Working Alone: A Challenge to Responsible Research Conduct
By Marianne Evola
"I was alone in my tastefully furnished office at the University of Groningen. I'd taken extra care when closing the door, and made my desk extra tidy. Everything had to be neat and orderly. No mess. I opened the file with the data that I had entered and changed an unexpected 2 into a 4; then, a little further along, I changed a 3 into a 5. It didn't feel right. I looked around nervously" (Stapel, 2012).
This excerpt comes from Nicholas Brown's English translation of Diederick Stapel's autobiography "Ontsporing" or "Derailment," which addresses his research misconduct. Brown's translation is freely available online. For those unfamiliar with the case, Diederick Stapel is a social psychologist and is one of the most notorious case studies of research misconduct to date. Recently, "Retraction Watch" announced the 55th retraction of Stapel's work due to misconduct. The above excerpt describes the moment when Stapel's questionable research conduct, or "grey" practices, as Stapel describes them, progressed to outright fabrication of research. It is an incomparable example on how working alone can impact professional decision making and responsible research.
"No one will know." When I present to students on responsible research conduct, I tell them to seriously question any decisions that prompt the notion: "No one will know." I warn students to recognize this as a big red flag regarding the ethics of any decision. However, working alone often predicates this thought, because only when you are alone are you truly in a situation where you can make bad decisions and have no one know. Which brings us to the ethical challenges associated with working alone.
Responsible decision making for daily tasks is much easier in a bustling lab than it is when you are working alone, late at night, when your lab mates are gone from the lab. When I was working on my dissertation, I faced a daily challenge related to a research protocol that demanded continuous cleaning of heavy equipment. I worked with rats in behavioral tasks and my equipment would track the movement of a rat while it was contained in a behavioral compartment. However, rats have a phenomenal olfactory system and can efficiently follow the scent trail of another rat. Thus, my equipment, had to be cleaned every time I placed a new rat in the behavioral compartment or the scent trail from the previous rat would impact the behavior of subsequent rats. So, I spent much of my day with a spray bottle and paper towels in hand, cleaning and moving equipment. The experiments were physically demanding because of the constant cleaning, sufficiently so that one graduate student (and a few undergrads) opted out of participating in these experiments because they were just too laborious. Each day, it was very easy to clean the equipment at the start of the day, when the lab was alive and busy, but as the hours passed, it became more and more exhausting to clean the equipment. Furthermore, it became more and more difficult to make the responsible decision to clean, especially as my experimental sessions progressed into the late night hours and the lab was largely deserted, except me and my rats.
Luckily, I had a defense for poor decision making built into my experimental design. Drug treatments and rat placement in the compartments were all randomized. I had to follow my daily step-by-step research plan because of the complexity of the design. It would have been too damaging to my dissertation data if I neglected to clean the equipment. In other words, even though I was the designer, I was running so many experiments with so much randomization, I was largely blind to experimental treatments. Thus, failing to follow protocol would only have contributed variability to the data, something that I was working hard to minimize. Oddly, I did not design the experiments in this way to protect myself from cutting corners, but rather the randomization was the appropriate research design. Regardless, there were many late nights that I was thankful for the design complexity and how it protected me from making bad decisions.
It is the seemingly small daily decisions in the lab that are most at risk for negligence. Especially for inexperienced students that are new to the lab and do not fully understand the procedural rational, laboratory pressures or the need for experimental consistency. Many processes can seem persnickety to new students until they realize their pertinence. Our experimental protocols used to require the frequent transport of animals. One of the things that I was very picky about was how animal cages were loaded on a cart for transport. White rats, generally look quite similar to one another and when given the opportunity, they love to hop into one another's cages to visit, which often would happen during transport. Thus, students were required to keep rats marked for identification and to load transport carts in a very precise order so that the first through last animal in a group were always placed on the same position on the transport cart. The reason for this was to minimize the likelihood that animals would be mixed-up during transport. In other words, if the cages were always placed in the same position on the cart, then rat #1 would be in position #1 and then placed in experimental chamber #1. There was little room for error once students learned the motor tasks of transporting animals. Invariably, new students would always find these requirement nitpicky, until they mixed up the rats and potentially damaged an ongoing research project.
Over the years, I learned to keep an eye on new students until they realized the importance of these small tasks. I also became very adept at sorting out a seemingly unidentifiable cluster of rats. That being said, the student's rational for violating protocol was always the same, they were running late and were rushing to get their experiment started. They did not think it was critical to check the ID markers on their rats or fuss over how they loaded the transport cart. The students had permitted time pressures to negatively influence their responsible research decisions.
