Free Trial

Safari Books Online is a digital library providing on-demand subscription access to thousands of learning resources.

  • Create BookmarkCreate Bookmark
  • Create Note or TagCreate Note or Tag
  • PrintPrint

3.6 Summary

  • Data can be described graphically and numerically.

  • Continuous and nominal data require different graphs and numerical summaries.

  • JMP uses appropriate tools depending on whether a column is either continuous or nominal.

  • Continuous data can be graphed by histograms, box plots, time plots, and CDF plots.

  • Continuous data histograms show the distribution of values by tabulating the frequency of occurrence per class (or bar) width.

  • Box plots summarize the median, the lower and upper quartile, as well as minimum and maximum values.

  • Box plots of few data values can distort the data distribution.

  • Box plots of data where a large portion of values is concentrated in a narrow band may give undue focus on values outside the concentrated data range.

  • Time plots graph data over time. They can be used to discover temporal patterns.

  • CDF plots show the cumulative frequency of occurrence.

  • Continuous data can be summarized by measures of location (mean, median, and quantiles) and measures of spread (standard deviation, variance, and range).

  • The mean is a simple average of the data values.

  • The median is the value with at least 50% as large as and at least 50% as small as it.

  • The upper and lower quartile split the data into the lower 25%, the middle 50%) and the upper 25% respectively.

  • JMP uses interpolated quantile values. The median, quartiles, or other quantiles, except the maximum or minimum, need not be data values.

  • Nominal data can be graphed in frequency histograms, mosaic plots, and pie charts.

  • Nominal data can be numerically summarized by frequency counts and relative frequencies per nominal category. In a nominal frequency histogram, the horizontal axis represents outcome categories.

  • Relative frequencies stress the proportional occurrence of each nominal category.

  • Relative frequency can be graphed in a mosaic plot or a pie chart.

  • A good graph has all axes labeled and does not contain unnecessary information.

  • A good graph has easily distinguishable marker symbols.

  • A good graph has properly spaced axes.


You are currently reading a PREVIEW of this book.


Get instant access to over $1 million worth of books and videos.


Start a Free 10-Day Trial

  • Safari Books Online
  • Create BookmarkCreate Bookmark
  • Create Note or TagCreate Note or Tag
  • PrintPrint