ISSRM_Report_Public Innovations in Method of Research

ESRC National Centre for Research Methods
Innovations in Social Science Research Methods1
Maria Xenitidou & Nigel Gilbert
University of Surrey
May 2009
1 DO NOT CITE WITHOUT PRIOR PERMISSION FROM THE AUTHORS

ACKNOWLEDGEMENTS
This work was carried out with the co-operation of the NCRM Hub which assisted us in identifying
research participants. The project, however, was undertaken solely by the authors and they are
responsible for the findings presented hereafter.
We would like to thank the people who responded to our call, participated in our survey and
interview sessions and provided information or fora for networking and disseminating the project.
2
EXECUTIVE SUMMARY
One of the aims of the National Centre for Research Methods (NCRM) is to identify and foster
methodological innovation in the UK. The aim of this project was to identify methodological
innovations outside the UK and draw NCRM’s attention to them. The project sought out research
practices that have not yet filtered through to typical research methods courses or that impact on the
research process in novel ways. These usually entailed (i) technological innovation, (ii) the use of
existing theoretical approaches and methods in new ways and (iii) interdisciplinarity.
The project’s focus on innovative research practices ranged from data collection to analysis and
covered disciplines such as (social) psychology, sociology, social work, socio-legal studies,
political science (including public health and public policy) and international studies, (social)
geography (area studies, demography, environmental and urban planning), (social) anthropology,
(socio-)linguistics, education, communication studies, economic and social history, economics
(management and business studies), science and technology studies, statistics, methods and
computing.
The work was conducted between October 2008 and March 2009 and written up in April and May
2009. The project gathered evidence by reviewing previous reports, carrying out desktop research,
conducting an e-mail survey with academics, practitioners, research methods experts and others
(N=215) - registering data entries in the form of nominations of experts, institutions and links to
explore (N=670) - and holding interviews with gatekeepers (N=36) and telephone interviews with
nominated experts (N=40).
The project concluded, firstly, that innovative methodologies usually entail the use of one or more
technological innovation(s) (visual, digital or online). This could be the advent of new software or
the development of online methods and the use of the Internet to conduct research. Secondly,
innovative methodologies often entail crossing disciplinary boundaries. This is observed in
combinations of disciplines and methods such as in ethnography, anthropology and psychology.
Thirdly, innovative methodologies often entail the use of existing theoretical approaches and
methods in reformed or mixed and applied ways. This is observed in participatory methods, action
research, professional work, social and consultancy work. Finally, innovative methodologies reside
both inside traditional academic institutions (universities) and outside (research centres, institutes,
consultancy agencies and organisations), yet even in the latter methods developers and experts
usually have academic backgrounds and previous or current affiliations, status or posts.
Overall, psychology figured prominently in methodological innovations and developments followed
by survey methodology, ethnography, sociology and management. These developments were
classified into mixed (N=8), qualitative (N=7) and quantitative (N=7) types of research. The
institutional structures identified as ‘hosting’ these developments are primarily Academic followed
by both Academic and Professional, then Research Centres and finally Professional and
Consultancy institutions. The majority of the innovations are a consequence of working across
disciplinary boundaries, followed by developments within methods and disciplines and then by
developments in technology. Innovations were mainly spotted in North America – the USA and
Canada – Italy, Germany and the Netherlands.
The report includes summary descriptions of the methodological innovations located by the project.
As a follow up to this project a workshop will be organised to bring together some of the developers
and experts identified of these innovations. The workshop is planned to be adjacent to the NCRM
Research Methods Festival to be held in July 2010.
