A Cross-Cultural Study of Risky Online Self-Presentation

Abstract

The use of social media is pervasive among young adults. However, not all posted content is beneficial to their self-presentation, but can have negative and damaging consequences. This study investigated how individual differences in self-monitoring and impulsiveness influence risky online self-presentation in British and Italian samples. British participants (n = 88) were more likely to post comments and images related to their alcohol and drug use, whereas Italian (n = 90) participants posted more offensive content and personal information. High self-monitoring and high impulsiveness was positively predictive of risky self-presentation online regardless of nationality, highlighting the normative influence of social media culture, and the influence of both spontaneous and deliberative behavior on posting inappropriate content online. These novel insights regarding the way young adults present themselves on social network sites could help explain differences in self-presentation.

Introduction

Social networking sites (SNSs) are extremely popular among adolescents and young adults, providing them with a unique platform to enhance their social development,¹ increase social capital,² and find academic and employment opportunities. However, not all user-generated content on SNSs is appropriate or even legal. Young adults often use SNSs to share images of alcohol and drug consumption,^3,4 disseminate personal information (e.g., credit card details),^5,6 and post (semi-)nude selfies.⁷ Since young Internet users from different European countries have been shown to behave differently and experience different risks online,⁸ this study investigated individual and cultural differences in risky online self-presentation in the United Kingdom and Italy.

Most users report that they would be happy for their friends and family to view their SNS posts. However, many worry about future employers or strangers gaining access to this information.⁹ In fact, almost 40 percent of British, Canadian, and U.S. companies now use SNSs to check candidates' suitability.^10–12 Individuals have been fired from jobs,¹³ resigned from public office,¹⁴ and suspended from higher education¹⁵ because of disparaging social media posts. At the same time, researchers¹⁶ have argued that positive self-presentation on SNSs is more vital than ever due to the “nonymous”¹⁶ nature of these sites. Indeed, self-presentation management, successfully portraying a positive image of oneself, while avoiding creating an unfavorable one, appears to run counter to posting potentially damaging information online.¹⁷ Therefore, it is vital to understand the processes that might underlie the propensity to self-disclose personal and unfavorable information on SNSs.

It is debated how much (cognitive) effort individuals invest in online self-presentation. Some suggest that postings on sites such as Instagram or Twitter are spontaneous¹⁶ and may be linked to impulsivity.³ Risky online posts on SNSs, therefore, might be driven by individuals not spending time and cognitive efforts on thinking about the (negative) effects of those posts. Others¹⁶ indicate that online personas, particularly on dating sites, are carefully crafted and edited until an ideal-self is presented, suggesting a fully deliberated approach. One fundamental factor in such a deliberate approach to online self-presentation might be self-monitoring, typically defined as an individual's ability to regulate their physical and emotional self-presentation such that situationally appropriate, favorable self-images are maintained.¹⁸ Individuals high in self-monitoring adapt the information they present of themselves based on social and interpersonal cues and norms. Thus, high self-monitors adjust their self-presentation to fit with what they perceive to be favored by others in a particular situation. Conversely, low self-monitors maintain a consistent self-image more akin with their true selves, personality, and beliefs.⁹ Individuals low in self-monitoring are also typically more impulsive,²⁰ probably because they do not have to adapt their self-image to different situations.

In this study, we investigated whether risky online posting on SNSs are associated with spontaneous (i.e., impulsive) or deliberate (i.e., self-monitoring) processes. Previous research indicates that impulsivity is positively related to risky online self-disclosure.³ However, since high self-monitors strive to amend their self-presentation in line with perceived social and situationally appropriate norms,¹⁸ people high in self-monitoring might also be more likely to post risky information on SNS, because they perceive this to be the “right thing to do” in these situations.

