A Second Project: Basic Data Analysis

• In any project, the next step is basic data analysis.
• We need have five fields, so we need to analize these fields.
• This is just the work we have done before.
  • For numeric data
    • Five number summary
    • Possibly box and whisker plot
    • Frequency distribution and histogram
    • Count or identify bad data.
  • For text data
    • Identify unique words
    • Frequency
    • Perhaps rank
    • List of most common
  • I understand that this is a repeat of what we have done before but you need to do it to identify problems with the data.
• In addition, this work will help you find problems in the data and indentify where it needs to be "cleaned"
• I will build a new tab for each field.
  • Since this is a small data set, I will copy the data to the new tab.
    • This will allow me change the Sorted Data sheet without interfering with the other tabs.
    • If the data were large, I would not do this.
• Homework
  • Make a new tab for homework.
  • Copy the homework data to that tab.
  • Sort this new columnm
  • Build a five number summary
    • I use quartile.exe because that matches the box and whisker plots.
  • 
  • Build a box and whisker plot
    • Use the data in column A, not the computed summary data.
    • Select Layout 2
    • 
    • The 1 at the bottom is annoying, and useless.
      • Turn it off by selecting
        • Axes, Primary Horizontal
        • 
    • Add a title as well.
    • 
    • I find the 120% at the top distracting as well.
      • Select More Axis Options
      • 
      • Using the Axis Options menu , select Vertical Value Axis
      • 
      • You have three choices at the top
        • Fill & Line
        • Effects
        • Axis Options
        • 
        • Select Axis Options
        • And change the Maximum to 1.0
        • 
      • 
    • We need to make sure we pay attention to the graphs we create.
      • What does this graph tell us?
      • Half of the homework grades are above an 81.
      • The average homework grade was a 70.
      • The top 75% of the class has lots of near 100% scores
      • The bottom 25% is very spread out.
      • We can verify this by looking at the data.
      • But in a very large dataset this might not be possible.
    • While we are here, build a frequency distribution.
      • By grades, so start at 59% (grades 0 - 59)
      • Go up by 10%
      • 
      • Add in a column for Grade.
      • 
      • Produce a column chart based on grade and frequency
      • 
      • By the way, the width of the columns can be adjusted with series Options
      • 
    • We can insert a histogram of this data as well
      • Select column A
      • Insert a histogram
      • 
      • Change the number of bins
        • More axis format
        • Axis options
        • Change number of bins.
        • 
      • Fix the title, lables and vertical axis
      • 
      • Why are the two counts off?
    • We should probably count the number of valid data as well.
      • I just tacked this on top of the 5-number summary.
  • Do we think there are any errors in the data?
  • Do we think that the data needs cleaned in any way?
• Now that we have a basic analysis of the homework field, we should go to the report and add the information.
  • In the data dictionary, I would add
    • A description of the data
  • In the Summary of data used section I would include
    • The five number summary.
    • The histgram.
    • 
  • I would save the other graphs.
    • They could go into the Summary section.
    • You might use them in answering the question.
  • We really did not do any data cleaning, so there is noting to note here.
  • We really did not do any advanced computations, so there is nothing to note in the Analysis Methods section.