It’s becoming increasingly acknowledged that big data and new computational methods are changing the world of social science. However, some social researchers are still uncertain about learning data science skills - and about the benefits it can bring. SAGE Campus learners have told us that they “wish they’d learnt this years ago” but weren’t aware of the true benefits of these skills.
That’s why we asked Matt Denny, Research Scientist and author of the SAGE Campus Practical Data Management with R online course, to write about what he’s found to be the benefits of learning data science.
Matt did his PhD in Political Science at Penn State, where he was an NSF Big Data Social Science IGERT fellow, and is affiliated with the Massive Data Institute at the McCourt School of Public Policy at Georgetown University. He’s worked as a data scientist at Skopos Labs, and cofounded Cerenetics, Inc. while in grad school.
Below, Matt writes about what he found to be the key benefits of learning data science.
What’re the benefits of learning data science skills?
I always find it’s easiest to learn something when I enjoy doing it. Many aspects of data science fit this description - and I have personally enjoyed the overwhelming majority of the data science work that I do, both in academic and industry settings. At the same time, it’s good to have a broader understanding of the value in learning data science skills on top of a social science background. The way I see it, there are three main benefits of developing these skills, especially if you are earlier in your career, which I detail below.
For me, a good data scientist is someone who can (mostly) independently follow a research project from question formation, through data collection and cleaning, and on to analysis and interpretation. In other words, a data scientist possesses the basic skills to ask a question and answer it using data.
I would usually just get discouraged and try to find some other project to work on where "nice" data was available. After a few years of learning how to program and how to collect and manage complicated datasets, I noticed that I could pretty much just pick something I wanted to study, and gather and clean the data I needed if it did not already exist.
This kind of "self-sufficiency" is also very important in industry, where you may be the only person in your organization working on a problem, and where I have found people place a high value on being able to "make things work" on your own, since you are often the "expert" compared to anyone else in the company.
One thing I’ve noticed since I started feel proficient as a data scientist is how much more varied the projects I work on are than they used to be. Some of this is simply due to my interests becoming broader, but I think another part is that I have felt more comfortable exploring new domains because my data collection and management skills have made it feasible for me.
For example, at one of the data science start-ups I worked for, I once ended up working on large scale data collection and analysis projects in four totally separate domains within the course of a single year. While, I had to develop the substantive expertise for each of these domains, I found I could apply the same programming skills in each one, which made the transition much easier.
Whether you are looking for an academic job, or interested in a career in industry, the efficiency gains you can realize from applying the programming skills you develop through your data science training in your daily work are a huge advantage.
To start with, learning to program can allow you to automate manual tasks in a way that can save hundreds of hours of your time. For example, one task I often have to do is to correct spelling or punctuation errors in the names of organizations I have collected data on over a long period of time, before I can analyze the data.
Additionally, learning data science skills can help you to deal with much larger datasets in much less time, which can open up new projects that otherwise would have taken too long to clean and analyze. Particularly in industry settings where there may be external time pressures to answer a research question, these efficiency gains can be an absolute necessity.
Whether you are considering a job in industry or academia, developing skills as a data scientist can dramatically extend your capacity to do research, and broaden the range of topics you are capable of working on. And my personal experience is that the demand in industry and academia for data scientists and researchers with social science domain expertise is continuing to grow.
Hopefully this encourages you to get started in data science with the course offering at SAGE Campus. Good Luck!
- Matt Denny, Research Scientist and course author of Practical Data Management with R.