At the suggestion of a visitor yesterday, I compiled some student achievement data from FSU’s Studio Physics Program to compare to typical results from traditional lecture classes, online courses and evidence-based pedagogies (like ours). The compilation is shown here:
The green bars correspond to “normalized learning gains” on the Force Concept Inventory typical for different types of classes taken from the research literature. The red bars are normalized learning gains from FSU’s studio physics classes since Spring 2015.
Now for some explanation.
First, the assessment instrument. The Force Concept Inventory (FCI) is a multiple-choice assessment first released in 1992 by researchers at Arizona State University. It measures understanding of Newton’s Laws, and it is considered the gold standard. The version we use here was released in 1995.
In 1998, Hake published a 6,000-student study using the FCI in which he compared student learning gains in traditional lecture classes and “interactive engagement pedagogy” classes (like studio physics) in terms of the normalized learning gain, which is extracted from pre- and post-testing. The normalized learning gain for each student is the difference between the post-test and pre-test scores as a percentage of the number of questions the student missed on the pre-test.
Hake found that in traditional lecture classes, average normalized learning gains were 23% +- 4%. That is, the average student in a traditional lecture class learned 23% of what she or he didn’t know at the beginning of the semester. In interactive engagement classes, the normalized learning gain was 48% +- 14% – about double the gain in traditional lecture classes (although with a significantly larger spread in results).
In 2014, researchers at MIT published their study of an introductory physics MOOC and reported a normalized learning gain of 31% +- 2%. That result probably gives the bottom line for a high quality online course – somewhat better than a traditional lecture class but still far short of an interactive engagement class.
All three of those results – traditional and interactive engagement results from Hake and the MOOC result from MIT – are shown in the figure.
Since our studio physics classes are taught in the SCALE-UP interactive engagement format, we’d expect to match Hake’s IE result. And we do consistently. In fact, we’ve been doing so since 2008 when our early efforts were cited as a success story in SCALE-UP creator Bob Beichner’s commissioned paper for the National Academy of Sciences.
It’s important to say this every time it comes up: Our success is due in large part to the tremendous support we’ve gotten from leadership in our Physics Department and at higher levels.
If FSU wants to improve the physics learning of students from all backgrounds and open the doors of STEM opportunity wider to those students, the university will expand the studio physics program. If the university does not expand the studio physics program, then it will be a clear signal that improving undergraduate science teaching is not a priority of our institution.
It’s that simple.