3
CONTENTS
Acknowledgements .................................................................................................................2
Executive Summary.................................................................................................................. 3
1. Introduction......................................................................................................................... 6
2. Background to the Project .................................................................................... 8
3. Methodology ...................................................................................................................... 10
Data Collection.................................................................................................................................10
Analysis ............................................................................................................................................10
4. Summaries of innovative methods .................................................................... 11
Method Descriptions.........................................................................................................................11
Technological Innovation..........................................................................................................................12
Managing Historical Longitudinal Data: Intermediate Data Structure ......................................................13
WeSearch ..............................................................................................................................................15
Virtual Agents (Avatars) .........................................................................................................................17
Internet Survey Software Tools: Human-Survey Interaction Framework...................................................19
Methods for Achieving Equivalence of Samples in Cross-National Surveys: The European Social Survey
Experience .............................................................................................................................................21
Item Response Theory – Tests without Items...........................................................................................23
Orbital decomposition ............................................................................................................................24
Crossing Boundaries .................................................................................................................................26
Folkbiology............................................................................................................................................27
Ethnography as a Non-linear Dynamic System........................................................................................29
Institutional Ethnography .......................................................................................................................32
Water Cooler Logic: Participatory Ethnography at Work.........................................................................34
Genetic Transmission of Political Orientations ........................................................................................36
Kinship Algebra Expert System (KAES)..................................................................................................39
Netnography ..........................................................................................................................................42
Journalism as both Reporting and Research ............................................................................................46
Appreciative Inquiry as a Research Method.............................................................................................48
Reflecting Interview................................................................................................................................50
Intuitive Inquiry .....................................................................................................................................51
Situational Analysis................................................................................................................................55
Towards a New Approach in Social Simulations: Meta-Language............................................................57
The Research Poem (in Social Work) ......................................................................................................60
Research-based drama ............................................................................................................................63
Transformative ........................................................................................................................................66
The impact of multi-sited ethnography on researcher involvement............................................................67
Participatory Learning and Action ..........................................................................................................68
Narrating Sensitive Topics: Applied and Basic Research ..........................................................................69
Software ..................................................................................................................................................71
Intelligence-Led Policing Software ..........................................................................................................72
Dialectometry: Measures Structure in Geographical Dialect Networks: Rug/Lo4 Software.......................73
HyperResearch: Computer-assisted programme for analysing qualitative data ..........................................74
HyperTranscribe: transcription tool.........................................................................................................74
Personal Network Analysis & EgoNet Software ......................................................................................75
Verbal Autopsy Software ........................................................................................................................77
Others .....................................................................................................................................................79
Facility for online data collection: Advanced Multi-disciplinary Facility for Measurement and
Experimentation in the Social Sciences (MESS).......................................................................................80
Integration and Implementation Sciences (I2S) ........................................................................................81
Critical Demography ..............................................................................................................................83
4
5. Discussion and Reflection ...................................................................................... 84
6. Conclusion ........................................................................................................................... 94
7. References...............................................................................................................................95
5
1. INTRODUCTION
The ESRC National Centre for Research Methods (NCRM) was established in April 2004 to
enhance the range and quality of research methods used by the social science community in the UK.
A main objective of the NCRM is to advance methodological understandings and practice by
developing a research agenda that attends to methodological innovation and assesses the training
needs created in the UK social science and professional social research communities.
After pursuing these goals in the UK for the past five years, the Centre reached a point at which an
outward-looking approach was considered timely. In light of this, a collaborative project was
undertaken in October 2008 with the aim of constructing a ‘map’ of the ‘hot spots’ of research
methods around the world. This map had two aspects: first, to locate the innovations in method
currently being developed within social science disciplines; and second, to locate the sites and
institutional contexts of these innovations, that is, who and where they are being developed. In so
doing, it aspired to situate the National Centre among its peers and highlight the people, places, and
ideas outside the UK to which it should pay attention.
The objectives of the project were:
• To identify the main developments in social science research methods (SSRM) and the most
significant growth points occurring outside the UK;
• To survey the different structures and processes that support research on SSRM in some of
the major countries of the world;
• To map concentrations of SSRM expertise and activity;
• To prepare and disseminate a report and to propose a plan for an international workshop.
The overall focus on innovative methodologies is inspired by the growing interest in identifying
who and what acts as the ‘impetus’ for innovation in social science research methods. Innovative
methodologies do not occur within a vacuum and context affects their diffusion. Therefore, the
project was expected to reflect on the underlying processes leading to innovative methodologies as
well as the processes, mechanisms and networks involved in their diffusion.
As authors in the research methods literature have noted, defining ‘innovation’ is not a simple task
as the concept itself is ambiguous and attention should be paid to the substantive area of innovation
(see inter alia Taylor & Coffey, 2008). Standard dictionary definitions or definitions from the
knowledge transfer literature are not immediately transferable to social science research methods.
Recent attempts to identify forms of innovation in qualitative research for example classified
innovations into new designs or methods (including methods of data collection and analysis,
techniques and software, representation of research), new concepts (including methodological
concepts and frameworks) and new ways of doing research (including new applications and
crossing disciplines) (ibid).
