Measurement Scales in Research

In this article, we will discuss about Measurement Scales in Research. Research involves studying qualitative and quantitative variables. Qualitative variables cannot be measured through counting or using physical quantitative measures such as miles, kilogram, or any other unit. Quantitative variables can be measured either through counting or using physical quantitative measures. For example, temperature can be measured in Celsius or Kelvin, and distance can be measured in kilometer or miles. There are four different types of scales that are used in research. These scales are known as nominal, ordinal, interval, and ratio. Also, these scales are used to measure different types of variables depending upon the objective of the research and nature of the variable. However, a brief description of each scale is provided below.

Nominal Scale: Measurement Scales in Research

Nominal scale is used to measure a variable using different categories. Also, it is used to classify data into different groups or categories. Nominal scale does not allow performing many statistical processes on data. Such as, an example of nominal scale is “Gender”. There are only two responses to gender in a questionnaire. Nominal scale classifies sample to two different categories on the basis of gender. A researcher can only examine the percentage of male and female respondents in the total sample size. Frequency distribution tables are most commonly used to present data related to measurement of variables using nominal scale. Furthermore, Dichotomous questions are mostly used to measure variables using nominal scale.

Ordinal Scale: Measurement Scales in Research

Ordinal scale has characteristics similar to nominal scale. However, variables measured using ordinal scales can be ranked using a continuum. It is not only possible to categorize variables but also to rank them. Letters or numbers can be assigned to rank variables. Also, Rank questions are mostly used in surveys to use ordinal scale for measurement of variables. Variables can be ranked from worst to best or vice versa. For example, a customer is asked to rank factors that affect the perception of a customer about service quality of a restaurant positively.

Rank the following factors on a scale of 1 to 4 depending upon their importance in service quality

The above four factors will be ranked by customers on a continuum using numbers 1 to 4. It shows how ordinal scale can be used in a survey.

Interval Scale

The above example of ranking service quality factors using ordinal scale is based on logical ranking by a customer. However, the distance between these four numbers on a continuum can never be measured. In an ordinal scale, it is not possible to measure the distance between two points on the continuum. Furthermore, Interval scale has properties of nominal and ordinal scale, with an additional property of having equal interval or distance between two points. All points on a continuum have an equal distance. Unlike ordinal scale, interval scale is not based on logical ranking with unequal distance. Interval scale is mostly used to measure discrete variables. These are variables that can only assume whole number values within a range. Mean and median of data collected using interval scale can be calculated.

However, interval scale has a short coming. It does not have a true zero point or an absolute zero value. Examples of interval scale include temperature, GAT score, and credit score. These variables are measured on a scale without absolute zero.

Ratio Scale

Ratio scale have all properties of nominal, ordinal, and interval scale. It also overcomes the short coming of interval scale by providing an absolute zero value. Most of quantitative variables, such as weight, speed, and distance, are measured using ratio scale. Many statistical techniques can be used to analyze data collected using ratio scale. Measures of central tendency, measures of dispersion, correlation, and regression can be used to examine relationship between variables measured using ratio scale.

The table below shows the major differences between the four scales.

Scale	Application
Nominal	Used to categorize variables in different groups on the basis of similarity
Ordinal	Used to categorize variables and rank them in an order is based on logical ranking with no equal distance
Interval	Used to categorize and rank variables using equally distanced intervalsLacks an absolute zero point
Ratio	Used to categorize and rank variables using equally distanced intervals and has an absolute zero point

References