently?• Which one do you use more often?• How is your group of friends different on each SNS?• What kind of information do you have on your Facebook profile?• Can you tell me about the privacy settings of your accounts? How are they differentfrom each other and why?Appendix A: Exploratory Study 65• who can see you posts? Do you review all the posts and things that you’re taggedin? Who can send you friend requests? Who can post on your timeline? Who cansee what others post on your timeline?• Do you share your passwords with anyone? Why?• How do you compare your online behaviour with your friends ( or people you know)?Are they more active than you? More private/public compared to you?Appendix BConfirmatory StudyB.1 Instagram Recruitment AdsFigure B.1: Instagram ads posted on July 24 and August 1,2018.66Appendix B: Confirmatory Study 67Figure B.2: Instagram ads posted on August 6 and August 14, 2018.Appendix B: Confirmatory Study 68B.2 Confirmatory Study Participant RecruitmentB.2.1 Initial Contact MessageProject Title: Understanding Influential Factors on Teenagers’ Online ConcernsHi [Participant’s Name]Thank you for your interest! The aim of this research is to learn about teenagers’ experi-ences with socializing online. Through an online survey that will take about 30 minutes,you will be asked questions about your interactions with social networking sites. Thissurvey is anonymous. Your responses will stay confidential. There is no link between youand your responses, and they are only accessed by the research team.Attached, please find a consent form that explains the interview process and goals inmore detail to you and your parents/guardians. Please read carefully. Once you decideand agree to participate in the study, you are required to pass the consent form to yourparents/guardians. Please get back to us no later than [Date, 5 days after this message issent] to inform us about their decision. Once they grant you permission, we can set a datethat works best for you in order to send you the link to the survey.You will be compensated for your time with a $5 Amazon gift card.If you have further questions, please do not hesitate to ask. We appreciate your help!B.2.2 Parental and Teen/Participant Consent FormProject Title: Understanding Influential Factors on Teenagers’ Online ConcernsPrincipal Investigator: Konstantin Beznosov, Professor, Dept. of Electrical and ComputerEngineering, [phone number],[email address] Co-Investigator: Salma Kashani MSc., Dept.of Electrical and Computer Engineering, [phone number], [email address]Appendix B: Confirmatory Study 69You (the teenager) are invited to participate in a research study conducted by the Univer-sity of British Columbia that will be occurring in the coming weeks. Please take a momentto review this information about the study.Purpose: Social networking sites (SNSs) are an integral part of many teens’ lives. Asreported by MediaSmarts in 2014, an overwhelming majority of Canadian students (9 to17 years-old) are Internet users. Of those, 57% reported Facebook as their favourite website.More than 40% indicated some social networking related activities, such as posting contentand following friends, as their most frequent online activities. Our study investigates howteens socialize online.Procedure: This study involves having you (the teenager) complete an online survey aboutyour online behaviours. You will be asked about your access to the Internet, what devicesyou use, and which SNSs you use often. You will be asked to rate different aspects of yourexperiences on SNSs. To gain a better sense of who you are, we ask demographic questionssuch as age (year and month of birth), gender and parents’/guardians’ education levels.The survey will take approximately 30 minutes to complete. You can access the surveyfrom any location of your choice, using any computer or mobile device.While there are no known risks associated with this study, should you feel uncomfortable,you have the right to withdraw from the study without any penalty, before submitting thesurvey. Once submitted, due to the anonymous nature of the survey, you will not be ableto withdraw your responses.To ensure that you are feeling safe to respond to the survey as honestly as possible, allthe information provided will be kept confidential. All the identifying information thatincludes the social networking communications will be kept separate from the main surveydata in different password-protected electronic folders. Only individuals in the researchteam will have access to the Facebook, Instagram or Snapchat messages, and the surveydata. Pseudonyms and modified information will be used to reference you during the studyand when reporting the completed study. Communication through Facebook, Instagramor Snapchat will be retained for five years and deleted after. All electronic files will bestored in password-protected folders in secure computers or a USB portable drive that isAppendix B: Confirmatory Study 70kept in locked filing cabinets