In view of the above and in order to ‘tailor’ what is considered innovative in social science research
methods, innovative research practices in the social sciences were initially defined as those that
have not yet filtered through to typical research methods courses or that impact on the research
process in ways which are novel (inventions) or different to existing ones. In the course of the
project, however, an emerging definition was formed influenced by the data. This defined
methodological developments and innovations in three broad ways: driven by advances in
technology (Technological Innovation); driven by interdisciplinarity (Crossing Boundaries); and
owing to intra-disciplinary or intra-methodological developments and extensions (Intra-disciplinary
Innovation). This emerging definition is in line with recent literature on emergent methods (see
inter alia Hesse-Biber & Leavy, 2008; Forbes, 2003). By the end of the project therefore, we came
6
to define innovative research practices as those which involve technological innovation, cross
disciplinary boundaries and / or extend existing methodologies and methods. In addition, we
adopted an inclusive approach with regards to whether the methods and methodologies identified
and included in this report have been applied, exploited and diffused (cf. Taylor & Coffey, 2008).
Instead of treating application, exploitation and diffusion of a method or methodology as criteria for
claiming innovation – and thus as criteria of inclusion in this report – we discussed such issues
when they became relevant in the description of a method or methodology.
The project was also attentive to action, interventionist and transformative research as it was
observed that the methodological flexibility used outside academic research enabled novelty and
ingenuity in the collection, analysis or presentation of data. Within these broad categories of
methodological innovations further classifications applied. These referred to (new) tools,
techniques, methods, methodologies, approaches (see Figure 1) and others (including software,
dataset structures, panel structures and disciplines).
Approach
Methodolog
Method
Technique
Tool
Figure 1. Hierarchical classification of Research Methods Categories2
For the purposes of the project, the focus on innovative research practices included research design,
data collection, analysis and presentation and covered disciplines such as (social) psychology,
sociology, social work, socio-legal studies, political science (including public health and public
policy) and international studies, (social) geography (area studies, demography, environmental and
urban planning), (social) anthropology, (socio-)linguistics, education, communication studies,
economic and social history, economics (management and business studies), science and
technology studies, statistics, methods and computing.
This report presents the methods used for the project, the innovative methodologies identified, and
then a discussion, reflections and future plans.
2 See Blaxter, Hughes & Tight (2001); Hussey & Hussy (2003); Saunders, Lewis & Thornhill (2007).
7
2. BACKGROUND TO THE PROJECT
The project was informed by key literature in Social Science Research Methods (SSRM) (see inter
alia Biber-Hesse & Leavy, 2008; Gilbert, 2008; Creswell, 2009) in order to frame social sciences
and research methods. The main points taken on board were: first, Social Science Research
Methods are tools which both enable understanding social life but also interact with social life in
dynamic ways. For example, different social problems may require different methods but also
different methods illuminate different aspects of social life or ‘construct’ different perspectives on
social problems. Second, the distinction between qualitative and quantitative research appears
infertile in the discussion on methodological innovation – that the distinction is not absolute in
research practice in general has been addressed elsewhere (Gilbert, 2008; Creswell, 2009). Overall,
emergent methods identified in the project rendered this distinction irrelevant theoretically and
restrictive in practice and treated it as a continuum since the focus was on the ‘problem’ driving the
‘method’. This focus was also argued to account for interdisciplinarity (Biber-Hesse and Leavy,
2008). Thirdly, crossing disciplinary boundaries produces interdisciplinary models for conducting
research which involve the development or intersection of methods. The fourth point was that the
development or intersection of methods is not only a result of crossing disciplinary boundaries but
methodological ones as well as it often occurs as a result of the failure of ‘traditional’ methods
(ibid). Fifth, the distinction between academic and professional research in terms of the locus and
agents of innovative research practices seems also to be frequently transgressed as addressing social
problems may take place in ‘contact methodological zones’ (ibid) inside and/or outside traditional
methodological institutions (Universities). Overall, emergent methods identified as part of this
project rendered this distinction problematic on the grounds that understanding social life is not
necessarily institutionally bound (yet this came from experts with an academic background and
former or current academic affiliation).
One might conclude from this literature that one should look to disciplinary, methodological and
institutional transgressions to identify innovative research practices. Section 4 discusses this in light
of the results, which indicate that the first two transgressions – disciplinary and methodological –
are more explicit and acute than the others.
Methods may be difficult to pinpoint chronologically and to associate to particular individuals.
Methods may re-appear modified, under a new name or discipline. Unfortunately, it was beyond
the scope of this project to attempt a historical account of innovative methods. In addition, while
some methods may be more clearly associated with particular experts, it is difficult to pinpoint
methodological developments to individuals ignoring inspiration, sources, networks and colleagues
involved (see Morse, 2006). Nevertheless, institutional affiliations, inspiration, colleagues and
relevant research were sought for each of the methods presented. Finally, diffusion of innovative
methods seems to be influenced by factors such as specialisation and reluctance to engage into
‘new’ learning, skills and training, funding, academic and institutional support and networking. It is
not the purpose of this project to consider why some methods become core and others remain
marginal or if institutional, geographical and network context matter in this process. Yet, these
details are noted for each of the methods presented and are summarised at the end.
