I should really learn how to do a longitudinal study. A longitudinal study is a long-term project where data points are gathered at more than two time points, spanning from months or even years. The difficulty of such study is cost in general, participants, money, manpower, etc. The benefit is we can see changes of one variable over a period of time and make good inferences about causal relations.
Daniel L. King (University of Adelaide), Paul H. Delfabbro and the videogame addiction expert Mark D. Griffiths (Nottingham Trent University) published a longitudinal study in Cyberpsychology, Behavior and Social Networking.
A three-wave, longitudinal study examined the long-term trajectory of problem gaming symptoms among adult regular video gamers. Potential changes in problem gaming status were assessed at two intervals using an online survey over an 18-month period. Participants (N=117) were recruited by an advertisement posted on the public forums of multiple Australian video game-related websites. Inclusion criteria were being of adult age and having a video gaming history of at least 1 hour of gaming every week over the past 3 months. Two groups of adult video gamers were identified: those players who did (N=37) and those who did not (N=80) identify as having a serious gaming problem at the initial survey intake. The results showed that regular gamers who self-identified as having a video gaming problem at baseline reported more severe problem gaming symptoms than normal gamers, at all time points. However, both groups experienced a significant decline in problem gaming symptoms over an 18-month period, controlling for age, video gaming activity, and psychopathological symptoms.
I am running an online survey on gamers’ online experiences, if you happen to have some time go ahead and fill it out, it takes 15 minutes.
The authors noted that there are many published videogame addiction research, but mostly on adolescents and very little on adults. They also noted the paucity of longitudinal studies, as of this posting, there are no more than 6 longitudinal studies I have tagged. Thus, they conducted a 18-month longitudinal study on adult gamers and examined the trajectory of videogame addiction symptoms.
Participants: The online survey had initial 393 Australian adult respondents on the first survey. However, it is expected that respondents drop out over the course of 18-months. By the third and final survey, they had a total of 117 respondents, an attrition rate of 71%. The authors examined whether their final sample data differed in any way with their initial sample and found none. The average age is 24 years (SD= 5.5) and it is mostly men (92%).
At the first survey, respondents reported basic demographical information, video gaming history and answered this question on yes/no: “Do you think that currently you may have a problem video gaming habit?” This question will serve to differentiate those who reported having a problem vs. those who don’t. The authors argued that because there are no gold standards in recognizing videogame addiction, a self-diagnosis is most meaningful since those who don’t admit it won’t look for help.
Video Game play Survey: A grid where respondents can report how much time, on each day of the week, they spent playing games on multiple platforms (e.g., computers, console, handheld console, arcade).
Problematic Video game Playing Test (PVGT): an adaption of the internet addiction test. A 20-item questionnaire.
Depression Anxiety Stress Scale (DASS): A 42-item questionnaire measuring three negative emotional states of depression, anxiety and stress.
The second and third survey were sent at 6-months and 18-months following the first survey, respectively.
At the first survey, there were 37 respondents who reported having problems with their video game habits and 80 respondents who did not. This is confirmed by their responses on the Problematic Video game playing survey and the number of hours of gameplay on all surveys.
The authors conducted a 2 (problem status) X 3 (time) repeated measures ANCOVA, with age, gameplay time, and DASS scores as controls. They examined how initial problem status affected the trajectory of the Problematic Video Game Playing Test scores across time. Their analysis revealed significant main effects with time and problem status and a rather weak interaction effect.
Post-hoc analyses showed that among the problematic group, their scores on the PVGT decreased at each time point. From an initial average score of 48.7 to 44.1, 6 months later, to 41.0 at the 18-month mark.
The take home message is that videogame addiction or rather in this study’s context, problematic video gaming and among adults, can decline without intervention. For those in the addiction research, this is not surprising to them as people can mature over time and recover from their addiction, as was the case for gambling addicts.
The authors cautioned several alternatives that maybe respondents overestimated their symptoms at the beginning and as they became more familiar with the survey, they are less prone to overestimate. Maybe participants felt obligated to report that they were improving. Maybe their participation in this longitudinal survey prompted them to be more self-aware and motivated them to improve their lot in life. Some limitations include self-report of problematic symptoms, the sample was self-selected (and perhaps a particular community might set-up some sort of researcher-gamer collaboration) which includes that some problematic gamers may not want to participate in an online survey about addiction, the long time gap between survey might have reduced the sensitivity of the measures and they should’ve asked more questions, such as personality or social support, to control for other confounding variables.
King, D. L., Delfabbro, P. H., & Griffiths, M. D. (2013). Trajectories of problem video gaming among adult regular gamers: An 18-Month longitudinal study. Cyberpsychology, Behavior, and Social Networking, 16 (1), 72-76. DOI:10.1089/cyber.2012.0062