at the University of British Columbia. These files will bepermanently deleted after five years.The results of this research will contribute to Salma Kashani master’s thesis, which will beavailable publicly.To express our appreciation for participation in this study, each participant will be of-fered an online Amazon gift card worth $5. You will receive the gift card independent ofcompleting or withdrawing from the study.If you have any concerns or complaints about your rights as a research participant and/oryour child’s experiences while participating in this study, contact the Research ParticipantComplaint Line in the [University] at [phone number] or if long distance e-mail [emailaddress] or call toll free [phone number].To Parents: If you DO NOT wish your child to participate in this study, please emailus within 5 days to [email address], providing the name of your child and indicating yourunwillingness for your child to participate in this study. We will confirm your email message.If you have questions or desire further information about this study, please contact SalmaKashani.Appendix B: Confirmatory Study 71B.3 Online Survey QuestionsConsent Statement Thank you very much for agreeing to participate in our studyby completing this survey. This anonymous survey gathers information on your onlineconcerns and behaviours when using social networking sites (SNSs). While you are notobligated to respond to every question, answering as many questions as possible will helpus gain a better understanding on how teens interact online. The provided informationwill not be used in a manner which would allow identification of your individual responses.Please note that completing this survey will demonstrate that you give your consent fordata to be analyzed, and the results be published at scientific conferences and journals.You acknowledge that your participation is voluntary and that your refusal to participateinvolves no penalty. You may discontinue participation at any time before submitting thesurvey. The incomplete surveys will be deleted from the server. However, due to theanonymous nature of this survey, you will not be able to withdraw your responses oncethey have been submitted. AgreeSection 1: Your Social Networking Activities The following questions help usunderstand how much time you are spending online and what social networking sites youuse.1. During the past 6 months, how frequently have you used social networking sites (forexample, Facebook, Messenger, Snapchat, Instagram, and Twitter) for things suchas browsing, watching videos, posting updates, or texting and chatting with friends?(Select the most appropriate choice.)© I don’t use social networking sites, © Almost never,© A moderate amount, © Almost alwaysAlthough it is difficult to estimate, approximately, how much time do you spend onsocial networking sites? [ ]Appendix B: Confirmatory Study 722. What devices do you use to access social networking sites? (Select all that apply) A smartphone,  A tablet (for example, iPad, Samsung Galaxy), A laptop,  A desktop computer,  Other[ ]3. Which of the following social networking sites have you been active on (for example,browsed through or posted on) in the past 6 months? (Select all that apply) Instagram,  Twitter,  Google+, Tumblr,  Vine,  Facebook,  Snapchat,  Other[ ]4. During the past 6 of months, on average, how frequently have you shared or postedcontent (for example, photos, videos, articles, opinions, media news) on your socialnetworking accounts for people in your network to see? This includes posting sto-ries on Snapchat, photos and videos on Instagram, status updates on Facebook, orretweeting on Twitter.© Never, © Once every few months, © Once a month, © Few times a month, ©Few times a week, © Few times a daySection 2: Friends & Followers The following questions help us understandwhom you interact with through social networking sites.5. Think about the social networking sites where you have your own online profile. Listbelow the number of friends or followers you have on your profile (your best guess). Facebook [ ],  Twitter [ ],  Snapchat [ ],  Instagram [ ]6. Of the friends and followers you have on the following social networking sites, howmany do you communicate online regularly with?SNSs No one 10% 30% 50% 70% 90% Everyone Don’t have an accountFacebook © © © © © © © ©Snapchat © © © © © © © ©Instagram © © © © © © © ©Twitter © © © © © © © ©7. Of the friends and followers you have on the following social networking sites, howmany have you also met in person?Appendix B: Confirmatory Study 73SNSs No one 10% 30% 50% 70% 90% Everyone Don’t have an accountFacebook © © © © © © © ©Snapchat © © © © © © © ©Instagram © © © © © © © ©Twitter © © © © © © © ©8. Internet users continue to spend more time on social networking sites than on anyother type of site. Observers have noted a range of positive and negative impactson users’ behaviours. Now, think about the last time you used a social networkingsite to connect with your friend. In order to demonstrate that you have read theinstructions, please ignore the items below. Instead, simply click the other optionand in the corresponding box, enter the text: I read the instructions.