Ethics: Hints and Tips for Testing and Research

Copyright 1995 Richard Halstead-Nussloch, Ph.D.
All Rights Reserved

Example Informed Consent Form

The purpose of the study has been described to me so that I understand it. I voluntarily agree to participate in the study. I realize that I may withdraw at anytime without penalty, merely by informing the researcher of my desire to do so.

Signature ______________________________________________ Date________

Print name ______________________________________________

Witness ________________________________________________ Date________

Ethical Principles of Research and Testing Involving Human Participants

Informed consent:

Researchers must inform human participants of the research procedures they will experience, ensure they understand what they will experience, ensure they understand the actions the researcher is taking to protect them, ensure they understand the voluntary nature of their participation throughout the study, and obtain written confirmation of their consent to participate.

Protection from harm:

Researchers must take steps to protect human participants from harm during the research study. This includes physical harm and psychological harm, e.g., ridicule and stress, as well. Risks of physical harm can be minimized by taking conventional precautions in safety, health, etc. Risks of psychological harm can be minimized by fully informing the participant of the study, and then adopting the scenario of the participant serving as an evaluator on your research team. The simplest statement of this scenario is the product is being evaluated, not the user. It is strongly recommended that researchers test the precautions they take with colleagues before the actual study. For example, run through the study with colleagues or family and ask about safety, stress, ridicule, etc.

Protection of privacy:

Researchers must take steps to protect the privacy of human participants. This includes ensuring they serve anonymously, that is, their responses and data cannot be linked back to specific individuals, and that their raw data is kept in confidence, that is, only collated or scrubbed data are publicly reported.

Freedom to withdraw without penalty:

Researchers must take steps to maintain the voluntary nature of each participant's involvement in the research. This can be accomplished by always providing the opportunity to ask questions, make suggestions to change processes or take a break, and ultimately to withdraw from the research study. Any of these actions can be done without penalty to the participant.

Debriefing Hints and Tips

In computer-science research using human participants, the debriefing after the study is often the most "information rich" part of the study. It is during the debriefing that researchers can obtain the full story of reactions, thoughts, and what really came to the evaluator's mind during the study.

Questions

Questions are a critical component of debriefing. But, they must be well-crafted in real time in order to work properly. Here are some Dos and Don'ts as tips for debriefings, which focus on questions during debriefing.

Don'ts
- Don't ask leading questions. For example, "Don't you agree that entering the product code is difficult?" "I've always thought this application was poor--Don't you agree?" Leading questions are actually statements of the researcher's bias disguised as questions. They must be avoided in research.
- Don't ask questions defensively. For example, "We designed the application to work so that you choose an action and then an object. You obviously didn't see that. I want to know why you didn't see that." Putting the participant on the defensive or into a stressful situation of having to justify an action is almost always counterproductive. A defensive question must be avoided.
Do's
- Do ask open-ended questions. For example, "How do you feel now that the study is over?" "What is your overall reaction to the application?" "Could you describe what you were thinking about when you were choosing an action and then an object?" Open-ended questions allow the participant to comment freely on a somewhat limited area of the study.
- Do ask multiple-choice or yes-no questions. For example, "Would you prefer an action-object or an object-action choice sequence?" "Did you have difficulty entering the date?" Multiple-choice and yes-no questions serve to formulate the participants thoughts in a form that provides researchers with a great deal of information.
- Do ask short-answer questions. For example, "What are the first three or four words that pop into your mind about the procedure to install the application?" Short-answer questions serve researchers by informing them what paths of thought the participants are taking about the study.
- Do provide reflective feedback. For example, "OK, let me make sure I understand. You are happy with the application, except for having to repeatedly enter your customer number, and this is frustrating." (If the participant is overly frustrated, the researcher needs to take steps to ensure they do not leave with that frustration. This is beyond the scope of this set of tips.)
- Do express appreciation for the participant's assistance. For example, "We appreciate your taking time out of your busy schedule to help us make sure our design is the best possible."