Abstract
Organizational research often relies on surrogate variables. By "surrogate" we do not refer to family of construct, factor, or latent variables. Rather, we address the situation where one variable is literally the substitute for another variable that is generally unavailable. Consider, for example, the use of "intent to turnover" or "intent to transfer" variables commonly used when actual turnover or transfer data are unavailable. We demonstrate that reliance on such surrogate variables may lead to some misinterpretation. This tendency may be particularly apparent when the relationship between the surrogate and the actual variable is low. This may be further exacerbated when the relationship between the surrogate variable and a third variable is modest as well.
Keywords
Get full access to this article
View all access options for this article.
