ADAPT’s 10 common learning elements encompass data science and AI-specific goals that support interdisciplinary learning and participation in the professional data science and AI communities.

A single course cannot usually address all of these learning elements, but the data science and AI education program aims to address these as the students walk through the sequence of data science and AI courses. This approach supports a coherent course development process that takes into account the interdisciplinary and collaborative nature of data science and AI. The 10 elements are organized into three categories: Data Perspectives, Data Practices and Data Discoveries. 

Data Perspectives

1. Recognizing data as information – not truth – with error, variability and missing information

“A lot of my students had never considered that AI could be wrong or biased… reading about facial recognition not working until a pale mask was used was eye-opening for them… It’s important for students to understand that AI and data are not perfect or true; they have to always interrogate them and be somewhat skeptical without just dismissing everything.”

Mahmoud Harding

Intro to R/Python for Data Science

2. Explaining what it means to be a data scientist and AI expert

“One of the things I’ve been very intentional about, since we’ve been having those discussions about a year ago, is talking about the role of the data scientist after running code. Just because there are no errors doesn’t mean the code is useful or that it did what you intended. A data scientist is curious—they check, verify and ensure they understand the context. To support this, I’ve added more opportunities for students to write about what the code should do and articulate their thought processes.”

Mahmoud Harding

Intro to R/Python for Data Science

3. Observing a variety of data scientist role models and careers

“One thing I was really pleased about was being able to bring in a guest speaker from the Secretary of the Department of Revenue, who is African American and has a legal background. It was interesting for the class to see that data scientists can come from various backgrounds and that data science is complementary to many other fields.”

Andrew Peterson

Intro to APACHE Spark Using Big Datasets

Data Practices

4. Examining how data are created and the related assumptions and collection practices

“My students gather their own data and they have to figure out the data gathering process. We discuss source bias and what happens when an interview pool is limited, as well as the impact that can have. One of the things students also discover is just how difficult it is to get people to agree to take a survey or participate in an interview. It’s something that many of them need to realize—you can’t just send out a survey and expect to get responses in a day.”

Christin Phelphs

Intro to AI Ethics

5. Practicing data curation, wrangling and cleaning

“We spend a lot of time on this learning element because getting the metric we want from the dataset often requires preparation. The way data is captured and stored doesn’t make it easy to just ask a question and get an answer. So, this learning element is something I spend a lot of time on because it’s most of what the job involves when preparing data.”

Justin Rowland

Introduction to Business Analytics at Wake Tech Community College

6. Assessing validity of data, methods, results and communication

“We discuss the validity of data in the context of assignments and projects, emphasizing critical assessment.”

Natalie Kraft

Intro to R/Python for Data Science

7. Employing design practices such as documenting work, considering whether broadband is required for applications, including color palettes that are visible to people who are color blind, adding captions to video and adding descriptive text to images

“This learning element is incorporated into assignments where students must consider accessibility in their data presentation.”

Natalie Kraft

Intro to R/Python for Data Science

8. Investigating ethical issues and ways to approach them

“Students read articles on AI biases and discuss potential biases in their projects, which is crucial for understanding the ethical implications of data science.”

Mahmoud Harding

Intro to R/Python for Data Science

Data Discoveries 

9. Articulating current issues or open questions in data science and AI

“One of my assignments that is actually new this semester—I didn’t use it the last time I taught the class—is having students do a two-minute news talk on something that has happened within the past six months related to AI. It’s only two minutes. They don’t even get out of their seat. They just develop one single slide that I integrate into my slide deck. It’s pretty easy to find something to talk about, and then we have a discussion on it. Usually, it’s not very long, but it keeps whatever the topic is fresh each day.”

Christin Phelphs

Intro to AI Ethics

10. Specifying exciting discoveries or impacts of data science and AI

“This is woven into the curriculum as students explore the real-world impact of data science in various sectors.”

Natalie Kraft

Intro to R/Python for Data Science