Chapters 4 and 5

• Some Sweet Data
  • FiveThirtyEight.com has some interesting data and articles.
  • They used a on line poll to allow users to compare fun-sized candy types.
  • They collected this data.
  • And produced this article.
    • They claim 8,371 different ip addresses -> probably means users.
    • They claim 269,000 matchups -> data points.
    • average = 32 and media 11 matchups from each ip.
  • They do not claim this is a scientific study in any way.
• Let's grab the data
  • There are many ways, but let's try Excel's import data function.
  • Start excel.
  • Grab the data
    • Data tab, Get External Data select From Web
    • 
    • Tell it to stop worrying about errors on the edinboro page.
    • Enter the url https://github.com/fivethirtyeight/data/blob/master/candy-power-ranking/candy-data.csv
    • It would have been cool if it found the table, but it didn't so select the arrow at the top of the page.
    • 
    • Select Import
    • After a brief delay, this will import the entire web page.
  • Clean up.
    • In my download, I needed to delete rows 1-65 and another set at the bottom of the page.
    • In needed to delete column A
    • I replaced Õ with a ' (as in HersheyÕs Kisses became Hershey's Kisses)
    • 
    • I formatted the Sugar Percent, Price Percent columns nicely
    • 
    • I added a column for a derived Win Percent and converted column M to a percent.
    • 
    • I hid column M as It is no longer needed.
    • I renamed the headers, made them bold and word wrapped
    • 
    • Note: This should all be added to the methods document, Data Cleaning section.
• Documentation
  • Multiple sources state that you should provide an "About" or "Welcome" worksheet
    • I kind of like this document
    • I have also seen this disputed, but for now, let's do it.
  • Insert a About worksheet.
    • Author
    • Date
    • Reason
    • Source of data.
    • A list of other sheets in the workbook and what they contain.
    • Data Dictionary
  • 
  • I am not the best at enforcing a particular style on anyone
    • But you should have an about page on anything you are going to tun in
• Save this somewhere you will have access to it later. We will start with it on the next set of notes.