Photo by Etienne Girardet on Unsplash

MEF BDA 503 Fall 2022

Another year of BDA 503 is behind us. It was the sixth time I gave this course and each year we are tweaking it to a better version of itself with my students.

This post is an overview of the Fall 2022 course which was taught completely online. I aim to create a written trail of our (both my and my students’) work. In this post, I will briefly explain what we did during Fall 2022 semester, what we did new and what we did the same. For this semester’s course webpage, please visit this link.

What is MEF BDA 503?

MEF University has a Big Data Analytics Masters Program aimed to teach “data science” to professionals from diverse backgrounds. BDA 503 is the code of the must course “Data Analytics Essentials”. See detailed explanation here.

What did we do?

Datasets we used in assignments are

• Startup Deals 2021: At each quarter, KPMG and 212 compiles a report about investments in Turkish startups. First group assignment was to preparing a report using Turkish Startup Investments Review 2021 data. One group wrote a Medium blog post titled “Turkey’s gaming sector is anticipated to continue expanding.”
• Foreign Students by Nationality: Students are asked to build a dashboard using data of foreign students from Higher Education Council’s (YÖK) statistics portal.
• Projects: This semester we had 5 groups. Each group selected a theme from Turkish National Statistics Institution (TUIK) and used the data to present their analyses. Four groups summarized their findings in Medium blog posts. Here they are.

Group Projects

1. “Discussion about Women Participation in Turkish Labor Force” by group Summer Moon.
2. “Analyzing Pandemic Effects on Turkey’s Tourism with R” by group Ramble On.
3. “Sectoral Analysis of the Covid-19 Pandemic in Turkey” by group Time and Space
4. “Analyzing the Factors Impacting Foreign House Sales in Turkey” by group Honey Drip
5. “Agriculture in Turkey” by group Peace and Trust.

What did we do new?

• We transitioned from RMarkdown to Quarto to publish “code-first” reports. Quarto is the most recent development from Posit (aka RStudio) to generate various documents such as reports, books and presentations. We used Quarto Book format. Main difference is in RMarkdown every Rmd file was individual and independent. In Quarto Book, students managed a “project”.
• We covered fundamental Operations Research topics. Even though it is not this course’s main topic, I cover basics of machine learning each year. This year we covered basic OR topics such as Linear Programming and Mixed Integer Linear Programming.
• We started a Medium publication: EDA Journal. BDA 503 students do not only learn but they also produce valuable content. I always “envy” grad course practice of writing a research paper in Stanford and other universities. EDA Journal is an organized way of publishing student work to public.

What did we do the same?

• We kept Github Pages practice. GitHub is by far the most favorite learning feature of this course. At first intimidating, students always get to like what GitHub is capable of (mainly versioning and GitHub Pages). It is also a best practice to have students check other students’ work.
• We kept using Slack as main communication tool. In a class of 20–30, it is hard to keep up with student questions and most of them have similar problems. They get to ask their questions in the channel and other students are usually able to give a proper answer. Also this way information is disseminated to the whole class, not just the students who asked the question.
• We kept having guest lecturers from industry to talk about their experience in the data science domain. It is really necessary to learn the theory but in practice stuff might be a bit different. Students slightly “despair” when they learn that there is much lo learn. I personally thank all our guest lecturers for contributing to this course.
• We kept using Datacamp for Classroom. It provides a certain amount of confidence to students.
• I kept getting student feedback at the end of semester with a very short and simple survey. Student sentiment about the course is consistent with previous years and participation rate is really high. One of the questions I ask is “What is your recommendation for next year’s students?”. Each year answers converge on the same line of “Start early, pay attention to assignments”.

Conclusion

We finished another semester of BDA 503 successfully. Even though the course became a very fine example of flipped learning, there are still areas for improvement. Spoilers…