Transcript Microsoft PowerPoint - the NCRM EPrints Repository

Social science methods for the
21st century
Mike Savage
University of Manchester
Uncertain times
There have been huge investments, and a considerable
enhancement of social scientific methodological
capacity, which has delivered undoubted good
This has occurred at a crisis moment when digitalisation
offers profoundly challenging ways of gathering,
analysing and deploying new kinds of social data (see
Savage & Burrows 2007).
These new currents are redrawing boundaries between
natural and social science, and between academic and
commercial research.
In this conjuncture the academic social sciences need to
appealing to technical authority alone and recognise
they are paradigm wars going on around methods
The remarkable rise of social science
The growth of the post-war social sciences is one of the
most important, yet un-researched, aspect of post-war
change.
– Social scientists are 3% of UK academics 1948 > c. 45% 2001
– Social scientists play pioneering role in elaborating new
‘technologies of the social’ in post-war years
• National sample survey (from 1930s)
• The qualitative interview (from 1950s)
• The ‘case study’ (from the 1960s)
– Social scientists generate ‘epoch descriptions’ which come to
define the meaning of social change itself
• ‘Affluence’ (1960s)
• Post-industrialism (1970s)
• Globalisation (1990s)
• Social science disciplines largely replace the humanities
in defining the nation through a focus on its ordinary
qualities (see Identities and Social Change in Britain
since 1940: the politics of method)
UK Social Sciences 1948 – 2003
1800
1600
1400
1200
1000
1948
2003
800
600
400
200
0
Soc
Pol
Anth
Econ
SP
The prospects for social science
today….
There is now an argument that in the 21st century, new
kinds of social research might eclipse the academic
social sciences. This is based on:
– Problems with conventional social science research repertoires
(e.g. falling response rates, the routinisation of interview
methods)
– The growth of huge commercial research agencies, who (sort of)
do social research though with little academic input
– The way that social research agendas are increasingly
intertwined with natural sciences (e.g. climate change; medical
research), so making it more difficult to champion narrowly
‘social’ research repertoires. (RCUK emblematic of larger
changes?)
– The challenge of digital data, which means that social traces are
routinely left, without the need for social scientists to conduct
specialist fieldwork to extract data.
Straws in the wind?
•
•
•
•
•
•
The future of the Census
E-borders
Debates in marketing research.
The politics of MPs expenses
The end of Social Trends
‘crowdsourcing’ devices
2: A new kind of social science?
• Over the past decade, social data has
proliferated beyond our previous imagination,
and we now routinely leave remarkable data
traces.
• Consider Tesco loyalty cards, google, geodemographics, etc
• Does this offer the prospect of new kind of social
research? See the following example
• MIT Media Lab: Reality Mining
Social science reactions
• Reality mining could be dismissed as
– Lacking theoretical awareness
– Gimmicky
– Having too much ‘noise’ and lacking necessary
information (e.g. on individual respondents)
• However, this kind of research is strongly in
keeping with current social theory, with its
concerns with
– The relational and ‘post-human’
– Transactions, interactions and complexity
– Dynamism, movement, and flow
• We need to recognise that different kinds of
paradigms of social research are being
mobilised and that we need to be alive to other
ways of doing social research
Paradigm battles (1): ‘depth’ vs ‘surface’
In defining their identities and activities, academic social
scientists invoke ‘depth models’, implicit in positivist,
realist, and hermeneutic methods
• Both the interview and the sample survey are
championed as a means of delving into, and revealing,
‘hidden’ social processes.
• Both allow ‘inference’, ‘abstraction’ and the search for
regularities, a ‘causal’ social science in which
particularities are subsumed to ‘underlying’ forces
• Digital data, by contrast
– works through surfaces using data on whole (sub-)populations.
– is concerned not with revealing the hidden, but with arraying
surface data in visible and accessible form.
– It concerned with particularising, as much as generalising.
– Is implicated in an audit and commercial ‘neo-liberal’ climate.
– Can be seen as part of ‘descriptive turn’.
Paradigm wars (2): individuals & networks
• The rise of the sample survey in the mid 20th century
depended on discrediting the ‘field analysis’ approach, in
which it was deemed essential to study whole
populations.
• Sociological focus was placed on, and remains with,
individuals (even despite the intellectual prominence of
post-humanist theoretical currents).
• Hence, despite the early prominence of British network
research in the 1950s (Barnes, Bott, Mitchell) this
tradition was largely dispersed (Burawoy 2000).
• But, transactional data allows the radical revival of
network methods, where understanding the links
between transactions, and not the attributes of the
individual ‘transactor’ becomes a central research issue.
• E.g. Amazon; Tesco loyalty cards; marketing research, etc.
• Here, there is also potential for cross fertilisation with
academic social network analysis expertise
Number of articles containing the phrase "social
network analysis"
16000
14000
12000
10000
8000
6000
4000
2000
0
1960
1970
1980
1990
Year
2000
2010
2020
My parting shot
• The second half of the 20th century was
the golden age of the social sciences.
• It is too soon to assess how knowledge
will be transformed during the 21st century.
• But there are important developments
taking place in the natural sciences and in
commercial sector and social scientists
cannot afford to be complacent…