INTRODUCTION: Aphasia tests validated according to the brain injury side are necessary, especially for Spanish instruments. OBJECTIVES: To study the concurrent validity of this Brief Aphasia Evaluation (BAE) to differentiate patients with left cerebral lesions (LC) from patients with right cerebral lesions (RC) as well as LC from healthy participants (HP). To study, through an unrestricted-sub-test-factor analysis, the BAE conceptual and content validity to generate a verbal homogeneous construct. MATERIALS AND METHODS: Data were obtained from a sample of 109 right-handed volunteers: 37 LC, 34 RC and 38 HP. The three groups were matched according to gender, age and education. RESULTS: Both groups of patients were similar in type and site of lesion, time since onset of condition, risk factors, presence of hemianopsia and hemiparesis and number of hospital visits. The Cronbach's alpha coefficient indicated an internal consistency of 0.99 for the total score and 0.88 or above for any of the sub-tests. All sub-tests (with loadings of 0.65 or above) grouped in one factor which explained 78% of the variance. The BAE showed a sensitivity and specificity of 0.84 or above to identify the LC (median as cut-off point). CONCLUSIONS: This test of free distribution demonstrated a satisfactory validity.
INTRODUCTION: Aphasia tests validated according to the brain injury side are necessary, especially for Spanish instruments. OBJECTIVES: To study the concurrent validity of this Brief Aphasia Evaluation (BAE) to differentiate patients with left cerebral lesions (LC) from patients with right cerebral lesions (RC) as well as LC from healthy participants (HP). To study, through an unrestricted-sub-test-factor analysis, the BAE conceptual and content validity to generate a verbal homogeneous construct. MATERIALS AND METHODS: Data were obtained from a sample of 109 right-handed volunteers: 37 LC, 34 RC and 38 HP. The three groups were matched according to gender, age and education. RESULTS: Both groups of patients were similar in type and site of lesion, time since onset of condition, risk factors, presence of hemianopsia and hemiparesis and number of hospital visits. The Cronbach's alpha coefficient indicated an internal consistency of 0.99 for the total score and 0.88 or above for any of the sub-tests. All sub-tests (with loadings of 0.65 or above) grouped in one factor which explained 78% of the variance. The BAE showed a sensitivity and specificity of 0.84 or above to identify the LC (median as cut-off point). CONCLUSIONS: This test of free distribution demonstrated a satisfactory validity.