© Facebook [ ], © Twitter [ ], © Snapchat [ ], © Instagram [ ], © Other [ ]Section 3: Online Concerns & Private Information The following questionshelp us understand what you are concerned about and consider as private when usingsocial networking sites.9. What information do you have in your social networking profiles such as your Face-book, Instagram, and Twitter? Full name,  Favorite TV shows/ movies,  Photos of yourself,  Political views, Photos of your friends,  Bio (a short description of who you are),  Photos ofyour family,  Where you live (city name),  Age,  Places you’ve lived,  Phonenumber,  Gender,  Email address,  Birth year,  Website,  Birthday (year,month, day),  School name,  Relationship status,  Work experiences,  Familymembers,  Credit Card Information,  Check-ins (places you’ve visited such asrestaurants, cafes, museums).10. Indicate how concerned you are about people other than your friends seeing thefollowing things online:Concern levels: © Not concerned at all (anyone could see it), © Not very concerned,© Somewhat concerned, ©Quite concerned, © Extremely concerned (nobody shouldsee it).(a) My entire profile on Facebook (or other social networking site).(b) Photos and videos that I have posted of myself online.(c) Photos and videos I have posted of others online.Appendix B: Confirmatory Study 74(d) Please select ”quite concerned” option.(e) Photos or videos that others have posted of me online (and tagged me in).(f) Comments to other people that I have posted online.(g) Comments to me from other people that are posted online.(h) General information about myself (like gender, birth date, hometown).(i) Personal information about myself (like my interests, activities, favorite movies,and relationship status).(j) Contact information for myself (like my email, home address, and phone num-ber).(k) Direct messages between my friends and I.11. Please rate how private is the following information:Levels: © I’d NEVER share this with anyone, © I’d share ONLY with my closefriends, © I’d share with all my friends, © I’d share with all people on my friends listor followers, © I’d share with friends of friends. © I’d share this with EVERYONE.Types of Content: Full name, City I live in, Home address, SIN, and medical informa-tion, Age, Sexual orientation, Email address, School name, Phone number, Romanticor family issues (for example, crushes, breakups, family dramas), Financial informa-tion (for example, credit or debit card number), Account passwords, Phone passwords(to unlock the screen), Political, religious, and other personal point of views on var-ious matters, Photos of myself and family, Nude photos, Favorite TV shows, booksand snacks, Favorite places to hang out, Name and identity of friends or people Ispend time with, Activities I’ve participated inSection 4: Content Sharing The following questions help us understand yourthought processes when posting content on social networking sites.12. How important for you are the following factors for determining which social net-working site to use?Levels: © Not at all important, © Not very important, © Somewhat important, ©Very important, © Extremely important(a) My expectations of how my friends access and use various social networking sites(e.g., constantly using Snapchat, rarely using Facebook, don’t have Twitter).Appendix B: Confirmatory Study 75(b) Type of the content that I am planning to share with my friends (e.g., text,images, videos, articles, media news, opinions).(c) Sensitivity of the content that I am planning to share with my friends (e.g.,mundane content on Snapchat, lifestyle photos on Instagram, media news onFacebook, my romantic troubles via Facebook Messenger).(d) Whether I want my parents, relatives, and/or other adults to see what I postonline (e.g., whether my parents are my Facebook friends).(e) How close I am with the people that I intend to share my content with (e.g.,close friends on Snapchat, acquaintances on Facebook).(f) What the social networking site allows me to do online (e.g., group chatting,direct messaging, video calls).(g) The urgency of the interaction (e.g., calls for asking for a ride, Facebook Mes-senger for daily updates).13. Please rate how much each of the following statements describes your online be-haviours:Levels: © Never true for me, © Sometimes true for me, © Often true for me, ©Always true for me(a) When I share things online, I hide the meaning so that only my friends canunderstand what I am saying.(b) For some social networking sites, I have multiple accounts that I use for differentpurposes (e.g., 2 Instagram accounts).(c) I provide fake or inaccurate personal information online, such as a fake name,address, phone number, and email address.(d) I keep my social networking accounts (e.g., Facebook, Instagram) deactivatedwhen I am not using them, to stay private.