Privacy in Social Network Sites (SNS)


COMPUTER PRIVACY IN

SOCIAL NETWORK WEBSITES

(Author note: This Thesis is published here anonymously)

  1. Abstract

This paper studies online information disclosure and privacy implication among teenagers on social network websites such as FaceBook and MySpace. Within Anonymous-Country context, this study analysed survey results from 97 secondary school going teenage students (age betweens 13 to 19). The study found significant gender variations in regards to several types of information they disclose online. In comparison with similar previous study on adult’s behaviour towards online information privacy, the statistics results show significant pattern in terms of the intentions of what information they publish online, including their preferences on privacy features they use during online social interaction as well as personal profiles information. Some recommendations, online safety tips and solutions can be found before the conclusion and study limitation which concludes this report.

KEYWORDS

Privacy, social network, Facebook, MySpace, teenagers, teenagers, danger, photos, videos, Anonymous-Country

Introduction

As Tim Berners-Lee had pointed out, “The web is more a social creation than a technical one“. He emphasized the social effect the Web brought into human daily life – “to help people work together – and not as a technical toy“. With the invention of social network applications, we can enjoy the interactivity by taking advantage of World Wide Web. However, even before Web was invented, Howard Rheingold (1985) already foresaw the massive challenge due to the application of social software and its potential for change: “Nobody knows whether this will turn out to be the best or the worst thing the human race has done for itself, because the outcome of this empowerment will depend in large part on how we react to it and what we choose to do with it.

Social Network Websites (or SNSs) as a new form of social media are the great invention to form interactive online communities. Due to their worldwide availability, they have gained extensive popularity among Internet users from all over the world regardless of race, religion, age, and geographic location. The success of social network websites is largely attributed to the need for people to connect with others. For many people, they provide a way for them to find other people who have similar interests, meet with romantic partners, or simply expand their existing friendship with friends in cyberspace through existing friends and acquaintances.

Despite social network websites are able to mimic how people tend to make friends in real world, social network websites draw increasing attention from public because “with the radical shift in the number of users worldwide onto online social networks, there are new and significantly higher privacy leakage concerns as compared to traditional Web websites” (Krishnamurthy, 2008).

A significant privacy threat is raised by an increasing amount of media content posted by users’ in their profile” (Squicciarini, 2009). Therefore, to what extent the information should be revealed to others becomes the increasing concern amongst the social network users. The unrestricted view to others’ profile and information an unwanted audience doesn’t own may compromise the privacy of others. And even worse, the social network websites operators who are the only ones having the full view to users’ data (Bhagat, 2010) breach their own terms and conditions related to the user’s privacy, where they promise not sharing user data with third parties (Jacobsson, 2010).

Privacy issue starts from the beginning when a user sign up for a social network account. Once a profile is ready, the member can invite the real-world friend to join MySpace as well. Once they sign up as members, they become the friends in MySpace. In turn, these MySpace friends can do the same to their friends. The network of friendship expands gradually as it can be endorsed by common friends. In this sense, it is not difficult to declare someone a friend (through a friend’s friends), but would be difficult to know each other well enough as a real endorsement through the number of friends would not be feasible.

Despite the arising problem with respect to the privacy concerns, people still rely on social network websites for social interactions. This is simply because humans are innately social, that is humans interact with each other for almost every need – foods, water, shelter, technology, friendship, learning, fun, sport, etc. A simple example would be, when we are in doubt, we turn to our social connections for help. Social network websites change the way people think, learn and communicate. If it is inevitable that people participates online social interaction, preparing them with privacy protection tactics become crucial during online interaction, especially for the youth.

The popularity of social network media has attracted many scholars to conduct extensive research works pertaining to social network privacy. In the later part of this paper, we will be reviewing some of these works in relation to the youth. In the mean time, we will be focusing on two most popular SNS i.e. Facebook and MySpace to understand teenagers’ behavior related to their privacy and tactics adopted to protect themselves. We will discuss about the procedure and methodology we adopt to conduct the research. Afterwards, we will analyze our research result and also compare with previous study that was done for the similar topic. Finally, we will be trying to come up with the solution for address the dilemma.

Literature review

As technologies evolve, Web becomes powerful enough to mirror the society we live in and provide the possibility of having the second life in the virtual world. As Berners (1999) described, in the virtual online communities, users develop the trustworthy relationships among family members, friends, and colleagues beyond the geographic location. He commented: “What we believe, endorse, agree with, and depend on is representable and, increasingly, represented on the Web.

Trust and Privacy

Trust is defined as “the willingness of a party to be vulnerable to the actions of another party based on the expectation that the other will perform a particular action important to the trustor, irrespective of the ability to monitor or control that other party” (Mayer, Davis, and Schoorman, 1995).

The real issue here is not only applicable to teenagers but also to general population as a whole, it is about trust. How can you trust someone online which you could not determine the real identity of the person, which you never met before? He or she could be using the identity or account the person that you trust in real world. There are Obama, Michael Jackson or Dalai Lama online in Facebook or MySpace. Are they the real President Obama from the US, or the late Michael Jackson (he is already dead, how can he be alive online) or the famous Dalai Lama from Tibet? In online world, anyone can be or claim whoever they want to be.

The other issue is privacy, privacy within social networking sites is often not expected or is undefined (Dwyer, 2007). One of the good thing (for some) about social networking websites is “mining” information (data). All the information entered, be it personal data, photos, videos or whatsoever, no one can guarantee the existence, can it be erased or removed permanently. Former President Bush mentioned that any online information that we put on Facebook or Myspace could one day resurfaced and hunt us down or be used by prospective employer to filter out the good from the bad candidates (the wolves among the herd of sheeps). Scary though, considering most of the social networking websites leave no trace behind. This brings up the notion about proper policies and data protection mechanism in social networking websites.

In this study, we find that teenagers trust their online friends and the media (eg. Facebook or MySpace) easily by giving out privacy data and lack of consideration as compare to adults.

Teenagers’ Behaviors in Using Social Network Websites

According to Boyd, the age limit on the social network websites was 18+ at the beginning. Over time, it dropped to 14 (Boyd, 2006). This suggests that the rising desire of youth pushes them to create their own virtual space. According to the report from Pew Internet & American Life Project (Albrechtslund, 2008), the majority of teenagers keep in touch with others they rarely see in their daily life through the online social network websites. While they are at the age of curious about the world they are living in. They want to know and want to experience as much as they could through the interactions with others. Teenagers learn to use social network features by interacting with friends rather than learning from their parents or teachers (Susan, 2006). Social network websites become indispensable for them.

The teenagers would access websites such as MySpace at least once a day for the average duration of one hour 22 minutes when the access is possible. They access to MySpace or Facebook more often than Yahoo and MSN. For most of them, social network websites play a key role in their culture. The things that keep attracting them sticking to social network websites is the set of features offered by the websites that allow them to converse with friends, share digital cultural ideas, and connect to vast networks of people (Boyd, 2007). MySpace provides the space for young people to blog, flirt, share pictures, videos, or creative artwork, and meet new people. Similarly, Facebook connects students together from same schools. Youth put their minds into cyberspace voluntarily. They do it for stowing their thoughts and gathering feedback from peers. Therefore, having access to social network websites become part of their lives. The dynamics of social network websites is another factor that makes them popular among the teenagers. For example, MySpace’s success doesn’t solely depend on its flexibility. Tapping into music industry, and making more than half a million bands sign up on MySpace, such as Back Street Boys, Fall out Boy, and West Grand, etc, is the reason behind attracting teenagers.

Teenagers not only reveal their thoughts and demonstrate their behaviors online. About 82 percent of teenagers provide their real first name in their profiles, according to the report from Pew Internet & American Life Project (Albrechtslund, 2008). While adults are worried about their privacy being invaded, the teenagers freely give up their personal information. Susan noted that this was due to teenagers in America were not aware of the public nature of the Internet (Susan, 2006) Lenhart (2005) reported that 79 percent of teenagers did not take the consequences into the consideration prior to unveiling their personal information online. The situation seems improved few years later. Alyson (2009) suggested that students know part of their profile should be falsified to protect themselves. Meanwhile, they also employ other protective measures to avoid privacy threats.

Common Properties and Problems with Social Network Websites

The privacy threat starts from the bit of information available online. The different type of privacy information hosted in the majority of social network websites can be categorized as followings (Krishnamurthy, 2008):

  • Thumbnail: the least profile about one’s privacy such as name, photo, email, etc. Please be noted that in Facebook, a user’s thumbnail as well as the list of a user’s friends are publicly accessible by anyone.
  • Greater profile: including interests, gender, age, relationships
  • Friend list: containing the list of friends that are deemed befriended by a user
  • Content generated by a user: such as photos, links, comments, blogs, videos, status update, tags (associated with photos or with some other forms of contents) etc.

There are different levels of access to these information can be set by any individual user, including one’s friends, the friends of one’s friends (or one’s network – a feature used in Facebook to control the access to various settings related to privacy) or the mass – all of users. However, this is still not good enough to protect one’s privacy.

Balachander (2008) realized a common problem with all social network websites, where the manner in which information being gathered by various parties is often invisible to users. This makes users very difficult to be aware who can actually gain access to their information therefore have little idea what it would take to control their access while they still retain their ability to take full advantage of the various features through online social interaction. For example, in Facebook, if a user wants to add an external application to his profile, he has to grant full access to that application in order to use it. Though the application may not use complete information about the user as some of information is completely irrelevant, there is no minimum-usage principle to be possibly applied to this situation. Even SNS provide grouping feature, however, it is fixed and fully controlled by social network websites. There is no choice to customize one’s profile by categorizing his/her friends into different groups with respective access rights.

Boyd (2007) argued that there are common properties in Social Network Websites altering social dynamics fundamentally. The social network websites have been complicating the ways in which people interact (Boyd, 2007). These properties include:

  • Persistence: What have been posted online will stick around and always tie to an originator. The teenagers might feel encouraged to post online to prove they live ‘there’.
  • Replicability: it would be difficult to differentiate whether the information available on social network websites is original or has been altered. The authenticity that represents the origins and legitimacy is very hard to obtain due to this property.
  • Scalability: through acceptance (sometimes careless) of another’s request, friends list and one’s network on social network websites grow exponentially in a short time. However, what spreads may not be ideal. For example, identity related personal information is supposed to be spread amongst those on the friends list only.
  • Searchability: the powerful search engine enables one to find anything they want so long as the engine has access to that bit of information. The searchability increases the possibility of information scalability and replicability as it provides a new mean to tap the information.
  • Invisible audiences: who we are actually interacting with becomes hard to judge as in most of cases users of social net work websites always have assumption in mind that the people having access to the information one published must be on the friends list. Without other way for identification, the impersonation can easily invalidate the assumption.
  • Blurring of the boundary of public information and private information: some argued that privacy becomes dead. “In networked publics, attention becomes a commodity.” In order to draw attention for the sake of personal gain or embarrassing others purposely, one can deliberately disclose some information which deemed very private.

Dangers to the Teenagers

In the analysis on the participation in social network websites among American teenagers, Boyd highlighted that those teenagers were able to and do develop their own strategies in managing the social complexities; and realized that “In some ways, teenagers are more prepared to embrace networked publics because many are coming of age in a time when networked affordances are a given.Lenhart and Madden (2005) reported that 57 percent of online teenagers create Internet content.

No matter how well the teenagers are prepared, in the complete picture of any social network websites, there is always a sleazy side – online perpetrators or practical jokers use social network websites to deceive others for either their own personal gain. This happens partly because of the anonymity on the Internet – one can never be absolutely sure about others met on Facebook, or MySpace or any other social network websites. Though the members of social network websites are able to decide what to reveal to others, it is not uncommon that some of the profiles don’t reflect its owner genuinely. For example, a married man pretended to be single; or men pretended they are pretty women by featuring beautiful photos. On the other hand, MySpace profiles are also being used for selling questionable products or for marketing scams.

