Everyday Applications of Data Science

Featuring Insights from Professor Dustin Tingley

Published March 21, 2024

Series Mentioned in this Post: Harvard on Digital 

Courses Mentioned in this Post: Data Science Principles, Data Science for Business

Ever wondered why airline and hotel prices fluctuate depending on when and where you book? Curious about how media outlets predict the outcomes of major events like the World Series or the Super Bowl? Puzzled by the accuracy of political poll predictions during election seasons?

Data science quietly but significantly shapes many facets of our daily lives, and these questions are just the tip of the iceberg when it comes to the many applications of data science in our everyday experiences and decision-making processes.

Image

Harvard Online Professor Dustin Tingley helped us understand the impact data science has on our everyday experiences in a few key areas: 

Data Science and Sports

Fantasy sports and bracket challenges have become popular ways for friends and colleagues to engage in friendly competition – but they’re also a great example of data science at work in our everyday lives.

Everyone has their own strategy for making big game and tournament picks. From the logical (analyzing past point values) to the illogical (picking only animal mascots), it’s not always obvious which strategy will win. Dustin Tingley, Harvard Professor and Data Science Principles instructor, helps provide clarity on how to understand this everyday sports fan challenge:

“The thing about March Madness and other tournament bracket challenges is that mistakes that happen earlier in the tournament can have huge impacts later on. In some sense, that is the fun of it. It is very different from betting on a single game. An unexpected upset incident that happens early on does not have an isolated impact in these situations. The challenge of picking the ‘right’ teams gets even harder when similarly ranked teams are matched up, making the outcome closer to a random outcome.”

In addition to helping fans pick the winning team, data science helps sports teams analyze player statistics and game data to inform strategy and player recruitment, while sports media outlets use predictive models to forecast match outcomes. However, Tingley emphasizes that data analysis often has to anticipate the strategic nature of sports. If the data suggests three-point shots (longer distance shots) are more valuable in basketball, and a team starts completing more three pointers, then the opponent can try to redesign their defense. Critical thinking in data science is crucial. 

Data Science and Travel

Finding the best travel deals and prices has become a fun pastime for avid travelers. From booking with credit card points to finding discounts on travel deal sites like Kayak or Expedia, there are many options for planning the best trip for a great value. If you ask Professor Tingley, he’ll say, “I think I have spent way too much of my life trying to find the best airline or hotel price and could have used that time doing something else.” But with two out of every three Americans budgeting to travel in 2024, many will turn to airline, hotel, and travel sites to book their trips and wonder why the prices vary so widely.

“I understand why people get frustrated with dynamic pricing, but these sites and companies are responding to changes in demand given that they have a pretty fixed supply of the product they offer,” says Professor Tingley.

He explains further that, “There are only so many flights that can run out of an airport, and hotels have only so many rooms. They also know peak travel times pretty handily, because vacation travel is very predictable. They definitely know when my kids are on break from their public school system. They also get to observe what people are searching for on their websites, so they can build up data sets about that and then try and predict how many of those searches will turn into purchases, which allows them to set the highest price they can to get you to purchase their service.”

While dynamic pricing can definitely feel frustrating, it's simply a reflection of supply and demand within the market, ultimately shaping the way we plan trips. Whether it is too frustrating or unethical is considered more a question for voters and politicians. 

Data Science and Elections

Data science tools and practices have become critical components of modern political campaigns and elections. Campaign strategists use data science insights to target and mobilize voters, tailor messaging, and optimize campaign strategies. Voters and engaged citizens often turn to sites like 538 to stay informed on campaign trends and potential results. But how accurate are these predictions?

Professor Tingley tells us that, “Many modern predictive analytics operations are not just using one source of data to make a prediction but instead combine many separate pieces of data together. They then try to weigh each of these data sources to predict an outcome. For example, when attempting to predict an election winner, they will input the information gathered by separate polls that are completed by different polling organizations. Each of those polls and polling organizations separately might not quite reflect what actual voters do (even though they will often say they do), but if you combine them in the right way, you can often come up with a relatively reliable prediction.”

When considering which outlet to trust when it comes to political and election data and insight, Professor Tingley suggests, “Focusing on a couple of things, like how well do they report their level of uncertainty about their own predictions, how transparent they are about how they make their predictions, and whether the data being used is high quality and suited for the prediction rather than just the data that might have been easily available.”

Data Science and Your Work 

“In today’s business environment, data is a necessity,” says Professor Yael Grushka Cockayne, instructor of Data Science for Business.

In every industry, data science is transforming the way we work, from improving productivity and efficiency to enabling data-driven decision making. Whether you're a marketer analyzing customer behavior, a supply chain manager optimizing inventory levels, or an HR professional predicting employee turnover, data science skills are increasingly valuable in today's job market.

If you’re interested in learning more about data science from Professor Dustin Tingley or Professor Yael Gruska-Cokayne and how to apply it in your own everyday life, join the next cohort of Data Science Principles or Data Science for Business.