1 Year Into the Program: Georgia Tech OMSA vs UC Berkeley MIDS
Which investment is worth your time and money?
I am teaming up with my good friend and former colleague, John Lee, to compare our experiences after 1 year into 2 different online master programs in data science/analytics.
We break this article down into 3 sections: 1) share a little bit about ourselves, 2) summarize 2 programs in 2 minutes, and 3) answer 10 questions about what we know now about our programs.
About Your Contestants
Starting school at the same time in the Fall of 2019, John enrolled in the University of California, Berkeley Master in Data Science (MIDS) program while I enrolled in Georgia Tech Online Master of Science in Analytics (OMSA) program.
To set the stage…
- We are both very similar in age (28–30) and stage of life (married to a spouse in different cities, no kids).
- We both shared similar career paths (previously as environmental engineer students in undergrad turned mechanical engineers in HVAC/energy efficiency). At some point, we were coworkers at two different organizations. We have about 6–7 years of working experience.
- We both study part-time on top of our day-job (not a data scientist).
- We both have little programming or data science experiences when starting the program.
- We both completed the majority of the foundational courses in our programs.
Why are these similarities relevant? This warrants some context about the biases from our experiences.
What does 1 year into the Programs look like in 2 minutes?
We consider “1 year into the program” equates to 2 semesters (Fall & Spring). Although, MIDS technically has 3 semesters per year. Summer classes are required for MIDS, but not for OMSA.
Disclaimer: Please keep in mind that you should refer to the official school’s website (GaTech OMSA, UCB MIDS) for the most current information.
GaTech OMSA
Courses Taken: 3 out of 12 courses
a. ISYE 6501 — Introduction to Analytics Modeling
(Difficulty: 3/5, Avg Hrs/week: 10–15)
b. MGT 8803 — Introduction to Analytics in Business
(Difficulty: 2/5, Avg Hrs/week: 5–10)
c. CSE 6040 — Introduction to Computing for Data Analysis
(Difficulty: 3/5, Avg Hrs/week: 10–15)
There are two additional courses that are considered part of the “must-take” course list for OMSA students: d. MGT 6203 — Data Analytics for Business, and e. CSE 6262 — Data Visualization & Analytics.
Duration: You have up to 6 years to complete the program. You can take up to 2 classes per semester (contact the program if you want to take 3+ classes). I find that 1 class per semester is most suitable for my work-life balance. If you need to opt-out for one semester (Fall or Spring), you need to contact the program.
Cost: (3 classes) *(3 units/class)* ($275/unit) + ($304 campus fee/semester) * 2 semesters = USD $3,083.
Resources: Typical online resources tools to GT students (library, Programs, tech discount, etc.), Slack Channel (need gatech.edu email address or email the slack owner), Reddit
Acceptance Rate: The on-campus version has a lower acceptance rate. OMSA acceptance rate was about ~24–27% in 2017 and moved to ~70% in 2019 (more GaTech stat).
Community Size: There are about ~2.8k enrolled students between 2019 and 2020. We have about ~3.7k in our OMSA-Study Slack Channel and Reddit as of May 2020.
Preparation before Joining the Program: I took UCSD edX Python for Data Science and paid $350 for Verification Certificate. I also did some projects on the side. Projects + edX class probably totaled around 80–100 hours.
UC Berkeley MIDS
Courses Taken: 4 courses out of 9
a. W200 — Python Fundamentals for Data Science
(Difficulty: 3/5, Avg Hrs/week: 18)
b. W201 — Research Design Applications for Data and Analysis
(Difficulty: 2/5, Avg Hrs/week: 10)
c. W203 — Statistics for Data Science
(Difficulty: 4/5, Avg Hrs/week: 20)
d. W205 — Fundamentals of Data Engineering
(Difficulty: 3/5, Avg Hrs/week: 8)
There is an additional foundational course: e. Applied Machine Learning. See list of courses here.
Duration: MIDS requires the summer semester as part of the maximum of 8 semesters (including summer semesters) to graduate. MIDS allows students to take semesters off for emergencies.