(e) Please select ”sometimes true for me” option.(f) I delete some comments and messages that I receive online after I read them.(g) I delete some comments and other content that I leave on others’ pages sometime after I post them.(h) I delete content that I’ve posted online, if it doesn’t gather enough attention.(i) I don’t post much information online.(j) I don’t share too much about myself online.(k) I don’t share something online that I would regret later.(l) I use privacy settings to protect myself online.Appendix B: Confirmatory Study 76(m) I don’t accept requests from online strangers or respond to their messages.(n) To share personal matters with friends, I use private modes of communications(such as Facebook Messenger, texting, and calling), rather than sharing thecontent on my social networking sites.(o) My family and friends help with configuring my privacy settings on social net-working sites.(p) I carefully consider who I share my online content with and the consequences ofsharing it.14. When sharing content on social networking sites, please rate how much you careabout each of the following statements:Levels: © Don’t care at all, © Don’t care much, © Somewhat care about, © Careabout, © Extremely care about(a) The ability to choose what information I share online and what I keep to myself.(b) Once I posted my information online, the ability to control who can access thatinformation.(c) The ability to control what information about me is shared online by others.(d) The ability to browse web pages and say things online without other peoplebeing able to track me down or look at my activity history.(e) The ability to control how I am perceived online.15. Please rate how well each statement applies to you when sharing content on a socialnetworking site.Levels: © Never true for me, © Sometimes true for me, © Often true for me, ©Always true for me(a) I don’t believe I have anything to hide.(b) I believe I am unlikely to have my privacy invaded online.(c) If privacy invasions happen online, I am confident that I can deal with theconsequences and stay away from harm.(d) I feel like when I am interacting online, I take the necessary precautions to staysafe.(e) I do not worry about having my privacy being invaded online by authorities andstrangers that I don’t know.Appendix B: Confirmatory Study 77(f) I do not worry about having my privacy being invaded online by my parentsand other adults.(g) I do not worry about having my privacy being invaded online by my friends/-classmates/other peers.(h) If there is no direct influence on my personal life, I do not mind other peopleaccessing my online content.Section 5: About You The following questions help us learn about who you are.16. How old are you? [ ]17. What grade are you in? [ ]18. Which country do you live in? [ ]19. Which city do you live in? [ ]20. Think about your relationships with your peers and rate how much you agree ordisagree with each of the following statements:Levels: © Never true for me, © Sometimes true for me, © Often true for me, ©Always true for me(a) I go along with my friends just to keep them happy.(b) I think it’s more important to be myself than to fit in with the crowd.(c) For me, it’s pretty easy for my friends to get me to change my mind.(d) I would do something that I knew was wrong just to stay on my friends’ goodside.(e) I hide my true opinion from my friends if I think my friends will make fun ofme.(f) I wouldn’t break the law just because my friends say that they would.(g) I change the way I act when I’m with my friends.(h) I take more risks when I’m with my friends than I do when I’m alone.(i) I say things that I don’t really believe because I think it will make my friendslike me more.(j) I think it’s important to stand up for what I believe, even if people might getangry at me for going against the crowd.(k) I find it hard to make friends.(l) I have a lot of friends.(m) I am very hard to like.(n) I am popular with other kids my age.Appendix B: Confirmatory Study 78(o) I feel that I am socially accepted (liked by people).(p) I am able to make really close friends.(q) I have a close friend that I can share secrets with.(r) I wish I had a really close friend to share things with.(s) I find it hard to make friends that I can really trust.(t) I don’t have a friend that is close enough to share really personal thoughts with.21. What do you identify your gender as?© Male, © Female, © Transgender, © Other [ ], © Prefer not to answer22. What is the highest level of education that your parents/guardians have?© Less than high school, © High school, © Some college, © Undergraduate degreefrom a university, © Graduate (Master’s or PhD), © Not sure, © Prefer not toanswerBibliography[1] Annual Demographic Estimates: Canada, Provinces and Territories, 9 2018. URL Visited on 2019-01-27.[2] Valerie Steeves. Young Canadians in a Wired World, Phase III-Life Online. Me-diaSmarts, 2014. URL Visited on2019-04-06.