SRS 2398 Lecture Notes - Lecture 9: Longitudinal Study, Statistical Significance, Confounding
Document Summary
In a longitudinal (prospective) study, certain characteristics or behaviours (such as insomnia) are associated, over the course of the study (sometimes many years), with certain outcomes (such as increased mortality [higher rate of death]). In longitudinal studies (as well as other studies), the characteristic or behaviour being studied is called the independent variable, and the outcome being measured is called the dependent variable. Although prospective studies may indicate an association between a characteristic or behaviour and a health outcome, they do not explain the mechanism (the psychosomatic pathways) whereby a particular behaviour may affect one"s health. Mediating variables: variables that can mediate the relationship between the independent variable and the dependent variable. Confounding variables: variables that can have an influence on the dependent variable but are not part of the theory (e. g. age, gender). Researchers control for these variables in order to determine the statistical significance of the independent variable.