In a recent article, Gottman and Krokoff (1989) presented a fascinating and counterintuitive portrayal of a temporally related dynamic relationship of marital conflict and marital satisfaction. Unfortunately the results from which their conclusions spring are quite problematic. In a 2-wave, extreme-groups design, Gottman and Krokoff used raw change in satisfaction as a correlate of Time 1 conflict variables for purposes of analysis. Serious psychometric questions arise with this approach to the causal assessment of change. Given the considerations raised, there is a high likelihood that Gottman and Krokoff's data result from statistical artifact. Alternative approaches to the measurement of change are discussed in the context of illustrating the problematic components of Gottman and Krokoff's analysis.