Culture, corruption, suicide, happiness and global social media use: A cross-cultural perspective

Culture, corruption, suicide, happiness and global social media use: A cross-cultural perspective

Abstract: This study was conducted to answer this simple question: ‘Can cultural values explain global social media use?’ Along with cultural dimensions introduced by past studies we also added several demographic, socio-economic and personality variables into this study that generated quite interesting findings. We found that there are low levels of suicide, more happiness and more corruption in societies that use social media heavily. We also observed that GDP per capita and median age are negatively related with social media use. Self-esteem stood out as important variable related to social media use intensity along with emotional expressiveness, openness and conscientiousness. Contrary to the common view, nation-level social capital achievement was negatively related with social media use and there was absolutely no relationship between high-context and low-context communication characteristics and local social media use. Some other findings also indicated that conservative and collectivistic countries use social media more often than do individualistic and developed countries. Schwartz’s cultural dimensions and the results of the GLOBE study accounted for a considerable amount of variation in country-level social media use where Hofstede and Trompenaars’ cultural dimensions were insignificant. Since most of the cultural values failed to explain the intensity of social media use, we also developed a cross-cultural online communication framework called cross-cultural self and others’ worth.

Culture Suicide Social Media Happiness

Keywords: culture; corruption; suicide; happiness; privacy; cross-cultural; Japan; social media; Eastern; Facebook.

Reference to this paper should be made as follows: Acar, A. (xxxx) ‘Culture, corruption, suicide, happiness and global social media use: a cross-cultural perspective’, Int. J. Web Based Communities, Vol. X, No. Y, pp.000–000.

Biographical notes: Adam Acar is an Associate Professor at Kobe City University of Foreign Studies where he teaches Global Business, Marketing and Communications.

1 Introduction

Studies show that more than half the population in developed countries has experienced online social networking and a quarter of the world citizens now have a profile in social media, whose users number 1.47 billion (eMarketer, 2013). Although English-speaking countries top the list of active social media-using nations, people from all around the world are represented on online social network channels (Leetaru et al., 2013). Social

Copyright © 200x Inderscience Enterprises Ltd.

9-1, Gakuen-higashi-machi,

Nishi-ku, Kobe 651-2187, Japan

2 A. Acar

media obviously is a global phenomenon; however, we do not know much about how each nation uses this tool and whether cultural values and demographic factors impact the usage behaviour (Acar and Deguchi, 2013). Even though many scholars indicated the need for additional studies on this topic (Acar and Deguchi 2013; Acar et al., 2013), currently none of the top twenty academic papers on Google Scholar and the top twenty books on Amazon in the social media category has anything to do with culture. On the other hand, this may also be the result of the fact that popular social media platforms such as Facebook, Twitter, Google+, and LinkedIn do not localise their interfaces, processes, and formats. People naturally may consider culture to be irrelevant, as social media users from different countries are getting along on these mediums and happily using the platforms, contrary to past studies that suggested culture impacts online interface preferences and online behavioural tendencies (Hermeking, 2005).

I believe there is a specific reason to explore the relationship between culture and social media because culture itself is related to sociality and socialisation. A number of well-known scholars suggested that the most important dimension of culture is the relationship between individuals and society. Hofstede (2001) focused on individualism versus collectivism (whether an individual’s priority is self-achievement or group achievement); Trompenaars and Hampden-Turner (1998) similarly drew on communitarianism versus individualism (whether people socialise as a group or as individuals); Schwartz (1999) pointed out embeddedness versus autonomy (whether people’s lives are influenced more by social relationships or individual pursuits); and Markus and Kitayama (1991) talked about how certain cultures promoted independent-self versus interdependent-self (whether the concept of self depends on one’s own judgment or reactions from others). This particular aspect (socialisation) of culture explains many things, including why people in some cultures stand very close to each other when talking, why people in some cultures call their family members every day, and why in some cultures people take many risks by thinking that their network members will take care of them (Hofstede, 2001). Since social media, by definition, is about building and maintaining relationships with one’s network members and influenced by collective self-esteem and the need to belong (Acar, 2008), we must clarify what role culture plays in what people do in social media and how intensely they use this particular communication tool.

2 Literature review

Since this is an exploratory study that includes more than 35 variables, it is difficult to run a rigorous literature review in this limited space. The following paragraphs give a glimpse of the relationship between social media and culture and brief definitions of the concepts discussed in this paper.

Culture, corruption, suicide, happiness and global social media use 3 Table 1 A summary of past studies about culture and social media

Individualistic people are more concerned about pictures they are tagged on Facebook then collectivistic people.

People from individualistic societies post more self-promoting content, more photos and are more likely to have online friends that they have not met face to face.

Collectivistic people are more likely to post tweets addressed to someone or mentions someone (@user).

People from societies with a higher pace-of-life post tweets in a more predictable frequency and pattern than those from countries with low pace-of-life scores.

American college students are more likely to post obscene content in social media compared to German college students.

Americans are more concerned about online privacy than Chinese and Indians.

Compared to American social media users, Chinese users care less about being popular online and less likely to post self-promoting content. Chinese internet users spend less amount of time for social media than Americans.

Chinese brand messages that are shared in social media usually focus on popularity, symbolism, and social status while American brand messages use the appeals of individualism and hedonic consumption.

RenRen users in China are more likely to digitally customise their profile pictures compared to American Facebook users.

Americans are more likely to post Facebook messages that are about themselves and what is happening in their immediate environment versus Indians who post about interests, ideas, religions etc.

Tokyoites tend to post pictures on Instagram that have red-yellow tones versus New Yorkers who post pictures with blue-grey tones.

Facebook photos posted in Namibia predominantly focus on individuals and show less of any background and object.

Americans in general do not have any problem posting their own pictures in social media while one third of Indians abstain to do so.

Middle Eastern students are less likely to show their faces in their profile photos compared to American college students.

The most positive tweets in the world are posted by Brazilians and (as a result of the country’s collectivistic nature) people in collectivistic countries (e.g., Indonesia) are more densely connected on Twitter than people from individualistic countries (e.g., Australia).

Online event schedule platform indicated that Colombians in general schedule events 12 days before the events while Germans usually schedule events 28 days prior to the actual activity.

A huge majority (87%) of Norwegian brands chose to answer questions asked to them on Facebook while this ratio was only 45% brands in the UK.

Rui and Stefanone (2012)

Rosen et al. (2010)

Garcìa-Gavilanes et al. (2013)

Poblete et al. (2011)

Peluchette et al. (2010)

Wang et al. (2011)

Jackson and Wang (2013)

Tsai and Men (2012)

Zhao and Jiang (2011)

Sharrock (2013)

Hochman and Schwarts (2012)

Peters et al. (2012)

Marshall et al. (2008)

Auter and Elmasry (2012)

Poblete et al. (2011)

Reinecke et al. (2013)

Eve (2012)

4 A. Acar
Table 1 A summary of past studies about culture and social media (continued)

Turkish people are the most active social media users (93% report sharing something online in the past month) and Japanese have the lowest engagement rate, (70% of Japanese report not posting anything in social media in the past month.

Compared to the social media activities of American top 100 brands, Japanese top 100 brands ask fewer questions, post less frequently, do not address their fans directly, do not initiate conversations with their fans, reveal less info on their profiles, and do not allow fans to post on their walls.

Japanese college students tweet more about themselves and TV and less about news, sports and family compared to American college students

25% of Japanese Twitter accounts are protected (tweets are not public) while this ratio is lower than 7% all around the world.