Deviants are not so obvious, and sometimes much more clever to make them up on purpose. There nearly isn’t such a perfect enough system that it cannot be broken into by hackers though some of them are ethical and inoffensive whereas others choose to go too far. One of great examples is the famous hack in MySpace created by its member Samy, called “popular guy” hack also known as Samy Worm. Whenever people viewed Samy’s profile in MySpace, they made Samy a friend of theirs automatically, i.e. without their permissions. Furthermore, when anyone viewed a member profile where Samy was a friend also made Samy a friend automatically. Obviously, it is easy for Samy to become the most popular man through this way. Eventually, the exponential increase in the number of Samy’s friends overloaded MySpace servers which had to be shut down in order to disable the worm and restore the services. A lesson learnt from this incident unveiled the problem with MySpace profile design which provides the flexibility to allow its users to “hack” into the profile freely with programming codes such as HTML and Cascading Style Sheets.

The study done by Andrew et al. at The Berkman Center for Internet & Society at Harvard University shows that, online harassment appears highest among those who are around 13-14 years old compared to older youth. Apart from age differences, girls tend to be at more risk of becoming victims in the crimes such as online solicitation and harassment. Also online youth victims are also found to have other problems, such as depression and offline victimization. The offenders committing these crimes are mostly “anonymous, globally distributed“.

A lot of attention needs to be paid especially by the youth when their network expands dramatically and list of friends becomes longer and longer. No doubt that the youth is the most vulnerable group among all of users because they have less experience, as compared to the users in other age groups, with decision on if they should get in touch with strangers, or if they should post personal information online, or if they can share password with others even friends, etc. These are just few examples of risks that could make teenagers become victims very easily.

Given the situation that Ai Ho (2009) highlighted in the study with regard to three major problems with existing websites:

  • No service providers inform users of the dangers of divulging their personal information.
  • The existing tools in social network websites are not flexible enough to protect user data.
  • While users of SNS can control who can have access to their own profile, they cannot control what others especially those on the friends list or the powerful search engine reveal about them.

Even if there are warnings or guidance to inform teenagers of the dangers of divulging their personal information, or perhaps even in systematic way to enforce that, the needs that drive the social interaction among teenagers or between teenagers and others may yet be able to control their behaviors. Many forms of online crime such as child pornography, violent content, sexual solicitation and cyberbullying will find the ways out to reach to teenagers.

Inescapable Identity – A Forced Move

Herbert Simon (1971) offered his insightful view to the inevitable outcome of this information onslaught. He noted that in an information-rich world, the wealth of information means a scarcity of whatever it is that information consumes (Simon, 1971).

Indeed, we are living in the Information era. There is more information than we know what to do with, much more than we could ever reach, digest, and even more than we could imagine. It is the first time, human have to deal with a situation for not drowning in the information pool created by ourselves. The amount of information about products or services or whatever else looks infinite, simply no one has enough time to consider all of options thoroughly. This symptom is called information overload. To fight this overload by information, we’re turning more and more to trusted sources, whether they are in our own household or in other social circles. Instead of trying to sort, filter, and weed through endless sources of information, we’re focusing our attention on those we already trust, or those we have reason to believe might be trusted. This forces us to seek only authentic interactions among all available sources since we don’t have much time and we must make a choice.

To make the interaction and conversation authentic, all users need to identify themselves before using the services. There is vast variety of personal identity-related information such as physical and cultural attributes to be disclosed not only to the service providers but to the friends, even friends of friends. Social network websites users have to be certain that his/her profile is recognized as uniquely as possible by the real friends by providing some valid information. That is how one can gain the recognition and reputation. Consequently it is impossible to achieve this by retaining anonymity.

By definition, privacy means the quality of being secluded from the presence or view of others. For the sake of complete privacy, one could choose not to be seen or hidden from the view of others. However, this would make his/her online ‘presence’ unidentifiable. The essential element of social interactions simply is eroded. Therefore, no social activities would be able to be carried out. On the other hand, prior to being identifiable, one has to be sure what is truly trustworthy, that is not only who are involved in the interactive socialization activities, but also the service of social networking as well as its provider.

So, here is the baseline: the minimum amount of information that adequately makes one’s presence identifiable online has better be determined. As socialization goes along, the amount of information shared with people involved in the interaction varies greatly. Constantly monitoring who is across the boundary and judging what to share out will also help one to achieve the satisfied level of privacy.

Existing Features That Could Help Protecting Privacy on Social Network Websites

Boyd recognized that youth value the socialization and see the benefit outweighing the harm. Social network websites also acknowledged that the concern with privacy for not only teenage users. Hence they enable certain features to ease such fears.

  • Users can choose to stop using the services and have their account revoked from the service providers, at any point of time. However, there is confusion with the terminology associated with account closure. For example, in MySpace, a user can choose to click on the ‘Cancel account’ button under ‘Account Settings’. There is no instruction found from Help center if this will result in the same effect as you choose to ‘Delete Profile’.
  • With respect to the visibility to one’s profile, there are options available for users to decide who can see their personal profile, from ‘friends only’, or ‘only friends and anyone over 18’, to ‘everyone’. MySpace indeed allows its users to control if they want their profiles publicly accessible or not. When publicly accessible, it means anyone either with or without MySpace account could see the person’s profile. If one chooses to keep it private, then only the people on his/her articulated list of Friends can access to it.
  • Application setting can help users to block all or selected applications from accessing their profile information.

The feature that has made Facebook different from other social network websites is the way in which the privacy is being managed (Boyd, 2010). At the beginning, Facebook did not allow users to make any of their content accessible by everyone. By restricting the access to only those who have valid harvard.edu email addresses, no one else was unable to join the network. During the expansion from Harvard university to other colleges, even high schools and corporate users in USA, the concept of network that Facebook created and required the members to be part of their respective networks (i.e. regional network, corporate network, high school network, etc.) doesn’t make any sense. Therefore, Facebook minimized the importance of the “network” concept by stopping asking users to join any network.

Noteworthy Points

The range of privacy settings seems to be very permissive. Neither Facebook nor MySpace seems to restrict users to unveil the information about individual subscriber to the public by default. Perhaps, they would have expected that the subscriber him/herself should take precaution before signing up. That is to say that the users of social network websites wouldn’t put them into the position of embarrassing themselves and causing being harmed. Though, research suggested that users are not naive completely in their disclosure practices. However, they may not have clear idea who has access to their information while willingly putting them online. From users’ point of view, they believe in the accessibility to the information they publicize online within social network websites are restricted to their friends as well as the owner of respective websites. In reality, every bit of information can be disclosed to and be accessed by the third parties, such as advertisers, data aggregators, members who are the friends of users, as well as external applications, if the privacy settings for user profiles are controlled properly.

There are default settings in privacy setup of a user’s profile which suit users’ desire of protecting the information that they don’t want to be accessible to others. ‘by default’ does matter in the context of privacy since most of users overlook these default settings and never change or are not even aware them.

For example, there was a case in 2006 when Facebook introduced ‘News Feed’ feature to its users. With this feature, a user will be notified about what their friends were doing on Facebook through a stream of data. The interactivity from a user within Facebook was broadcast to all who are his/her friends list. This triggered the outrage that led to more than 700,000 people forming a group called ‘Students Against Facebook News Feed’ to pronounce their dissent with this new feature. In the end, the protest forced Facebook to provide the privacy setting to control the information distribution. In 2009, Facebook users were prompted with two options to reconsider their privacy settings for the content they publish to Facebook before logging into the site, one is ‘old setting’, the other is ‘Everyone’ or ‘Friends of Friends’ or ‘Friends’, which is the default option. For those who clicked through without carefully evaluating what to be shared with ‘Everyone’ or ‘Friends of Friends’, the information that used to be tightly controlled and restricted for access are now available to Everyone’s finger tip because they chose default settings. There was no clear sign if Facebook did not make ‘Old setting’ as default option deliberately. However, the inconsideration indeed resulted in certain implications to some of users. In the middle of 2010, Facebook released a new privacy setting. Compared to the previous setting, the new setting allows users to tighten the control on each piece of information put up in Facebook for sharing.

Figure 1 (in Appendix A) depicts the information available by default in all Facebook users’ profiles as of April, 2010. The amount of information in default profile settings has increased since 2005. In addition, the trend observed from the image suggests that the number of people have visibility to those information by default are no longer as restricted as they used to be. We are not sure about what result in to this trend. Perhaps, it is due to the origin motivation of social interactivity, which is in effect to encourage users to exchange or unveil information without the restriction so that they can focus on the socialization, friendship making rather than trying to isolate from the others. However, for those, who care about their privacy or are in doubt of how safe their private information is, if they wish to tighten access to information gathered in social network websites, they may opt the more restrictive setting to certain part of their profile.

RESEARCH METHODOLOGY

First we need to highlight the important of choosing the correct survey subject. There are many security related report related to financial such as banking and shopping, which mainly aim at working adult but not many on teenagers because they are not heavy users of this activity. But as far as the security is concerned, personal data such as real name, date of birth, age, gender, etc are common properties in any online activities regardless of ages whether banking or social networking.

Social network websites users are mostly teenagers between the age of 13 and 19. They frequently use popular social network websites such as Facebook, MySpace, Friendster and a couple of less known websites. The trend has continued for a couple of years, since social network websites emerged since 3 years ago. Nonetheless there are some social network websites that are for more matured audience age above 40s. Although these websites are less popular, it shows the trend that social networking encompasses many age groups.

There are many procedures and analytical techniques and methodologies being used to collect information and data, the most common one is survey, which can be in the form of online survey which is made available on the internet and the second method of survey is through hard copy distribution. Both methods have their own set of challenges.

The following summarized the pros and cons of both methods through our own experiences.

Online Survey Method

Initially when we embarked on this project, we decided to use online survey. The first challenge we faced was monetary, the popular online survey such as SurveyMonkey provides free online survey for a limited number of questions. Otherwise, payment is required. We sourced around and looked for free online survey then we came upon KwikSurvey.

Creating the survey for online platform has its own set of experience. One of the advantages of online survey as compared to paper-based survey is the computation which can be controlled, such that certain question can be made compulsory or the question has certain conditions to be made. We even thought that the result can be computed easily and convert to nicely format diagram and best of all is the need to distribute the hard copy to survey participants is minimized to zero. All these and more make online survey very promising, so we did online survey.

We sent a few requests to friends and relatives to request for their children’s participation in our online survey. We also sent similar request to school principal to ask for permission to survey the students on campus. We had waited for up to three weeks, and eventually the school principal gave final disapproval (indirectly). We tried another method and asked the children’s parents directly, the parents gave a go ahead but the children did not respond to the survey. More than one month passed, we got zero response in our online survey.

Distribution of Survey

The greatest disadvantage of online survey is lack of intention from the participant or no respond. Why would we participate in the online survey if it doesn’t give me any advantage, this is the normal perception. Thus I realized why some professional survey give freebies in exchange for survey input. And we come to realize too that survey is a tough job, now I would appreciate those people who stand in public places begging people to taking survey.

After our failed mission on online survey, we switched tactic. We used the online survey format and print the hardcopies (many of them) and get ready to go school and distribute it. But we realized that it exceeded 10 pages for one survey (consist of 26 questions). With 10 pages, not many people would want to help us with our survey. We re-wrote the whole survey and shrank it to 1 page from 10pages. Initially 1 question consist of 2 or 3 sentences, we reduced and rephrased it to 1 sentence, many using short form. After much hard work, we managed to get 97 surveys collected though we could not get the online survey result. Please refer to Appendix C for survey questionnaires.

How do we convert those hard copies to softcopy so can be processed or computed is another challenge. We scanned the hard copy and convert them to soft copy. The other method was manual conversion to Microsoft Excel format (csv format) and following the convention for analysis.

SURVEY FINDINGS

This study attempts to answers the following research questions:

  1. What are the behaviors of teenagers in social network websites?

  2. How do teenagers use privacy features in social network websites?

We conducted a survey with 97 school-going teenagers in Anonymous-Country, age from 13 to 19 years old (see Figure 3 in Appendix A). About 86% of respondents are between the age of 15 and 18. Perhaps these ages are more mature compared to 13-14 years old who have little idea what to respond appropriately.

