Creating opportunities for the unexpected (without the chaos) in self-design labs.

Among all the challenges we face in transforming large introductory science courses, one of the biggest is managing the laboratory. This is particularly true in settings where labs are taught by graduate teaching fellows with varying degrees of teaching experience. Often pressured to focus their energies on research and writing (for example Educating integrated professionals: Theory and practice on preparation for the professoriate), there is little incentive for graduate students to be innovative in their teaching. This is enforced by an emphasis on providing standardized experiences for students across all sections.

Introductory laboratory.

As is often the case, my attempts in making laboratories more “authentic”  early in my journey were limited to having TAs guide discussions of “what hypothesis are we testing,” “what variables are being controlled for,” or “what do you predict will happen.” Not that these are bad scaffolding, but if the lab is entirely planned and the student has read the material and attended to content, then there is a single best answer for these questions, and a correct outcome for the lab exercise.

In other words, these labs cannot be anything other than demonstrations of known phenomena. But if there is no room for unexpected results, can the exercise be authentic inquiry? Can it even be educational if the only variation comes through doing something “wrong” and not getting the “right” answer? I think it is important to recognize that such activities reinforce the misconception that scientists only do experiments when they know the outcome (which even some K12 science teachers believe e.g. Antink-Meyer & Meyer, 2016). With my growing conviction of the need for students to experience and cope with meaningful “failure,” creating learner-centered lab activities that allowed variation became very important. By meaningful failure, I mean unexpected results that can be used to further student understanding of underlying phenomena. The osmosis example below is an example of this. To counter the discomfort many students expressed around science, I  sought to design labs that would engage these students and help them learn that success meant learning, not getting the “right answer.” In terms of my three goals, my goal in this case was to engage students in authentic practice so that they would understand that they could “do” science.

Classroom chaos.

First, I do want to discuss a common misconception of  open inquiry laboratories. We tend to assume that any inquiry – based lab requires students to develop their own hypothesis, identify or design an appropriate protocol, and collect, analyze, and interpret the data. Even with guidance, the end result  can be the chaos of 20 or more distinct projects, frazzled laboratory coordinators and managers, frustrated students, and graduate teaching assistants who are at a loss on how to mentor research – particularly if they are newly admitted graduate students. What I want to emphasize is this is not necessary.

The myth busting of this view of “student-design” is in our own practice. How often in actual research do we develop a completely novel protocol? Most of the time, we adapt protocols from those published in the field. What makes research “inquiry” is not a novel protocol, but that we chose to manipulate and measure particular variables to test an idea, and we do not know for certain what the outcome will be.

If the protocol is a “fixed” element of the laboratory, then the number of permutations is greatly reduced. Yet the laboratory is still authentic inquiry: students are unlikely to know what the outcomes will be no matter how well they understand the content. Importantly, not one of the laboratory activities I adapted was new to the course. The lab manager had all the equipment on hand and the basic protocols existed. What was different is that I created opportunities for students decide how to implement the protocols.

Key to this was a philosophy of less-is-more. I reduced the number of labs to seven: three introductory labs focused on foundational processes and skills, then four self-design labs. Each self-design lab took three sessions. The first lab session, we would discuss the underlying concepts, brainstorm protocols with a focus on what was being manipulated and what would need to be controlled for. The second lab session, students would collect their data. The third lab session we would talk about their results and possible interpretations.

So how do these labs work? Here are two examples, osmosis from the “foundations” set and antibiotic resistance from the “self-design” set.

dialysis apparatus
The apparatus for observing osmosis.

Osmosis demonstration. This lab was originally designed as a classic demonstration lab, and I will admit that it was pure serendipity that led to unexpected results. The students constructed their apparatuses as illustrated.  Since time and space were limited, I instructed the students to work in groups of 4, with one pair setting up the low concentration demo, and the other the high concentration demo, and compare results. Of course, we all expected osmosis to happen faster with the higher concentration sucrose solution. But most groups did not find that to be true. When we started unpacking this, I found that despite my instructions, haphazard variation in set-up had produced variation in the results. For instance, in one group the the pouch of low-concentration sucrose was much larger than the pouch of high-concentration sucrose. In another group, the pair setting up the low-concentration demo had used warm tap water while their partners had used cold tap water. [the full lab as it is now run can be found here]

Three things happened because of this “failed” lab. First, as we unpacked the sources of variation, students realized that they had learned things despite the “failures.” Second, these students understood at a personal level the importance of identifying and controlling for confounding variables. I did not have to constantly remind this group of the need to identify and control for variables during their self-design labs. Lastly, when we moved on to self-design labs, these students had greater self-efficacy in the lab than any previous group I’d worked with. They’d had a “failure” and turned it into a learning experience, and no longer feared failures.

Antibiotic test dots

Antibiotic resistance. This self-design lab involves planned sampling of surfaces around campus, culturing a single bacterial colony from each sample, and testing the cultures against one to four antibiotics. The basic protocol is provided: students are given a full flow-chart of the process. What they do is make choose sampling locations based upon their understanding of what drives evolution of antibiotic resistance and their conception of where bacteria exposed to different amounts of antibiotics might be found. Working in pairs, they sample two surfaces, manage the colony selection and sampling, choose which antibiotics to use (from six available), and analyze and discuss their results. Examples of comparisons include the hands of students with different histories of childhood antibiotic use, men’s versus women’s restrooms, cracked versus uncracked cell phone cases. The results were often surprising – no difference in antibiotic resistance between men’s and women’s restrooms – and sometimes not – the student with a history of antibiotic use had bacteria resistant to a wider range of antibiotics. In each case, they were required to unpack their predictions in terms of antibiotic use and bacterial evolution.

In these labs, students are engaged in authentic inquiry: although they are not designing full investigations, they are making choices that have significant impact on the outcomes. The labs are no longer simply cook-book recipes to confirm phenomena, but actual investigation into the biological world. And while some students find this very unsettling, most rise to the occasion and become more deeply engaged than I was accustomed to seeing in standard lab activities. The spill over was obvious in classroom settings, where they became very adept at picking out confounding variables, unclear descriptions of protocols, and other flaws in media reports. In other words, for many, their experiences in the lab lead them to more critical consumption of science reports, as well as increased confidence in science as a human activity.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s