Choosing your research approach

1
Choosing your research approach

• Research Methods

2
First

• Before getting down to questions of research
  methodology
• Some issues people asked about after last weeks
  session

3
Dates

• Dissertation proposal. Week 4 Friday 7th March
  2003 by 5.00pm.
• Dissertation approval and allocation of
  supervisors. Week 6 Friday 21st March 2003.
• Dissertation submission. Monday 1st September
  2003 by 5.00pm.
• Coursework for this module. Friday 23th May 2003
  by 5.00pm.

4
Ethics

• Linked to good practice more generally.
• If your research concerns people, you need
• No harm Informed consent Confidentiality/anony
  mity Access to data and results.
• Universitys Good Practice Website
• http//www.shef.ac.uk/r/researchoffice/RO/Shef-onl
  y/principles.html
• http//www.shef.ac.uk/r/researchoffice/RO/Shef-onl
  y/guidelines.html
• Any research concerning NHS staff or patients
  must go to a formal ethics committee.
• State any ethical issues in your proposal.

5
Coursework

• 1. A revised version of your initial dissertation
  proposal, encompassing the preliminary work that
  you have carried out during Semester 2 (as
  detailed in the section in the Dissertation
  Handbook entitled Initial Progress of
  Dissertation).
• You should include a bibliography of the
  relevant literature identified in your literature
  review thus far (not a full literature review).
• This section should be 2,000-2,500 words long
  in total, not including the bibliography.

6
Coursework

• 2. A critical review of a previous dissertation
  produced by a student on the Masters programmes,
  on the same (or a closely related) area as your
  own research project. 1,250-1,500 words long in
  total, not including the bibliography.
• In some cases, an externally published report
  or paper(s) may provide a more appropriate,
  alternative source for your critique.
• However, whatever you choose must be agreed
  with your dissertation supervisor.

7
Coursework

• Typical criteria
• Aims/objectives/scope. Are these clearly spelt
  out? Are the research problems and/or rationale
  well defined?
• Research context. Is the project related to
  previous work? Does it draw on the literature in
  the field? Is the research or organisational
  environment described?
• Methodology. Is there a full account of the
  research methods adopted? Are they justified and
  are the constraints and limitations assessed?

8
Coursework

• Typical criteria
• Methodology (continued)
• In the case of a survey, what is the sample
  size, how representative is it, is the
  questionnaire or interview schedule explained?
• Alternatively, in the case of an application
  project, is there a justification for the design
  approach or the tools selected? What constraints
  affected the design. Was an appropriate
  evaluation built in to the design?

9
Coursework

• Typical criteria
• Results and findings. How well are the results
  presented? Are they appropriately interpreted and
  discussed? In the case of an application is the
  final system appropriately documented, and to
  what extent did the final implementation meet the
  specification?
• Conclusions/recommendations. What are the
  conclusions? To what extent do they relate to the
  aims and objectives and are they generally valid?
  Are the recommendations for future work
  achievable and prioritised?

10
Coursework

• Typical criteria
• Presentation. Is the dissertation well
  presented, clearly written and appropriately
  structured? Is the abstract informative? Are the
  references correctly cited?

11
Research Methods Dissertation Preparation

• Research paradigms abound - e.g. Burrell and
  Morgan's (1979) functionalist, radical humanist,
  radical structuralist and interpretative
  paradigms in information systems and
• Olaisen's (1991) empirical, materialistic, action
  and clarified subjectivity paradigms in
  information science.

12
Research Methods Dissertation Preparation

• But arguably we can think broadly of 2 main
  approaches
• positivist and
• interpretative.
• These are often aligned with
• quantitative and
• qualitative approaches.

13
Traditionally research in the physical sciences
has concentrated on...

• analysing complex situations into component
  parts,
• studying them, then
• reassembling the parts into
  the original whole with
  increased understanding.

14
A number of assumptions underlie this approach,
namely...

• we can increase our understanding of complex
  wholes by analysing them into component parts,
  better understanding the parts, then
  re-assembling them to form the whole
• we can discover universal laws of behaviour that
  is, we can identify variables which when
  subjected to the same conditions behave in
  exactly the same way in similar samples.