While previous research highlighted cultural differences in the perception and use of social media^21–25 and Internet performance and ability,²⁶ cultural variations in risky online self-presentation and its underlying processes have rarely been considered. Karl et al.¹⁹ argued that cultural variations, based on Hofstede's²⁷ six cultural dimensions, could elucidate differences in online risky self-presentation. American, compared with German, students were more likely to post inappropriate material (e.g., sexual content) on their profiles, due in part to the lower Uncertainty Avoidance and higher Individualist culture in America.²⁸ We compared the behavior of young adults from Italy and the United Kingdom. British culture scores low on Uncertainty Avoidance resulting in a relaxed attitude toward uncertainty and an acceptance to take things as they come.²⁷ Conversely, Italian culture scores high on Uncertainty Avoidance, indicating intolerance for beliefs and behaviors outside the norm and more rigid codes of conduct. Additionally, the British high score on the Indulgence dimension is associated with an inclination to gratify desires for the purposes of fun and enjoyment, whereas Italy's lower score on this dimension is associated with a suppression of gratification to preserve social normative expectations.²⁷

In sum, we hypothesized that (a) people high in impulsiveness would display higher rates of risky online self-presentation; (b) participants high in self-monitoring should engage in higher rates of risky online self-presentation; (c) there would be an interaction between self-monitoring and impulsiveness; (d) due to their higher cultural scores on Indulgence and lower scores on Uncertainty Avoidance British participants would score higher on impulsivity compared with Italians. Therefore, impulsivity would be a stronger predictor of risky online self-presentation for British participants; (e) due to their higher cultural scores in Uncertainty Avoidance and lower scores in Indulgence, Italians should show higher self-monitoring than British participants. Consequently, self-monitoring should be a stronger predictor of risky online self-presentation for Italian participants. We further included a Self-Presentation measure and time spent online as control variables.

Methods

Participants

One hundred seventy-eight British (N = 88, M_age = 20.87 years, SD = 4.92, 73 Female) and Italian (N = 90, M_age = 22.37 years, SD = 2.06, 57 Female) participants were recruited to complete an online questionnaire. All were undergraduate students, who received course credit for their participation.

Materials

Social network use

Participants indicated which of the top 10 SNSs in Britain and Italy²⁹ they frequented, and how many hours per week they used each site.

Online risk exposure

To measure risky online self-presentation, we designed a risk exposure scale containing 19 items relating to potentially risky images or texts that individuals could post online, such as drug and alcohol use, sexual content, personal details, and offensive material. This scale was pilot tested in the United Kingdom and Italy, and any ambiguous items were reworded for clarity. Participants indicated whether they had engaged in these activities in the past by responding No (0), Don't Know (1), or Yes (2). If individuals responded “Don't Know” or “Yes” they were asked to state which SNSs these postings were on. A risk exposure score, engagement × number of SNSs, was calculated. The items were then categorized by five independent coders into four content areas: Alcohol/Drugs, Sexual, Personal, and Offensive Content Exposure (Cohen's κ = 0.84).

The values survey module

The values survey module³⁰ assessed cultural differences on six dimensions: Power Distance, Individualism versus Collectivism, Masculinity versus Femininity, Uncertainty Avoidance, Long- versus Short-Term Orientation, and Indulgence versus Restraint. The 24 items were scored on a 5-point scale (scored 1–5), and country scores on each dimension calculated using specific index formulae (see Ref.³⁰).

Self-presentation

The Psycho-Social Aspects of Facebook Use (PSAFU) Scale³¹ evaluates a range of psychological behaviors on Facebook. We utilized only the Self-Presentation subscale, which contained eight items. We tailored some items to represent social media use in general by removing reference to Facebook specifically. Participants responded on a 5-point scale (1 = It doesn't refer to me at all to 5 = It completely refers to me) and scores for the eight items were summed (α = 0.87).

The self-monitoring scale

The self-monitoring scale¹⁸ measured individuals' active control of their behavior and the way they presented themselves to others. Participants answered “True” or “False” to 18 statements. Each statement was predefined as requiring a specific response to reflect a high self-monitoring individual. As such, 10 statements were keyed as False and 8 statements were keyed as True. High self-monitors answered in the keyed direction (1), whereas low self-monitors answered in the opposite direction (0). Because the answer options on this scale were binary, we calculated the polychoric ordinal alpha (α = 0.80).³²

The Eysenck impulsivity inventory

Impulsiveness subscale asked participants to answer Yes (1) or No (0) to 19 items (α = 0.82).³³

Procedure

The questionnaire was first produced in English before being translated and back-translated from Italian to English. All participants provided consent before completing the questionnaire online.