[3] Monica Anderson and Jingjing Jiang. Teens, Social Media & Technology 2018. 2018.Washington, DC: Pew Internet & American Life Project. Retrieved on March 10,2019 from[4] Laurence D. Steinberg. Age of Opportunity: Lessons from the New Science of Ado-lescence. Houghton Mifflin Harcourt, 2014. ISBN 9780544279773.[5] Sonia Livingstone. Taking risky opportunities in youthful content creation: teenagers’use of social networking sites for intimacy, privacy and self-expression. New Media& Society, 10(3):393–411, 2008. ISSN 1461-4448. doi: 10.1177/1461444808089415.[6] Monica Anderson and Jingjing Jiang. Teens’ Social Media Habits and Ex-periences. Washington, DC: Pew Internet & American Life Project, 6, 2018.Washington, DC: Pew Internet & American Life Project. Retrieved on April 06,2019 from 80[7] Elisabeth Staksrud, Kjartan O´lafsson, and Sonia Livingstone. Does the use of socialnetworking sites increase children’s risk of harm? Computers in Human Behavior,29(1):40–50, 2013. ISSN 07475632.[8] Anirban Sengupta and Anoshua Chaudhuri. Are Social Networking Sites a Sourceof Online Harassment for Teens? Evidence from Survey Data. Children and YouthServices Review, 33(2):284–290, 2011. ISSN 01907409. doi: 10.1016/j.childyouth.2010.09.011.[9] Justin W Patchin and Sameer Hinduja. Cyberbullying prevention and response: Ex-pert perspectives. Routledge, 2012.[10] Trevor Hughes. Kik messaging app scrutinized in wake of Va. teen’s mur-der, 2016. URL Visitedon 2017-08-01.[11] Gustavo S Mesch. Parental mediation, online activities, and cyberbullying. CyberPsy-chology & Behavior, 12(4), 2009. ISSN 1557-8364. doi: 10.1089/cpb.2009.0068.[12] David Goldman. Campus uproar over Yik Yak app after sex harassment, mur-der - May. 7, 2015, 2015. URL Visited on 2017-08-01.[13] Lisa M Jones, Kimberly J Mitchell, and David Finkelhor. Trends in Youth InternetVictimization: Findings From Three Youth Internet Safety Surveys 2000 – 2010.Adolescent Health, 50(2):179–186, 2012. ISSN 1054-139X. doi: 10.1016/j.jadohealth.2011.09.015.[14] Kimberly J Mitchell, Lisa M Jones, David Finkelhor, and Janis Wolak. Trendsin Unwanted Online Experiences and Sexting. Crimes against Children ResearchCenter, 2014.[15] Janis Wolak, David Finkelhor, and Kimberly Mitchell. Internet-initiated sex crimesagainst minors: Implications for prevention based on findings from a national study.Journal of adolescent health, 35(5):424–e11, 2004.Bibliography 81[16] Li-Hui Chen, Susan P. Baker, Elisa R. Braver, and Guohua Li. Carrying Passengersas a Risk Factor for Crashes Fatal to 16- and 17-Year-Old Drivers. JAMA, 283(12):1578–1582, 2000.[17] Thomas D. Marcotte, Nicole M. Bekman, Rachel A. Meyer, and Sandra A. Brown.High-Risk Driving Behaviors among Adolescent Binge Drinkers. The American Jour-nal of Drug and Alcohol Abuse, 38(4):322–327, 2012.[18] Sangmi Chai, Sharmistha Bagchi-Sen, Claudia Morrell, H.R. Rao, and ShambhuJ. Upadhyaya. Internet and Online Information Privacy: An Exploratory Study ofPreteens and Early Teens. Professional Communication, IEEE Transactions on, 52:167 – 182, 07 2009. doi: 10.1109/TPC.2009.2017985.[19] Joyce M. Wolburg. The ”Risky Business” of Binge Drinking among College Students:Using Risk Models for PSAs and Anti-Drinking Campaigns. Journal of Advertising,30(4):23–39, 2001.[20] Jason Chein, Dustin Albert, Lia O’Brien, Kaitlyn Uckert, and Laurence Steinberg.Peers increase adolescent risk taking by enhancing activity in the brain’s rewardcircuitry. Developmental science, 14(2):F1, 2011.[21] Valerie Steeves. Chapter VI. “Pretty and Just a Little Bit Sexy, I Guess”: Publicity,Privacy, and the Pressure to Perform “Appropriate” Femininity on Social Media,chapter 5. Les Presses de l’Universite´ d’Ottawa — University of Ottawa Press, 2017.ISBN 9780776622576;0776622579;.[22] Alice E Marwick and Danah boyd. Networked privacy: How teenagers negotiate con-text in social media. New Media & Society, 16(7):1051–1067, 2014. ISSN 1461-4448.URL Visited on2019-04-06.[23] Adrienne Felt and David Evans. Privacy Protection for Social Networking Platforms.Workshop on Web 2.0 Security and Privacy, 2008. URL Visitedon 2019-04-06.Bibliography 82[24] Vadivel Sithira and Yok-Yen Nguwi. A Study on the Adolescent Online SecurityIssues. International Journal of Multidisciplinary and Current Research, 2:596–601,2014.[25] Alice E Marwick, Diego Murgia Diaz, and John Palfrey. Youth, Privacy and Repu-tation. Harvard Law School Public Law & Legal Theory Working Paper Series, 10-29(2010-5):1–82, 2010. URL Visited on 2019-04-06.[26] Florence Martin, Chuang Wang, Teresa Petty, Weichao Wang, and Patti Wilkins.Middle School Students Social Media Use. Journal of Educational Technology & So-ciety, 21(1):213–224, 2018. URL Visitedon 2019-04-06.[27] Emily Christofides, Amy Muise, and Serge Desmarais. Information disclosure andcontrol on Facebook: are they two sides of the same coin or two different processes?Cyberpsychology & behavior : the impact of the Internet, multimedia and virtualreality on behavior and society, 12(3):341–345, 2009. ISSN 1094-9313. doi: 10.1089/cpb.2008.0226.