In the previous sections, you learned about the different types of variables. Specifically, you learned that there are two main types of variables you will encounter in statistics: qualitative and quantitative variables. While quantitative variables are numerical, qualitative variables tell us about the properties of any unit, individual or object.
While in the previous section you may have learned about the two types of quantitative variables, discrete and continuous variables, in this section you will be introduced to the two types of qualitative variables: ordinal and nominal.
Types of Qualitative Variables
Every few years, in a massive effort of coordination, the European Commission publishes the results of a survey called the Eurobarometer. This survey, which started almost 50 years ago, touches upon many aspects of the lives of people living in Europe: culture, health, climate change, poverty and more.
While quantitative information about populations is very important, such as the average income of the elderly or the number of people in a country, many decision-makers often need qualitative data instead. Surveys like the Eurobarometer are the perfect example of the different types of qualitative data that can be gathered from a population. These include questions like the items and electronics people have available at home or tasks like rating life satisfaction.
Understanding the differences between different qualitative variables is important for understanding the data and interpreting its meaning. Let’s start by understanding the two types of qualitative variables.
Ordinal and Nominal Variables
A nominal variable is another name for qualitative, or categorical, data. Nominal variables have two or more categories to describe a person or thing and do not fall under any kind of order. Meaning, nominal variables are not ordered according to any sort of preference, satisfaction or worth. For example, in our eye colour example we had a nominal variable with five different categories:
These categories are mutually exclusive, which means that someone with brown coloured eyes has only brown coloured eyes and not any other colour in any other category. Someone with blue eyes doesn’t have grey eyes, just like someone with hazel eyes doesn’t have green eyes. If something is mutually exclusive, it can either be one thing or another, but it can’t be both. Nominal variables don’t have any intrinsic order, which means that they do not have a predetermined order based on worth or preference.
If this is confusing, it might be helpful to think of the opposite situation, which is captured in ordinal variables. Ordinal variables are qualitative, or nominal, variables that have a predetermined order. This can be something like level of satisfaction, ordered into three simple categories:
- Very happy
- Very unhappy
These categories are also mutually exclusive, but in this case they do reflect a certain order. Order, in this case, can be misleading because it doesn’t mean that changing the order of our three categories will change their meaning. A better word to use when thinking about ordinal variables is “scale.” On a scale of satisfaction, we have three different levels that have a predetermined order, or a predetermined place on this scale.
This scale, going from left to right, reflects the order of our categories. Either someone is very happy, high on the scale, okay, at the midpoint of the scale, or very unhappy, which is at the lowest part of the scale. Thought about in this way, we can see that nominal and ordinal variables are very different. Here are some other examples to help you get a better picture of these differences.