Do You Need a Degree to Be a Data Scientist?
I often get asked, "Do I need a degree to break into Data Science?"
Short answer: no, but it certainly helps.
Let me explain exactly what I mean by this in this article. Hopefully, it will shed some light on this question or give you some examples backed by data and studies.
As a disclaimer, everything in this article is my opinion, and it is not professional advice.
Background of Data Scientists
The question, "Is university or college worth it anymore?" has been answered and peddled to death.
The same response is always given: "It is not worth it unless you need it for the job you want to do, like a doctor or lawyer."
I agree with that statement, as it is technically correct, but it is way more nuanced than that.
According to this article from Coursera, based on a study from Zippia, 51% of data scientists have bachelor's degrees, 34% have master's and 13% have a PhD. That leaves 2% with an undefined or other degree category.
Another study by Bob Moore, who analysed data scientists' LinkedIn profiles, found that "Over 79% of data scientists that list their Education have earned a graduate degree, and 38% have earned a PhD."
And another study by 365DataScience said that "According to our data, the typical data scientist in 2020 holds either a Master's degree (56%), a Bachelor (13%), or a Ph.D. (27%) as their highest academic qualification."
We must take all summary statistics with a pinch of salt as we can't fully know the data collection process or how the numbers were generated. The job "data scientist" is still quite ambiguous and varies quite a bit between companies, so it is important to keep this in the back of your mind.
Even though the exact statistics disagree somewhat, it's clear that most data scientists hold a degree, and many have advanced degrees like master's or PhD's.
From personal experience, this makes sense. I have never met a data scientist without a degree, and the majority of those I work with always at least have a master's and several I know have PhD's (they are all really clever!).
What's more interesting is that the Zippia article goes further and explains the degree subjects that are most popular among degree holders.
We see that ~50% of data scientists with degrees are in computer science, statistics, math, and physics. Other majors are not defined, but they will likely include subjects like engineering, chemistry, and economics. Basically subjects with significant maths and coding components.
Therefore, the majority of data scientist with degrees come from a STEM background, which are arguably some of the most difficult subjects to study.
Again, this aligns with my personal experiences where pretty much every data scientist I know has a degree in either maths, physics, or computer science. There are of course others, as shown by the above plot, but these three subject clearly dominate.
So, what does this all mean?
Well, most current data scientists have degrees in very technical disciplines and often have further degrees than just a bachelors.
However, this doesn't fully answer our question of whether you need a degree to be a data scientist or not. Let's dive into the typical job requirements to dig a bit deeper.
Job Requirements
According to indeed (#1 job site in the world):
"Data scientists are generally expected to have a minimum of a bachelor's degree in data science, computer science or another closely related field. Many employers prefer a master's degree to further hone skills in computer programming languages, probability and statistics."
They also share data on data science job requirements within the last three years in the United States.
We see that 95% of employers want a bachelor's degree or higher for data scientist positions and only 3% of employers require a high school diploma (equivalent to A levels in the UK).
If I look at my LinkedIn jobs for data scientist positions, most appear to be after STEM master's degrees, which aligns with the data we saw from Indeed and Zippia above.
The point is that it's clear that employers want a technical degree for their data science jobs, and many also specifically want a master's or PhD, not just a bachelor's degree.
I would note that the PhD requirement is often for more research-based roles and normally a master's is more than sufficient.
Personal Take
We have just seen that the majority of data scientists have degrees, and the majority of job descriptions require candidates to have degrees in subjects like maths, statistics, computer science, and physics.
So, from a pure data and statistical perspective, you need a degree to become a data scientist, and it should be in a STEM subject if you want to boost your chances.
Are we done? Well, no.
Like I said in the beginning, the answer is more nuanced than that.
Employers are not necessarily after the degree itself but rather the skills those degrees give and the abilities those candidates posses.
I have written several articles on the maths, statistics, and coding knowledge you need to be a data scientist. You need a large array of skills, some of which are only really taught in the first and second years of many STEM subjects.
High school maths classes hardly teach stuff like eigenvalues, tensors, and complex Bayesian statistics. However, these topics come up a lot within data science and Machine Learning, and as I mentioned in previous articles, you need to be good at maths to be a top data scientist, that's just the honest truth.
These STEM degrees give you a demonstrable example of your proficiency in these areas through the degree and the grade given to you. They also show that you can study and learn complicated topics and apply them, which is the back bone of a data science job.
The advanced master's and PhD degrees show independent research and the ability to dive deep into a single problem and really explore it. These are fundamental skills of an upper-echelon data scientist and why companies are willing to pay them more than candidates with bachelor degrees.
According to the Zippia study, a data scientist with a bachelor's degree earns an average of $101k, but a data scientist with a master's earns about $110k. Sure, the difference isn't massive, but it does show that an advanced degree will increase your earning potential.
The degree is a stamp of approval of your abilities in maths, coding and problem solving, and at the moment, it's the best and most used system we have for getting these stamps on our resume or CV.
Sure, you can technically learn everything online, but imagine trying to self-motivate yourself to learn tensor calculus in your spare time. I am afraid it's not quite the same motivation as failing an exam for a degree for which you are paying thousands in tuition fees.
Now, don't get me wrong, landing a data science job is much more complex than simply having a degree and like we have seen with software engineering over the last decade, a degree may become less of a requirement over time as data science becomes a more standardised role.
Projects, demonstrable interest and excellent soft skills also go a long way in making you a well-rounded data scientist. Like anything nowadays, a degree may teach you some things, but it's unlikely to fully prepare you for the whole job. You need to up-skill in other areas if you want to stand out.
Another thing to remember is that job requirements are more of a list of things for a "perfect" candidate, which never truly exists in practise. So, if you have all the skills and no degree, you should still apply as there is a strong chance you may still get the job. There are a few people I know who have degrees in subjects like history and English and are now data scientists and really good ones at that.
Overall, as it stands today, it seems that to increase your chances of landing a data science role, it's preferable to have a degree in a STEM subject. However, this is not to discourage you as you can learn all the required skills without a degree, which I would argue is even more impressive and really shows your passion for the subject!
If you want a roadmap to learn the required maths and coding to become a data scientist, check out my previous posts.
What Statistics To Learn For Data Science
Another Thing!
I have a free newsletter, Dishing the Data, where I share weekly tips for becoming a better Data Scientist, my general experience in the field, and some thoughts I have had over the past week.
Connect With Me!
- LinkedIn, X (Twitter), or Instagram.
- My YouTube Channel to learn technical data science and machine learning concepts!
