The ggeffects package in R is designed to calculate marginal effects for a wide range of regression models, making it easier to interpret complex models. Marginal effects provide a more intuitive understanding of how predictors relate to the outcome, especially in models with interactions, transformations, or non-linear relationships. The package's functions are type-safe and return consistently structured data frames, facilitating integration into workflows and custom functions. The output follows a "tidy data frame" philosophy, making it compatible with ggplot2 for visualization. The data frames include x, predicted, conf.low, conf.high, and group values, which can be used for plotting with confidence intervals and grouping. The package supports numerous models, including linear, generalized linear, mixed-effects, and survival models, as well as interaction terms, splines, and polynomials. It also supports labelled data, allowing for human-readable annotations in plots, which is particularly useful for users working with Stata or SPSS data. The package includes a plot() method that automatically handles common plotting characteristics. Examples demonstrate how to use ggpredict() to calculate marginal effects and plot them. The package has been used in various publications to create marginal effects plots, which are often preferred over tables for their clarity in representing complex relationships.The ggeffects package in R is designed to calculate marginal effects for a wide range of regression models, making it easier to interpret complex models. Marginal effects provide a more intuitive understanding of how predictors relate to the outcome, especially in models with interactions, transformations, or non-linear relationships. The package's functions are type-safe and return consistently structured data frames, facilitating integration into workflows and custom functions. The output follows a "tidy data frame" philosophy, making it compatible with ggplot2 for visualization. The data frames include x, predicted, conf.low, conf.high, and group values, which can be used for plotting with confidence intervals and grouping. The package supports numerous models, including linear, generalized linear, mixed-effects, and survival models, as well as interaction terms, splines, and polynomials. It also supports labelled data, allowing for human-readable annotations in plots, which is particularly useful for users working with Stata or SPSS data. The package includes a plot() method that automatically handles common plotting characteristics. Examples demonstrate how to use ggpredict() to calculate marginal effects and plot them. The package has been used in various publications to create marginal effects plots, which are often preferred over tables for their clarity in representing complex relationships.