28% of Japanese Facebook users did not show their faces in their profile photos while only 8% of Americans chose to do so.

For Japanese college students the most popular Facebook content category to ‘like’ is ‘landscape/scenery’ unlike international students who chose the ‘funny/humorous’ as their most favourite.

Japanese Facebook users are less likely to share a picture of their partners and their families than do American college students. On the other hand, they are more positive about friending their professors on Facebook.

Japanese Facebook users are about ten times less likely to share a content they liked on Facebook compared to Americans and ten times less likely to comment on a content they liked on Facebook compared to Germans.

IPSOS (2013)

Acar et al. (2013)

Acar and Deguchi (2013)

Acar (in press)

Despite the fact that there are more than two dozen studies on culture and social media, so far no study has investigated culture and social media usage intensity. Therefore, we decided to assess how cultural dimensions and widely accepted cultural values influence how intensely people use social media platforms.

2.1 High-context and low-context communication

Edward Hall, who introduced the concepts of high-context and low-context communication, made it very clear that it is just a continuum and countries do not have index scores ranking their place on the context continuum with reference to other nations. He also indicated that we should be careful using these high-low context labels for large countries like the USA, because obviously there would be many different ethnicities within the same country that communicate differently (Hall and Hall, 1990). At the same time, he implied that English-speaking, German-speaking, and Scandinavian countries are usually low-context nations and that Japan, China, and the Arab and Mediterranean countries tend to be high-context cultures. Based on this, we coded the USA, UK, Australia, New Zealand, Germany, Austria, the Netherlands, Canada, Norway, Sweden, Denmark, and Switzerland as low-context and China, Taiwan, Hong Kong, Japan, Saudi Arabia, the United Arab Emirates, Spain, Italy, Brazil, and Mexico as high-context countries.

Culture, corruption, suicide, happiness and global social media use 5 2.2 Schwartz’ dimensions

Shalom Schwartz, researcher from Hebrew University, identified several dimensions of culture, including conservatism, hierarchy, harmony, egalitarianism, mastery, affective autonomy, and intellectual autonomy (Schwartz, 1999). According to Schwartz, conservatism is very similar to ‘collectivism’ introduced by Hofstede (2001), where members of society pay more attention to being ‘embedded’ in their social groups. “In conservative societies personal interests are not seen as different from those of the group and high value is placed on preserving the status quo and avoiding individual actions and attitudes that might undermine the traditional order of things.” [Gutterman, (year unknown), p.2, n.d.). Conservative societies also tend to be more traditionalist and conservative. Out of the seven dimensions we only picked conservatism (respect for tradition), hierarchy (existence of power relationships), egalitarian commitment (concern about others’ welfare), affective autonomy (individual pleasure seeking) and intellectual autonomy (individual information seeking) that were provided by Basabe and Ros (2005).

2.3 Trompenaars’ dimensions

Fons Trompenaars, another Dutch scholar, is mostly known for his universalism and particularism concepts that stand for whether people think ‘rules are rules’ or ‘rules are just rules’ that may change according to different circumstances (Trompenaars and Hampden-Turner, 1998). He also claimed that cultures can be contrasted based on how people display emotions (shown or hidden), how people perceive success; ascribed (given by the authority), achieved (gained by individual achievements), how people perceive time (linear or circular), how people perceive their environment (inner or outer locus of control), how people perceive their society (group orientation vs. individual autonomy) and how people perceive their destiny (everything is linked or nothing is linked). He reduced all of these aspects of culture into two major dimensions that are egalitarian commitment – success is gained by individual achievements and relationships are not deep – and utilitarian involvement – family and social relations are more important than everything.

2.4 World Values Survey

A large-scale study conducted in more than thirty countries showed that most nations in the southern hemisphere, particularly Middle Easterners, Latin Americans, Africans, and South East Asians, tend to have strong religious values (http://www.worldvaluessurvey.org). Countries in the northern hemisphere and the eastern bloc usually have a loose view of religion and overall choose rationalism over traditional values. The split between the self-expressive and survival values is seemingly related with a country’s development level, except for Latin America. Industrialised nations place a heavier value on freedom of speech, environmental protection, and social equality, while underdeveloped and developing nations give priority to job security and one’s own safety (http://www.worldvaluessurvey.org).

The World Values Survey, led by a famous cross-cultural scholar Ronald Inglehart, found that there are two major cultural dimensions, namely traditionalism vs. rationalism

6 A. Acar

and survival vs. self-expressiveness (http://www.worldvaluessurvey.org). There seems to be a significant relationship between traditionalism and social media use. According to the official website of the institution that manages these surveys, traditionalism is similar to collectivism and religiosity. It is not a surprise that people in Israel (strongly Jewish), Russia (strongly Orthodox), Argentina, Philippines and Chile (strongly Catholic) have been using social media heavily as they feel an urge to be in touch with their society and make sure their conservative values are preserved. Of course, there can be other explanations as religious and traditionalist societies tend to be more collective and collectivism results with a higher need for connectedness.

2.5 The GLOBE dimensions

Similar to other cross-cultural scholars, a group led by Robert House from University of Pennsylvania, conducted several cross-cultural studies to identify the relationship between leadership and culture in 61 countries (House et al., 2002). The project Global Leadership and Organizational Behavior Effectiveness (GLOBE) identified six aspects of culture that may impact how leaders are perceived around the world. It was found that leader effectiveness and leader acceptance were culture bound that the same leader may be perceived incompetent in one culture and very competent in another culture. Additionally, Dr. House and his group believed that organisations may have different practices from ideal social norms and that’s why they measured both values (how people are supposed to behave) and practices (how people actually behave) in each country (House et al., 2002).

2.6 Cross-cultural emotional expressiveness

David Matsumoto, a famous cross-cultural psychologist from San Francisco State University, wanted to see if people show their emotions in the same intensity all around the world. He postulated that emotions regulate social adaptation and social interactions, particularly when it comes to conveying messages to those from outside one’s own group (Matsumoto et al., 2008). His findings showed that people from individualistic cultures and English-speaking countries showed their emotions more intensely and Asians by and large suppressed their emotions. After surveying more than six thousand respondents from thirty-three countries, he managed to compare different nations on his emotional expressivity scale (Matsumoto et al., 2008). In this study we also attempted to assess whether emotionally expressive countries use social media more or less.

2.7 Self-esteem

Self-esteem is defined as “a positive or negative attitude toward a particular object, namely, the self” [Rosenberg, (1965), p.30]. There are currently some conflicting views about social media use and self-esteem. While sharing one’s own identity on the internet may boost his/her self-esteem (Toma, 2010), comparison with others and uploading content that is usually ignored may negatively impact one’s esteem. This debate is getting more intense as recently a research team from Sweden surveyed 1,000 people and concluded that using Facebook reduces self-esteem (Denti et al., 2012). To clarify the relationship between self-esteem and social media we decided to analyse the average time

Culture, corruption, suicide, happiness and global social media use 7 spent for social networking and the country level self-esteem scores (Schmidt and Allik,

2005).

2.8 Social capital

Social capital is defined as the “resources (such as information, ideas, support) that individuals are able to procure by virtue of their relationship with other people” (Grotoaert et al., 2004). Social capital in one country is usually measured by the levels of social participation, trust, and solidarity; citizen engagement with social groups; the frequency of cooperation and collective action among the members of society; and how much information is shared among the public (Grotoaert et al., 2004). Social capital is also measured on an individual level. It is argued that we rationally choose to use online social networks to generate bridging or bonding social capital, which in turn satisfies our utilitarian and affective needs in life. Researchers suggest that social media, especially Facebook, can help individuals increase their bridging (benefits from strong ties) and bonding (benefits from weak ties) social capital as it can turn latent ties into close friends (Ellison et al., 2011). Based on these assumptions, one may expect that countries with higher levels of social media engagement will have higher levels of social capital.

