Measuring dysgraphia: a graded-difficulty spelling test

In this paper we describe the construction of a graded-difficulty spelling test for adults consisting of two alternative forms each containing 30 words (GDST, Forms A and B). The spelling test, together with background tests of verbal and non-verbal skills, was administered to 100 control patients with orthopaedic injuries. The two forms of the spelling test were highly correlated (0.92). Spelling was highly correlated with reading (0.75, 0.77) and moderately correlated with vocabulary (0.57) and naming (0.39, 0.40). There was no correlation between spelling skills and non-verbal reasoning. The test was validated in a group of 26 patients with left hemisphere and 20 patients with right hemisphere lesions. Spelling was shown to be lateralized to the left hemisphere and there appeared to be a shift in scores of the left hemisphere group towards the lower quartile, with 65% of the left hemisphere group falling within this band. The most severe spelling impairments were invariably associated with other language disorders but a number of dissociations were documented at spelling levels falling between the 5th and 25th percentile band. Two patients with left hemisphere lesions (8%) were identified as having selective dysgraphias. The lack of overlap between the anatomical sites of the two patients with specific lexical dysgraphia argues against a single site for this type of dysgraphia and argues for further refinement of this classification of spelling disorder.