ارائه در فیس بوک و خطر قربانی مزاحمت سایبری
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|30402||2014||7 صفحه PDF||سفارش دهید||5572 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Computers in Human Behavior, Volume 40, November 2014, Pages 16–22
Facebook is an environment in which adolescents can experiment with self-presentation. Unfortunately, Facebook can also be an environment in which cyberbullying occurs. The aim of the current study was to investigate whether specific self-presentation behaviours in Facebook were associated with cyberbullying victimisation for adolescents. The contents of 147 adolescent (15–24 years) Facebook profile pages were recorded and used to predict cyberbullying victimisation. Coded contents included the presence or absence of Facebook profile features (e.g., relationship status) and the specific content of certain features (e.g., type and valence of wall posts). Participants completed measures of cyberbullying victimisation and traditional bullying victimisation and perpetration. More than three out of four participants reported experiencing at least one victimisation experience on Facebook in the preceding 6 months. A series of Facebook features and experiences of traditional bullying victimisation/perpetration were found to be associated with an increased risk of cyberbullying victimisation. Number of Facebook friends and traditional bullying victimisation were also significant predictors of cyberbullying victimisation. These results support the hypothesis that self-presentation on Facebook can increase the likelihood of eliciting negative attention from potential perpetrators. This has important implications for the development of cyberbullying prevention and education programs that teach adolescents about measures they may take to decrease their risk for cyberbullying victimisation within social networking sites like Facebook.
The use of social networking sites (SNS) such as Facebook, Twitter and Instagram is prolific amongst young people (Duggan and Smith, 2004 and Madden et al., 2013). Self-presentation is a central feature of SNS because their interface is based around the creation of visible personal profiles that display a friends list, personal information, and photos. Unfortunately, SNS have also become environments in which users can target and harass other users. This phenomenon is typically called cyberbullying (Smith et al., 2008). Consequently, the associations between the ways in which young SNS users manage their online self-presentation and risk of cyberbullying, has recently begun to attract the interest of researchers. Cyberbullying has been defined in the research literature as “an aggressive, intentional act carried out by a group or individual, using electronic forms of contact, repeatedly and over time against a victim who cannot easily defend him or herself” (Smith et al., 2008, p. 376). Published reports of cyberbullying prevalence rates in teens (generally below 18 years of age) have ranged from 6% to 30% (Sabella, Patchin, & Hinduja, 2013) and victimisation experiences have been associated with multiple emotional, cognitive and behavioural impacts such as social anxiety (Dempsey, Sulkowski, Nichols, & Storch, 2009), poor concentration (Beran & Li, 2005), suicidal thoughts and behaviours (Hinduja & Patchin, 2010), and lower school grades and poor school attendance (Price & Dalgleish, 2010). Considering the associated negative outcomes, it is important to identify factors that influence the risk of cyberbullying victimisation. Victimisation has been defined as an individual’s “self-perception of having been exposed, either momentarily or repeatedly, to aggressive actions emanating from one or more other persons” (Aquino & Bradfield, 2000, p. 172). There are multiple factors that may influence the risk of victimisation. Victimologists have suggested that these may include perpetrator characteristics, environmental factors, or victim behaviour (Elias, 1986). Identifying the role that victim behaviour may play in the likelihood of being targeted by others, as suggested by the victim precipitation model (Timmer & Norman, 1984), is as important as focusing on perpetrator and environmental factors. According to the victim precipitation model, victim behaviour may, whether intentionally or unintentionally, elicit a response in perpetrators that leads to victimisation (Kim & Glomb, 2010). It is important to note that this perspective does not blame the victim for the victimisation; rather the model identifies behavioural factors that are related to an increased risk of being targeted. The victim precipitation model has been used extensively within the criminal victimology literature (Aquino & Byron, 2002) and has been applied empirically in studies investigating the role of personality characteristics (Coyne, Seigne, & Randall, 2000), conflict management style (Aquino & Bradfield, 2000), and other organisational variables (Aquino & Thau, 2009) on risk of workplace victimisation. Therefore the victim precipitation model may also provide a framework for the study of victim-specific risk factors that increase the likelihood of being cyberbullied. To date, research investigating factors that influence the risk of cyberbullying victimisation has focused on individual differences of young information and communications technology (ICT) users. Conflicting results regarding the role of gender as a predictor of victimisation have been reported. While some studies have found no significant difference between males and females (e.g., Patchin and Hinduja, 2006 and Slonje and Smith, 2008), other studies have found that females are more at risk than males (e.g., Li, 2007 and Wang