2.9 Privacy

The first study ever published on Facebook was about privacy (Gross and Acquisti, 2005) and in the past eight years several studies also looked at the relationship between privacy and online social network usage (Bulgurcu et al., 2010; Cavusoglu et al., 2013). It was found that people from individualistic countries disclose more information and worry more about their privacy than those in collectivistic countries (Thomson, 2013). Recently, two researchers from North America argued that the concept of social media privacy was so loose that in some countries people unknowingly shared some sections of their profiles as default and involuntarily uploaded content to see others’ profiles (Ur and Wang, 2013). The researchers called for more research in this underexplored area by pointing out that in each society there are different cultural norms of privacy, different regulations about invasion of privacy, and different user attitudes about protecting online content (Ur and Wang, 2013). We wanted to see if there is a relationship between country-level privacy protections and country-level social media engagement.

Privacy International (http://www.privacyinternational.org) gives a large list of countries an index score based on several criteria including constitutional protection, identity cards, data-sharing, workplace monitoring, democratic safeguards, visual surveillance, and so on. A high privacy index score is good and indicates low invasion of privacy. A low privacy score means there is a great deal of official or unofficial privacy invasion.

2.10 Subjective well-being

When people use social media, they perhaps feel more connected and happier, although this may not apply to those who have social problems. Past research also showed that people not only gained emotional and social support (Toma, 2010) by using Facebook but also reduced their loneliness (Große Deters and Mehl, 2012). Does social networking

8 A. Acar

really make people happy? Or is it simply happy people do more social networking? In other words, is there any connection between social networking and happiness? To answer these questions, we looked at the relationship between the Happy Planet Index and comScore’s social media use index. Since it is known that lonely people are more likely to commit suicide (Koivumaa-Honkanen et al., 2003) and cultural norms have an impact on suicides (Arsenault-Lapierre, 2004), we also added country-level suicide rates to the analysis.

2.11 Pace of life, income distribution and corruption

Pace of life index (Levine and Norenzayan, 1999) assigns each country a score that indicates whether life is fast- or slow-paced. These scores were obtained after the measurement of several indicators including how fast people walk in public areas, how punctual the public trains are, how punctual people are and etc. Recently Garcia-Gavilanes et al. (2013) found that pace of life along with GDP and income distribution impact how people tweet. Although there is no research available about corruption and social media use, we included corruption in this study as it may be related with GDP, income distribution and pace of life and we used transparency international’s corruption perception index scores reported by Lattin and Young (available online).

2.12 Big 5 personality variables

It does not make much sense to measure personality on the country level, but in the past researchers conducted several surveys that used the Big 5 personality-measurement scales in various countries and provided mean scores for each country (Schmitt et al., 2007). Since personality is a good predictor of whether one uses social media or not (Acar et al., 2012), we wanted to see how country-level personality characteristics compare to country-level social networking scores. We expected that social media penetration would positively correlate with extroversion, openness, while negatively correlating with neuroticism (Correa et al., 2010).

2.13 Other variables

The methodology section explains how we obtained country level social media use scores. Most of the other variables were obtained from World Bank’s database that is publicly available. Additionally, Appendix 1 lists all of the databases and resources used in this study.

3 Methodology

This study was inspired by an original report published by comScore on December 21, 2011, with the title It’s a Social World. The study reported that three out of every four minutes spent online globally were for social networking, women were more active social media users than men, local social networks were on the rise, and the future of social media was to be determined by smartphones and tablets. The study also provided local social media usage intensity information for 40 different countries around the world along with index scores. In each country comScore measured the percentage of people

Culture, corruption, suicide, happiness and global social media use 9

who have a social media account, the average time people spend for social networking, and the proportion of total time online spent on social networking. The data was adjusted for internet penetration in each country and then every country was given an index score for these measurements. For instance, on an index for the percentage of online population visiting social networks, Spain scored highest with 119 and Japan lowest with 70. These are not nominal percentage values but index scores according to the world average. Having these scores publicly available made us realise that these country-level usage intensity scores – for which there are already many datasets – can be used just like an individual-level data to understand the relationship between cultures and social media.

We used three simple analyses in this study. As a first step we looked at the difference between the high-context and low-context countries by using an independent sample t-test. As a second step we correlated all the values and socio-economic variables with comScore social media use scores. After the correlation analyses we generated data plots for some important relationships by using IBM SPSS 21. These data plots not only illustrate relationships between two variables but also produce a regression line and an R-square value presuming that the horizontal axis in each graph predicts the vertical axis. Lastly we ran six separate multiple regression analyses in which each cross-cultural value dimensions predicted the total social media use globally.

Note: The comScore report used the ‘term social networks’ throughout; however, the company treated all time spent on Twitter, Tumblr, Facebok and local networks the same as time spent on social networks. Since Twitter, Facebook, Tumblr, RenRen, Mixi and blogging platforms are simply considered as ‘social media’, the word ‘social networks’ in the comScore report means exactly the same thing as ‘social media’. Additionally, comScore measured the social media use intensity by using its Media Metrix method that relies on the internet panel recruited by the company. Advertising Research Foundation explains the methodology of Media Metrix as follows:

“The information retrieved from the panelists registered and active on the Internet includes all Web sites visited, viewing durations, and click streams. It captures delivery of pages, frames, banners and other display ads, videos and search engine queries, keywords used, and search ads served (although not all of this data is included in the Media Metrix offering.) All personally-identifiable information (PII) is stripped from web pages on panelists’ machines before being transferred to the comScore collection server.” [Cook and Pettit, (2009), p.5]

Additionally throughout this paper we use the terms social media use, social media use intensity, social media use behaviour interchangeably. We calculated this variable by summing up the three different values for each country reported by comScore and divided by 3 (% of social media users + proportion of social networking time + average time spent for social networking) / 3). This variable was named social_total in our data file and the data points are shown in Figure 1.

10 A. Acar
Figure 1 Social media use around the world

4 Findings

4.1 High-context and low-context communication

When we compared the average social media intensity scores of high-context and low-context countries we saw that there was no difference (mLow-Context: 105.13, std: 7.41 versus mHigh-Conext: 102.52, std: 33.71, p > .05, see Appendix 2). This means, high-context and low-context communication do not have anything to do with social media engagement. Despite the small sample size, we think this is a very clear finding, because both East Asian and Latin countries are considered high-context; the Japanese and Chinese usually have the lowest level of engagement whereas Latin countries have the highest. Additionally, nowadays social media has become a social norm among all different types of people in different cultures. Therefore, we should not expect any difference between high-context and low context cultures in terms of usage intensity.

Culture, corruption, suicide, happiness and global social media use 11 Figure 2 High-context and low context countries’ share of time spent for online social

networking

4.2 Hofstede’s dimensions

As seen in Figure 3, there is a negative relationship between individualism and the proportion of online time spent for social networking. This negative relationship however does not hold true for the average time spent for social networking or the percentage of social media users in a country. Latin American countries consistently score higher than the world average in terms of time they use for browsing social media sites or engaging with their friends on social platforms compared to the time they spend online for other activities (e.g., reading mails, news, etc.). Latin Americans also spend more time on social media when compared to the rest of the world. Although one may speculate that Westerners spend less time for online social networking because they are busy with their jobs, we should remember that Latin Americans spend more time on social media than do other developing countries as well.

