CSIRO PUBLISHING
Sexual Health, 2014, 11, 332–339 http://dx.doi.org/10.1071/SH14032
Prevalence, correlates and attitudes towards sexting among young people in Melbourne, Australia Timothy H. Yeung A,B, Danielle R. Horyniak A,C, Alyce M. Vella A, Margaret E. Hellard A,C and Megan S. C. Lim A,C,D A
Burnet Institute, Centre for Population Health, 85 Commercial Road, Melbourne, Vic. 3004, Australia. The University of Melbourne, Royal Melbourne Hospital Department of Medicine, Royal Parade, Parkville, Vic. 3050, Australia. C Monash University, School of Public Health and Preventive Medicine, 99 Commercial Road, Melbourne, Vic. 3004, Australia. D Corresponding author. Email: [email protected] B
Abstract. Background: ‘Sexting’ is the exchange of sexually explicit material via communication technologies. Despite significant media attention, there has been little examination of sexting in the Australian setting. This study aimed to provide insight into sexting behaviours and attitudes among young Australians. Methods: A cross-sectional survey was conducted with a convenience sample of people aged 16–29 years attending a music festival (n = 1372). Correlates of lifetime sexting were determined using multivariate logistic regression. Attitudes towards and perceived consequences of sexting were explored in focus group discussions (FGDs) with 39 young people. Results: Forty percent of survey participants reported that they had ever sent or received a sext (48% of males, 36% of females), most commonly with a regular partner. Lower levels of education, greater recreational spending, greater number of sexual partners, inconsistent condom use with a regular partner, identifying as being nonheterosexual and risky alcohol consumption were all independent correlates of sexting. FGD participants made a clear distinction between consensual creating, sending and possessing of sexts, and nonconsensual sharing of sexts. Positive outcomes of consensual sexting included flirting and sexual experimentation, with sexting perceived as a normalised aspect of sexual interaction. Conclusions: Sexting is a common and normalised practice among young Australians. Our findings highlight the distinction in young people’s minds between consensual sexting and the nonconsensual sharing or circulation of sexts, which is not currently well recognised in sexuality education, the media or the law. Additional keywords: behaviours, communication, Internet, mobile phone, sexually explicit images, text messaging. Received 13 February 2014, accepted 2 July 2014, published online 4 August 2014
Introduction ‘Sexting’ is the transmission of sexually explicit material via mobile phones or online, a practice that has attracted concern from media and the public, due to the potentially severe legal consequences. Australian law states that creating, possessing or forwarding sexually explicit images of a person under the age of 18 is illegal and a child pornography offence under the 1958 Crimes Act. As an unintentional consequence of this law, sexting that involves an image of someone under the age of 18, whether or not the image is consensually produced, possessed or shared, is a crime, and those participating can be charged and registered as a sex offender. The legalities of sexting have been previously examined by researchers and policymakers, and legislative change has occurred in several jurisdictions. In contrast, there has been limited exploration of sexting as a social and health issue, although several notable Journal compilation CSIRO 2014
social outcomes of sexting have been identified. Young people report that the most serious negative outcome of sexting is the dissemination of images via mobile phones or online without the consent of the person depicted.1–6 This exposure of sexts to peers or the wider community can lead to social shame, stigma and harassment, which have been associated with mental health problems for young people, particularly in the high school setting.1,6–10 Furthermore, sexting has been linked to suicide in a small number of cases, which has been one of the main drivers of media attention to the practice.11–14 Internationally, several studies have explored the prevalence of sexting; however, most have been convenience studies conducted by media companies rather than academic peerreviewed studies and almost all are from the United States.4,5,8,15–19 Most have reported prevalences between 15% and 30% for both sending and receiving sexts. A review of www.publish.csiro.au/journals/sh
Sexting: prevalence, correlates and attitudes
sexting prevalence suggested that the wide range of results may be a reflection of the definitions of sexting used across studies and the varying age ranges of participants.19 There are only limited data on correlates of sexting, with most being reported in the grey literature only. The association between age and sexting is contentious, with some studies suggesting that ‘older’ young people are more likely to sext than ‘younger’ young people, but this association may be due to measuring any sexting event in a lifetime, thus allowing older people more time to sext.16,17,20 Females have been reported to be more likely than males to create and send sexually explicit material, and males are more likely to receive such material.1,3–6,16,20,21 Research has suggested these findings may reflect gender roles and expectations presented in public discourse.22,23 Studies have also found that alcohol, other recreational drugs and high-risk