15
These basic assumptions themselves involve other
assumptions, namely...

• we can measure individual variables in isolation
  from one another (as opposed to defining and
  knowing them through their relations with each
  other)
• having done so, we can profitably re-link them
  using statistical relationships
• we should control variables and avoid the
  intrusion of uncontrollable elements.

16
These assumptions require a strong emphasis on...

• quantitative objectively measurable data
• statistical significance testing in order to
  predict to other samples in a search for
  universal laws
• searching for discrete relationships,
  particularly cause and effect relationships

17
Limitations

• This approach may be highly productive in
  studying e.g. plant biology.
• But has limitations when it comes to studying
  e.g. plant ecology, or the effects of gardens on
  human moods!

18
Interpretative approaches rest on other basic
assumptions...

• the "whole" is more than simply the sum of the
  parts, and cannot be fully understood by means of
  isolating, analysing, testing, then reassembling
  the parts
• complex situations must also be studied in their
  entirety, in a relatively holistic way

19
Interpretative approaches rest on certain basic
assumptions...

• human behaviour is too complex necessarily to
  allow us to reduce it to universal laws. It is
  also necessary to study complex interacting
  phenomena which may interact differently in
  different contexts
• instead of separating and defining variables in
  isolation from one another, phenomena must be
  studied within the context of their interactions

20
Interpretative approaches rest on certain basic
assumptions...

• relationships cannot solely be conceived of as
  discrete e.g. causes and effects
• rather, relationships must also be seen as
  complex patterns of mutual interaction
• findings that relate to restricted contexts, as
  opposed to those having universal applications,
  are valid.

21
Interpretative approaches rest on certain basic
assumptions...

• it is possible/desirable to place more emphasis
  on qualitative data and analysis, even if, as is
  usually the case, this means reduced quantity
  (i.e. the number of cases studied).

22
Interpretative approaches

• Olaisen (1991) considers that this pole of the
  research paradigms dimension is also
  characterised by high-complexity problems, an
  emphasis on social-intuitive as opposed to more
  logico-mathematical analysis, and "sensitising"
  as opposed to "definitive" concepts.

23
Interpretative approaches

• Sensitising concepts Olaisen 1991 254 are
  tentative and speculative concepts that
• "... offer a general sense of what is relevant
  and will allow us to approach flexibility in a
  shifting, empirical world to 'feel out' and 'pick
  one's way in an unknown terrain. ... In sum, the
  on-going refinement, formulation, and
  communication of sensitizing concepts must
  inevitably be the building block of our
  exploratory theory."

24
Interpretative approaches

• Approaches located towards this pole are better
  able to address problems in the "what we don't
  know that we don't know" as opposed to the "what
  we know that we dont know" category.
• This arguably entails relatively divergent
  thought in comparison with so-called "scientific"
  approaches.

25
(No Transcript)
26
Differences

• Understanding
• Its nature what it is you are explaining
• Its generalisability

27
Differences

• Evidence
• Its nature
• Its qualities
• validity reliability neutrality
• Procedures for obtaining it
• sampling data collection
  data analysis

Well discuss these separately in relation to
positivist and interpretative research later
28
Interpretative approaches

• Relatively holistic understanding of a large
  number of complex interactions amongst possibly
  fuzzy aspects of a low number of cases with
  inclusion of rich context.
• Generalisable in the sense of representing deep
  understanding of a complex whole enabling us to
  understand other cases where similar features and
  conditions apply.
• Evidence based on the internal coherence and
  plausibility (perceived explanatory power) of
  their findings.

29
Positivist approaches

• Relatively narrow understanding of the incidence
  of simple common denominator relationships
  between a small number of well defined isolated
  variables in large numbers of cases, entailing
  minimal extraneous context.
• Generalisable in the sense of being statistically
  predictable to occur in the wider population of
  which your cases were a representative sample.
• Evidence based essentially on statistical
  probabilities.

