Ethics Boards Harming Survey Research, says York Professor

October 2005

The proof is in, says a York University professor: overzealous research ethics boards and university legal departments are undermining survey-based research. They’re doing this by requiring researchers to include with their surveys unfriendly and overly legalistic cover letters or consent forms, says sociologist J. Paul Grayson. These types of letters and forms are unwarranted for simple survey research and scare off respondents, lowering participation rates, he says.

Dr. Grayson has made these claims before, but now he has data to back them up. Frustrated by the situation, Dr. Grayson decided to conduct a study-within-a-study to demonstrate how ethics boards are undermining survey research involving students.

The problem stems from research ethics boards applying a “one-size-fits-all” dictum to university research involving humans, says Dr. Grayson. According to the guidelines of the Tri-Council Policy Statement: Ethical Conduct for Research Involving Human Subjects, adopted by the three main federal research granting agencies in 1998, all research involving human subjects at Canadian universities must be reviewed by research ethics boards to ensure that, among other things, no harm will befall the research participants. Dr. Grayson says there is no evidence to suggest that survey research has any negative impact on subjects.

In 2003, researchers at York, the University of British Columbia, McGill University and Dalhousie University received funding from the Social Sciences and Humanities Research Council to investigate the experiences of domestic and international students during their first three years of study. Research ethics boards at the four universities reviewed the project, including the proposed letter of introduction inviting students to participate.

The researchers designed the letter to provide the required information without discouraging participation. “This is the first contact you make with the person, and it’s very, very important to establish a proper contact,” says Dr. Grayson. An unfriendly or legalistic letter “creates the wrong kind of environment.”

The proposed letter was approved by the ethics board at York and with minor changes at Dalhousie. At McGill, it was also approved, but the legal department insisted that a “highly legalistic” consent form be included. At UBC, the letter was rejected in favour of a “highly detailed and legalistic letter,” he says.

The letters were sent to randomly selected first-year students at the four universities, with students at each institution receiving the letter approved by its ethics board. Consistent with Dr. Grayson’s expectations, the response rates were highest at York (43 percent) and Dalhousie (38 percent), and lowest at UBC (33 percent) and McGill (20 percent). The differences were statistically significant.

To be certain that response rates were not due to differences in student characteristics between the universities, Dr. Grayson decided to send the UBC letter and the McGill letter and form to a randomly selected sub-sample of students at York. Again, as expected, the response rates were lower for the York students who got the UBC letter (24 percent) and the McGill letter and form (29 percent) than the 43-percent response rate for students who received the original York letter. The implications for survey research are troubling, says Dr. Grayson. Lower response rates may mean insufficient sample size of the need for more follow-up.

“This is costing us money and it’s costing us quality research.”

He calls for survey research to be taken out of the hands of research ethics boards. “There is no reason to believe that survey researchers themselves are incapable of designing and implementing ethical research.”

Bruce Clayman, chair of the Interagency Advisory Panel on Research Ethics, countered that while surveys present no physical risk, there is always the risk to participants’ privacy if their responses aren’t properly protected. “So even the work that may seem very innocuous can bear a potential threat to the person’s well-being,” he said, and should therefore be reviewed.

Nevertheless, he is sympathetic to the concerns of social scientists. The advisory panel has a working group that’s looking at ways of modifying the Tri-Council guidelines “to try to minimize the amount of difficulties social science researchers have while maintaining the highest ethical standards in the conduct of that research.”

There is also provision for appealing an ethics board decision, Dr. Clayman noted. But Dr. Grayson said he doesn’t see the point. “You basically are appealing to the same people who rejected you in the first place.”