The development and validation of a new resilience inventory based on inner strength

Abstract

There are a number of resilience scales with good psychometric properties. However, the various scales differ in their item content in accordance with the model of resilience the developer had in mind. Culture is one of the reasons for the difference. Thailand, one of the Buddhist cultures, has a different view on resilience compared with Western culture. This study aimed to develop and validate a resilience inventory created based on the inner strength concept using a confirmatory factor analysis (CFA) and Rasch measurement model. The resilience inventory (RI) was developed by creating new items representing inner strengths attributed to resilience. The inner strength was adopted to form the resilience construct, including perseverance, wisdom, patience, mindfulness, loving-kindness and equanimity. In addition, face and content validity were examined by experts in both mental health and Buddhism. The final RI comprised nine items with a 5-point Likert-type scale. The RI-9 was completed by 243 medical students who participated in the study, along with other measurements, i.e., Inner Strength-Based Inventory (iSBI), measuring the ten characteristics of perfection or inner strength, and the Core Symptom Index, measuring anxiety, depression and somatization symptoms. CFA, internal consistency and the Polytomous Rasch rating model were used to investigate the RI-9 construct validity. The mean age of the participants was 22.7 years (SD, 0.8); one-half were male (50%). The RI-9 construct demonstrated item hierarchy as follows: perseverance, patience (tolerance), mindfulness and equanimity, wisdom and loving-kindness. CFA showed that the unidimensional model fitted the data well. Rasch analysis showed no misfitting items and local dependence. The reliability of the person and item was good, and no disordered threshold was observed. Two items were found to exhibit differential item functioning due to sex. RI-9 scores were significantly related to all ten strengths from the iSBI, whereas they were negatively related to depression, anxiety, somatization and interpersonal difficulties. The RI-9 demonstrated validity and reliability. It constitutes a promising tool for outcome assessment in nonclinical populations. Further investigation on external validity as well as psychometric validation in other different cultures, should be encouraged.

Figure 1

Person-item Wright Map of RI-9. Legend: Histogram representing person distribution, more able persons with high level of resilience are located at the top of the map. More difficult (strong resilience) items are located at the top of the map, X1 = Male group, X2 = Female group, Circles are item threshold, item 3 is the easiest item; item 5 is the hardest. The mean of the items is much lower than the person’s ability, indicating that RI-9 is relatively easy for this sample.

Figure 2

Categorical probability curves for RI-9.