We also divide the respondents into two groups based on their genders, i.e. boys and girls (see Figure 4 in Appendix A), to see if gender plays vital roles in the study. In general, girls seem more approachable than boys since they have much higher response rate compared to boys.

The response to the number of users on different social network websites shows that Facebook and even Friendster are more popular than MySpace in Anonymous-Country (see Figure 5 in Appendix A). Overall, there are about 93.8% of subjects having Facebook accounts. This may be due to the fact that Facebook has much more unique visitors over the past one year compared to MySpace according to Compete.com (see Figure 2 in Appendix A).

Among the respondents, most of teenagers (about 67% in total) have been using social network websites between 1 and 3 years, regardless of gender (see Figure 6 in Appendix A). Another one quarter of teenagers has even longer access to social network websites for more than 3 years. Considering the majority of the respondents are between the age 15 and 18 years old, this would mean that most of them have been using the social network websites since the beginning of their journey in their secondary schools. On the other hand, the standard study period of secondary school in Anonymous-Country is 4 years. The fourth year of secondary school is rather crucial for all teenagers since they need to prepare for O level examination.

Nearly half of teenagers (about 46%) surveyed access as often as a few times every day to their favorite social network websites (see Figure 7 in Appendix A). The majority of boys do not access to the social network websites like girls do, about only 36.4% of boys as compared to 46.7% of girls use the websites only a few times each day; boys access those websites frequently only on weekly basis (see Table 7 in Appendix B). The largely available Internet access infrastructure in Anonymous-Country could be the main reason for that frequent access to social network websites so easily.

What attract both girls and boys in social network websites include seeking for fun and entertainment, as well as interacting with their friends in real life (see Figure 8 in Appendix A). Very little use it for showing their online presence, or finding online friends. The majority of subjects don’t see participating online social network is the trend they need to follow. This may imply that they have clear objective in mind prior to start online social activities.

And the most of them (81.4%), particularly the boys, do not look at social network websites as the source of the information (see Table 8 in Appendix B).

As to whom the teenagers would add into their friend list for online social activities, we find (see Figure 7)

  • The most of teenagers (about 95% overall) choose their friends in real life.
  • About four fifth of them socialize with their family members; while another one fifth choose NOT to do so in particular.
  • Two third of girls interacts with their schoolmates on social network websites; while boys have NO preference in this regard – about 60% of boys choose not to do so.
  • What is noticeable is that about one quarter of subjects is willing to socialize with total strangers.
  • Apart from the aforementioned options, there is NO other category appearing in the friend list.

Figure 7 Who is on friend list

About 86% of teenagers regardless of gender claimed they have more than 100 friends online (see Figure 9 in Appendix A). Given that most of people the teenagers interact with online are their classmates and family members, it is not hard for them to have more than 100 friends.

The top 5 information the teenagers would like to include in their online profiles are as follows (see Figure 10):

  1. Real name (87.63%)
  2. Date of birth (77.32%)
  3. Education (62.8%)
  4. Videos (55.67%)
  5. Photos (43.3%)

Figure 10 What are in teenagers’ profiles

Comparing the different gender groups, boys like videos the most since there are more that 70% of them choosing to have videos in their profiles, whereas only 50% of girls are willing to do the same. Another noticeable difference between girls and boys is photos. Slightly more than 50% of girls add photos to their profiles, while less than 20% of boys choose to do so (see Figure 11).

Figure 11 Comparison of profiles between girls and boys

We employ Chi-Square or Fisher’s exact test to find if there is any significant difference between girls and boys on their intention of including what type of information in their online profiles (see Table 1):

Information Chi-Square/Fisher’s exact test value Asymp. Sig. / Exact Sig. Degree of freedom Statistically significant difference Reference
Real name .026 1 YES see Table 11 in Appendix B
DOB .039 1 YES see Table 12 in Appendix B
Education 5.888 .015 1 YES see Table 13 in Appendix B
Interests 5.732 .017 1 YES see Table 14 in Appendix B
Religion .287 .592 1 NO see Table 15 in Appendix B
Relationship status .848 .357 1 NO see Table 16 in Appendix B
Address .147 1 NO see Table 17 in Appendix B
Email .958 .328 1 NO see Table 18 in Appendix B
Sexual orientation .730 1 NO see Table 19 in Appendix B
IM name 3.407 .065 1 NO see Table 20 in Appendix B
Phone# .128 1 NO see Table 21 in Appendix B
Your website 1 1 NO see Table 22 in Appendix B
Your photos 7.311 .007 1 YES see Table 23 in Appendix B
Your videos 5.381 .020 1 YES see Table 24 in Appendix B
Others .128 1 NO see Table 25 in Appendix B

Table 1 Chi-Square/Fisher’s exact test for Q9 between girls and boys

Among all information teenagers put up online, girls and boys take very different approach in unveiling real name, DOB, education, interests, photos, and videos.

  • Real name: girls tend to provide their real name in their profile; while boys are very conservative in this regards. There is similar tendency in DOB.
  • Personal interests: though both girls and boys are not willing to provide this information, still girls tend to share it more than boys do. Photos have the similar trend.
  • Videos: majority of boys like to share videos, whereas girls do not.

As to what make teenagers feel safe on their account information, none of the following factors matters: privacy policy, certification assurance, site reputation. Instead, only trusted people who can view their account information make them feel their account information is in safety (see Figure 13).

Figure 13 Q16 – What helps safety

Both girls and boys have the same intention in this regard (see Figure 12 in Appendix A).

Overall, about 67% of teenagers use the privacy features provided by social network websites. There are 35% of girls while only 27% of boys using such features all the time. However, 45.5% of boys use it for sure for important information. About 15.5% of subjects find the privacy feature ineffective (see Table 9 in Appendix B).

Upon comparing girls and boys on their intention of using privacy features provided by social network websites, the result of independent sample t-test shows that there is NO statistically significant difference between different gender groups (t = -.104, p = .917) (see Table 10 in Appendix B). The gender may not matter since they all are coming from similar education background, and having similar experiences with social network websites in terms of how long and how often they use.

As for which privacy features teenagers would use, controlling who can view photos is the most welcome feature comparing to the rest (50.67%), videos come next (35.05%) (see Figure 14).

Figure 14 Q18 – Privacy features

The observations from the Figure 14:

  • About 73.3% of girls and 77.3% of boys do NOT take control over who can view their real name.
  • One fourth of girls (25.3%) and slightly more than one third of boys (or 36.4%) take action to limit who can view their age. Overall, only 72.2% of teenagers choose not to do so.
  • Similarly, as to education, 76.3% of teenagers do not mind to let public see their educational background. But it is more sensitive to girls (26.7%) than boys (13.6%).
  • 20.6% or one fifth of teenagers care who can see their interests. While the rest (or 79.4%) leave it open to everyone. Boys and girls have equivalent level of concern in this regard.
  • Even less, about only 15.5% of respondents unveil their religion to public, and the majority does not bother too much. Boys and girls have equivalent level of concern in this regard.
  • 80.4% of teenagers leave their online status visible to anyone. While the girls seem less worried about this – only 16% of them control this bit of information, compared to the boys which are near 32%, feeling a bit sensitive to this piece of information.
  • As to address, girls behave more protectively than boys do. 26.7% of girls vs. 13.6% of boys control who can view their address. Still, the majority of respondents leave this out.
  • 7 out of 10 (or 71.1%) of teenagers share their email address publicly, i.e. do not control who can view their email addresses.
  • Sexual orientation and IM name are two of the least concerns among teenagers, and only 1 out of 10 (or 10.3%) control who can view these information.
  • 74.2% of teenagers would not restrict the access to their phone numbers.
  • 83.5% of respondents do not restrict the access to their websites.
  • Photo is the utmost concern among teenagers. Unlike other features or information, more than 53% of respondents express their worry about the access to their photos. Surprisingly, boys are more concerned about their photos than girls are.
  • Videos are the second most concerned – 35% of respondents control who can view their videos. This means the majority of them do not control the access to it. Similar to photos, boys are more concerned who can view their videos than girls are.
  • More than 80% of teenagers do not block other users. Girls slightly tighten the blockage than boys do.
  • More than 71% of respondents do not block applications used in social network websites. There are more girls using this feature than boys.
  • Slightly less than one fourth of teenagers control who can make comments. Similar to the aforementioned two features, the number of girls using this feature is almost as twice as that of boys, i.e. 28% vs. 13.6%. But overall 75.3% of teenagers do not use this feature.
  • Conversely, there are more boys controlling who can send message than girls. However, when looking at all respondents as a whole, more than 82% of them do not use this feature at all.
  • Slightly above 20% of respondents choose to hide their online status, which means the majority of them do not.
  • Another least concern from teenagers is that they do not control if they can be found through search engine.
  • In summary, the most concern to teenagers is photos, videos are the second; the least of concerns include sexual orientation, IM name as well as being found through search.

Using Chi-Square t-test/Fisher’s exact test, we find out that phone number is the only statistically significant difference between girls and boys, i.e. girls would not restrict the access to their phone numbers. Here is the summary of test result (see Table 2).

Information Chi-Square/Fisher’s exact test value Asymp. Sig. / Exact Sig. Degree of freedom Statistically significant difference Reference
Real name .138 .710 1 NO see Table 26 in Appendix B
DOB 1.03 .310 1 NO see Table 27 in Appendix B
Education 1.597 .206 1 NO see Table 28 in Appendix B
Interests .770 1 NO see Table 29 in Appendix B
Religion 1 1 NO see Table 30 in Appendix B
Status .128 1 NO see Table 31 in Appendix B
Address 1.597 .206 1 NO see Table 32 in Appendix B
Email .121 .728 1 NO see Table 33 in Appendix B
Sexual orientation 1 1 NO see Table 34 in Appendix B
IM name .447 1 NO see Table 35 in Appendix B
Phone Number 6.702 .010 1 YES see Table 36 in Appendix B
Your website 1 1 NO see Table 37 in Appendix B
Your photos 1.151 .336 1 NO see Table 38 in Appendix B
Your videos 1.353 .245 1 NO see Table 39 in Appendix B
Block user .550 1 NO see Table 40 in Appendix B
Block application .522 .470 1 NO see Table 41 in Appendix B
Make comments 1.885 .170 1 NO see Table 42 in Appendix B
Send message .526 1 NO see Table 43 in Appendix B
Online status 1 1 NO see Table 44 in Appendix B
Search 1 1 NO see Table 45 in Appendix B

Table 2

We use Independent Sample T-test for the comparison of intention between girls and boys to see if there is any statistically significant difference between these two different gender groups. The result shows that there is no significant difference between different gender groups in this regard.

Question

T value

P value

Degree of freedom

Statistically significant difference

Reference

19. I am concerned that my information may be inappropriately viewed by others without my knowledge.

1.465

.146

95

NO

see Table 46 in Appendix B
20. I am concerned that my information may be inappropriately

forwarded to others without my knowledge.

.667

.507

95

NO

see Table 47 in Appendix B21. I am concerned that the people I only know online are not who they claim to be.

-.452

.653

92

NO

see Table 48 in Appendix B22. I am concerned that others may post information about me online without my knowledge.

.638

.525

94

NO

see Table 49 in Appendix B23. I am concerned about threats to privacy (e.g. identity theft, stalking etc.).

1.046

.298

94

NO

see Table 50 in Appendix B24. I am sufficiently warned by my chosen social network website about the risks of posting information about myself online.

1.489

.140

94

NO

see Table 51 in Appendix B25. I will provide real information about myself on my chosen social network website if required.

