Interactive Course

Regression Modeling in R: Case Studies

Learn how to practically implement GLMs and mixed effect models and assess these models to answer research questions.

  • 4 hours
  • 14 Videos
  • 49 Exercises
  • 360 Participants
  • 4,000 XP

Loved by learners at thousands of top companies:

roche-grey.svg
uber-grey.svg
credit-suisse-grey.svg
forrester-grey.svg
whole-foods-grey.svg
ea-grey.svg

Course Description

Now that you have a foundational understanding of generalized linear models (GLMs) and mixed effect models, a critical skill is knowing when and how to practically implement and assess these models to answer research questions. In this course, you will apply these modeling techniques to an ecological data set containing information about dragonfly habitat and behaviour. Here, you'll find that some questions can be answered using GLMs, while other questions require insight that can be gained using random intercept or random intercept and slope models. At each step, you will learn how to determine if your model is appropriate using model diagnostics, use the estimated parameters to make predictions about dragonfly abundance and behavior, visualize the relationships described by the models, and learn how to address problems that arise when the model you have built is not the right one for the job!

  1. 1

    GLMs

    Free

    Let's get started! In this chapter you'll learn what data exploration steps should be taken before you even start modeling. You'll apply and visualize predictions from both Gaussian and Poisson Generalized Linear Models (GLMs) in R and run diagnostic tests to determine if the models are appropriate for answering the research question being addressed.

  2. Mixed Effect Models I

    In this chapter, you'll move into the world of Mixed Effect Models, leaning how to practically apply a random intercept model to explore both fixed (population level) and random (within-site) effects. You'll extract key values from model output, and also learn how model fit methods influence the type of model comparisons that can be made during the model selection process.

  3. Extending GLMs

    Increase the complexity of your GLM by adding factors and offsets and examine the impact of this complexity on model diagnostics. You'll also learn how to apply Negative Binomial GLMs and the process and techniques of model selection. Being able to defend your chosen model is key!

  4. Mixed Effect Models II

    Let's extend our Mixed Effect Model to the random intercept and slope model! You'll learn why and how these models can be used and how to interact with the model output to extract key values and generate visualizations. This course wraps up with an overview of how modeling can be used as a tool and what pieces of information are essential to include in model reporting.

  1. 1

    GLMs

    Free

    Let's get started! In this chapter you'll learn what data exploration steps should be taken before you even start modeling. You'll apply and visualize predictions from both Gaussian and Poisson Generalized Linear Models (GLMs) in R and run diagnostic tests to determine if the models are appropriate for answering the research question being addressed.

  2. Extending GLMs

    Increase the complexity of your GLM by adding factors and offsets and examine the impact of this complexity on model diagnostics. You'll also learn how to apply Negative Binomial GLMs and the process and techniques of model selection. Being able to defend your chosen model is key!

  3. Mixed Effect Models I

    In this chapter, you'll move into the world of Mixed Effect Models, leaning how to practically apply a random intercept model to explore both fixed (population level) and random (within-site) effects. You'll extract key values from model output, and also learn how model fit methods influence the type of model comparisons that can be made during the model selection process.

  4. Mixed Effect Models II

    Let's extend our Mixed Effect Model to the random intercept and slope model! You'll learn why and how these models can be used and how to interact with the model output to extract key values and generate visualizations. This course wraps up with an overview of how modeling can be used as a tool and what pieces of information are essential to include in model reporting.

What do other learners have to say?

Devon

“I've used other sites, but DataCamp's been the one that I've stuck with.”

Devon Edwards Joseph

Lloyd's Banking Group

Louis

“DataCamp is the top resource I recommend for learning data science.”

Louis Maiden

Harvard Business School

Ronbowers

“DataCamp is by far my favorite website to learn from.”

Ronald Bowers

Decision Science Analytics @ USAA

Danielle Quinn
Danielle Quinn

PhD Candidate, Biology

Danielle Quinn is a PhD Candidate at Memorial University in Newfoundland, Canada, where she is developing and testing computational approaches to solving common challenges in marine conservation. She has been using R for a wide variety of tasks since 2009 and, more recently, is using Python's scikit-learn to apply machine learning techniques to marine conservation research. She has an MSc and BScH in Biology from Acadia University in Nova Scotia, Canada, with a primary focus in fisheries and ecology, and is the President and co-founder of Terranaut Club, a non-profit organization dedicated to providing hands-on science and nature exploration opportunities for girls.

Icon Icon Icon professional info