Why the world of data?
My most direct path to the study of data science was through work experience in digital marketing. For some 15 years, I have been regularly tracking the success of digital marketing campaigns I have designed and ran for Google Adwords, social media and web-based content marketing campaigns. While putting together reports on these campaigns, it didn’t take much imagination to see the value of adding forecasts to these reports. I began looking around for tools to help expand the scope of my work.
About 4 years ago, I discovered a few graduate programs in Chicago called, variously, predictive analytics, business analytics, or business analysis. I don’t remember coming across any called data science then. I thought serious about enrolling, but I was unsure if these programs taught the skills I was looking for and if I was ready for the challenge.
My first step was to take a 3-month online course on data analytics last summer through Northeastern University. The course wasn’t data science, it used Excel instead of Python, but the bulk of it focused on statistics, which I am interested in but until then had little knowledge of. Success in that course encouraged me to continue, so I began to scout out data science boot camps. I wanted to get enough experience to become a data professional, but wait on graduate, i.e. ‘night’, school until I had some real-world experience in the field.
The two programs in Chicago that intrigued me most were the data science immersive bootcamps offered by Metis and General Assembly. My impression was that Metis had a little more of a theoretical bent, while General Assembly struck me as having a slightly more practical approach. The programs were both impressive, but my interest in data has always been pragmatic, how to help a company get an edge in business or how to help any type of organization improve its operations. I held my breath and signed on at General Assembly.
Besides working through the prep materials for the course which focused on statistics and Python, I found an online course on Coursera that helped prepare me conceptually. Titled “Computational Thinking for Problem Solving,” the course is essential for anyone without a STEM degree who wants to learn data skills. I particularly enjoyed the section on the “Four Pillars of Computational Thinking,” which are decomposition, pattern recognition, data representation and abstraction, and algorithms. Translated into regular English: 1) break a big problem into smaller problems, 2) look for the patterns, 3) translate the messiness into a rational, organized representation, 4) define the step-by-step process to arrive at a solution.
All of this enlightened thinking somehow brought me back to my youth studying jazz drums. I had an amazing professor, Professor Milford Graves, who could play any style of percussion from around the world on the drum set. That was beyond me, but I still learned valuable lessons from him. One was the importance of breaking things down to the basics (decomposition) so you can see and master the concepts (pattern recognition & abstraction), then put the basics together into complex drumming styles (algorithms). Professor Graves emphasized that this process was essential to learning anything throughout life. He was right.
When I was in graduate school for art history, focusing on painters in early 20th century Europe, I was frustrated by what I saw as lack of a coherent process for cultural history. Theory, or the -isms (postmodernism, postcolonialism, Marxism, etc.) reigned in the 1990s liberal arts world. Often, it seemed we were abstracting to these theories when they weren’t necessary a good fit for the ‘data’. Many of us longed for a better approach. It seems to me that the processes of computational thinking have a lot to offer to the humanities. It was time to jump into boot camp and learn more.
