Print

Print
Muki, thanks for this overview and certainly agreed. I am using multiple data sources so am, fortunately, not relying on weblogs alone. I think if we can acknowledge bias in the results, weblogs still have some value in qualitative analysis.


Dr. Paul B. Williams, BAH, MPhil, PhD
Instructor,
Dept. of Geography and Environmental Studies,
Carleton University, Ottawa, ON, Canada



 
Date: Thu, 3 Mar 2011 21:25:28 +0000
From: [log in to unmask]
Subject: Re: Critical use of Blogs as research data
To: [log in to unmask]

While there are all sort of ways to 'mine' or 'mechanically process' information from twitter, blogs and other sources, this should be done with a lot of caution and awareness to who is participating and all sort of biases.

First, all these media is suffering from 'participation inequality' which means that few people contribute a lot, while most people contribute very little - and because of the big numbers or the size of the contribution from the vocal sources, you notice them more than the others, multiple but much more quieter voices. 

Second, there is a gender imbalance in some sources - blogs for example, but also in Wikipedia contribution, OpenStreetMap mappers (my pet subject) or twitter.

Third, the digital divide or just digital ignorance means that you will notice the voices of the affluent and not people who are poor.

Fourth, there is some tendency for clustering around similar points of view (best demonstrated in the climate change bloggosphare with little connection between camps).

Fifth, you're exluding voices that are confident in verbal communication but don't feel secure in their writing.

There are other factors ...

So just use critical thinking when you approach such sources - a lot of the current analysis glosses over these biases.

On 03/03/2011 21:12, Bruce D'Arcus wrote:
[log in to unmask]>
On Thu, Mar 3, 2011 at 2:55 PM, Dr. Paul B. Williams
<[log in to unmask]> wrote:

I have been enjoying this thread on blogging and am amazed at how many
colleagues are using this medium.

The thread got me thinking about issues associated with the critical use of
blogsites as sources of primary research data. I am beginning to use
blogsites associated with places, meanings and sense of identity in my own
research. I have collected a number of these - not generally academic blogs
- over the last couple of years with an interest not only in the blog itself
but also opinions/ views within the accompanying commentaries. I am aware of
the concerns about objectivities/ subjectivities associated with such sites,
but I do think they give a snapshot in time.

So, has anyone else been mining such sites for data? If so, what issues have
you encountered? Have you come across any literature about their use in
geographical research?

I've not (but am curious).

I gather people in the "Digital Humanities" world might have something
to say on this. Dan Cohen has an extensive list on Twitter:

<http://twitter.com/dancohen/digitalhumanities>

I'm more curious about mining twitter data, but haven't figured out a
productive way to do that.

Bruce

--
Dr. Muki Haklay: Senior Lecturer in GIS
Department of Civil, Environmental & Geomatic Engineering
University College London (UCL)
Gower St. London WC1E 6BT
T: +44 20 7679 2745
E: [log in to unmask]
W: http://www.ge.ucl.ac.uk/~mhaklay/
Towards Successful Suburban Town Centres - www.sstc.ucl.ac.uk
Bridging the Gaps - www.ucl.ac.uk/btg
Mapping for Change - www.mappingforchange.org.uk
Po ve Sham blog - povesham.wordpress.com