30
To summarise...
31
Limitations

• Positivist studies may often suffer from a
  relative lack of ecological validity, due to the
  complexity they are often forced to eschew in the
  interests of experimental control and precise
  numerical measurability.

32
Limitations

• Olaisen 1991, p. 260, for example, attributes
  fragmentation to a paradigmatic imbalance
• "... information science thought has been
  imprisoned by the dominant quantitative empirical
  metaphors which have drawn the attention to some
  quantitative phenomena while neglecting other
  more qualitative phenomena. The result is a
  cumulation of trivial findings."

33
Limitations

• Interpretative studies tend to offer a relatively
  subjective type of evidential support, due often
  to smaller samples and the preservation of
  complexity resulting in data which is not
  particularly susceptible to objective measurement
  and statistical testing.

34
Limitations

• Ford 1999, p. 1151 has drawn attention to the
  dilemma posed by these differences
• "Overly-analytic states of knowledge are
  characterised by fragmentation - at worst,
  isolated facts lacking integration into any
  coherent wider conceptual picture... As a result,
  much research in information science has arguably
  provided highly reliable answers to highly
  meaningless questions.

35
Limitations

• The take-up of qualitative research approaches is
  now widespread in user-oriented research. But
  without critical interaction with complementary
  perspectives the increasing use of subjective
  analysis of introspections using small samples of
  information users threatens to supply highly
  meaningful questions with highly unreliable
  answers. Some balance and integration must be
  achieved between the two extremes.

36
Limitations

• The limitations associated with research may be
  thought of as a curtain preventing us from
  viewing the reality beyond, that we seek to
  understand. Our existing knowledge ranges between
  two extremes, which to some extent mirror the
  different research approaches discussed above.

37
Limitations

• One may be characterised as scattered pinpricks
  in the curtain, allowing clear and deep, but
  narrow and unconnected views through to the
  reality beyond.
• The other may be characterised as more extensive
  areas where the curtain is thinned, allowing
  complex, inter-connected but hazy shapes to show
  through, inviting us to trace them onto the
  curtain, elaborating their detail to represent
  what we imagine to be their reality.

38
Limitations

• It is arguably all too easy to ignore the extent
  of curtain and our consequent lack of clear view,
  due to a variety of (sometimes unconscious) forms
  of over-optimism as to how securely and how
  widely we can generalise research findings to
  form a reliable and relevant picture.

39
Forms of over-optimism include

• falsely equating constrained research contexts
  with more complex reality is a form of
  over-generalisation most often associated with
  quantitative research in which experimental
  control and accurate measurement of variables are
  paramount.

40
The danger is in...

• assuming that what is observed in experimental
  conditions also applies in non-experimental, more
  natural conditions (much positivist research)
• and/or confusing internal coherence and
  plausibility with generalisability (much
  interpretative research).

41
Different views...

• Sometimes a broad holistic view can paint a
  picture...

42
(No Transcript)
43
Different views...

• Sometimes a broad holistic view can paint a
  picture...
• that does not stand up to more precise objective
  measurement and calculation

44
(No Transcript)
45
(No Transcript)
46
(No Transcript)
47
Different views...

• Sometimes a narrow analytic view of parts of a
  whole paint a clear and logical picture...

48
(No Transcript)
49
Different views...

• Sometimes a narrow analytic view of parts of a
  whole paint a clear and logical picture...
• that requires a different interpretation when a
  broader more holistic view is taken

50
(No Transcript)
51
(No Transcript)
52
(No Transcript)
53
Problems

• Sometimes it is difficult to see things from
  different perspectives at the same time...

54
(No Transcript)
55
(No Transcript)
56
(No Transcript)
57
(No Transcript)
58
(No Transcript)
59
Problems

• As Marton and Svensson 1979, p. 484 note in
  relation to research approaches
• "What we can see from one point of view we cannot
  see from another. . . . With one kind of
  observation certain aspects become visible with
  another kind of observation we see something
  else. We cannot arrive at a procedure of
  observation which makes all the various aspects
  visible simultaneously."

