Alicia, I think this is complicated very much by the uneven distribution of traffic across the web. Do you want a sample of what patients are LIKELY to see? Then you need to develop a traffic model of parental engagement- generally speaking, though, people go via search, and recommendation from a trusted source which includes friends on social media as well as official sources, I would guess. Which sources these are may well be mediated by locale, and perhaps by the viewers' position in a community. I don't think necessarily there is any point evaluating advice that no one will ever find. Our colleagues in health info have run hundreds of such reviews. I do not recommend emulating some of them and just picking the top x search results. Search results have their own challenges and should be used with care! But others may have developed more sensitive sampling frames since I last looked. Good luck! Sent from my iPhone
On 15 Nov 2016, at 17:01, Alicia BR <alicialorna@gmail.com> wrote:
Hello all,
I am looking for reading suggestions for how to develop a robust sample for qualitative content analysis of web content that is not specific to a particular platform or period of time.
Here's the scenario: We are looking to create a comparative analysis of 'screen time' advice available for parents. This includes advice from commercial providers, government, NGOs, peer-to-peer discussion sites, bloggers and beyond.
We can create a typology of these different categories of advice-givers and once the sample is generated will be using a deductive analysis framework based on the parental mediation literature (in part looking to see if popular advice and research literature match up). But before we get there we are struggling to generate a sample that is both purposive (representing the different categories of institutions and individuals acting as advice-givers) and yet also representative within those categories.
Sonia Livingstone and I conducted a short version of this exercise for a policy brief <http://eprints.lse.ac.uk/66927/> on screen time earlier this year, but now want to extend this into a more detailed study.
Any suggestions for methods readings and/or examples of work that have achieved something similar are most welcome!
Best, Alicia
Dr Alicia Blum-Ross Research Officer, Parenting for a Digital Future Department of Media & Communications London School of Economics and Political Science www.parenting.digital _______________________________________________ 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/