ANALYSIS OF RELATIONSHIPS BETWEEN PROBLEMATIC INTERNET USE AND ATTACHMENT STYLES

The purpose of this study is to determine university students' problematic internet use levels and analyze some variables that predict problematic internet use. This study, which was designed according to the relational survey model, was conducted at a large university in Turkey. The research data were obtained from a total of 217 university students, 101 female and 116 male who volunteered to participate in the study. Two different scales were used to collect the research data. In order to determine university students' problematic internet use levels, Problematic Internet Usage Scale, which was developed by Ceyhan, Ceyhan and Gurcan (2007), was used. The scale was a 5-point likert type scale that contained response choices ranging from (1) "totally inappropriate" to (5) "totally appropriate" and consisted of 33 items in 3 factors. In order to measure university students' attachment styles, Relationship Scales Questionnaire, which was developed by Griffin and Bartholomew (1994) and whose validity and reliability was tested and proven on Turkish university students by Sumer and Gungor (1999), was used. The scale, which was used to determine four different attachment styles, was a 7-point likert type scale that had choices ranging from (1) "it does not describe me at all" to (7) "it totally describes me". Before analyzing the data obtained from the scales, first the normality and linearity of the data were tested. To this end, central distribution, skewness and kurtosis values were examined on the distributions of the total scores obtained from the scales. Since the calculated skewness and kurtosis values were between +1 and -1, it was decided that normality of the distribution had been ensured. According to this, descriptive statistics, correlation analysis and regression analysis were used in the analysis process. Pearson Product Moment Correlation Coefficient technique and stepwise regression analysis technique were used to investigate the relationships between problematic internet use and its sub-dimensions (excessive usage, social benefit/social comfort and negative consequences of the internet), which were the dependent variables of the study, and attachment styles and its sub-dimensions (secure, fearful, preoccupied and dismissing), which were the independent variables of the study. The findings of the study indicate that university students have a medium level problematic internet use. As a result of the stepwise multiple regression analysis, in which factors that predicted university students' problematic internet use, the preoccupied, secure and dismissing attachment styles were found to be the strongest predictors of problematic internet use respectively.