Israel and Japan stand out in the graph, as the Japanese spend very little time on social networking even though they are not an idiocentric society and Israelis spend the highest amount of time on social media even though they are not a highly collectivistic society. Furthermore East Asian countries and Singapore clearly have low individualism scores and low ratios of social networking that contrast with the rest of other collective cultures from the Middle East and South East Asia. This implies that, in line with Professor Nisbett’s arguments about limiting Easterners to East Asians, collectivistic countries that are influenced by Confucianism have a dramatically different social media behaviour compared to those that are not influenced by Confucianism.

Readers should note that Western countries like the USA or the UK do not necessarily spend less time for social networking; apparently people in these cultures do

12 A. Acar
others things in addition to social networking on the internet. We found no significant

relationship between other variables (see Appendix 3 for details). Figure 3 Individualism and share of time spent for social networking

4.3 Schwartz’ dimensions

Despite the small sample size, most of the relationships were statistically significant particularly intellectual autonomy and conservatism. Broadly speaking, intellectual autonomy is a culture’s emphasis on mentally engaging individual activities, such as, the arts, creativity, and so on, and seems to be negatively correlated with the percentage of social media users in a country. This makes perfect sense as Northern European countries, which tend to be curious about other types of challenging activities and want to gather information about various things online, are less likely to spend a lot of time using social networking channels.

This graph makes it clear that Israel, Turkey, Malaysia and Mexico, the top social media users, also have very conservative and collectivistic values.

Culture, corruption, suicide, happiness and global social media use 13 Figure 4 Conservatism and global social media use

4.4 Trompenaars’ dimensions

Our correlation analysis (Appendix 5) indicated that neither of these two dimensions could explain social media use. It may be because the dimensions became too broad when reduced down to two. See appendixes for more details.

4.5 World Values Survey

One may expect that the survival vs. self-expressiveness dimension can successfully explain whether there is a country level interest in social media. However, usually countries that score high on this dimension tend to be underdeveloped and they were not included in the data analysis (see Appendix 6 for more details).

14 A. Acar
Figure 5 Traditionalism and global social media use

4.6 The GLOBE dimensions

Since these six elements of culture not only explain the leadership behaviour but also reflect how people from different cultures communicate, we ran correlations between comScore’s usage intensity scores and GLOBE’s country index scores. Not surprisingly, we observed that collectivism was the best predictor of social media use globally. Similar to Hofstede’s, Shwartz and Inglehart’s findings, once again, collectivism, group attachment in broader terms, had the highest correlation coefficient among all the variables. It also appeared that the GLOBE dimensions may clarify our confusion over what aspects of collectivism predicts social media use.

As can be seen in Appendix 7, in-group collectivism – need to express attachment, loyalty and pride – had a more significant relationship with social media use compared to societal collectivism (collective action practices in organisations). Furthermore, when we focus on practices, rather than values, we see that societal collectivism actually has a negative relationship with social media use. As counterintuitive as it may sound, obviously in societies where social media use is common there are not many organisation level incentives to encourage collective action. The fact that in-group collectivism had a solid relationship with social media also explains why social media use is very low in East Asian countries like Japan, Korea and Hong Kong. These countries have collectivistic values but they do not necessarily feel a need to express their pride and attachment via computer mediated communication.

Culture, corruption, suicide, happiness and global social media use 15 4.7 Cross-cultural emotional expressiveness

The results show that emotionally expressive countries are more likely to adopt social media than less emotionally expressive countries. Obviously, those who intensely express themselves feel a higher need to be in social media or people who are emotionally expressive feel more comfortable having social media presence.

Figure 6 Emotional expressiveness and global social media use

4.8 Self-esteem

As can be seen in Figure 7, the results implied that there must be a positive relation between self-esteem and social media use. Although Israel, Turkey and Chile may have little in common, their citizens have high self-esteem and spend a lot of time for social networking. More strikingly, Israel has the highest nation-level self-esteem and spends the highest amount of time for online social networking while Japan has the lowest nation-level self-esteem and spends the lowest amount of time for online social networking. This finding strengthens the argument that people with high self-esteem are comfortable sharing many things about themselves and/or their ideas thus do not mind spending time online while people with low self-esteem are worried about how others will react to their posts and spend a shorter time on social networking sites (it is more time consuming to post content compared to browsing through what others posted).

16 A. Acar
Figure 7 Self-esteem and global social media use

4.9 Social capital

Contrary to our expectations that more social media use means higher social capital, the data showed that it is completely the opposite (Appendix 10). The higher the social capital, the lower the levels of social media use we found. However, this finding may have several explanations. First of all, rich countries tend to have higher social capital and be individualistic. Individualistic countries do not use social networks as much as those in collectivistic countries, which tend to be poorer. Additionally, people in rich countries may participate in many face-to-face collaborative activities (volunteering, organising civic events, etc.) that reduce their time for online social networking. Lastly, this study did not measure country-level social media engagement and social capital over several years; a longitudinal study may provide a better answer to the question of whether social networks increase our social capital or not.

4.10 Privacy

By correlating the privacy index scores provided by Privacy International with the social media use index from comScore, we found absolutely no relationship between high-surveillance and low-surveillance countries in terms of time spent online for social networking (see Appendix 11).

Culture, corruption, suicide, happiness and global social media use 17 4.11 Personality

Interestingly, extroversion did not correlate well with online social media use indexes (Appendix 12). This is an interesting finding because extroversion usually distinguishes users from nonusers and predicts usage intensity. However, conscientiousness (a trait related to being careful and working hard to finish tasks) was positively correlated with average time spent social networking. This finding is in line with past studies that found that conscientious individuals spend time online perfecting their social media profiles and carefully evaluating the activities of network members. At the same time, openness was also positively related with time spent for social media, as apparently nations that are open to new ideas and different experiences enjoy using social media.

4.12 Subjective well-being

To our greatest surprise, there was a crystal-clear relationship between time people spend for online social networking and their country’s score on the global happiness index. Of course, correlation does not mean causation so we cannot say social media makes people happy. We can, however, say there seems to be a connection between happiness and social media use at least on the country level. Most of the countries with high social media use intensity also scored high on the global happiness index except Turkey which had a relatively low happiness score.

Figure 8 Happiness and global social media use

18 A. Acar

What is more, in countries where people had higher rates of time spent for social networking there were lower suicide rates, particularly for females (Appendix 13). It is known that women are more susceptible to mood disorders and more frequently experience depression compared with men (Stravynski and Boyer, 2001). Additionally, male suicides are more likely to be related to substance abuse and childhood disorders and female suicides to social disorders (Canetto and Sakinofsky, 1998). Since spending time in social media may create an illusion of togetherness, reduce loneliness, create opportunities to share problems with loved ones, and bring opportunities to be consoled by network members, we can conclude that in countries where people spend a great deal of time for online social networking less female suicides are likely to happen. However, readers should take this finding with a pinch of salt because, as we mentioned above, most of those countries with high social media use are relatively more religious and conservative where suicide is usually uncommon.

Figure 9 Female suicide rates and global social media use

4.13 Pace of life, income distribution and corruption

Past studies suggested that income inequality and pace of life must be taken into account to understand global social media behaviour. As can be seen in Appendix 14, pace of life (whether life is fast- or slow-paced, people walk fast or slowly on the street, people are punctual or not, etc.) had a weak but positive relationship with social media use intensity. Income inequality, along with GDP, however, was strongly related to social media use.

