How To Start Learning Data Science

More than ever before, companies are relying on data to make business decisions. However, in an increasingly digital world, there has also never been more data available. With such an influx of information, where do companies even start looking for insights to make sound decisions that impact the bottom line? 

They rely on data scientists. 

You’re probably wondering, what is a data scientist, exactly? Data science, the process of extracting insights from data, is one of the fastest-growing fields in tech, with demand for data scientists ranking third on LinkedIn’s 2020 Emerging Jobs Report and placing on their report for the third year in a row. And while you’re probably most familiar with the title “Data Scientist,” similar roles are often also categorized as Data Engineers, Data Analysts, or Machine Learning Engineer.

So, what do data scientists do? The role involves a few things, but mainly math and programming.

First, data scientists use math skills, like algebra, calculus, and statistics, to build models that extract insights from a group of data. To build these models, they work in Python to clean data sets. They then leverage Machine Learning and predictive modeling to get at insights from the data set. Math knowledge allows data scientists to understand how to effectively use algorithms in their models and iterate on the modeling process.

If this sounds interesting to you, why not get started developing your own data science skillset? We’ve put together a list of resources for you that bring together helpful guides for brushing up on math skills, understanding Machine Learning, and more. Browse, explore, and study the list below to help you jumpstart your journey to becoming a data scientist.  

Still not convinced? Just remember: the average data scientist salary is $113,309, according to Glassdoor as of August 2020.

Step One: Brush up on your math skills.

You need to be comfortable with basic algebra and should know what an equation of the form y = mx + c looks like on a graph. You don’t need to be a math expert, but you will need to be familiar with calculus and linear algebra and have a solid understanding of statistics. These three areas are the kind of mathematics that scientists use to analyze data.

• Check out free, online statistics and probability courses with Khan Academy. Work through units that build upon each other and complete quizzes to obtain mastery points.

• Explore different statistics courses on Coursera. They even have specific materials for getting started in data science, like a Mathematics for Data Science Specialization course.

• Read Naked Statistics by Charles Wheelan, a “lifesaver for those who slept through Stats 101.”

• Discover the mathematical foundations of data science with – you guessed it – more statistics. The Elements of Statistical Learning covers data mining, inference, and prediction.

• Refresh on linear algebra, the backbone of matrices and how companies like Netflix, Spotify and more deliver recommendations to their customers. This Khan Academy course offers a great starting point.

• Don’t forget calculus. EdX runs a Pre-University Calculus course, plus offers a verified certificate for a $50 fee.

• Download an overview of all the different types of math data scientists need to know in this blog post from Elite Data Science.

Step Two: Introduce yourself to coding.

Before any formal data science training, you don’t have to be an experienced programmer, but you have to be comfortable with breaking down bigger problems into smaller ones that a computer can understand. Some experience with coding will help you to get started, but ultimately, to become a data scientist, you’ll need to know Python and SQL.

• Navigate the basics of coding. Knowing variables, conditionals and loops will help. This article on Open Book Project provides some guidance.

• Check out free Python for Dummies resources and explore various articles related to the history and uses of Python.

• Learn some SQL. Start with w3schools Introduction to SQL or Udacity’s Learn SQL course.

• Explore Codecademy’s Data Science resources, career paths, and skill paths. Bonus: they also have (more!) stats courses.

• Get familiar with other frameworks and languages you may encounter. Free Code Camp’s article explaining the importance of JavaScript in Data Science is a good place to begin.

Step Three: Explore data science courses and resources to combine your math and programming skills.

By the time you feel comfortable with math concepts and understand the foundations of coding, it’s time to put it all together. Whether you follow the path of formal training with a bootcamp or paid course, or choose to keep up with data science as one of your personal passions, the below resources will teach you how to become a great data scientist.

• Familiarize yourself with the key concepts in Data Science. 

• Subscribe to FiveThirtyEight. Known for their data-driven sports analyses, this data-focused news outlet covers topical news from the lens of a data scientist.

• Get comfortable sharing insights visually. Check out a few examples in this Harvard Business Review article, Visualizations That Really Work. 

• Learn via online lessons in Flatiron School’s free intro-level data science workshop. The course includes hours of preparatory material to get yourself ready for a bootcamp, including Flatiron School’s career change Data Science bootcamp-style course. 

• There are other Data Science bootcamps out there like Galvanize, General Assembly, Metis, NYC Data Science Academy, and Thinkful.

• Take a look at an even more exhaustive list of data science bootcamps on Course Report, SwitchUp, and Career Karma.

• Join social media communities to meet data science professionals and fellow learners. Here are some popular Facebook groups.

• Or, follow data science thought leaders on Twitter like Andrew Ng (former head of Google Brain) or Hilary Mason (GM for Machine Learning at Cloudera).

• Read Andriy Burkov’s The Hundred-Page Machine Learning Book.

Disclaimer: The information in this blog is current as of September 23, 2020. Current policies, offerings, procedures, and programs may differ.