Back to School

Selection criteria

Now that I was (mostly) sure I wanted to take data science as major, it was time to start shopping programs.I tried to be thorough.

I searched a number of schools and types of programs. I started out looking at “boot camps”, since I already had a master’s degree and several years of experience. It seemed unusual to go for a second master’s, and a short, intense program seemed appealing at first. There were a number of problems though. I was seeing mixed reports of how well graduates performed in new roles and from graduates who had a hard time finding a job. The cost was much higher than I expected, ranging from around $3,000 from schools I’d never heard of to upwards of $12,000 from a name-brand school.

What really turned me away from then is that they were almost purely technical. How to code in this language, how to code in that language, how to code for databases… with little to no theory or background on it.

Of course, there are all kinds of online courses I could take. EdX, Coursera, and Udemy have more courses than I could ever take in a lifetime. It’s true, in 2020 the information is there… but I really wanted something a little more structured, something that would give me a roadmap of how to get from where I am to where I want to be. There were some “specializations” (groups of courses on a related topic that build to a final project) that seemed like good prerequisites for a master’s, but they didn’t seem like enough to me.

That led me back towards master’s degree programs. I had a few criteria for the schools I looked at:

  1. The degree had to be available online. There are some stellar degrees if I wanted to move to Pittsburgh or Boston, but that’s not an option right now.
  2. I specifically wanted “Master of Science in Data Science.” I found several programs that were master’s in other things, like statistics, applied mathematics, or computer science with specializations or focuses or tracks on data science. Those were all ruled out. I found a few variations, like “applied data science”, that still met other criteria so I didn’t rule that out.
  3. Courses. I read through the course catalogs of dozens of schools to see what was required and what electives were available. I counted how many required or elective courses I was actively excited about and noted how many required courses I had no interest in.
  4. Prerequisite options. Obviously, data science requires some solid knowledge of statistics and programming, and even more for certain specializations. Since I don’t have that, either from school or professional experience, I needed to meet the entry requirements. Since this was already going to take a couple years and cost a lot (and I’m not young), the ones that were more flexible were the most appealing to me.┬áThere were a few options that I found:
    1. Require candidates to take college-level courses from accredited institutions and get at least a B-, which adds a year and several thousand dollars to the entire endeavor.
    2. Let candidates take any course they wanted, from structured courses with certificates to free open online courses from MIT, as long as you can pass a test or convince the selection committee you know the material.
    3. Offer a skills test and offer their own online courses to help you prepare for it.
  5. Student stories. I read Reddit posts, blog posts, and anything else I could find with people talking about different programs, their experiences, and both pros and cons they saw. There were so many people recommending Georgia Tech’s analytics degree that I started to think they had a referral program, but it’s primary selling point seems to be “it’s cheap.” I wasn’t overly excited by the courses in the program.
  6. Name brand. Yeah, I’ll admit it. I wanted a school that looked good on my resume.

With those in mind, I applied to four schools:

  1. University of California, Berkeley. Their Data Science program looks absolutely amazing and it is by far my top choice. I spent hours going over the capstone projects that their graduates created and I am just dying to go there.
  2. Johns Hopkins University. They passed all my selection criteria, though a few reviewers said the program felt like two majors crashed together and not a cohesive program. (after I applied and was provisionally accepted, they told me their requirements for prerequisites (see 4A above), and I declined).
  3. Northwestern University. Another solid program that I liked a lot and couldn’t find any downside to. They have a LOT of cool looking AI courses.
  4. Harvard. Believe it or not, the least expensive of the bunch. They have a unique application process… take two classes, get a B in them and you’re in! Of course, the hard part is pulling off Bs in Data Modeling and Advanced Python for Data Science.

So that’s it!

Next time, I’ll talk about the prerequisites and what I’m doing to get ready (plus any updates on my applications)