Culture, corruption, suicide, happiness and global social media use 19

Similar to the Gini index, we calculated income inequality by dividing the income shared by the highest 20% with income shared by the lowest 20%. A higher ratio of rich to poor income share indicates a bigger gap between rich and poor. We can speculate that in slow paced countries like Brazil, Mexico and Indonesia, people put more emphasis on enjoying online (or offline) social interactions with their peers and social relationships are more important than efficiency at work. Note: A positive pace index score means routine activities take longer time to complete.

The data shows that people from countries where national income is unequally distributed spend more time for social media (Appendix 14). On the contrary, the higher a country’s GDP, the lower the time people spend for social networking (Appendix 10). One may argue that people from rich countries have jobs and do not have much time using Facebook or Twitter but according to the data, unemployment had no significant impact on time spent for social networking. So, the most logical argument then could be that rich countries tend to be more individualistic and autonomous, thus use the internet for individual purposes such as entertainment or information browsing. People from developing countries, however, tend to be more collectivistic and use social media to get emotional and social support. Apparently, as countries get richer, people spend less time on social networking.

Figure10 Corruptionperceptionindexandglobalsocialmediause

It was also obvious that social media use intensity had a positive relationship between country-level corruption. People may automatically think that this finding is not a

20 A. Acar

surprise, as there is usually more corruption in countries with huge income inequality. However, there was no relationship between income inequality and corruption. This finding indicates that not all countries with income inequality are corrupt; the corrupt ones tend to be those who have high social media engagement and those that are more collective. We cannot conclude that social networking gives people more opportunities to take advantage of the system or gain benefits without working hard but it seems like people feel a need to use social media more if there is corruption.

4.14 Demographics

GDP growth but not population growth are related to social media use, as in general developing countries tend to have more room to grow their GDP. Population growth may not predict social media use because countries that have flexible immigration policies such as Canada, Australia, Hong Kong and Singapore have a growing population but at the same time their median age is still higher than the average. Not surprisingly, median age (whether the country has a young population or not) correctly predicted country level social media use similar to past studies which found that young people use social media more frequently and spend more time for social networking. Russia, however, stood out as an outlier as it has a relatively older population who uses social media very actively.

Figure11 Medianageandglobalsocialmediause

Culture, corruption, suicide, happiness and global social media use 21

On the other hand, the percentage of urban population, and cultural diversity did not have any relationship with social media use (Appendix 14). One may expect that countries with a higher percentage of urban population spend more time for social networking but this was not the case. Additionally, social network researchers commonly emphasise homophily (similar people tend to interact more); however, in our findings we observed that cultural diversity and ethnic fractionalisation had neither a positive nor negative impact on social media use intensity (see appendixes). Large countries, which presumably have larger cultural diversity, use social media more because people can make different connections with people from different backgrounds and consume different online content. On the other hand, smaller countries, which presumably have more homogenous populations, may use social media frequently as well, to gain a feeling of solidarity. Obviously, we observed no difference because different types of countries use social media with high frequency for different compelling reasons.

Finally, we wanted to answer our research question by running separate standard multiple regression analyses where total social media use is the dependent variable and cultural dimensions are independent variables. We ran a separate analysis for each set of cultural values introduced by Hofstede, Hall, Trompenaars, Schwartz, the GLOBE study and World Values Survey in order to see which set of values does the best job explaining the variance in the target variable. This is a practical method to identify the best set of values by checking the R-square values but it is not ideal because not all individual variables in each model have significant beta coefficients (e.g., Schwartz’s set generated the highest R-square but two out of five variables have insignificant Beta coefficients).

After testing the six different cultural dimensions introduced by the past studies, we confirmed that culture does influence social media use intensity as three out of six models were statistically significant. However, Hofstede’s and Hall’s dimensions, perhaps the most popular cross-cultural frameworks, failed to explain a significant amount of variation in social media use. When we ran linear regression analyses where social media use intensity ((% of social media users+ proportion of social networking time + average time spent for social networking) / 3) was predicted by the six sets of cultural dimensions, we found that Schwartz’s dimensions did the best job by explaining 42% of the variation (R-square: 0.26, F20: 3.85, p < .05) followed by the GLOBE dimensions (R-square: .31, F32, 2.57, p < .05) and the World Values Survey (F37: 7.43, p < .05). On the other hand, the total explained variance accounted for Hofstede’s four major dimensions was only about 13%, meaning 87% of cross-cultural social media behaviour can be explained by other factors. Figure 12 illustrates the total variance explained by each cross-cultural set of dimensions.

22 A. Acar
Table 2 Regression models, dependent variable: social media use (social_total)

Model 1: Hofstede’s dimensions

Model 2: Schwartz’s dimensions

Model 3: Trompenaars’ dimensions

Model 4: GLOBE dimensions

Power distance Individualism Masculinity
Uncertainty avoidance R-square
F
Affective autonomy Intellectual autonomy Conservatism
Hierarchy
Egalitarian commitment R-square
F
Egalitarian commitment Utilitarian involvement R-square
F
Assertiveness values Societal collectivism values
In group collectivism values Future orientation values Gender egalitarianism values Humane orientation values Performance orientation values Power distance values Uncertainty avoidance values R-square
F
Traditionalism vs. rationality Survival vs. self-expressiveness R-square
F
High vs. low context communication R-square
F

Beta

0.142 –0.192 0.031 0.284 0.131 2.431 0.119 0.639 1.586 0.277 0.829 0.417 3.857 –0.365 0.37 0.088 2.2 –0.089 0.023 0.4 –0.024 0.036 –0.207 0.021 –0.35 0.337 0.307 2.574 –0.5 –0.137 0.258 7.43 0.358 0.0744 2.357

t

0.695 –0.935 0.202 1.808

0.38

1.79 3.667 1.047 2.702

–1.747 1.769

–0.506 0.109 1.928 –0.119 0.157 –1.122 0.094 –1.991 1.53

–3.456 –0.948

1.535

Sig. 0.492 0.356 0.841 0.08

0.066 0.71 0.094 0.002 0.312 0.016

0.019 0.094 0.09

0.134 0.618 0.914 0.066 0.906 0.876 0.274 0.926 0.058 0.14

0.033 0.001 0.35

0.002 0.144

0.144

Model 5: World Values Survey

Model 6: Hall’s context dimensions

Culture, corruption, suicide, happiness and global social media use 23 Figure12 Totalvarianceexplainedbyeachcross-culturalsetofdimensions

Although this paper was not written to develop a cross-cultural social media communication framework we were quite disappointed to see how some popular theories failed to explain this important phenomenon. Inspired by this graph, we thought it’s better for researchers to focus on the dimensions of Schwartz. Since it is difficult to combine all five of Schwartz’s dimensions, we focused on the one that had the highest positive relationship with social media: conservatism. According to Schwartz, conservatism is very similar to collectivism (group and other orientation) with some emphasis on traditions. We also thought the predictive power of conservatism can be strengthened by a personal trait like self-esteem which also explained a huge amount of variation in social media as illustrated by Figure 4. After all ‘self-esteem’ can play a differentiating role between collectivistic countries that use social media heavily (e.g., Argentina, Mexico, etc.) and East Asian countries that have the lowest usage rates (e.g., Japan, Korea, etc.).

