The stability of symptom measures: Implications for stress research

Forty-nine female undergraduates completed life events and stress ratings on a daily basis over a 4-week period. They also completed a checklist of physical and psychological symptoms at the end of each week. Data from this study as well as others reported in the literature showed that physical and psychological symptom scores are highly stable over time. Although life events and stress ratings predicted future symptoms, the correlations diminished after initial symptom levels were controlled. The correlation between stress measured at time 1 and symptoms measured at time 2 is largely a function of their concurrent correlation at time 1 as well as the stability of the symptom measure. Hence, the value of computing lagged correlations between stress and symptoms has to be reexamined. Possible reasons for the stability of symptom measures are discussed.