There are different types of variables which we encounter when dealing with business analytics. The type of data we are going to work with are divided into two main ways.
1. Type of data
2. Measurement levels
1. Type of data
It is further divided into numerical and categorical data.
Categorical data
It defines categories or groups. Car brands like Tata, Honda, Toyota etc.. are considered to be different categories. Another example is the Yes/No questions. When someone asks us whether we are above 18 years of age or not, we have only two answer categories (Yes and No).
Numerical data
As the name suggests, this data represents numbers. It is further divided into two types , discrete and continuous.
- Discrete numerical data :
Discrete data can be counted in a finite manner. For example, you can count the change in you pocket, you can also count the money in your bank account. Even if we want to count the money in everyone's bank account, it is feasible and only thing is it would take a considerable amount of time. It is actually the opposite of continuous data.
- Continuous numerical data :
Continuous numeric data is infinite and impossible to count. It is never possible to count them in a particular time interval. For example age, we cannot count age. Why? Because it would take more time. You would be 11 years, 11 months, 11 days, 11 hours, 11 minutes, 11 seconds etc.... __, and the __time factor is continuous and it keeps on changing. So at any fixed point of time, we cannot calculate our age.
How to calculate age then ?
The above age value is a continuous value and it keeps on changing, the only way to calculate the age of a person is to convert the age value to a discrete variable and then count it. We remove the time factor (continuous value) from the age.
Time on a clock is discrete, but in general it is continuous. Now it becomes a discrete variable and we can calcul ate the age. The age value will be 11 years or 300 months.
Other examples of continuous data include the height, weight, age , time, area etc..
2. Measurement levels
Levels of measurement are split into two groups, qualitative and quantitative data.
QUALITATIVE DATA
Qualitative data can be categorised based on traits and characteristics. For example: cars can be categorised based on their brand name and a food can be categorised based on their reviews. Qualitative data can be nominal or ordinal.
- Nominal data
Nominal variable are like the car categories, they aren't numbers and cannot be ordered . Here the options can be chosen individually but they don't have any order here. So the measurement level here is qualitative and nominal.
- Ordinal data
Ordinal data consists of groups and categories that follow a strict order. For example when we are asked to rate the food in a restaurant, the options to rate are stale, good ,very tasty, must try etc... Although these options does not represent numbers, these options are ordered from negative to positive values in terms of rating. Thus the measurement level here is qualitative and ordinal.
QUANTITATIVE DATA
They are data represented by numbers. Usually there are also measurement units associated with data , eg. 12 metres, 10 kg etc.. It is data expressing a certain quantity, range or amount. They are further split into interval and ratio.
- Interval data
Interval allows you to measure all quantitative attributes. Any measurement of interval can be ranked, added, subtracted etc.. and equal intervals separate each quantitative value. It will also have its lower bound less than zero. For example: a temperature value is a quantitative data and it can have both negative and positive values, so this is said to be interval data.
- Ratio
It also has the same properties as interval data. It can also be used to add, subtract, count etc.. Ratio values never fall below zero. Height and Weight never fall below zero, and they always measure from zero.
Interval ratio scale measurements
Understanding the different scales of measurements allows you to identify the different types of data you can gather. These differences will help us to identify the kind of statistical research required for your research.
Interval level of measurement
It specifies the distance between each interval of a scale is equivalent, from low interval to high interval. For example , the difference between 10 degree C and 20 degree C is the same as the difference between 20 degree C and 30 degree C.
Ratio level of measurement
Ratio levels can have a value of zero. It states that the cost difference between two cars that are 100000 dollars and 150000 dollars is the same as the cost difference between two cars that are priced 200000 dollars and 250000 dollars respectively. However, you won't be able to find cars priced less than zero.
Top comments (0)