[28] Danah boyd and Alice Marwick. Social Privacy in Networked Publics: Teens’ At-titudes, Practices, and Strategies. A Decade in Internet Time: Symposium on theDynamics of the Internet and Society, September 2011, pages 1–29, 2011. ISSN0003-066X. URL Visited on 2019-04-06.[29] Katie Davis and Carrie James. Tweens’ conceptions of privacy online: implicationsfor educators. Learning, Media and Technology, 38(1):4–25, 2013. ISSN 1743-9884.doi: 10.1080/17439884.2012.658404.[30] Chris Hoofnagle, Jennifer King, Su Li, and Joseph Turow. How Different are YoungAdults from Older Adults When it Comes to Information Privacy Attitudes andPolicies? SSRN, 4(19):10, 2010. ISSN 15565068. doi: 10.2139/ssrn.1589864.[31] Mary Madden, Amanda Lenhart, Sandra Cortesi, Urs Gasser, Maeve Duggan, AaronSmith, and Meredith Beaton. Teens, Social Media, and Privacy. Pew ResearchCenter, 21:2–86, 2013.Bibliography 83[32] Danah boyd. Why Youth (Heart) Social Network Sites: The Role of NetworkedPublics in Teenage Social Life. MacArthur Foundation Series on Digital Learning– Youth, Identity, and Digital Media, 7641(41):1–26, 2007. ISSN 15292401. doi:10.1162/dmal.9780262524834.119.[33] Zeynep Tufekci. Grooming, Gossip, Facebook and Myspace. Information, Commu-nication & Society, 11(4):544–564, 2008. doi: 10.1080/13691180801999050.[34] Jo Bryce and James Fraser. The role of disclosure of personal information in theevaluation of risk and trust in young peoples’ online interactions. Computers inHuman Behavior, 30:299–306, 2014. ISSN 07475632. doi: 10.1016/j.chb.2013.09.012.[35] Chen Wei Chang and Jun Heo. Visiting theories that predict college students’ self-disclosure on Facebook. Computers in Human Behavior, 30:79–86, 2014. ISSN07475632. doi: 10.1016/j.chb.2013.07.059.[36] Amanda Lenhart, Mary Madden, Sandra Cortesi, Urs Gasser, and AaronSmith. Where Teens Seek Online Privacy Advice. 2013. Pew Re-search Center, Internet & Technology. Retreived on April 6, 2019 from[37] Denise E Agosto and June Abbas. “Don’t be dumb–that’s the rule I try to live by”: Acloser look at older teens’ online privacy and safety attitudes. New Media & Society,2015. ISSN 1461-4448. doi: 10.1177/1461444815606121.[38] Pamela Wisniewski, Arup Kumar Ghosh, Heng Xu, Mary Beth Rosson, and John MCarroll. Parental Control vs. Teen Self-Regulation: Is there a middle groundfor mobile online safety? Proceedings of the 2017 ACM Conference on Com-puter Supported Cooperative Work and Social Computing, pages 51–69, 2017. URL Visited on 2019-04-06.[39] Faye Mishna, Charlene Cook, Michael Saini, Meng-Jia Wu, and Robert MacFadden.Interventions to Prevent and Reduce Cyber Abuse of Youth: A Systematic Review.Research on Social Work Practice, 21(1):5–14, 2011. ISSN 1049-7315. doi: 10.1177/1049731509351988.Bibliography 84[40] Ross D Parke and Douglas B Sawin. Children’s Privacy in the Home: Developmental,Ecological, and Child-Rearing Determinants. Environment and Behavior, 11(1):87–104, 1979. ISSN 0013-9165. doi: 10.1177/0013916579111004.[41] Maxine Wolfe and Robert Laufer. The Concept of Privacy in Childhood and Ado-lescence. In D. H. Carson (Ed.), Man-Environment interactions: Evaluations andapplications. Part II. Stroudsberg, PA.:Dowden, Hutchinson & Ross, 1975.[42] Jeffrey J Arnett. Emerging adulthood: A theory of development from the late teensthrough the twenties. American Psychologist, 55(5):469–480, 2000. ISSN 0003-066X.doi: 10.1037//0003-066X.55.5.469.[43] John P. Caughlin and Tamara D. Golish. An analysis of the association between topicavoidance and dissatisfaction: Comparing perceptual and interpersonal explanations.Communication Monographs, 69(4):275–295, 2002. ISSN 0363-7751. doi: 10.1080/03637750216546.[44] Thomas J. Dishion and Robert J. McMahon. Parental Monitoring and the Preventionof Child and Adolescent Problem Behavior: A Conceptual and Empirical Formula-tion. Clinical child and family psychology review, 1(1):61–75, 1998. ISSN 1096-4037.doi: 10.1023/A:1021800432380.[45] P.C. Fletcher, F. Happe´, U. Frith, S.C. Baker, R.J. Dolan, R.S.J. Frackowiak, andC.D. Frith. Other minds in the brain: A functional imaging study of “theory ofmind” in story comprehension. Cognition, 57(2):109–128, 1995. ISSN 00100277. doi:10.1016/0010-0277(95)00692-R.[46] Monica Anderson. Parents, Teens and Digital Monitoring. Pew Internet and Amer-ican Life Project, pages 1–26, 2016.[47] Sonia Livingstone and Ellen J Helsper. Parental Mediation of Children’s Internet Use.Journal of Broadcasting & Electronic Media, 52(4):581–599, 2008. ISSN 08838151.doi: 10.1080/08838150802437396.Bibliography 85[48] Jennifer D. Shapka and Danielle M. Law. Does One Size Fit All? Ethnic Differences inParenting Behaviors and Motivations for Adolescent Engagement in Cyberbullying.Journal of Youth and Adolescence, 42(5):723–38, 05 2013.[49] Pamela Wisniewski, Haiyan Jia, Na Wang, Saijing Zheng, Heng Xu, Mary BethRosson, and John M Carroll. Resilience Mitigates the Negative Effects of Ado-lescent Internet Addiction and Online Risk Exposure. Proceedings of the ACMCHI’15 Conference on Human Factors in Computing Systems, 1:4029–4038, 2015.doi: 10.1145/2702123.2702240.