Copyright 2013 1 The Supervisor's Dilemma Roger Clarke Xamax Consultancy, Canberra Visiting Professor in Computer Science, ANU, Canberra Visiting Professor.

Copyright2013

1

The Supervisor's Dilemma

Roger ClarkeXamax Consultancy, Canberra

Visiting Professor in Computer Science, ANU, CanberraVisiting Professor in Cyberspace Law & Policy, UNSW, Sydney

http://www.rogerclarke.com/SOS/CAHI {.html, .ppt}

26th Bled eConference – 10 June 2013

Copyright2013

2

The Supervisor's DilemmaAgenda

1. General Statement of the Dilemma

2. eInteraction Research Realities

3. Specific Dimensions of the Dilemma

4. Resolution? Reconciliation? Avoidance?

Copyright2013

3

Dilemma

A condition in which the alternatives to be chosen among

are equally unfavourable

resulting in a position of doubt or perplexity

Copyright2013

4

The Supervisor's Dilemma

TheCandidate's

vs. Needs

TheSupervisor

'sIntegrity

Is Reconciliation Possible?

Copyright2013

5

The Supervisor's Dilemma

Contemporary disciplinary norms lead the candidate to adopt a

a research perspective, a theoretical basis for research,

a research design and/or rigour-relevance balance

that may conflict with the supervisor's

academic and professional values and personal integrity

Copyright2013

6

2. Some Research Realities

The Focus of the Bled eConference The eInteraction Research Domain

• Inter-Entity communications facilitated by electronic tools

• Entities may be human or machine

• Purpose is not limited, e.g. economic, social, hedonism

• It is irrelevant whether the communications occur within an organisational context or otherwise

• It encompasses technologies, technologies-in-use, and the politics of technologies and their uses

Copyright2013

7

Key Characteristics ofthe eInteraction Research Domain

• Its Immaturity• The Instability of the Phenomena• Driven by Marketing Hype,

Misinformation• Innovators' 'Casino Mentality'• The Vast Numbers of Confounding

Variables• The Interconnectedness among

Phenomena• The Inadequacies of Available Theories

Copyright2013

8

Approaches to Research

• Pure Research (or 'Basic' Research)'because it’s there'

• Applied ResearchTheory-driven or tool-driven 'normal science'

Copyright2013

9

Approaches to Research

• Pure Research (or 'Basic' Research)'because it’s there'

• Applied ResearchTheory-driven or tool-driven 'normal science' 'With a hammer in your hand, everything looks like a nail'

• Instrumentalist ResearchProblem-driven'I need to firmly connect two pieces of timber, so find out how to do it'

Copyright2013

10

Practical Realities• IS and the eInteraction domain are not 'industrialised'

• Tightly-focussed training in 'normal science' isn't appropriate

• Instead, broad background needs to be developed in:

• the philosophy of research• the full range of research techniques

• Much time and effort are invested in the learning process

• Considerable formalism is involved in satisfying examiners

• The dissertation must make a sufficient 'contribution'

• So the topic is very tightly focussed by the time of submission

Copyright2013

11

The Candidate's Motivations

• Attain education and training relevant to the conduct of research

• Make a contribution to a discipline and/or a research domain

• Gain access to the research profession• Gain a prenominal (Dr) or postnominal (PhD)

as evidence of advanced university education• ...

Copyright2013

12

The Supervisor's Motivations

Develop the next generation of researchers, for reasons ofservice / altruism or deferred and indirect reciprocity

Gain intellectual stimulation

Have a junior colleague undertake work closely related toand contributing to the supervisor's own research

Contribute to the supervisor's broader research program

Project the supervisor's theories or methods to the world

Have a research assistant

Satisfy an expectation of the supervisor's employer or peers

Copyright2013

13

3. Specific Dimensions of the Dilemma

• Researcher Perspective• Theoretical Basis• Empirical Research

Design• Rigour-Relevance

Balance

Copyright2013

14

Unit of Study / Unit of Analysis

A technology, an event, an object, a person, a group, an organisational sub-unit, a legally-defined organisation, a relationship between entities (e.g. a dyad), an aggregation of organisations (e.g. an industry sector), a region, a nation

Copyright2013

15

Unit of Study / Unit of Analysis

A technology, an event, an object, a person, a group, an organisational sub-unit, a legally-defined organisation, a relationship between entities (e.g. a dyad), an aggregation of organisations (e.g. an industry sector), a region, a nation

A technology, an event, an object, a person, a group, an organisational sub-unit, a legally-defined organisation, a relationship between entities (e.g. a dyad), an aggregation of organisations (e.g. an industry sector), a region, a nation