et al., 2009). Conflicting results have also been found regarding the relationship between age and victimisation with some studies finding no relationship (e.g., Patchin and Hinduja, 2006 and Smith et al., 2008), and others a positive (Kowalski & Limber, 2007) or negative relationship (Slonje & Smith, 2008). Research has also focused on the relationship between the risk of cyberbullying victimisation in young people and the extent and nature of internet and computer use. For example, time spent online and computer proficiency were significant positive predictors of victimisation among participants under 18 years of age (Hinduja & Patchin, 2008). It has also been shown that likelihood of being a cyberbullying victim was higher for those who (1) were more dependent on the internet (e.g., would surf on the internet at the expense of other activities; Vandebosch & Cleemput, 2008), (2) were more likely to chat with older online acquaintances (Walrave & Heirman, 2011), or (3) who gave passwords to others and shared personal information on a blog (Walrave & Heirman, 2011). Other studies have found a relationship between being a cyberbullying victim and being a traditional bullying victim or perpetrator in samples of young people. Cyberbullying victims (12–18 years old) have been found to be more than six and a half times more likely to have been a cyberbullying perpetrator (Walrave & Heirman, 2011) and more than two and a half times more likely to be a traditional bullying victim (under 18 years; Hinduja & Patchin, 2008). Results from other studies have confirmed the strong relationship between both cyber and traditional bullying victimisation in children and adolescent samples (e.g., Juvonen and Gross, 2008, Li, 2007, Twyman et al., 2010 and Vandebosch and Cleemput, 2009). One issue regarding previous research on cyberbullying victimisation risk factors is that samples have been recruited from different populations (e.g., under 18 years old, 12–15 years, middle school students only) which makes cross study comparisons of risk factors and prevalence rates difficult. More recently, the role of ‘risky SNS practices’ in online risk was investigated in 9–16 year olds (Staksrud, Olafsson, & Livingstone, 2013). Participants were asked to report the time they spent online daily, how much they knew about the internet (digital competence), whether their SNS profile was set to public/private, whether they had more than 100 SNS contacts, and whether they included specific personal information on their profiles (e.g., last name, address, phone number, school, and correct age). Cyberbullying was measured dichotomously (yes/no) in the last 12 months. Results showed that overall, 8% of participants who use SNS had experienced cyberbullying, while 10% of participants who use SNS and have more than 100 friends had experienced cyberbullying. Those with public SNS profiles and those who displayed their mobile phone number or address on SNS were also more likely to be cyberbullied. However, these differences were not statistically significant. These results support the victim precipitation model in that self-presentation behaviours account for some degree of the risk in cyberbullying victimisation. While the results are interesting, this study relied on participants’ self-report of SNS behaviours, which is subject to potential memory and self-presentation biases. In an effort to avoid these problems, researchers who investigate self-presentation behaviour in SNS directly view and code users’ profile pages. Numerous studies have implemented this approach (e.g., Boyle and Johnson, 2010, Mehdizadeh, 2010 and Zhao et al., 2008). The current study extended the Staksrud et al. (2013) study that investigated the role of only a small selection of self-presentation behaviours in SNS as predictors of cyberbullying victimisation, by coding each profile page feature and the content of specific features. This study also focused on risk in adolescence as this period is considered to be critical in the development of a personal, individuated identity (Erikson, 1968). Furthermore, how adolescents choose to present in SNS may be a key part of identity development (Gonzales & Hancock, 2011). The current study was exploratory due to a lack of previous related research. The main objective was to understand the victim related factors that increase the risk of cyberbullying victimisation so that successful interventions for the prevention of cyberbullying victimisation can be developed and safer SNS environments can be constructed. More specifically, this study aimed to determine the frequency that cyberbullying victimisation occurred in Facebook in the preceding 6 months and what specific features of a Facebook profile page, that when used or used in a certain way, were associated with an increased risk of cyberbullying victimisation in adolescents.
نتیجه گیری انگلیسی
