Test-retest reliability of the Edinburgh Handedness Inventory and Global Handedness preference measurements, and their correlation.

Test-retest reliability of the Edinburgh Handedness Inventory and of self-reported Global Handedness (left, right, ambidextrous), administered 18 mo. apart, was evaluated for 735 adult subjects. Conventional methods for evaluating test-retest reliability were compared with cross-tabulation and the kappa statistic, using as criteria both the degree of agreement and the degree and direction of disagreement. Inventory-derived laterality scores were correlated with Global Handedness responses, yielding a characteristic distribution of laterality scores for each global response which permitted the partitioning of the J-shaped frequency distribution plot of laterality scores into right, left, and ambidextrous regions. Test-retest reliability of the correlation between global responses and laterality scores on retest was demonstrated.