It’s been about a month since I had my summer break from work and started my newest learning journey – Data Science (and/or Analytics).
I’ve been keeping a learning journal and thought it might be interesting to write something in longer form—as well as crosspost this to LinkedIn where this journey takes on a meta-angle with my job in LinkedIn Learning.
I’ll start at the beginning: I’ve gotten more interested in Data in my year at LinkedIn. Within my space (Learning), the data we get describes the way that employees are learning within organizations, and how those organizations align with macro trends in employee learning (for example: the rise of learning on how to work from home in March and April of this year).
Through my role, I’ve gotten to expand my skills in Excel/Sheets and in analyzing large data sets about learning behaviors. I’ve loved taking on that role, which has expanded my analytical abilities, and now I have set my sights on growing far past my current efficacy.
After talking to a dozen or so people within or around the Data Science space, I set out to learn some popular Analytics skills: SQL, for instance (and advised by nearly all to be the best starting place).
When it comes to starting new learning journeys, I’m lucky to have access to LinkedIn Learning and to know the tool well. I know, for instance, we have dozens (if not hundreds) of learning paths on all sorts of Data skills and hundreds (if not thousands) of courses on the subject. Coming from the learning and development space, coupled with my current role as a Customer Success Manager for LinkedIn Learning, gives me this access and knowledge but it also gives me a good idea on how to approach a long-term and in-depth learning goal like this.
Last month, when it came start to really begin, I actually started with a director competitor of ours: Udemy. Specifically this course. I started with this because it was the most recommended course I found in my research (mostly internet forums/Reddit/LinkedIn posts on learning SQL). It was said to be friendly enough to a complete beginner (though I had some exposure to SQL at a previous company) and I had never used Udemy so it doubled as work research. (The course was also on sale for something like $9.99).
The course was great. (The Udemy platform was fine for the consumer, though I do prefer some of the functionality of LinkedIn Learning – no bias :)). It was well thought-out and linear. The instructor did a great job explaining and then showing—and course quizzes reviewed lessons taught and challenged me to think beyond what had been in the lectures.
I finished the course (9 hours of content plus hands-on work) in less than two weeks and then immediately started with part two of my journey. This part included:
- Reviewing concepts from the Udemy course with LinkedIn Learning courses — specifically the Master SQL for Data Science learning path
- Putting my SQL skills into practice — I used three sites for this: HackerRank, SQLZoo, and db.grussel.org. All three of these sites utilize fake databases and give you prompts and direct and immediate feedback on your work. This is what is called a “kind” learning environment because of the immediate feedback, though none go so far as telling you what went wrong in a query if that happens. They just confirm a right answer when they see one.
- Preparing data sets from my own work (anonymized across clients and LiL analytics) to be put into a table to run queries on (much more on this later)
I’ll write about the progress of these steps and the ones coming after in my next post.
For now, my thoughts on starting this journey:
- I remain energized by the idea here: finding ways to search, view, and query a database to make sense of what’s in its tables
- I like the problem-solving aspect of SQL (as many have told me comes with learning coding languages). When a prompt asks for something specific (example: find all nobel prize winning authors in Economics who are from countries that have never had a nobel prize winner in Chemistry), your brain has to start untangling the mess for a starting point and a strategy to uncover the answer. That strategy is then written out in coded language.
- By far the most rewarding part of this learning journey (like many others) is to be able to put skills into practical application. I was able to take what I had learned and query on anonymized learning data—sorting the data in ways that Excel would not have allowed (or would have been immensely more complicated). I can’t wait for more of this.
Okay, all for now. More to come!