[50] Andrea Duerager and Sonia Livingstone. How can parents support children’s internetsafety? EU Kids online. London, UK, 2012. URL Visited on 2019-04-06.[51] F Hunter. Adolescents’ perceptions of discussions with parents and friends. Develop-mental Psychology, 21(3):433–440, 1985. ISSN 0012-1649. doi: 10.1037/0012-1649.21.3.433.[52] Susan B. Silverberg and Laurence Steinberg. Adolescent autonomy, parent-adolescentconflict, and parental well-being. Journal of Youth and Adolescence, 16(3):293–312,1987. ISSN 00472891. doi: 10.1007/BF02139096.[53] R. A. Sanders. Adolescent Psychosocial, Social, and Cognitive Development. Pedi-atrics in Review, 34(8):354–359, 2013. ISSN 0191-9601. doi: 10.1542/pir.34-8-354.[54] Amanda Lenhart. Teens, Social Media & Technology Overview 2015: Smartphonesfacilitate shifts in communication landscape for teens. Pew Research Center, pages1–47, 2015. ISSN 07475632. doi: 10.1016/j.chb.2015.08.026.[55] Global Snapchat teen usage penetration 2016 — Statistic, 2016. URL Vis-ited on 2019-04-06.[56] Anabel Quan-Haase and Alyson L. Young. Uses and Gratifications of Social Media:A Comparison of Facebook and Instant Messaging. Bulletin of Science, Technology& Society, 30:350–361, 2010. ISSN 0270-4676. doi: 10.1177/0270467610380009.Bibliography 86[57] Shelley Goldman, Angela Booker, and Meghan McDermott. Mixing the Digital,Social, and Cultural: Learning, Identity, and Agency in Youth Participation. Youth,identity, and digital media 216, 216, 2008.[58] Facebook Terms of Service, 2015. URL Visitedon 2017-03-07.[59] Amanda Lenhart. Cyberbullying. Pew Internet & American Life Project, 2007. URL Visited on 2019-04-06.[60] Deborah M Moscardelli. Teens Surfing The Net: How Do They Learn To ProtectTheir Privacy? Journal of Business & Economics Research (JBER), 2(9):43–56,2004. ISSN 2157-8893. URL Visitedon 2019-04-06.[61] Lilian Edwards and Ian Brown. Data Control and Social Networking: IrreconcilableIdeas? Stanford University Press, 2009.[62] John Palfrey and Urs Gasser. Born Digital: Understanding the First Generationof Digital Natives. Hedgehog Review, 198(2008):288, 2008. ISSN 15279677. doi:10.1097/NMD.0b013e3181cc549e.[63] Kim B. Sheehan and Mariea G. Hoy. Dimensions of privacy concern among onlineconsumers. Journal of Public Policy & Marketing, 19(1):62–73, 2000.[64] Steve Jones, Camille Johnson-Yale, Sarah Millermaier, and Francisco Seoane Perez.Everyday life, online: US college students use of the Internet. First Monday, 14(10),2009.[65] Seounmi Youn. Teenagers’ Perceptions of Online Privacy and Coping Behaviors: ARisk-Benefit Appraisal Approach. Journal of Broadcasting & Electronic Media., 49(1):3–22, 2005. ISSN 0883-8151. doi: 10.1207/s15506878jobem4901.[66] Seounmi Youn. Determinants of Online Privacy Concern and Its Influence on PrivacyProtection Behaviors among Young Adolescents. Journal of Consumer Affairs, 43(3):389–418, 2009. ISSN 00220078. doi: 10.1111/j.1745-6606.2009.01146.x.Bibliography 87[67] Alan F. Westin. Privacy and Freedom. American Sociological Review, 33(1):173,1968. ISSN 00031224. doi: 10.2307/2092293.[68] Irwin Altman. The Environment and Social Behavior: Privacy, Personal Space,Territory, Crowding, volume 1. Brooks/Cole Pub. Co, Monterey, Calif, 1975. ISBN0818501685. doi: 10.2307/2065073.[69] Tony Doyle. Helen Nissenbaum, Privacy in Context: Technology, Policy, and theIntegrity of Social Life. The Journal of Value Inquiry, 45(1):97–102, 2011. ISSN0022-5363. doi: 10.1007/s10790-010-9251-z.[70] Danah boyd. Facebook privacy settings: Who cares? First Monday, 15(8):13–20,2008. ISSN 13960466. doi: 10.1177/1354856507084416.[71] Catherine Dwyer. Digital Relationships in the “Myspace” Generation: Results froma Qualitative Study. 2007 40th Annual Hawaii International Conference on SystemSciences (HICSS’07), pages 19–19, 2007.[72] Michelle Kwasny, Kelly Caine, Wendy Rogers, and Arthur Fisk. Privacy and Tech-nology: Folk Definitions and Perspectives. CHI’08 Extended Abstracts on HumanFactors in Computing Systems, pages 3291–3296, 2008.[73] Susannah Fox. Trust and Privacy Online: Why Americans Want to Rewritethe Rules. Pew Internet American Life Project, 20:1–29, 2000. URL Visited on2019-04-06.[74] Hanna Krasnova and Paula Kift. Online privacy concerns and legal assurance: Auser perspective. Workshop on Information Security and Privacy, 2(1):1–23, 2012.doi: 10.14763/2013.1.107.[75] Deirdre O’Brien and Ann M. Torres. Social networking and online privacy: Facebookusers’ perceptions. Irish Journal of Management, 2012.[76] Yang Feng and Wenjing Xie. Teens’ concern for privacy when using social networkingsites: An analysis of socialization agents and relationships with privacy-protectingBibliography 88behaviors. Computers in Human Behavior, 33:153–162, 2014. ISSN 07475632. doi:10.1016/j.chb.2014.01.009.[77] Carly Lejnieks. A Generation Unplugged, 2008. URL Visited on 2019-04-06.