Standardization, validity and reliability study of gulhane aphasia test-2(GAT-2)

STANDARDIZATION, VALIDITY AND RELIABILITY STUDY OFGULHANE APHASIA TEST-2 (GAT-2)ABSTRACTGülhane Aphasia Test-2 (GAT-2) has been developed to show the presence ofaphasia and to give the clinician implications for the accompanying speechdisorders such as apraxia and dysartria.132 healthy individual was included to the standardization study and 30 aphasicand 11 right brain injured individual was included to the validation study. For thetest-retest and interobserver reliability research, 32 healthy individuals wereincluded. Healthy individuals were selected from a private physical therapy clinicin Eskişehir. Aphasic and right brain injured patients were selected from theNeurology and Physical Therapy Departments of various hospitals in Eskişehirand zmir. All individuals were selected by predetermined inclusion criteria.All individuals were given GAT-2 (175 total). To find out the GAT-2 scoredifference among healthy, aphasic and right brain injured individuals, one-wayanalysis of variance (ANOVA) and post-hoc Tukey tests were used. The effects ofage, years of education, stroke time post on-set and sex variances over GAT-2scores of healthy and aphasic individuals were calculated by Spearmann?scorrelation and T-tests were used. Med-calc program was used to find the cut-offscores from the GAT-2 scores of healthy individuals. Cronbach α statistic wascalculated to find internal consistency of GAT-2. For the test-retest reliability, 32healthy individuals were tested 2 times (7 days interval between two tests) andSpearmann?s correlation was used to find the difference between two assessments.For the inter-observer reliability, 32 healthy individual were assessed by twoobservers and Spearmann?s correlation was used to find GAT-2 score differencebetween two observers.Healthy individuals? GAT-2 scores were significantly different from the GAT-2scores of aphasic patients, but not from right brain injured patients?. Healthyindividuals? GAT-2 scores were not affected from the sex, age variances but fromyears of education so cut-off scores were calculated by this variance. GAT-2scores of aphasic patients were not affected from the age, sex, years of educationand stroke time post onset. Test-retest and inter-observer reliability and internalconsistency results show that GAT-2 is highly reliable.As a result, GAT-2 was found to be a standardized, highly reliable and a validaphasia test for Turkish staroke patients.Keywords: GAT-2, aphasia test in Turkish, standardizastion, reliability andvalidity