It is important to understand different types of data because it is one of the most important topics od statistics. Knowing about the different types of data will help you to apply statistical measurements on your data and hence, rightly conclude specific assumptions about it. Datatypes are an important topic to learn.
The reason behind this is that statistical methods can only be used with certain types of data. You have to analyze each data type different to than another data type, otherwise, you would arrive at wrong conclusions. Therefore, having knowledge about different types of data enables you to choose an accurate technique of analysis.
Having an understanding of different data types is a key condition for performing Exploratory Data Analysis (EDA) because you can use only a specific statistical measurement for certain data types. In addition to this, one should be aware of the type of data to select the right visualization method.
Let’s learn about different types of data in this article ahead.
Table of Contents
1) Nominal Data :
This type of data is used to represent discrete units and are used to tag variables. The variables tagged using nominal data have no quantitative value. Nominal data don’t follow any particular order. Consequently, the change in the order of its values does not alter the meaning. The data that represents more than two categories are called nominal data.
2) Categorical Data :
This type of data is used to represent characteristics. Therefore, using this type of data one can represent things like language, citizenship, gender etc. in addition to this, categorical data can also receive numerical values as input. For example, 1 for physical disability and 0 for physically fit. But keep one thing in mind that these numerical values don’t have any mathematical meaning.
3) Ordinal Data :
This type of data is a combination of Nominal and Categorical. Hence, it represents values which are both ordered and discrete in nature. One limitation of ordinal data is that it does not tell about the difference between the values. Because of this reason, ordinal data is used to determine non-numeric entities such as customer’s satisfaction, happiness, and joy etc.
4) Dichotomous Data :
This type of data uses binary values (1 and 0) to represent a value. most studies use dichotomous data for meta-analysis. Because of its limited values, it can only be used to represent information with only two values. You can’t represent data where you have more than two units. For those values, nominal data can be used. the example of dichotomous data are questions like what is your gender? For this type of question two categories are given such as Male/Female.
5) Continuous Data :
This type of data is used to represent measurements hence, values of this type of data cannot be counted but can only be measured. For example, the height of a person can only be explained using intervals on a real number line. The continuous data can be of two types.
a) Interval data :
this type of data represents ordered units which have the same difference between them. Hence, interval data is used when we want to measure a variable that has numerical values which are ordered and the exact difference between the values is known. An appropriate example of this type of data is the temperature of a place.
But there is one problem with this type of data. The problem is that it does not take “true zero” value. which mean in our example no temperature does not exist. Interval data can be added and subtracted but cannot be multiplied or divided. In addition to this, you can also not calculate ratios of interval data and because of the lack of true zero value many inferential and descriptive statistics can’t be applied.
b) Ratio Data :
This type of data is used to represent ordered units and these units have the same difference. Ratio values are similar to interval values with an exception that there is a true zero in ratio data. Therefore, this type of data is suitable to represent weight, height, and length etc.
6) Discrete data :
This type of data is used to represent data whose values are distinct and separate. But there is only one limitation that is that discrete data can only represent a certain type of values. Discrete data can only be measured and can’t be calculated or counted. The information that can be categorized into a classification can only be represented using discrete data.
For example, to calculate chances of heads to appear when a coin is flipped 100 times. one can identify that one is dealing with discrete data or not by answering a simple question that if the outcomes of the experiment can be counted or not. If the answer is no that means you are dealing with discrete data.
