International Journal of Virtual Worlds and Human Computer Interaction (VWHCI)

ISSN: 2368-6103

1. Introduction

In recent years, social media has become widely used around the world. Anywhere, and at any time, social media users can post in various forms of expression, including text, audio, images and moving images, using mobile devices. As in other industrialised countries, many people in Japan use social media, and LINE, Twitter and Facebook are popular especially amongst the younger generations. However, it is well known that social media companies collect and store as much personal data as possible, which they utilise and share for various business purposes. Personal data including user names, e-mail addresses, phone numbers, dates of birth, genders and affiliations are often required for social media user registration. Social media users appear to voluntarily provide personally identifiable information to the companies. In addition, social media sites are designed to encourage active communication between users, prompting disclosure of personal data at every opportunity, as these data are crucial for the profitability of social media companies. Data analyses on users' habits of online activities and user-to-user relationships are conducted by them utilising user personal data such as search keywords, posts, names/handles of friends/followers, photos with name tags and communication metadata, which are automatically collected and stored in their database in a nearly real-time fashion.

Personal data obtained from a social media site are utilised by its operating company for executing their business models (Krombholz et al. [10]). For example, social media companies sell advertising placements on their sites and/or provide advertising platform services to their customer organisations. Social media advertisements are then personalised based on user preferences or behavioural patterns online. Profiling data is derived from an analysis of user personal data stored in the social media company databases. Social media companies may also sell users' personal data to other organisations. These data are usually anonymised and/or aggregated to avoid prosecution for violation of personal data protection laws. Buyers then use these data to analyse their (potential) customers' preferences, real opinions and consumption propensities, and the results of the analysis may be sold to third-party businesses. Various kinds of personal data provided by social media users are collected, stored, processed and distributed by social media companies, and those data are now considered a crucial component of 'big data' analysis and a precious resource for streamlining the business operations of many companies.

On the other hand, typical Japanese social media users seem unaware of how personal data they post online are utilised by social media companies and other organisations, although it has been repeatedly alleged that societal awareness of privacy and personal data protection has been enhanced in Japan (ECOM [4]; JIPDEC [8]; Public Relations Department of Cabinet Office [20]; Quality of Life Council [21]). Orito et al. [19] and Murata et al. [15] investigate privacy awareness of Japanese online shoppers and find that their concerns about privacy invasion and personal data misuse are centred on financial and/or physical damage caused by, for example, misuse of credit card data or cyberstalking.

Therefore, how do the users of "free" online services such as social media recognise the privacy issues related to their service usage? It is not clear whether even the users who are cautious about personal data handling by social media companies recognise the full extent of their (and other companies') use and circulation of personal data those users provide on social media sites. However, it is plausible that knowledge of personal data handling by social media companies could influence users' disclosure of their and others' personal data online.

With these recognitions, this study attempts to investigate the attitudes of Japanese social media users with regard to their personal data provision on social media sites. A considerable number of studies have examined social media users' awareness of privacy issues and/or their attitudes towards personal data disclosure on social media (e.g. Beye et al. [2]; Krasnova et al. [9]; Madden and Smith [12]). In particular, a lot of surveys on Facebook users' attitudes and behaviour have been conducted (e.g. Acquisti and Gross [1]; Debatin et al. [3]; Govani and Pashley [5]; Gross and Acquisti [6]; Krombholz et al. [10]; Tow et al. [22]; Stutzman et al. [23]). Hoy and Milne [7], among others, investigate young-adult Facebook users' privacy beliefs and their concern about third party data usage beyond the original purpose and behavioural advertisement techniques from a perspective of gender differences. O'Brien and Torres [16] also discuss Facebook users' privacy concerns exploring the role of trust in a social networking environment and third parties' access to Facebook user information. They find out that "Facebook users seem willing to push aside trust issues to achieve social interaction" (p. 93) although they tend to have low levels of trust in the social networking environment.

However, as far as we are aware, no research has examined Japanese social media users' understanding of how social media companies and their related third party companies utilise personal data that users post on social media sites such as Facebook and Twitter. It is difficult to confirm that Japanese social media users have similar beliefs or attitudes to those in other countries in terms of privacy issues involved with their social media usage, or they understand the current situations generated by the operation of business models social media companies deploy. To make up for the lack of research, this study conducts an exploratory survey on Japanese social media users' attitudes towards online privacy and recognition of personal data handling by social media companies and their business models. To attain the research purpose, a questionnaire survey and follow-up interviews were conducted in November and December 2013, respectively. The study also discusses the social responsibility of social media companies in terms of personal data handling, taking the results of the authors' related studies (Murata et al. [15]; Orito et al. [19]) into account.