Results

Descriptive analyses

British participants used significantly more SNSs, but did not spend more time on these sites each week compared with Italian participants (Table 1). Italians scored considerably higher on Masculinity and Uncertainty Avoidance, whereas the British showed a higher score for Long-term Orientation and Indulgence (Table 2).

Table 1.

Mean Scores (and Standard Deviation in Parenthesis) and Results of the Independent Samples t-Test for Risky Online Self-Disclosure in the Four Content Areas for the British and Italian Participants

Risky content type	British	Italian	t, df, p
Alcohol/drug content	7.00 (6.83)	3.82 (5.71)	3.35, 167.59, 0.001
Sexual content	7.65 (10.90)	5.08 (8.18)	1.77, 157.80, 0.079
Personal content	1.46 (2.60)	3.61 (3.95)	−0.425, 155.40, <0.001
Offensive content	12.99 (15.50)	24.93 (26.45)	−3.68, 144.84, <0.001

Table 2.

Country Scores for SubScales of the Values Survey Model for British and Italian Participants

VSM subscale	Britain	Italy
Power distance	31.99	22.42
Individualism	38.58	38.08
Masculinity	8.01	45.38
Uncertainty avoidance	13.81	38.57
Long-term orientation	40.44	8.07
Indulgence vs. restraint	72.52	62.64

VSM, values survey module.

A series of independent samples t-tests (Table 1) with risky self-presentation (alcohol/drug; sexual; personal; offensive) as the dependent variable and nationality (British; Italian) as the independent variable was conducted. British participants posted significantly more images/comments containing alcohol and drug content than Italian participants. Italian participants posted significantly more personal information and offensive content than British participants. There was no significant cultural difference for sexual content postings.

A series of independent samples t-tests (Table 3) revealed no significant cross-cultural difference on the self-monitoring scale. However, British participants scored significantly higher on Self-Presentation and marginally significantly higher on impulsiveness.

Table 3.

Mean Scores (and Standard Deviation in Parenthesis) and Independent Samples t-Test Results for Self-Monitoring, Psycho-Social Aspects of Facebook Use, and Impulsiveness for British and Italian Participants

Scale	British	Italian	t, df, p
Number of SNS used	5.30 (1.47)	4.24 (1.34)	4.98, 176, <0.001
Time weekly on SNS	15.72 (8.58)	13.96 (7.21)	0.15, 176, 0.140
Self-monitoring	9.78 (2.83)	9.01 (3.12)	1.77, 177, 0.079
PSAFU	26.70 (6.57)	20.77 (7.35)	5.72, 177, <0.001
Impulsiveness	8.21 (4.75)	6.87 (3.87)	1.94, 176, 0.054

PSAFU, Psycho-Social Aspects of Facebook Use; SNS, social networking site.

A full breakdown of correlations for each country can be seen in Tables 4 and 5. For both British and Italian participants, weekly time spent on SNSs was correlated with the number of SNSs used and with sexual content disclosure, and alcohol/drug content disclosure for the Italian participants. For both samples, posting offensive content was significantly related to posting risky content in the other three content areas. The impulsiveness scale was significantly positively correlated with alcohol/drug, personal information, and offensive content postings for the British participants, and with alcohol/drug postings and offensive content for the Italian participants. Scores for Self-Presentation were negatively significantly correlated with offensive content postings for the British sample.

Table 4.

Correlations Between Risky Self-Disclosure (Alcohol/Drugs, Sexual, Personal, Offensive), Self-Monitoring, Psycho-Social Aspects of Facebook Use, and Impulsiveness for the British Sample

	1	2	3	4	5	6	7	8
1. Weekly time on SNS
2. No. of SNSs	0.56^**
3. Alcohol/drug disclosure	0.17	0.13
4. Sexual disclosure	0.27^*	0.18	0.41^**
5. Personal disclosure	0.20	0.14	0.33^*	0.15
6. Offensive disclosure	0.05	−0.001	0.30^**	0.23^*	0.30^**
7. Self-presentation	0.13	0.20	−0.04	−0.05	0.01	−0.39^**
8. Self-monitoring	0.09	−0.02	0.31^**	0.23^*	0.30^**	0.01	0.18
9. Impulsivity	0.16	0.07	0.29^**	0.09	0.30^**	0.28^**	0.10	0.33^**

p < 0.05, ^**p < 0.01.

