# Separating Great Data Scientists From OK Data Scientists: Statistics

**Separating Great Data Scientists From OK Data Scientists: Statistics**

January 26, 2018 by Editorial Team Leave a Comment

Data science and analytics are incredibly popular fields right now, and for good reason. As more organizations and brands launch targeted forms of customer and data collection, they’ll need to put teams in place to extract actionable intel from their databases. Even with automation and AI tools to do the filtering, highlighting and categorizing, data scientists must still make sense of the information.

There’s plenty of work available in the industry, and that’s not likely to change anytime soon. But at some point, you’ll need to up your game to be competitive. That means boosting or improving your training and coming up with new skills, knowledge and experience to do your job. What sets the truly great data scientists apart from the rest? How can you ensure you’ll remain competitive on many levels?

The answer is by building — or already having — a strong background in statistics. Wait, what? Why is statistical knowledge and experience so influential in data science?

**Crushing Misconceptions**

Before diving in, it’s important we talk about current misconceptions and mumblings about the industry of data science. Many so-called experts claim you don’t need much probability and statistical knowledge if you plan to work in the field. While you can certainly get by — mostly — without a decent grasp of these concepts, it will be much more difficult for you. In fact, you could argueyou’re missing a crucial component of data science as a whole if you lack experience with statistics. To correctly enter the field, if you haven’t already, start with a strong foundation in statistics and move on from there.

**Why Does Statistics Matter?**

A handful of statistical methods and concepts prove influential to working with modern data. Some of the more common methods include linear regression, classification, shrinkage, dimension reduction and more. But there’s also an important distinction to make: Data science and statistics are two separate things. Statistics is a critical element of data science, but related departments now have the risk of becoming irrelevant, thanks to the widespread adoption of big data and analytics. That’s technically because data science is a bit more involved, and as a result more useful to many organizations and teams.

In data science, you collect and organize troves of data, which you then sort, filter and analyze by creating models and visualizations. Data scientists, thanks to their background in statistics, can look at a set of information and come up with important trends and patterns. Data scientists can take their collected insights and communicate them, in full, to decision-makers to help them choose a proper course of action or carry out their work. It’s incredibly rare to see this process streamlined and moving in one direction, however. Even skilled data scientists must reboot and revisit certain collections to come up with new methods for extracting or identifying new sets of data. It’s a constant series of updates, revisions and improvements — so much so, the process of data science is continuous, almost endless in nature.

The more important aspect of data science, however, is that the relevant insights and content must be translated into a more viewable form. Industry outsiders, for instance, are not going to see or understand the same things a skilled scientist will. As you delve deeper, the concept of data science becomes much more complex and involved. Though statistics are just a fraction of what goes into making a proper data scientist, it’s crucial for data scientists to have a working understanding of statistical models.

**Where Do I Start?**

Having read this far, you likely have questions about where to start with your data science training. Assuming you’re not already heavily involved in the field — though training is still a good idea, if you are — it would be wise to begin with courses on statistics and probability. You can find a list of recommended courses and online training curricula here. Any of the courses mentioned are suitable for a data science beginner, and will help you grasp the knowledge and information needed to enter the field. One thing to note is that you’ll want to gain an understanding of both statistics and probability to get the total picture, as there’s a difference between them.

About the Author

*Contributed by:** **Kayla Matthews, a technology writer and blogger covering big data topics for websites like Productivity Bytes, CloudTweaks, SandHill and VMblog.*