In research, we ask people for their consent so that we can be sure they understand and agree to any risks. A typical consent process includes four parts:
  1. An explanation of the study and why it is being done, and what information we are asking for
  2. What the risks and benefits are of the study; for example, one of the risks of a survey that asks for personal information could be discomfort with the topic
  3. Ways that the participant can gain more information or ask questions about the study
  4. An opportunity for people to opt out of the study
People can give their consent in two ways: actively, by signing off that they consent; or passively, by letting the research happen without any interference. Consent procedures can go wrong in a few ways. They can demand a level of consent exceeding the associated risks of the study. For example, we have experience with a school district that required active consent from parents, which costs a great deal of time and money to manage, even though the risk to students is seemingly very low, with a short, 10-minute survey focused on various topics in school accompanied by a picture drawn by the student; student grades (not linked to names) are also accessed. We learned that this school district experienced problems related to poor consent procedures in the past, so has tried to prevent future problems by requiring rigorous consent procedures. Consent can be way too lax, too. In one study we heard of, evaluators did not obtain any consent, either active or passive, to survey and interview young people in juvenile detention facilities. This is one of the most vulnerable populations; young people, below the age of consent, who have little control or choice in their activities, and probably felt the study was required (this crosses a line of coercion which is a clear no-no in evaluation ethics). The evaluator who was later hired to take over the study had to go back and re-initiate consent procedures. Finally, the consent process itself could put people at risk – for example, letting non-custodial parents know where their children are. The consent procedures must remain focused on the study, and not the people involved. To avoid these problems, just consider a few basic things: What risks are there of my study itself, and what kind of care should I take to inform respondents? What information do I need to share? How prepared will the respondent be to give consent?

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to Yahoo BuzzAdd to Newsvine