Regression analysis is one of the most frequently used statistical methods in the field of animal science, as it is in many branches of science. However, it is not possible to provide linear regression assumptions for many biological studies. It is known that especially in cases where discrete data should be used as explanatory variables, the models obtained using linear regression can produce statistically incorrect results. For such cases, it is recommended to use non-parametric or semi-parametric methods instead of linear regression. In this study, it is aimed to introduce the generalized additive models proposed in the presence of discrete explanatory variables, using a dataset obtained from the field of animal science (n=166). While lactation milk yield was used as the outcome variable, lactation period and maternal age taken from dairy goat were used as explanatory variables. It was determined that the variables of lactation milk yield, lactation period and maternal age did not show normal distribution. It was determined that the obtained linear regression model was statistically significant (P0.05), the variables were not normally distributed. In this case, it can be said to recommend to use of generalized additive models. As a result, it has been determined that when it is necessary to create a model with discrete variables in animal science studies, using generalized additive models instead of linear models can be recommended as it will increase the reliability of the model to be obtained.