2.1. Participants
Participants were N = 336 individuals who identified as female, aged 14-19 years old (Mage =15.98 SD = 1.27), and were recruited from the general community and the university of the research team. All participants outside of the age bracket, who did not identify as female, or those that did not provide explicit consent, were excluded from the study. The majority of participants were attending secondary school (85.4%) at the time of completing the survey.
2.2. Materials
2.2.1. Cyberbullying and Cybervictimisation
The Berlin Cyberbullying-Cybervictimisation Questionnaire (BCyQ; Schultze-Krumbholz & Scheithauer, 2011) was used to assess experiences of cyberbullying (20 items) and cybervictimisation (19 items). The BCyQ has been validated and has good psychometric properties in young people aged 9-17 years. Participants were asked if they had experienced a list of behaviours over the previous six months, as well as if they had acted in that way. The scale ranged from 1 (has not happened to me at all) to 5 (several times a week). Combined scores were then created to allow a total score for both cyberbullying (score range = 18–90) and cybervictimisation (score range = 17–85), with higher scores indicating greater frequency of either bullying or victimisation. These scales were shown to have good internal consistency in this study (cyberbullying α = 0.883; cybervictimisation α = 0.94).
2.2.2. Appearance-related cyberbullying (ARC)
Due to the limited research on ARC, there is a scarcity of tailored questionnaires that adequately measure its dimensions. Thus, this study adopted an approach to examine the various dimensions of ARC through three distinct measures, based on the frameworks established in previous research (Berne et al., 2014; Frisén & Berne, 2020; Wang & Ding, 2023).
2.2.2.1. ARC roles. The first measure of ARC utilised six items from the BCyQ to evaluate the different roles in ARC dynamics. From the BCyQ's original set of 17 victimisation and 18 bullying items, we carefully selected and adapted three items from each measure based on their applicability to ARC. This selection process facilitated the categorisation of participants as either ARC-bullies, ARC-victims, ARC-bully-victims, or those with no experience of ARC. A key modification in this adaptation was the integration of the phrase "my appearance or body" into the text of the six chosen statements, thereby tailoring the measure to focus specifically on appearance-related aspects of cyberbullying. For example, the BCyQ statement 'Others spread embarrassing, insulting, or humiliating video clips/photos of me without my permission on the Internet or by mobile phone' was revised to 'Others spread embarrassing, insulting, or humiliating video clips/photos of my appearance or body without my permission on the Internet or by mobile phone.' The instructions and response scales are the same as the BCyQ. Participants were asked if they had experienced a list of behaviours over the previous six-months, as well as if they had acted in that way. The 5-point Likert scales ranged from 1 (has not happened to me at all) to 5 (several times a week). Combined scores were then created to allow a total score (analogous to the scoring methodology of the BCyQ) for ARC-bullies (score range = 4–15) and ARC-victims (score range = 4–15) or ARC-bully-victims (score range = 4–15 on both victim and bully scale), with higher scores indicating greater frequency of ARC. Cronbach's alpha for this category was calculated at .76, indicating acceptable internal consistency.
2.2.2.2. Types of ARC experienced. The second measure comprised ten items formulated by the research team to delineate specific types of ARC. Drawing from established body image, eating disorder, and appearance-related questionnaires (Duarte et al., 2015; Kling et al., 2019; Welch et al., 2012) these items explored several aspects of appearance, including body shape, body size, specific body features (e.g., breasts), disability, changes to appearance (e.g., scars), facial features, clothing or style, skin colour, age, and body tone or muscularity. Participants were asked if they had ever experienced these types of ARC. An example question is, “Have you ever been made fun of or teased online because of your body shape?” The 5-point Likert scales ranged from 1 (has not happened to me at all) to 5 (several times a week). The internal consistency for this measurement was robust, with a Cronbach's alpha of .92.
2.2.2.3. Impact of ARC on the desire to change physical appearance. The third measure aimed to investigate the emotional and psychological impact of ARC. Informed by research that suggests an association between experiences of appearance-related bullying and the desire to change physical appearance (Fardouly & Vartanian, 2015), three items were utilised to evaluate the extent to which individuals desired to alter their physical appearance due to their experience with ARC. Participants were asked whether and how often they had felt a certain way about their appearance because of their experience with ARC. An example question is, “Have you ever felt the need to change your appearance through cosmetic procedures (e.g., nose job, boob job) because of appearance-related cyberbullying?” The 5-point Likert scale ranged from 1 (never) to 5 (frequently). The internal consistency of this category was evidenced by a Cronbach's alpha of .86.
2.2.3. Body esteem
The Body Esteem Scale for Adolescents and Adults (BESAA; Mendelson et al. (2001) is a 23-item tool designed to assess an individual's cognitive, emotional, and behavioural responses to their body image, providing insights into their self-perception and body esteem. The BESAA has three subscales: BE-Appearance (general feelings about one's appearance; e.g., “I like what I see when I look in the mirror”) (ten items), BE-Weight (satisfaction with one's weight; e.g., “I really like what I weigh”) (eight items) and BE-Attribution (evaluations attributed to others about one's body and appearance (e.g., “People my own age like my looks”) (five items). Using a five-point a five-point Likert-scale ranging from 0 (never) to 4 (always), participants were asked to indicate the degree to which they agreed with each statement on. The BESAA is a widely used measure in research and clinical settings and was shown to have good internal consistency in this study (α = 0.94) and in past research (Kling et al., 2019).
