Hi. I teach at a four-year school, but I'd like to join in. We offer a bachelor's and master's in Data Analytics. Data Analytics is a separate department, although the dean who is in charge of math is also in charge of data analytics. For the bachelor's we cover a range of Python, R, MySQL, PowerBi, probably others in electives. We require data analytics majors to take one semester of calculus, a math course in statistics, followed by a math course in multiple regression, but we are not what I would consider a math-heavy data analytics program.
I suspect that where the program is housed makes a lot of difference. If it is housed in the computer department, then it will probably be a more computer-oriented program than if it is housed in the math department. A third possibility is housing it in the business department.
I personally had felt the whole data analysis thing had somehow passed me by, so I went back to school and earned a masters in data analytics at the same school. My master's included more data analysis tools such as decision trees, random forests, market basket analysis, and principal component analysis (PCA). (By the way, PCA finally answers my age-old question of what is a practical use of eigenvalues and eignevectors.)
One thing my school's bachelor's and master's program omits is computer simulation. I think computer simulation is an extremely versatile tool that helps address problems that have no simple closed-form solution.
I continue to learn by subscribing to blogs of people who post their R code, and then by trying out their code. I very occasionally post myself on one R blog. R is incredibly powerful. I can do so much more in R than I can in Excel. (The same can probably be said for Python, but I have concentrated on R.) For example, I cheat in Wordle because I wrote a program in R that narrows down the possibilities after each guess and tells me the frequency of letters in the remaining words. Yes, I know this is cheating, but I am pleased with myself for having written the program - it would be so much harder to do in Excel.
There are seemingly unlimited real-world databases available online to experiment with, and I encourage suggesting students explore these. You can play with credit card default, housing prices, baseball, speed dating, politicians' tweets, and the list is seemingly endless.
Happy data sciencing.
------------------------------
Jerome Tuttle
Online adjunct instructor
Southern New Hampshire University
Manchester NH
------------------------------
Original Message:
Sent: 07-04-2022 12:25:38
From: Rachel Saidi
Subject: IMPACT in Action - Questions about Data Science Pedagogy
- What kind of program do you offer or will be offered? A certificate, A.A. degree, A.A.S degree, or A.S. degree?
- Where is the data science program housed? In the math, statistics, computer science, information technology, or other department?
- What types of data tools and skills will the program stress? Will the program include proprietary software or open source programming languages? (Examples: R and R Studio, Python, SQL, Git/Github, LaTex, Power BI, or Excel, to name a few)
------------------------------
Rachel Saidi
Associate Professor and Data Science Program Coord
Montgomery College
Rockville MD
------------------------------