Nominal, ordinal, or numerical variables?
Posted on 24th July 2015 by Alyssa G. Ricci
When determining what statistical method to use to analyze your data, it is essential to know the type of variable you are working with. What is the key to success? Make sure you look at the data in context to determine how the study calculated the results of a specific variable. Did the study look at pain as a number from 1 to 100? Did they just ask participants a simple yes or no question? Or did they convert a numerical answer to fall in a particular category?
Nominal data is classified without a natural order or rank, whereas ordinal data has a predetermined or natural order. On the other hand, numerical or quantitative data will always be a number that can be measured. In the example previously alluded to, the presence or absence of pain would be considered nominal data, while the severity of pain represented by categories such as none, mild, moderate, or severe would be ordinal data. If pain was analyzed on a Visual Analog Scale from 0-100, where 100 would be two times as painful as 50, this would be numerical data.
In a study by Pochapski, et al. titled Effect of pretreatment dexamethasone on postendodontic pain, pain is analyzed in two different ways, yielding different statistical tests used to analyze the appropriate variables. In figure 1, the numeric rating scale is used to record pain for each group at each time point in the study. In this case, pain is a numerical variable. However, in figure 2, pain intensity is analyzed in different categories—none, mild, moderate, severe. In this case, pain is an ordinal variable. Because the variable types are different in each case, the statistical test used to calculate results will be different as well. It is very important to properly identify the type of variables used to analyze data in order to choose the correct statistical tests when calculating results.
Pochapski, Márcia Thaís, et al. “Effect of pretreatment dexamethasone on postendodontic pain.” Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and Endodontology 108.5 (2009): 790-795
Nominal, ordinal, or numerical variables? by Alyssa G. Ricci is licensed under a Creative Commons Attribution-NoDerivatives 4.0 International License. All images used within the blog are not available for reuse or republication as they are purchased for Students 4 Best Evidence from shutterstock.com.