Ordinal level of measurement is the second level of measurement and builds upon the nominal level. Variables at the ordinal level possess all the characteristics of nominal variables but also have an inherent order or ranking among the categories. However, the differences between the categories are not necessarily equal or quantifiable. In other words, ordinal variables allow for comparisons of the order or rank, but not precise measurement of the magnitude of the differences.
Key characteristics of ordinal variables are:
1. Distinct Categories: Like nominal variables, ordinal variables have distinct categories or groups.
2. Order or Ranking: Ordinal variables have an inherent order or ranking among the categories, indicating that some categories are higher or lower than others. However, the size or magnitude of the differences between categories may not be uniform.
Examples of ordinal variables include:
• Likert Scale: A rating scale used in surveys where respondents indicate their level of agreement or disagreement on a statement, such as strongly agree, agree, neutral, disagree, and strongly disagree. The order of the responses represents a ranking of agreement.
• Educational Level: Categories representing different levels of education, such as elementary school, high school, bachelor's degree, master's degree, and doctoral degree. The categories have an order, but the difference between each level is not necessarily equal.
• Socioeconomic Status: Categories representing different socioeconomic groups, such as lower class, middle class, and upper class. The categories have an order, but the difference between each group is not quantifiable.
In statistical analysis, ordinal variables can be summarized using frequencies and proportions, and they can be displayed using bar charts or histograms. Additionally, non-parametric statistical tests, such as the Mann-Whitney U test or the Kruskal-Wallis test, are commonly used for analyzing ordinal data. These tests are suitable for comparing groups based on the ordinal variable while considering the order of the categories, but they do not assume equal intervals between the categories.