.568

.571

95

NO

see Table 52 in Appendix B

 

Table 3 Comparison between girls and boys for Q19 – 25

  • As to whether or not teenagers are concerned about if their information is viewed inappropriately, most of respondents agree or strongly agree. And no boys would disagree with it. However, it is noticeable that there are 3 girls out of 97 respondents expressing their disagreement.
  • As to the concern with if information is forwarded to others without prior knowledge, more than 82% of teenagers agree. There is no disagreement at all.
  • Nearly 66% of teenagers are worried about the people they know online is not who they claim to be. This group of teenagers seems aware of danger of the impersonation. Only slightly more than 10% of respondents are not worried.
  • Only 1% of teenagers is NOT concerned if other post information about them without their knowledge, while the majority (or 81.5% of respondents) does not seem to like others posting information about them without their knowledge.
  • Overall, about 83.5% of teenagers are worried about the threat to their privacy. While no boy disagrees with that, about 7.2% or 7 girls out of 97 respondents do not worry at all.
  • 85.6% of respondents agree that they were warned about the risks of posting information online. While no boy disagrees with that, there are only 2 girls expressing their disagreement.
  • Only 65% of teenagers would like to provide the real information about them when it is required, while slightly more than one fourth of teenagers remain neutral, only 7 out of 97 do not provide the real information.

Comparison between teenagers and adults

Both teenagers and adults are using social network websites very often – from a few times a week to a few times a day. This is common for both age groups.

However, there are significant differences in other aspects of social network websites. On what information is included in the online profile between adults and teenagers, below table (see Table 4) is the summary of test result:

Information Chi-Square/Fisher’s exact test value Asymp. Sig. / Exact Sig. Degree of freedom Statistically significant difference Reference
Real name 2.715 .099 1 NO see Table 53 in Appendix B
DOB .085 .771 1 NO see Table 54 in Appendix B
Education .102 .749 1 NO see Table 55 in Appendix B
Interests 13.581 .000 1 YES see Table 56 in Appendix B
Religion .027 .869 1 NO see Table 57 in Appendix B
Relationship status .000 .986 1 NO see Table 58 in Appendix B
Address .084 .772 1 NO see Table 59 in Appendix B
Email 11.237 .001 1 YES see Table 60 in Appendix B
Sexual orientation 36.889 .000 1 YES see Table 61 in Appendix B
IM name 4.273 .039 1 YES see Table 62 in Appendix B
Phone number 5.209 .022 1 YES see Table 63 in Appendix B
Your website 1.938 .164 1 NO see Table 64 in Appendix B
Your photos 20.529 .000 1 YES see Table 65 in Appendix B
Your videos 28.875 .000 1 YES see Table 66 in Appendix B
Others 1.322 .250 1 NO see Table 67 in Appendix B

Table 4 Chi-Square/Fisher’s exact test FOR Q9 between adults and teenagers

  • Personal interests: while adults are more open to provide the personal interests information in their account profiles, teenagers tend not to do so. The reason of this significant difference could be due to the differences of their experiences. Unlike adults who are more experienced and clear what they are interested in, teenagers still are the age of developing their interests and many things remain unknown to them. Therefore, they are not very clear on what are really interested them at this age yet.
  • Email addresses: teenagers intend to share their email addresses in their profiles, as compared to adults. Similarly, teenagers also intend to include IM name in their profiles. Both are tightly integrated into social network websites, and are the most important means for keeping in touch with friends. Adults seem reluctantly in this regard may be due to they spend more time on their work. Thus, they will have less time to check those emails and chat with friends via IM, but rather through other means of communication such as mobile phones to stay in touch. Whereas teenagers have plenty of time, as compared to adults, in using email and IM most of time.
  • Sexual orientation: Teenagers, unlike adults, less than 15% of them treat this one of important components in their profiles. Most of them may not be well aware of the dynamics of the world. Comparatively, adults are mature and aware of the different sexual orientations that exist. In order to avoid unexpected disturbance, they choose to clearly state their sexual orientation in their profile.
  • Phone number: among the people who choose to put their contact numbers in their online profiles, including both adults and teenagers, the majority of them are adults. Perhaps unlike adults, not every teen owns a phone. Therefore, they are not contactable easily via phone while adults do not have such constraint. Or maybe it is contradicting against their interest – to seek for fun. Despite this observation, it is noted that neither the most of adults nor the most of teenagers are willing to do so.
  • Comparing photos and videos, adults are more conservative to putting up videos than teenagers are. On the other hand, teenagers become less open to share their photos in their online profiles than adults do. Undoubtedly, videos can tell more about a person than photos can, and videos are much more fun than photos. Perhaps this is correlated to the different purpose of participating online social networking. Teenagers seem to find the fun part of sharing videos since 70% of them participating social networking is intent for fun and entertainment (see Figure 15). While, adults want to connect with the friends in real life, photos are sufficient to fulfill such purpose.


Figure 15 Purposes of online social networking from teenagers and adults perspectives

Table 5 below shows if there is any significant difference between adults and teenagers on what privacy feature would be used during online social interaction.

Information Chi-Square/Fisher’s exact test value Asymp. Sig. / Exact Sig. Degree of freedom Statistically significant difference Reference
DOB 6.595 .010 1 YES see Table 68 in Appendix B
Education .621 .431 1 NO see Table 69 in Appendix B
Interests .947 .330 1 NO see Table 70 in Appendix B
Religion .454 .501 1 NO see Table 71 in Appendix B
Status .084 .772 1 NO see Table 72 in Appendix B
Address .621 .431 1 NO see Table 73 in Appendix B
Email 2.821 .093 1 NO see Table 74 in Appendix B
Sexual orientation .283 .595 1 NO see Table 75 in Appendix B
IM name 2.982 .084 1 NO see Table 76 in Appendix B
Phone Number 4.637 .031 1 YES see Table 77 in Appendix B
Your website .084 .771 1 NO see Table 78 in Appendix B
Your photos .003 .954 1 NO see Table 79 in Appendix B
Your videos 6.093 .014 1 YES see Table 80 in Appendix B
Block user 9.901 .002 1 YES see Table 81 in Appendix B
Block application .020 .889 1 NO see Table 82 in Appendix B
Make comments .047 .829 1 NO see Table 83 in Appendix B
Send message .010 .920 1 NO see Table 84 in Appendix B
Online status 7.200 .007 1 YES see Table 85 in Appendix B
Search 4.337 .037 1 YES see Table 86 in Appendix B

Table 5 Chi-Square/Fisher’s exact test FOR Q18 between adults and teenagers

We use Independent Sample T-test for the comparison of intention between teenagers and adults and we find many statistically significant differences between these two age groups.

Question

T value

P value

Degree of freedom

Statistically significant difference

Reference

19. I am concerned that my information may be inappropriately viewed by others without my knowledge.

2.659

.009

189

YES

see Table 87 in Appendix B
20. I am concerned that my information may be inappropriately

forwarded to others without my knowledge.

2.220

.028

189

YES

see Table 88 in Appendix B21. I am concerned that the people I only know online are not who they claim to be.

2.734

.007

186

YES

see Table 89 in Appendix B22. I am concerned that others may post information about me online without my knowledge.

4.294

.000

188

YES

see Table 90 in Appendix B23. I am concerned about threats to privacy (e.g. identity theft, stalking etc.).

1.265

.208

188

NO

see Table 91 in Appendix B24. I am sufficiently warned by my chosen social network website about the risks of posting information about myself online.

7.865

.000

188

YES

see Table 92 in Appendix B25. I will provide real information about myself on my chosen social network website if required.

1.955

.052

189

NO

see Table 93 in Appendix B

 

Table 6 Chi-Square/Fisher’s exact test for Q19-25 between adults and teenagers

According to the result shown in the Table 6, we find out that

  • Adults are not so concerned with their information being viewed by others without their knowledge, compared to teenagers.
  • 10% of adults are not concerned with their information being forwarded to others without their knowledge. Whereas none of teenagers disagree sin this regard.
  • Teenagers are more worried about the people they know online are not who they claim to be. 11% of adults do not worry at all.
  • Some of adults are not concerned with others posting their information, while teenagers seem unable to tolerate such incidents.
  • However, the threat to privacy becomes the common concern to both age groups.
  • Compared to teenagers, approximately 22% of adults are not satisfied with the existing services that social network websites offer to warn their users on the risk of posting information.
  • As to if the real information would be provided to social network websites when required, teenagers are more honest than adults are. Perhaps, teenagers are not well aware of the adverse impact or negative consequence of information misuse or other fraudulent activities.

Possible Solutions

We can look at three different approaches to protect privacy in online social network websites (SNS) such as:

Social Solutions

Parents, schools, and social networking service providers can work together on various social solutions to the privacy issue. Experts (Sullivan, 2005) agree that the first step in building protections for teenage starts with parents. Parents need to involve and monitor teenagers’ online intention in a friendly manner. We know most teenagers are advance in computer usage than their parents but parents should spend time to learn these differences.

Government and non-profit organization play a part to educate parents and teenagers on the online danger and privacy issues such effort exits in US for an example of “The Federal Bureau of Investigation and the National Center for Missing and Exploited Children offers parent advice for detecting whether their children are engaging in appropriate behavior” (Susan, 2006).

Schools can play their part to educate students about ethics on online posting and how to safe guard their privacy in online. Schools can share real-world example for privacy misuse case studies, organize seminar, symposium and bulletin board to run awareness campaign.

Commercial social networking websites providers like MySpace and Facebook should work with different educational and law enforcing agencies to protect teenagers from privacy misuse related danger. Service providers also can run online campaign by posting safety ads, banners, advertise on Interactive media as well as news coverage on Radio, TV and Teenagers magazines and newsletter.

In a nutshell protecting of teenagers is a part of parent responsibilities but the education growing privacy problem for both parent and teenagers will require an educational effort that involves social network service providers, education institutes and government agencies (Susan, 2006).

Technical Solutions

On top of social awareness, social networking site providers should explore technical solutions to protect their users sensitive information. Tightening the privacy setting may give users more controls over who can sees what information or design user profile such a way that it is secure by default. That is to say no one can see any information unless users authorize to publish their information. When user enables such feature, web websites should also prompt for possible warning messages for possible consequences for such disclosure. Installation of age and identity verification system to enforce privacy setting and response to inappropriate contents could help to tackle such issue. Schools, libraries, home computer to install internet filter engine to control and monitor teenagers online behaviors also could be a part of the solution.

Legal Solutions

There are growing concerns over the misuse of social networking websites. A legal solution to the privacy problem encompasses the monitoring of social networking websites and technical solutions. Appropriate legislation by the government body and effective law enforcement is essential to solving privacy issues. Many counties already have special law to protect private information such as ‘The Deleting Online Predators Act’ (Fitzpatrick, 2006) in US and ‘Data Protection Act’ in UK (ICO, 2010). However we are unable to find current specific legislation for the protection of personal data in Anonymous-Country. Perhaps this kind of matters can be dealt with other civil or criminal law (Privacy International, 2003).

Lately, social networking websites such as Facebook and MySpace have been caught for sharing private data with advertisers (Jacobsson, 2010). This is clearly breaking their privacy policies. Therefore every country should have proper law and enforcement to deal with violation of privacy issue caused by the social networking service providers.

CONCLUSION

Our findings highlighted insights to the online socializing intention of two groups (adult and teenagers) of users and found teenagers are more open to share information online and less awareness about the privacy issue. we have also analyze the information in different dimension to find gender preferences based on Teenagers and found female are higher engagement rates on social network sties.

Now Legislation to protect teenagers against the misuse of their personal information from predators is a serious concern. Root of the privacy paradox is a control for the personal information and appropriate steps is necessary to resolve this paradox. Currently legal system and industry responses to focus on issue of predators and undermine the social response to privacy issue in social networks (Susan, 2006).

The remedy to the paradox is not very simple. We need all levels of society engagement to tackle the social issues related to teenagers and online privacy. Awareness is the key to solving the issue. We all have to do our part to educate each other and protecting our privacy on the social networking websites.

All these details above shows interaction of trust and privacy issues in social networking websites is not fully understood yet to be a sufficient degree to allow accurate modeling on behavior and activities. The results of this report encourage further study on the effort to understand online social environment and relationship (Dwyer, 2007).