sexual behaviour are associated with sexting.24–27 Studies of young people have identified a range of motivations for engaging in sexting, including peer and partner pressure, sexual experimentation and development, revenge, attention seeking and humour.1,4,6,17,25 Despite growing public and legal attention, there have been few research studies investigating sexting as a social and health issue. To date, most studies have focussed on the negative outcomes and legal aspects of the practice, and there has been limited research into why young people sext and what they perceive to be the positive outcomes of doing so. Furthermore, no studies have examined sexting prevalence or correlates in the Australian setting. The aims of the present study were to measure the prevalence and correlates of sexting, and to explore attitudes towards sexting among young people in Australia. Methods Prevalence study Data on the prevalence and correlates of sexting were obtained from an annual cross-sectional study conducted at a music festival in Melbourne (described previously by Lim et al.28). In January 2013, festival attendees were recruited at a market stall within the festival grounds, where they either approached the stall themselves or were approached by researchers. Participants self-completed a paper questionnaire containing questions on sexual health, sexting behaviours (ever sexted, sexting partner type (regular partner, casual partner, friend and/ or stranger), sext format (text and/or image), alcohol and drug use, and sociodemographic information. As reimbursement, participants received a showbag containing sexual health, alcohol and drug information materials, condoms and snacks. The primary outcome of interest, sexting, was defined as having ever sent or received a sexually explicit picture or text message online or by mobile phone. Text sexting was included, as we believed it to be a potential avenue for flirtation among young people. Potential correlates of sexting included sociodemographic, sexual practices and risk behaviours, and alcohol and drug use. Sexual identity was dichotomised as heterosexual versus non-heterosexual (gay, bisexual, questioning or queer). ‘Binge drinking’ was defined as having consumed six or more alcoholic drinks in one session,
Sexual Health
333
at least weekly. Partner types were defined on the questionnaire as regular (boyfriend, girlfriend or in a relationship) or casual (all other partners). Higher risk sexual behaviour included inconsistent condom use with new or casual partners, or reporting multiple sexual partners in the past year. Sexually transmissible infection knowledge was assessed with a series of statements that could be answered as ‘true,’ ‘false’ or ‘don’t know’; ‘don’t know’ was considered incorrect. Univariate logistic regressions were conducted for all potential correlates. Variables significantly associated with sexting (P < 0.05) were subsequently included in a multivariate logistic regression model. Additionally, approximately half of the questionnaires (n = 597) contained an additional mental health module. This variable was added to a multivariate model in a separate analysis so as not to reduce the sample size of the overall model. Analysis was conducted using Stata ver. 11 (StataCorp, College Station, TX, USA). Focus group discussions Focus group discussions (FGDs) explored the attitudes and opinions of sexting among young people. Participants were aged between 16 and 25 years, and were recruited from the prevalence survey participant pool, through advertising on social media and university websites, and via purposive snowball recruitment. In total, 39 participants (21 females, 18 males) participated in FGDs, which were segregated by sex or age: female only, all ages (FG1); mixed sex, all ages (FG2); male only, all ages (FG3); and mixed sex, ages 16–18 years (FG4). FGDs were semistructured and covered key topics including what is considered a sext, how sexting is spoken about, gender differences in sexting practices, and the positive and negative outcomes of sexting. Key topics for inclusion were informed by previous research and consultation with experts in the field. FGDs lasted ~90 min and participants were reimbursed $30 for their time. Discussions were audio recorded and transcribed verbatim. FGD transcripts were managed using Nvivo ver. 10 (QSR International, Doncaster, Vic., Australia) and were analysed thematically. Both studies were approved by the Alfred Hospital Human Research Ethics Committee. Results Prevalence and correlates of sexting Of the 1403 participants who completed the questionnaire, 31 were excluded from the analysis because of missing data. In the final sample (n = 1372), 65% were female and the median age was 19 years. Characteristics of the sample are shown in Table 1. Overall, 552 (40%) reported ever sexting (48% of males, 36% of females). Of those who reported ever sexting, sexting with a regular partner was the most common context (67%; Table 2). Text sexting was found to be more common than image sexting with all partner types, except strangers. In multivariate regression, sexting was independently associated with being male, having lower levels of education, greater recreational spending, a greater lifetime number of sexual partners, inconsistent condom use with a regular partner, identifying as being nonheterosexual and excessive alcohol consumption resulting in either injury to self or others or regular memory loss (Table 3). Reporting having
