Validity and Reliability of the Instrument DOA : A Dialogue about Working Ability

• Main Author: Linddahl, Iréne
• Format: Others
• Language: English
• Published: Hälsohögskolan, Högskolan i Jönköping, HHJ, Avd. för rehabilitering 2007
• Subjects: Client dialogue; Goal setting; Occupational therapy; Participation; Rasch analysis; Unidimensionality; Work assessment
• Online Access: http://urn.kb.se/resolve?urn=urn:nbn:se:hj:diva-783; http://nbn-resolving.de/urn:isbn:978-91-85835-00-3

The overall aim of this thesis was to examine the construct validity and the reliability of the instrument DOA; Dialogue about working ability. The theoretical foundation for DOA is the Model of Human Occupation. The instrument is divided into two sections; client self-assessment and professional assessment focusing on the individual’s working ability. The assessments are followed by a dialogue to distinguish goals for the return to-work process based on the client’s own preferences. In line with the theoretical base, the concept ‘work’ refers to both paid work and other productive activities related to work. In Study I, 21 professionals and 126 clients participated from three clinics in Sweden. All clients were involved as outpatients in psychiatric work rehabilitation in county councils or community services. The results from the Rasch analysis test indicated that items are well separated from each other and generally work together in five unidimensional continuums with the objective to measure abilities related to work. Twenty-nine of 34 items showed goodness-of-fit statistics, which means acceptable infit MnSq values from >0.6 to <1.4 in association with Z values from –2 to 2, and that the five misfit items have to be revised. Thirty-four outpatient clients and 14 professionals from four clinics within psychiatric work rehabilitation in Sweden participated in Study II. Two reliability aspects were studied: test-retest and inter rater reliability. The statistical analyses used were Spearman Rank Correlation test and Percentage of agreement (PA). Generally, the Spearman Rank Correlation test showed acceptable significant correlations between test retest since a greater part of items correlated “moderate to good” (r = 0.51 – 0.75). In the inter raters´ test PA was good, ranging from 93.1 % to 96.5 %, which is well above the acceptable criteria of 80 %. There were no significance differences between the raters. The study has shown that one item has to revised. The findings verify that DOA has potential as a means of making valid and reliable assessments of working ability, as described by the Model of Human Occupation, out of both clients´ and occupational therapists´ perspective.