Measuring attitudes to science: Unidimensionality and internal consistency revisited

Summated ratings attitude scales commonly consist of numerous items whose scores are summed to yield a total score. An assumption underlying this technique is that the items in the scale reflect a common construct. If this is not met, the procedure produces uninterpretable data. Although this psychometric principle has been known for a long time, numerous studies in the literature demonstrate a neglect of it. Some make no attempt to conceptualise the construct to be measured; others conceptualise the construct but then ignore the possibility that it may be multidimensional; still others contain evidence indicating that the construct is multidimensional and then proceed to ignore that evidence. A possible contributor to the confusion is the misunderstanding of the related yet distinct concepts of internal consistency and unidimensionality. This paper presents examples of poor and good instrument design, in the hope that clarification of the issues might make a difference in the future.