The Open Case Studies project is an educational resource of
experiential guides that demonstrate how to effectively derive knowledge
from data in real-world challenges.
Our case studies cover many concepts! Statistical
topics: from correlation to the influence of multicollinearity
on linear regression Data science topics: from
making simple plots to building dashboards and performing machine
learning (ML), a type of artificial intelligence (AI)!
What problem are we addressing?
Despite unprecedented and growing interest in data science on
campuses, there are few courses and course materials that provide
meaningful opportunity for students to learn about real-world
challenges. Most courses frequently fail to frame the lectures around a
real-world application and provide unrealistically clean datasets that
fit the assumptions of the methods in an unrealistic way. The result is
that students are left unable to effectively analyze data and solve
real-world challenges outside of the classroom.
Problems with previously suggested solutions
In 1999, Nolan and Speed
argued the solution was to teach courses through in-depth case studies
derived from interesting problems, with nontrivial solutions that leave
room for different analyses. This innovative framework teaches the
student to make important connections between the scientific question,
data and statistical concepts that only come from hands-on experience
analyzing data. However, these case studies based on realistic
challenges, not toy examples, are scarce.
What are we proposing as a solution?
To address this, we are developing the Open Case
Studies educational resource of case studies, which demonstrate
illustrative data analyses that can be used in the classroom to teach
students how to effectively derive knowledge from data. This approach
has successfully been used to teach
data science courses at many universities, including: