I was so busy blogging about everybody else’s presentations that I haven’t had a chance to write about my own talk at AAPT! I’ve been working madly for the past few months to pull together a monstrosity of data on the outcomes and lessons learned from our work in the Science Education Initiative at the University of Colorado. In my poster and talk I outlined the current status of that project (in collaboration with my wonderful colleagues Danny Caballero and Carl Wieman). These results are very preliminary, and will be worked into a series of papers – including a special issue of Physical Review this year. Please ask me any questions about this work that you think of – I’ll be happy to answer them, and they will help inform me about things I should emphasize in this work.
Our model of institutional change
The SEI was a $5 million dollar, university funded program aimed at improving undergraduate STEM instruction at the University of Colorado (and, later, its’ sister program at the University of British Columbia). The main emphasis of the program was the department as the unit of change – resources and power were localized in the departments, by way of a competitive grant program. Departments had to submit proposals to the SEI, indicating what they would do with the funds and what courses would be transformed. Several departments at CU were funded, and most of the money went towards hiring Science Teaching Fellows (STFs) with a PhD in the discipline, and an interest in education. I was the first STF in Physics, though I’m now the Associate Director of the program, having stuck around long enough that I’m doing more top-level type of stuff.
The aim of the SEI was twofold – one, to transform STEM courses using a backwards design model, creating materials and learning goals that instructors could use moving forward. The second was really to serve as an experiment in institutional transformation.
I’m still analyzing the data for more nuance, but in broad-brush strokes, I can tell you that this model – focusing on the departmental level, and providing postdoctoral support – did work. Some of the main results:
- 135 faculty across all departments have modified their teaching in some way, representing about 50% of teaching faculty
- 92 courses were impacted in some way
- When considering only those courses with which the postdocs had a deep involvement, it cost about $145K per course
- About 20,000 students enroll per year in courses that have been affected by the SEI in some way (though those changes aren’t necessarily sustained)
What factors affected success in a department?
I’m still teasing this out, but some early indicators:
- Who is the STF? Do they work well with faculty?
- Who is the department chair? Are they supportive of the SEI? Do they assign faculty to SEI-transformed courses who are likely to use the changes, at least for a few semesters?
- How much do faculty rotate among courses? In physics, our faculty only teach a course once or twice before rotating out – so focusing on a particular course is likely to impact many faculty, but be difficult to sustain.
- Institutional incentives for teaching are still lacking, and are a barrier to instructional change
- Lack of departmental ownership and accountability for the changes that they say that they’ll make (in the proposal) is a problem. “We can’t tell our faculty how to teach,” they would often say. One solution to this is to offer incentives to the department that will be removed if they don’t follow through, or to have higher-level administrators indicate that they do not look favorably upon departments that do not carry out their commitments.
- While it seemed natural to start with creating learning goals for large introductory courses, this may not always be the best approach. Creating learning goals is a long, arduous process, and it is difficult to get consensus. It also doesn’t seem to be motivating to faculty. Additionally, large introductory courses – while impacting many students – are difficult to change due to multiple sections and instructors. Thus, it may be better to start with small, noticeable changes (as in to a single lab), and use those to motivate faculty to undertake bigger changes.
- SEI administrative oversight is important, in order to create clear expectations of the STFs, provide high-quality sustained training of the STFs, and provide oversight and consequences for departments to make sure that they follow through.
Again, please let me know if you have any questions or comments!
View my slides below.