Thanks Elijah... I really appreciate your thoughts. I did some further research that might help others in the future. I was considering asking members of the entire (or possibly just my own) LinkedIn network to complete an online survey. I don't think this would entail any programming (beyond the survey itself). LinkedIn has a Question/Answer area and I found previous requests for survey participants for dissertation studies (one made reference to their IRB). However, when I look at the intended use of the area, this type of request doesn't seem to fit (but the previous requests were not flagged for improper use). I would be a little hesitant given this. I think the self-selection issue may be the most concerning of all as the membership appear to be biased in favor of executives, entrepenuers, and people who are looking for work (or recruiting for jobs). I would need a more balanced sample requiring current employment by an organzation at various levels. As far as sampling... in I/O Psychology I've noticed that many dissertations (and published studies) use non-experimental designs and convenience samples. But I'm being stubborn and wondering if I couldn't figure out a way to get a better sample. I'll have to talk to my chairperson and get his thoughts. In the end, because my 2nd degree network has thousands of people (not to mention the 11 million LinkedIn claims), I found the idea intriquing and wondered if anyone had experience with it (or had discounted it). This might be outside the realm of the aoir group, as I would not be studying the network itself (but I'm actually not sure about this). Thanks again for your time! Elena
Subject: [Air-l] Using LinkedIn for survey sample
Has anyone recruited survey participants from a social network
(e.g.,
LinkedIn)? I'm looking at recruitment possibilites for a dissertation on organizational behavior (motivation to lead and moral philosophy). It occurred to me that LinkedIn might be a good resource for a sample of professionals, but I'm not sure of the privacy/ethical aspect. Any thoughts or experiences?
less from the privacy/ethical aspect, but more from the practical aspect...
1) does LinkedIn's terms of service (TOS) allow you to use their data in this fashion? Your IRB should probably study that document carefully as part of your project approval...
2) LinkedIn is somewhat self-selecting - in that you get a very particular set of folks, with a particular set of goals, signing up for it.
3) difficulty of accounting for bias in your sample. how're you going to explain your sampling strategy, and what defenses will you use in discussing it with people who care about that sort of thing?
4) data collection issues - how're you going to extract this data that you want, and where're you going to stick it? IIRC LinkedIn doesn't have an API as such, and so hiring a programmer to write a cool little crawler for it in an afternoon is a pretty substantial investment of time, energy, and dollars.
5) can you really get at organizational structures from the (very limited, in terms of information available at linkedin.com...) data available on the folks that you sample out?
Anyway... these are the things that come to mind fairly immediately. There's research out there about sampling from incomplete networks; some of that should prove most instructive as you work out your ideas...
--elijah _______________________________________________ The air-l@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/
--------------------------------------------- This message was sent using Monmouth Internet MI-Webmail. http://www.monmouth.com/