The Beetle Mortality Experiment

Let’s get a brief overview of binomial counts and proportion data.

We'll cover the following

R packages

We’ll use the following R packages in this chapter:

  • ggplot2
  • arm
  • AICcmodavg

Binomial counts

Binomial counts appear when we have a known number of occasions (binomial trials) where something of interest may occur. We also know how many times something happened (successes) and how many times something didn’t happen (failures). The successes and failures summed together give us the binomial denominator. Binomial count data can be expressed as proportions, particularly, as the proportion of successes. However, if the binomial denominator (the size of the trials) varies, then it’s better to work with the binomial count data. This is because our analysis can weigh the larger trials more than the smaller ones instead of treating all proportions equally. However, sometimes the size of the trials isn’t recorded and we have to work with the proportions.

Get hands-on with 1200+ tech skills courses.