I’ve blogged previously about one of my favorite courses at Allegheny — Investigative Approaches in Biology (FSBio 201), a course in which students rotate through multiple research modules and conduct research projects in small groups. The course emphasizes hypothesis testing, experimental design, and publicly presenting results. I think this type of course is intriguing and has tremendous potential for teaching undergrads the foundations of good research… but it also has its limitations. In this particular post, I’d like to describe my module (which I’ll be teaching for the first time Spring 2014) in a bit more detail, where I see it succeeding, and have a discussion about what I might do to minimize the shortcomings. This class meets 5 hours/ week (2 hrs class 1; 2 hrs class 2).
My Module: The Ecology of Infectious Diseases
This topic shouldn’t be terribly surprising given my area of expertise. My module will present students opportunities to examine the context-dependent nature of host-parasite interactions. The module starts off with an intro to hypothesis testing, how to find primary scientific literature, and tips on how to read a scientific paper (i.e., interpreting figures). After presenting them with a brief (2 class periods) on host-parasite ecology, which will end with a brief introduction to my study system (amphibian-chytrid fungus). We then conduct a staged experiment (in my module, I am looking at the relationship between pathogen burden and some behavioral metrics) From here, they will work in groups of 4 to develop a research project in which they can manipulate any number of biotic or abiotic conditions and test how those influence host resistance or pathogen transmission. They will spend the next class setting up and starting their experiment and the next couple of weeks will be a combination of data collection (e.g., learning DNA extractions and qPCR techniques) and small group activities (e.g., preparing PowerPoint presentations, discussing scientific papers). They’ll finish up the module by collect their final data, prepare a 12 minute group presentation, and move along to their next module (they turn in their “lab report” 1 week after the module ends). I’ll get a new crop of students and start fresh.
Why I Like This Course
There are many, but I’ll highlight a couple.
First, this course is necessary for our curriculum because ours is very research oriented. Students are required to take Junior and Senior research courses and are expected to hit the ground running when they take these courses. We would not meet our educational goals in the upper-level courses if we did not provide them with a hands on experience in conducting independent research and developing hypotheses prior to those courses.
This course sidesteps the structure in which we ask students to learn by exclusively lecturing to them. A good module walks a fine line — it provides enough information to students so that they can proceed in asking their own questions without getting lost but allows them to take risks, fail, and learn. This latter part is often lost in traditional lecture style courses.
It just so happens that they also learn that scientists take risks, fail, and learn in the process. I often remind students in lecture that concept X, Y, or Z that is summarized nicely in a color figure was the product of many years of questions and sometimes failures. I think that FSBio provides students an opportunity to do this and it opens their eyes to the fact that bar graphs in their textbooks are the product of real experiments.
This course is far from perfect. I foresee a couple of problems in the way the course is currently designed and I’m anxious to have my first stab at teaching my module. I’m also excited about the vision that I have for my module and some of the ways in which the course might change in the future.
First and foremost, we only have 1/2 of a semester to work. Learning how to design, start, and complete an experiment usually takes graduate students an entire semester… we’re trying to do the same with Sophomores (sometimes Juniors). This leaves us with little time to really make it through all of the course outcomes.
Like all scientists, we are financially limited. Research is expensive and these limitations set boundaries in what we can do. For example, I calculate that it costs approximately $6 for me to run each qPCR sample. Each module has 4 groups of 4 students and I run 2 modules per semester, so I have 8 groups of students working with tadpoles. I estimate that each group of students will use approximately 25 tadpoles for a “bare bones” experiment (see below), which will cost ~$1200 in qPCR. I also will have to run samples for the 1st staged experiment, so I’m around the $1500 mark in qPCR. Consider that we generally offer 4-6 different sections of this course each semester… so these costs can grow pretty quickly. I’ve considered switching to another host-parasite system (e.g., Daphnia-fungus), but I lose the molecular component to my module, which is one of the things that the faculty like about my module — it is integrative.
This often results in poor replication which influences whether/if we can even discuss statistics. My plan is to pseudoreplicate data from two of the more successful experiments in the module and introduce students to very basic statistical analyses with those data (e.g., ANOVA, MANOVA, linear regression) so that they can at least see the connection between experimental design and statistical analysis.
I think this course has amazing potential. In my opinion, it goes above and beyond traditional canned laboratory activities and gives students an early experience with research. I’m hoping that some of our readers have experience in a course like this (either taking one themselves or teaching one) and might chime in with some suggestions. For example, are the benefits from switching to a Daphnia-fungus system (less cost and increased replication) worth losing my molecular component? Are there any better ways I can introduce students to basic statistics such that they can really connect experimental design –> data –> analysis?