Introduction

There are two concepts whose meaning is not related but completely related; this is Data Analysis and Scales of measurement. Refer to the blog on “from raw data to meaningful findings” on the explanation of data analysis. The two concepts are related because you cannot conduct any meaningful statistical analysis if you DO NOT KNOW the scale of measurement or the level at which you measured your variables. The type of data collected helps to determine the scale of measurement.

NOTE: The variable view of the SPSS requires you to assign the correct measurement level

What is Data?

Data can be defined as the quantitative or qualitative information of a variable that is obtained from the subjects of the research. Data can be in numbers, images, words, figures, pieces of information, facts or ideas. Do not confuse quantitative or qualitative information to refer to quantitative and qualitative research respectively. Data is quantitative (numerical) or qualitative (non-numerical) an under the quantitative we have discrete and continuous data.

What is measurement and scaling?

Measurement is the assignment of numbers or other symbols to characteristics of objects (variables) according to certain pre-specified rules while scaling is the generation of a continuum upon which measured objects are located.

There are four data measurement scales: nominal, ordinal, interval, and ratio.

• Nominal and Ordinal variables exist as categories while Interval and Ratio exist as numbers. Categories are counted while numbers are the actual count. For instance,
• What is your age? 40 years
• What is your age group?
• Below 20 years
• 21-30 years
• 31-40 years – x
• Above 41 years

In the above example, 40 (age) is a number (actual count) while 31-40 (age group) is a category. We can only count the number of people in that category and assign a number to each category (1- Below 20 years, 2- 21-30 years, 3 – 31-40 year and 4- Above 41 years). When entering these two variables on SPSS, 40 will be entered as it is because it is an actual number; it carries quantitative value while numbers 1-4 assigned to the age group categories are referred to as codes and they carry no quantitative value.

• Therefore, numbers assigned to nominal and ordinal scale carries no quantitative value while the numbers assigned to interval and ratio scale carries quantitative value.

Nominal Scale

Nominal deals with mathematical numbers assigned to a category and the number carries no quantitative value. These numbers assigned to these variables are just “tags” or “labels” to identify or classify an object; that’s why we say nominal names. Nominal tells whether the object belongs or doesn’t belong to a category. For instance, you are either a smoker or not a smoker; either a student or not. Therefore, nominal measures variables that can be categorized and the categories are mutually exclusive. Examples of variables measured at nominal level are gender, marital status, type of drink etc

Ordinal Scale

Ordinal data measures variables that are ordered according to quantity though difference between one level and the other is not equal i.e. adjacent ranks need not be equal in their differences. We normally say ordinal orders or rank but ranks are not equal. Just as nominal, ordinal deals with mathematical numbers assigned to data that is ordered according to quantity and these numbers carry no qualitative value. Examples of variables measured at ordinal level are educational level, age group, salary scale, grading system etc.

Interval Scale

Interval scale measures variables whose intervals between them are equal but do not have absolute zero i.e. one begins to measure not from an absolute zero point that represents total absence of a property but from an arbitrary zero e.g. attitude, satisfaction levels, temperature levels. 0oC does not mean absence of temperature. Interval data are therefore more powerful than ordinal scale due to equality of intervals. To remember interval scale, take note of no absolute zero.

Ratio Scale

Ratio scales measures variables from an absolute zero point. In other words, ratio variables are interval variables with a true zero point. For example, a scale designed to measure height, weight, age, income, would be a ratio scale

Summary

• Nominal and ordinal variables are called categorical variables. They exist as categories. This type of data is non-numerical.
• Interval and ratio variables are called continuous variables. They exist as numbers. This type of data is numerical.

Assignment:

Identify the measurement level of different variables.

Grading system, Gender, marital status, performance scores, standardized scores, age, temperature, salary scale, height, religion, gender, career, personal opinion, distance, age group, motivation level