thank you once again to all who helped with the non-profit project led forward by the German University of Dusseldorf, whose aim it was to create a diagnostic questionnaire which does not confuse normal, engaged play with a clinically conspicuous use of MMORPGs. I’ve now come back to (finally!) present you the results of the aforementioned study, which have been made possible by your help:
191 anglophone players took part in this study, of which 166 were male. The players’ age lied in between 13 and 56 years (24 years on average). At the time of the study, participants stated they played 2 MMORPGs on average, having played their primary MMORPG for 3 years on average. The mean number of hours per week spent on playing MMORPGs was 25; the standard deviation was 21 hours, which shows that, regarding the hours of gameplay, players tend to differ notably from each other.
High (Pearson) correlations (statistical relations) could be detected between the questionnaire score and the weekly amount of hours spent on playing MMORPGs (r = .53, p < .001) as well as between the questionnaire score and the relative amount of spare time spent on MMORPGs (r = .51, p < .001). r describes the correlation, p is the level of statistical significance, according to which an effect can be described as statistically relevant if p is lower than .05. In this case, both correlations are statistically significant. In summary it can be said, therefore, that the more (spare) time was spent on playing MMORPGs, the higher were the obtained questionnaire scores.
The obtained questionnaire scores lie in between 19 and 95, with an average of 40. The score distribution is “shifted to the left”, which means that most participants rather rejected the questionnaire items and obtained a rather low score. On a scale from 1 (does not apply at all) to 5 (fully applies), the average answer lies at 2.12. As a conclusion, the bigger part of participants can most likely not be referred to as “dependent” on MMORPGs.
The ability of questionnaire items to distinguish “excessive” from “non-excessive” players (the first should clearly agree, the latter disagree) lies in between .46 and .70 (corrected item-scale correlations), which is to be rated as high discriminatory power. That’s gratifying because it was our goal to develop a questionnaire which clearly discriminates between these two groups and does not, for example, wrongly categorize people as “addicted”, when they enjoy playing, and play a lot, without having any problems.
The reliability of our questionnaire, which describes the accuracy of its measures, lies at .91 (Cronbach’s alpha) and therefore, it is to be rated as high.
The principal component analysis results in one dominant main factor (eigenvalue: 7.31, explained variance: 38%), which indicates that just one common construct underlies all items of the questionnaire. In this case, this is the “clinically conspicuous use of MMORPGs”.
The 90th percentile was used as a preliminary benchmark in order to classify a “critical use”. In this case, it lies at a score of approximately 58. However such a “cut-off score” should not be misinterpreted as a definite criterion for addiction. This criterion is rather to be a first reference point and make an objective comparison possible. The German version of the questionnaire is currently being tested in a clinical setting.
The questionnaire is still available online: http://www.piaorcs.uni-duesseldorf.de/25questions/
Further results will also be published online on the above mentioned website.
Once again thanks a lot for your help! Without the contribution of the player community, such a project would not have been possible.
(sabine.breuing at uni-duesseldorf.de)