Skip to main content

Online interviews 2: Purposive sampling and digital divides

Ten years ago, the “predicted trajectory of future penetration” of the Internet generated optimism on behalf of online research (Couper 2000). The rise of the Internet as a pervasive communication technology heralded unprecedented reach and ever lower costs of social science.

This second post on conducting qualitative interviews online is concerned with exactly this reach - the recruitment of respondents. Borrowing from quantitative lingo, one might call it sampling. But it should be made clear that theoretical purpose rather than probability must guide the choice of who to interview. Thus, I suggest the term 'purposive sampling'.

A major obstacle to purposive sampling of interviewees for online questioning is the so-called digital divide. The divide exists both between Internet users and non-users, and within the flock of Internet users.

According to Couper’s early meta-study, between 27-53% of adult Americans used the Internet in 1998-99. A recent US survey said the figure was now 74%, with growth in penetration flattening out since 2006 (Rainie 2010). To put that into perspective, the 74% is dwarfed by the 83% who claimed to own a cell phone.

Although not directly comparable, these figures suggest that using online methods reduces the number of people that can be accessed. This is termed coverage error in quantitative research. In qualitative research it means that the people needed to inform the research are simply not reachable via the Internet.

This is extra problematic, because Internet access is clearly associated with social stratification. Even without confronting the digital divide between the global North and the global South, there are major domestic discrepancies between Internet users and non-users. To put it crudely, Internet users tend to be too white, too well off and too educated to be representative of the population as a whole.

Apart from simple access, lack of skills can also be a barrier to participation in online interviews. For example, something as fundamental as literacy is a potential issue when conducting interviews in a text-based environment.

Another related issue is technological constraints. Owning and using a computer does not mean that it has the required hardware or can run the software necessary for conducting interviews. This is a real threat, and a reason for choosing text-based interviewing over voice or video technology.

A final issue that even recent textbooks seem to underestimate is the fact that online communication might be subject to easy wire-tapping. This could intimidate vulnerable respondents. Another security-related problem is that recruitment might be hindered by the lack of accessible directories of online contact details.

The next post in the mini series on online interviews deals in more detail with the case of interviewing through text as opposed to speech.

Couper, M. P. (2000). Web surveys: A Review of Issues and Approaches. The Public Opinion Quarterly , 64 (4), 464-494.

Rainie, L. (2010). Internet, broadband, and cell phone statistics. Pew Internet & American Life Project. Washington, DC: Pew Research Center.


Popular posts from this blog

Official statistics: 51% of 16-74 year old Danes use Facebook

In making a case for why my MSc dissertation here at the Oxford Internet Institute should be concerned with something as hyped and mundane as Facebook, I've been looking for numbers on the Danish social media landscape.

On the English-language web, the commercial SocialBakers Facebook statistics suggest that 49% of the Danish population are on Facebook.

This rather non-transparent number can now be compared with a recent report by Statistics Denmark, suggesting that 51% of 16-74 year old Danes have a Facebook account. The second-largest online social network service in Denmark, LinkedIn, is trailing far behind at 8%. Most surprisingly perhaps, a mere 3% of the surveyed age cohort use Twitter.

As such, there are compelling quantitative reasons for choosing Facebook over e.g. Twitter for a case study of how social media reflect life in Denmark. Another recent survey produced for a Danish daily confirms this: A tiny elite of the 319 most active Twitter users in Denmark write half of …

Two (used) comments on Gillespie's new chapter "The Relevance of Algorithms"

I'm in Paris this semester, as a visiting doctoral student at the Center for the Sociology of Innovation (CSI) at Ecole des Mines and at the médialab at Sciences Po. 
Apart from finding myself in the middle of two very lively research communities, I've also been so lucky that a series of cross-institutional seminars on Digital Methods are taking place in Paris this spring.
The last seminar was on "Transformative interaction: web effects on social dynamics", for which I volunteered to prepare a brief comment on one of the selected readings, namely Tarleton Gillespie's chapter "The Relevance of Algorithms", forthcoming in an edited volume on "Media Technologies" to be published by MIT Press. (The full chapter has been uploaded by Gillespie here).
Since I prepared the comments in writing, and since they did in fact spark some discussion, I've decided that it might be appropriate to recycle them as a blog post. Here goes:

Introducing: The Twitter-thing!

Context: The Twitter-thing is the (awkward?) translation into English of 'Twittertinget' - a project I worked on last year with two Danish colleagues, Irina Papazu (CBS) and Tobias Bornakke (Uni. of Copenhagen) in collaboration with the Danish newspaper Politiken. The Twitter-thing is a tool that draws on TCAT in order to build a network visualisation of how Danish MPs use hashtags on Twitter. Here follows my abstract for the upcoming Data Publics conference in Lancaster, where I'll be exhibiting the Twitter-thing.

Parliaments could seem to be highly issue-agnostic places. All sorts of problems move in and out of these large and expensive devices (Dányi 2015), while the membership stays more or less the same in-between elections. But as issues are taken up and left behind by parliaments, they also make cuts in the parliament in the sense that specific sets of parliamentarians become attached to specific issues. The aim of the Twitter-thing tool is to trace these cuts and v…