The goal of a canonical correlation analysis (CCA) is to find the linear combinations in two sets of variables that maximize the correlations. By analyzing the variance-covariance structure of these two sets of variables, uncorrelated pairs of linear combinations from the two sets of variables (called canonical variables) are created and their correlations (called canonical […]
Loops can be used in a variety of ways to tackle statistical analyses that would be overwhelming otherwise. For instance, loops can be used to run a particular model or produce a series of figures over a sequence of variables. In this workshop, we will cover several examples in which loops are utilized to make […]
Nonparametric statistical methods or distribution-free tests are statistical procedures that do not rely on assumptions about the underlying distribution of the data. These methods are useful when the assumptions of parametric tests are not met or when little is known about the population parameter of the variable of interest in the population. They are particularly […]
Multiple linear regression is a time-tested and powerful tool for inferring relationships in multivariate data. However, when the observations are associated with spatial locations, the independence assumption is often not appropriate. This workshop will motivate the use of models with spatially correlated residuals and provide hands on experience fitting and interpreting models in the R […]
Randomized controlled trials are considered to be the gold standard in research design due to removal of confounding through randomization. However, in many instances true experimental designs are not always practical, or even ethical. Propensity score analysis methods can reduce bias in estimating treatment effect that is introduced due to non-random assignment of experimental units […]
This workshop is designed for people who need to analyze their data using SAS. We will focus on the workflow, best practices, and procedures that any researcher should consider when faced with a new dataset and illustrate them in a hands-on fashion using SAS. Researchers should have access to a copy of SAS if they […]
Multinomial logistic regression is an extension of binary logistic regression. This statistical method can be implemented when modeling a dependent variable that is a categorical variable with more than two levels. This workshop will cover the mathematics of the multinomial logistic regression model, the interpretation of coefficients, model fit, and post-hoc tests.
In many fields of research, it is common to have several individuals rate a common set of study participants or objects. These measurements are almost always prone to various sorts of errors. Agreement statistics gauge how close the repeated measurements are by estimating the measurement error. Reliability statistics assess how well study participants\objects can be […]
Principal Component Analysis (PCA) is a popular multivariate technique used for data reduction. By analyzing the variance-covariance structure of a set of variables, uncorrelated linear combinations of the variables (called principal components) are calculated that maximize the amount of variance explained. The resulting principal components can then be used in further analyses such as regression […]
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 […]