Research Tips

Multi-Item Scales

Have you ever found yourself answering a questionnaire and wondering why many questions appear to be similar to one another? They are probably part of a multi-item scale. These scales are often used to measure unobservable characteristics such as attitudes. Asking several questions and combining the responses tends to provide a more accurate representation of these characteristics than asking a single question does.

For example, suppose a retailer is interested in finding out how it is perceived by consumers. The retailer could ask this in a single question, yet consumers’ impressions of the retailer are likely to be a composite of a number of things (range of products offered, quality of products, standard of service, etc.). Asking these questions individually then combining the responses would give a better overall indication of how the retailer is doing, and also identify areas that deviate from the overall pattern (in this example, consumers might like the quality and range of products offered, but think service is poor).