In recent years, psychology, like many areas of science, has faced a credibility problem, as scientists and the public have realized that many common research practices can lead to a literature that includes questionable or nonreproducible findings. What has come out of these discussions are a set of practices—including preregistering studies and sharing one’s materials and data with other scientists—designed to make research more transparent and replicable (see "Ways to embrace transparency").
These open science practices may be new, but for many students, they seem intuitively appealing. “My undergraduates believe research transparency is the way science has always been done,” says Morton Ann Gernsbacher, PhD, a professor of psychology at the University of Wisconsin–Madison, who gave a workshop on teaching research transparency at APA 2019 in Chicago. It’s important to reinforce those beliefs early, she says, because for open science practices to become mainstream, they need to be taught to the next generation of psychologists and other scientists. Psychology faculty share their advice on how to integrate open science into teaching and mentoring.
1. Start early
It’s important to talk about transparency to students at every level, even to introductory psychology students, most of whom will not become scientists, says Julia Strand, PhD, an assistant professor of psychology at Carleton College who teaches a course called “Psychology’s Credibility Revolution.” Even nonscientists “are going to spend the rest of their lives evaluating scientific claims,” Strand says. “Having a sense of how scientists collect their data and the flexibility that they have in reporting enables students to be more critical consumers of scientific information.”
2. Assign readings on replication
A simple way to introduce open science practices is to read papers related to replication issues with your students as part of a seminar course or ongoing lab meetings. Strand suggests incorporating review papers (e.g., Simmons, J.P., Psychological Science, Vol. 22, No. 11, 2011opens in new window and Flake, J.K., PsyArXiv, 2019opens in new window) at first rather than research papers, which may require familiarity with specific methods and analytic techniques. Once students become familiar with transparency topics, you can point out places in the research process where bias could enter and discuss how to combat these biases.
3. Lead by example
For more advanced undergraduate research assistants and graduate students, work together to preregister a study or submit a registered report. In addition to engaging students with the mechanisms of transparent research, another benefit to preregistering studies—which involves writing the introduction and methods section before collecting data—is that it relieves the burden of writing those sections later. Meanwhile, registering reports—which involves submitting for peer review a research plan to a journal before collecting the data—does even more to facilitate students’ progress and careers because as soon as the research plan has been accepted, students can list it on their resumes and graduate school applications. The Center for Open Science maintains a growing list of more than 200 journals that accept registered reports.
4. Teach students how to share data
“Of all the steps toward transparency that researchers can take, sharing data is one of the most beneficial, but also one of the most fear inducing,” Gernsbacher says. To overcome the anxiety she feels about sharing her own data, Gernsbacher asks colleagues to check it for errors before making it public—and does the same for them. She eases students into this practice by requiring them to share their data—collected as part of a graduate student thesis or undergraduate class-based project—with other students, who in turn have to analyze the data and attempt to draw the same conclusions as the student who shared the data with them. Gernsbacher also recommends helping students apply Creative Commons licenses to their research materials, such as visual stimuli and computer codes. Creative Commons licensing allows researchers, including students, to specify whether, where and when their materials can be copied, distributed or altered.
5. Plug students into the network
There are now multiple ways to plug into open science communities online, and Twitter is one of the most effective, according to Strand. “There is active and spirited discussion about open science on Twitter,” she says. “We have found it to be very effective for staying up to date with issues and advancements, discovering new papers, getting rapid answers to questions and networking.” Research transparency is also the subject of numerous blogs, such as the one hosted by the Berkeley Initiative for Transparency in the Social Sciencesopens in new window, and podcasts, such as The Black Goat and ReproducibiliTeaopens in new window. In addition, attending the Society for the Improvement of Psychological Science’s annual meeting and related meetups at other conferences can be an important source of information and network building around research transparency.
6. Teach a transparency course
In addition to working open science principles into existing psychology coursework, some schools are also developing specific classes on the topic. In spring 2019, Benjamin Le, PhD, a professor of psychology at Haverford College, taught the inaugural session of his course “Open Science and Inclusive Psychology,” one of the first of its kind for undergraduates. Le and his students tackled case studies of debunked findings, explored p-hacking (trying lots of analyses until you get a significant effect) and other tricks that game the research process and discussed the influence that pressure to publish has on research outcomes. “This was the most fulfilling course I’ve ever taught,” Le says. “Seeing students’ eyes light up when realizing studies in the intro psych canon may be flawed, and helping them understand how they can make science better, was really rewarding.”
7. Pave the way
After you’ve shown students the path to open science practices, help them along it. In a 2018 Trends in Cognitive Sciences articleopens in new window, Gernsbacher suggested emphasizing research transparency when you’re on search committees that are drafting job announcements or evaluating candidates. She also advised telling graduate students when they’re on the tenure-track job market to highlight their research transparency bona fides in their cover letters and annotate their vitae to indicate which of their studies were preregistered or used open access materials and data. Then, encourage your institution to reward transparency during academic promotion and tenure evaluation. When researchers are rewarded for these practices, their habits will trickle down to trainees, Gernsbacher says.
8. Overcome your own hang-ups
If you’re having a hard time with transparency in your own research, do some selfintrospection, says Strand. A concern that psychologists often cite for not getting on board with transparent research practices is the perception that they take too much time. “In my experience, that’s just not true,” Strand says. “They just change where you invest the time. Rather than figuring out how I’m going to analyze my data after I collect it, I just make the plan before.”
Another common worry is the fear of making mistakes and being “found out.” “That’s a valid fear that you can pretty easily get around by asking colleagues to proof your materials before making them public,” Gernsbacher says. Strand agrees and offers a lesson that she thinks all budding psychologists should hear. “As scientists, we should not be in the business of hiding when we’re wrong,” she says. “We should be looking for ways to be less wrong, but if we are wrong, it should be out there so that other people can find it. That helps make science self-correcting.”