The project consulted past reports on SSRM expertise and innovation in the UK (see inter alia
Beissel-Durrant & Lang, 2004; Wiles, Durrant, De Broe & Powell, 2005; Bardsley & Wiles, 2006;
Wiles, Bardsley & Powell, 2008) in order to be informed by existing evidence in exploring effective
methods of gathering data about SSRM expertise and innovation outside the UK. The main items
taken on board were: the main recent methodological innovations and developments identified in
the UK are modelling techniques (panel data analysis, structural equation modelling, multivariate
8
data analysis, time series, social network analysis), quantitative techniques (Bayesian methods,
survey design and analysis, simulation methods, spatial analysis, Geographical Information Systems
(GIS) and STATA), and qualitative methods (discourse and narrative analysis, analysis of
composite data, qualitative data analysis software). Innovative data collection methods cited are
web-based, computer-assisted, audio and video technology, while approaches cited are meta-
analysis, action research, evaluation research and evidence-based policy and practice (see Beissel-
Durrant & Lang, 2004; Bardsley & Wiles, 2006). These data and the NCRM actions taken towards
clustering research and training around them have been the starting points of contact for this project
(see Section 3 below).
9
3. METHODOLOGY
The project has been guided by a mixed methods approach. It started with a triangulated design
(interviews and email survey) to identify growth points and participants. It then moved to an
explanatory design (in-depth interviews) to collect detailed data on innovative research practices.
Data Collection
Both secondary and primary data were located, collected, used and analysed. We conducted
extensive desk research looking at the Web of Science, Methodspace, Intute, search engines,
portals, wikis, blogs, indexes, directories, publishers, SSRM online handbooks and encyclopaedias,
journals, archives, associations, teaching material, university websites, research centres,
government websites, commercial websites, business/professional networks, and events, such as
conferences, workshops, and their proceedings. This enabled the compilation of lists of contacts and
links through the NCRM Networks of Methodological Innovation (NMIs) (including their
associations), SSRM publications (contributors etc.), international networks and associations.
In terms of primary data, two sets of interviews and one e-mail survey were conducted. The first set
of interviews was either face-to-face or conducted by email depending on the availability of the
respondent. The target population for these interviews, identified through the NCRM mailing lists,
was NCRM Representatives and Methods Gatekeepers3 (N=36). These interviews were conducted
concurrently with the email survey and both were premised on the snowball technique, asking
participants to nominate experts and links to explore for methodological developments. The email
survey was conducted with academics, practitioners, research methods students and experts
(N=215) and yielded nominations of further experts, institutions and links to explore (N=670). In
addition, the email text appeared as a Call at Interact, the newsletter of the Society for the
Advancement of Games and Simulations in Education and Training, in the Sustainable
Development Research Network mailing, and in the Social Research Association mailing and was
widely circulated. The second set of interviews was either carried out by telephone (using Skype) or
by email and was addressed to nominated Experts (N=40).
Analysis
Data collected through desk research were analysed for theme(s) and cross-referenced. Data from
the first set of interviews were analysed for content. The email survey data were processed by
applying a combination of thematic analysis, cross-referencing and analytic induction. Finally, the
second set of interviews were preliminarily coded using Atlas.ti into type of method, background,
description and explanation, application and applicability, diffusion and dissemination, associates
and networks.
3Methods Gatekeepers refers to pioneers of particular research methods or key links associated to particular research
methods in the UK.
10
4. SUMMARIES OF INNOVATIVE METHODS
Method Descriptions
In this section, each of the innovative methods that were located are summarised using a standard
format. The methods are classified into three broad categories: Technological Innovation; Crossing
Boundaries; and Intra-disciplinary/Intra-methodological. Technological innovation refers to
advances in technology (usually in the form of software and/or internet-related) that either
constitutes a method in itself or pushes the boundaries of a methodology in contributing to or
amending previous research practice. Crossing boundaries refers to methodological innovations that
develop as a result of addressing research questions beyond or across disciplinary boundaries. This
usually involves a combination of theoretical and methodological approaches. Intra-
disciplinary/Intra-methodological refers to developments in methodology that are observed within
disciplines or that extend a method or apply it in different ways.
