# Types of measurement scales in survey questionnaires

Measurement scales while framing a questionnaire play an important role to understand the characteristics of the variables. A questionnaire is a collection of a set of questions. A survey, on the other hand, includes the process of making the questionnaire, collecting the responses, aggregating the data and analysing it (SurveyMonkey, 2019). Therefore, the questionnaire is a subset of a survey. A previous article discussed the important points to remember while making a questionnaire. It also mentioned about two different types of questionnaires.

These are:

- closed ended or structured and,
- open ended or semi-structured or unstructured questionnaires.

This article focuses on the types of measurement scale used in a closed ended questionnaire.

A closed-ended questionnaire developed for a research purpose typically consists of a number of sections. The sections depend on the study topic and its objectives. But the questions contained in these sections can broadly be classified into two categories. These are

- Questions for descriptive analysis.
- Questions for inferential analysis.

The questions for descriptive analysis collect information on the demographic profile (age, gender, educational qualification among others) and general background of the survey participants. In contrast, the questions for inferential analysis are framed to collect data on the dependent and independent variables of the study. However, each question in a questionnaire uses one of the specific measurement scales.

## Importance of measurement scales

Take the example of research to investigate the effect of organisational factors on job satisfaction among employees working in the IT industry in Delhi-NCR. For the primary study make a closed-ended questionnaire to collect data. The following measurement scales can be used for different variables in the questionnaire.

## Nominal scale

A nominal scale classifies variables that do not involve any numerical value. The following questions use nominal scale. The corresponding variables are gender, place of residence and area of work.

What is your gender?

- Male
- Female

Where do you live?

- Delhi
- Noida
- Faridabad
- Ghaziabad
- Gurgaon

What is the area of your work in the IT industry?

- Analytics
- Web development
- Applications development
- Cyber security
- Consultancy
- Sales
- Software
- Hardware
- Others

## Ordinal scale

An ordinal scale is used to measure variables that involve an order or a rank. However, these variables are not measurable by a standardized unit. Likert scale is an example of ordinal scale. It is typically a 5 point or a 7 point scale that includes a number of options. The following statements use a Likert scale. The corresponding variables are organizational factors and job satisfaction.

The rate of scale involves choices of:

- strongly disagree = 1,
- disagree = 2,
- slightly disagree = 3,
- neither agree nor disagree = 4,
- slightly agree = 5,
- agree = 6,
- strongly agree = 7.

Please put a tick mark in the corresponding box against your choice.

- disagree = 2,
- slightly disagree = 3,
- neither agree nor disagree = 4,
- slightly agree = 5,
- agree = 6,
- strongly agree = 7.

Please put a tick mark in the corresponding box against your choice.

Statement | Scale | ||||||
---|---|---|---|---|---|---|---|

1 | 2 | 3 | 4 | 5 | 6 | 7 | |

Your compensation affects your job satisfaction. | |||||||

Your relation with employers affects your job satisfaction. | |||||||

A healthy environment at workplace affects your job satisfaction. | |||||||

Workload and hours of work affect your job satisfaction. |

## Interval scale

An interval scale is used when the variable involves an order or a rank. In addition, the variable involves numerical values and is measurable by a standardized unit. The following question uses an interval scale. Here, the corresponding variable is annual income which is measurable in a currency unit.

What is your annual income in Indian Rupees?

- Less than 4,00,000
- 4,00,000 – 6,00,000
- 6,00,001 – 8,00,000
- 8,00,001 – 10,00,000
- More than 10,00,000

## Ratio scales

A ratio scale is a numerical scale. It is an interval scale with the additional property that the variable can take a value of zero. The following question uses a ratio scale. The corresponding variables are age and duration of employment.

What is your age?

- Below 25
- 25 – 35
- 36 – 45
- Above 45

For how many years have you been employed in the IT industry?

- Less than 5 years
- 5 years – 10 years
- 11 years – 15 years
- 16 years – 20 years
- More than 20 years

## Use of these scales

The table below summarizes the cases of when to use which scales.

Nominal | Ordinal | Interval | Ratio | |
---|---|---|---|---|

When the variable has no specific sequence. | Yes | – | – | – |

When the variable has a specific sequence but the difference between two values cannot be calculated. | – | Yes | – | – |

When the variable has a sequence, can be expressedin units of measurement and in intervals. | – | – | Yes | Yes |

When the variable has a sequence, can be expressed in units of measurement as well as in intervals analso can take a value of zero. | – | – | – | Yes |

## Important points to remember

- Use the nominal scale when only labeling the variable is required.
- An ordinal scale has all the properties of a nominal scale. In addition, it involves the ranking of variables.
- An interval scale has all the properties of an ordinal scale. In addition, it involves proportional interval of values of variables and the values are measurable.
- A ratio scale has all the properties of an interval scale. In addition, the variables that are expressed in ratio scales can take values of zero.

#### Reference

- SurveyMonkey. (2019). Survey vs Questionnaire: What’s the difference? Retrieved January 24, 2019, from https://www.surveymonkey.com/mp/survey-vs-questionnaire/

## Discuss