The validity and reliability of the Turkish form of recovery assessment scale

ABSTRACT The study was designed to evaluate the validity and reliability of Turkish form of Recovery Assessment Scale developed by Corrigan et al. (1999) with the aim of evaluating subjective recovery processes of individuals who have been diagnosed with mental illness. The research was carried out in two institutions as Ege University Medical Faculty Hospital Community Mental Health Center and Manisa Mental Health and Disease Hospital Community Mental Health Center. The sample of the research consists of 180 people; who have registered these institutions, who have met the research criteria and who have volunteered to participate in the research and who have been diagnosed with mental illness. The research was planned in a methodological way. The data of the study were obtained from February 2017 to May 2017 using a face-to-face interview technique by the Introductory Information Form, the Recovery Assessment Scale (RAS) and the Psychological Well-Being Scale (PWBS). For the validity studies of the research, the linguistic validity, the content validity, the construct validity and the criterion-related validity methods were used. For the validity of the language, the translation-re-translation method is used, and the CVI values of items were found using the Davis technique for the content validity. Confirmative Factor Analysis were used construct validity. For the criterion-related validity, the correlation coefficient, regression coefficient and explanatory coefficient values between RAS and PWBS were examined. Time-invariance and internal consistency methods were used for the reliability study of the research and for internal consistency, item analysis and Cronbach's alpha coefficient value were examined. For the linguistic validation, RAS was translated to Turkish and then into English. After the linguistic validity was given, the scale was presented to the expert opinion and the content validity was evaluated according to expert opinions. According to expert opinions CVI value of all items found to be over 0.80, which is the accepted value (CVI> 0.80). The Turkish version of the scale was created after the preliminary application was made to the individuals who would not be included in the sample later. For the construct validity of the scale, the values of the fit indices examined Confirmative Factor Analysis (CFA) were found to be among the accepted values (χ2 / SD = 1.98, RMSEA = 0.07, SRMR = 0.06, GFI = 0.98, CFI = 1.00, NFI = 1.00, AGFI = 0.97). When factor loads related to the subscales of the scale were examined, it was found that the factor loadings of the items varied between 0,34 and 0,98 and between the desired values (p<0.01). When the criterion-relation validity of the scale was examined, it was found that there was a positive and significant relationship between the Recovery Assessment Scale and the Psychological Well-Being Scale (r = 0,864, p <0,01). The regression coefficient (β) was 0.71 and the explanatory coefficient (R2) was 0.50 between the Recovery Assessment Scale and Psychological Well-Being Scale. For the reliability evaluation of the Recovery Assessment Scale, item scale total correlation values (r) were examined and it was seen that these values changed between 0.078 and 0.806. Item-total score correlation values were then examined. The p-values calculated for the item-total score correlation of Item 2, Item 3, Item 10 which are below the lowest desired value of 0.30, were found to be statistically insignificant (p> 0.05). Subsequently, the item-subscale total score correlation value of the scales was examined and only the value examined for Item 10 was found to be below 0.30. It has been seen that both the item- total score correlation value of item 10 and the item-subscale total score correlation value were found below the desired value. It was discussed with the scale owner and the item extraction was not carried out in order to the fact that the item factor load of item 10 higher than 1.96 which is the lowest value in DFA, and not to disturb the conceptual structure of the scale. The Cronbach's alpha coefficient of the Recovery Assessment Scale was found 0.90. The Cronbach's alpha coefficient of the subscales of the scale ranged from 0.74 to 0.89. To test the test-re test reliability of the Recovery Assessment Scale, the scale was applied again to 35 individuals two weeks after the first application and the correlation between the first and last measurements was examined and it was found that there was a significant correlation between the two measures (p <0,01). As a result of the validity and reliability analyses, it was determined that the Recovery Assessment Scale is a valid and reliable scale to measure subjective recovery in individuals who are diagnosed with mental illness. It has been suggested that the Recovery Assessment Scale should be applied to wider sample groups to investigate the instability of individuals who are diagnosed with mental illness.