I was recently on a call with a new client, and they seemed a bit surprised when I told them that they needed to recruit considerably more participants than their enrollment goal calls for. Their enrollment goal was 400, but their research was going to be spread over eight months. I explained how they were naturally going to experience attrition over that period, and they needed to be prepared for it.
If you’re not new to research studies and trials, then you’re likely already familiar with the importance of planning for no-shows and drop-outs. No matter how effective your recruitment is, you’re going to experience both. You’re dealing with people, after all, and people are never easy.
However, a question we’re commonly asked by both new and experienced researchers alike is this: how many extra participants should we recruit?
Let’s walk through how we approach this at Rekrewt for the government agencies, universities, and businesses that hire us. Let’s identify the “risk factors” that will influence how large your buffer needs to be.
First, let me be clear: there is no simple formula that you can use to answer this question. My answer is going to be more artistic than scientific, as there are quite a few factors that will influence your attrition rate.
One of the most impactful, as I alluded to earlier, is the length of your research. If one study consists of only a thirty-minute zoom interview, while another study consists of months’ worth of questionnaires and check-ins, then obviously the second study is going to experience many more drop-outs.
Another factor is whether your research is in-person or remote. Again, like the last factor, not much of a surprise here, but obviously in-person research that requires scheduling, transportation, etc. is going to have no-shows and drop-outs at higher rates than studies that allow a person to participate from just about anywhere.
Additionally, the amount of your compensation, and how it is disbursed, can also have impacts on attrition rates. Many of our clients structure incentives to be back-loaded, so that the majority of the incentive is only received after participation is completed. I’m a big fan of this approach.
Now, of these factors I’ve discussed, they likely haven’t been big surprises to you. However, the more of these factors that are present in your study, the more additional participants you need to recruit.
Telling you a flat percentage is tough, as some of you are hoping to enroll a few dozen, while others are hoping to enroll a few thousand. However, our general rule of thumb is to recruit double the number of participants you need.
I’m not suggesting that you completely enroll double the number of participants that you need. That would get very difficult to manage. However, you absolutely should have plenty more participants in the hopper that you can enroll to compensate for no-shows and drop-outs.
As far as how many extra participants you should enroll, I’d suggest 15-30%. I know that the thought of using your budget to potentially compensate more participants than you need isn’t appealing, but trust me: this is much better than scrambling to enroll additional participants once you’re already weeks or months or years into your study. I spoke with someone else last week that had recruited the 30 participants she needed, with no extras, but 27 of them had failed to complete their participation. Now she is having to start over with recruitment. That is not a position you want to find yourself in.
So, those are my rules of thumb: recruit 200% of your ideal participants, and enroll a total of 115% – 130% of your ideal enrollments. Adjust those percentages depending on how many of the “risk factors” I discussed are present in your research.
If you’d like to discuss planning for your specific research, let me know.