Measuring resilience with the RS-14: a tale of two samples

The purpose of this study was to systematically examine the psychometric properties of the 14-item Resilience Scale (RS-14; Wagnild, 2009a). Exploratory and confirmatory factor-analytic methods were employed, including an analysis of measurement invariance models by sex and race/ethnicity. Descriptive statistics, reliability, and validity data were also calculated. Analyses were based on 2 samples, one made up of individuals seeking mental health services following the 2010 Gulf oil spill (N = 1,032) and the other made up of university students (N = 1,765). For both samples, all items loaded (>.30) onto 1 factor, indicating cohesive structure for a 1-factor model explaining 53.2% of the variance in the clinical sample and 67.6% of the variance in the undergraduate sample. Further, the examination of measurement invariance indicated that the RS-14 was similarly structured for sex and race/ethnicity. Reliability coefficients exceeded.90 for both samples and also when data were examined by comparison groups. The RS-14 correlated significantly and as expected with measures of positive concepts (such as perceived meaning in life and satisfaction with life) and indexes of psychological distress (such as depression, anxiety, stress, and posttraumatic stress). These data support the utility of the RS-14 with clinical and undergraduate student samples. Implications for these data are discussed.