[78] Joy Peluchette and Katherine Karl. Social networking profiles: An examination ofstudent attitudes regarding use and appropriateness of content. Cyberpsychology &Behavior, 11(1):95–97, 2008. ISSN 1094-9313. doi: 10.1089/cpb.2007.9927.[79] Claire Balleys and Sami Coll. Being publicly intimate: teenagers managing onlineprivacy. Sage Publication, 2016. doi: 10.1177/0163443716679033.[80] Barney Glaser and Anselm Strauss. The discovery of grounded theory. Weidenfield& Nicolson, London, 1967.[81] Juliet Corbin and Anselm Strauss. Basics of Qualitative Research (3rd ed.): Tech-niques and Procedures for Developing Grounded Theory. Sage Publications, Inc.,2008. doi: 10.4135/9781452230153.[82] Catherin Marshall and Gretchen B. Rossman. Designing Qualitative Research.SAGE, sixth edition, 2016. ISBN 9781452271002.[83] Seounmi Youn and Kimberly Hall. Gender and online privacy among teens: riskperception, privacy concerns, and protection behaviors. Cyberpsychology & Behavior,11(6):763–765, 2008. ISSN 1094-9313. doi: 10.1089/cpb.2007.0240.[84] Wonsun Shin and Hyunjin Kang. Adolescents’ privacy concerns and information dis-closure online: The role of parents and the Internet. Computers in Human Behavior,54:114–123, 2015. ISSN 07475632. doi: 10.1016/j.chb.2015.07.062.[85] Yang Wang, Pedro Giovanni Leon, Gregory Norcie, Alessandro Acquisti, and Lor-rie Faith Cranor. “I regretted the minute I pressed share”: A Qualitative Study ofRegrets on Facebook. Symposioum on Usable Privacy and Security 2011, Pittsburgh(USA), pages 1–16, 2011. doi: 10.1145/2078827.2078841.Bibliography 89[86] Kim Thomas. Teen Online Safety & Digital Reputation Survey. Cox Com-munications in partnership with the National Center for Missing & Ex-ploited Children, 2010. URL Visited on 2019-04-06.[87] Milton Friedman. The Use of Ranks to Avoid the Assumption of Normality Implicitin the Analysis of Variance. Journal of the American Statistical Association, 32(200):675–701, 1937.[88] Ricky Jay Weible. Privacy and Data: An Empirical Study of the Influence of Typesof Data and Situational Context Upon Privacy Perceptions. PhD thesis, MississippiState, MS, USA, 1993. UMI Order No. GAX93-31661.[89] Lorrie F. Cranor, Joseph Reagle, and Mark S. Ackerman. Beyond Concern: Under-standing Net Users’ Attitudes About Online Privacy. The Internet upheaval: raisingquestions, seeking answers in communications policy, pages 47–70, 2000. Cambridge,MA: MIT Press.[90] Christine A Varney. Consumer Privacy in the Information Age: A View from theUnited States, 1996. URL Visited on 2019-03-10.[91] Catherine Dwyer, Starr Hiltz, and Katia Passerini. Trust and Privacy ConcernWithin Social Networking Sites: A Comparison of Facebook and MySpace. AM-CIS 2007 proceedings, page 339, 2007.[92] Petter Bae Brandtzæg. Social Networking Sites: Their Users and Social Implications— A Longitudinal Study. Journal of Computer-Mediated Communication, 17(4):467–488, 07 2012. ISSN 1083-6101.[93] Joe Phua, Seunga V. Jin, and Jihoon Kim. Uses and gratifications of social network-ing sites for bridging and bonding social capital: A comparison of Facebook, Twitter,Instagram, and Snapchat. Computers in Human Behavior, 72:115–122, 2017.Bibliography 90[94] Nicole B. Ellison, Jessica Vitak, Rebecca Gray, and Cliff Lampe. Cultivating SocialResources on Social Network Sites: Facebook Relationship Maintenance Behaviorsand Their Role in Social Capital Processes. Journal of Computer-Mediated Commu-nication, 19(4):855–870, 2014.[95] Williams Dmitri. On and off the ’Net: Scales for social capital in an online era.Journal of computer-mediated communication, 11(2), 2006.[96] Danah boyd. It’s Complicated: The Social Lives of Networked Teens. Yale UniversityPress, 2014.[97] Alyson Young and Anabel Quan-Haase. Information revelation and internet privacyconcerns on social network sites: a case study of Facebook. In Proceedings of thefourth international conference on Communities and technologies, pages 265–274,2009.[98] Zeynep Tufekci. Can You See Me Now? Audience and Disclosure Regulation inOnline Social Network Sites. Bulletin of Science, Technology & Society, 28(1):20–36,2007. ISSN 0270-4676. doi: 10.1177/0270467607311484.[99] Alexander W. Astin. What Matters in College?: Four Critical Years Revisited.Jossey-Bass, San Francisco, 1st edition, 1993. ISBN 1555424929.[100] Nicole L. Muscanell and Rosanna E. Guadagno. Make new friends or keep the old:Gender and personality differences in social networking use. Computers in HumanBehavior, 28(1):107–112, 2012. doi:[101] S Guzin Mazman and Yasemin Koc¸ak Usluel. Gender differences in using socialnetworks. Turkish Online Journal of Educational Technology-TOJET, 10(2):133–139, 2011.[102] Alireza Moghaddam. Coding Issues in Grounded Theory. Issues in educationalresearch, 16(1):52–66, 2006.[103] Norman K. Denzin, editor. Sociological Methods: A Sourcebook (Methodological Per-spectives). Transaction Publishers, 2006. ISBN 9780202308401.


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items