LIMITATION

Despite how comprehensive we have tried our survey to be, there are bound to some limitations, from the way we carried out the survey, how the survey being conducted and the data collected. The following highlight some of the limitations of the survey:

Number of Respondents

Due to other factors, we can’t have a comprehensive survey covering large quantity of respondents. Large quantity of data will provide more advantages, which are explained as follows:

  • Invalid survey responds can be removed or eliminated with less impact on the overall survey result.
  • More data can be analyzed which will give more accurate result and dependency.
  • Result is more guaranteed which shows rate of accuracy towards certain trends.
  • Some minor corrections can be made without affecting the overall survey result.
  • Error can be minimized or eliminated from indicative responses given from majorities of responses.
  • Future survey can be done in less error based on earlier experience.

The disadvantages of having lesser quantity of data are the opposite of the advantages listed above which this survey has. Should we have a larger population, our survey rate result would probably been better.

Honesty and Accuracy

The accuracy rate of the survey is higher if only the respondents give an honest and true answer, but there is no way that we can validate the genuine from the fake responses, of course except for the obvious one. Just like an older man claimed to be young innocent girl on the internet chat room. This issue in fact is common to all surveys and questionnaires. We understand that there are many reasons or factors that caused the respondents giving invalid or untrue responses, such as limited knowledge of the subject, lack of attention, rushing for time and completely ignore the question.

One way to minimize the invalid respond is to use the online survey with some conditions attached to the answer, for example having compulsory question which the respondent will not be able to proceed without giving answer. This is one advantage of online survey as compared to hard copy survey, which computation can avoid or minimize error such typo error, invalid response, etc. But again online survey sometimes does not guarantee error free responds.

Geography

This survey is conducted on specific geography region, the respondents population are based in Anonymous-Country. As we all know that computer privacy and security issue encompasses and applicable to everyone (users of the Internet regardless of their country or region. The survey result can vary from country (or region) to country (or region), factors that affecting the survey result such as custom, culture, tradition, etc can have significance differences. Such that this survey can be best describe based on Anonymous-Country teenagers, it may or may not applicable to teenagers in other country for example USA, China or Indonesia.

Acknowledgements

This work is supervised by Assistant Professor Na Jin Cheon. We thank him for the valuable comments that help improve the paper.REFERENCES

[1]. Sarah Jacobsson (2010). Facebook and MySpace sharing private data with advertisers. Retrieved 31 Aug, 2010, from http://news.techworld.com/security/3224231/facebook-and-%20%20%20myspace-sharing-private-data-with-advertisers/?olo=rss

[2]. Ajith Abraham, Aboul-Ella Hassanien, Vaclav Snásel (2010). Computational Social Network Analysis: Trends, Tools and Research Advances. Springer

[3]. Retrieved 30 Aug, 2010 from http://en.wikipedia.org/wiki/Facebook_features

[4]. Heather Havenstein (2007). Facebook, MySpace users will trade privacy for features. Retrieved Aug 30, 2010 from http://www.computerworld.com/s/article/9037379/Facebook_MySpace_users_will_trade_privacy_for_features

[5]. Balachander Krishnamurthy, Craig E. Wills (2008). Characterizing privacy in online social networks. In Proceedings of the first workshop on Online social networks. New York, ACM

[6]. Anna Cinzia Squicciarini, Mohamed Shehab, Federica Paci (2009). Collective privacy management in social networks. In Proceedings of the 18th international conference on World Wide Web. New York, ACM

[7]. Smriti Bhagat, Graham Cormode, Balachander Krishnamurthy, Divesh Srivastava (2010). Privacy in dynamic social networks. In Proceedings of the 19th international conference on World Wide Web. New York, ACM

[8]. Alyson L. Young, Anabel Quan-Haase (2009). Information revelation and internet privacy concerns on social network websites: a case study of facebook. In Proceedings of the fourth international conference on Communities and technologies. New York, ACM

[9] Dwyer S. Hiltz S. Passerini K. (2007). Trust and Privacy Concern Within Social Networking Websites: A comparison of Facebook and MySpace. Americas Conference on Information Systems (AMCIS)

[10] Susan B. B (2006). A privacy Paradox: Social networking in the United States. Retrieved September 16, 2010 from http://firstmonday.org/htbin/cgiwrap/bin/ojs/index.php/fm/article/viewArticle/1394/1312

[11] Boyd, d. m., & Ellison, N. B. (2007). Social network websites: Definition, history, and scholarship. Journal of Computer-Mediated Communication, 13(1), article 11. Retrieved September 10, 2010 from http://jcmc.indiana.edu/vol13/issue1/boyd.ellison.html

[12] Simon, H. A. (1971), “Designing Organizations for an Information-Rich World”, in Martin Greenberger, Computers, Communication, and the Public Interest, Baltimore, MD: The Johns Hopkins Press. Retrieved on 21 October, 2010, from http://en.wikipedia.org/wiki/Attention_economy.

[13] Tim Berners-Lee (1999). Weaving the Web: The Original Design and Ultimate Destiny of the World Wide Web. Retrieved 10 October, 2010, from http://www.w3.org/People/Berners-Lee/Weaving/Overview.html

[14] Howard Rheingold (1985). Tools for Thought. Retrieved 10 October, 2010, from http://www.rheingold.com/texts/tft/.

[15]Howard Rheingold (1993). Virtual Communities. Retrieved 10 October, 2010, from http://www.rheingold.com/vc/book/.

[16] Danah Boyd, Eszter Hargittai (2010). Facebook privacy settings: who cares. Retrieved 10 October, 2010, from http://firstmonday.org/htbin/cgiwrap/bin/ojs/index.php/fm/article/view/3086/2589.

[17] Danah Boyd’s (2006). Identity Production in a Networked Culture: Why Youth Heart MySpace. American Association for the Advancement of Science, St. Louis, MO. February 19. Retrieved 20 October, 2010 from http://www.danah.org/papers/AAAS2006.html.

[18] Andrew Schrock, Danah boyd, (2008). Online Threats to Youth: Solicitation, Harassment, and Problematic Content. Retrieved on 30 October, 2010 from http://cyber.law.harvard.edu/pubrelease/isttf/.

[19] Danah Boyd (2010). Social Network Websites as Networked Publics: Affordances, Dynamics, and Implications. In Networked Self: Identity, Community, and Culture on Social Network Websites (ed. Zizi Papacharissi), pp. 39-58.

[20] Danah Boyd (2007). Why Youth (Heart) Social Network Websites: The Role of Networked Publics in Teenage Social Life. MacArthur Foundation Series on Digital Learning – Youth, Identity, and Digital Media Volume (ed. David Buckingham). Cambridge, MA: MIT Press.

[21] Joshua Porter (2008). Design for the social web. New Riders Publishing.

[22] Ai Ho, Abdou Maiga, Esma Aïmeur (2009). Privacy protection issues in social networking websites. AICCSA 2009. IEEE/ACS International Conference on, vol., no., pp.271-278, 10-13 May 2009

[23] Retrieved 10 October, 2010, from http://mattmckeon.com/facebook-privacy/.

[24] Anders Albrechtslund (2008). Online Social Networking as Participatory Surveillance. First Monday, Volume 13, Number 3 – 3 March 2008. Retrieved 11 November 2010, from
http://firstmonday.org/htbin/cgiwrap/bin/ojs/index.php/fm/article/viewArticle/2142/1949

[25] Dwyer, C.,Hiltz†, S,. Passerini, K. (2007). Trust and Privacy Concern Within Social Networking Websites: A Comparison of Facebook and MySpace. AMCIS 2007 Proceedings. Association for Information Systems. Retrieved November 10, 2010, from http://aisel.aisnet.org/cgi/viewcontent.cgi?article=1849&context=amcis2007

[26] Wikipedia (2010). The deleting online predator act (DOPA). Retrieved November 10, 2010, from http://en.wikipedia.org/wiki/Deleting_Online_Predators_Act_of_2006#cite_note-15

[27] ICO (2010). Personal information online – code of practice. Retrieved November 17, 2010, from http://www.ico.gov.uk/for_organisations/data_protection/topic_guides/online.aspx

[28] Privacy International (2003). Privacy and Human Rights. Retrieved October 21, 2010, from http://www.privacyinternational.org/survey/phr2003/countries/Anonymous-Country.htm

[29] TechWorld (2010). Facebook and MySpace sharing private data with advertisers – Social networks may be breaking their own privacy policies. Retrieved November, 2010, from http://news.techworld.com/security/3224231/facebook-and-myspace-sharing-private-data-with-advertisers/?olo=rss

[30] Wikipedia (2010). The deleting online predator act (DOPA). Retrieved November 10, 2010, from http://en.wikipedia.org/wiki/Deleting_Online_Predators_Act_of_2006#cite_note-15

[31] Mayer, R. C., J. H. Davis, and F. D. Schoorman (1995) “An Integrative Model of Organizational Trust,” The

Academy of Management Review (20) 3, pp. 709-734

APPENDIX A

Figure 1 Facebook privacy growth

From http://mattmckeon.com/facebook-privacy/.

Figure 2 Comparison between Facebook and MySpace

Figure 3 Age Profile

Figure 4 Gender

Figure 5 Number of users by Social Network Websites


Figure 6 Duration of access to Social Network Websites

Figure 7 Frequency of access to Social Network Websites

Figure 8 Reason for being part of Social Network Websites

Figure 9 Number of friends

Figure 12 Comparison of perception of safety between girls and boys

APPENDIX B

Frequency * Gender Crosstabulation

Gender

Total

Female

Male

Frequency A few times/day Count

35

8

43

% within Gender

46.7%

36.4%

44.3%

A few times/week Count

22

9

31

% within Gender

29.3%

40.9%

32.0%

When there is new post Count

4

2

6

% within Gender

5.3%

9.1%

6.2%

When I get notification Count

7

0

7

% within Gender

9.3%

.0%

7.2%

Others Count

7

3

10

% within Gender

9.3%

13.6%

10.3%

Total Count

75

22

97

% within Gender

100.0%

100.0%

100.0%

Table 7 Frequency * Gender crosstab

find_information * Gender Crosstabulation

Gender

Total

Female

Male

find_information 0 Count

59

20

79

% within Gender

78.7%

90.9%

81.4%

1 Count

16

2

18

% within Gender

21.3%

9.1%

18.6%

Total Count

75

22

97

% within Gender

100.0%

100.0%

100.0%

Table 8 Find informatoin * Gender crosstab

Use_privacy_features * Gender Crosstabulation

Gender

Total

Female

Male

Use_privacy_features Always Count

26

6

32

% within Use_privacy_features

81.3%

18.8%

100.0%

% within Gender

34.7%

27.3%

33.0%

yes, for important info Count

23

10

33

% within Use_privacy_features

69.7%

30.3%

100.0%

% within Gender

30.7%

45.5%

34.0%

no, I don’t know Count

10

0

10

% within Use_privacy_features

100.0%

.0%

100.0%

% within Gender

13.3%

.0%

10.3%

no, I don’t know how to use Count

4

3

7

% within Use_privacy_features

57.1%

42.9%

100.0%

% within Gender

5.3%

13.6%

7.2%

no, they are not effective Count

12

3

15

% within Use_privacy_features

80.0%

20.0%

100.0%

% within Gender

16.0%

13.6%

15.5%

Total Count

75

22

97

% within Use_privacy_features

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Table 9 Privacy feature * Gender crosstab

Independent Samples Test

Levene’s Test for Equality of Variances

t-test for Equality of Means

95% Confidence Interval of the Difference

F

Sig.

t

df

Sig. (2-tailed)

Mean Difference

Std. Error Difference

Lower

Upper

Use_privacy_features Equal variances assumed

.046

.830

-.104

95

.917

-.036

.344

-.718

.647

Equal variances not assumed

-.105

34.667

.917

-.036

.341

-.729

.657

Table 10 T-test for Privacy feature between girls and boys

Crosstab

Gender

Total

Female

Male

Real_name No Count

6

6

12

% within Real_name

50.0%

50.0%

100.0%

% within Gender

8.0%

27.3%

12.4%

Yes Count

69

16

85

% within Real_name

81.2%

18.8%

100.0%

% within Gender

92.0%

72.7%

87.6%

Total Count

75

22

97

% within Real_name

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

5.828a

1

.016

Continuity Correctionb

4.186

1

.041

Likelihood Ratio

5.008

1

.025

Fisher’s Exact Test

.026

.026

Linear-by-Linear Association

5.768

1

.016

N of Valid Cases

97

a. 1 cells (25.0%) have expected count less than 5. The minimum expected count is 2.72.
b. Computed only for a 2×2 table