Table 5.

Correlations Between Risky Content Area Postings (Alcohol/Drugs, Sexual, Personal, Offensive), Self-Monitoring, Psycho-Social Aspects of Facebook Use, and Impulsiveness for the Italian Sample

	1	2	3	4	5	6	7	8
1. Weekly time on SNS
2. No. of SNSs	0.33^**
3. Alcohol/drug disclosure	0.26^*	0.001
4. Sexual disclosure	0.25^*	0.12	0.19
5. Personal disclosure	0.21	−0.01	0.05	0.21^*
6. Offensive disclosure	0.15	0.04	0.22^*	0.46^**	0.32^**
7. Self-presentation	0.10	0.17	0.09	0.03	0.19	0.20
8. Self-monitoring	0.07	−0.08	0.26^*	0.07	0.24^*	0.21^*	0.11
9. Impulsivity	0.20	0.07	0.27^*	0.07	0.17	0.24^*	0.14	0.21^*

p < 0.05, ^**p < 0.01.

Self-monitoring was significantly related to risky online postings in both samples. For British participants, significant correlations were found for alcohol/drug content, sexual content, and personal content. For Italian participants there was a significant relationship between self-monitoring and alcohol/drug content, personal content, and offensive content.

Predicting risky online self-disclosure

Generalized estimating equations were used to fit a Poisson regression with a natural log link function with risky online postings on SNSs as the dependent variable. The predictor variables were Nationality (Italy, U.K.), Risk Type (Alcohol/Drug Use, Sexual Content, Personal Information, Offensive Content), Self-Monitoring, Impulsiveness, Self-Presentation, and Weekly Time Spent Online. We included the predicted main effects of Impulsiveness, Self-Monitoring, and Nationality, as well as the predicted interaction effects of Impulsiveness × Self-Monitoring, Nationality × Impulsiveness, and Nationality × Self-Monitoring. Furthermore, we added Risk Type, Self-Presentation, and Weekly Time Spent Online as control variables. Since our descriptive analysis revealed country differences in risky online postings by risk type, we additionally entered the interactions of Nationality × Risk Type × Impulsiveness and Nationality × Risk Type × Self-Monitoring.

Table 6 displays the parameter estimates and the 95% Wald confidence intervals for all main and interaction effects. Impulsiveness (OR = 1.07, p = 0.04) and self-monitoring (OR = 1.16, p = 0.01) positively predicted risky online postings. Overall, Italians (OR = 1.54, p = 0.02) posted more risky content than U.K. participants. Those participants who spent more time online showed more risky online self-presentation (OR = 1.02, p = 0.02). Risky online postings differed by risk type; participants took significantly less risks when giving out personal information (OR = 0.51, p < 0.01), and significantly more risks when posting offensive content (OR = 3.86, p < 0.01). A three-way interaction also revealed that U.K. participants who scored higher in self-monitoring posted significantly less offensive content (OR = 1.18, p = 0.01).

Table 6.

