Study Recruitment

[Ollie123]

---

Members don't see this ad.

For those doing clinical research with community samples (i.e. not undergrad subject-pool studies), curious to hear what your main recruitment strategies are these days. We've been struggling more and more to get people in the door. Main strategies right now are craigslist, flyering and university listservs. Mixed success with facebook advertisements. Television/radio are expensive enough it doesn't seem to be worth what it gets us. Direct recruitment through university clinics is hit or miss depending on the specific population. Newspaper is (perhaps not surprisingly) a complete failure.

Friend had good success with a tech startup that used an algorithmic approach to do targeted facebook ads, but they appear to be shutting down that line of business. We've tried a couple other similar companies without success. Curious what other folks might be trying. Considering a snowball sampling approach, even though its uncommon in my area and introduces some statistical concerns due to interdependency coupled with the relatively small samples we end up with in laboratory studies.

Also - if anyone is aware of graphic design companies familiar with developing recruitment materials and working with IRBs, I'd appreciate a PM with their info.

 

[ToneTone]

I used Nextdoor and Reddit for my online study. When I did a qualitative pilot study, I recruited through Reddit, in the subreddit for the area I am in (which is a large area) with decent success.

 

[MCParent]

I’ve used mturk with great success lately but its usefulness is highly dependent on your project. With some validity checks you can get good data from hard to reach populations. I got really wonderful brief qual data for a pilot study and lots of good quant data on shorter, simpler things and very simple experiments.

I’ve used reddit with mixed success. I don’t think subreddits for special populations (eg folks with depression) like people popping in to get data from them. I’ve had much more success there with studies about things highly relevant to the group or clearly non-pathologizing.

Just call snowballing “respondent driven sampling,” which is totally in vogue right now (I’m being flip, it is a littler different).

 

[futureapppsy2]

I’ve used mturk with great success lately but its usefulness is highly dependent on your project. With some validity checks you can get good data from hard to reach populations. I got really wonderful brief qual data for a pilot study and lots of good quant data on shorter, simpler things and very simple experiments.

I’ve used reddit with mixed success. I don’t think subreddits for special populations (eg folks with depression) like people popping in to get data from them. I’ve had much more success there with studies about things highly relevant to the group or clearly non-pathologizing.

Just call snowballing “respondent driven sampling,” which is totally in vogue right now (I’m being flip, it is a littler different).

+1 on mturk
Also, snowball sampling

(PS. @Ollie123 , this reminds me that I owe you an email)

 

[G Costanza]

I had a very specific criteria and posted in forums where those people were likely to belong.

Sounds like you’re doing everything right. Can you increase incentive to participate?

If you’re asking about a specific sample and not sampling in general, perhaps we could brainstorm something more tailored to what you need.

 

[Ollie123]

Many thanks for the responses so far. Reddit is a great idea - I haven't tried that one yet.

RE: mTurk - are you actually using it for "recruitment" for clinical trials or data collection for online surveys? We've done the latter some, but hadn't even heard of it being used for the former. I poked around and couldn't find any examples of others doing so...is this a thing now? How are people doing it?

Population for this one isn't rare at all (smokers). ~15% of the population in our county - higher in surrounding counties but they are also more rural so much smaller absolute numbers. I don't think incentives are the biggest issue, we pay a fairly sizable sum ($400 - works out to ~$25/hour) for what we're asking and once we contact people they rarely turn us down...just not generating enough leads. Once no-shows happen and ineligibles are ruled out due to medical contraindications, wonky bloodwork or MRI contraindications they didn't disclose over the phone, we're going at a snail's pace.

Anyways, may be something we just have to deal with and I can "design" around it to some extent though I hate feeling limited in that way. Curious if there are strategies/sources others have tried that we haven't tapped yet.

 

[Ollie123]

Bump.

Other thoughts/responses? Anyone had success with an external recruitment agency or have other strategies they are using?

 

[temppsych123]

No specific thoughts on recruiting smokers, but one thing I might recommend would be to look up who else has funded grants to do similar types of research as you do (NIH Reporter is great) and reach out to those individuals about strategies that have worked. Especially if they're not in the same city/metro area as you (e.g. they have no reason to be "competing" with you for participants). Although even people in your area might be helpful, insofar as you could refer folks to each others' studies.

 

[MAClinician]

After undergrad I was an RA for a year on a smoking cessation study in the NYC/NJ area. We worked with residents at teaching hospitals and would sit in the waiting room at pediatric clinics to recruit participants (the study examined second hand smoke in babies and kids under 2 I think). Have you already reached out to doctor offices, health clinics and hospital placements? I know you said you've done flyers but wasn't sure where they were placed. Just an idea to throw out there.

 

[bcliff]

what about doing brief workshops, lectures, meetings, etc. at groups/agencies that would be interested in learning more about what you do? or on general themes of health behaviors, psychology, etc. — i worked on a longitudinal study once (with no shortage of newly enrolled participants) and the primary recruitment strategy was workshops/lectures/meetings, etc. at community groups, recruitment at those group meetings, and then subsequent word of mouth — we offered no incentive other than a brief description of basic cognitive functioning (e.g., “no red flags”). Maybe doing something on the front end like that is contraindicated given your design (e.g., pre/post design re: the effect of psychoed on smoking specific health literacy), but if not, then it could be worth a shot.


Sent from my iPhone using SDN mobile

 

[tr]

Nextdoor has been our most fruitful source by far for recruiting from the general population. A distant second has been recruiting through flyering on campus and at local health clinics.

For recruiting pregnant women I had good success using a dropbox in Ob clinic, where people could write their name and contact info on a card and drop it in the box if interested. This required some babysitting (an RA has to drop by clinic weekly to pick up cards, refresh blanks, restore stolen pens etc) but was high yield because it didn't put the onus on the subject to make first contact. This could work in smoking cessation groups or primary care offices for your population.