Cost: (4 classes) * (3 units/class)* ($2,573/unit) + ($728.50 campus fee/semester) * 2 semesters + $95 (1-time)= $32,428. (Official Site)
Resources: Typical online tools available to any Berkeley students (library, Programs, tech discount, etc.), WeWork Access (super cool), Slack Channel (once accepted into the program), Zoom, In-person Conference available 3 times per year normally (however, students are only required to attend once, $500/registration + travel costs)
Acceptance Rate: I don’t know. A MIDS Program Status Report in 2018 revealed some numbers such as the program accepts about 150 students per year, aiming to grow to 450.
Community Size: We have about 880 members in our Slack channel and 2.5k in the official School of Information channel as of May 2020. You’ll need ischool.bekeley.edu account to join once you’re accepted into the program.
Preparation before Joining the Program: I took online courses on Data Science Fundamentals, Python programming, Intro to Computer Science, and Intro to Algorithms. In addition to studying for the GRE (not required anymore), the total time of personal preparation was about 200 hours.
10 Things We Learned ~1 Year Later
Having completed the Fall and Spring semesters, our understanding of our respective program is definitely clearer than we first applied.
1. How are the courses “experience”?
- GaTech OMSA: Classes size ranges from 400 to over 1000 students per class (if you count the Edx MicroMaster/Verified Students). This number might be lower after the foundational classes. Piazza (official course forums) was sometimes hard to follow through every post, but luckily there are separate forums for GaTech and Edx students. Lectures are pre-recorded. Grades are heavily based on exams (take this with a grain of understanding there are some variations). In general, I think this is similar to the traditional classroom experience I had in undergrad.
- UC Berkeley MIDS: Classes size is below 20 students/class. Lectures are live and recorded. Grades are heavily based on Projects.
2. How much time do you spend each week?
- GaTech OMSA: OMScentral is a great class review site where you can get a broader sense of time commitment and difficulty varies by class. Personally, I am around the averages of most classes, where I average around 10–15 hours/wk/course.
- UC Berkeley MIDS: Classes discussions are mainly on Slack. I average around 15 hours/wk/course while reading through at least 50% of my assigned reading materials.
3. What is the most surprising thing we learned about the program?
- GaTech OMSA: There is definitely an “art” to analytics as much as the “science”. Additionally, not all the courses advertised are available every semester, especially for the Computational Track. You also can not take OMSCS classes unless they are offered through the OMSA program, which may change every semester.
- UC Berkeley MIDS: I’m amazed at how classmates manage to work full time and manage the workload. This online program is also a lot better than I had expected. For one, I assumed an online program is not much different from a typical online course from Udemy where you watch lectures and you do a few assignments (without caring much about the quality). However, live sessions and collaboration with classmates on projects get you motivated. Additionally, struggling through the work with others really teaches you more than taking a class on your own as questions/topics that you would’ve never thought about are discussed.
4. Were you able to apply what you learned?
- GaTech OMSA: I was able to apply what I learned more than what I expected to. For example, CSE 6040 enabled me to apply regex in parsing pdf files. MGT 8803 helped me understand the financial lingo in accounting and budgeting that supports my manager role. ISYE 6501 prepped me to constantly think about how to apply machine learning methods to real problems.
- UC Berkeley MIDS: A few classmates of mine collaborated on an exploratory analysis of the COVID-19 data set. The statistics course (W203) has taught me not to blindly trust any study. You really have to look into its methodology and decide for yourself if there are any false positives or negatives.
5. What do you enjoy most about the program?
- GaTech OMSA: I enjoyed that I was exposed to a lot of the fundamental stats and machine learning theory. This was important because I could understand mathematical proofs and papers. I felt more confident in understanding the materials and motivated to find potential applications. I really enjoyed the active slack channel too and connecting with people outside my current career domain.
- UC Berkeley MIDS: What I really appreciate about being a part of this program is gaining confidence in employing data manipulation and getting values from the data. It is a total 180 from before I started the program. I like how learning was progressive, you learn to set up your environment, learn concepts, then apply it on a project.