Results of Regression Analysis Predicting Risky Online Self-Presentation

Predictors	B (standard error)	95% Wald confidence interval
Intercept	1.08 (0.21)^**	[0.67 to 1.48]
Nationality
United Kingdom	0	0
Italy	0.43 (0.18)^*	[0.07 to 0.79]
Impulsiveness	0.07 (0.03)^*	[0.004 to 0.14]
Self-monitoring	0.15 (0.06)^*	[0.03 to 0.27]
Self-presentation	−0.01 (0.01)	[−0.03 to 0.02]
Weekly time online	0.02 (0.01)^*	[0.003 to 0.03]
Risk type
Alcohol/drugs	0	0
Sexual content	0.21 (0.14)	[−0.06 to 0.47]
Personal information	−.68 (0.15)^**	[−0.98 to −0.39]
Offensive content	1.35 (0.13)^**	[1.10 to 1.61]
Nationality × impulsiveness
United Kingdom × impulsiveness	0	0
Italy × impulsiveness	−0.004 (0.04)	[−0.09 to 0.08]
Nationality × self-monitoring
United Kingdom × self-monitoring	0	0
Italy × self-monitoring	−0.07 (0.06)	[−0.19 to 0.06]
Impulsiveness × self-monitoring	−0.005 (0.01)	[−0.02 to 0.01]
Country × risk type × impulsiveness
United Kingdom × alcohol/drugs × impulsiveness	0
United Kingdom × sexual content × impulsiveness	−0.06 (0.04)	[−0.14 to 0.02]
United Kingdom × personal information × impulsiveness	−0.03 (0.04)	[−0.11 to 0.06]
United Kingdom × offensive content × impulsiveness	−0.01 (0.04)	[−0.08 to 0.06]
Italy × alcohol/drugs × impulsiveness	0
Italy × sexual content × impulsiveness	−0.05 (0.04)	[−0.13 to 0.04]
Italy × personal information × impulsiveness	−0.04 (0.04)	[−0.12 to 0.05]
Italy × offensive content × impulsiveness	−0.02 (0.04)	[−0.09 to 0.06]
Country × risk type × self-monitoring
United Kingdom × alcohol/drugs × self-monitoring	0
United Kingdom × sexual content × self-monitoring	0.04 (0.07)	[−0.12 to 0.18]
United Kingdom × personal information × self-monitoring	−0.05 (0.08)	[−0.20 to 0.11]
United Kingdom × offensive content × self-monitoring	−0.16 (0.06)^*	[−0.28 to −0.05]
Italy × alcohol/drugs × self-monitoring	0
Italy × sexual content × self-monitoring	−0.05 (0.04)	[−0.14 to 0.03]
Italy × personal information × self-monitoring	0.01 (0.06)	[−0.12 to 0.13]
Italy × offensive content × self-monitoring	−0.02 (0.05)	[−0.12 to 0.09]

Regression analysis predicting risky online self-presentation was modeled using generalized estimating equations assuming a Poisson distribution for the outcome.

p < 0.05, ^**p < 0.001.

Discussion

Social media use is pervasive among young adults,³⁴ yet with so much emphasis on maintaining a good online reputation, little is known about why some individuals post potentially negative or damaging comments and images. To address this important question, we investigated psychological factors, which may influence risky online activity, namely self-monitoring and impulsiveness. We expected that higher impulsiveness and self-monitoring would predict higher rates of risky online postings. Concordant with our predictions, and with earlier findings,^3,6 our data indicated that impulsiveness was predictive of online risky postings. This is very much in line with previous research on impulsivity and online (e.g., posting illegal content,³ problematic Internet use,^35,36 and Internet addiction³⁷) and offline risk taking (e.g., alcohol and drug use, smoking, risky sexual behavior³⁸). Our findings extend this research to the study of risky online self-presentation.

Self-monitoring was also positively predictive of risky online posting activities. Superficially, posting details of drug consumption or sexually provocative images may not appear appropriate when considering that an individual's post is visible to current and/or potential employers.¹⁶ However, people high in self-monitoring behave in what they perceive is a situationally appropriate way,¹⁸ and online identity is argued to be a product of the online social environment.¹⁶ Consequently, if individuals perceive risky postings as common, or the norm, on SNSs they may follow these normative expectations.⁷ Furthermore, people are often driven by the pleasure related to their self-disclosure (i.e., likes) despite (or maybe due to) the potential risks involved.³⁹ Some¹⁶ have argued that individuals present themselves on SNS in ways that are congruent with both the standards of the online spectators as well as the value that those spectators can bring to the individual. High self-monitoring SNS users often experience “audience segregation difficulties”⁴⁰; however, and are unable to effectively distinguish between groups of spectators and what is appropriate self-presentation. Thus, posts that may be highly inappropriate on a career networking site may seem situationally appropriate on Facebook, where this behavior may be the norm. Indeed, many Facebook users utilize provocative pictures to be noticeable on SNSs,¹⁶ or to gain positive attention from friends.⁴¹ Our high self-monitoring participants clearly used SNSs as platforms to self-present themselves as “cool,” where this behavior was valued and rewarded. Consequently, future research should more closely investigate how different risky posting behaviors are exhibited across different SNSs in relation to self-monitoring.