As a first step we plotted conservatism against self-esteem and tried to understand what kind of relationship there is between these two variables globally. Dividing the graph into four quadrants helped us better understand how each group of countries may behave online. Obviously in some countries people think others are more important than them and in some countries people think others are less important. Based on this graph we concluded that there is no country with low self-esteem and low collectivism and only countries that were influenced by Confucianism (China, Japan, Hong Kong, Taiwan, etc.) have low self-esteem and high collectivism.

Culture, corruption, suicide, happiness and global social media use 25

Encouraged by the perfect relationship between self-esteem/conservatism and global social media use we thus propose a new cross-cultural online communication framework called cross-cultural self and others’ worth which divides the world into three regions based on the perception of one’s own self value and the value of his/her social group. As explained in Table 4, these three regions are

1. a  East Asian countries that are influenced by Confucianism (Japan, Korea, Hong Kong, China, Taiwan, and Singapore)
2. b  Western countries (English speaking countries and most of Europe)
3. c  the rest of the world (particularly Latin America, the Middle East, Eastern Bloc countries, South East Asia and Africa).

However, readers should note that Table 4 is not directly based on the findings of this study but a derivation of Nisbett’s work and our literature review on how different countries use social media. This framework hasn’t been tested yet.

Table 4 Cross-cultural self and others’ worth framework

Group I: Low self-esteem and high conservatism

• Not confident about his/her own content
• Careful about not bothering others online
• Thinks that others may not care about him/her
• Not comfortable expressing personal opinions publicly
• Suppresses his/her (positive or negative) emotions
• Does not care much about being popular online

  Example: Japan, Korea, Hong Kong, China, Taiwan and Singapore.

  5 Conclusions

Group II: High self-esteem and low conservatism

• Confident about own content

• Concerned about self-image

• Emotionally expressive • Comfortable expressing

personal opinions publicly

• Would like to be popular online

• Not so interested about sharing community/group related messages

• Not so concerned/curious about what everybody is up to

Example: Germany, France, Netherlands, Italy, Portugal, Spain, etc.

Group III: High self-esteem and high conservatism

• Confident about own content

• Concerned about self-image

• Emotionally expressive • Comfortable expressing

personal opinions publicly

• Would like to be popular online

• Interested in sharing community/group related messages

• Concerned/curious about what everybody is up to

Example: Turkey, Egypt, Israel, Mexico, Argentina, Malaysia, Indonesia, Poland, etc.

This study was conducted to answer one simple research question that is if culture plays a role in social media use. We also included some socio economic, demographic and personality variables in addition to commonly known cultural dimensions in this study because some past studies indicated a partial relationship between socio economic variables and social media use. Our analyses have shown that culture is directly related to

26 A. Acar

country-level social media use which may also be related with country-level self-esteem, pace-of-life, happiness, suicide rates, GDP per capita, median age and corruption. The findings indicated that in countries where people use social media heavily there is low suicide, high corruption, low GDP, high self-esteem and high respect for traditions. At the same time societies with low social media use rates tend to be older, less emotionally expressive, less happy, score low on openness and conscientiousness, have higher GDP and higher social capital.

We observed that Schwartz’s cultural dimensions, the GLOBE study and the World Value Survey successfully explained more than 25% of the variation in the global social media behaviour. On the other hand, Hall’s high-low context communication theory was totally ineffective. We believe this was mostly because Hall grouped East Asian nations – who do not want to bother others online, who do not want to show their emotions publicly and who relatively have low self-esteem – the same way as Arabs and Latin Americans who have relatively high self-esteem and who are emotionally expressive. Similarly, Trompenaars’ dimensions could not explain social media use intensity in different countries perhaps because of the business aspects of the dimensions. That is why we introduced the cross-cultural self-versus others’ worth framework that can contribute significantly to the future cross-cultural social media use studies.

As explained in the findings section, we do not speculate that social media increases happiness, openness, national self-esteem and corruption. By the same token we do not claim that social media use reduces suicides. By no means have this paper talked about direct or indirect causality as we all know that correlation, what most of this paper is about, does not mean causation. For instance if there is a positive correlation between social media use intensity and country-level suicide rates this means in countries where people use social media more there is relatively low suicide record; however the low suicide may have nothing to do with the social media use or both low suicide and high social media use may be impacted by one common factor such as collectivism. In order to make any causal conclusion there needs to be a longitudinal study where two comparable countries with similar suicide rates but dissimilar social media use are tracked for a certain time. Additionally most of these concepts may be related (e.g., it is known that there is a negative relationship between individualism and GDP per capita) a comprehensive structural equation models can explain these relationships in a more meaningful way.

6 Limitations

Despite the fact that most of the findings made sense and were in line with the past research, this study has many limitations and the findings should be considered explanatory. First, all the data was gathered from secondary sources, some of which may be outdated or not validated. More importantly, most importantly, the data we gathered from secondary resources had different levels of measurement (e.g., ordinal data, interval data, normalised (index) data, nominal data, etc.) that are likely to have different error distributions, a possible cause of heteroscedasticity. Secondly, the sample size was very small (e.g., we only had the index scores of 20 countries for Schwartz’s dimensions and 32 countries for the GLOBE dimensions). Lastly, comScore’s measurement of social media use is different than how academicians measure individual social media use (weekly access, number of activities, etc.) and comScore’s report always uses the word

Culture, corruption, suicide, happiness and global social media use 27 ‘online social networking’ instead of social media that also refers to interest networks and

interactive entertainment platforms such as Pinterest and Youtube.

References

Acar, A, Rauschnabel, P., Lin, C. and Fukui, M. (2012) ‘The relationship between personality and social media preferences’, Proceedings of the 45th Annual Conference of Japan Association of Consumer Studies.

Acar, A. (2008) ‘Antecedents and consequences of online social networking behavior: the case of Facebook’, Journal of Website Promotion, Vol. 3, Nos. 1–2, pp.62–83.

Acar, A. (in press) Culture and Social Media.
Acar, A. and Deguchi, A. (2013) ‘Culture and social media usage: analysis of Japanese Twitter

users’. International Journal of Electronic Commerce, Vol. 4, No. 1, pp.21–32.

Acar, A., Takamura, D., Sakamoto, K. and Nishimuta, A. (2013) ‘Culture and brand communications in social media: an exploratory analysis of Japanese and US brands’, International Journal of Web Based Communities, Vol. 9, No. 1, pp.140–151.

Arsenault-Lapierre, G., Kim, C. and Turecki, G. (2004) ‘Psychiatric diagnoses in 3275 suicides: a meta-analysis’, BMC Psychiatry, Vol. 4, No. 1, p.37.

Auter, P. and Elmasry, M. (2012) ‘A cross-cultural content analysis of student Facebook use’, paper to be presented at the Saudi Association for Media & Communication 6th Annual Forum: Social Media and Public Opinion, Riyadh, Saudi Arabia.

Basabe, N. and Ros, M. (2005) ‘Cultural dimensions and social behavior correlates: individualism-collectivism and power distance’, International Review of Social Psychology, Vol. 18, No. 1, pp.189–225, 223–224.

Bulgurcu, B., Cavusoglu, H. and Benbasat, I. (2010) ‘Understanding emergence and outcomes of information privacy concerns: a case of Facebook’, Proceedings of the 31st Annual International Conference on Information Systems (ICIS), Saint Louis, MO, Paper 230.

Canetto, S.S. and Sakinofsky, I. (1998) ‘The gender paradox in suicide’, Suicide and Life-Threatening Behavior, Vol. 28, No. 1, pp.1–23.