Table 11 Chi-Square/Fisher’s exact test for Q9 between girls and boys

Crosstab

Gender

Total

Female

Male

DOB No Count

13

9

22

% within DOB

59.1%

40.9%

100.0%

% within Gender

17.3%

40.9%

22.7%

Yes Count

62

13

75

% within DOB

82.7%

17.3%

100.0%

% within Gender

82.7%

59.1%

77.3%

Total Count

75

22

97

% within DOB

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

5.391a

1

.020

Continuity Correctionb

4.131

1

.042

Likelihood Ratio

4.928

1

.026

Fisher’s Exact Test

.039

.024

Linear-by-Linear Association

5.336

1

.021

N of Valid Cases

97

a. 1 cells (25.0%) have expected count less than 5. The minimum expected count is 4.99.
b. Computed only for a 2×2 table

Table 12 Chi-Square/Fisher’s exact test for Q9 between girls and boys

Crosstab

Gender

Total

Female

Male

education No Count

23

13

36

% within education

63.9%

36.1%

100.0%

% within Gender

30.7%

59.1%

37.1%

Yes Count

52

9

61

% within education

85.2%

14.8%

100.0%

% within Gender

69.3%

40.9%

62.9%

Total Count

75

22

97

% within education

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

5.888a

1

.015

Continuity Correctionb

4.734

1

.030

Likelihood Ratio

5.726

1

.017

Fisher’s Exact Test

.023

.016

Linear-by-Linear Association

5.828

1

.016

N of Valid Cases

97

a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 8.16.
b. Computed only for a 2×2 table

Table 13 Chi-Square/Fisher’s exact test for Q9 between girls and boys

Crosstab

Gender

Total

Female

Male

interests No Count

44

19

63

% within interests

69.8%

30.2%

100.0%

% within Gender

58.7%

86.4%

64.9%

Yes Count

31

3

34

% within interests

91.2%

8.8%

100.0%

% within Gender

41.3%

13.6%

35.1%

Total Count

75

22

97

% within interests

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

5.732a

1

.017

Continuity Correctionb

4.580

1

.032

Likelihood Ratio

6.434

1

.011

Fisher’s Exact Test

.021

.013

Linear-by-Linear Association

5.673

1

.017

N of Valid Cases

97

a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 7.71.
b. Computed only for a 2×2 table

Table 14 Chi-Square/Fisher’s exact test for Q9 between girls and boys

Crosstab

Gender

Total

Female

Male

religion No Count

49

13

62

% within religion

79.0%

21.0%

100.0%

% within Gender

65.3%

59.1%

63.9%

Yes Count

26

9

35

% within religion

74.3%

25.7%

100.0%

% within Gender

34.7%

40.9%

36.1%

Total Count

75

22

97

% within religion

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

.287a

1

.592

Continuity Correctionb

.080

1

.777

Likelihood Ratio

.284

1

.594

Fisher’s Exact Test

.620

.384

Linear-by-Linear Association

.284

1

.594

N of Valid Cases

97

a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 7.94.
b. Computed only for a 2×2 table

Table 15 Chi-Square/Fisher’s exact test for Q9 between girls and boys

Crosstab

Gender

Total

Female

Male

status No Count

49

12

61

% within status

80.3%

19.7%

100.0%

% within Gender

65.3%

54.5%

62.9%

Yes Count

26

10

36

% within status

72.2%

27.8%

100.0%

% within Gender

34.7%

45.5%

37.1%

Total Count

75

22

97

% within status

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

.848a

1

.357

Continuity Correctionb

.449

1

.503

Likelihood Ratio

.834

1

.361

Fisher’s Exact Test

.453

.250

Linear-by-Linear Association

.839

1

.360

N of Valid Cases

97

a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 8.16.
b. Computed only for a 2×2 table

Table 16 Chi-Square/Fisher’s exact test for Q9 between girls and boys

Crosstab

Gender

Total

Female

Male

address No Count

62

15

77

% within address

80.5%

19.5%

100.0%

% within Gender

82.7%

68.2%

79.4%

Yes Count

13

7

20

% within address

65.0%

35.0%

100.0%

% within Gender

17.3%

31.8%

20.6%

Total Count

75

22

97

% within address

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

2.181a

1

.140

Continuity Correctionb

1.385

1

.239

Likelihood Ratio

2.027

1

.155

Fisher’s Exact Test

.147

.121

Linear-by-Linear Association

2.158

1

.142

N of Valid Cases

97

a. 1 cells (25.0%) have expected count less than 5. The minimum expected count is 4.54.
b. Computed only for a 2×2 table

Table 17 Chi-Square/Fisher’s exact test for Q9 between girls and boys

Crosstab

Gender

Total

Female

Male

email No Count

46

16

62

% within email

74.2%

25.8%

100.0%

% within Gender

61.3%

72.7%

63.9%

Yes Count

29

6

35

% within email

82.9%

17.1%

100.0%

% within Gender

38.7%

27.3%

36.1%

Total Count

75

22

97

% within email

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

.958a

1

.328

Continuity Correctionb

.527

1

.468

Likelihood Ratio

.988

1

.320

Fisher’s Exact Test

.450

.236

Linear-by-Linear Association

.948

1

.330

N of Valid Cases

97

a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 7.94.
b. Computed only for a 2×2 table

Table 18 Chi-Square/Fisher’s exact test for Q9 between girls and boys

Crosstab

Gender

Total

Female

Male

sexual_orientation No Count

65

18

83

% within sexual_orientation

78.3%

21.7%

100.0%

% within Gender

86.7%

81.8%

85.6%

Yes Count

10

4

14

% within sexual_orientation

71.4%

28.6%

100.0%

% within Gender

13.3%

18.2%

14.4%

Total Count

75

22

97

% within sexual_orientation

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

.324a

1

.569

Continuity Correctionb

.050

1

.823

Likelihood Ratio

.309

1

.578

Fisher’s Exact Test

.730

.395

Linear-by-Linear Association

.320

1

.571

N of Valid Cases

97

a. 1 cells (25.0%) have expected count less than 5. The minimum expected count is 3.18.
b. Computed only for a 2×2 table

Table 19 Chi-Square/Fisher’s exact test for Q9 between girls and boys

Crosstab

Gender

Total

Female

Male

IM_name No Count

59

13

72

% within IM_name

81.9%

18.1%

100.0%

% within Gender

78.7%

59.1%

74.2%

Yes Count

16

9

25

% within IM_name

64.0%

36.0%

100.0%

% within Gender

21.3%

40.9%

25.8%

Total Count

75

22

97

% within IM_name

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

3.407a

1

.065

Continuity Correctionb

2.461

1

.117

Likelihood Ratio

3.192

1

.074

Fisher’s Exact Test

.095

.061

Linear-by-Linear Association

3.372

1

.066

N of Valid Cases

97

a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 5.67.
b. Computed only for a 2×2 table

Table 20 Chi-Square/Fisher’s exact test for Q9 between girls and boys

Crosstab

Gender

Total

Female

Male

phone# No Count

74

20

94

% within phone#

78.7%

21.3%

100.0%

% within Gender

98.7%

90.9%

96.9%

Yes Count

1

2

3

% within phone#

33.3%

66.7%

100.0%

% within Gender

1.3%

9.1%

3.1%

Total Count

75

22

97

% within phone#

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

3.416a

1

.065

Continuity Correctionb

1.318

1

.251

Likelihood Ratio

2.737

1

.098

Fisher’s Exact Test

.128

.128

Linear-by-Linear Association

3.380

1

.066

N of Valid Cases

97

a. 2 cells (50.0%) have expected count less than 5. The minimum expected count is .68.
b. Computed only for a 2×2 table

Table 21 Chi-Square/Fisher’s exact test for Q9 between girls and boys

Crosstab

Gender

Total

Female

Male

your_website No Count

68

20

88

% within your_website

77.3%

22.7%

100.0%

% within Gender

90.7%

90.9%

90.7%

Yes Count

7

2

9

% within your_website

77.8%

22.2%

100.0%

% within Gender

9.3%

9.1%

9.3%

Total Count

75

22

97

% within your_website

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

.001a

1

.973

Continuity Correctionb

.000

1

1.000

Likelihood Ratio

.001

1

.972

Fisher’s Exact Test

1.000

.668

Linear-by-Linear Association

.001

1

.973

N of Valid Cases

97

a. 1 cells (25.0%) have expected count less than 5. The minimum expected count is 2.04.
b. Computed only for a 2×2 table

Table 22 Chi-Square/Fisher’s exact test for Q9 between girls and boys

Crosstab

Gender

Total

Female

Male

your_photos No Count

37

18

55

% within your_photos

67.3%

32.7%

100.0%

% within Gender

49.3%

81.8%

56.7%

Yes Count

38

4

42

% within your_photos

90.5%

9.5%

100.0%

% within Gender

50.7%

18.2%

43.3%

Total Count

75

22

97

% within your_photos

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

7.311a

1

.007

Continuity Correctionb

6.048

1

.014

Likelihood Ratio

7.902

1

.005

Fisher’s Exact Test

.007

.006

Linear-by-Linear Association

7.236

1

.007

N of Valid Cases

97

a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 9.53.
b. Computed only for a 2×2 table

Table 23 Chi-Square/Fisher’s exact test for Q9 between girls and boys

Crosstab

Gender

Total

Female

Male

your_videos No Count

38

5

43

% within your_videos

88.4%

11.6%

100.0%

% within Gender

50.7%

22.7%

44.3%

Yes Count

37

17

54

% within your_videos

68.5%

31.5%

100.0%

% within Gender

49.3%

77.3%

55.7%

Total Count

75

22

97

% within your_videos

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

5.381a

1

.020

Continuity Correctionb

4.308

1

.038

Likelihood Ratio

5.679

1

.017

Fisher’s Exact Test

.028

.017

Linear-by-Linear Association

5.325

1

.021

N of Valid Cases

97

a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 9.75.
b. Computed only for a 2×2 table

Table 24 Chi-Square/Fisher’s exact test for Q9 between girls and boys

Crosstab

Gender

Total

Female

Male

others No Count

63

15

78

% within others

80.8%

19.2%

100.0%

% within Gender

84.0%

68.2%

80.4%

Yes Count

12

7

19

% within others

63.2%

36.8%

100.0%

% within Gender

16.0%

31.8%

19.6%

Total Count

75

22

97

% within others

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

2.702a

1

.100

Continuity Correctionb

1.791

1

.181

Likelihood Ratio

2.487

1

.115

Fisher’s Exact Test

.128

.093

Linear-by-Linear Association

2.674

1

.102

N of Valid Cases

97

a. 1 cells (25.0%) have expected count less than 5. The minimum expected count is 4.31.
b. Computed only for a 2×2 table

Table 25 Chi-Square/Fisher’s exact test for Q9 between girls and boys

Crosstab

Gender

Total

Female

Male

Control_who_can_view_my__real_name No Count

55

17

72

% within Control_who_can_view_my__real_name

76.4%

23.6%

100.0%

% within Gender

73.3%

77.3%

74.2%

Yes Count

20

5

25

% within Control_who_can_view_my__real_name

80.0%

20.0%

100.0%

% within Gender

26.7%

22.7%

25.8%

Total Count

75

22

97

% within Control_who_can_view_my__real_name

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

.138a

1

.710

Continuity Correctionb

.009

1

.925

Likelihood Ratio

.141

1

.708

Fisher’s Exact Test

.789

.473

Linear-by-Linear Association

.137

1

.712

N of Valid Cases

97

a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 5.67.
b. Computed only for a 2×2 table