British participants scored higher on impulsiveness, lower on self-monitoring, lower on Uncertainty Avoidance, and slightly higher on Indulgence, compared with Italian participants. However, our data did not support our hypotheses that the processes underlying risky online posting (i.e., impulsiveness, self-monitoring) differed by country. Thus, we can cautiously conclude that the psychological processes affecting risky online behavior might be similar across culture. This would be in line with research on offline risk-taking, which has shown strong similarities in the factors influencing risk-taking across cultures.^42–44

Our data did, nonetheless, reveal differences by country for the types of risky self-presentation. U.K. participants were more likely to post images/comments of alcohol/drug use, whereas Italian participants posted personal information and offensive content. These findings could be attributed to the binge drinking culture in the United Kingdom⁴⁵ and by the Italian's high score on the Masculinity dimension of the values survey module (VSM)²⁷ which, coupled with low Uncertainty Avoidance, produce individuals who are highly passionate, emotional, and expressive of their opinions. As such, these social norms are expected to migrate to SNSs. However, the lack of differences between the U.K. and Italian participants in terms of what influences risky online self-presentation points to the pervasiveness of cyberculture⁴⁶ and the possibility that Internet cultures exact more influence than one's nationality.⁴⁷ This is certainly a promising area for future research.

There are some limitations to our findings. First, our samples were not representative of all British or Italian Internet users. Research with participants from other cultures could determine if there are more widespread cultural differences in risky online posting behavior. Additionally, the self-monitoring and impulsiveness scales were focused on offline behavior and, therefore, may not reflect how individuals regulate their behavior online. Since no online self-monitoring scale appears to exist this is a further area of potential future research.

What our results nicely reveal is that young people can behave both spontaneously and deliberately in their risky online postings on SNSs depending on the situation.^48,49 Furthermore, postings that may be viewed as impulsive (i.e., drug consumption), may turn out to represent deliberate choices that are driven by people's self-monitoring strategy. More deliberative risky decision making has been shown to result in higher rates of risk-taking in online situations by adolescents and young adults.^50,51 Our findings support these previous studies, highlighting that the deliberate consideration of risks and rewards can result in potentially negative outcomes. These important revelations about young adult's online self-presentation behavior have not previously been considered.

While young adults tend to focus less on being employable and are, therefore, less concerned about the potential future use of the information that can be harvested online,^5,28 many individuals come to regret previous online disclosures.^49,52 Consequently, further research will not only enable better understanding of this counterintuitive behavior, but also help to develop educational and technological strategies to enable young people to more appropriately manage their online self-presentation to avoid future regret and unfavorable consequences.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Yang

, Brown

. Online self-presentation on Facebook and self-development during the college transition. Journal of Youth and Adolescence, 2016; 45:402–416.

Moll

, Pieschl

, Broome

. Trust into collective privacy? The role of subjective theories for self-disclosure in online communication. Societies, 2014; 4:770–784.

Drouin

, Miller

. Why do people record and post illegal material? Excessive social media use, psychological disorder, or both?. Computers in Human Behavior, 2015; 48:608–614.

Morgan

, Snelson

, Elison-Bowers

. Image and video disclosure of substance use on social media websites. Computers in Human Behavior, 2010; 26:1405–1411.

Nosko

, Wood

, Molema

. All about me: disclosure in online social networking profiles: the case of Facebook. Computers in Human Behavior, 2010; 26:406–418.

Peluchette

, Karl

. Examining students' intended image on Facebook: “what were they thinking?!.”. Journal of Education for Business, 2010; 85:30–37.

Sarabia

, Estevez

. Sexualized behaviors on Facebook. Computers in Human Behavior, 2015; 61:219–226.

