Social media privacy protection skills scale for adolescents

In this study, it was aimed to develop a valid and reliable scale in which the privacy protection skills of adolescents can be evaluated. The study was carried out with two working groups between 11-18 years. 640 people (343 girls, 297 boys) in the first working group, 150 people (92 girls, 58 boys) in the second working group. The research data were collected 2018 - 2019 academic years. In the scale development process, first, the item pool was formed in line with the related literature, other similar measurement tools and interviews with 10 students aged 15-18. The item pool consists of 58 item. Online privacy scale, online security scale, personal information form and social media privacy protection skills scale for adolescents were used to collect data. The construct validity of social media privacy protection scale for adolescents; item total correlation was analyzed by exploratory and confirmatory factor analysis. Criterion related validity of the scale was investigated by revealing relationships between social media privacy protection skills scale for adolescents and online privacy scale, online security scale inventory scores. Reliability analyses were carried out with internal consistency Cronbach Alpha and test-retest reliability methods. As a result of the factor analysis, it was found that the total variance explained by the 27-item scale was 46.33% and exhibited a three-factor structure. These three factors titled as privacy protection trend, self-efficacy and privacy protection behavior. In the first order confirmatory factor analysis, the three-factor construct was confirmed, results showed that the confirmatory factor analysis provided good fit indices (X2/sd= 1.82, NFI = .93, NNFI = .97, CFI = .97, IFI = .97, RMSEA = .056, GFI = .86, AGFI = .83). In the second order confirmatory factor analysis, the three-factor construct was found to merge with an upper-construct. It was also observed that the fit indices of the second order confirmatory factor analysis demonstrated good fit values (X2/sd = 1.82, NFI = .93, NNFI = .97, CFI = .97, IFI = .97, RMSEA = .056, RFI=.93). Correlations coefficients between online privacy scale, online security scale and the social media privacy protection skills scale for adolescents scores supported criterion related validity. The Cronbach Alpha coefficient of the online privacy scale .42. The Cronbach Alpha coefficient of the online safe behavior factor of the online security scale .50 and the Cronbach Alpha coefficient of the online unsafe behavior factor is .63. Test-retest reliability coefficients of the scale were .90 and .78. For the privacy protection trend, self-efficacy and the privacy protection behavior factors, Cronbach Alpha coefficients were .85, .85 and .65, respectively. In addition, test-retest reliability coefficients were found .55, .78 and .54, respectively. In conclusion social media privacy protection skills scale for adolescents was considered reliable and valid instrument.