The R package 'coin' provides a unified approach to permutation tests, enabling a wide range of independence tests for various data types, including nominal, ordered, numeric, and censored data, as well as multivariate data. It is based on a flexible conceptual framework that embeds different permutation test procedures into a common theory, allowing for a common S4 class structure with associated generic functions. This framework enables the computational tools in 'coin' to inherit the flexibility of the underlying theory and facilitates the implementation of conditional inference functions for important special cases. The package allows for the easy implementation of conditional versions of classical tests, such as tests for location and scale problems, independence in contingency tables, and association problems for censored, ordered categorical, or multivariate data. The paper provides a detailed exposition of the internal structure of the package and the user interfaces, along with examples of how to extend the implemented functionality. The 'coin' package is implemented in R and serves as a computational counterpart to the theoretical framework developed by Strasser and Weber (1999). The package includes a function 'independence_test()' that provides a convenient user interface for performing independence tests. The function allows for flexible specifications of data, transformations, and distributions, and can be used to perform a wide range of tests, including the Wilcoxon-Mann-Whitney test, Cochran-Mantel-Haenszel test, and others. The package also includes tools for computing exact and approximate p-values, as well as for handling categorical data. The 'coin' package is designed to be flexible and extensible, allowing for the integration of new methods and algorithms. The package is used for a variety of applications, including statistical inference, data analysis, and hypothesis testing. The package is available in R and is widely used in the statistical community for its flexibility and ease of use.The R package 'coin' provides a unified approach to permutation tests, enabling a wide range of independence tests for various data types, including nominal, ordered, numeric, and censored data, as well as multivariate data. It is based on a flexible conceptual framework that embeds different permutation test procedures into a common theory, allowing for a common S4 class structure with associated generic functions. This framework enables the computational tools in 'coin' to inherit the flexibility of the underlying theory and facilitates the implementation of conditional inference functions for important special cases. The package allows for the easy implementation of conditional versions of classical tests, such as tests for location and scale problems, independence in contingency tables, and association problems for censored, ordered categorical, or multivariate data. The paper provides a detailed exposition of the internal structure of the package and the user interfaces, along with examples of how to extend the implemented functionality. The 'coin' package is implemented in R and serves as a computational counterpart to the theoretical framework developed by Strasser and Weber (1999). The package includes a function 'independence_test()' that provides a convenient user interface for performing independence tests. The function allows for flexible specifications of data, transformations, and distributions, and can be used to perform a wide range of tests, including the Wilcoxon-Mann-Whitney test, Cochran-Mantel-Haenszel test, and others. The package also includes tools for computing exact and approximate p-values, as well as for handling categorical data. The 'coin' package is designed to be flexible and extensible, allowing for the integration of new methods and algorithms. The package is used for a variety of applications, including statistical inference, data analysis, and hypothesis testing. The package is available in R and is widely used in the statistical community for its flexibility and ease of use.