Student Notebook

Do No Harm (Unless It’s to a Research Assistant?)

Research assistants (RAs) across the different subfields of psychology may experience psychological, social, and physical risks when carrying out their assigned tasks (for a review, see Naufel & Beike, 2013). As someone who investigates such risks, I often am told stories pertaining to RA harm. Recently, a graduate student shared such a personal anecdote with me. In one of the studies for which she was serving as an RA, she had the responsibility of observing participants react to a graphic video featuring torture. Unlike the participants, who only saw the video once, she witnessed the torture repeatedly — sometimes for several consecutive hours, multiple times per week, and over multiple weeks. “I was having nightmares,” she said. “It was horrible.” She later followed up, “I can still see every single scene in my head.”

The experience clearly had distressed her, and although she told me her supervisor took steps to mitigate her stress, this story — and the other stories of harm that RAs share with me — sting my heart. With a little awareness and planning, much of the trauma that she and others have experienced could have been reduced or prevented altogether.

RAs in psychology can experience psychological risks when carrying out their assigned tasks. Those include vicarious traumatization or symptoms similar to those induced by post-traumatic stress disorder from repeatedly seeing stressful stimuli, transcribing personal essays, or interviewing study participants about distressing topics like sexual assault.

RAs also can experience social risk when they act as rude, incompetent, or sexually inappropriate confederates. For example, Naufel and Beike (2013) discuss an anecdote in which an RA’s responsibilities entailed watching sexually explicit media in front of female participants, some of whom were students in his other classes. The debriefing was withheld until the end of the semester, and the RA explained that he felt as if his classmates perceived him as being perverted during that time (p. 9).

Though prompt debriefings may help reduce social risk, they may not be enough to diminish it completely. When describing the distress and embarrassment that he felt as a confederate, Olianksy (1991) mentions the part of the debriefing process when he met with participants. He noted that, although participants seemed okay during the debriefing time with him, he discovered that some participants “had not taken the news of being fooled as well as they had indicated with [him]” (p. 256).

Amplifying this risk is the fact that participants are often college students in psychology courses. RAs may be peers of participants — or worse, an RA may later become a teaching assistant or instructor for the participants (Naufel & Beike, 2013). Graduate students may agree to act in videos or model in photographs used as stimuli in research. Later, when they become instructors, they may feel embarrassment when they realize students are participating in studies still using those stimuli (for a specific example, see Naufel & Beike, 2013).

RAs also face several physical risks, including the risk of being assaulted by angry participants; for a review and discussion, see Naufel & Beike, 2013. Recently, when the Georgia state legislature discussed the possibility of a new campus carry bill, my lab members and I talked about potential increased physical risks for certain research protocols. We had been discussing paradigms similar to those used when studying the southern culture of honor. In such research, a confederate runs into participants in the hall and calls them a derogatory name, an action which yielded an aggressive response in southern participants (Cohen, Nisbett, Bowdle, & Schwartz, 1996). Though we acknowledge it would probably be unlikely, we questioned the extent that study paradigms that intentionally provoke participants could put our RAs in grave danger.

After I present information about RA risks at conferences, graduate students or faculty members often approach me to share personal anecdotes. Our conversations often focus on identifying and describing the risks; rarely do I hear how they are actually addressed. After a near-decade of investigating this issue, I feel that it is unacceptable to simply continue recognizing the risks — it is time to prevent them from occurring.

Currently, institutional review boards (IRBs) are not required to have formal policies that protect RAs from the previously stated risks (Naufel & Beike, 2013). On the other hand, graduate students are in a unique position to protect RAs from harm, as they can mentor undergraduate RAs, teach, or serve as RAs themselves. Resources such as the Research Assistant Bill of Rights (Naufel & Beike, 2013) and the Table for Identifying Risks to RAs (Naufel & Le, in press; see supplemental table online) exist to help graduate students identify, prevent, or reduce risks to students and themselves.

The best way to prevent such risks to RAs is to think preemptively. When designing a study, consider the extent that a procedure could harm an RA; if such harm seems likely, seek a safer approach. The supplemental table provided online offers a questionnaire to determine if a method might pose a risk to RAs with alternative methods to curb that risk (from Naufel & Le, in press). Preemptive planning not only can curtail risks to RAs, it also can result in improved research quality. For instance, my graduate student was examining how college students reacted to either stigmatizing or nonstigmatizing comments about people with mental illness. The student first considered creating a video for this manipulation; however, we realized that the people who could act in the videos were either graduate students teaching or about to teach Introduction to Psychology courses or undergraduate RAs. This created a social risk for RAs: We worried that participants might see their instructors or peers as actually endorsing the stigmatization of mental illness. To counter this risk, we decided to use text paired with audio recordings. Participants now had less of a chance of identifying the person from voice alone, which protected the assistants who made the recording. Additionally, the use of text made it easier to standardize the message across the two conditions.

Unforeseen harm, however, can occur even with preemptive planning. Therefore, it is important for principal investigators to consider how such fallout will be handled. The Research Assistant Bill of Rights, which can be found at, is a document that outlines the protections and benefits for RAs, and it provides guidelines for how to address situations should unanticipated harms occur (Naufel & Beike, 2013). It also prompts thinking about how to handle unanticipated risks. Data collection could be slowed if an RA drops out unexpectedly for any reason. The Bill of Rights prompts supervisors to plan for conditions in which an RA may need to stop or slow the process of research.