Haddon

, Livingstone

; the EU Kids Online Network. (2012). EU kids online: national perspectives. London, U.K.: EU Kids Online and the London School of Economics & Political Science.

Peluchette

, Karl

. Social networking profiles: an examination of student attitudes regarding use and appropriateness of content. Cyberpsychology & Behavior, 2008; 11:95–97.

10.

Beeger

. (2007). Soziale Netzwerke: Karrierekiller im Internet. www.stern.de/digital/online/soziale-netzwerke-karrierekiller-im-internet-3272102.html (accessed May 23, 2017)

11.

Cerasaro

. (2008). Employers Defy Privacy by Using Facebook. Tennessee Journalist. http://tnjn.org/2008/apr/08/employers-defy-privacy-by-usin/ (accessed May 23, 2017)

12.

Simpson

. (2015). Maryland prison officials fired for Facebook joke about being groped. Huffington Post. www.huffingtonpost.com/2015/02/20/maryland-prisons-official_n_6722366.html (accessed May 23, 2017)

13.

Shaw

. (2013). Paris Brown: kent PCC resigns after Twitter row. www.bbc.co.uk/news/uk-england-22083032 (accessed May 23, 2017)

14.

Kingkade

. (2015). Colorado college suspends student for six months over Yik Yak post. www.huffingtonpost.com/entry/colorado-college-yik-yak_us_56718ab1e4b0648fe301b019 (accessed May 23, 2017)

15.

Subrahmanyam

, Reich

, Waechter

, et al. Online and offline social networks: use of social networking sites by emerging adults. Journal of Applied Developmental Psychology, 2008; 29:420–433.

16.

Marder

, Joinson

, Shankar

, et al. Strength matters: self-presentation to the strongest audience rather than lowest common denominator when faced with multiple audiences in social network sites. Computers in Human Behavior, 2016; 61:56–62.

17.

Snyder

, Gangestad

. On the nature of self-monitoring: matters of assessment, matters of validity. Journal of Personality and Social Psychology, 1986; 51:125–139.

18.

Snyder

. (1987) Public appearances/private realities: the psychology of self-monitoring. New York: Freeman.

19.

Karl

, Peluchette

, Schlaegel

. Who's posting Facebook faux pas? A cross-cultural examination of personality differences. International Journal of Selection & Assessment, 2010; 18:174–186.

20.

Rosenberg

, Egbert

. Online impression management: personality traits and concerns for secondary goals as predictors of self-presentation tactics on Facebook. Journal of Computer-Mediated Communication, 2011; 17:1–18.

21.

Kobayashi

, Boase

. Tele-cocooning: mobile texting and social scope. Journal of Computer-Mediated Communication, 2014; 19:681–696.

22.

Brandtzæg

, Heim

. Why people use social networking sites. Lecture Notes in Computer Science, 2009; 5621:143–152.

23.

Jackson

, Wang

. Cultural differences in social network use: a comparative study of China and the United States. Computers in Human Behavior, 2013; 29:910–921.

24.

Al Omoush

KHS

, Yaseen

, Alma'aitah

. The impact of Arab cultural values on online social networking: the case of Facebook. Computers in Human Behavior, 2012; 28:2387–2399.

25.

Recabarren

, Nassbaum

, Leira

. Cultural divide and the internet. Computers in Human Behavior, 2008; 24:2917–2926.

26.

Hofstede

, Hofstede

, Minkov

. (2010) Cultures and organizations: software of the mind (Rev. 3rd ed.). New York: McGraw-Hill.

27.

Hofstede

. (2001) Culture's consequences: comparing values, behaviors, institutions and organizations across nations. Thousand Oaks, CA: Sage.

28.

Chau

PYK

, Cole

, Massey

, et al. (2002). Cultural differences in the online behaviour of consumers. Communications of the ACM, 2002; 45:138–143.

29.

Ten most searched for social networking sites by country (2015). www.digitalstrategyconsulting.com/intelligence/2013/12/top_10_most_searched_for_social_networks_by_country.php (accessed May 23, 2017)

30.

Hofstede

, Minkov

. (2013) Value Survey Module (VSM) Manual.

