[Air-L] Facebook and Twitter user recruitment?

Wed Aug 3 15:18:59 PDT 2016

Tanja, just wanted to belatedly follow up and thank you for your
suggestions. I ended up leaning on MTurk and Twitter ads to get me to the N
I needed for my study, but I did contact Qualtrics and Survey Monkey at
your suggestion, too. Qualtrics reached out but never actually got back to
me with a quote and Survey Monkey said they won't work with Qualtrics
surveys, which is how my study is implemented.

I would highly recommend Twitter ads to others on this list. The
acquisition costs went down over time as people favorited and retweeted the
ad, which lent credibility to it and (maybe? not sure) free distribution to
their followers. I was able to bring costs down for MTurk, too, by forcing
workers to manually check a box next to each eligibility requirement to
certify that they did, in fact, meet the requirements, which many otherwise
breezed by. The usual confirmation codes and attention checks were
essential, and one worker also helpfully suggested I use a timer in
Qualtrics to force workers to sit on my Consent Form long enough to have
plausibly read it. I also used a worker qualification for the first time to
prevent Turkers from accidentally retaking my HIT when I re-posted it. The
how-to is here:
https://experimentalturk.files.wordpress.com/2009/10/excluding-workers-from-previous-studies-using-excel.pdf

If I'd felt like I could have, paying more per participant would have been
ideal especially, perhaps, when it came to recruiting from my university's
subject pool, which was not as successful as I'd hoped. But I tried to make
up for the lowish pay karmically by making the study as seamless for
participants as possible (apart from the mental effort).

Anyway, just reporting back in case this is helpful to anyone. Thanks again
to everyone who offered suggestions.

Sincerely,
Galen

On Tue, Jul 19, 2016 at 11:13 PM, Tanja Aitamurto <tanja.aitamurto at gmail.com
> wrote:

> How about subject pools run by Qualtrics, Survey Monkey, etc? We once
> asked for a quote for an online survey from their subject pools. We didn't
> end up using those though.
> I'd be curious to know what makes your MechTurk participants disinterested
> in the study? 90 minutes is a long time though. In user-testing studies, 90
> minutes participation is often compensated with more than $25. It may be
> worthwhile considering distributing the payment so that there's a draw to
> the follow-up sequence too.
>
> Dr. Tanja Aitamurto
> Brown Fellow, postdoctoral
> The Brown Institute for Media Innovation <http://brown.stanford.edu/>
> School of Engineering
> Stanford
> www.tanjaaitamurto.com <http://brokenfence.flavors.me/>
> ~ examining collective intelligence in journalism, governance and design ~
>
> On Tue, Jul 19, 2016 at 4:33 PM, Galen Panger <gpanger at gmail.com> wrote:
>
>> Do others know, perchance, of any subject pools that are open to Internet
>> researchers, where a lowly researcher such as me might be able to list a
>> study? I'm working with my own university's subject pool, but it's summer
>> and they don't anticipate I'll recruit the number of subjects I need. It's
>> going well with this pool so far, but I probably won't hit my numbers
>> there. Looking for any other possible sources.
>>
>> Anyway, thanks to Fabio, Scott and to a couple of others who responded
>> off-list. Fabio, I'll try out the Facebook lead ads!
>>
>> Sincerely,
>> Galen
>>
>> On Fri, Jul 15, 2016 at 11:22 PM, Fabio Giglietto <
>> fabio.giglietto at uniurb.it
>> > wrote:
>>
>> > Hi Galen,
>> > have you also tried Facebook lead ads (
>> > https://www.facebook.com/business/a/lead-ads)?
>> >
>> > Best, Fabio Giglietto
>> >
>> > Il sab 16 lug 2016, 3:36 AM Galen Panger <gpanger at gmail.com> ha
>> scritto:
>> >
>> >> Thanks -- to be clear, I'm not looking for my study to be blasted out,
>> at
>> >> least not without talking first about (1) what the population looks
>> like
>> >> and (2) working on the recruitment language. I shared the URLs in case
>> >> people wanted to get a better sense of what the study/ies entail.
>> Thanks
>> >> for any and all ideas/leads/advice, though!
>> >>
>> >> On Fri, Jul 15, 2016 at 6:30 PM, Scott MacLeod <scott at scottmacleod.com
>> >
>> >> wrote:
>> >>
>> >> > Hi Galen and AoIR,
>> >> >
>> >> > I just shared your request, Galen, as a Minute as part of World
>> >> > University and School's open monthly business meeting process, as
>> you'll
>> >> > see here -
>> >> >
>> >> >
>> >>
>> http://worlduniversityandschool.blogspot.com/2016/07/minutes-for-wuass-july-9-2016-monthly.html
>> >> > - along with some of WUaS's rationales for this, which include as
>> you'll
>> >> > see: "In seeking to become the online Harvards of the Internet in all
>> >> > ~204 countries main languages (accrediting for free CC MIT OCW in 7
>> >> > languages and CC Yale OYC BS/BA, Ph.D., law and M.D. degrees, as
>> well as
>> >> > I.B. diplomas), this is also a model for how WUaS (and if I become an
>> >> > assistant professor in the MIT Media Lab too) may further seek social
>> >> > science samples for research in both courses and in various
>> languages."
>> >> >
>> >> > Thanks for this interesting and rigorous online social media study.
>> >> >
>> >> > Sincerely,
>> >> > Scott
>> >> >
>> >> > https://twitter.com/WorldUnivAndSch
>> >> >
>> >> >
>> >> >
>> >> > On 7/15/16 2:12 PM, Galen Panger wrote:
>> >> > > Hi all, for my dissertation I'm recruiting separate samples of
>> Twitter
>> >> > and
>> >> > > Facebook users for a multi-part study (see
>> >> > http://facebookstudy.berkeley.edu
>> >> > > and http://twitterstudy.berkeley.edu). I'm having some luck with
>> >> Twitter
>> >> > > ads, and with my university's subject pool for Facebook, but am
>> >> wondering
>> >> > > if folks here might have ideas for recruiting more participants to
>> >> help
>> >> > put
>> >> > > me over the top (I'm hoping to end with usable samples of over 300
>> >> each).
>> >> > >
>> >> > > I'm paying about $25 (plus a $500 Apple Gift Card drawing) for a
>> total
>> >> > time
>> >> > > of around 90 minutes, with the catch that the second part of the
>> study
>> >> > > involves downloading and using an app, which is a step that results
>> >> in a
>> >> > > fair amount of drop-off (despite my follow-up efforts, which only
>> >> help on
>> >> > > the margin). I've tried Mechanical Turk but there's too much
>> drop-off.
>> >> > > Craigslist also appears to be taking down anything that's not a
>> >> > > face-to-face transaction, and my study is entirely
>> >> online/Qualtrics/app.
>> >> > > Facebook ads haven't worked at all for recruitment for me.
>> >> > >
>> >> > > Anyway -- if your university subject pool is open to outsiders, or
>> if
>> >> you
>> >> > > have a tip that's worked for you in the past, I'm all ears for any
>> >> ideas!
>> >> > >
>> >> > > Sincerely,
>> >> > > Galen
>> >> > >
>> >> >
>> >> > --
>> >> > - Scott MacLeod - Founder & President
>> >> > - http://worlduniversityandschool.org
>> >> > - 415 480 4577
>> >> > - PO Box 442, (86 Ridgecrest Road), Canyon, CA 94516
>> >> > - World University and School - like Wikipedia with best STEM-centric
>> >> > OpenCourseWare - incorporated as a nonprofit university and school in
>> >> > California, and is a U.S. 501 (c) (3) tax-exempt educational
>> >> > organization, both effective April 2010.
>> >> >
>> >>
>> >>
>> >>
>> >> --
>> >> galen.website
>> >> _______________________________________________
>> >> The Air-L at listserv.aoir.org mailing list
>> >> is provided by the Association of Internet Researchers http://aoir.org
>> >> Subscribe, change options or unsubscribe at:
>> >> http://listserv.aoir.org/listinfo.cgi/air-l-aoir.org
>> >>
>> >> Join the Association of Internet Researchers:
>> >> http://www.aoir.org/
>> >
>> >
>>
>>
>> --
>> galen.website
>> _______________________________________________
>> The Air-L at listserv.aoir.org mailing list
>> is provided by the Association of Internet Researchers http://aoir.org
>> Subscribe, change options or unsubscribe at:
>> http://listserv.aoir.org/listinfo.cgi/air-l-aoir.org
>>
>> Join the Association of Internet Researchers:
>> http://www.aoir.org/
>>
>
>

-- 
galen.website