2. The Overview of the Survey

The questionnaire survey was conducted in November 2013 using an online survey site. The respondents were university students from the School of Commerce at Meiji University in Tokyo and from the Faculty of Law and Letters at Ehime University in Matsuyama. A total of 368 valid respondents (110 at Meiji University, 258 at Ehime University) were included in the study, and there was no significant difference between the two sample groups. There are several reasons why university students were selected as respondents to this survey. One of the most important reasons was their experience of social media or social networking services (SNS). Because the major social media such as Facebook and Twitter have been used in Japan since the late 2000s, it was expected that the respondents to the survey (university students aged 18-23 years) had gained their experiences of social media usage during their lower and upper secondary school years. In addition, because this survey was conducted as part of the coursework at each university, the students could be expected to take it seriously.

The questionnaire's title was "Questionnaire about Social Media Usage" to avoid priming, and the respondents had the option of providing their real names to express their consent for accepting follow-up interviews. Respondent attributes and their social media usage are shown in Table 1. The complete questionnaire is provided in the Appendix.

Respondents who used social media less than three times per week (Q1) and those who did not have any particular social media they usually used (Q2) were excluded from the analysis. According to the results of the survey, over 70% of respondents always connect with social media or connect with social media more than three times per day (Q1), and almost all respondents are using smart phone to access social media (Q1-2). Facebook, Twitter and LINE are the major social media the respondents usually use (Q2). Over a half of respondents post articles, photos and video and over 60% of the respondent view posts by other users and click "Like" or repost (Q5). The questionnaire response tendencies and relationships between the responses to questions were examined through statistical tests including Pearson's chi-squared test and Fisher's exact test. After the online survey, 14 respondents at Meiji University were interviewed in December 2013.

3. Survey Results

3.1. Recognition of Privacy Protection and Privacy Policies

According to the survey results, whereas more than 90% of respondents recognised the importance of privacy protection when they used social media (Table 2), a majority of respondents were unaware of the right to privacy, and nearly a half of respondents could not describe why protection of the right to privacy was important (Table 3). In addition, more than 70% of the respondents who used Facebook and also of those who used Twitter hadn't read the terms of service and/or privacy policies of the social media sites (Table 4).

These findings are similar to the results of previous studies on young Japanese online shoppers' attitudes towards online privacy the authors engaged in ( Murata et al. [15]; Orito et al. [19]). Many of the respondents did not regard the terms of service and privacy policies of social media as a good source of information on privacy protection or invasion. Even those respondents who had read the terms of service and/or privacy policies might not read them with caution. In fact, several interviewees stated that they had read only a part of the terms or policies concerning, for example, charging and cancellation or termination of their membership.

3.2. Personal Data Protection Behaviour on Social Media Sites

Whereas many respondents de-emphasised the importance of terms of service and privacy policies, the majority of respondents seemed to take care in the disclosure of their personal data on social media sites. As shown in Table 5, 56.7% of respondents published their true names, but more than 80% of respondents refrained from revealing personal data unnecessarily on social media sites. On the other hand, 36.4% of respondents used handles (pseudonyms) and tried to control the disclosure of their personal data. In addition, more than 60% of respondents had changed their privacy settings to limit accessibility to the personal data and/or articles they posted on social media sites (Table 6), more than 80% of respondents did not post information with location data, and nearly 80% didn't post photos with tags of their and/or others' names (Table 7).

The results of the tests of independence between the responses to Q9 and Q10 (Table 8) and between those to Q9 and Q11 (Table 9) statistically confirmed that, for both Facebook and Twitter users, the frequency distribution of responses to Q10 and Q11 was not affected by those to Q9 (chi-squared (4) = 4.792, p > 0.1 in the relationship between the responses to Q9-1 and Q10; chi-squared (4) = 5.339, p > 0.1 in the relationship between the responses to Q9-2 and Q10; chi-squared (3) = 6.481, p > 0.05 in the relationship between the responses to Q9-1 and Q11; and chi-squared (3) = 5.691, p > 0.1 in the relationship between the responses to Q9-2 and Q11). The majority of respondents seemed to consider personal data protection when they used social media sites, though they tended not to read the terms of service or privacy policies.