Table 26 Chi-Square/Fisher’s exact test for Q18 between girls and boys

Crosstab

Gender

Total

Female

Male

Control_who_can_view_my__DOB No Count

56

14

70

% within Control_who_can_view_my__DOB

80.0%

20.0%

100.0%

% within Gender

74.7%

63.6%

72.2%

Yes Count

19

8

27

% within Control_who_can_view_my__DOB

70.4%

29.6%

100.0%

% within Gender

25.3%

36.4%

27.8%

Total Count

75

22

97

% within Control_who_can_view_my__DOB

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

1.030a

1

.310

Continuity Correctionb

.554

1

.457

Likelihood Ratio

.993

1

.319

Fisher’s Exact Test

.417

.226

Linear-by-Linear Association

1.020

1

.313

N of Valid Cases

97

a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 6.12.
b. Computed only for a 2×2 table

Table 27 Chi-Square/Fisher’s exact test for Q18 between girls and boys

Crosstab

Gender

Total

Female

Male

Control_who_can_view_my__education No Count

55

19

74

% within Control_who_can_view_my__education

74.3%

25.7%

100.0%

% within Gender

73.3%

86.4%

76.3%

Yes Count

20

3

23

% within Control_who_can_view_my__education

87.0%

13.0%

100.0%

% within Gender

26.7%

13.6%

23.7%

Total Count

75

22

97

% within Control_who_can_view_my__education

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

1.597a

1

.206

Continuity Correctionb

.958

1

.328

Likelihood Ratio

1.747

1

.186

Fisher’s Exact Test

.263

.164

Linear-by-Linear Association

1.580

1

.209

N of Valid Cases

97

a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 5.22.
b. Computed only for a 2×2 table

Table 28 Chi-Square/Fisher’s exact test for Q18 between girls and boys

Crosstab

Gender

Total

Female

Male

Control_who_can_view_my__interests No Count

60

17

77

% within Control_who_can_view_my__interests

77.9%

22.1%

100.0%

% within Gender

80.0%

77.3%

79.4%

Yes Count

15

5

20

% within Control_who_can_view_my__interests

75.0%

25.0%

100.0%

% within Gender

20.0%

22.7%

20.6%

Total Count

75

22

97

% within Control_who_can_view_my__interests

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

.077a

1

.781

Continuity Correctionb

.000

1

1.000

Likelihood Ratio

.076

1

.783

Fisher’s Exact Test

.770

.495

Linear-by-Linear Association

.077

1

.782

N of Valid Cases

97

a. 1 cells (25.0%) have expected count less than 5. The minimum expected count is 4.54.
b. Computed only for a 2×2 table

Table 29 Chi-Square/Fisher’s exact test for Q18 between girls and boys

Crosstab

Gender

Total

Female

Male

Control_who_can_view_my__religion No Count

63

19

82

% within Control_who_can_view_my__religion

76.8%

23.2%

100.0%

% within Gender

84.0%

86.4%

84.5%

Yes Count

12

3

15

% within Control_who_can_view_my__religion

80.0%

20.0%

100.0%

% within Gender

16.0%

13.6%

15.5%

Total Count

75

22

97

% within Control_who_can_view_my__religion

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

.073a

1

.787

Continuity Correctionb

.000

1

1.000

Likelihood Ratio

.074

1

.785

Fisher’s Exact Test

1.000

.543

Linear-by-Linear Association

.072

1

.789

N of Valid Cases

97

a. 1 cells (25.0%) have expected count less than 5. The minimum expected count is 3.40.
b. Computed only for a 2×2 table

Table 30 Chi-Square/Fisher’s exact test for Q18 between girls and boys

Crosstab

Gender

Total

Female

Male

Control_who_can_view_my__status No Count

63

15

78

% within Control_who_can_view_my__status

80.8%

19.2%

100.0%

% within Gender

84.0%

68.2%

80.4%

Yes Count

12

7

19

% within Control_who_can_view_my__status

63.2%

36.8%

100.0%

% within Gender

16.0%

31.8%

19.6%

Total Count

75

22

97

% within Control_who_can_view_my__status

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

2.702a

1

.100

Continuity Correctionb

1.791

1

.181

Likelihood Ratio

2.487

1

.115

Fisher’s Exact Test

.128

.093

Linear-by-Linear Association

2.674

1

.102

N of Valid Cases

97

a. 1 cells (25.0%) have expected count less than 5. The minimum expected count is 4.31.
b. Computed only for a 2×2 table

Table 31 Chi-Square/Fisher’s exact test for Q18 between girls and boys

Crosstab

Gender

Total

Female

Male

Control_who_can_view_my__address No Count

55

19

74

% within Control_who_can_view_my__address

74.3%

25.7%

100.0%

% within Gender

73.3%

86.4%

76.3%

Yes Count

20

3

23

% within Control_who_can_view_my__address

87.0%

13.0%

100.0%

% within Gender

26.7%

13.6%

23.7%

Total Count

75

22

97

% within Control_who_can_view_my__address

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

1.597a

1

.206

Continuity Correctionb

.958

1

.328

Likelihood Ratio

1.747

1

.186

Fisher’s Exact Test

.263

.164

Linear-by-Linear Association

1.580

1

.209

N of Valid Cases

97

a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 5.22.
b. Computed only for a 2×2 table

Table 32 Chi-Square/Fisher’s exact test for Q18 between girls and boys

Crosstab

Gender

Total

Female

Male

Control_who_can_view_my__email No Count

54

15

69

% within Control_who_can_view_my__email

78.3%

21.7%

100.0%

% within Gender

72.0%

68.2%

71.1%

Yes Count

21

7

28

% within Control_who_can_view_my__email

75.0%

25.0%

100.0%

% within Gender

28.0%

31.8%

28.9%

Total Count

75

22

97

% within Control_who_can_view_my__email

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

.121a

1

.728

Continuity Correctionb

.006

1

.936

Likelihood Ratio

.119

1

.730

Fisher’s Exact Test

.791

.460

Linear-by-Linear Association

.120

1

.730

N of Valid Cases

97

a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 6.35.
b. Computed only for a 2×2 table

Table 33 Chi-Square/Fisher’s exact test for Q18 between girls and boys

Crosstab

Gender

Total

Female

Male

Control_who_can_view_my__sexual_orientation No Count

67

20

87

% within Control_who_can_view_my__sexual_orientation

77.0%

23.0%

100.0%

% within Gender

89.3%

90.9%

89.7%

Yes Count

8

2

10

% within Control_who_can_view_my__sexual_orientation

80.0%

20.0%

100.0%

% within Gender

10.7%

9.1%

10.3%

Total Count

75

22

97

% within Control_who_can_view_my__sexual_orientation

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

.046a

1

.831

Continuity Correctionb

.000

1

1.000

Likelihood Ratio

.047

1

.829

Fisher’s Exact Test

1.000

.595

Linear-by-Linear Association

.045

1

.832

N of Valid Cases

97

a. 1 cells (25.0%) have expected count less than 5. The minimum expected count is 2.27.
b. Computed only for a 2×2 table

Table 34 Chi-Square/Fisher’s exact test for Q18 between girls and boys

Crosstab

Gender

Total

Female

Male

Control_who_can_view_my__IM_name No Count

66

21

87

% within Control_who_can_view_my__IM_name

75.9%

24.1%

100.0%

% within Gender

88.0%

95.5%

89.7%

Yes Count

9

1

10

% within Control_who_can_view_my__IM_name

90.0%

10.0%

100.0%

% within Gender

12.0%

4.5%

10.3%

Total Count

75

22

97

% within Control_who_can_view_my__IM_name

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

1.022a

1

.312

Continuity Correctionb

.375

1

.540

Likelihood Ratio

1.200

1

.273

Fisher’s Exact Test

.447

.286

Linear-by-Linear Association

1.012

1

.314

N of Valid Cases

97

a. 1 cells (25.0%) have expected count less than 5. The minimum expected count is 2.27.
b. Computed only for a 2×2 table

Table 35 Chi-Square/Fisher’s exact test for Q18 between girls and boys

Crosstab

Gender

Total

Female

Male

Control_who_can_view_my__phone# No Count

51

21

72

% within Control_who_can_view_my__phone#

70.8%

29.2%

100.0%

% within Gender

68.0%

95.5%

74.2%

Yes Count

24

1

25

% within Control_who_can_view_my__phone#

96.0%

4.0%

100.0%

% within Gender

32.0%

4.5%

25.8%

Total Count

75

22

97

% within Control_who_can_view_my__phone#

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

6.702a

1

.010

Continuity Correctionb

5.344

1

.021

Likelihood Ratio

8.544

1

.003

Fisher’s Exact Test

.011

.006

Linear-by-Linear Association

6.633

1

.010

N of Valid Cases

97

a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 5.67.
b. Computed only for a 2×2 table

Table 36 Chi-Square/Fisher’s exact test for Q18 between girls and boys

Crosstab

Gender

Total

Female

Male

Control_who_can_view_my__your_website No Count

62

19

81

% within Control_who_can_view_my__your_website

76.5%

23.5%

100.0%

% within Gender

82.7%

86.4%

83.5%

Yes Count

13

3

16

% within Control_who_can_view_my__your_website

81.3%

18.8%

100.0%

% within Gender

17.3%

13.6%

16.5%

Total Count

75

22

97

% within Control_who_can_view_my__your_website

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

.169a

1

.681

Continuity Correctionb

.007

1

.933

Likelihood Ratio

.175

1

.676

Fisher’s Exact Test

1.000

.483

Linear-by-Linear Association

.167

1

.683

N of Valid Cases

97

a. 1 cells (25.0%) have expected count less than 5. The minimum expected count is 3.63.
b. Computed only for a 2×2 table

Table 37 Chi-Square/Fisher’s exact test for Q18 between girls and boys

Crosstab

Gender

Total

Female

Male

Control_who_can_view_my__your_photos No Count

37

8

45

% within Control_who_can_view_my__your_photos

82.2%

17.8%

100.0%

% within Gender

49.3%

36.4%

46.4%

Yes Count

38

14

52

% within Control_who_can_view_my__your_photos

73.1%

26.9%

100.0%

% within Gender

50.7%

63.6%

53.6%

Total Count

75

22

97

% within Control_who_can_view_my__your_photos

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

1.151a

1

.283

Continuity Correctionb

.688

1

.407

Likelihood Ratio

1.165

1

.280

Fisher’s Exact Test

.336

.204

Linear-by-Linear Association

1.139

1

.286

N of Valid Cases

97

a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 10.21.
b. Computed only for a 2×2 table

Table 38 Chi-Square/Fisher’s exact test for Q18 between girls and boys

Crosstab

Gender

Total

Female

Male

Control_who_can_view_my__your_videos No Count

51

12

63

% within Control_who_can_view_my__your_videos

81.0%

19.0%

100.0%

% within Gender

68.0%

54.5%

64.9%

Yes Count

24

10

34

% within Control_who_can_view_my__your_videos

70.6%

29.4%

100.0%

% within Gender

32.0%

45.5%

35.1%

Total Count

75

22

97

% within Control_who_can_view_my__your_videos

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

1.353a

1

.245

Continuity Correctionb

.826

1

.363

Likelihood Ratio

1.320

1

.251

Fisher’s Exact Test

.311

.181

Linear-by-Linear Association

1.339

1

.247

N of Valid Cases

97

a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 7.71.
b. Computed only for a 2×2 table

Table 39 Chi-Square/Fisher’s exact test for Q18 between girls and boys

Crosstab

Gender

Total

Female

Male

who_can_block_user No Count

59

19

78

% within who_can_block_user

75.6%

24.4%

100.0%

% within Gender

78.7%

86.4%

80.4%

Yes Count

16

3

19

% within who_can_block_user

84.2%

15.8%

100.0%

% within Gender

21.3%

13.6%

19.6%

Total Count

75

22

97

% within who_can_block_user

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

.640a

1

.424

Continuity Correctionb

.244

1

.621

Likelihood Ratio

.682

1

.409

Fisher’s Exact Test

.550

.321

Linear-by-Linear Association

.633

1

.426

N of Valid Cases

97

a. 1 cells (25.0%) have expected count less than 5. The minimum expected count is 4.31.
b. Computed only for a 2×2 table