60
Paradigm wars

• Entwistle and Hounsell 1979 363 note that
• "As each paradigm marks out boundaries and
  establishes its own rules of discourse, there is
  a danger that territorial advantage will be
  sought through confrontation rather than mutual
  understanding-and the outcome of a pitched battle
  is more likely to be schism than synthesis."

61
Paradigm wars

• Burrell and Morgan 1979 397-8 propose that, in
  the interests of self-preservation and to avoid
  emasculation
• "Contrary to the widely held belief that
  synthesis and mediation between paradigms is what
  is required, we argue that the real need is for
  paradigmatic closure."

62
Paradigm wars

• Bradley and Sutton 1993 407 - are of the
  opinion that the conflicts are artificial
• "The paradigm debate has, in some senses, created
  an artificial polarisation based on abstractions
  that can easily harden into misunderstanding,
  caricature, and an attitude of superiority on
  both sides."

63
Integration of approaches

• Entwistle and Hounsell 1979, p. 361 note
• "The two paradigms ... contain the tension of
  opposites - a thesis and antithesis out of which
  a fruitful synthesis might be anticipated, but is
  still far from being achieved ... Yet the
  methodologies of competing paradigms could be
  used alongside one another, each providing
  distinctive yet equally valid types of evidence."

64
Complementary strengths
When we build a bridge, we need both what might
be thought of as the relatively narrow strength
of individual components e.g. the ability of
individual stones to fit with others and take the
compacted forces, and relatively broad holistic
strength, deriving from more complex
multi-directional interactions between structural
components.
65
Complementary strengths
Whilst the individual components may be
constructed in isolation from one another, the
placing of the horizontal stones at the apex of
the bridge requires a more holistic approach
requiring interaction between multiple components.
66
Strategies for methodological pluralism include

• the mapping of different paradigms onto different
  kinds of problem
• the use of different paradigms within a single
  study
• critical dialogue relating to a common phenomenon
  from different paradigmatic perspectives.

67
Vertical horizontal strength

• Commitment to particular research approaches may
  often centre on decisions as to how deep is "deep
  enough". As Popper 1968, p. 111 noted
• "... if we stop driving the piles deeper, it is
  not because we have reached firm ground. We
  simply stop when we are satisfied that the piles
  are firm enough to carry the structure, at least
  for the time being."

68

• Triangulation
• For some phenomena, combining narrow analytic and
  broad holistic perspectives may enhance and
  strengthen understanding
• like shining lights on the same phenomenon from
  different angles to give a clearer picture

69

• Triangulation
• Other phenomena, may only be truly understandable
  from one perspective and not be susceptible to
  the triangulation of approaches.
• Alternatively, different approaches may lead to
  very different types of understanding.

70
References

• Bradley, J. Sutton, B. (1993). Reframing the
  paradigm debate. Library Quarterly, 63 (4),
  405-409.
• Burrell, G. Morgan, G. (1979). Sociological
  paradigms and organisational analysis. London
  Heinemann.
• Ellis, D. (1996). The dilemma of measurement in
  information retrieval research. Journal of the
  American Society for Information Science, 47 (1),
  23-36.
• Entwistle, N.J. Hounsell, D. (1979). Student
  learning in its natural setting. Higher
  Education, 8, 359-363.
• Ford, N. The growth of understanding in
  information science towards a developmental
  model. Journal of the American Society for
  Information Science, 50(12), 1999, 1141-1152.

71
References

• Marton, F. Svensson, L. (1979). Conceptions of
  research in student learning. Higher Education,
  8, 471-486.
• Olaisen, J. (1991). Pluralism or positivistic
  trivialism important trends in contemporary
  philosophy of science. In H.E. Nissen, H.K. Klein
  R. Hirschheim (Eds.). Information systems
  research contemporary approaches and emergent
  traditions. Amsterdam Elsevier. pp. 235-265.
• Popper, K. (1968) The logic of scientific
  discovery. New York Harper and Row.

72
(No Transcript)
73
(No Transcript)