A lot has changed recently. While individuals practice physical distancing to limit the spread of COVID-19, many in-person activities have been suspended. To maintain a research program during a time of physical distancing, many psychologists are transitioning all of their data collection online. For those who have never conducted a study online, this can be a daunting prospect.
Before diving into the technology to determine which tools are best suited for your study and situation, take a moment to step back, look at the bigger picture and consider the implications of transitioning from a real-world operation of recruiting participants and conducting a study to a virtual one.
Here are five tips for making the switch to an online research platform.
1. Update the IRB protocol
Before you begin the process of collecting data, ensure that the IRB protocol includes any new procedures that will need to be implemented as a result of moving to an online environment. For example, if recruitment methods or the informed consent process have changed, these should be reflected in the IRB protocol.
Franki Kung, Assistant Professor of Industrial-Organizational Psychology at Purdue University, also recommends taking extra care in re-developing debriefing procedures, with the safety of participants in mind. Kung said, "…I'd say if you are studying sensitive topics or vulnerable populations, I'd be extra careful in updating the debriefing procedures to make sure that it's adequate, as compared to in-person debriefing which usually gives you more insights into how the participants reacted to the study."
2. Include checks to improve data quality
Moving studies online involves giving up some control over experimental conditions, leading to concerns about data quality. However, there are steps you can take to mitigate data quality concerns:
- Recruit more participants than you think you will need in the event that some need to be excluded from your analysis.
- Before looking at the data, outline any exclusion criteria. What in a participant's data will lead you to exclude that participant from analysis? Common criteria include failing attention checks, taking an abnormal amount of time to complete the study (i.e. at least three standard deviations from the mean), and nonsensical or plagiarized responses to open-ended questions.
Kung recently learned as a result of pilot testing to consider how participants interpret questions in light of the pandemic. He therefore recommends being clear in study instructions if you're asking about their behavior before or during the pandemic, when applicable.
Kung also suggests including pandemic-related questions in all research studies being conducted right now. "Even if the pandemic is not the focus of a research project, questions about the impact of COVID on participants—whether and how participants' job/family/wellbeing is affected etc.—is important to include," he said. "At the very least [the questions] can serve as an informative covariate/exploratory variable when testing your major research question. Better yet, they may be useful if later your research evolves and find the questions useful (e.g., a longitudinal study on COVID recovery)."
3. Remember that online participants are people too
When study participants are reduced to a list of usernames, it can be easy to forget that there are human beings on the other side of the screen. Some steps you can take to keep the process professional and engaging include:
- Make it easy for participants to generate high-quality data.
- Keep studies brief to boost engagement.
- Ask a friend or colleague to pilot test a study to minimize any technological problems.
- Keep a close eye on your inbox after first launching a study in the event that there is a technical issue. If there is a problem, a participant will typically notify you and you can fix the problem before it impacts all of the responses.
- If you choose to compensate with money, do so fairly.
- Estimate payment based on the maximum time you expect a participant to take, rather than the average time.
- Pay participants promptly (i.e. within a week).
- Just as with in-person studies, pay all participants who complete the study in good faith, even if not all of their data is useable.
4. Be creative
Just like psychologists have always flexed their creativity in lab studies, the same can be done in an online environment.
- Avoid using common paradigms with which participants may have practice.
- Many online tools can be utilized for psychological research, even if they were not specifically designed for psychological research.
5. Don't be afraid to ask for help
Shifting your research to an online environment for the first time can be intimidating, but support is readily available to help with all parts of the process.
- Others may have had a similar problem and posted a solution online.
- The support pages for a specific tool may have guides on using it.
- Many online tools have support teams available to answer questions.
- If the solution to a problem is not online, you can ask your question on a crowdsourced help website, such as Stack Overflow or Stack Exchange.
Maria Garay, a Ph.D. student at Tufts University, often finds support from social media. "Twitter was my best friend when it came time to find the resources that I needed for my studies," she said. "For example, I needed to create stimuli for a study and the software I used was only accessible on our lab computers. I sent out a tweet asking if anyone knew of a free online version of that software and got an answer almost immediately. Any time I need help in finding a resource, I always turn to Twitter!"
Ready to get started? Check out this overview of the technological tools available and some of the pros and cons of each.
Barbara Toizer is a PhD student at the University of Kansas.