In marketing, metrics provide a clear picture of how a campaign is performing. They also help businesses make informed decisions that improve performance.
In mathematics, a measure is a set function that assigns to a set of sets the value 1 if any of its subsets is countable and adds up to 1. Examples of such measures are Lebesgue measure, counting measure, and circular angle measure.
Nominal Scale
A nominal scale, also called a naming scale, is the simplest of all measurement scales. It consists of categories without any inherent order or hierarchy (think of it like putting things into different buckets, for example). A nominal scale is a good way to categorize qualitative answers when conducting surveys and gathering non-numerical responses, such as gender, nationality, or eye color.
This is a useful scale for measuring demographics, as it allows for the exhaustive classification of people into their respective groups. It can be used for a wide range of purposes, including customer demographics, which allow digital marketers to tailor their content and ads to specific groups of individuals. Understanding the levels of measurement enables researchers and data analysts to choose the appropriate analytical methods, ensuring the integrity of their research results. For more information, see CUNY’s Measurement Guide or the helpful resources listed below. Often, these resources compare the differences between nominal and ordinal scales.
Categorical Scale
The categorical scale is a type of measurement scale that represents different categories. It can be further divided into two types: nominal and ordinal. Both have their own unique properties and applications.
Nominal data has no inherent order and ranking, while ordinal data can be ranked in a definite order. Some examples of this type of data include Likert scales, school grades (Elementary, High School, College and Postgraduate), or movie ratings (1-5 stars).
However, these data are limited in their ability to provide quantitative analysis as they do not have a standard interval scale. For example, respondents cannot gauge their options before responding to a question such as “Do you agree or disagree?” Categorical data also has low measurement sensitivity and requires large samples to obtain accurate results. This limit on statistical analysis limits its use for certain research questions and objectives. Continuous level measurement will always increase statistical power and accuracy. Therefore, choosing the right measurement scale is crucial to ensure accurate and meaningful results.
Hierarchical Scale
Hierarchy is a structure that places one element above another. It is used in a variety of fields, including business, organizational theory, systems analysis and computer science. Hierarchies are also prevalent in art, where artists use it to convey importance.
Artists use hierarchical scale when they purposely enlarge a certain form in a composition to communicate its relative importance. This is most often seen in the artwork of ancient cultures, where kings and other significant figures were depicted larger than their attendants or lesser figures to demonstrate their power and authority.
Michelangelo’s “David” is an example of a work that uses hierarchical scale to evoke feelings of awe and reverence in viewers. The large scale of the statue emphasizes David’s strength and defiance, while his youthful appearance contrasts with his size, evoking emotion and respect in spectators. Hierarchical scale can also be found in religious artwork, where saints and other divine figures are depicted as larger than human figures to reinforce their spiritual significance.
Complex Scale
Understand the four research scales (Nominal, Ordinal, Interval, Ratio) to interpret and present data accurately. This is important because the ASWB exam often includes scenario-based questions that require knowledge of these scales. Using the wrong type of scale can lead to inaccurate conclusions about client progress or program effectiveness, which could have ethical implications.
Nominal scales classify data into distinct categories without a specific order or ranking, such as gender or colors. Although they can be labelled with numbers, these numbers do not have a specific numerical value or relationship and therefore cannot be used in arithmetic computations. Examples include satisfaction ratings from a survey where respondents rate their level of happiness from one to five.
The interval scale has the properties of the ordinal and the categorical scales, as it shows both the order of data points and their exact differences from each other. However, unlike the ratio scale, interval data points can be added and subtracted but cannot be multiplied or divided. This is because interval scale variables have a true zero point, like degrees or a weight measurement.