How to Compare Nominal Data

Save

Sometimes your statistical task involves presenting nominal data: That is, data that consists of counts or tallies of one or more named variables. Election results are an example of this kind of data, where the candidates and the polling places are the named variables, and you're tallying votes for each candidate at each polling place. When you're dealing with nominal data, you can use a name, or a code, such as a number or a letter of the alphabet, to label what you're counting. For example, you may compare the numbers of people who have bought various makes of car.

Things You'll Need

  • Calculator
  • Graphing software (optional)
  • Statistical software (optional)
  • Ruler (optional)
  • Colored pencils (optional)
  • Count the number of dimensions in your data. If you are comparing sales of different makes of cars, the Torpedo, the Lightning, and the Jazzcar, you have one dimension: Make of Car. If you are comparing sales of these makes of cars in four cities -- New York, Atlanta, Seattle, and Los Angeles -- you have two dimensions: Make of Car, and City.

  • Tabulate your data. Make a one-dimensional table if you have one dimension of data. List the three car makes in three columns across the top -- one column labeled "Torpedo," one labeled "Lightning" and one labeled "Jazzcar." List the sales figures under each make.

  • Construct a two-dimensional table if you have two dimensions of data, such as "Make of Car" and "City." Label the three columns as described in Step 2. List the four cities as labels for four rows down the left side of the table. Place the appropriate sales figures for each make of car in each city in the cells formed by the intersections of the columns and the rows.

  • Illustrate your data. Use bar charts to show the relative numbers of cars sold by make for one dimension of data, or by make within each city for two dimensions of data. You may draw these charts by hand, using a ruler and colored pencils, or use graphing software.

Tips & Warnings

  • If you are considering additional nominal data, such as gender of the car buyer (female or male), you could make a three-dimensional table of data. Three dimensions can be tricky to tabulate and chart, however. Comparing four or more dimensions is sometimes impractical.
  • Make sure you understand what “nominal data” means, and how this differs from other types of data. This all comes under the general topic of “levels of measurement,” or “measurement of variables.”
  • Photo Credit Leaning Cars image by Michael Wuelfrath from Fotolia.com
Promoted By Zergnet

Comments

You May Also Like

Related Searches

Check It Out

Are You Really Getting A Deal From Discount Stores?

M
Is DIY in your DNA? Become part of our maker community.
Submit Your Work!