Variable selection, or the selection of subsets of relevant variables, is an important aspect of statistical regression modeling. Well implemented variable selection techniques yield interpretable models that are characterized by high prediction accuracy. In this workshop we briefly cover various approaches to variable selection, including stepwise and penalization based approaches. Model selection approaches such as cross-validation and information criterion approaches will also be discussed. The focus of the workshop will be to provide advice on implementation of variable selection techniques in regression models. Approaches will be implemented on an illustrative example using R software.