The validation and reliability of a Japanese version of the Problematic Online Gaming Questionnaire (POGQ-J)

Participants

A total of 414 undergraduate students completed the questionnaires. The participants were recruited from Ritsumeikan University, Meisei University, Waseda University, Hokusyo University, and Hokkai-Gakuen University by distributing a flyer with a Google Forms survey link in the classrooms between December 1, 2019, and June 30, 2020. The participants from these five universities were taking a psychology course. They responded to the survey using their smartphones or PCs via the Google form URL on the flyer. The flyer stated the purpose of the study as “This survey is being conducted to determine daily life mood and thoughts about online games.” The flyer also contained the following statement: “The results of this study will be used for research purposes only. No individual will be identified. Your responses will not affect your academic performance. It will take approximately 10 min to complete.”

The target sample of this study was university students. One graduate student and one person over the age of 60 responded to the questionnaire and were excluded from the data analysis as they did not fit our criteria for inclusion. In addition, 127 respondents who answered that they did not play online games during the past week were excluded, resulting in 285 respondents (128 males, 157 females, Mage = 19.66, SD = 1.36) in the final analysis of the present study. This study was conducted in accordance with the COSMIN checklist [11], following the detailed guidelines of the preferred reporting style for the development of patient-reported outcome measures [12]. In this study, we did not use statistical methods to calculate the sample size. Mokkink et al. [13] state 7 times the number of items and ≥ 100 as a criterion of excellence as a guide for sample size when conducting factor analysis. Since the POGQ-J has 18 items, it was determined that a minimum of 126 participants would be needed. Since 285 subjects were included in the final analysis of this study, Mokkink’s [13] criteria were considered to be met. It was also necessary to gather over 100 people for the analysis criteria of test–retest reliability [13], and we collected as many samples as possible, considering the withdrawal rate of the second response for test–retest reliability of the POGQ-J. Moreover, no incentives were provided in this study.

MeasuresDemographics

Participants were asked questions to elicit information about their sex, age, and year of study in the university. Moreover, they were asked to report subjectively about the time and amount of money spent on online gaming in the previous week, the primary online games played, and overall Internet usage time in the previous week. The criterion of the past week was based on the original version of the paper [7]. All these items were answered via a Google form.

The Japanese version of the Problematic Online Gaming Questionnaire (POGQ-J; see Additional file 1)

The POGQ developed by Demetrovics et al. [7] comprises 18 items rated on a 5-point Likert scale (1 = never, 5 = always), with higher scores reflecting a greater tendency toward online problematic gaming. This scale consists of six factors (preoccupation: 2 items, overuse: 3 items, immersion: 4 items, social isolation: 3 items, interpersonal conflict: 2 items, and withdrawal: 4 items). The translation of the POGQ was conducted in accordance with the International Society for Pharmacoeconomics & Outcomes Research (ISPOR) task force [14]. First, a forward translation from English to Japanese was performed independently by two of the authors (KY and TI). Then, a professional English translator who was English-Japanese bilingual and blind to the original POGQ translated the provisional POGQ-J back into English. The two English versions of the POGQ (i.e., the original and back-translated versions) were reconciled by KY and TI, and only minor discrepancies were found. These discrepancies were discussed until a consensus was reached. The original author (DZ) evaluated the finalized English version of the POGQ-J and confirmed that the original meanings of each item, instructions, and responses were maintained throughout the translation procedure. As with the original POGQ, each item in the POGQ-J was rated by respondents on a 5-point Likert scale to indicate the extent to which they agreed with the values expressed in each item [7].

The Japanese version of the Game Addiction Scale for Adolescents (GAS7-J [10])

The GAS7-J comprises seven items rated on a 5-point Likert scale (1 = never, 5 = very often), with higher scores reflecting a greater tendency toward game addiction. These items include “Did you think about playing a game all day long?”; “Did you spend increasing amounts of time on games?”; “Did you play games to forget about real life?”; “Have others unsuccessfully tried to reduce your game use?”; “Have you felt bad when you were unable to play?”; “Did you have fights with others (e.g., family, friends) over your time spent on games?”; “Have you neglected other important activities (e.g., school, work, and sports) to play games?” The Japanese version of the GAS7-J has been shown to have good internal consistency (Cronbach’s α = .87) and a one-factor structure. The GAS7-J has been found to be correlated with game usage time (r = .32), the UCLA Loneliness Scale (r = .17), and the “Physical Aggression” subscale of the Japanese version of the Buss-Perry Aggression Questionnaire (r = .11). The GAS7-J was used in this study to examine the convergent validity of the POGQ-J. We assumed that the POGQ-J and GAS7-J would show a moderate positive correlation, as they are comprised of almost similar items (Items of POGQ-J: preoccupation, overuse, immersion, social isolation, interpersonal conflict, and withdrawal; items of GAS7-J: salience, tolerance, mood modification, relapse, withdrawal, conflict, and problems). Although there is a lot of overlap between the items, we expected that there would not be a high correlation because some factors are not common, such as preoccupation and social isolation.

