How to recruit for a scientific study?

July 17, 2012 12:17 PM

Calling all scientists! Conducting a research study...how best to recruit my target population?

I'm a Ph.D. student, currently collecting data for a dissertation on women's sexual issues (specifically lack of orgasm and low desire). In addition to recruiting people from the local medical center/doctor's offices/university, I've been trying to get people from online sources. Specifically, I've been using Craigslist, Reddit (because of their targeted subreddits), and the Village Voice Classifieds. I tried using Mechanical Turk, but I can't seem to get that to work.

I'm not sure why, but most of my participants to date (40 so far) have been in their early-to-mid 20's...which is not so good from a diversity perspective. I'd love to get a hold of some older women, but: a)I'm not sure as to where else I can recruit locally; and b) where older women hang out online.

Any ideas? (and no, I'm not trying to recruit from here...I'm legitimately stumped!)

posted by amyshmamy to Science & Nature (19 answers total)

Have you considered putting an ad in the newspaper? Maybe run it for a week?

Facebook? Put it out in individual messages that you need anonymous (and maybe post your IRB privacy schpiel, generalization about what you're hoping to study, they can quit any time, subject identity is protected, etc) random subjects for your PhD, and you're hoping they can send you some people to _survey link_.

Do this in individual messages. Include the recipient name in each message, and something about how they maybe know you're working on your PhD, or the last time you saw them, or something. Do not spam your friends. I repeat, Do not spam your friends. By that, I mean, do not write one message and then send it to 100 people.

Finally, have you asked your adviser about this?
posted by bilabial at 12:25 PM on July 17, 2012

Continuing ed programs, particularly those offered through traditional four-year schools, tend to draw more mature populations. If you have a few universities near you that offer night classes, you might contact department secretaries to see about postering/listserv/whatever other ways there might be to reach those populations.

If the program itself is somewhat relevant to your field-- nursing? social work? science ed MA?-- then you might even have luck simply by emailing the professors and asking if they'd mind forwarding a blurb to their classlist.
posted by Bardolph at 12:26 PM on July 17, 2012

Parent groups of any kind (online, real life) or really anything associated with kids would be a source of late twenties-early thirties women (you'll miss the child-free people, of course, but parents are a pretty wide section of the population).
posted by randomnity at 12:27 PM on July 17, 2012

Locally, you could post on church bulletins (with their blessing), in nursing homes, and at hobbyist/group sites (dancing, speaking Spanish, AA, yoga, etc).
posted by vegartanipla at 12:29 PM on July 17, 2012

Snowball sampling...ask your participants if they can name others who might also be interested in participating in the study, particularly women who are in your target age groups.
posted by _cave at 12:33 PM on July 17, 2012 [1 favorite]

You don't mention what kind of incentive you're offering. Is it something that everybody would be interested in (cash, Amazon gift card, etc.)?
posted by Wordwoman at 12:35 PM on July 17, 2012 [1 favorite]

Urbanbaby.com has an audience of mothers in the NYC area -- most in their 30s and 40s. I see a fair amount of study links posted there.
posted by xo at 12:42 PM on July 17, 2012

Response by poster: Part of the problem is that I don't have a lot of money to get this recruitment off the ground. For example, I'm aiming for a sample size of 60 - it would be hard for me to pay out of pocket, even $20 per interview. (yay for being a poor grad student with no grants!) I'm entering everyone into a raffle for a $150 Amazon gift card, but I'm not sure if that's enough.

I have tried a snowball sample...the problem with a sexuality study is that a lot of these women don't talk about their sexual issues with others and vice-versa. Asking others to participate would require them to self-disclose. So people have been a little wary of that.
posted by amyshmamy at 1:11 PM on July 17, 2012

I'd look to see how others have recruited.
posted by k8t at 1:17 PM on July 17, 2012 [1 favorite]

How about women-centric places such as gyms like Curves, salons, gynecologist offices, etc.

Also, you probably have this covered but just in case, make sure the recruitment materials/places are ok'ed by your IRB.
posted by Katine at 1:18 PM on July 17, 2012

Response by poster: Oh! That's a great idea to post at the women-centric places. I've got gynecologist's offices covered, but the others are great suggestions.

I also like the Facebook idea, although I'm afraid I've already tapped most of the people I know for possible leads. It wouldn't hurt to contact acquaintances, though.

Also, would it be possible to "market" my study by creating a Facebook page for it and then doing...something with it? (I'm ridiculously non-proficient in the ways of Facebook.)
posted by amyshmamy at 1:27 PM on July 17, 2012

I think you'd get more takers if you made the odds on the raffle better, 15 $10 Amazon instead of 1 $150 Amazon. If you have 60 people, that would be a one out of 4 chance to win instead of a 1 in 60.
If you post on one of the freebie sites that list surveys like slickdeals, you'll get takers. The only problem I'd worry about is you might get bad data from people willing to say they have whatever you're looking for to get in the survey. Most surveys use a pre-screen to weed these people out with questions that don't hint which way they need to answer to be accepted.