Table 40 Chi-Square/Fisher’s exact test for Q18 between girls and boys

Crosstab

Gender

Total

Female

Male

block_application No Count

52

17

69

% within block_application

75.4%

24.6%

100.0%

% within Gender

69.3%

77.3%

71.1%

Yes Count

23

5

28

% within block_application

82.1%

17.9%

100.0%

% within Gender

30.7%

22.7%

28.9%

Total Count

75

22

97

% within block_application

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

.522a

1

.470

Continuity Correctionb

.207

1

.649

Likelihood Ratio

.540

1

.462

Fisher’s Exact Test

.596

.331

Linear-by-Linear Association

.517

1

.472

N of Valid Cases

97

a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 6.35.
b. Computed only for a 2×2 table

Table 41 Chi-Square/Fisher’s exact test for Q18 between girls and boys

Crosstab

Gender

Total

Female

Male

make_comments No Count

54

19

73

% within make_comments

74.0%

26.0%

100.0%

% within Gender

72.0%

86.4%

75.3%

Yes Count

21

3

24

% within make_comments

87.5%

12.5%

100.0%

% within Gender

28.0%

13.6%

24.7%

Total Count

75

22

97

% within make_comments

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

1.885a

1

.170

Continuity Correctionb

1.192

1

.275

Likelihood Ratio

2.072

1

.150

Fisher’s Exact Test

.261

.136

Linear-by-Linear Association

1.865

1

.172

N of Valid Cases

97

a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 5.44.
b. Computed only for a 2×2 table

Table 42 Chi-Square/Fisher’s exact test for Q18 between girls and boys

Crosstab

Gender

Total

Female

Male

send_message No Count

63

17

80

% within send_message

78.8%

21.3%

100.0%

% within Gender

84.0%

77.3%

82.5%

Yes Count

12

5

17

% within send_message

70.6%

29.4%

100.0%

% within Gender

16.0%

22.7%

17.5%

Total Count

75

22

97

% within send_message

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

.533a

1

.466

Continuity Correctionb

.169

1

.681

Likelihood Ratio

.508

1

.476

Fisher’s Exact Test

.526

.329

Linear-by-Linear Association

.527

1

.468

N of Valid Cases

97

a. 1 cells (25.0%) have expected count less than 5. The minimum expected count is 3.86.
b. Computed only for a 2×2 table

Table 43 Chi-Square/Fisher’s exact test for Q18 between girls and boys

Crosstab

Gender

Total

Female

Male

online_status No Count

59

18

77

% within online_status

76.6%

23.4%

100.0%

% within Gender

78.7%

81.8%

79.4%

Yes Count

16

4

20

% within online_status

80.0%

20.0%

100.0%

% within Gender

21.3%

18.2%

20.6%

Total Count

75

22

97

% within online_status

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

.103a

1

.748

Continuity Correctionb

.000

1

.983

Likelihood Ratio

.106

1

.745

Fisher’s Exact Test

1.000

.505

Linear-by-Linear Association

.102

1

.749

N of Valid Cases

97

a. 1 cells (25.0%) have expected count less than 5. The minimum expected count is 4.54.
b. Computed only for a 2×2 table

Table 44 Chi-Square/Fisher’s exact test for Q18 between girls and boys

Crosstab

Gender

Total

Female

Male

search No Count

67

20

87

% within search

77.0%

23.0%

100.0%

% within Gender

89.3%

90.9%

89.7%

Yes Count

8

2

10

% within search

80.0%

20.0%

100.0%

% within Gender

10.7%

9.1%

10.3%

Total Count

75

22

97

% within search

77.3%

22.7%

100.0%

% within Gender

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

.046a

1

.831

Continuity Correctionb

.000

1

1.000

Likelihood Ratio

.047

1

.829

Fisher’s Exact Test

1.000

.595

Linear-by-Linear Association

.045

1

.832

N of Valid Cases

97

a. 1 cells (25.0%) have expected count less than 5. The minimum expected count is 2.27.
b. Computed only for a 2×2 table

Table 45 Chi-Square/Fisher’s exact test for Q18 between girls and boys

Group Statistics

Gender

N

Mean

Std. Deviation

Std. Error Mean

Agree Female

75

1.92

.850

.098

Male

22

1.64

.581

.124

Independent Samples Test

Levene’s Test for Equality of Variances

t-test for Equality of Means

95% Confidence Interval of the Difference

F

Sig.

t

df

Sig. (2-tailed)

Mean Difference

Std. Error Difference

Lower

Upper

Agree Equal variances assumed

.525

.470

1.465

95

.146

.284

.194

-.101

.668

Equal variances not assumed

1.794

50.069

.079

.284

.158

-.034

.601

Table 46 T-test for Q19 between girls and boys

Group Statistics

Gender

N

Mean

Std. Deviation

Std. Error Mean

Agree Female

75

1.80

.735

.085

Male

22

1.68

.716

.153

Independent Samples Test

Levene’s Test for Equality of Variances

t-test for Equality of Means

95% Confidence Interval of the Difference

F

Sig.

t

df

Sig. (2-tailed)

Mean Difference

Std. Error Difference

Lower

Upper

Agree Equal variances assumed

.000

.990

.667

95

.507

.118

.177

-.234

.470

Equal variances not assumed

.676

35.038

.503

.118

.175

-.236

.473

Table 47 T-test for Q20 between girls and boys

Group Statistics

Gender

N

Mean

Std. Deviation

Std. Error Mean

Agree Female

73

2.16

1.106

.129

Male

21

2.29

1.007

.220

Independent Samples Test

Levene’s Test for Equality of Variances

t-test for Equality of Means

95% Confidence Interval of the Difference

F

Sig.

t

df

Sig. (2-tailed)

Mean Difference

Std. Error Difference

Lower

Upper

Agree Equal variances assumed

.000

.988

-.452

92

.653

-.121

.269

-.655

.412

Equal variances not assumed

-.476

35.103

.637

-.121

.255

-.639

.396

Table 48 T-test for Q21 between girls and boys

Group Statistics

Gender

N

Mean

Std. Deviation

Std. Error Mean

Agree Female

74

1.76

.791

.092

Male

22

1.64

.727

.155

Independent Samples Test

Levene’s Test for Equality of Variances

t-test for Equality of Means

95% Confidence Interval of the Difference

F

Sig.

t

df

Sig. (2-tailed)

Mean Difference

Std. Error Difference

Lower

Upper

Agree Equal variances assumed

.168

.682

.638

94

.525

.120

.189

-.254

.495

Equal variances not assumed

.668

37.055

.508

.120

.180

-.245

.485

Table 49 T-test for Q22 between girls and boys

Group Statistics

Gender

N

Mean

Std. Deviation

Std. Error Mean

Agree Female

74

1.93

1.051

.122

Male

22

1.68

.716

.153

Independent Samples Test

Levene’s Test for Equality of Variances

t-test for Equality of Means

95% Confidence Interval of the Difference

F

Sig.

t

df

Sig. (2-tailed)

Mean Difference

Std. Error Difference

Lower

Upper

Agree Equal variances assumed

.254

.615

1.046

94

.298

.251

.239

-.225

.726

Equal variances not assumed

1.281

50.545

.206

.251

.196

-.142

.643

Table 50 T-test for Q23 between girls and boys

Group Statistics

Gender

N

Mean

Std. Deviation

Std. Error Mean

Agree Female

74

1.86

.799

.093

Male

22

1.59

.590

.126

Independent Samples Test

Levene’s Test for Equality of Variances

t-test for Equality of Means

95% Confidence Interval of the Difference

F

Sig.

t

df

Sig. (2-tailed)

Mean Difference

Std. Error Difference

Lower

Upper

Agree Equal variances assumed

.159

.691

1.489

94

.140

.274

.184

-.091

.639

Equal variances not assumed

1.751

46.189

.087

.274

.156

-.041

.589

Table 51 T-test for Q24 between girls and boys

Group Statistics

Gender

N

Mean

Std. Deviation

Std. Error Mean

Agree Female

75

2.32

1.002

.116

Male

22

2.18

1.006

.215

Independent Samples Test

Levene’s Test for Equality of Variances

t-test for Equality of Means

95% Confidence Interval of the Difference

F

Sig.

t

df

Sig. (2-tailed)

Mean Difference

Std. Error Difference

Lower

Upper

Agree Equal variances assumed

.006

.938

.568

95

.571

.138

.243

-.345

.621

Equal variances not assumed

.567

34.171

.575

.138

.244

-.357

.634

Table 52 T-test for Q25 between girls and boys

Crosstab

Group

Total

Adults

Teenagers

Real_name No Count

20

12

32

% within Real_name

62.5%

37.5%

100.0%

% within Group

21.3%

12.4%

16.8%

Yes Count

74

85

159

% within Real_name

46.5%

53.5%

100.0%

% within Group

78.7%

87.6%

83.2%

Total Count

94

97

191

% within Real_name

49.2%

50.8%

100.0%

% within Group

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

2.715a

1

.099

Continuity Correctionb

2.114

1

.146

Likelihood Ratio

2.736

1

.098

Fisher’s Exact Test

.122

.073

Linear-by-Linear Association

2.700

1

.100

N of Valid Cases

191

a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 15.75.
b. Computed only for a 2×2 table

Table 53 Chi-Square/Fisher’s exact test for Q9 between adults and teenagers

Crosstab

Group

Total

Adults

Teenagers

DOB No Count

23

22

45

% within DOB

51.1%

48.9%

100.0%

% within Group

24.5%

22.7%

23.6%

Yes Count

71

75

146

% within DOB

48.6%

51.4%

100.0%

% within Group

75.5%

77.3%

76.4%

Total Count

94

97

191

% within DOB

49.2%

50.8%

100.0%

% within Group

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

.085a

1

.771

Continuity Correctionb

.015

1

.904

Likelihood Ratio

.085

1

.771

Fisher’s Exact Test

.865

.452

Linear-by-Linear Association

.084

1

.772

N of Valid Cases

191

a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 22.15.
b. Computed only for a 2×2 table

Table 54 Chi-Square/Fisher’s exact test for Q9 between adults and teenagers

Crosstab

Group

Total

Adults

Teenagers

education No Count

37

36

73

% within education

50.7%

49.3%

100.0%

% within Group

39.4%

37.1%

38.2%

Yes Count

57

61

118

% within education

48.3%

51.7%

100.0%

% within Group

60.6%

62.9%

61.8%

Total Count

94

97

191

% within education

49.2%

50.8%

100.0%

% within Group

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

.102a

1

.749

Continuity Correctionb

.029

1

.864

Likelihood Ratio

.102

1

.749

Fisher’s Exact Test

.768

.432

Linear-by-Linear Association

.102

1

.750

N of Valid Cases

191

a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 35.93.
b. Computed only for a 2×2 table

Table 55 Chi-Square/Fisher’s exact test for Q9 between adults and teenagers

Crosstab

Group

Total

Adults

Teenagers

interests No Count

36

63

99

% within interests

36.4%

63.6%

100.0%

% within Group

38.3%

64.9%

51.8%

Yes Count

58

34

92

% within interests

63.0%

37.0%

100.0%

% within Group

61.7%

35.1%

48.2%

Total Count

94

97

191

% within interests

49.2%

50.8%

100.0%

% within Group

100.0%

100.0%

100.0%

Chi-Square Tests

Value

df

Asymp. Sig. (2-sided)

Exact Sig. (2-sided)

Exact Sig. (1-sided)

Pearson Chi-Square

13.581a

1

.000

Continuity Correctionb

12.534

1

.000

Likelihood Ratio

13.745

1

.000

Fisher’s Exact Test

.000

.000

Linear-by-Linear Association

13.510

1

.000

N of Valid Cases

191