A common question asked by researchers using regression models is, What sample size is needed for my study? While there are formulae to estimate sample sizes, their assumptions are often not met in the collected data. A more realistic approach to sample size determination requires more information such as the model of interest, strength of the relations among the variables, and any data quirks (e.g., missing data, variable distributions, variable reliability). Such information can only be incorporated into sample size determination methods that use Monte Carlo (MC) methods. The purpose of this article is to demonstrate how to use a MC study to decide on sample size for a regression analysis using both power and parameter accuracy perspectives. Using multiple regression examples with and without data quirks, I demonstrate the MC analyses with the R statistical programming language.