What is Nominal Data?
Nominal data are divided into mutually exclusive categories that do not have a natural order, nor do they provide any quantitative information. The definition of nominal in statistics is “in name only.” This definition indicates how these data consist of category names—all you can do is name the group to which each observation belongs. Nominal and categorical data are synonyms, and I’ll use them interchangeably.
For example, literary genre is a nominal variable that can have the following categories: science fiction, drama, and comedy.
Nominal Data Examples
For the following examples, remember that the nominal definition means “names.” Each variable has category names, observations can fit into only one of these groups, and you can’t meaningfully order the groups.
Social science research often collects a lot of nominal data in the form of demographic information: gender, race, place of residence (e.g., state, country, county, zip code), marital status, employment status, and so on.
| Nominal Variable | Category Names |
| Gender |
|
| Blood Type |
|
| College Major |
|
| Area Code for Phone Calls |
|
Note that when categorical data use numbers, such as area codes, they do not provide numerical information. They’re still only names of groups.
Nominal Scale of Measurement
The nominal scale is the lowest type on the hierarchy of measurement scales. Consequently, these data provide the most imprecise information because you can only assign labels to the observations. Unlike ordinal data, you can’t even rank nominal data meaningfully—much less perform many of the usual mathematic operations on them.
| Nominal Data Analysis | Valid? |
| Frequency Distributions | Yes |
| Percentile Ranges, Median, Ranks | No |
| Subtraction, Addition, Mean, Standard Deviation | No |
| Multiplication, Division, Ratios | No |
Learn more about the Nominal, Ordinal, Interval, and Ratio Scales.
Analyzing Nominal Data
Even though nominal data limit the types of analyses you can perform, you still have options! Frequency tables using counts or relative frequency tables with proportions and percentages are great for this data type. Below is the relative frequency distribution for favorite ice cream flavors.
| Favorite Ice Cream Flavor | Percent |
| Chocolate | 53% |
| Strawberry | 32% |
| Vanilla | 15% |
Learn more about Frequency Tables and Relative Frequency Distributions.
Alternatively, bar charts and pie charts are great ways to graph categorical data, as shown below.
This clustered bar chart effectively graphs two categorical variables, ice cream preference and gender.
Learn more about Bar Charts and Pie Charts.
You can even perform hypothesis tests on nominal data. In fact, statisticians designed chi-square tests specifically for categorical data. Several different forms of this test have distinct goals. Click the links to learn more about each one and see example analyses.
The Chi-Square Goodness of Fit test determines whether the frequency distribution of a categorical variable differs from a hypothesized distribution.
The Chi-Square Test of Independence evaluates whether two nominal variables are correlated.
Multinomial logistic regression evaluates nominal outcome variable when you have at least one independent variable.
By definition, nominal data have many limitations, but they can still provide a great deal of insight!
Comments and Questions