Should I pay survey respondents? Is it ethically acceptable to do so? Is it ethically acceptable not to do so? Will it affect my research?
I’ve heard this question enough that I felt it was time to address the topic in a blog post. That said, there is no simple answer here. Whether or not you pay survey respondents matters — and it isn’t necessarily a bad thing to do.
As with many aspects of data collection and analysis, the most important thing here is to make an informed decision. Markus Goldstein of the World Bank has written an excellent piece on why paying respondents isn’t as bad as people think — but that doesn’t mean there are no issues at all. Let’s take a look at both sides of the argument.
Pro: You can attract more respondents.
I’ve done a lot of work designing surveys and collecting survey data from respondents. Sometimes, these surveys are long. Like really long. Asking people to give their time to answer a survey can range from inconvenient and off-putting to downright impossible. Offering to pay survey respondents provides an incentive, especially if you’re working in places where an hour spent answering questions is an hour of work lost.
Con: You might appear to incentivize specific answers.
We all want our data to be as unbiased as possible. While paying survey respondents can encourage more people to provide you with their data, doing so is not entirely without risk. Beware of designing your survey (and compensation) in a way that suggests you are looking for specific answers. To obtain accurate data, respondents must understand that they will be paid equally, regardless of their responses.
Pro: It’s ethically responsible.
Survey respondents are providing you with something you would not otherwise have access to — their data. We don’t think twice about purchasing data access elsewhere, and we shouldn’t here. Again, as many of our respondents often belong to vulnerable populations, it seems only right they should be legitimately paid for their contribution to our work.
Con: It can yield more extreme responses.
A study (gated) on incentivizing survey responses conducted in India by Guy Stecklov, Alexander Weinreb and Gero Carletto found that respondents who were paid for their answers tended to provide more extreme answers. The researchers posit that, when survey respondents were paid, they were more inclined to provide honest answers rather than sticking to the easy middle ground.
Pro: It can help build trust.
Establishing a relationship with your survey respondents may not matter much if you’re asking people what toppings they like on their pizza, but it’s critical if you’re asking for personal data. (Especially sensitive personal data that could expose respondents to embarrassment, discomfort, or — in specific cases — physical harm.) When you pay survey respondents, you demonstrate your commitment to the research you’re doing and show you value what they can provide.
Con: It appears to drive respondents to report lower income.
Stecklov and company’s study also revealed that paid survey respondents seem to report having lower income, spending less, and having few assets than unpaid respondents with similar assets and circumstances. It’s difficult to ascertain exactly what is occurring here, but it does seem that respondents attempt to appear poorer when they are paid to respond.
So, Should You Pay Survey Respondents?
Short answer: it depends.
As I’ve spelled out there, there are both pros and cons to the decision to pay survey respondents. Whether or not it’s the right move for you will depend on your research and the population you’re attempting to survey. Who are you trying to get responses from? What does it cost them to answer your questions? The important thing is to know that paying respondents can affect your results. You must design your surveys and analyze any data you collect with that in mind.
Want help developing a survey or collecting data? The experts at Datassist are here to assist you. Talk to us today about your project.