Obtaining the short form of the internet addiction scale according to Classical Test Theory

The aim of this research is to obtain the short form of the "Internet Addiction Scale" (IAS) according to the Classical Test Theory. In this correlational study, there are a total of 394 people selected by convenient sampling method. The Turkish version of the IAS developed by Nichols and Nicki (2004) and adapted to Turkish by Kayri and Günç (2009) was used in the study. The "Interaction Anxiety Scale" adapted into Turkish by Coşkun (2009) was used to check the criterion validity of the IAS. SPSS 24 for the normality assumption of the scale used, exploratory factor analysis, Pearson correlation coefficient, Cronbach alpha (α) coefficient used to determine reliability; SPSS AMOS was used for confirmatory factor analysis and R package program was used to determine the item combinations of the short forms that gave the highest correlation with the scores obtained from the long form of the scale. Fisher Z statistics were calculated in Excel 2016 program for the significance of the difference between the obtained correlation scores. According to the results obtained, it was determined that obtaining the short form of the IAS by factor analysis and selecting the item combinations that gave the highest correlation with the scale total scores increased the total variance explained by the scale and preserved the structure of the long form. The short forms obtained by factor analysis and the combination of items with the highest correlation with the total score obtained from the scale showed similar psychometric properties. Finally, it was deemed appropriate to use the IAS with 6 items according to the method of factor analysis and the items with the highest correlation with the total score.