These categories are then classified into subsections based on the methodological cluster or primary
discipline. Each summarised description includes the following information:
• Innovator’s Details: Name, Affiliation, Country
• Title of Innovative Research Practice
• Introduction/Background
• Description and Explanation of the Methodological Development
• Examples of Application (Published if available or short descriptions)
• Inspiration and Rationale
• Application and Applicability
• Colleagues and/or Relevant/Related Research
• Material (Usually in the form of References)
These categories were derived from those developed in coding the interviews conducted with
methods experts. It should be noted that summarised descriptions draw on material derived from the
interviews conducted as well as from innovators’ own work derived from text available in published
(papers, articles, book chapters, website information) and unpublished material (internal documents,
reports, proposals, conference presentations).
Towards the end of the section, transformative work (a different genre fostering ingenuity in the
research process) and software development (a method development or extension) are also
described.
11
Technological Innovation
This section presents innovative research practices driven by technological development. The areas
and disciplines these developed in are mainly history, information systems and communication
studies, survey methodology and statistics in psychology (psychometrics).
12
George Alter
Inter-university Consortium for Political and Social Research (ICPSR), University of Michigan, Ann Arbor, Michigan,
USA
Managing Historical Longitudinal Data: Intermediate Data Structure

Introduction
George Alter has been working with colleagues in Europe, Asia, and the U.S. to make historical
longitudinal data more easily available. These data have great value, because they show life
histories unfolding over time in a range of places and time periods. The challenge has been to make
data sets available to a wider community of researchers. Currently, specialized skills are required to
transform the data into rectangular arrays that can be used with standard statistical packages.
Standardisation in both format and in programming techniques will encourage many more
researchers to use these data.
Description and Explanation of the Method
The ‘method’ here refers to an Intermediate Data Structure (IDS) which is a new format based on
the entity-attribute-value model. In the new strategy databases will transfer their data into this
common format. Researchers will use open-source programs to extract data from the IDS for
statistical analysis. This process encourages standardization, transparency, and comparative
research and reduces duplicate effort.
The IDS requires two kinds of computer programmes: 1. Data transfer: Data must be reformatted
for transfer from the original database to the IDS. This is the responsibility of the data producer,
because each database is different. Transferring information from the source database into the IDS
format also implies the generation of descriptive metadata to document the source and construction
of all data. 2. Extraction. The extraction process moves data from the IDS into rectangular file
formats suitable for statistical packages such as SPSS. Since the requirements of every type of
analysis differ, many specialised extraction programmes are expected. All extraction programmes
will start with the IDS, and they will work on any dataset that includes the necessary attribute types.
Alter has written two model extraction programmes to illustrate this process. One creates birth
intervals for studying fertility, and the second counts co-resident siblings for studying the effects of
household composition.
Example of Method Application
Family reconstitutions, which are derived from parish registers of births, marriages, and deaths, can
be used to study fertility and mortality in the period before regular population censuses. These data
are available in some countries from the sixteenth century.
Inspiration and Rationale for developing the Method
The rationale has been to standardize the format of a complex form of data. In this way researchers
can make use of data that has been too difficult for most people to process. Ultimately, as data
becomes more accessible we get more science. One source of inspiration has been the success of the
Integrated Public Use Microdata Series (IPUMS; Ruggles et al. 2008) project in encouraging new
research with historical data. By providing data in a consistent and easy to use form, IPUMS has
generated thousands of studies with data that were already available in less user-friendly versions.
With this in mind, the Inter-university Consortium for Political and Social Research (ICPSR)
worked with the Historical Sample of the Netherlands at the International Institute of Social History
to launch the IDS project in 2003.
13
Application and Applicability
More than a dozen databases have been involved in planning, and several databases have begun
implementing the IDS. This promises to result in much more cross-national comparative research.
Beyond historical demography, this approach may be applied to contemporary data with similar
characteristics, such as data from the demographic surveillance sites operating in more than thirty
places in Africa and Asia.
Colleagues and/or Relevant Research
Kees Mandemakers, Historical Sample of the Netherlands, International Institute of Social History,
Amsterdam and Erasmus University Rotterdam
Myron Gutmann, Inter-university Consortium for Political and Social Research (ICPSR),
University of Michigan
References
Alter, G., K. Mandemakers, and M. Gutmann. Forthcoming. “Defining and Distributing
Longitudinal Historical Data in a General Way through an Intermediate Structure.” Historical
Social Research
14
Luca Rossi
University of Urbino, Urbino, Italy
WeSearch
Introduction
WeSearch ‘exploits’ user generated content as data source for sociological research. The research
team working on it were Professor Giovanni Boccia Artieri, Professor Laura Gemini, Dr. Luca
Rossi and Dr. Fabio Giglietto.
