How much Math do you need for Data Analysis?

Share on facebook
Share on twitter
Share on linkedin
Share on whatsapp
Data Analysis vs Maths-4

While data analysts need to be good with numbers, and a foundational knowledge of Math and Statistics can be helpful, much of data analysis is just following a set of logical steps. As such, people can succeed in this domain without much mathematical knowledge.

If you are considering a career in data but you’re worried about your mathematical skills not being good enough, our Head of Data Delivery at CodeClan, Stephanie Boyle,  is here to ease your mind. Here’s her experience with Data Analysis, CodeClan graduates and Maths:

One of the most frequent things I hear from people considering doing the data course is: “I don’t have a lot of experience with Maths, so I don’t think I can do the course.”

And I get why. Chances are if you google “What maths do I need to know to become a data scientist?” you’ll end up with a list as long as your arm, which includes things like probability, calculus, linear algebra, trigonometry, differential equations, functions, geometry, statistics… and so on. 

Fortunately, although there are things on that list you will need to be familiar with to work in data, the honest answer is that you actually don’t really need to have a full understanding of “maths” as a field to get started. The big three you do need to be familiar with tend to be calculus, linear algebra, and statistics. And in my experience – as someone who has a PhD in data science and who teaches data science as a job – once you become familiar with those concepts, you’ll have a decent understanding of and ability to do a lot of the work data analysts do. 

Theory vs Applied 

Sure, if you are creating new methods, approaches and algorithms, you’ll need to be an expert in both mathematics and statistics. But the truth is, data science is much more about understanding business questions and being able to decide which tools and algorithms you require and use them, rather than the mathematical details of those tools.

Most data scientists are the applied type: they aren’t creating algorithms or statistical tests, they’re using them. And for this all you really need is an understanding of the fundamentals of statistics (some calculus, some linear algebra, basic probability, experimental design, descriptive statistics and inferential statistics). You don’t necessarily need any more complex math than that.

Junior level vs Senior level 

It’s also important to make the distinction between how much you need to know to get a role as a junior vs. a senior. Juniors in any field don’t need the same depth of knowledge as a senior. When you get your foot in the door at a place, chances are you will be tasked with doing more of the low level data cleaning work, rather than being responsible for creating models and deploying them.

You won’t be expected to train staff up on model building, or explain the underlying maths to anyone. As a junior, your role is to learn, and work up the way to a more high level position. Whatever the role, it’s pretty likely you’re not going to need to have advanced knowledge to work on advanced projects right away.

And your advancement to a senior role isn’t going to depend on what degree you have (if any), or what your entry level maths knowledge was. It’s dependent on what you did in your junior / mid level role and nothing more. And by the time you’re moving up, you will have consolidated a lot more understanding than you would have when you started. 

The struggle is real, but that’s ok 

I wholeheartedly understand the struggle people have with maths, because I have it too. Memories of confusing maths classes, struggling with equations, and feeling stupid in class are part of my entire education. I bailed on maths early on, and didn’t even take it to standard grade as I was not good at it.

When I eventually ended up in university, I was made to do beginners maths classes alongside my degree, which didn’t really go much better. Until I got into the statistics courses. Here was math, but it was hidden inside practical applications. For the first time I was using the concepts to *do* things, rather than just for the sake of it. What a game changer. 

And I have taught enough students data and statistics now over the last 8 years to know often this is the turning point as well. The theory that seems complex, suddenly gets context. We see how values like means, medians and modes appear on plots, and in the context of real life data. Lines are fit to data, and we can actually see what changing the data does to a line. The resounding feedback we’ve had is that statistics isn’t as scary once you get taught it. 

And I guess that is the key point to remember. When it comes to maths and statistics, you can learn it. I can only speak to my experience with CodeClan grads, but we’ve had many people come through the course with no numerical background, who then go onto data analyst and sometimes junior data science roles. Hiring partners and companies come to codeclan to find people who can learn anything, not to find people who know everything. Don’t let a bit of maths put you off a very rewarding and often not very maths-heavy career. 

Now you know, it is possible to launch a career in data without being a “mathematical wizard”!
If you’re interested in learning more about our data analysis courses, take a look at our course page or join us at an info session, we’d love to see you there!

Subscribe

Receive our newsletter to hear about courses, events and more.

Search

Don’t be a stranger... Stay in touch!

Why not sign up to our newsletter? We will keep you informed about new courses and also send you information from the world of digital.

X