Lesson 6: Percentiles and Boxplots

September 6, 2016

Wed, Sep 7

Review:

Measuring Center with Mean and Median
Problems with Median – SORT!
Tutoring available Thursdays 4-6pm in GCB 201 with David Mould

Presentation:

Percentiles
- What are Percentiles?
- How to calculate Percentiles
- To find the xth percentile
  - calculate x/100 * count (rounded to the nearest integer)
  - result is the position of the value in an ordered (smallest to largest) data set
  - 25th percentile = 25/100 * count
  - 50th percentile = 50/100 * count = Median
  - 75th percentile = 75/100 * count
- Example
Boxplots
- What are Boxplots
- Five number graphic summary
  - 25th Percentile
  - 50th Percentile = Median
  - 75th Percentile
  - Minimum value
  - Maximum value
- Examples
  - Stacking Bar Charts, Histograms and Boxplots
  - Boxplot paired with Histogram
- Video: Boxplots
Quartiles and Inner Quartile Range (IQR)

Example: Kick/Punt returns by Charles Nelson during Oregon vs UC Davis game Sat, Sep 3
- {-2, 16, 27, 31, 33, 38, 46, 62}
- n = 8
- mean = 31.4
- median = 31 (position 4)
- 25th percentile = 16 (position 2)
- 75th percentile = 38 (position 6)
- Min = -2
- Max = 62

Activity:

Product Boxplots (by hand)
- Find the five number summary for Female height and Male height from the Student Survey data
- Draw 2 boxplots, side by side, for comparison purposes.
- Use the same scale for both box plots
- Repeat using the World Population estimate data
  - Draw 4 boxplots, one per time period, side by side
  - Use the same scale for all 4 boxplots

Assignment:

Read pp. 34-38 (Quartiles, Five number summary, Boxplots)
Watch Boxplots with Google Sheets
Use Google Sheets
- Make a Copy of RE East Pueblo
- For each of the 3 neighborhoods: Belmont, Eastside and University
  - Use the “Selling Price” variable (see Column I)
  - Generate the 5 number summary
    - Calculate the mean, median, 25th and 75th percentiles
    - Identify the min and max values
- Produce 3 boxplots
  - One boxplot per neighborhood, i.e., Belmont, Eastside and University
  - Plot together on the same chart for side-by-side visual comparison
  - This means using the same scale for all 3 boxplots
  - You’re welcome to use Sheets or you can do it by hand (if done neatly)

Justin

Justin Holman is CEO of Aftermarket Analytics, where he leads efforts to develop cutting edge sales forecasting and inventory optimization technology for the Automotive Aftermarket. Prior to joining Aftermarket Analytics, Justin managed corporate consulting for the Strategy & Analytics division at MapInfo Corporation, leading major projects for retail clients including The Home Depot, Darden Restaurants, Bridgestone-Firestone, Sainsbury’s and New York & Company. Before that, Justin served as Vice President of Software Development at LogicTools, now part of IBM's supply chain application software group. Justin holds a B.A. from Claremont McKenna College, a Ph.D. from the University of Oregon and an Executive Management certificate from Northwestern University's Kellogg School of Management.

2 Comments


Rique Lucero 8 years ago

I used sheets on my android phone. To be able to use the candlestick trend chart, I only get to use 4 arguments, not 5 as per the book. It asks for the min, open, close and max. So I used the min value, Q1 for open, Q3 for closed and max value. I then created the mean and median values separately so they will be charted next to their respective data group (belmont, mean, median….eside,mean,median) would this be okay for tests, or do I HAVE to have the median in the candlestick?
- Reply
  
  Justin 8 years ago
  
  Sounds like you found a good trick for making boxplots. If the visualization works well in terms of showing the spread of the data and how the different neighborhoods compare then I’m good with it.