6. What do you least enjoy about the program?
- GaTech OMSA: It can be isolating sometimes.
- UC Berkeley MIDS: High tuition cost.
7. How do you feel about the overall quality of the program and/or classes you took so far?
- GaTech OMSA: The foundational courses are beginner-friendly as long as you put in the work. I do gain value from each class that I took. On the other hand, I do perceive that the program class offerings can be improved after a lot of reviews from prior students. I have a difficult time picking classes due to availability and sequential design. You have to ask around to know which classes are a good idea to take first before the other.
- UC Berkeley MIDS: The foundation courses are doable for a beginner in the data science field. But I always meet classmates (who know so much more) that make me feel like I am not supposed to be here (imposter syndrome). For example, I had completed a data streaming to a data analysis project neatly, clearly, and per instructions. After submitting, another classmate went above and beyond and shared their same project but instead, they created a game that attacks zombies to generate streaming data. Seeing what the other students have done really motivates me to attempt more difficult tasks.
8. Would you pick this program again if you can go back in time?
- GaTech OMSA: Georgia Tech Online Master of Computer Science (OMSCS) is a very popular “older-brother” online program. In hindsight, I think I should have picked OMSCS because it aligns more with my personal goals and outside my comfort zone. For example, if you want to become a Machine Learning Engineer or be more involved with the production pipeline, opinions on Reddit suggest OMSCS may be a better path because you are better prepared for the development pipeline. I am a little worried I won’t be taken more seriously in a more technical role in the future. Nevertheless, I really enjoy what I’m learning and plan to stick with the program.
- UC Berkeley MIDS: Yes. I still don’t know whether I’d prefer other programs more because I don’t know how the other programs would be. Regardless, I am proud of and happy to be a part of this program. It is a good fit for me, someone who did not know where to begin to learn data science.
9. What was the worth it factor about your program?
- GaTech OMSA: The worth-it factors are definitely the affordability and highly reputable school. There are some courses that are really challenging (20+ hrs/wk) and others that are more flexible courses (<20 hrs/wk). If you select your courses strategically, you can finish the program faster if that is your goal (~ranges from 1,200 hours to 2,000 hours). It’s an interdisciplinary program mixed with Business classes as well, which may be adding brownie points for certain employers to fund your advanced degree.
- UC Berkeley MIDS: Berkeley MIDS definitely has a strong network, a community of motivated peers, and a prestigious school brand that is well known in the U.S. West Coast. Although I haven’t discussed with my network more than just course-related material, a few classmates I interact with work at Nvidia, FANG, and other silicon tech companies. The benefits of having such class cohorts have really been seeing their take on topics such as homework problems, how their workplace would have approached similar problems, and just their knowledge in the field.
10. What would you recommend to prospective students for each respective program?
- GaTech OMSA: Garbage in = garbage out. That goes with everything from data analysis to what you get out of this program. I recommend all prospective students to read this Reddit’s FAQ guide. Save it. Book it. Read it multiple times. Take the prerequisites seriously (coding, calculus, prob/stat, linear algebra) and try out simple projects (1, 2, 3) with Jupyter Notebook (or whatever tool you prefer) will help you feel less frustrated in the beginning. If you’re thinking about applying to OMSA, here is my blog on how to get in.
- UC Berkeley MIDS: Take advantage of your network. You can learn a lot from them both in class and professionally. Many are highly accomplished working professionals from the California Bay Area and tech firms. Treat your projects as if they will be showcased in your portfolio for hiring managers. Don’t worry about the grades, if you work hard, the grades will follow. Dive into simple DS projects earlier. Learn some Linux commands and git. Once classes start, it gets pretty busy.
Each program offers a different experience. Whether they’re right for you, it’s ultimately something you have to decide for yourself.
If you’re contemplating between either school, we hope this offers valuable insights for your research and decision-making process. Clap if you find this helpful or would like to see more questions answered about these online master programs! :)
Feel free to connect with us on Social Media:
