In program evaluation, we often don't have to go through and IRB or ethics review before we conduct a survey. That means we are responsible for conducting an ethical survey ourselves. The consent process is one area where it's easy to make mistakes that can have real ethical implications.
Here, I outline some common pitfalls and approaches to ensure an ethical survey.
What is informed consent?
Informed consent means that the person taking the survey knows what the purpose of the survey is, how they might be affected by participating, and most important, that they don't have to participate.
Before you ask any questions, you should present an introduction that allows participants to opt in or out of completing your survey. This section should describe who you are, why the survey is being conducted, why they might benefit (directly or indirectly) from completing the survey, why they might not want to complete it (for example if there are questions that might be upsetting), how to get in touch with you if they have questions, and what to do if they choose not to complete the survey. It's a good practice to also let participants know how long you expect the survey to take.
If you are administering the survey in person, you'll want to summarize this information and also give the participant a written copy to sign and one to keep. If you're giving the survey electronically, you will want to describe this on an introductory page and have participants click an "I agree" or "OK" button to continue on to the survey.
If the participants are children, their parents will have to consent for them. Then, the participants should receive the same information and have an opportunity to assent. If the child is 7 or older, they should go through the same process as an adult, with a verbal introduction and a form to sign. Their form should be written in language that they can understand.
Problems arise when we inadvertently contradict ourselves through our actions.
There are many instances when we, as program evaluators, can give mixed mesages about whether completing a survey is really optional. One is that surveys are often distributed at programs where the participants need services. Asking a participant to complete a survey before they get their service can imply that it's required to receive the service they need. It's best to avoid this kind of confusion by delivering the service first. For example, if the survey is being conducted at a food pantry, make sure you do it at the exit when people already have their food and not as they enter. If the survey is being conducted in a classroom, it is very important to make sure students understand that their teachers will not know whether or not they completed the survey, and that it would not affect their grade.
Even in cases where you are giving something away, be careful that you don't mix your survey with your give-away. For example, if you have bookbags to distribute to students on the same day that you would like them to fill in a survey, be careful not to have your bookbag giveaway station too close to your survey station. You can easily create the impression that people should fill in the survey in return for the bookbag.
When conducting program evaluation with disadvantaged or minoritized populations, it is crucial that you consider all the implications of your choices. So here are the takeaways:
- Provide multiple opportunities to opt out of the process, including stopping the survey in the middle
- Make sure that any written materials are in language that the participants can read and understand -- that means having everything translated if your participants speak other languages
- Remember that children can and should assent to participating in a survey
- Make sure that participants know that completing your survey is not connected to the services they are receiving and is not a requirement for receiving services
While collecting data is important, it is never worth collecting it in a unethical way or in a way that takes advantage of vulnerable people. In program evaluation, the participants are often economically or socially disadvantaged and we have to be extremely thoughtful about how we engage them in the program evaluation process.