Modeling Acquiescence in Measurement Models for Two Balanced Sets of Items

This article concerns a number of conditions that must be met before reaching a conclusion about the existence of the agreement component of a response style called acquiescence. The focus is on the specification of a common style factor behind at least 2 independent theoretical concepts, each measured by a balanced set of items. The proposed approach is explored with a random subsample (N = 986) from a large-scale survey, and subsequently confirmed in another subsample (N = 992) from the same population (Flanders) and in a random sample (N = 1,055) from a different population (Wallonia). The strong relation in both populations of the latent style factor with a variable "sum of agreements" supports the idea that it is possible to control for acquiescence in structural equations.