This chapter discusses the proper interpretation of study results in terms of both statistical and causal inference. Any epidemiological analysis, whether descriptive or analytic, generates statistics, and it is up to the researcher to understand how to interpret the results from the statistics. Crude statistics represent a bivariate comparison between an exposure and an outcome and are not adjusted for potential confounding effects. The independent variables' effects are interpreted as an average change in the outcome measure. For a categorical independent variable, the effect is based on a relative comparison of the levels of the categories to some baseline, or referent, level. The estimate for the smoking variable, known as the coefficient in the regression equation, tells us that on average, in referent group of women, smoking was associated with a 335 g lower birth weight compared to not smoking, controlling for maternal age, race, previous preterm labor, and number of first trimester doctor visits.