You Ask, I Answer: AI, Data Science, and What To Study in College?

In today's episode, Paul asks, if you were entering college, knowing what you know now, what would you study? Well, I mean, if I knew what I knew now, I wouldn't need to study anything, would I? Uh at least I wouldn't need to study things I already know. Let's put it that way. Um, this is an interesting question because I think the way college is structured, uh, it would actually be detrimental to the way I personally learn.

Now, I will say n of one, sample of one. Uh, everybody is different, everybody has a different learning style. My learning style is much more uh I guess the trendy term in education would be unschooling, where you pursue a line of inquiry and you pick up the skills along those lines of inquiry as you're going down whatever investigation you're going towards, with the understanding that you're gonna have gaps in your knowledge because you didn't have the formal frameworks for whatever those things were. When I think about what AI and data science folks need in terms of skills, it comes down to those big four areas that we've been talking about a lot for uh a couple of years now. You need those business skills and the domain expertise in whatever it is that you're you're working in.

You need technical skills, coding and such, uh data engineering, you need mathematical skills, uh, which are uh, you know, statistics, probability, linear algebra, and you need uh scientific skills, the ability to understand the and execute the scientific method. And you need all four in relatively equal balance, and that's one of the reasons why I say this idea of you know the college major where this is your single area of study, can be a bit misleading in terms of where it takes your education because it's not at least in AI and data science, you you you need to be equally strong in four different areas. And so instead of one area of focus, you might have uh call them mini major in each. You might have a mini meet major in probability. You might have a mini major in psychology uh or anthropology a mini major in uh programming or databases and a mini major in uh some other hard science although psychology is actually be a good a good fit there and that those four disciplines what you're looking to get are the frameworks it well what I would be looking to get I think it's really important to clarify that that's how I learn I don't learn like the standard educational system teaches it's actually a uh not a great student uh because the education system said we're gonna we want you to follow this very linear path from A to Z and don't skip steps along the way and don't get diverted and if you've ever had a conversation with me for more than 10 minutes you realize we go down rat holes and rabbit holes all the time my brain works differently.

For some other folks they might need that linear path but data science and AI are such that they are such broad disciplines and they require so many different prerequisite skills that you would still need a sampling of each of those. Now it's possible to to create that but you're probably gonna end up leaning towards one of those four areas more heavily I think the mathematics and statistics are important the computer science is important psychology is important and anthropology is important. And the psychology and anthropology are for two very important reasons. One learning the scientific method. But two, when you look at how we collect data and we use it for artificial intelligence and machine learning, and we look at the people who are doing it and how they're doing it, they're not always using the best practices, particularly if your AI team has a a bias towards the coding side.

They have not learned sample sizes. They have not learned statistical uh relevance and peace and P scores uh and and P hacking, unfortunately. Uh they have not learned margins of error and all these things if they come from a pure coding background. And they need to learn them, and you do learn them over time, but it's not the same as having frameworks and stuff up in advance. Now, the other thing that I would change in my own education is I think in terms of frameworks and structures.

So uh for those of you who have been on the Trust Insights website over TrustInsights.ai, you'll see this thing called instant insights in the resources section, and it's a whole bunch of one pages of essentially like PowerPoint slides of frameworks that I use a lot, and some of them are classic textbook frameworks like SWOT analysis, and others are ones that I've come up with in my own work. And the reason I like those is that it gives me sort of a quick reference handbook of uh a particular set of processes and the steps I need to take in order to to do the process well and repeatedly. And there are so many of these frameworks in anthropology and psychology and computer science and mathematics and statistics. And the way I was taught was I was taught more rote in the sense of um just memorization of facts and things without those containers, those frameworks for me to organize my knowledge in. And so it took me a really long time to relearn some of these disciplines.

I failed statistics in college. I did. I got on the final exam, I scored a 37 out of 100 because I had a teacher who was a brilliant researcher, avid publisher, prolific publisher, uh, and so he did great things for the university, getting published papers and stuff, couldn't teach to save his life, couldn't talk and it couldn't step down his teaching anywhere close to where a beginner would need to be. And so I didn't learn statistics until much later in life when I retaught it to myself using frameworks that I've Googled for and stuff and and read some textbooks to fill in the blanks where uh my education had uh totally failed me. And so part of that college education, knowing what I know now, it would be going back and filling in those frameworks.

I would take necess uh, you know, 100 and 200 level courses in each of these four areas. I wouldn't necessarily need to go beyond the 200 level, uh, but I would want to gather as many frameworks as possible so that I had them as references, and I know like this is when you use Porter's Five Forces, this is when you use SWAT, this is when you use Pest, uh, this is when you use P scores to measure error rates, or RMSEs or R squared, or uh or ROC. And so there's all these different rules and codes and frames of reference that uh I need to be able to do my work well. Again, I'm a sample of one. There are so many different ways to learn that I would encourage anybody entering college right now to first and foremost figure out how you learn.

Before you do anything else, before you take a single course, spend some time self-reflecting on how do you learn best? What are the methods and techniques that l that you're able to acquire information well? Um, is it linear? Is it nonlinear? Is it a line of inquiry?

Is it, you know, someone just giving you the information? Is it frameworks? Is it notes? Even something as simple as how you take notes is different for everybody. I think in mind maps, you know, the ability to drag different pieces around and see interconnected branches.

Other people look at that and go, I that that no, and they need that linear bulleted list, or they need pros, or they need to hear it, or they need to see it. Uh so more than anything, if you are entering college or you're in college, figure out how you learn. And then tailor your learning as best as you can to how your brain works, find mentors, or even just find people on YouTube who are subject matter experts in the disciplines you're studying. And if you find one that you understand, you listen to the person, you're like, ah, this person can explain it to me. Stick to it like glue.

Grab onto that and hold on to it tight because that's what's going to help you be successful is an understanding how you learn. Uh something I wish colleges would teach more. So really good question. Very, very uh self-awareness focused. Um, not a whole lot of technology in it, but that's what I would do if I was entering college now.

And it's what I would encourage anybody, no matter where they are in their career to do right now. Figure out how you learn, how you learn best, find people to follow that you can learn from, that you do learn from, stick to them like glue and acquire as much knowledge as you can from them. Worry less about the formal categories of learning and worry more about how you can accelerate your learning for yourself. As always, please leave your questions in the uh comments box below. Subscribe to the YouTube channel and the newsletter.

I'll talk to you soon. Take care. Want help solving your company's data, analytics, and digital marketing problems? Visit trustinsights.ai today and let us know how we can help you.