Analyzing Survey Data2

Numerical data: Types and Characteristics

Improve decision-making by gathering customer data.


Table of Contents

Numerical data helps you test customer satisfaction & experience with your brand, products, services, and other aspect of the business in numbers which are easy to analyze. Let’s dive in to understand how numerical data is useful in research to accelerate business decision. 

What is Numerical Data?

Numerical data refers to the data that is in the form of numbers, and not in any language or descriptive form. Often referred to as quantitative data, numerical data is collected in number form and stands different from any form of number data types due to its ability to be statistically and arithmetically calculated. 

Example: You have a total count of your employees. You take a count of male employees and subtract that from the total number of employees to get the count of female employees. This characteristic of numerical data to be manipulated arithmetically makes it a best suit for statistical data analysis.

Explore all the survey question types possible on Voxco

Types of Numerical Data

The two forms of numerical data you will get to see are discrete data and continuous data. Both of these variants are explicitly used in statistical and research purposes and are proven to give the best data through research methods. 

Let’s take a deeper look at how they are different from each other:

  • Discrete data 

Discrete data is used to represent countable items. It can take both numerical and categorical forms and groups them into a list. This list can be finite or infinite too. 

Discrete data basically takes countable numbers like 1, 2, 3, 4, 5, and so on. In case of infinity, these numbers will keep going on. 

Example: counting sugar cubes from a jar is finite countable. But counting sugar cubes from all over the world is infinite countable.

  • Continuous data

As the name says, this form has data in the form of intervals. Or simply said, ranges. Continuous numerical data represents measurements and their intervals fall on a number line. Hence, it doesn’t involve taking counts of the items. 

Example: in a school exam, students who scored 80%-100% come under distinction, 60%-80% have first class and below 60% are second class. 

Continuous data is further divided into two categories: Interval and Ratio.

  • Interval data – interval data type refers to data that can be measured only along a scale at equal distance from each other. The numerical values in this data type can only undergo add and subtract operations. Example: body temperature can be measured in degree Celsius and degree Fahrenheit and neither of them can be 0.
  • Ratio data – unlike interval data, ratio data has zero point. Being similar to interval data, zero point is the only difference they have. Example: in the body temperature, the zero point temperature can be measured in Kelvin. 

Numerical Data Variables

A numerical variable is something that inhibits any value that is finite or infinite. Like length, age, weight, exam scores, etc. numerical variables can be called a continuous variable when it has continuous data characteristics. 

  • Interval variables 

It has values with interpretable differences but never zero. These values can be added or subtracted but can never be multiplied or divided. Interval variables have standard difference between them and are an extension of ordinal variables. 

Interval variables have two distributions: Normal distribution and non-normal distribution.

    • Normal distribution 

A random variable having a real value is said to be normally distributed when its distribution is unknown. Two different tests are carried on two different sample such as;

      • Matched sample tests 

        • Paired t-test: to compare two sample population means.
        • Repeated measures ANOVA: to compare means of 3 or more variables. It is based on repeated observations. 
      • Unmatched sample tests

        • Un-paired test: to compare two sample population means.
        • ANOVA: to compare means of 3 or more variables based on single observation. 
  • Non-normal distribution 

A random variable having a real value is said to be normally distributed when its distribution is known. Two different tests are carried on two different sample such as;

    • Matched sample tests
      • Wilcoxon rank-sum test: to compare two groups of matched samples.
      • Friedman 2-way ANOVA: to compare the difference in means across 3 or more groups.
    • Unmatched sample tests
      • Wilcoxon rank-sum test: when the requirements for the t-test of two unmatched samples are not satisfied.
      • Kruskal-Wallis test: to see whether three or more groups of unmatched samples start from the same distribution.


  • Ratio variable

It is an extension of the interval variable but the difference is that it has a true zero value. These variables can undergo any operation like addition, subtraction, multiplication as well as division. 

Talking about the tests that are performed on ratio variables, refer to those of interval variables as they are the same. 

Analysis of Numerical Data

There are two ways to interpret the data collected in numerical form. Depending on your data and the way you results, you can make use of the following methods:

  1. Descriptive statistics – this makes use of the datasets to describe a sample of population. These datasets are collected from the population itself. The methods used in descriptive statistics are: mean, median, mode, standard deviation, variance, etc.
  2. Inferential statistics – this method involves making inferences or predictions with regards to a population depending on the data collected from the sample of that population. These are some ways of performing inferential statistics:
  • Trend analysis: to draw trends and insights by collecting the survey data over a certain period. 
  • SWOT analysis: stands for Strengths, Weaknesses, Opportunities, and Threats. Strengths and Weaknesses perform internal analysis, whereas Opportunities and Threats perform external analysis of an organization.
  • Conjoint analysis: it determines how people make their choices. This is a market research analysis technique.
  • TURF analysis: stands for Total Unduplicated Reach and Frequency analysis. It is used to assess the market potential for a combination of products.

See Voxco survey software in action with a Free demo.

Characteristics of Numerical Data

  • Numerical data has two categories: discrete data and continuous data, where the latter is further classified into interval data and ratio data.
  • Numerical data is quantitative in nature as it takes quantitative values for data. 
  • Numerical data allows us to perform arithmetic operations on them like add and subtract. It can also use any statistical analysis calculations. 
  • It can be estimated and enumerated. When the numerical data is precise, it is enumerated, or else it is estimated. 
  • The interval difference between each numerical data when put on a number scale, comes out to be equal. A clock, a thermometer are perfect examples for this. 
  • Numerical data can be analysed using two methods: descriptive and inferential analysis. 
  • Numerical data makes it easy to be visualized. It uses data visualisation techniques like scatter plot, dot plot, stacked dot plot, histograms.

5 Examples of Numerical Data

  • Age – age of an individual is counted under numerical data since it can take countable numeric values. Example: a child of age 10 years started walking 7 years ago. 
  • Time – time is a numerical data and is countable, finite. Example: the time in which a runner runs 10 laps of a ground. 
  • Height – the height of a person can take any value that is countable and it keeps on growing with time. Example: a person’s height can be measured in meters, inches, feet or centimetres. 
  • Income – a person’s income or a family’s income is a numerical data. In the market, this is used to determine the purchase capability of the customers. 
  • Test scores – students’ test scores are marked in numbers and then are further ranked based on their score. Example: students with scores from 80-100 are considered a distinction class. 60-80 are first class and students below 60 are second class. 
Online Surveys

Advantages of Numerical Data

  • Population prediction – researchers can use numerical data to collect the birth of newborns in a certain period of time and then use that data to predict the population of the country.
  • Marketing and Advertising – before starting tier marketing and advertising strategies, researchers make use of SWOT analysis to determine external and internal variables that can affect those strategies. 
  • Research – numerical data is a common practice among the researchers due to its ease of statistical calculation. 
  • Product development – researchers use TURF analysis in product development stages to determine the scope of the new product in the market. 
  • Education – as given various examples above, interval data is used in schools and colleges to grade students’ performance in exams. 
  • Medicine – doctors make use of the thermometer on a regular basis to study a patient’s body temperature. This also comes under interval data.

Disadvantages of Numerical Data

  • Numeric data from people does not define their feelings towards certain topics. 
  • Results are short and limited.
  • Generic questions from researchers can lead to structural biases.


This sums up everything you need to know about numerical data before you start collecting them. Whether you use surveys or experiments to collect numerical data, it can help you get validate the hypothesis. 

If you are ready to start collect numerical data but don’t know which method to use or what questions to ask you can contact our experts

Read more