3.3. Awareness of the Secondary/Tertiary Usage of Personal Data

The survey outcome that a significant number of respondents were cautious about personal data protection on social media sites did not signify their full understanding of the use of their personal data by social media companies. While 33.0% (99/282) of respondents presumed that the personal data from posted information would be processed for secondary/tertiary use by social media companies or other companies (Q16), 95.7% of respondents (266/278) presumed that the information they themselves had posted on social media sites had not been put into secondary/tertiary use (Q17).

Table 10 is the cross tabulation between Q9-1 (Facebook)/Q9-2 (Twitter) and Q16. Similar outcomes to those described in the above section were derived from the statistical analysis on this table. As shown in it, approximately one third of respondents assumed secondary/tertiary use of personal data by social media companies or other companies (31.7% of Facebook users and 31.3% of Twitter users). However, respondent recognition of the secondary/tertiary usage of their personal data did not significantly influence their behaviour as to whether they read the terms of service and/or privacy policies on social media sites (Q9-1. Facebook: chi-squared (1) = 0.037, p > 0.1; Q9-2. Twitter: chi-squared (1) = 2.544, p > 0.1).

The following two tables describe the relationships between respondent recognition of corporate secondary/tertiary use of personal data and behavioural dimensions of social media usage. Chi-squared analysis on Table 11 revealed that respondents' recognition of secondary/tertiary use of personal data did not significantly influence their disclosure of personal data (chi-squared (4) = 3.268, p > 0.1). In Table 12, the relationship between respondents' recognition of secondary/tertiary use and their privacy settings is described. There was no statistically significant relationship between the two variables (chi-squared (3) = 6.245, p > 0.1). Respondents' recognition of secondary/tertiary use of personal data did not influence them to modify the privacy settings to control their personal data disclosure.

There was also no statistically significant relationship between respondents' recognition of social media-posted personal data utilisation by the social media companies (Q22) and attitudes towards personal data disclosure (Table 13; chi-squared (4) = 2.942, p > 0.1), and between the recognition and respondents' modification of privacy settings (Table 14; chi-squared (1) = 2.773, p > 0.1). If social media users assume that companies utilise their personal data for profit, it is plausible that they tend to adopt a cautious attitude towards posting personal data, or to be motivated to modify their privacy settings. However, the survey results revealed that respondents' recognition of personal data handling by social media companies had no influence over the attitudes of respondents towards personal data disclosure and privacy settings.

3.4. Recognition of Social Media Business Models and the User Risks Involved

Many of respondents did not have a clear understanding of social media business models and the user risks involved. Only 4.3% of respondents (12/278) recognised that they had experienced secondary/tertiary use of information they posted on social media (Q17). However, it is a fact that almost all user-provided personal data are utilised by the social media companies and/or their related organisations as a precious resource central to their business models. In addition, many of the responses to the open questions "How do the social media companies and/or other companies utilise the personal data posted on social media sites for secondary/tertiary purposes?" (Q16-1) and "How do social media companies use your posts on the social media sites?" (Q22-1) were ambiguous, suggesting a lack of respondents' understanding.

One interviewee acknowledged that it was difficult for her to imagine the risks of personal data provision on social media because she had not suffered at all as a result of her personal data disclosure online. Several respondents feared misuse of personal data by individual users, rather than business organisations. Thus, they did not want other users to know details of their personal life. Conversely, one interviewee stated that it was personally beneficial to reveal factual personal data on the social media sites to enhance the quality of the online personalised services provided.

In addition, large differences were observed in terms of recognition of social media company accessibility to personal data. Almost 70% of respondents (188/280) thought their posts would be by social media companies over time, with deletion at some future date (Table 15). However, there was a range of their expectations of the durations (Table 16). Thus, many respondents may not fully comprehend the social media business models and the user risks involved.

4. Implications of the Survey Results

The perceptual and behavioural dimensions of social media usage were investigated through statistical analysis of the survey results, and the relationships between them were confirmed. Many respondents recognised the importance of privacy protection, but their perception was not supported by knowledge of the right to privacy, or by thoughtful judgment regarding privacy protection. The inconsistent personal data protection behaviour of respondents was also confirmed statistically. In particular, there was no statistically significant relationship detected between the three behavioural dimensions: whether or not respondents read privacy policies, how they attempted to properly control personal data provision on social media sites, and how they modified privacy settings on social media sites. Moreover, there was no significant relationship between the behavioural dimensions and respondent perception of secondary/tertiary use of personal data. Considering that over 90% of respondents recognised the importance of privacy protection, the authors expected that most of those who acknowledged secondary/tertiary use of their personal data would take appropriate action to protect their online privacy. However, this was not the case. Although almost all of respondents recognised the importance of privacy protection and a majority were concerned about the use and protection of online personal data by social media companies and other companies, such recognition and concern did not motivate them to understand how their personal data posted online were (mis)used or (un)protected.