Cavusoglu, H., Phan, T. and Cavusoglu, H. (2013) ‘Privacy controls and information disclosure behavior of online social network user’, Proceedings of the Workshop on Economies of Information Security (WEIS), Washington, DC, 11–12 June 2013.

CIA World Fact Book (2013) Median Age [online] http://world.bymap.org/MedianAge.html.

comScore, It’s a Social World: Top 10 Need-to-Knows About Social Networking and Where It’s Headed, 21 December 2011, retrieved 15 November 2013 [online] http://www.comscore.com/content/download/12135/231287/file/Top_10_Need-to- Knows_About_Social_Networking_and_Where_it_is_headed.pdf+&cd=1&hl=en&ct=clnk.

Cook, W.A. and Pettit, R.C. (2009) comScore Media Metrix US Methodology: An ARF Research Review, retrieved 15 November 2013 [online] thearf-org-aux- assets.s3.amazonaws.com/downloads/ research/comScore-RReview.pdf.

Correa, T., Hinsley, A.W. and De Zuniga, H.G. (2010) ‘Who interacts on the web?: the intersection of users’ personality and social media use’, Computers in Human Behavior, Vol. 26, No. 2, pp.247–253.

Denti, L., Barbopuolos, I., Nilsson, I., Holmberg, L., Thulin, M., Wendeblad, M. and Davidsson, E. (2012) Sweden’s Largest Facebook Study, Gothenburg Research Institute, retrieved 15 November 2013 [online] https://gupea.ub.gu.se/bitstream/2077/28893/1/gupea_2077_28893_1.pdf.

Abdallah, S., Michaelson, J., Shah, S., Stoll, L. and Marks, N. (2012) The Happy Planet Index:

2012 Report, A Global Index of Sustainable Well-being, The New Economics Foundation,

UK.

28 A. Acar

Ellison, N.B., Steinfield, C. and Lampe, C. (2011) ‘Connection strategies: social capital implications of Facebook-enabled communication practices’, New Media & Society, Vol. 13, No. 6, pp.873–892.

eMarketer (2013) Social Networking Reaches Nearly One in Four Around the World, 18 June, retrieved 26 December 2013 [online] http://www.emarketer.com/Article/Social-Networking- Reaches-Nearly-One-Four-Around-World/1009976#HEAXV1Ujbjd9jo9o.99.

Eve, P. (2012) ‘15 Interesting facts about international social media use’, EConsultancy Web Log, 26 October [online] http://econsultancy.com/us/blog/10955-15-interesting-facts-about- international-social-media-use.

Fearon, J.D. (2003) ‘Ethnic and cultural diversity by country’, Journal of Economic Growth, Vol. 8, No. 2, pp.195–222.

Garcìa-Gavilanes, R., Quercia, D. and Jaimes, A. (2013) ‘Cultural dimensions in Twitter: time, individualism and power’, in Proceedings of the The 7th International AAAI Conference on WebLogs and Social Media (ICWSM).

Gross, R. and Acquisti, A. (2005) ‘Information revelation and privacy in online social networks’, in Proceedings of the 2005 ACM Workshop on Privacy in the Electronic Society, ACM, pp.71–80.

Große Deters, F. and Mehl, M.R. (2012) ‘Does posting Facebook status updates increase or decrease loneliness? An online social networking experiment’, Social Psychological and Personality Science.

Grotoaert, C., Narayan, D., Jones, V.N. and Woolcock, M. (2004) ‘Measuring social capital: an integrated questionnaire in Karen Healy and Anne Hempsire’, Beyond the Local: Using Social Networks as a Policy Resource for Building Resilient Communities, The World Bank, Washington, DC.

Gutterman, A.S. (year unknown) Shalom Schwartz’s Dimensions of Societal Cultures, retrieved 15 November 2013 [online] http://alangutterman.typepad.com/files/cms—schwartzs-cultural- dimensions.pdf.

Hall, E.T. and Hall, M.R. (1990) Understanding Cultural Differences, Intercultural Press.

Hermeking, M. (2005) ‘Culture and internet consumption: contributions from cross-cultural marketing and advertising research’, Journal of Computer-Mediated Communication, Vol. 11, No. 1, pp.192–216.

Hochman, N. and Schwarts, R. (2012) ‘Visualizing instagram: tracing cultural visual rhythms’, in

Proceedings of the 6th International AAAI Conference on Weblogs and Social Media (ICWSM).

Hofstede, G.H. (2001) Culture’s Consequences: Comparing Values, Behaviors, Institutions and Organizations Across Nations, Sage.

House, R., Javidan, M., Hanges, P. and Dorfman, P. (2002) ‘Understanding cultures and implicit leadership theories across the globe: an introduction to project GLOBE’, Journal of World Business, Vol. 37, No. 1, pp.3–10.

House, R.J., Hanges, P.J., Javidan, M., Dorfman, P.W. and Gupta, V. (2004) Culture, Leadership, and Organizations, Sage.

IPSOS OTX News and Polls (2013) 71% of Global Internet Users ‘Share’ Social Media Content Monthly: Pics (43%) plus Opinions, Status Updates and Links to Articles each Top Out at 26%, 13 August [online] http://www.ipsos-na.com/news-polls/pressrelease.aspx?id=6216.

Jackson, L.A. and Wang, J.L. (2013) ‘Cultural differences in social networking site use: a comparative study of China and the United States’, Computers in Human Behavior, Vol. 29, No. 3, pp.910–921.

Koivumaa-Honkanen, H., Honkanen, R., Koskenvuo, M. and Kaprio, J. (2003) ‘Self-reported happiness in life and suicide in ensuing 20 years’, Social Psychiatry and Psychiatric Epidemiology, Vol. 38, No. 5, pp.244–248.

Culture, corruption, suicide, happiness and global social media use 29

Lattin, R. and Young, S., Country Ranking: Social Capital Achievement [online] https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=0CC4QFjAA &url=http%3A%2F%2Fwww.cauxroundtable.org%2Fview_file.cfm%3Ffileid%3D43&ei=Ff FGUr20EMjMkwXNqICIDA&usg=AFQjCNGXqaXgLdyAWHiNuP41E1Fk1M8clg&sig2=o 4GaAHWDr43mrS-cb88mkA&bvm=bv.53217764,d.dGI.

, V ol. 18, No. 5 [online] http://firstmonday.org/ojs/index.php/fm/article/view/4366 (accessed 15 January 2014).

Levine, R.V. and Norenzayan, A. (1999) ‘The pace of life in 31 countries’, Journal of Cross- Cultural Psychology, Vol. 30, No. 2, pp.178–205.

Markus, H.R. and Kitayama, S. (1991) ‘Culture and the self: implications for cognition, emotion, and motivation’, Psychological Review, Vol. 98, No. 2, p.224.

Marshall, B.A., Cardon, P.W., Norris, D.T., Goreva, N. and D’Souza, R. (2008) ‘Social networking websites in India and the United States: a cross-national comparison of online privacy and communication’, IIS, Vol. 9, No. 2.

Matsumoto, D., Yoo, S.H. and Fontaine, J. (2008) ‘Mapping expressive differences around the world: the relationship between emotional display rules and individualism versus collectivism. Journal of Cross-Cultural Psychology, Vol. 39, No. 1, pp.55–74.

Peluchette, J.V., Karl, K. and Schlagel, C. (2010) ‘A cross-cultural examination of student attitudes and gender differences in Facebook profile content’, Faculty of Commerce-Papers, pp.11–31.

