Multimodel inference (MMI) is a model selection framework that has recently gained some popularity as an alternative to null hypothesis significance testing. This type of inference favors stepwise approaches (forward and backwards model selection) to determine a single best “final” model. This workshop gives an overview of the MMI framework, in which researchers generate a number of possible candidate models based on literature knowledge or current hypotheses, and fit every model to the data. Researchers then use information criteria (e.g. AIC, BIC…) to rank the models and assign each a weight according to how well it is supported by the data. Model inference can then be made in a number of ways: a few of the best-ranking models can be discussed, a variable importance score can be calculated for each variable considered in the set of models, and predictions and parameter coefficients can be averages over the set of model predictions or coefficients.