The method (software tool) has been developed in order to investigate media-related generational
discourses on the web, and has since been applied to investigate holiday destination preferences and
EU identity symbols. The WeSearch software methodology works only with blogs. It has been
articulated in four steps:
a) identification of generational products (or other depending on research topic): in this case the
developers used conventional methods conducting group interviews;
b) definition of query strings for online analysis searching Google blogs to obtain RSS feeds. The
outcome of this phase was a standard RSS feed generated by Google’s service, composed by every
retrieved blog entry containing selected keywords. The first problem that the research team
encountered was that sometimes blog services did not provide an RSS feed containing the full text
of the blog entry. To solve this problem it was necessary to develop a tool able to fetch the full text
content of every single article. The team chose to use the ‘Yahoo pipes’ services in order to develop
a small tool to carry out this work. The final output of this phase was an RSS feed containing the
full text of every single blog entry containing the selected keyword;
c) storage and retrieval of blog author’s information: The RSS feeds was stored in an incremental
way using software which was specifically designed by the research team. The web application is
able to store an RSS feed, sorting the stored information by authors. Another feature of this
software is the ability to retrieve biographical information about authors’ entries. Biographical
information (age, gender and location) is retrieved using a scraping technique specifically
developed for every blogging or sharing service. Currently there are 7 supported platforms:
Blogger, Flickr, I1 Cannocchiale, Libero, Splinder, Windows Live Space, You Tube;
d) content analysis of collected data using NVivo: All the collected information (both the posts and
the biographical data about authors) may be exported to textual files (.doc format). The exported
files are compatible with the NVivo7 (qualitative analysis software). Biographical information
about authors is exported in a comma separated data file which can be imported as an NVivo
casebook retaining the link between the biographical information and the text entries.
This may help produce qualitative research with large amount of data, yet no statistical
representativeness is sought (at this stage) and, thus, generalising on the basis of the results is
problematic.
Media Products, Tourism, EU identity
The developers were inspired by the fact that online conversations are searchable, which allows the
unfolding of a whole new set of research possibilities. The developers claim that while virtual
ethnography observes a specific online place (e.g a single newsgroup, a chat room, a single virtual
15
world, etc.) it seems to have difficulty when there is a need to observe how a specific discussion
develops on the Internet.
It can be applied on anything having RSS feed and any general topic making use of it.
Colleagues
Boccia Artieri, G, Gemini, L., Rossi, L. & Giglietto, F.
References
Boccia Artieri, G, Gemini, L., Rossi, L. & Giglietto, F. (2008) Analysing User Generated Content
for Social Science. Generational “we sense” in the Italian blogosphere. Paper presented at Fourth
International Conference on e-Social Science, University of Manchester, 18-20 June
16
Fred Conrad
University of Michigan, Ann Arbor, Michigan, USA
Virtual Agents (Avatars)

Introduction
The method was developed as part of an exploratory three-year project funded by the National
Science Foundation (NSF), US. The main developers were Fred Conrad and Michael Schober. The
developers evaluated the use of virtual interviewers (computer-animated agents in the user
interface) for asking sensitive questions and questions that may be complex and require
clarification.
The virtual interviewers were developed using motion capture techniques to animate 3D models of
human-like faces. The amount of motion in the faces was experimentally varied as a way to affect
the visual realism of the virtual interviewers:
In the study about complex questions and clarifications the authors used the ‘Wizard of Oz’
approach in which the experimenter (the wizard) simulates the behaviour of an intelligent computer
application in a laboratory setting, in this case playing video files at the appropriate time, e.g.,
offering a definition of an ambiguous term. Participants communicated with the virtual interviewer
(Derek) and Derek responded accordingly (with the help of the wizard). The main experimental
question was whether visual realism (amount of facial movement) or dialog capability (whether or
not Derek could clarify survey concepts or was restricted to giving only neutral probes like “let me
repeat the question”) had more impact on the accuracy of answers (data quality).
Dialogue capability seems to correlate positively with response accuracy, while visual realism did
not. However, a more realistic looking interviewer might elicit more requests for help and the
respondent is generally more socially engaged.
In the second study, the Virtual Interviewer is called ‘Victoria’. Victoria delivered spoken questions
and respondents answered by clicking. Victoria was autonomous in that there was no human
operator; this was facilitated by her lack of dialogue capability making it possible to play a single
video file in order to deliver each question. Just as with Derek, there were two versions of Victoria
varying in visual realism, one higher (more facial movement) and one lower (less facial movement).