Peters, A., Oren, M. and Bidwell, N. (2012) ‘Namibian and American cultural orientations toward Facebook’, CHI EA.

Poblete, B., Garcia, R., Mendoza, M. and Jaimes, A. (2011) ‘Do all birds tweet the same?: characterizing twitter around the world’, in Proceedings of the 20th ACM International Conference on Information and Knowledge Management, ACM, October, pp.1025–1030.

Privacy International (2007) National Privacy Ranking 2007 – Leading Surveillance Societies Around the World [online] https://www.privacyinternational.org/sites/privacyinternational.org/files/filedownloads/phrco mp_sort_0.pdf.

Reinecke, K., Nguyen, M.K., Bernstein, A., Naf, M. and Gajos, K.Z. (2013) ‘Doodle around the world: Online scheduling behavior reflects cultural differences in time perception and group decision-making’, in Proceedings of the 16th ACM Conference on Computer Supported Cooperative Work and Social Computing (CSCW).

Rosen, D., Stefanone, M.A. and Lackaff, D. (2010) ‘Culturally unique social patterns in computer-mediated social networking’, in J-R. Park and E. Abels (Eds.): Interpersonal Relations and Social Patterns in Communication Technologies: Discourse Norms, Language Structures and Cultural Variables, IGI Global, pp.354–367.

Rosenberg, M. (1965) Society And The Adolescent Self-Image, Princeton University Press, Princeton, NJ.

Rui, J. and Stefanone, M.A. (2012) ‘Strategic self-presentation online: a cross-cultural study’, Computers in Human Behavior.

Schmitt, D.P. and Allik, J. (2005) ‘Simultaneous administration of the Rosenberg self-esteem scale in 53 nations: exploring the universal and culture-specific features of global self-esteem’, Journal of Personality and Social Psychology, Vol. 89, No. 4, p.623.

, Vol. 89, No. 4, p.623.

Schmitt, D.P., Allik, J., McCrae, R.R. and Benet-Martínez, V. (2007) ‘The geographic distribution of Big Five personality traits patterns and profiles of human self-description across 56 nations’, Journal of Cross-Cultural Psychology, Vol. 38, No. 2, pp.173–212.

Schwartz, S.H. (1999) ‘A theory of cultural values and some implications for work’, Applied Psychology, Vol. 48, No. 1, pp.23–47.

Leetaru, K., Wang, S., Cao, G., Padmanabhan, A. and Shook, E. (2013) ‘Mapping the global

Twitter heartbeat: the geography of Twitter’, First Monday

Schmitt, D.P. and Allik, J. (2005) ‘Simultaneous administration of the Rosenberg self-esteem scale

in 53 nations: exploring the universal and culture-specific features of global self-esteem’,

Journal of Personality and Social Psychology

30 A. Acar
Sharrock, J. (2013) ‘How Facebook plans to make us all get along’, BuzzFeed, January [online]

http://www.buzzfeed.com/justinesharrock/how-facebook-plans-to-make-us-all-get-along. Stravynski, A. and Boyer, R. (2001) ‘Loneliness in relation to suicide ideation and parasuicide: a

population-wide study’, Suicide and Life-Threatening Behavior, Vol. 31, No. 1, pp.32–40. Thomson, R. (2013) A Socio-Ecological Approach to Privacy Concern on Facebook: The Role of

Relational Mobility, Masters thesis, Hokkaido University, Sapporo, Japan.
Toma, C.L. (2010) ‘Affirming the self through online profiles: beneficial effects of social

networking sites’, in Proceeding of CHI 2010, pp.1749–1752.
Trompenaars, F. and Hampden-Turner, C. (1998) Riding the Waves of Culture, McGraw-Hill, New

York.

Tsai, W.H.S. and Men, L.R. (2012) ‘Cultural values reflected in corporate pages on popular social network sites in China and the United States’, Journal of Research in Interactive Marketing, Vol. 6, No. 1, pp.42–58.

Ur, B. and Wang, Y. (2013) ‘A cross-cultural framework for protecting user privacy in online social media’, in Proceedings of the 22nd International Conference on World Wide Web Companion, International World Wide Web Conferences Steering Committee, pp.755–762.

Wang, Y., Norice, G. and Cranor, L.F. (2011) ‘Who is concerned about what? A study of American, Chinese and Indian users’ privacy concerns on social network sites’, in Trust and Trustworthy Computing, Springer Berlin Heidelberg, pp.146–153.

World Bank (2013) Retrieved 1 September from World Data Bank Database. http://www.who.int/mental_health/prevention/suicide_rates/en/.

World Values Survey Group (2013) World Values Survey [online] http://www.worldvaluessurvey.org/wvs/articles/folder_published/article_base_54/files/ValueS cores_5_waves.doc.

Zhao, C. and Jiang, G. (2011) ‘Cultural differences on visual self-presentation through social networking site profile images’, in Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, ACM, May, pp.1129–1132.

World Health Organization (2013) Suicide Pates Per 100,000 by Country, Year and Sex (Table),

retrieved 1 September [online]

Culture, corruption, suicide, happiness and global social media use 31 Appendix 1

Data sources

Social media use index (% of social network users, average time spent for social networking, proportion of social networking activities)

High-context and low-context countries Hofstede’s dimensions
Schwartz’s dimensions
Tompenaars’ dimensions

World Values Survey

The GLOBE dimensions

Self-esteem values

Pace of life values

Emotional expression values

Big 5 personality variables

Social capital index, transparency international corruption perceptions index

Privacy index Happiness index Suicide rates***

GDP per capita, income distribution, GDP growth, population growth, urban population ratio, unemployment rates***

Cultural diversity, ethnic fractalisation Median age***

comScore

Hall and Hall (1990) http://geert-hofstede.com/

Basabe and Ros

Basabe and Ros
World Values Survey Group (2013)

and Allik
and Norenzayan Matsumoto et al. (2008) Schmitt et al. (2007)
Lattin and Young (available online)

Privacy International (2007)

World Bank (2013)

Fearon (2003)

CIA World Fact Book (2013)

House et al. (2004)

Schmitt

(2005)

Levine

Abdallah et al. (2012)

(1999)

World Health Organization (2013)

Notes: ***values from the year 2011 were used. If no value from 2011 was available, the 2010 values or the most recent values were used.

Appendix 2

High context-low context communication and social media use intensity
N Mean Std. deviation Significance

Group statistics

Social media use intensity

Context

Low context countries

12 7

105.1389 102.5238

7.41614 33.71276

.796

High context countries

Culture, corruption, suicide, happiness and global social media use 37 Appendix 8

Emotional expressiveness and global social media use

12345

percentage_of_social_media_users share_of_time_spent_social_networking average_time_spent_social_networking social_total

Emotional expressiveness
Note: Sample size: 21 (excluding Japan)

Appendix 9

1
0.266 1 .387* .706** .397* .923**

.590** –0.312

1
.922** 1
–0.087 –0.192 1

Self-esteem and global social media use

percentage_of_social_media_users

share_of_time_spent_social_networking average_time_spent_social_networking social_total
Self-esteem

Note: Sample size: 31 (excluding Japan)

1

.475** 1 .490** .748** .584** .936** .632** .512**

1
.928** 1
.685** .663** 1

12345

Culture, corruption, suicide, happiness and global social media use 43

Appendix 15

Average time spent for social networking all around the world

44 A. Acar
Appendix 16
Share of time spent for online social networking all around the world