Data science is the study of how to extract and apply insights from information. Under many circumstances, it entails working with large datasets, such as usage statistics for a mobile or desktop program, a database of addresses or a collection of financial trading trends. The real-world uses of data science range from fintech applications — e.g., building more consumer-oriented banking services backed by superior fraud detection — to the further development of technologies like artificial intelligence (AI) and machine learning.
A day in the life: A look at the everyday work of data scientists
Data science should not be mistaken for a field preoccupied with the pursuit of far-off and theoretical breakthroughs. Actual data scientists perform more practical tasks during their day-to-day work, including the following:
- Collecting and cleaning data, potentially across multiple disparate IT systems, so it’s usable in projects
- Creating visualizations that condense complex information into an accessible graph or chart
- Contributing to dashboards that contain those visualizations and help an organization’s stakeholders and end users make informed decisions
- Collaborating with these same groups to explain the features of data science projects and what conclusions to draw from their results
In fact, 80 percent of the typical data scientist’s time goes toward simply finding, cleaning and organizing key information, with only 20 percent devoted to analysis, according to an IBM assessment. That might sound surprising, but it’s a good reflection of what data science work really entails, namely securing the best information available for a dashboard, visualization or other project.
After all, any model a data scientist creates is only as good as the information underpinning it, so it pays to take the time to gather as many assets as needed and ensure they cohere and minimize common biases, such as:
- Sampling bias: Creating unrepresentative samples by favoring — often without intending to — the selection of certain criteria.
- Anchor bias: Settling on the datasets gathered at the outset of a project, even if there is subsequent information that should be collected and which might yield different results.
- Confirmation bias: Exclusively seeking out items that confirm an existing opinion, a practice that almost inevitably excludes data that would make a model more accurate.
When performed with the proper due diligence, rigorous statistical methodologies and the right supporting technical tools, data science is invaluable for organizations in fields as varied as technology, healthcare, finance and insurance. For example, the work of today’s data scientists is integral to making connected home speakers like Amazon’s Alexa or Google Home “smarter” and for continuously improving the directions provided by mapping services. At the same time, it’s important for quantitative research when creating insurance policies, conducting risk management and ensuring proper identity verification.
Trend lines and insights: The outlook for data science careers
Harvard Business Review once called data scientist “the sexiest job of the 21st century,” an occupation in short supply and high demand. In 2017, IBM released a study supporting that conclusion with the key finding that the total number of data science jobs was expected to increase 28 percent between 2017 and 2020, adding 364,000 positions for a 2020 total of more than 2.7 million in the U.S. alone.
The U.S. Bureau of Labor Statistics (BLS) has come to similar conclusions about the short- and medium-term trajectory for data science careers. The BLS does not break out data scientists into an exclusive category, instead grouping them with computer and information research scientists. For the period from 2016 to 2026, the BLS expects 19 percent growth in employment for these professionals, much faster than the average (below 10 percent) for all occupations. Median pay was estimated at more than $114,000 in 2017.
What’s fueling these high expectations for data science? One big reason is the multifaceted expertise data scientists possess, spanning not only machine learning and technological savvy, but also mathematical knowledge and analytical thinking. Accordingly, they can work on projects such as business intelligence initiatives that disseminate data to the right stakeholders, along with more technical undertaking such as improving a recommendation algorithm on an e-commerce platform.
Becoming a data scientist: How the UCR MSE puts you on the right track
At the University of California, Riverside (UCR), you can take advantage of a 100% online data science program to earn a Master of Science in Engineering (MSE) with a specialization in data science. This online engineering degree delivers comprehensive background in both the technical and managerial sides of being an engineer today, while providing unique flexibility for working professionals.
All UCR MSE students will complete a quartet of core courses — Engineering in the Global Environment, Technology Innovation and Strategy for Engineers, Introduction to Systems Engineering, and Principles of Engineering Management — followed by a selection of classes specific to their specialization. For data science students, the latter covers topics in data mining, computer vision, information retrieval, web search, machine learning and statistical computing.
There’s no required residency for the UCR MSE, plus admissions and starting dates are both flexible. Under an expedited completion track, students may finish their online engineering degrees in as few as 13 months. These program features make the UCR MSE experience highly accommodating without sacrificing any of its rigor. Graduates can be confident they have the credentials to compete for in-demand positions in data science, whether they’re looking to become a data scientist, data visualization specialist, director of analytics or another role.
Visit the main UCR MSE program page for a quick overview of the online engineering programs, including the data science track. You can also answer a few simple questions there to receive a free copy of our program brochure with additional information on data science.
Recommended Reading:
MSE or Master of Data Science?
Career spotlight: Data scientist
Sources:
Breaking the 80/20 rule: How data catalogs transform data scientists’ productivity
Data Scientist: The Sexiest Job of the 21st Centuryx
The Quant Crunch: How the Demand for Data Science Skills is Disrupting the Market
What is Bias in Analytics? Computer and Information Research Scientists