The Reliability and Validity of the Adolescent Subjective Well-Being Scale in Japan

[Aim] The Adolescent Subjective Well-Being Scale was developed to measure quantitatively adolescents’ subjective well-being; its reliability and validity were also examined. [Method] Drawing from the Philadelphia Geriatric Center Morale Scale developed by Lawton et al. in 1972, which is used widely in gerontology, 10 subjective well-being scale items were constructed. Being high in morale means that one is satisfied, has one’s own place, and has accepted inevitable truths. A self-administered questionnaire was completed by junior high school students (aged 13 to 15 years) residing in a rural village (T) near a large city in 2002, 2005, 2008, and 2011. The questionnaire was distributed by hand to each household and left there to be completed. The items were evaluated using a three-point scale. Factor analysis was undertaken to confirm the factor structure, and a reliability coefficient was determined for each sub-scale. The Spearman’s rank correlation coefficient was then calculated for the relationship between total scale score with sleep patterns, stress level, and satisfaction with life. [Results and discussion] A subjective well-being scale, comprised of 3 factors and 10 items, was developed. Data were obtained from 433 respondents (135 in 2002, 115 in 2005, 97 in 2008, and 86 in 2011). Of these, 387 (120 in 2002, 102 in 2005, 85 in 2008, and 80 in 2011; no missing data) were analyzed, along with the items that made up the scale. Factor contribution of the three-factor solutions was 1.86 for the first factor, 1.68 for the second factor, and 1.45 for the third factor. Based on the characteristics of the questionnaire items that constituted each factor, the first factor was named “affect,” the second “self-recognition,” and the third “adaptation to life.” The Cronbach’s alpha coefficient was .62 for the first factor, .57 for the second, .52 for the third, and .71 for the overall scale, which confirmed the scale’s internal consistency reliability. The correlation coefficient between total scale score and sleep patterns was r = .22 (p < .001), between total scale score and stress level was r = -.50 (p < .001), and between total scale score and satisfaction with life was r = .39 (p < .001). Thus, a significant positive correlation with sleep patterns and satisfaction with life, and a significant negative correlation with stress level were indicated. These results confirmed the scale’s criterion-related validity. Therefore, the scale, consisting of 3 factors and 10 items, could be used to examine young people’s subjective well-being.