What is Data?
When I applied for my first data analyst position, I was asked a very simple question. A question which is so obvious that we often choose to skim it through, or we doesn’t keep it in the immediate memory. The question was, “What are the different types of data?”
My immediate response to that question was to ramble about different data types that are there in Python, like string, float and integer! Even when the interviewer gave me a second chance, I ended up telling that there is numerical data &text data . I wasn’t selected for that position, although I kept thinking to myself that interviewer was asking useless questions as an excuse for not hiring me. I couldn’t have been more wrong!
“Data is any fact or quantity which is relevant for making the analysis and/or calculation for task at hand.”
Type of Data :
Qualitative : These data provides us characteristics and traits for our object of observation. It is descriptive through textual data. For example, essay written by all of the students, user comments on social media.
Quantitative : Data that can be counted, measured and provides us the power to perform multiple mathematical operations & computations on them. For example, height & weight of students.
Interval Data: It provides both the order and exact difference between the values which is meaningful. Example : Celsius temperature, pH scale. The only problem with interval scale is that they don’t have a true zero.
Ratio Data: It has all the feature like ordinality, meaningful difference, zero value. And so it has a lot of application in inferential and descriptive statistics. Example : height, weight.
Ordinal Data: Provides order to the values and that is the significant part. We can not know the difference between variables. Example : Survey Rating from 1 to 5.
Nominal Data: It is used for labelling variables without any quantitative value. Example : Gender(Male, Female), Direction (East, West, North, South).
“It is imperative that we pay due attention to the type of data that we are using. Misunderstanding can lead to the implementation of wrong statistical analysis to improper data type.”
