Ultimate FAQ to University of British Columbia Master of Data Science

Typically, I get two to three LinkedIn requests a week with questions about how to apply to the University of British Columbia Master of Data Science program (UBC MDS). From these requests, I have compiled some of my most frequently asked questions about the MDS. I will be linking to the official MDS website because they have the process pretty well documented and the rest will be based on my personal experience. 

Q: I don’t know if MDS is right for me and I don’t know where to start. 

A: If you don't know where to start I would consult the admission page, the last official Q&A video, the MDS official FAQ, and follow MDS on Linkedin. 

Q: Is the MDS program worth it/would you recommend it/would you do it again?

A: Absolutely, I recommend the MDS program. The program completely changed my career trajectory. I wanted to be a business analyst out of my undergrad. However, while my undergraduate education prepared me with the soft skill side of a career in data, I felt I was missing some of those hard skills. I was looking for additional schooling to gain these skills. However, I didn’t know what I wanted to do was called “Data Science”. It wasn’t until my best friend dropped a link to the MDS website in our group chat that I became obsessed with the idea of becoming a data scientist. Additionally, my best friend and I both got into the fourth cohort and since then our younger brothers have also done the MDS program. From my perspective, MDS was the best decision I have ever made. 

Q: Does the MDS program guarantee you a job?

A: Absolutely not. With every Master’s program, it is a matter of “measured risk” and I would say the MDS program is no exception. I would say the only consistent way to become a data scientist is to have experience as a data scientist. The industry as a whole is competitive and I have seen 200+ applications for a Data Scientist job within 24 hours of being posted on Linkedin. MDS facility has transparent statistics on alumni employment.

Q: How long did it take you to find a job as a Data Scientist?

A: It took me about four months, 52 job applications, and 25 cold calls, to find employment as a Data Science Intern. There are about 100 MDS students entering the job market at the same time and there aren’t that many positions in the summer. The graduates with experience in data science or consulting were the first to get hired. Some of my peers, including myself, took jobs as interns, data analysts, or data engineers as intermediate steps to gain employment as data scientists. 

Q: Did MDS prepare you for a job as a Data Scientist?

A: Yes, I use skills I learned from the MDS program every day in my job. However, it is impossible to learn every data science skill imaginable in ten months. MDS program gives a good foundation to the fundamentals of data science. However, there is always an opportunity to grow those skills and learn new ones for your specific area of interest. I am still constantly learning new techniques and software for my job. 

Q: Did MDS prepare you for a technical interview?

A: The MDS provided me with enough skills as a baseline for the technical interview I did for my current job. However, I would recommend you practice doing mock technical interviews to feel more confident during actual ones.  

Q: You graduated with an Undergrad in Business Administration, how much of a coding background did you have?

A: I met the MDS computer science requirement with a computer science class I took as an elective during my first year of undergrad and it was taught in a programing langue called C#. This was not enough preparation, so I took eight months to self-teach some of the fundamentals of R and Python before the program began. By the time I entered the program, I did several small personal projects such as visualizing Pokemon statistics, building a Craigslist web scraper to help find an apartment, and performing a random forest classifier on R Titantic. After all this practice, I would say I still came in with below-average computer science skills compared to others in the program. Only now after two years of using Python do I feel competent using the language.

Q: How do they decide on who is admitted? 

A: This is pure speculation on my part, but the MDS program wants a diverse cohort and there isn’t a way to “game the system”. I would say your acceptance depends on a mixture of your work experience, GPA, letter of intent, and academic/professional recommendation letters. Here is a link to the MDS admission statistics.

Q: How can I best prepare for the MDS program?

A: For fundamentals, I would recommend focusing on Python’s NumPy, Pandas and Seaborn, and R’s Tidyverse (swirl is a good resource). Other than that, I would practice by doing projects you find enjoyable. Data Science should be a fun process, I started out making visualizations for Pokemon statistics. 

Q: What will I be learning in the MDS program? 

A: I would estimate that one-third of the program is dedicated to statistics, one-third programming/machine learning, and one-third miscellaneous skills. Here is a link to the coursework.

Q: Is there an opportunity to work on research while doing the MDS program?

A: To the best of my understanding, I would say no. MDS is more of a professional master’s program rather than a traditional master’s program. If research is important to you, I would recommend you consider another program. There is a two-month capstone project at the end of the program which is a good opportunity to gain some real-world data experience before entering the job market. Link to the previous capstone projects. 

Q: What was your Capstone Project?

A: I worked with a team of four to help provide the mobile phone retailer GLENTEL Inc. with an analysis on the keywords in resumes and that correlated with future employee performance. It was a fun project that helped my transition to work on customer churn models as part of my internship with PagerDuty. You can read more about my capstone project here. 

Q: Would recommend UBC MDS over a data science boot camp, self-study, or other master programs?

A: UBC MDS was what I needed personally to go from data science as a hobby to a profession. I would say it depends on where you are in your personal data science journey. If you are already working in the data industry and have a lot of the prerequisite skills you may only need a 12-week Bootcamp to transition. If you have a lot of personal motivation and strong programing/mathematical skills you may be able to teach yourself the skills on your own and demonstrate it with a personal project. A lot of it comes down to your personal circumstances and understanding that MDS is not the only path to become a data scientist. 

Q: What are your key takeaways from the MDS program?

A: 

1. Interpretability over accuracy
2. Data storytelling is undervalued
3. Not everything you learn will be applicable to every problem you encounter
4. Start with simple models and pile on complexity i.e. don't start a neural network unless you absolutely need to
5. Find your niche in data science early and build your profile around it

Q: What does workload look like?

A: Glad you asked! I wrote an article about it and Jonathan Chen did a lot of awesome visualizations!

Q: What did you do for personal projects during your time as an MDS student?

A: My personal philosophy when it comes to the personal project is the story matters more than the difficulty of the work. My requirements for myself were that I had to be able to write a fun and engaging article at the end of the process. 

1. Man vs Machine Learning: Movie Poster Prediction
2. Oscar Machine Learning Model vs Data Scientist’s Girlfriend (1/2)
3. Oscar Machine Learning Model vs Data Scientist’s Girlfriend (2/2)
4. What is MST3K and How is it Affecting my IMDb Review Prediction Model?
5. TOM TALK - Walmart M5 Forecasting

I hope this acts as a good guide you help you start to think about whether the MDS program is the correct one for you. Feel free to connect with me on LinkedIn and ask me any further questions you may have. Who knows, maybe your question will end up in this guide :).