Hi,

In our everyday lives, we develop norms around how often we should ask people to do something for us. We don't want to pester people and we take their silence as a "no." Participant recruitment is different.

My first academic job was as a research assistant at the University of Chicago's Survey Lab. My job was to contact people to take our surveys online and over the phone. For one study, we emailed participants four times. Those who didn't respond, we called twice a week.

Not all studies should follow this approach, but the early lesson I learned was to believe in your research and your participants. People want to participate and they don't respond for a variety of reasons that don't mean "no, stop calling me."

In my earlier story, we initially sent four emails. I still believe the 3-4 email range is the right starting point for a study. At Volunteer Science, we typically send three emails to our pool when recruiting for a study. Our data reveal a camel hump. The first and third emails have roughly equal yield. The second email has the highest yield.

We have to give others time to respond before sending a reminder. In everyday life, we often wait a week before prodding someone. In our emails, we see most respondents come in within 48 hours. After that, the response drops to almost zero (though we still have people opening emails years after we've sent them). So, in our threemail burst, we wait roughly 3 days between emails.

The one exception is when we need to recruit many more people from the pool than we would get from the threemail. In this case, we send one email per month. This helps us maximize the total number of reminders while minimizing the annoyance.

These rules of thumb are good for a subject pool, people who've signed up to participate in studies in general or for cold contacts - people who you're reaching out to who don't know you. But, there are three reasons you should contact people more often:

1. They've already agreed to participate in your study. If someone has already said they will participate in your study, you should contact them until they participate or tell you that they've changed their mind. It's okay if they ghost you. It happens. But when they give you that initial signup, they are giving you permission to contact them until they participate. You can save yourself time by making it easy for them to change their mind and opt out. But don't be afraid to follow up a dozen times until they tell you to stop.
2. You have a very narrow sample. Sometimes you're recruiting a very special sample of people and there just aren't that many of them either in the world or in whatever list you've managed to construct. At the Survey Lab, we were recruiting Teaching Artists and there are only a couple thousand in any metropolitan area. In this case, you want to follow up often, but the follow up has to include an explanation as to why it's important they participate. We talked a lot about the importance of teaching artists and what our survey was trying to accomplish by surveying them.
3. Reducing selection effects is critical. Finally, there are some cases where selection effects can be very corrosive on the science. Probability-based sampling is one. But so too is any study where important segments of the population are under-sampled or there are certain arms of an experiment that may be difficult for participants to complete. In these cases, you must go the extra mile to get participants to do their thing. For example, the Survey Lab regularly hired people to drive to participants' houses and knock on their doors. You don't necessarily have to do that. But understand it's okay to do more outreach to people whose participation is particularly important and be transparent about this fact. People appreciate the explanation.

As always, I hope this advice is helpful. At the very least, I hope it gives you license to reach out more frequently.

If you have any questions, please feel free to reach out to us at support@volunteerscience.com. And if there are other topics you want us to cover, let us know. Like radio DJs, we'll take requests.