Perspective

Copyright2013

16

The Problem of the Singular Perspective

All observation has to be from a specific point of view

Stakeholders' perspectives are diverse

Stakeholders include:• Customers• Suppliers• Providers of similar goods and services• Providers of substitute goods or services• Employees and contractors• Usees• Regulators

The perspective adopted is usually that of one organisation

Other stakeholder interests are excluded, marginalised, or treated as a constraint rather than an objective

Copyright2013

17

Some Perspectives that are Under-Represented in eInteraction

Research• Individual Employees and Contractors• Usees

Affected individuals (e.g. suppliers, customers, licensees,

other people whose data are handled by an organisation, people adjacent to an organisation's facilities)

• Industry Sectors and Industry Segments• Economies• Societies• Humanity • The Biosphere

Copyright2013

18

Some Indicators of Under-Represented Perspectives

'Public Policy'The AIS Library – the term appears in 18 of 16,000 Abstracts The Bled Proceedings (2001-2013) – in 0 of 600 Abstracts

'Privacy'The AIS Library – appears in 360 of 16,000 Abstracts But from the perspectives of marketers, not of people

'Green' TopicsThe AIS Library – 9 in 1998-2008, 19, 20, 28 and 34 in 2009-12But almost all adopt the corporate viewVery few adopt the perspective of humanity or the biosphere

Copyright2013

19

Alternative Perspectives ==>> Additional Topics

skills formation; employment and other income-distribution mechanisms; the differential needs of, and impacts on, people and communitities in urban, regional, rural and remote areas; impact and implications for cultural and linguistic minorities; technology design for people with physical and cognitive impairments; impacts of data surveillance on users and usees; consumer needs and consumer rights; and consumer marketing philosophies other than conventional, predatory, mass-marketing approaches, e.g. permission-based, consensual, micro-, 1-to-1, and non-mass

Copyright2013

20

The Theoretical Basis for the Research

'Convenience Theories'or 'Fashion Theories'

• Cultural Factors TheoryHofstede 1980; but see McSweeney 2002

• Innovation Diffusion Theory Rogers 1962; but see McMaster & Wastell 2005

• Transaction Cost TheoryWilliamson 1981; but see Ghoshal & Moran 1996

• Technology Acceptance Model (TAM)Davis 1989; but see Benbasat & Barki 2007, Bagozzi 2007

Copyright2013

21

Weak Empirical Research Designs - 1

Cases• Semi-structured interviews aren't case studies• Neither are shallow cases / vignettes• Most cases are unauditable (the source is

undeclared even to examiners / reviewers let alone readers)

Surveys• Of what people say they do, not what they

actually do• Failure to complement with in-depth information• Proxy populations and sampling frames (esp.

students)• Over-surveyed populations, minuscule response

rates

Copyright2013

22

Weak Empirical Research Designs – 2

Convenience Samples'snowball sampling'open invitations to fill in a web-form

Convenience Data• long lists of highly ambiguous questions• forced choice• 5 or 7 highly ambiguous points

• highly ambiguous scales• assumption that the data is on a ratio scale

'Likert scales' are a contrivanceThey produce misleading and even dangerous results

Copyright2013

23

Rigour and IntellectualismDominate Relevance

• Clearly, cet. par., more rigour is to be preferred to less• However, all other things are seldom equal• The eInteraction domain creates enormous challenges

(Slide 6)• High levels of rigour make demands of researchers

that conflict with the need for the results to be useful• A trade-off has to be achieved between rigour and

relevance• But, instead, topics are favoured for their

researchability rather than for the value that they can deliver

Copyright2013

24

TensionsThe Candidate's Interest in the

Mainstreamvs. The Supervisor's Professional

Integrity

• Researcher Perspective• Theoretical Basis• Empirical Research

Design• Rigour-Relevance Balance

Copyright2013

25

4. Conclusions

• Resolution ?• Reconciliation

?• Avoidance ?• …............... ?

Copyright2013

26

'Design Science' + 'Critical Theory'?

• Design Science tries to draw the researcher back into contact with the needs of the real world, ground it, and deny abstract, intellectualist social science

• Critical Theory recognises the political dimension of IT applications – the interests and agendas of the players that exercise institutional and market power, and the tensions and the balances among competing forces in influencing the design of technologies and their applications

Copyright2013

27

The Supervisor's DilemmaAgenda

1. General Statement of the Dilemma

2. eInteraction Research Realities

3. Specific Dimensions of the Dilemma

4. Resolution? Reconciliation? Avoidance?

Copyright2013

28

The Supervisor's Dilemma

Roger ClarkeXamax Consultancy, Canberra

Visiting Professor in Computer Science, ANU, CanberraVisiting Professor in Cyberspace Law & Policy, UNSW, Sydney

http://www.rogerclarke.com/SOS/CAHI {.html, .ppt}

26th Bled eConference10 June 2013