334
Sexual Health
T. H. Yeung et al.
Table 1. Sociodemographic, sexual behaviour, and alcohol and drug-related characteristics of study participants by sex Data show the number of participants in each group, with percentages in parentheses. Percentage values for each variable do not always add up to 100% because of a small number (
Sexual Health, 2014, 11, 332–339 http://dx.doi.org/10.1071/SH14032
Prevalence, correlates and attitudes towards sexting among young people in Melbourne, Australia Timothy H. Yeung A,B, Danielle R. Horyniak A,C, Alyce M. Vella A, Margaret E. Hellard A,C and Megan S. C. Lim A,C,D A
Burnet Institute, Centre for Population Health, 85 Commercial Road, Melbourne, Vic. 3004, Australia. The University of Melbourne, Royal Melbourne Hospital Department of Medicine, Royal Parade, Parkville, Vic. 3050, Australia. C Monash University, School of Public Health and Preventive Medicine, 99 Commercial Road, Melbourne, Vic. 3004, Australia. D Corresponding author. Email: [email protected] B
Abstract. Background: ‘Sexting’ is the exchange of sexually explicit material via communication technologies. Despite significant media attention, there has been little examination of sexting in the Australian setting. This study aimed to provide insight into sexting behaviours and attitudes among young Australians. Methods: A cross-sectional survey was conducted with a convenience sample of people aged 16–29 years attending a music festival (n = 1372). Correlates of lifetime sexting were determined using multivariate logistic regression. Attitudes towards and perceived consequences of sexting were explored in focus group discussions (FGDs) with 39 young people. Results: Forty percent of survey participants reported that they had ever sent or received a sext (48% of males, 36% of females), most commonly with a regular partner. Lower levels of education, greater recreational spending, greater number of sexual partners, inconsistent condom use with a regular partner, identifying as being nonheterosexual and risky alcohol consumption were all independent correlates of sexting. FGD participants made a clear distinction between consensual creating, sending and possessing of sexts, and nonconsensual sharing of sexts. Positive outcomes of consensual sexting included flirting and sexual experimentation, with sexting perceived as a normalised aspect of sexual interaction. Conclusions: Sexting is a common and normalised practice among young Australians. Our findings highlight the distinction in young people’s minds between consensual sexting and the nonconsensual sharing or circulation of sexts, which is not currently well recognised in sexuality education, the media or the law. Additional keywords: behaviours, communication, Internet, mobile phone, sexually explicit images, text messaging. Received 13 February 2014, accepted 2 July 2014, published online 4 August 2014
Introduction ‘Sexting’ is the transmission of sexually explicit material via mobile phones or online, a practice that has attracted concern from media and the public, due to the potentially severe legal consequences. Australian law states that creating, possessing or forwarding sexually explicit images of a person under the age of 18 is illegal and a child pornography offence under the 1958 Crimes Act. As an unintentional consequence of this law, sexting that involves an image of someone under the age of 18, whether or not the image is consensually produced, possessed or shared, is a crime, and those participating can be charged and registered as a sex offender. The legalities of sexting have been previously examined by researchers and policymakers, and legislative change has occurred in several jurisdictions. In contrast, there has been limited exploration of sexting as a social and health issue, although several notable Journal compilation CSIRO 2014
social outcomes of sexting have been identified. Young people report that the most serious negative outcome of sexting is the dissemination of images via mobile phones or online without the consent of the person depicted.1–6 This exposure of sexts to peers or the wider community can lead to social shame, stigma and harassment, which have been associated with mental health problems for young people, particularly in the high school setting.1,6–10 Furthermore, sexting has been linked to suicide in a small number of cases, which has been one of the main drivers of media attention to the practice.11–14 Internationally, several studies have explored the prevalence of sexting; however, most have been convenience studies conducted by media companies rather than academic peerreviewed studies and almost all are from the United States.4,5,8,15–19 Most have reported prevalences between 15% and 30% for both sending and receiving sexts. A review of www.publish.csiro.au/journals/sh
Sexting: prevalence, correlates and attitudes
sexting prevalence suggested that the wide range of results may be a reflection of the definitions of sexting used across studies and the varying age ranges of participants.19 There are only limited data on correlates of sexting, with most being reported in the grey literature only. The association between age and sexting is contentious, with some studies suggesting that ‘older’ young people are more likely to sext than ‘younger’ young people, but this association may be due to measuring any sexting event in a lifetime, thus allowing older people more time to sext.16,17,20 Females have been reported to be more likely than males to create and send sexually explicit material, and males are more likely to receive such material.1,3–6,16,20,21 Research has suggested these findings may reflect gender roles and expectations presented in public discourse.22,23 Studies have also found that alcohol, other recreational drugs and high-risk sexual