Here's the link for slickdeals survey page if you want to give it a try.

http://slickdeals.net/forums/forumdisplay.php?f=39
posted by stray thoughts at 3:16 PM on July 17, 2012

Whatever way you recruit, PLEASE make sure that it is okay with your IRB and that your recruitment techniques are acceptable for your discpline/the journals you want to publish in/etc.
posted by k8t at 3:19 PM on July 17, 2012 [2 favorites]

Yes, what k8t said. You want to make sure your dissertation gets you where you want to go after you get your degree.

Having said that -- other ideas for places to recruit older women online: weight loss communities (I'm thinking of fitbit.com or loseit.com), alumnae association forums, craft or quilting forums, sorority listservs, any neighborhood association listservs you can find in your area. In fact -- maybe think listservs instead of forums. My Mom is just discovering them.
posted by OrangeDisk at 3:23 PM on July 17, 2012

Best answer: You can advertise on Facebook for a fee. They met you target specific genders and age groups.

I know someone who did this with some success.

As k8t says, run this by your IRB first.
posted by k8lin at 5:52 PM on July 17, 2012

Response by poster: Re: Facebook - No way! That's awesome! I'll definitely look into that.

Also, I'm very aware of the requirements of the IRB - I've modified my application more times than I care to remember. :-)
posted by amyshmamy at 6:00 PM on July 17, 2012

I think you'd get more takers if you made the odds on the raffle better, 15 $10 Amazon instead of 1 $150 Amazon.

Agreed. As someone that used to participate in studies, the chance to win $150 is unappealing.

I'm not sure what your study specifically involves, it seems like, since you're looking in the area, that you are interested in doing face-to-face interviews? This will involve these women having to take the effort of getting to the location, taking time for the interview, and going home, and for what? A CHANCE to win money? This is probably why you're getting so many 20 year olds - the majority of 20 year olds have an excess of time and energy. Spending some time for the chance of cash is not a big deal. Once you get older, though, your energies are more divided and the CHANCE of cash is not an incentive. Are you offering these women anything besides just the chance to talk? The only time I participated in a study for free I was given new treatments for my asthma.

I've always thought you pay scientific study subjects to compensate them for the time and effort of participating in the study. With your $150 gift card (or even 15 $10 cards) you are only compensating (or overcompensating) one or a few of your participants. As an "older" woman, I would probably read that an move on. ::shrug::

So if you want a wider base I would:

1) offer ALL the participants SOME kind of incentive - free parking, gift cards to locations, or even see if there are doctors working with low libido in women and advertise some participants will receive a referral to the program (or something, I dunno....but that would also have the downside of just attractive ONE type of woman - one who knows she has a low libido and is unhappy with it - which is not good from an experimental design perspective). I know you said you're strapped for cash, but these are the kind of decisions that need to be made. I can't fudge on using sterile equipment because it will put me in the red, I either do an experiment properly or I don't. If you don't offer better compensation, I don't know that you will get more than these 20 year olds, which, as you say, could affect how your results are interpreted. You may have to reevaluate your thesis and say you're looking at libido solely within a narrow age range.

2) Open up the questions to an online forum. If the questions do not NEED to be asked in person (do they require a physical examination? ...if so, going back to one, you can offer a free pap smear/blood work/whatever) then see about setting up online interviews. This would also help with the honesty of self-reporting.


Honestly, I don't know where "older" women hang out (if "older" is just above mid-twenties) - I've always seen pre-menopausal women in "melting pot" scenarios. The 20's crowd, 30's crowd, and 40's crowd blend together for the most part in terms of locations. I'm guessing the 20's are just going for your bait, but it's not good enough for beyond that.

I'm aiming for a sample size of 60
BTW I would not consider this to be an adequate number of participants, especially if you are dealing with self-reporting. I'm not in psych, though.
posted by Lt. Bunny Wigglesworth at 7:12 PM on July 17, 2012 [1 favorite]

Response by poster: Thanks for the feedback. I hadn't thought of refocusing on a narrower age range, but it's a good idea. After asking my advisor about it, we've agreed that's the way to go.

As for the compensation...I had done an earlier study on vulvodynia, and people (with a range of ages) were more than happy to talk to me without the promise of any compensation, much less a raffle. Perhaps that experience skewed my expectations a bit. By the time I realized there was a problem, I didn't know how I could just switch my compensation midstream - I felt that it was unfair to those who had already participated. Maybe not.

Finally, this is a qualitative study - textual, not statistical analysis. So a sample of 60-70 is acceptable in this instance, I think.
posted by amyshmamy at 6:22 AM on July 18, 2012 [1 favorite]

Just wanted to say, it's not unfair at all to re-evaluate your compensation mid-study. It happens all the time, especially in instances of recruitment troubles. It's not unfair to the previous participants; they consented to the terms of the study at the time of their participation. They'll likely never know the compensation was changed. If that's something you think would help recruitment, it just takes another mod to the IRB. :) No one would bat an eye.
posted by Katine at 10:38 AM on July 18, 2012