Data Science Bootcamp Reviews : What do Our Graduates Think?

We see this question a lot. It has been asked on Reddit and Quora forums. While we do have testimonials all over our site, we thought it may be useful to compile some information into a single blog post. The data that we used to generate this response is 100% from our students. After every bootcamp, we ask the graduates to anonymously write what they liked and disliked. We then used these free-form student responses to generate the word clouds shown in this answer.

This is a brief summary of how we created each word cloud for “what did you like most” and “what did you like least”:

  1. Created text files from the survey responses
  2. Read these files into R and removed punctuation, numbers, stop-words (like “a” and “the”), and our own personal stop-words (“lot,” “like,” “data,” “much,” “bit”)
  3. Stemmed the remaining words (removed the endings) This explains why our word clouds have nonsense words like “theori—“; they have been stemmed.
  4. Calculated the frequency of each word
  5. Plotted the words that appeared more than 5 times

First, here is our word cloud for “what did you like least”:

wordcloud_dislike_color

The most common complaints are “time” and “day.” From our experience, people are sometimes unhappy with the pace of Data Science Dojo, especially during the 5 day in-person bootcamp. You’ll notice that “cover” and “topics” are also complaints. Five days is a short amount of time to absorb all the topics that we want to teach, so we supplement it with pre- and post-bootcamp webinars and videos. However, when it comes down to it, the bootcamp is 50 hours in 5 days; that’s more hours in a week than a full-time job. It’s fast paced, but our instructors are there to help, both in-person and online.

Now for the more positive stuff, which we obviously prefer to talk about. This is the word cloud for “what did you like most”:

wordcloud_like_color

The most well-liked parts of the Data Science Dojo bootcamp are the instructors and the hands-on labs. These are consistent responses. Students rank the instructors from 1 to 10 as well, and the instructors’ ratings average between 8 and 10, as shown in the table below. You’ll notice that our CEO, Raja, even got his name in the word cloud because people mentioned him so often in “what do you like most.”

 Instructor Average Rating
Raja Iqbal 9.2/10
Phuc Duong 8.8/10
Brian Johnson 8.8/10
Overall Satisfaction 8.9/10

The other strength is the hands-on portions of the class. The curriculum includes a whole day of data engineering, building a model in Microsoft Azure (“azur—“), and using the model to enter a Kaggle (“kaggl—“) contest. People seem to like using these tools in addition to learning the theory.

Oddly enough, theories made both word clouds. We suspect that some people enjoy the theory and others would rather work in the labs all day, so Data Science Dojo does both.

We hope that this clearly and honestly shows our graduates experience with the program.