This course provides an overview of data science tools and best practices that can be used to create transparent and reproducible workflows when working with environmental data. Students learn how to translate raw data from field and lab studies into databases and “tidy” digital formats, which can then be used for plotting, statistical analyses, etc. Students learn how to track the history of file changes (version control), collaborate online with others, and generate “recipes” for re-creating one’s work. Students also learn how and where to ask for help when attempting something new (e.g., How do I create X from Y?), debugging or fixing code (e.g., What does this error message mean?), etc.
This course is intended to give students an overview of the theory and practical aspects of fitting time series models to fisheries and environmental data. The course covers topics ranging from autocorrelation and crosscorrelation, autoregressive (AR) and moving average (MA) models, univariate and multivariate state-space models, and estimating model parameters. Co-taught with Eli Holmes (NOAA) and Eric Ward (NOAA).
This course introduces students to a large class of statistical models commonly used for analyses in ecology and environmental science. These include simple linear models, such as analysis of variance (ANOVA) and regression models, along with Generalized Linear Models (GLMs) (e.g., logistic and Poisson regression) and mixed-effects models (LMM, GLMM). Students learn how to choose model forms and structures based upon questions being asked and the nature of the data.
This course includes the development of scientific ideas into coherent proposals, as well as addressing other forms of communication, and aspects of professional development to help graduate students during and after their time at the University of Washington.