choice-metrics.com

Dear choice modellers,

I have a general question regarding sampling procedure for choice experiments.

I am considering rewarding my participants (who are winegrowers) for accepting to answer, face to face, my choice experiment, the reward being of a value of 15 euros. Is it common to incentivize respondents to a DCE with money? I am worried about potential response and sampling bias. What would be your advice on this?

Thank you!

In online surveys using consumer panels the respondent is essentially always rewarded, although the reward is usually small (a few dollars). In my driving simulator studies I pay participants around $60 to come for an hour to a laboratory as otherwise it is very difficult to recruit participants.

You are right that preferably your participants are interested in the study and are willing to participate without being paid. Monetary rewards are generally used when you expect that people would not participate otherwise. So if you can recruit a sufficient number of winegrowers that are interested in the study, then you would not need to pay them, but it would not be unusual to provide monetary incentives, even if it is a simple "thank you" gesture for a long survey or interview.

Others may have different views on this, and there may be differences across disciplines, so let's see what others say.

Michiel

It is very common to reward survey participants for undertake surveys of all types. Often people bring people into labs to complete surveys for which they are paid for their time. Internet panels are frequently used which pay respondents if and only if they complete the survey. Focus group participants are also paid for their time, and depending on circumstances, so are people intercepted at locations. Payment is sometimes in the form of a lottery where participants enter a draw to win a prize of some value. There were lots of studies in transportation led by Peter Stopher many years ago that looked at incentives aimed specifically at survey non-responders which found that increasing payments may switch a few people who would otherwise not complete the survey. In some cases, payments are not made, but I would say that outside of health economics, paying or rewarding respondents is probably more of the norm.

As Michiel states, having engaged respondents is crucial.

John

Thank you both for your opinion on this!

Best,
Gaëlle