The article introduces the RMediation package, an R software tool designed for constructing confidence intervals (CIs) for mediated effects. Mediated effects are the product of two regression coefficients, and the distribution-of-the-product method is highlighted as the most statistically effective method for generating CIs. The RMediation package offers several methods, including the distribution-of-product approach, Monte Carlo simulations, and asymptotic normal distribution, to produce CIs, percentiles, and quantiles. It also includes features such as plotting the distribution and CI of the mediated effect. The package addresses limitations of existing programs like PRODCLIN, which are cumbersome to use and have restrictions in handling certain cases. A simulation study comparing different CI methods shows that the distribution-of-product method generally provides more accurate results, especially for smaller sample sizes. The article concludes by recommending the distribution-of-product method over other methods, particularly for smaller datasets.The article introduces the RMediation package, an R software tool designed for constructing confidence intervals (CIs) for mediated effects. Mediated effects are the product of two regression coefficients, and the distribution-of-the-product method is highlighted as the most statistically effective method for generating CIs. The RMediation package offers several methods, including the distribution-of-product approach, Monte Carlo simulations, and asymptotic normal distribution, to produce CIs, percentiles, and quantiles. It also includes features such as plotting the distribution and CI of the mediated effect. The package addresses limitations of existing programs like PRODCLIN, which are cumbersome to use and have restrictions in handling certain cases. A simulation study comparing different CI methods shows that the distribution-of-product method generally provides more accurate results, especially for smaller sample sizes. The article concludes by recommending the distribution-of-product method over other methods, particularly for smaller datasets.