What is nominal data
As mentioned before, nominal data includes different categories which can present some properties of respondents; e.g. gender, marital status, etc. Defining the categories is an important step and you need to carefully choose the ones which are meaningful for you and your work. It is necessary to determine your goal for defining such categories. In other words, you should always think about why you need these information and what you want to do with it.
As the first step you may identify the percentage of responses in each category. One common and efficient way to present nominal data is to use contingency tables (cross tab). This tables show how responses differ by each category.
As an example, let’s consider the following survey question and 100 fancy responses dealing with customer satisfaction:
Question: What do you like most about our product?
Now let’s split the responses into two groups, namely, male and female. Now let’s say we can we can generate the following contingency table for our responses:
|Total/share||Quality||Various color||Ease to use||Total|
|Male||72% (18/25)||4% (2/25)||20% (5/25)||25% (25)|
|Female||29,3% (22/75)||37,3% (28/75)||33,3% (25/75)||75% (75)|
|Total||40% (40)||30% (30)||30% (30)||100% (100)|
In case you like to go deeper in your data to investigate and measure the relationship between nominals, you can use some standard statistical test like chi-square test or multi-way tables. How to implement these methods and interpret the results, needs some college level knowledge of statistical inference, and therefore, are out of the scope of this article.