8 Lessons Learned: Udemy’s Intro To Data Analysis Course

Published by Klint Ciriaco on

Photo Credit: kreatikar

I decided against getting my master’s degree after doing some research. As much as I want to go to graduate school, I find it difficult to justify getting a $70,000 loan. 

I even asked for advice from a person I respected – he’s the head of growth of one of the largest marketing companies in the world – for further advice, and these are his exact words: 

“I don’t believe an MBA is necessary. And if you do go you have to go to a top 10 school which will cost you a lot. If you are professionally stagnating think of all the new ways of learning online or get a side gig or look for a new job. And if you do apply to go to school don’t go unless you get the full campus experience. Right now students are pissed as they are paying full tuition online.”

So I followed his advice and kicked off this journey by taking Intro to Data Analysis on Udacity

I experienced some ups and downs throughout the class, but it was overall rewarding. My hope in this article is to not only help you minimize your frustration when you take the course, but to also help you find ways to enjoy it. 

That said, here are the lessons I learned when I took the Intro To Data Analysis course on Udacity

Learn Python First

I could barely remember how to code in python because it had been years since I’ve used it. Instead of focusing on the exercises, I was spending hours on debugging. 

So, I decided to invest about a week relearning python before moving forward. It was a time well-spent because not only did I get less errors in my code, I started to have fun doing the exercises

If you don’t know how to code in python yet, I suggest learning it first before taking this class. 

Test Your Knowledge on Mini-Projects

After learning a new concept, find ways to test your skills on a small project. This is a fantastic way to find the gaps in your knowledge so you can patch them up. 

For example, I had a task at work where I needed to cross-check a list of email leads against a list of customers. In essence, I needed to calculate the number of email leads who became customers. 

There was no convenient way to track the conversions because the lists were stored in two different systems. There were also thousands of emails on each list which made them more difficult to process. 

When I learned how to use Pandas in python, I immediately thought, “I could use this for work!” And I did, but I made some mistakes along the way. I thought I knew the syntax at heart, and boy, I was wrong. I eventually fixed my mistakes and further deepened my understanding of the material. 

Give Yourself 30 Minutes to Solve A Challenge Problem

The temptation to look at the solutions is going to be high, but urge yourself to not check them until you’ve struggled for at least 30 minutes. Pushing your thinking skills to your limits will help you develop as a problem solver. 

If you get stuck, you can use some of the strategies below to help you get past an exercise. 

Ask yourself, “Why don’t I understand this problem?” 

This is one of the most powerful questions I asked myself because it helped me get unstuck. 

Whenever I couldn’t move past  an exercise, I asked myself the question above and tried to come up with an explanation as to why I was struggling. Some of those explanations were guesses, even. 

Here are some of those guesses: 

-Maybe I missed an important detail in the prompt (which was usually the case). 

-Maybe I don’t have the prerequisite knowledge to solve this exercise?

-Maybe I’m looking at this the wrong way? 

-Maybe the solution is simpler than I think?

-Maybe the solution has multiple steps? 

Let’s look at an example. There was an exercise where we had to standardize some data. I was confused because I didn’t know what that meant. So I asked myself the “magic question” above and answered it with, “It’s probably because I don’t quite get what standardization means.” So, I looked up “standardization.” An article showed an equation so I plugged it into my code, and bam. I got it right. 

Start With The Specific Before Solving the Broader Problem 

At the beginning of the course, there was a stark contrast between my problem solving skills and the instructor’s. 

I leaned towards tackling the problems head on. That approach worked on the easier exercises, but it didn’t scale on the bigger, more abstract ones, which caused me a lot of unnecessary headaches. 

The instructor? She took a small “slice” of the problems, solved it, gained some insight from it, and applied that insight to solve the bigger problem. 

I followed her approach, which made coming up with solutions easier.  

Take Short Breaks to Keep Yourself From Wasting Time 

There were several times when I was determined to finish an exercise. I was stuck to my chair for hours thinking up solutions to the exercises to no success.  But, I’d tackle the same problem the following day and then surprisingly come up with the solution in a short  time. The time wasted the previous day would’ve been prevented if only I took a break. 

Ever since, I’ve taken short breaks while in the course; I went for a  5- minute walk for every 25 minutes worth of work.

Don’t Think About The Finish Line

When you’re going through the grind and your brain hurts like it’s about to blow up, it’s easy to think, How many more sections do I have to go through until I’m done? Having this thought will tempt you to rush through the sections without gaining deep understanding of the material. 

To show why that’s bad, think of yourself as someone who’s building a chain. Each link represents your understanding of the data analysis discipline. If one link is weak, it’ll break, rendering the chain useless.

So, patiently tackle the sections one at a time. Yes, time is valuable, but you’ll end up wasting more of it if you finish the course without mastering the material. And besides, it’s cool to tell employers you’ve taken a data analysis course, but it’s cooler to have the ability to demonstrate mastery of the material. 

Get Good At Asking Questions

In my opinion, learning Pandas and Numpy isn’t the goal of the course because they’re just tools for gaining insights. The main objective of the class is for you to get good at getting answers from data. This means you have to get good at asking questions. 

How? I found Socratic Questioning to be helpful. 

Socrates, a philosopher, was thought to assume ignorance when exploring a problem and asked probing questions to gain a deep understanding of a subject. This is the same approach you need to take throughout the course. 

Read up on Socratic Questioning. It’ll help a lot. I find the article, Socratic Questioning: 30 Thought-Provoking Questions to Ask your Students to be a good read. 

Tackle a Project You’re Interested In

The final project involved analyzing data from either the Titanic or in baseball. Neither interest me so I went online to search for something else. 

Lo behold, I found data on the Crossfit Open. If you don’t know what the Crossfit Open is, it’s a competition where Crossfitters all over the world go against each other online. 

Since I’m a huge Crossfit nerd, I downloaded the csv’s, and then went to work. 

Don’t feel obligated to analyze the final project in the course. If you’re going to put yourself through weeks of brain hurt, you might as well work on data that interests you. 

Conclusion

I compare the whole experience to a beginner’s Crossfit workout program for the brain. It was challenging, but it was also rewarding because my “brain muscles” feel bigger and stronger than before. I didn’t just re-learn how to code, I developed some thinking skills that I could use in future courses. 

I’m just getting started though, so I’m looking forward to learning more in future courses. 

Categories: Application