Summary.A 20‐item questionnaire designed by Joyce and Hudson to measure convergence and divergence was given to a sample of 306 first‐year university students from various science and non‐science disciplines. From the data the scale was refined using factor analysis. To validate the scale, scores on intelligence and open‐ended tests were obtained from the sample and used to define three groups consisting of convergers, divergers and all‐rounders. The scale proved to be more closely related to science/non‐science bias than to convergence