Analyze Distributions

The Distribution platform illustrates the distribution of a single variable (univariate analysis) using histograms, additional graphs, and reports. The word univariate simply means involving one variable instead of two (bivariate) or many (multivariate). However, you can examine the distribution of several individual variables within a single report. The report content for each variable changes depending on whether the variable is categorical (nominal or ordinal) or continuous.

•

For categorical variables, the initial graph is a histogram. The histogram shows a bar for each level of the ordinal or nominal variable. The reports show counts and proportions.

•

For continuous variables, the initial graphs show a histogram and an outlier box plot. The histogram shows a bar for grouped values of the continuous variable. The reports show selected quantiles and summary statistics.

Once you know how your data are distributed, you can plan the appropriate type of analysis going forward.

Note: For details about the Distribution platform, see Distributions in the Basic Analysis book.

Scenario

You want to view the nutritional values of cereals so that you can eat a more healthful diet. Analyzing distributions of cereal data reveals answers to the following questions:

•

Which cereals are highest in fiber?

•

What is the average, minimum, and maximum number of calories?

•

What is the median amount of fat?

•

Which cereal contains the most fat?

•

Are there any outliers in the data?

Create the Distributions

Select Help > Sample Data Library and open Cereal.jmp.

Select Analyze > Distribution.

Press Ctrl and click Manufacturer, Calories, Fat, and Fiber.

Click Y, Columns and then click OK.

Figure .2 Distributions for Manufacturer, Calories, Fat, and Fiber

In the Fiber distributions, notice the following:

–

Fiber One and All-Bran with Extra Fiber contain the most fiber as shown in the Fiber box plot. These cereals are outliers in terms of fiber content.

The row that contains Fiber One in Cereal.jmp is labeled. This label shows the name of the cereal next to a data point in graphs. To see the entire label, drag the right-most vertical border to the right. Place your cursor over the unlabeled data point to see “All Bran with Extra Fiber”.

In the Fat distributions, notice the following:

–

Place your cursor over the top data point (the x marker) in the Fat box plot to see that 100% Nat. Bran Oats & Honey is the highest in fat.

–

In the Fat Quantiles report, the median amount of fat is 1 gram.

In the Calories Quantiles report, notice the following:

–

The maximum number of calories is 250.

–

The minimum number of calories is 50.

In the Manufacturer histogram, click on the bar for Nabisco.

Figure .3 Distributions for Nabisco Cereals