2.2.4. Body shame
The Body Image Shame Scale (BISS; Duarte et al. (2015) was used to measure experiences of body image shame. The BISS comprises a two-factor structure that assesses both externalised and internalised dimensions of body shame. The externalised dimension consists of seven items that assess judgments of being negatively evaluated or criticised by others based on one's physical appearance, such as feeling uncomfortable in social situations due to fear of criticism (e.g., " The relationship I have with my physical appearance makes it difficult for me to feel comfortable in social situations"). The internalised dimension includes seven items that focus on negative self-evaluations based on one's physical appearance, such as feeling like a defective person when seeing one's body in the mirror (e.g., "I choose clothes that hide parts of my body that I consider ugly or disproportional"). Additionally, a combined score for overall body shame was calculated. Participants rated each item based on the frequency of experiencing body image shame, ranging from 0 (never) to 4 (almost always). This scale was shown to have good internal consistency in this study (α = 0.95) and in past research with a female adolescent sample (Webb & Hardin, 2016).
2.2.5. Body appreciation
Feelings of body appreciation and acceptance were measured by the Body Appreciation Scale-2 (BAS-2; Tylka and Wood-Barcalow (2015). The BAS-2 is a 10-item scale consisting of a series of statements that capture positive attitudes and behaviours related to body image, such as valuing one's body for its functionality, expressing gratitude towards one's body, and treating one's body with kindness and respect. Using a 5-point Likert scale with responses ranging from 1 (never) to 5 (always), participants were asked to rate their level of agreement or disagreement with each statement (i.e., “I respect my body”) on. The BAS-2 has shown good internal consistency in this study (α = 0.95) and in previous studies (Tylka & Wood-Barcalow, 2015).
2.2.6. Eating disorder (ED) symptomology
The 8-item Child Eating Disorder Examination Questionnaire (chEDE-Q8; Fairburn et al., 2008) was used to assess ED symptomology. The chEDE-Q8 is based on the EDE-Q (a gold standard diagnostic interview for EDs) but is modified for use in younger populations. The chEDE-Q includes 8-items that assess key attributes of eating disorders over the past 14-days, such as dietary restraint, eating concern, shape concern, and weight concern. Participants were asked questions such as “How many times in the past 14-days have you been trying to cut down on food to control your weight or shape?” in which they responded to on a 7-point Likert scale ranging from 1 (no days) to 7 (every day). The chEDE-Q8 has been shown to have good internal consistency in this study (α = 0.92) and good convergent reliability, validity and internal consistency in young people aged 13-18 years (Kliem et al., 2017).
2.3. Procedure
This study was approved by the Human Research Ethics committee at home institution of the lead author (approval number S221703). Participants were recruited through social media paid advertisements, networks of the research team, and a first-year psychology student research participation scheme. The study was described as examining the relationship between ARC and body image outcomes. Participants were asked to carefully consider their participation and withdraw if topics of cyberbullying, body image, or disordered eating were triggering for them.
Participants entered the online survey (hosted by Qualtrics), read the project information sheet, and before commencing were required to provide consent. To ensure informed consent, participants were required to complete a brief questionnaire aimed at assessing their comprehension and understanding of their participation in the research project before commencing the survey (Mackenzie et al., 2021). To maintain both anonymity and informed consent, a two-question strategy was implemented. One example question inquired, "Is my participation in this study voluntary?" Participants were required to select one of two options: 1) "My participation is not voluntary, and I am expected to complete the study," or 2) "My participation is voluntary, and it is perfectly okay to decline or withdraw at any point." Importantly, participants needed to answer both questions correctly (i.e., “My participation is voluntary, and it is perfectly okay to decline or withdraw at any point") to proceed to the main survey, ensuring a robust verification of their understanding and consent. This approach aligns with ethical considerations and underscores the researchers' commitment to upholding participant rights and transparent research practices.
Once participants commenced the survey, they were required to first complete a set of demographic questions, followed by the above-mentioned self-report questionnaires in the following sequence: BCyQ, ARC measures (roles, types and impact), BSQ, BESSA, BISS, BAS-2 and chEDE-Q. After completing the self-report questionnaires, participants received a debriefing statement. Additionally, they were provided with links to relevant information from the Butterfly Foundation, a non-profit Australian organisation dedicated to supporting individuals with eating disorders and promoting body positivity (Butterfly Foundation, 2022), as well as links to other important mental health services. Participants were also offered a follow-up call with the research team member if required, and a list of resources were made available to participants to access further support if needed. Participants were provided with a separate link which allowed them to leave their details to enter a prize draw for six products worth AUD $50.
2.3.1. Statistical analysis
Data analysis was conducted using SPSS® (version 27, IBM Corporation, Chicago, IL, USA). Prior to analysis, a series of assessments were performed to ensure data integrity, including checking for normality, homogeneity of variance, and identification of any potential outliers. Additionally, descriptive statistics were used to assess the means and standard deviations (SD) for various body image and ED measures across different ARC groups (bullies, victims, bully-victims, and those with no experience of ARC). Notably, ARC-bullies were excluded from further analysis due to the small sample size. Pearson’s correlations were then used to assess relationships between these different measures, and an ANOVA was used to identify differences in body image and ED variable measures between ARC groups (i.e., ARC-victims, ARC-bully-victims and no experience of ARC). Finally, multiple regression analyses were conducted to explore the predictive relationship between specific ARC-victimisation types and adolescent females' inclination to consider changes in appearance, encompassing factors including dieting, altering style, or contemplating cosmetic procedures.