Why do empirical research?

Document Sample
Why do empirical research? Powered By Docstoc
					              Explanation and Quality

  George Gaskell, Ilina Singh and Martin W. Bauer
              Methodology Institute
          London School of Economics
 ESRC Research Methods Festival, Oxford July 2006

Methodology Institute
  1895: First prospectus of the LSE
 • ‘Training in methods of scientific investigation’.
   Scientific method to be cultivated, and not left to
   take care of itself’
 • The Webbs – in the Shaw Library painted by
   William Nicholson.

Methodology Institute
                        The Webbs
 • Beatrice: did not want to spend money on ‘mere
   propaganda of the shibolleths of collectivisim….Reform
   will not be brought about by shouting. What is needed is
   hard thinking….So Sydney has been planning to persuade
   the trustees to devote the greater part of the money to
   encouraging research and economic study’

 • The Fabian (intellectual left wing) mission: reform of
   society based on scientific knowledge – back to the
   ‘Enlightenment Project’ and anticipating ‘evidence based
   policy making’.

Methodology Institute
                 The science of society
 • “What we have to do in social science is to
   apply the scientific method to the facts of
   social life. There is only one scientific
   method - that used in the physical sciences.
   It consists of three parts – observation,
   conjecture as to the cause and effect of the
   facts observed, and afterwards verification
   by renewed observation. This is the
   scientific method”. B.Webb Diaries

Methodology Institute
                        The Zeitgeist
 • W.H.R. Rivers gave following advice to Bartlett
   for the study of social anthropology
 • “the best training you can possibly have is a
   thorough drilling in the experimental methods of
   the psychological laboratory”
 • For Bartlett this “gives one a sense of evidence, a
   realisation of the difficulties of human
   observation, and a kind of scientific conscience
   which no other field of study can impart so well”

Methodology Institute
                        The LSE traditions
 • The Cartesians
       – Wallas (political science) Geddes (sociology) Bowley
         and Moser (statistics)
 • And the post-Kantians
       – Malinowski (anthropology) and Hobhouse (sociology)
 • Philosophy
       – Karl Popper on falsificationism, conjectures and
       – Imre Lakatos and the Methodology of Scientific
         Research Programmes
       – Thomas Kuhn and the structure of scientific revolutions

Methodology Institute
                        The Science Wars
 • Alan Sokal, 1996. Transgressing the
   boundaries: Towards a transformative
   hermeneutics of quantum gravity. Social
   Texts, 46/47, pp.217-252.
 • A spoof on Derrida, Lacan and others
   immortalised in Sokal and Bricmont (1997)
   ‘Intellectual Impostures: Post-modern
   philosophers’ abuse of science’.

Methodology Institute
                             The wars
 • Scientific orthodoxy
     A world out there of objects, properties and processes; statements can
     be true of false but never know for certain; methods give a good
     handle on learning how things are – but are fallible; science is
     cumulative; knowledge based on sensory experience
 • Social constructivists
     Knowledge is socially constructed, it is the product of social factors
     and not the result of objective investigation of how things are
     independent of social interests. Science is one form of construction and
     other forms are equally legitimate.
 • The pragmatists
     Accept that the subject matter of the social is socially constructed but
     for practical purposes follow the lines of scientific orthodoxy

Methodology Institute
              Contrasts between quant and qual
                        quantitative               qualitative
 Epistemology           Cartesian                  Post-Kantian
                        Stimulus/response          Social construction
 Research question      What is distribution of    Mapping ‘common sense’ and
                        attitudes and opinions?    what lies behind it
 Design                 linear                     More flexible, quasi circular
 Respondents            Statistical sampling       Purposive/theoretical
                        Taxonomic groups           Natural groups
 N of cases             Sampling                   To point of saturation of meaning
                        theory/statistical power
 Data elicitation       Pre-specified questions    Open ended dialogue with probes
                        standardised responses
 Data and Analysis      Numerical inferential      Textual forms of discourse
                        statistics                 analysis
 Warrants               See next slide             Hotly debated

Methodology Institute
                        Quality indicators

 • Well developed for quantitative research
 • Experiments
       – Internal and external validity
 • Psychometrics –
       – Reliability and validity
 • Survey research
       – Sampling and measurement errors
 • Three canonical quality criteria
       – Reliability, validity and representativeness
 • But what of the qualitative tradition?
Methodology Institute

Shared By: