Subscale distance and item clustering effects in self-administered surveys: A new metric

The authors explore the effect of a form of question context on responses t o a computer-mediated marketing research survey. As an increasing proportio n of marketing research is conducted through computer interfaces, the pool of potential context effects is rapidly expanding. The authors conduct an e xperiment using a multi-item scale that consists of five dimensions and man ipulate three such context effects: explicit item labeling, item presentati on (alone/grouped), and subscate items presented contiguously or not. In a refined analysis of variance, the authors use a special one-dimensional cas e of the spatial and attribute-based distance metric proposed by Hoch, Brad low, and Wansink (1999) to explain subscale variance, replacing the indicat or variable for clustering used in a standard analysis of variance. This me tric provides a scalar measure of how much variation exists in the order of presentation of items within a subscale (the subscale distance). This anal ysis indicates a significant decrease in subscale variance (increased relia bility) with decreasing subscale distance but no-longer-significant effects due to labeling and grouping. The authors discuss implications of their fi ndings for researchers conducting surveys in computer-mediated environments .