Although graduate students are in a position to protect RAs, they also are sometimes RAs themselves, and thus similarly susceptible to risks. Encouragingly, many faculty members are beginning to take precautions to protect RAs. For instance, a colleague’s research involved having RAs code participants’ traumatic experiences; based on her personal experiences with this line of study, she recognized that RAs may develop psychological symptoms as a result of this coding. In her IRB proposal, she therefore proactively addressed how she would handle risks to both participants and RAs.

Unfortunately, not all faculty members consider reducing risks to RAs. For some, the perceived pressure to collect data may overshadow such consideration — for others, it may simply go unnoticed. A student could therefore make carefully-worded inquiries that encourage planning for possible harms. The question, “What is the protocol if a participant gets physically aggressive with me?” could prompt a mentor to think about the physical risks of a study. The question, “I teach a lot of these students, and their recognition of me may make the study less believable. How can we change this?” could give rise to a discussion about alternative methodologies that both protect the RA and strengthen the research design.

It is important to create a culture in which thinking about RA risks is the norm rather than the exception. When talking about study designs with others, include discussions of RA risks. When teaching about Milgram’s (1974) experiments in introductory psychology courses, have students consider the ethical dilemmas to both the participants and RAs. When writing IRBs, mention the steps you are taking to protect RAs even though it is not required. By repeatedly doing so, you are making the risks to RAs known — and thus making their safety a priority.

Table 1. Identifying risks to RAs and potential solutions for handling risk.

  Questions to consider:1 If yes to any of the questions, then consider doing the following:
Identifying physical risk Is there a risk that a participant could hurt the RA?

Is aggression or rudeness a dependent measure?

Is the RA violating a social norm?



Ensure a safety plan is in place

Train the RA on how to de-escalate aggressive behavior

Have another RA present or nearby

Install a panic button in the lab area, or have the RA carry a panic alarm

Have the supervisor onsite

  Is the study’s apparatus or materials dangerous?

Is the RA using dangerous chemicals?

Is it likely that the RA will encounter dangerous pathogens or biohazardous waste?

Is the RA being exposed to loud noises or pollution?

Is the apparatus dangerous?



Make sure both you and your RA are extensively trained in the proper use of chemicals or apparatuses

Ensure lab is equipped to handle emergencies (e.g., have an eye wash station or fire extinguisher)

Follow OSHA standards


Have the supervisor onsite when research is being conducted


  Is the setting safe?

Is the research being done in the field as opposed to the lab?

Is the research being conducted outside of normal business hours?

Is the research being conducted in a dangerous setting? (e.g., on a busy road, at a disaster area)


Have two RAs conduct research at
the same time

Consider alternative methods, if possible


Require RAs to take risk-related precautions (e.g., wear orange vests if near a busy road)

Identifying psychological risk Could the research evoke a stress response?
Will the RA be exposed to emotional content repeatedly?Is the research topic sensitive or unpleasant?Will the RA be interviewing, transcribing, or coding participants’ personal stories?

Have the RA take frequent breaks


Train RAs on how to cope with trauma
Limit the RA’s exposure to the stressful material.

Have a list of counseling resources premade and available to the RA

Check in frequently with RAs to see how their stress levels are

Have an alternative assignment ready

  Is deception involved? Allow the RA to take breaks from deception.

Limit the extent that an RA engages in deceptive tasks.

Check in frequently with RAs to see how the study is going

Offer an opportunity for an alternative assignment once the study begins

Identifying social risk Could participants view the RA unfavorably?

Will the RA be acting as a confederate?

Will the RA be acting in an unfavorable manner for the study’s purpose?

Will photographs, videos, or voices of RAs be used for research purposes?

Will the RA deceive participants?

Is it likely that the RA will interact with the participants outside of the lab?



Debrief participants fully and immediately

Make sure participants meet RA in a positive context


Consider having someone outside of the lab be the confederate


Could others view the RA’s participation in the study as questionable?

If the RA acted as a confederate, deceived participants, or engaged in any other questionable acts for the sake of research, will the RA’s role be acknowledged?

Would a potential employer find the RA’s role in a study questionable?




Inform RAs of the risks of acknowledgment, and allow RAs the option to decline (Naufel & Beike, 2013)


When writing letters of recommendation, exclude specific description any of the RA’s “questionable” responsibilities


1 Adopted with permission from Naufel & Le (in press); Based on the initial work of Naufel and Beike (2013)

References and Suggested Readings

Cohen, D., Nisbett, R. E., Bowdle, B. F., & Schwarz, N. (1996). Aggression and the southern culture of honor: An “experimental ethnography.” Journal of Personality and Social Psychology, 70, 945–960.

Milgram, S. (1974). Obedience to authority. New York: Harper & Row.

Naufel, K. Z., & Beike, D. R. (2013). The ethical treatment of research assistants: Are we forsaking safety for science? Journal of Research Practice, 9, Article M11. Retrieved from

Naufel, K. Z., & Le, J. (in press). Do we do no harm? Acknowledging the unacknowledged risks to the psychology student researcher. In A. Buddie & T. Pusatori (Eds.), Teaching Tips: A Compendium of Conference Presentations on Teaching. Retrieved from

Oliansky, A. (1991). A confederate’s perspective on deception. Ethics & Behavior, 4, 253–258.