The Missing Pieces of ANOVA, Regression and Other Linear Models
Details
🗓 Feb 24 | Mar 3 | Mar 10 | Mar 24
🕓 Tuesday, 3-4 pm Pacific Time
⎈ Zoom
What will be taught
- How to evaluate the model assumption of normality and how to handle deviations from this assumption.
- How to evaluate the model assumption of equal variance;
- How and when to model heteroscedasticity
- How to interpret common outputs from linear models
- Working with model estimates, aka “least squares means”, “estimated marginal means” and “emmeans” to answer your scientific questions
Workshop overview
You’ve taken the stats class, but what do you remember? Can you confidently run an ANOVA and interpret the results, even when things look wonky? Are p-values the only thing you look at? (Your model can tell you so much more!) Experiments require a massive effort to design, set-up and see through to the end. Make that effort worth it with a solid analysis. This is a general topical webinar and does not require knowledge of specific statistical program. Examples will be shown in R, but
Intended audience and how to prepare
This workshop is open to scientists, students, technicians and anyone else who analyzes data requiring a linear model. Previous experience using linear models to any extent is assumed, but deep mastery is not (this is a refresher webinar series, after all).
Attendees are not expected to follow along with coding, but all scripts will be made available after the workshop (when applicable). The webinars will be recorded and posted online.
Schedule
| Date | Topic | Links |
|---|---|---|
| Feb 24 | Model Assumptions: Normality. How do we identify this and what do we do when it’s violated? What can we do about outliers? An exploration of this and brief discussion of generalized linear models. | slides | code |
| Mar 3 | Model Assumptions: Equal Variance. How do we identify this and what do we do when it’s violated? How to identify and model heteroscedasticity and what to do it for repeated measures/longitudinal data. | slides | code |
| Mar 10 | Linear Model Outputs: What Does It All Mean? Understanding how to read and interpret output from a linear model. This will cover different model types (ANOVA, regression, repeated measures). | |
| Mar 24 | Making Sense of Your Estimates. Treatment effects are frequently the final outcomes of a statistical analysis. Let’s talk about how to leverage these results to answer your research objectives. |
Meet Your Instructor
Julia Piaskowski is an agricultural statistician at the University of Idaho, Software Carpentry Certified Instructor and a long-time R programmer.
This workshop is licensed under the Creative Commons Attribution-NonCommercial 4.0 International License.