Week 2: Data Exploration

January 15-16, 2018

Before you can fit models to data, you have to understand the data you’re plotting. In this lecture we discuss the concept of Data Exploration.

We will spend most of the time discussing the step-by-step model outlined in Zuur et al (2010).

*Note: We actually did this lecture on Jan 22-23

Lecture Topics

  • Understanding study design
  • Finding key points from the Methods section in a paper
  • Executing an eight-step data exploration:
    1. Outliers Y & X
    2. Homogeneity Y
    3. Normality Y
    4. Zero trouble Y
    5. Collinearity X
    6. Relationships X and Y
    7. Interactions
    8. Independence Y

In-class Activities

We will:

  • Do an exercise where we sketch out a study design
  • Execute two versions of a data exploration (simple two-variable. Harder full exploration)

Pre-class Prep

Please skim Loi et al (2017). We will have ten minutes to read it in class, so a detailed examination is not necessary.

Loi B, Guala I, Pires da Silva R, Brundu G, Baroli M, Farina S. (2017) Hard time to be parents? Sea urchin fishery shifts potential reproductive contribution of population onto the shoulders of the young adults. PeerJ 5:e3067 https://doi.org/10.7717/peerj.3067

If we are proceeding quickly, we will also try to sketch the study design of:

Peiffer F, Bejarano S, Palavicini de Witte G, Wild C. (2017) Ongoing removals of invasive lionfish in Honduras and their effect on native Caribbean prey fishes. PeerJ 5:e3818 https://doi.org/10.7717/peerj.3818

In addition, at some point you will need to read in detail Zuur et al (2010).

Zuur, A.F., Ieno, E.N., Elphick, C.S. (2010). A protocol for data exploration to avoid common statistical problems. Methods in Ecology and Evolution 1(1): 3-14.

Additional Resources

For historical interest:

Don’t worry, most data in ecology are pseudoreplicated to some extent and we will discuss ways to deal with that.

Files

Slides

Slides via speakerdeck