Abstract: Resilience is defined as individuals’ healthy and constructive adjustment after they have experienced serious, traumatic, or catastrophic events. Research on resilience was stimulated by the development of positive psychology. For research purposes, resilience is usually considered from three perspectives: personality, process, and outcome. These three components are different facets of a virtually consistent construct. Several resilience scales have been developed by Western researchers, such as the Resilience Scale (RS), Resilience Scale for Adults (RSA), and Connor-Davidson Resilience Scale (CD-RISC). However, the validity of these scales was not supported when the scales were tested on Chinese samples (Yu & Zhang, 2007). Therefore, the aim of the present study is to develop a localized resilience scale specifically for Chinese adolescents. For intervention purposes, in our research, resilience was defined as a coping process, since neither personalities nor outcomes can be easily altered.
We interviewed twenty-five middle school students who had undergone major negative life events but were still mentally well-adjusted. The interview was structured and included an investigation into the students’ efforts in coping with adversity. Using content coding and stepwise coding, the information abstracted from the interviews was formulated into 100 items.
Two hundred and eighty-three teenagers—including 126 males, 146 females and 12 unclear—from both junior and senior high schools completed the 100-item scale. After item discrimination analysis, 87 items were retained for exploratory factor analysis with oblique rotation. Items with low or double loadings were deleted. The final scale included 27 items, classified into five factors: goal planning, help-seeking, family support, affect control, and positive thinking, which accounted for 52.4% of the total variance; each factor individually accounted for more than 5% of the variance. Cronbach’s alpha coefficient of separate factor and total scale were all higher than 0.70. The discrimination of each item was greater than 0.30.
The second sample included 420 middle school students, consisting of 191 males, 209 females and 20 unclear. They completed the 27-item resilience scale and 2 relative scales. A confirmatory factor analysis was conducted on this sample. The fit indexes met the psychometric requirements, c2 = 787.85, df = 314, RMSEA = 0.07, and CFI = 0.92, which indicated that the five-factor structure of resilience was stable and achieved goodness of fit. Higher-order confirmatory factor analysis implied that goal planning, affect control, and positive thinking belong to a higher order factor called individual power, while help-seeking and family support belong to another factor called supportive power. The Resilience Scale (RS; Wagnild & Young, 1993) and the Quality of Life Scale for high school students (QLSH; Hu et al., 2002) were used as external criterion. The correlation coefficient with the RS was 0.53 and that with the QLSH was 0.49. No gender differences were found. The strategy of localization and the scope of application are also discussed in this study

Key words: resilience, scale, validity, adolescent