One plausible reason for these could be a lack of transparency regarding the business models adopted by social media companies. Social media are seemingly free-of-charge online services, and typical users may not fully comprehend the cost of usage and how social media companies profit from users' personal data. The outcomes of the follow-up interviews support this hypothesis. Users' lack of knowledge regarding the business models could result in underestimation of the risks of using social media. Another plausible reason could be the failure of privacy education in Japan. Japanese people are educated about the importance of protecting the right to privacy at both lower and upper secondary schools. However, the education emphasises, without clearly defining privacy and the right to it, that the protection of the right is necessary to keep the serenity of their daily lives and to protect their property, and the relationships between privacy and freedom, let alone privacy and democracy, are usually not mentioned at all in the educational environment where teaching staff are reluctant to deal with political subjects and current affairs in the classrooms. Such privacy education could make Japanese people underestimate the risks of invasions of their privacy made by social media companies and their related organisations. This rationale seems to be supported by the outcomes of the questionnaire survey and the follow-up interviews.

In addition, many of the respondents appeared to prefer online socialisation to privacy protection, even though they felt vaguely insecure about their privacy when using social media. Or, they might attempt to submit to the belief that infringement of their right to privacy is none for their business. Actually, many of the interviewees told that misuse of personal data or an invasion of privacy would never happen to them. The similar tendencies were observed in the results of the authors' previous surveys (Murata et al. [15]; Orito et al. [19]). Japanese social media users would continue to use social media site regardless of how they consider social media companies reap the benefit from their personal data, and social media would be able to survive users' privacy concerns.

It is difficult for anyone to completely erase personal data posted on the Internet. Once personal comments, photos and videos are uploaded on a social media site, those personal data are collected and stored by social media companies. The stored data may be used by them, their related organisations and/or individual users ignoring the context in which the personal data were uploaded. Users are required to understand the importance of careful personal data handling on social media sites. On the other hand, social media companies should provide users with comprehensive and clear explanations about how they collect, store, use, share and sell personal data. Considering the significant influence of social media over users' mental status (O'Keeffe et al. [17]), identity (Marwick and boyd [13]), autonomy (Murata and Orito [14]) and intellectual freedom (Orito [18]), social media companies should fulfil their social responsibility and accountability.

5. Conclusion

This study examined social media user recognition of personal data handling by social media and other companies, and its influence on user behaviour towards disclosure of personal data and privacy protection online. The survey results revealed that the respondents use social media as they like regardless of their recognition of social media companies' and other companies' ways of using personal data they provide on social media sites. In the meantime, because of the exploratory property of this research, the evidence supporting our conclusion was limited. The lack of consistency between the perceptual and behavioural dimensions of social media users was statistically confirmed by the results of chi-square tests that indicated there was no significant relationship between these variables. However, because these variables were measured by binary or qualitative scales, our choice of method of analysis was very limited. More positive and multi-dimensional evidence is needed to reinforce our findings. In particular, it is necessary to measure the perceptual and behavioural dimensions of the social media users quantitatively and analyse them via the multi-factor ANOVA and its post-hoc analyses in order to explore the complex linkages between and among these variables in detail.

The tremendous advances in big data technology have rendered social media business models unintelligible to typical users. There is also a lack of user understanding of the commercial value and sensitivity of personal data. Personal data should appropriately be protected in the current socio-economic and technological environment, as otherwise the potential danger of violations of human dignity and rights would be created. Ensuring the transparency of social media business models and establishing a proper privacy education system should be recognised as urgent social issues in Japan.

Acknowledgements

This study was supported by the MEXT (Ministry of Education, Culture, Sports, Science and Technology, Japan) Programme for Strategic Research Bases at Private Universities (2012-16) project "Organisational Information Ethics" S1291006, the MEXT Grant-in-Aid for Scientific Research (B) 25285124, and the MEXT Research Grant-in-Aid for Young Scientists (B) 24730320.

References