Respondents answers with the two versions of Victoria were contrasted with their answers in a face-
to-face interview and recorder audio (ACASI). Three patterns emerged:
1) Face-based conditions (i.e. Face-to-face, the two versions of the virtual interviewer) look
similar to one another, but different from Audio Computer-Assisted Self-Interviewing; in
the face-based conditions respondents tended to report more socially desirable answers than
under ACASI.
2) Linear trends, where answers in the virtual interviewing conditions fall in between those
given in Face-to-face and Audio Computer-Assisted Self-Interviewing (ACASI);
3) High social cue conditions (i.e. Face-to-face and higher realism virtual interviewers) and
low social cue conditions (i.e. lower realism virtual interviewers and ACASI) look different
from one another.
17
Already in use in some non-survey areas, e.g. Higher education: AutoTutor project; Help facilities
on some web sites, e.g IKEA’s ‘Anna’; Gaming; Virtual environments like Second Life.
People are becoming reluctant to participate in survey interviews or drop out. A next generation of
interviewing systems is likely to make use of paradata from respondents during the interaction, as a
way of diagnosing when help might be needed so that the system can intervene appropriately to
collect high quality data. Respondents’ typing errors, changed answers, response times, speech
disfluencies, or facial expressions may be explored in the future to assess confidence in answers,
misinterpretation, or even nonresponse. In general highly interactive, multimodal systems may well
be the way of the future in survey interviewing and virtual interviewers are a central example.
Virtual-human community: Virtual Interviewers essentially render the distinction between self-
administered and interviewer-administered interviews obsolete as there appears to be a continuum
between the two if viewed from the perspectives of everyone involved: communication
technologists, survey interviewers and participants. In addition, agent-like interviewing systems
blur the distinction between computer-mediated communication (CMC) and human-computer
interaction (HCI).
In addition there is a great promise in using paradata to facilitate the virtual interviewer to the
respondents, e.g. increasing facial/head movement or adjusting speech if the respondent seems
distracted. In addition, the attributes of the virtual interviewer, such as race, gender and age can be
adjusted to match (or not) those of the respondents.
Justine Cassell, a linguist, psychologist and computer scientist is an early developer of the
technology (she uses the phrase Embodied Conversational Agents) and her collaboration with
survey researcher Peter Miller is reported in a chapter in the book Envisioning the Survey Interview
of the Future, edited by Conrad and Schober. They applied the method to the study of interviewer-
respondent.
Other Information
1. An extension of the project would be to allow respondents to choose their interviewer’s
appearance.
2. Having this kind of choice at the researcher’s disposal is novel, as are the ways in which
these may be combined, which complicates the key issue in surveys: standardisation.
References
Conrad, F. G. & Schober, M. F. (2008) New Frontiers in Standardized Survey Interviewing. In
Hesse-Biber, S. & Leavy, P. (2008) (Eds.) Handbook of Emergent Methods. New York: The
Guildford Press, pp. 173-188
Schober, M. F. & Conrad, F. G. (2007) Survey Interviews and New Communication Technologies.
In F. G. Conrad and M. F. Schober (Eds.) Envisioning the Survey Interview of the Future. New
York: John Wiley
18
Lars Kaczmirek
GESIS - Leibniz Institute for the Social Sciences, Mannheim, Germany
Human-Survey Interaction Framework and Internet Survey Software Tools

The research involved developing tools by focusing on the Human-Survey Interaction Framework -
how applied usability increases response rates in surveys, mainly addressing error tolerance and
feedback.
The research deals with the problem of nonresponse: Not all contacted persons start participating in
a survey (unit nonresponse), and not all respondents complete a survey (partial response) or even
answer all questions (item nonresponse). This introduces a bias when respondents differ from
nonrespondents since population parameters are estimated on the basis of respondents’ answers.
The general goal was to develop instruments and methodological designs with which respondents’
behaviour could be observed and nonresponse could be reduced. The theoretical background was
shaped by combining concepts from human-computer interaction and survey methodology into a
framework which Kaczmirek termed human-survey interaction.
In particular, the human part of the framework incorporated theories of the response process. The
satisficing/optimising-approach was employed as it mediates the data quality of responses. While
increasing response rates is a major goal, this should not lead to an increase in satisficing behaviour.
