5 Lab Exercise 4
In 2012, a researcher in coastal California was looking for ways to better assess populations of larger mammalian predators for conservation purposes. Their solution? To determine how effectively morphometric traits (shape and size measurements) from animal dung could be used to distinguish particular species. To do this, they collected samples and identified the species using DNA, and then conducted a series of measurements on each sample like length and diameter, as well as shape characteristics such as tapering and segmentation.
Your academic supervisor came across this dataset and wants to know what sorts of things you would look for in this data before you both go off to spend the summer in the woods collecting animal scat. Like a good data scientist, you’re going to do a bit of EDA to take a look at the data first.
Using the predatorScat.csv data file, produce a script that demonstrates how you would look at this data. In particular, I want you to be able to show do the following:
Look at 2 different kinds of data in a univariate analysis, and describe the kinds of patterning you see (hint: be careful that numbers are not representing categories!)
Look at 2 different kinds of relationships in a bivariate analysis (e.g., two numerical, two categorical, numerical-categorical), and describe the patterning you see
Choose numerical summaries metrics to look at central tendency and spread of one of the numerical variables
Identify any outliers in the data and whether they appear to be genuine or suspect
Write your findings as comments in the script alongside the code used to produce the graph and submit the script on Canvas.