31.

Bodroza

, Jovanovic

. Validation of the new scale for measuring behaviours of Facebook users: Psycho-Social Aspects of Facebook Use (PSAFU). Computers in Human Behavior, 2016; 54:425–435.

32.

Gadermann

, Guhn

, Zumbo

. Estimating ordinal reliability for Likert-type and ordinal item response data: a conceptual, empirical, and practical guide. Practical Assessment Research & Evaluation, 2012; 17:1–13.

33.

Eysenck

SBG

, Pearson

, Easting

, et al. Age norms for impulsiveness, venturesomeness and empathy in adults. Personality and Individual Differences, 1985; 6:613–619.

34.

The demographics of social media users. (2015) www.pewinternet.org/2015/08/19/the-demographics-of-social-media-users/ (accessed May 23, 2017)

35.

Mottram

, Fleming

. Extraversion, Impulsivity and online group membership as predictors of problematic internet use. Cyberpsychology & Behavior, 2009; 12:319–321.

36.

Jeske

, Briggs

, Coventry

. Exploring the relationship between impulsivity and decision making on mobile devices. Personal & Ubiquitous Computing, 2016; 20:545–557.

37.

Zhang

, Mei

, Jingxin Chae

, et al. The relationship between impulsivity and internet addiction in Chinese college students: a moderated mediation analysis of meaning in life and self-esteem. PLoS One, 2015; 10:1–13.

38.

Zuckerman

, Kuhlman

. Personality and risk taking: common biosocial factors. Journal of Personality, 2000; 68:999–1029.

39.

Krasnova

, Kolesnikova

, Guenther

. “It won't happen to me!”: self-disclosure in online social networks. AMCIS 2009 Proceedings. Paper 343. http://aisel.aisnet.org/amcis2009/343 (accessed May 23, 2017)

40.

Leone

, Corte

. Concern for self-presentation and self-congruence: self-monitoring, Machiavellianism, and social conflicts. Social Behavior & Personality, 1994; 22:305–312.

41.

Petronio

. (2002) Boundaries of privacy: dialectics of disclosure. New York: State University of New York Press.

42.

Deardorff

, Gonzales

, Christopher

, et al. Early puberty and adolescent pregnancy: the influence of alcohol use. Paediatrics, 2005; 116:1451–1456.

43.

Kleop

, Guney

, Çok

, et al. Motives for risk taking in adolescence: a cross-cultural study. Journal of Adolescence, 2007; 32:135–151.

44.

Steinberg

. A social neurobiological perspective on adolescent risk taking. Developmental Review, 2008; 28:78–106.

45.

Measham

, Brain

. ‘Binge’ drinking, British alcohol policy and then new culture of intoxication. Crime, Media & Culture, 2005; 1:262–283.

46.

Bell

. (2007). Cyberculture theorists: manuel Castells & Donna Harraway. Abingdon, U.K.: Routledge.

47.

Macfayden

, Roche

, Doff

. (2004). Communication across cultures in cyberspace: a bibliographical review of intercultural communication online. LIT Verlag: Münster.

48.

Van Gool

, Van Ouytsel

, Ponnet

, et al. (2015). To share or not to share? Adolescent's self-disclosure about peer relationships on Facebook: an application of the Prototype Willingness Model. Computers in Human Behavior, 2015; 44:230–239.

49.

Wang

, Leon

, Chen

, et al. From Facebook regrets to Facebook privacy nudges. Ohio State Law Journal, 2013; 74:1307–1355.

50.

White

, Gummerum

, Hanoch

. Adolescents' and young adults' online risk taking: the role of gist and verbatim representations. Risk Analysis, 2015; 35:1407–1422.

51.

White

, Gummerum

, Hanoch

. Framing of online risk: young adults' and adolescents' representations of risky gambles. Decision, 2016; DOI: https://dx-doi-org.web.bisu.edu.cn/10.1037/dec0000066

52.

Dhir

, Kaur

, Chen

, et al. Understanding online regret experience in Facebook use—effects of brand participation, accessibility and problematic use. Computers in Human Behavior, 2016; 59:420–430.