Introduction
Descriptive analysis is the second part of the 4 parts of data analysis(Exploratory, Descriptive, Predictive, and Inferential). This blog post will focus on descriptive analysis however I do have other posts about the other 3 types which are all on my page. Descriptive analysis is right after Exploratory analysis and mainly builds on the statistical analysis portion of exploratory analysis. Except for this time, we go into much more detail and really apply a multitude of statistical tests and create a lot of statistical visualizations to get an overview of our data from a statistician’s perspective. This and so much more are talked about in the article An Overview of Descriptive Analysis by Ayush Singh Rawat (the article can be found here).
Summary of Article
First, the article gives a definition of what descriptive analysis is: “the type of analysis of data that helps describe, show or summarize data points in a constructive way such that patterns might emerge that fulfill every condition of the data.” The article then gets into the techniques for descriptive data analysis. Some of them are as follows:
- Constructing tables of quartiles and means
- Statistical measures such as segregation, discrimination, and inequality.
- Crosstab or two-way tabulation
The article then gets into the types of descriptive analysis(for only 1 variable). They are as follows: measures of frequency, measures of central tendency, measures of dispersion, and measures of position. After the article details some types of descriptive analysis which are useful for multivariable data such as scatter plots and contingency tables. Finally, the article gets into the advantages of descriptive analysis which are the fact that descriptive analysis is extremely vast while still maintaining its accuracy, allows for the use of quantitative and qualitative information, is extremely objective, and is extremely useful due to its ability to identify relationships and trends.
My Take
Overall, I liked the article. This one was written in a different format than the article detailing exploratory data analysis. It had fewer visuals, and less of a process, but explained more theory about the topic. While it did definitely have its advantages I feel that the format could have been like the exploratory data analysis article and that would have made this a more enjoyable read. I felt that since the article wasn’t following an actual iteration of this type of analysis being done with the same case study throughout the whole article, I got a bit bored reading it. However, in terms of the information it was a good article since it detailed a lot of aspects of descriptive data analysis. I also really liked how it explained every technique that it presented through some small examples as it really aided in my understanding of the technique. Another thing that the article did well in my opinion was the fact that it linked to other sources for further reading, especially when the article just touched on a certain topic instead of fully explaining it there was some sort of further reading that I could pursue if I wanted to.
Conclusion
All in all, this article was a good one. While it did have some ups and downs when compared to the other article, the information presented in the article cannot be denied its usefulness and the article concretely covered a lot of the aspects of descriptive data analysis so the pros outweigh the cons. This is why you should read the article when you get a chance(the article can be found here).