Undergraduate students that take the time to engage in research can be very susceptible to pressure. Often the students who seek out research opportunities are generally overachievers, often honors students. As a population, overachievers often put tremendous pressure on themselves to excel, to be highly productive and even to be perfect. Thus, even a small dose of external pressure can be hugely magnified internally by these students. In their efforts to be productive, and please their supervisor, they can neglect precision as they learn to balance these competing demands. As a mentor or supervisor, it is critical to provide students with plenty of supervision so that it is easy for them to make good research decisions while under pressure. Furthermore, supervision does not always have to be labor intensive for the supervisor. Creating an interactive team that is empowered to make intellectual contributions and respectful critiques to all of the research projects in a lab can markedly enhance overall supervision of the research while encouraging students to work as a team to efficiently advance all of the projects in the lab. The accountability associated with many eyes on your project enhances efficiency and makes students more comfortable in the lab. And if all students are encouraged to keep a critical eye on all projects, and provide constructive input on all projects, then it minimizes any negative perceptions of that input. Finally, regular respectful critique of your work encourages students to develop a "thick skin" and a healthy regard for critical input.
Creating a system of accountability with a team of students is surprisingly easy to do. From my experience, new students prefer an interactive environment and do not like to be alone in the lab. Just a few days ago I was having lunch with an undergraduate student. We were discussing her previous research experience and why her interests transitioned away from research toward the applied sciences. Like many students, although she found the theory interesting, she was not well suited for the monotony that accompanies good research practices. But one other issue she mentioned was how much she disliked working alone in the lab. Her first undergraduate research position required her to spend a great deal of time working alone and she did not like the isolation. It made her feel vulnerable because there was seldom anyone around with whom she could clarify research protocol. As a typical high achieving student, her vulnerability stemmed from a fear of making errors. When she had sought out research experience, she had imagined that the lab would have more camaraderie and less isolation.
Years ago, I realized how little students like being alone in the lab. We had accepted a new and exceptional student into the lab as a research assistant during a transitional time when we had a pretty small research group. Historically, I had always supervised a large team of undergraduates with my desk situated in the middle of the laboratory. My constant proximity to the research team resulted in great working relationship with all of the students. However, admittedly, the hustle, bustle and interruptions made it quite difficult to concentrate. Then, one day, a surprising thing happened, a faculty office opened up and I was asked to temporarily move into that office so that the space looked occupied to those looking for empty offices. Although I would eventually be bumped by a new faculty member, I could not believe my luck. I would finally have a quiet place to concentrate. What I did not realize was how difficult it would be to supervise and build a mentoring relationship with a new student while housed behind a closed door. Furthermore, I was amazed that even familiar students did not seek me out to answer questions if they had to cross the hall and knock on my door and the new student never sought me out. I quickly realized that I needed to maintain a strong presence in the lab to make sure that I remained approachable to the team and actively moved much of my stuff back into the lab. Interestingly, later, that new student informed me that she had been contemplating quitting the project just before I moved back into the lab. The lack of supervision had made her feel uncomfortable.
Although I am convinced that students would like a mentor's constant presence in the lab, I'm also aware that is not generally an option. Many research teams are much smaller than ours used to be. Often, the research team is in fact, one mentor and one student. And, the mentor is a faculty member with many more obligations than time. Thus, maintaining a constant presence in the lab is impossible. If that is the case, it is important to utilize other strategies to keep students focused on efficiency and responsibility in the presence of pressure. Regularly ask to see students' raw data as well as their summarized results. It should be common practice to request the raw data of all collaborators. In a wonderful discussion of the importance of reviewing the data of students or collaborators, CK Gunsalus (2015) asserts that you are not doing your due diligence if you fail to request the data of collaborators . It should be standard practice to take the time to review student's data and to regularly review the importance of following all research procedures. Furthermore, if it is standard practice, reviewing data will never be perceived as invasive or offensive by the team. Another suggestion to keep students focused on procedure is to consider research designs that demand focus on procedure, such as keeping students blind to expected research outcomes. If students do not know their mentor's theoretical expectations, they cannot feel pressured to produce a particular research outcome. Their success can only be measured by their ability to efficiently follow research protocol.
Now, to be clear, I do not think that a student working in the lab is likely to engage in outright research fabrication synonymous with that of Diederick Stapel. His misconduct is the product of years of inappropriate and progressively irresponsible and deviant research decisions. Rather, I feel that students are vulnerable to making small irresponsible research decisions. However, each time a student purposely violates a research protocol it increases the likelihood that he will do so again when under pressure. Furthermore, with each misstep, there is a possibility that future missteps will escalate, especially if there is no discovery or negative feedback regarding the violation of procedure. Do these violations make a student researcher unethical? Perhaps or perhaps not, but it definitely contributes to sloppy research.
Marianne Evola is senior administrator in the Responsible Research area of the Office of Research & Innovation. She is a monthly contributor to Scholarly Messenger.