From the course: Data Science Foundations: Data Assessment for Predictive Modeling

Unlock the full course today

Join today to access over 22,500 courses taught by industry experts or purchase this course individually.

How to be effective doing univariate analysis and data visualization

How to be effective doing univariate analysis and data visualization

From the course: Data Science Foundations: Data Assessment for Predictive Modeling

Start my 1-month free trial

How to be effective doing univariate analysis and data visualization

“

- There are rules for how to perform Univariate analysis and data Visualization. Univariate simply means looking at one variable at a time. These rules date all the way back to Stevens in his article, "Introducing Level of Measurement." Not only are they what others are going to expect, an increasing number of automated software packages will automatically generate output based on these rules. So you really should be familiar with them. Let's start with the highest level, Scale. And you really can describe it as the highest level because it has the most information content, and therefore you can run the most tests on it. The rules say to run just the Mean, but I always run both Mean and Median. Because if the Mean is much higher than the Median it indicates a skew to the data. So what about min, max and standard deviation? Well, with Scale variables we're looking at a continuous distribution of data and we want to see the…

Contents