behaviour are associated with sexting.24–27 Studies of young people have identified a range of motivations for engaging in sexting, including peer and partner pressure, sexual experimentation and development, revenge, attention seeking and humour.1,4,6,17,25 Despite growing public and legal attention, there have been few research studies investigating sexting as a social and health issue. To date, most studies have focussed on the negative outcomes and legal aspects of the practice, and there has been limited research into why young people sext and what they perceive to be the positive outcomes of doing so. Furthermore, no studies have examined sexting prevalence or correlates in the Australian setting. The aims of the present study were to measure the prevalence and correlates of sexting, and to explore attitudes towards sexting among young people in Australia. Methods Prevalence study Data on the prevalence and correlates of sexting were obtained from an annual cross-sectional study conducted at a music festival in Melbourne (described previously by Lim et al.28). In January 2013, festival attendees were recruited at a market stall within the festival grounds, where they either approached the stall themselves or were approached by researchers. Participants self-completed a paper questionnaire containing questions on sexual health, sexting behaviours (ever sexted, sexting partner type (regular partner, casual partner, friend and/ or stranger), sext format (text and/or image), alcohol and drug use, and sociodemographic information. As reimbursement, participants received a showbag containing sexual health, alcohol and drug information materials, condoms and snacks. The primary outcome of interest, sexting, was defined as having ever sent or received a sexually explicit picture or text message online or by mobile phone. Text sexting was included, as we believed it to be a potential avenue for flirtation among young people. Potential correlates of sexting included sociodemographic, sexual practices and risk behaviours, and alcohol and drug use. Sexual identity was dichotomised as heterosexual versus non-heterosexual (gay, bisexual, questioning or queer). ‘Binge drinking’ was defined as having consumed six or more alcoholic drinks in one session,
Sexual Health
333
at least weekly. Partner types were defined on the questionnaire as regular (boyfriend, girlfriend or in a relationship) or casual (all other partners). Higher risk sexual behaviour included inconsistent condom use with new or casual partners, or reporting multiple sexual partners in the past year. Sexually transmissible infection knowledge was assessed with a series of statements that could be answered as ‘true,’ ‘false’ or ‘don’t know’; ‘don’t know’ was considered incorrect. Univariate logistic regressions were conducted for all potential correlates. Variables significantly associated with sexting (P < 0.05) were subsequently included in a multivariate logistic regression model. Additionally, approximately half of the questionnaires (n = 597) contained an additional mental health module. This variable was added to a multivariate model in a separate analysis so as not to reduce the sample size of the overall model. Analysis was conducted using Stata ver. 11 (StataCorp, College Station, TX, USA). Focus group discussions Focus group discussions (FGDs) explored the attitudes and opinions of sexting among young people. Participants were aged between 16 and 25 years, and were recruited from the prevalence survey participant pool, through advertising on social media and university websites, and via purposive snowball recruitment. In total, 39 participants (21 females, 18 males) participated in FGDs, which were segregated by sex or age: female only, all ages (FG1); mixed sex, all ages (FG2); male only, all ages (FG3); and mixed sex, ages 16–18 years (FG4). FGDs were semistructured and covered key topics including what is considered a sext, how sexting is spoken about, gender differences in sexting practices, and the positive and negative outcomes of sexting. Key topics for inclusion were informed by previous research and consultation with experts in the field. FGDs lasted ~90 min and participants were reimbursed $30 for their time. Discussions were audio recorded and transcribed verbatim. FGD transcripts were managed using Nvivo ver. 10 (QSR International, Doncaster, Vic., Australia) and were analysed thematically. Both studies were approved by the Alfred Hospital Human Research Ethics Committee. Results Prevalence and correlates of sexting Of the 1403 participants who completed the questionnaire, 31 were excluded from the analysis because of missing data. In the final sample (n = 1372), 65% were female and the median age was 19 years. Characteristics of the sample are shown in Table 1. Overall, 552 (40%) reported ever sexting (48% of males, 36% of females). Of those who reported ever sexting, sexting with a regular partner was the most common context (67%; Table 2). Text sexting was found to be more common than image sexting with all partner types, except strangers. In multivariate regression, sexting was independently associated with being male, having lower levels of education, greater recreational spending, a greater lifetime number of sexual partners, inconsistent condom use with a regular partner, identifying as being nonheterosexual and excessive alcohol consumption resulting in either injury to self or others or regular memory loss (Table 3). Reporting having
334
Sexual Health
T. H. Yeung et al.
Table 1. Sociodemographic, sexual behaviour, and alcohol and drug-related characteristics of study participants by sex Data show the number of participants in each group, with percentages in parentheses. Percentage values for each variable do not always add up to 100% because of a small number (
