What is data science?

Making data understandable 

“With data science, you can transform large amounts of data into results that are understandable to the public while cutting down on human error and providing insights in new areas,” says Katrina (she/her), a fourth-year student in Waterloo's Data Science program. 

Nathaniel Stevens (he/him), professor of data science at Waterloo, adds that “human behaviour is very messy; we’re all very different and that heterogeneity translates into noisy data, so data scientists develop methods to see the signal through that noise and generate practical insight.”

What does a data scientist do? 

With a love for math, pattern recognition, and puzzle-solving, data scientists combine numerical expertise with an understanding of human behaviour to draw conclusions and contextualize numerical findings. 

With such broad applicability, data scientists work for organizations of all kinds, from medicine and engineering to public health, retail, physics, and streaming services. 

“There are all kinds of problems that arise, which is the fun of data science. As soon as you have the technical tools, you can apply them in a variety of areas,” professor Stevens explains. 

Data is everywhere — and so are data scientists! 

Data scientists work in many fields, but some major sectors where data scientists play a big role include health care, finance, marketing, and technology.

• Health care uses predictive analytics, improving patient outcomes by analyzing vast amounts of medical data to identify trends and health risks, enabling tailored treatments and interventions for more effective care.
• Finance uses data science to assess risks and detect fraud. Financial institutions leverage machine learning to analyze transaction patterns, identify anomalies, and safeguard assets.
• Marketers use consumer data for targeted advertising and engagement. By analyzing consumer behaviour, preferences, and trends, marketers tailor strategies for maximum impact, leading to more effective campaigns and improved return on investment.
• Technology companies use big data analytics to improve product development and user experience. 

Katrina, the Waterloo Data Science student mentioned earlier, spent a co-op work term at Environment and Climate Change Canada analyzing data from mining facilities across the country.  

"I worked on all aspects of a research project, including gathering and cleaning data, analyzing the data in Python, and visualizing the results in a dashboard. I learned the challenges working with messy data and the importance of collaborating with experts to produce meaningful results.”

Why study data science? 

Data is everywhere and data drives so many decisions in our everyday lives, whether it's the algorithms that underlie the things we do on the Internet, questions of climate change and policy that comes along with that, or policy decisions around vaccines or flu shots.  

“Every time you go on the Internet, you’re consuming and generating tons of data. Data scientists turn that data into insights which drive decisions and impact practical change,” says professor Stevens. 

Besides offering promising career opportunities, data science provides the data literacy needed to think critically and be an informed citizen, says Stevens.  

"You want to understand the limitations of what we can and can't do with data, such as understanding distinctions between correlation and causation or being able to spot misinformation and disinformation." 

How does data science work? 

Data science projects typically follow a life cycle with several key stages.

1. Problem definition – A clear formulation of the question being answered or the problem being solved is developed.
2. Data collection – Relevant data is gathered from sources like databases, surveys, or experiments. For example, a retail company might analyze transaction data to understand purchasing behaviours.
3. Data cleaning – Data scientists identify and fix errors and inconsistencies to ensure accurate analyses.
4. Data analysis – Using statistical techniques and algorithms, data scientists uncover patterns and insights. This stage often involves data visualization, modeling fitting, and hypothesis testing.
5. Solution deployment – Data scientists create solutions, such as models which offer predictions or recommendations, and share their insights. 

Waterloo's Data Science program

Data science programs traditionally blend statistics and computer science. Waterloo’s strengths in mathematics have led to a more interdisciplinary Data Science program.  

Professor Stevens says that most other universities don't have a Faculty of Mathematics.  

"We have multiple math departments with so many experts in so many different areas of math, but they can all contribute to data science because it's such a varied discipline. We can offer a program that provides all these interesting perspectives that other universities just can't." 

This unique academic experience extends beyond the classroom through the world's leading co-op program. Students complete multiple four-month work terms to apply their knowledge in various roles before graduating. 

With data science driving change and innovation across industries, data scientists have the unique opportunity to combine a love for numbers, puzzles, and storytelling with their passions to shape the future of business and society. 

Want to learn more? Check out Waterloo's Data Science program page or explore other programs in the Faculty of Mathematics.

This article was created with the support of Generative AI.