30  Building the Past

Readings

  1. geofacet for ggplot2 in R (strongly recommend experimenting with this type of graph)
  2. USA Boundaries if you haven’t read it yet

Guided Instruction

Review the Guided Instruction for the other tasks in this module if you haven’t yet.

You have been asked to support a story for the local paper (that has a web presence) that looks back on the housing collapse and the early effects of residential construction. You have data on residential building permits from 1980 through 2010 for each county in the United States. Your colleague that is writing the article would like a few maps and graphics that highlight the single family building permit patterns across your state as well as the patterns in the US.

Remember the big story is the collapse of new building permits at the initial stages of the 2007–2010 mortgage crisis. Make sure your graphics highlight the collapse in a clear and honest manner.

  1. Use the library(USAboundaries) package to get two spatial data sets. One for all US states, and the other for counties of a state of your choice (for example, counties of Idaho).

    • Look through your state data and county data and see if you can identify the columns containing the FIPS numbers. You will need these columns to merge this spatial data to our building permit data.

  2. Read in the permits data from the devtools::install_github("hathawayj/buildings") package. If you have trouble with devtools, try running remotes::install_github("hathawayj/buildings").

    1. Note: buildings is the name of the package. permits is the name of the dataset within the buildings package.

Wrangle the data into two separate data sets. One data set should look at single family permits at the state level for all US states. The other should look at single family permits at the county level for the state of your choice. -Look through your permit data and identify the FIPS columns.

  1. Use the FIPS number columns to join the state and county permit data to the state and county spatial data.

  2. Create a chart that shows, for each US state, the trend in single family building permits over time.

    • Make sure at least one (if not all) of your charts display the data using some sort of spatial/geographical representation (so geofacets, a chloropleth map, etc.).
    • Also, consider how you can effectively deal with the disparities of scale between geographies. (For example, the number of permits in California in a low year are much higher than the highest year for North Dakota.)

  3. Create another chart(s) that shows, for each county in your chosen state, the trend in single family building over time. Again, the chart should use some sort of spatial/geographical representation and you should consider how to deal with disparities of scale.

  4. Write a couple of paragraphs summarizing insights from your graphics for the reporter, and the choices you made during your visualization process.

  5. Render the .qmd file. Push all the files created in the rendering process into your GitHub repository.

Submit

In I-learn submit a link to the .md file on GitHub.