Abstrakt
The aim of this study was: a) to verify the diagnostic quality of the Edinburgh Handedness Inventory (EHI) items and b) to determine an appropriate number of hand preference categories in a children population. Sample and setting.
A total of 366 participants (178 boys and 188 girls) randomly selected from elementary schools in Prague, Czech Republic participated in this study. Children's age ranged from 8 to 12 years and none of the participants has reported any intellectual or language deficiencies.
Statistical analysis. Polytomous item response theory model (Rating scale model) was used to evaluate the diagnostic quality of particular items in the inventory.
Appropriate number of handedness categories based on the EHI results was determined using latent class analysis. Chisquare tests and intra-class correlation coefficient were used for further evaluation of the obtained handedness categorization.
Results. Two items (using a broom and opening a box) that violated the Rasch properties of the Rating scale model were identified.
Two classes that were characterized by the predominant use of either right or left hand (right-handers and left-handers) and a class with inconsistent hand-preference (mixed-handers) were stated. Results of the study suggest that values of laterality quotient LQ=0+-60 and LQ=0+-70 could be recommended for assigning individuals into the mixed-handedness class.
Study limitation. It was not assessed whether self-reported hand preference is in accordance with the real hand preference for the EHI items/ activities.
Using a response format with three categories could be seen as another limitation since the originally proposed instrument utilized five categories.