Testing the Measurement Equivalence of the Adapted Chinese Leisure Satisfaction Scale Between Schizophrenia and General Populations in Taiwan

PURPOSE: This study aimed to test the measurement equivalence of the adapted Chinese Leisure Satisfaction Scale (LSS) between the schizophrenia and general populations, exploring the major characteristics of people with schizophrenia to help them increase leisure satisfaction through leisure activity participation. Participating in leisure activity helps increase flow experience, well-being, and life satisfaction. Although occupational therapy practitioners encourage their clients to increase participation in leisure activities, leisure satisfaction should also be considered. The adapted LSS was developed to assess satisfaction with different leisure activities in isolation.

Given that the adapted LSS was developed with the general population, little is known about differences on the measurement compared with people with schizophrenia, who may have attention deficits and motivation problems when participating in activities. Testing measurement equivalence helps determine whether an instrument is equivalent across two independent groups, focusing on the unchanged operation of the factor loadings. In testing for the equivalence of factor loadings, the parameters for the first group are freely estimated, and then the factor loadings for the other group are constrained equal to those of the first group.

DESIGN: For this cross-sectional design study, we enrolled 204 adult participants in two groups. The schizophrenia group included 100 adults diagnosed with schizophrenia or schizoaffective disorder from psychiatric outpatient clinics, day hospitals, and vocational rehabilitation programs in Taipei, Taiwan; the control group included 104 general population participants without severe mental illness from the same settings (e.g., staff, volunteers, interns, community residents). Participants were encouraged by the instructor with autonomy support to determine the speed of and play with an interactive ping pong ball launcher for 50 balls (right and left sides alternately with 3-s intervals). They then completed the Chinese Flow State Scale–2 and the adapted Chinese Leisure Satisfaction Scale, and participants in the schizophrenia group completed the Positive and Negative Symptom Scale.

Lisrel 9.2 (Scientific Software International, Skokie, IL) was used to analyze data. We first conducted exploratory factor analysis and confirmatory factor analysis (CFA) for each group. Then we conducted multigroup CFA with the CFA model for the control group as a standard, including factor loadings, factor correlations, and error variances invariant, and then held parameters of the schizophrenia group equal to the control group with Wald’s t test to drop an insignificant structural parameter and modification index to add an additional meaningful structural parameter. Root mean square (RMR), standardized RMR (SRMR), and goodness of fit index (GFI) were administered to assess the adequacy of fit. Then we alternatively applied the schizophrenia group as the standard to do multigroup CFA on the same procedures and criteria.

RESULTS: With the control group as the standard, the results of multigroup CFA indicated that the cross-group equivalence was poor (RMR = .252, SRMR = .243, GFI = .533); five items for the schizophrenia group were added on this constrained model. Moreover, with the schizophrenia group as the standard, the results of multigroup CFA showed that the cross-group equivalence was better (RMR = .219, SRMR = .215, GFI = .987); two items for the control group were added on this constrained model.

CONCLUSION: We found that the items related to self-confidence, leisure experience, and motivation may have caused major differences between the schizophrenia and control groups in leisure satisfaction.

IMPACT STATEMENT: This study suggests that occupational therapy practitioners should address the problems of deficiencies in self-confidence, leisure experience, and motivation in people with schizophrenia when facilitating their leisure satisfaction through leisure activities. The results of this study may serve as part of the evidence to establish the foundations for further inquiry regarding occupational therapy intervention outcomes and cross-cultural differences.

References

Byrne, B. M. (2008). Testing for multigroup equivalence of a measuring instrument: A walk through the process. Psicothema, 20, 872–882.

Csikszentmihalyi, M. (2014). Flow and the foundations of positive psychology: The collected works of Mihaly Csikszentmihalyi. New York: Springer.

Di Bona, L. (2000). What are the benefits of leisure? An exploration using the Leisure Satisfaction Scale. British Journal of Occupational Therapy, 63, 50–58. https://doi.org/10.1177/030802260006300202

Kline, R. B. (2015). Principles and practice of structural equation modeling (4th ed.). New York: Guilford Press.