The Japanese version of Young’s Internet Addiction Test (IAT) by the Kurihama Medical and Addiction Center (Treatment of Internet Addiction and Research: TIAR)

The IAT comprises 20 items rated on a 5-point Likert scale (1 = rarely, 5 = always), with higher scores reflecting a greater tendency toward Internet addiction. The IAT has been widely used in previous studies [4, 15]. We used the items translated by the Kurihama Medical and Addiction Center (TIAR) [16]. The IAT was used in this study to examine the discriminant validity of the POGQ-J. We assumed that the POGQ-J and IAT would show a moderate positive correlation, as people with tendencies of online game dependence have been shown to have higher IAT scores [17]. These items include “How often do you find that you stay online longer than you intended?” and “How often do you neglect household chores to spend more time online?”.

The Japanese version of the EuroQol 5 Dimension 5-level (EQ-5D-5L) [18]

This is a generic instrument used to assess the quality of life. It comprises five items on mobility, self-care, usual activity, pain/discomfort, and anxiety/depression. For each item, there are five possible levels of response, namely “no problems,” “slight problems,” “moderate problems,” “severe problems,” and “extreme problems,” with higher scores reflecting a greater tendency toward a poor health state. For mobility, the items are: “I have no problems in walking about,” “I have slight problems in walking about,” “I have moderate problems in walking about,” “I have severe problems in walking about,” and “I am unable to walk about.” For self-care, the items are: “I have no problems with washing or dressing myself,” “I have slight problems with washing or dressing myself,” “I have moderate problems with washing or dressing myself,” “I have severe problems with washing or dressing myself,” and “I am unable to wash or dress myself.” For daily or usual activities, the items are: “I have no problems doing my usual activities,” “I have slight problems doing my usual activities,” “I have moderate problems doing my usual activities,” “I have severe problems doing my usual activities,” and “I am unable to do my usual activities.” For pain/discomfort, the items are: “I have no pain or discomfort,” “I have slight pain or discomfort,” “I have moderate pain or discomfort,” “I have severe pain or discomfort,” and “I have extreme pain or discomfort.” The items for anxiety/depression are: “I am not anxious or depressed,” “I am slightly anxious or depressed,” “I am moderately anxious or depressed,” “I am severely anxious or depressed,” and “I am extremely anxious or depressed.”

The EQ-5D-5L was used in this study to examine the convergent validity of the POGQ-J. We assumed that the POGQ-J and EQ-5D-5L would show a moderate positive correlation as game dependence is associated with a variety of psychiatric problems [3], and among people with these problems, it is assumed that the quality of life is lower.

Procedure

Participants were considered to have provided written informed consent for the use of the information provided for research purposes by completing the research questionnaire using Google Forms. To verify the test–retest reliability of the POGQ-J, participants who responded were asked to retake the POGQ-J approximately two weeks after their initial response.

Statistical analysis

The analysis of the descriptive statistics, convergent and divergent validity, and intraclass correlation coefficients (ICCs) were conducted using IBM SPSS statistics software version 26. For the ICC criteria, .40–.59 was considered fair, .60–.74 good, and over .75 excellent [19]. A confirmatory factor analysis was carried out to confirm the factor structure of the POGQ-J. We hypothesized that the POGQ had a six-factor model. Because the POGQ-J items are rated on a 5-point Likert scale (an ordinal and categorical scale), a weighted least squares mean-variance (WLSMV) estimation was conducted. Four fit indices were employed: chi-square (χ2), comparative fit index (CFI), root mean square error of approximation (RMSEA), and standardized root mean square residual (SRMR). Confirmatory factor analysis was conducted using the Lavaan package with the statistical software package R 3.5.2 (R Core Team, 2019, Vienna, Austria). The goodness of fit criteria were set as CFI > .95, RMSEA < .05, and SRMR < .05 [20]. In this study, a p-value of less than 5% was considered significant.

留言 (0)

沒有登入
gif