Does hand proficiency determine hand preference?

There is debate in the handedness literature over proficiency and preference measures are indicators of common underlying factors, or separate dimensions of behavior with different causes. It has been argued that hand preference has different origins from hand proficiency on the grounds that (i) the two types of measure are imperfectly correlated and (ii) distributions of relative hand proficiency are normal or nearly so, whereas distributions of hand preference are J-shaped. A model is described in which the probability that one hand will be preferred for a given activity is directly proportional to the relative proficiency of the two sides, the latter being normally distributed. It is shown that imperfect correlations between proficiency and preference measures, and J-shaped distributions of preference, can be predicted by such a model. These facts about hand preference and proficiency do not, therefore, go against the view that preference is determined by proficiency.