3. Results 3.1. Frequency of cyberbullying victimisation In the preceding 6 months, 51% reported having experienced more than one of the 14 behaviours (M = 2.54, SD = 2.89), 25.9% reported having experienced one of the 14 behaviours, and 23.1% of participants reported that they had not experienced any of the SNS victimisation behaviours. Table 2 shows the observed frequencies of each target behaviour. The most prevalent reported behaviour was deliberate blocking of participants (“defriending”) from a social networking site. Participants also indicated the frequency with which they had experienced each behaviour in the preceding 6 months. A total cyberbullying victimisation score was computed by summing the total number of behaviours experienced weighted by their frequency. Therefore, total cyberbullying victimisation could range from 0 to 84. The observed range in the sample was 0–20 (M = 3.09, SD = 3.90) indicating that few participants had experienced multiple behaviours at high frequencies. No significant relationships between total cyberbullying victimisation and age, gender, or daily SNS use were found. 3.2. Relationships between Facebook features and cyberbullying victimisation Due to positive skewness in most of the variables of interest, Spearman correlations between continuous Facebook features and total cyberbullying victimisation in the preceding 6 months were calculated. Point-biserial correlations were conducted between dichotomous (yes/no) Facebook features and total cyberbullying victimisation. Results are provided in Table 3 and Table 4. Following other Facebook users, the number of days until first wall post, the number of negative wall posts, and traditional bullying victimisation and perpetration were significantly associated with cyberbullying victimisation. Table 3. Correlations between Facebook Features and Cyberbullying Victimisation. Variable N (%) r Cover photo 138 (93.9) −.01 Profile picture 147 (100) School 123 (83.7) −.06 Universitya 99 (79.8) .13 Employment 88 (59.9) .12 Current city 123 (83.7) .20 Relationship status 108 (73.5) .12 Family 125 (85) .04 About me 54 (36.7) −.01 Gender 121 (82.3) −.04 Interested in 70 (47.6) .01 Languages 28 (19) .10 Religion 46 (31.3) .05 Political views 18 (12.2) .15 Email 0 (0) Mobile number 14 (9.5) −.01 Other phone number 1(0.7) −.07 Instant messenger 11 (7.5) .10 Address 9 6.1) −.06 Website 5 (3.4) .15 Networks 17 (11.6) −.10 Quotes 40 (27.2) −.02 Following others 40 (27.2) .20⁎ ∗∗ p < .05. a Those under 18 years old were not included as they could not be at university. ⁎ p < .01. Table options Table 4. Correlations between content of Facebook features, related concepts, and cyberbullying victimisation. Variable M (SD) r Number of friends 511.00 (237.37 .16 Number of photos 305.00 (297.80) .01 Number of check-ins 103.00 (107.08) .1 Number of liked pages 517.00 (700.87) .15 Number of notes .78 (4.69) .09 Number of days to post 10 WPs 170.00 (210) −.14 Number of days to first WP 19.86 (40.12) −.17⁎ Number of status updates (WP) 3.67 (2.58) .09 Number of check-ins (WP) .67 (1.06) .04 Number of check-ins with photo (WP) .31 (.95) −.02 Number of links (WP) .87 (1.56) .09 Number of videos (WP) .27 (.72) .01 Number of shared photos (WP) 4.09 (2.55) −.12 Number of negative WP 1.05 (1.49) .30⁎⁎ Number of positive WP 5.24 (2.25) −.12 Number of neutral WP 3.60 (2.16) −.08 Traditional bullying perpetration 24.43 (9.25) .27⁎⁎ Traditional bullying victimisation 29.99 (13.05) .38⁎⁎ WP = wall post. ⁎ p < .01. ⁎⁎ p < .05. Table options The relationship between coded features with more than two possible values (e.g. type of relationship status and the content of profile/cover photos) and cyberbullying victimisation were examined through a series of one-way between groups analyses of variance (ANOVA). Cyberbullying victimisation in the preceding 6 months was significantly related to type of relationship status, F(3, 143) = 3.78, p = .012, η2 = .073. Post hoc analyses with Tukey’s HSD (with α = .05) revealed that those who stated that their relationship status was ‘married’ reported significantly more cyberbullying victimisation (M = 20.80, SD = 5.96) than those who did not report their relationship status (M = 16.31, SD = 2.95), those who reported they were ‘single’ (M = 16.88, SD = 4.24), and those who reported they were ‘in a relationship’ (M = 17.12, SD = 3.51), with no significant differences between the latter three groups. 3.3. Logistic regression analyses for cyberbullying victimisation Because the total cyberbullying victimisation variable was positively skewed, it was dichotomised. Those who reported no experience with any of the 14 victimisation behaviours were given a score of zero and those who reported experiencing one or more of the victimisation behaviours were given a score of one. Point-biserial correlations were calculated to test the relationship between continuous Facebook features, APRI scores and the dichotomous cyberbullying victimisation variable. Phi-coefficients were calculated for dichotomous (yes/no) Facebook variables. Number of friends was standardised because there was a large disparity across participants (Range = 87–1064; M = 511, SD = 237.37). Variables with significant associations were included in a backward stepwise logistic regression to examine the risk factors for cyberbullying victimisation. These variables included number of friends, traditional bullying victimisation, and whether or not the profile owner reported their city of residence. Table 5 shows the results of this analysis. The final model included two significant predictor variables. Each increase in one standard deviation of number of friends was associated with nearly a twofold increase in risk of cyberbullying victimisation. Traditional bullying victimisation was associated with an 11% increase in cyberbullying victimisation risk. Table 5. Stepwise logistic regression model of an adolescent experiencing a negative behaviour on Facebook. B SE Wald Sig. Odds ratio 95% Confidence intervals for EXP(B) Lower Upper Step 1 Number of friends .60 .28 4.61 .03⁎ 1.82 1.05 3.16 Traditional victimisation .10 .04 6.66 .01⁎⁎ 1.10 1.02 1.19 Current city −.85 .52 2.65 .10 .43 .16 1.19 Constant −1.12 .97 1.35 .25 .33 Step 4 Number of friends .67 .28 5.94 .02⁎ 1.96 1.14 3.34 Traditional victimisation .10 .04 6.65 .01⁎⁎ 1.11 1.02 1.19 Constant −1.35 .98 1.91 .17 .26 Note. R2 = .19 (Hosmer & Lemeshow), .11 (Cox & Snell), .17 (Nagelkerke). Model χ2(8) = 11.21, p = .19. df = 1 for each predictor. ⁎ p < .05. ⁎⁎ p < .01.