請炫文在確認黃色底線料的正確性。
title:
The Chinese version of the 27-item revised Social Functioning Scale (CSFS-27): Confirmatory factor analysis
Abstract
Objectives: The Chinese version of the 27-item revised Social Functioning Scale (CSFS-27) assesses multidimensional social functioning. The CSFS-27 is consists of 4-item social withdrawal, 5-item sociality, 5-item interpersonal communications, 4-item leisure activities, 4-item competence of independence and 5-item performance of independence. Total scores of these 6 subdomains represent overall social functioning. This study aimed to apply confirmatory factor analysis to validate the CSFS-27 in individuals with schizophrenia.
Methods: A total of 342 subjects living in the community were assessed by the CSFS-27. A confirmatory factor analysis was performed to test the six 1st order factors model comprising 6 domains. We also examine a single 2nd order factor representing overall to social functioning of an individual.
Results: The six 1st order factors met the criteria of the four fit indices (χ2/df = 1.88, CFI = 0.92, TLI = 0.91, and RMSEA = 0.06), demonstrating good model fits. The single 2nd order factor showed generally good model fit indices (χ2/df = 2.17, CFI = 0.89, TLI = 0.88, and RMSEA = 0.08).
Conclusion: The six 1st order factors and single 2nd order factor of the CSFS-27 for the social functioning of schizophrenia appear to reflect a statistically coherent construct. Implications for the quantitative application of the CSFS-27 are discussed.
Keywords: Social functioning; schizophrenia; confirmatory factor analysis.
1. Methods
1.1. Participants
Individuals with schizophrenia were recruited from community psychiatric rehabilitation centers, day centers and chronic wards in Taiwan from November 2014 to July 2015. Inclusion criteria were as follows: (1) diagnosis of schizophrenia or schizoaffective disorder according to the Diagnostic and Statistical Manual of Mental Disorders, Text Revision, 4th edition (DSM-IV-TR); (2) age of 18 to 65 years; (3) at least an elementary level of education; and (4) stable clinical condition (no psychiatric emergency treatment records and consistent dose of antipsychotic medication received for at least 3 months). Exclusion criteria were as follows: (1) severe brain injuries and impaired cognitive function and (2) with diagnoses of substance use. This study was approved by the Institutional Review Board of Taipei City Hospital (TCHIRB-10406117). Informed consent for participation was obtained from the participants personally.
1.2. Procedures
Participants who met the recruited criteria were assessed the CSFS in a quiet environment. Demographic and clinical information of the participants were collected from clinical records.
1.3. Instrument
The CSFS is revised from Social Functioning Scale (Birchwood, Smith, Cochrane, Wetton, & Copestake, 1990). We deleted the subdomain of employment, due to the complexity and jobless status of most schizophrenic persons. In addition, we took Chinese culture into consideration and revised some items, such as revised “Church activity” to “Religious activities”. Thus, the CSFS contains 6 subdomains: 10-item social withdrawal, 10-item sociality, 10-item interpersonal communication, 13-item leisure activities, 15-item competence of independence, and 15-item performance of independence. Each subdomain is assessed on a 4-degree scale of a frequency from almost never, rarely, sometimes, and often, in which 1 means a frequency of almost never of daily living in recent 3 months and 4 means often. The final result includes 6 subdomains and a single overall rating, where higher scores indicate better social functioning.
1.4. Data analysis
All statistical analyses were performed using Software IBM SPSS and AMOS (version 21, IBM, Armonk, NY). The factorial structure of the CSFS was assessed through Confirmatory Factor Analysis, using Structural Equation Modeling with Maximum Likelihood estimation method.
We assessed the assumption of normality of the items using skewness (Sk) and kurtosis (Ku), Sk>|3| and Ku>|10| indicating severe deviations to normal distribution (Markus, 2012). Model fit was assessed using the normed chi-square (χ2/df), goodness of fit index (GFI), confirmatory fit index (CFI), Tucker Lewis index (TLI), and root mean square error of approximation (RMSEA). We determined the good model fit of the data if χ2/df≦2.00, CFI≧.95, TLI≧.95, and RMSEA≦.06. Additionally, model fit was determined acceptable when χ2/df≦3.00, CFI≧.90, TLI≧.90, and RMSEA≦.08 (Dwinger, Kriston, Härter, & Dirmaier, 2015).
Items’ standardized loading (λ) and individual reliabilities (R2) were analyzed. When model fits were poor, items were iteratively deleted based on low factor loading and low R2 values. Additionally, the items with a large modification index were deleted and then CFA has conducted again till a sufficient model fit was shown. We first performed a 1st order CFA to verify the structures of CSFS. A 2nd order CFA was conducted to confirm the existence of a higher-order social functioning factor.
CRAVE, keep writing.
