Learning Objectives

Following this assignment students should be able to:

• understand the syntax of a ggplot2 object
• graph raw data to diagnose QC and completeness
• match a ggplot2 geometry function to the appropriate QC task

Reading

• Topics

  • Purposes and principles of data visualization
  • ggplot

• Readings

  • A protocol for data exploration to avoid common statistical problems
  • A Layered Grammar of Graphics
  • R for Data Science - Data visualisation

Lecture Notes

• Joins
• Principles and practice of data visualization
• ggplot
  • introduction & syntax
  • grouping & faceting
  • handling & plotting time series

---

Exercises

1. -- Join fish tables --

   # see what's available in this week's data folder
   list.files("DataViz-26March2018/data/", pattern = '.csv')
   
   

   Using the fish families data set in the data folder for this week, create a table that includes as columns:

   • fish species
   • fish family
   • count

   and show only those species with more than 7 counts.

   [click here for output]

2. -- Chlorophyll A histogram --

   Using the lakes dataset, generate a histogram of chlorophyll A levels in the samples. (Hint: use geom_histogram)

   You may see a message in the console that reads:

   `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
   

   Use the binwidth option in the geom function to create a histogram with data by bins of 10.

3. -- Scatterplot --

   Using the lakes dataset and ggplot2, create a bivariate scatterplot with algal species richness on the x axis and chlorophyll a on the y axis.

   What else would you add to this plot if you were trying to understand this data more deeply?

4. -- Grouping ggplot --

   Using the lakes dataset and ggplot2, create a bivariate scatterplot with algal species richness on the x axis and chlorophyll a on the y axis.

   Using the shape grouping argument in aes(), assign a different shape to each lack in the scatterplot.

   What does this plot look like?

   What if anything does the message in the console say?

   What might be a better way to plot this relationship for the different lakes?

5. -- Faceting ggplot --

   Using the lakes dataset and ggplot2, create a factor plot with treatment on the x axis and chlorophyll a on the y axis.

   Make this a faceted plot with an individual panel for each lake. Make sure to display all the data.

   Which lake is the most variable in its treatment response?