The survey part of the framework subsumed theories of questionnaire design. The response burden
which is imposed by the survey can be seen as one of the most important reasons for dropout. The
interaction part of the framework was concerned with visual changes in online surveys due to
respondents’ actions. A review of human-computer interaction and usability principles led to the
conclusion that principal aspects of successful interaction are error tolerance and feedback.
One of the main contributions of the study was that it developed a data box model for the concept of
paradata. Paradata are process data in surveys which can be collected while respondents answer a
questionnaire. Paradata provide insight into respondents’ behaviour such as response times, changes
in answers, and mouse clicks. In addition, paradata serve as performance indicators for surveys:
dropout patterns, item nonresponse, and response rates are key measures for the success of a survey.
Paradata is already widely used in survey research, yet they seem to lack a concise
conceptualisation and systematised use and collection. The data model developed (the universal
client-side paradata approach) distinguishes four levels of paradata in the dimensions respondents,
variables and time.
The first data level consists of single actions such as mouse clicks. The second level aggregates
over actions, resulting in content which is conceptually relevant for researchers, for example
interaction failure measured as the number of mouse clicks which missed an answer option. The
third level aggregates over variables to summarise measures such as the overall number of items
missing. The third level also includes aggregations over respondents, resulting in measures such as
the average response time on each page. The fourth level is the most abstract aggregation and
summarises across respondents, variables and time to yield a single measure such as the average
time to complete the survey or the response rate.
In exploring the role of feedback, the study developed a dynamic calculation method for calculating
progress feedback. The dynamic calculation method can be implemented with either a progressive
or a conservative estimation of progress, yet higher completion rates are observed with a
progressive algorithm.
19
The need identified was for greater formalisation of procedures and systematisation of techniques to
manage nonresponse in online surveys and form completion.

It is applied in all online surveys or form completion tasks and is not therefore bound to any
particular discipline.
In terms of inspiration, this work cites Mick Couper, Don Dillman, Michael Bosnjak, and Wolfgang
Bandilla as the pioneers in online survey research.
Other Information
The research indicated that further study of online survey methodology with the paradata box model
and the universal client-side paradata approach may prove insightful with regards to developing
methods to reduce nonresponse and generally understand respondent reaction and treatment of
online tasks.
The development of the human-survey interaction framework provides new insight in what makes
respondents leave or complete an online survey.
References
Kaczmirek, L. (2008) Human-Survey Interaction: Usability and Nonresponse in Online Surveys.
Mannheim: MADOC
Kaczmirek, L. (2008) Internet Survey Software Tools. In N. Fielding, R. Lee, & G. Blank (eds.)
The Handbook of Online Research Methods. London: SAGE, pp. 236-254
20
Sabine Haeder
GESIS - Leibniz Institute for the Social Sciences, Mannheim, Germany
Methods for Achieving Equivalence of Samples in Cross-National Surveys:

The European Social Survey Experience
Introduction
Cross-national sampling frames are rare whilst large-scale cross-national surveys are tantalised in
trying to ensure ‘equivalence’ of samples. The approach presented below is developed by four
sampling experts, Sabine Haeder, Peter Lynn, Siegfried Gabler, and Seppo Laaksonen. It is
presented as a ‘research process’ which may produce new insights in sampling for large cross-
cultural surveys.
The methodological approach refers to ways of ensuring sampling equivalence in cross-national
surveys. The experts propose two criteria as fundamental:
• The study population must be equivalent in each nation. In practice, this will usually mean
that the same population definition is applied in each nation and that no or only minimal
under-coverage can be permitted;
• Sample-based estimates must have known and appropriate precision in each nation. In
practice, “known” precision means that a strict probability sample design must be used, and
those aspects of sample design that affect precision (selection probabilities, stratum
membership, PSU membership) must be available in the microdata to permit estimation of
standard errors; “appropriate” precision may mean a) meeting some minimum precision
requirement in order for the estimates to be useful, and b) aiming for similar precision in
each nation, as this would represent an effective allocation of resources if a prime objective
was to make between-nation comparisons.
The goal is functional rather than numeric or stylistic equivalence.

The European Social Survey: Equivalence of sample designs in 22 nations in Round 1 of the
European Social Survey (ESS). The pre-sampling requirements were grouped into five categories:
population definition and coverage, response rates, sample selection methods, effective sample size
and documentation.
Defining the population takes into account cultural elements and the nature of the available
sampling frames in terms of coverage, update and access. In cases where lists were not available,
area sample designs (Särndal et al., 1992) were applied, in which the selection of a probability
sample of small geographical areas (e.g., Census enumeration areas within municipalities) preceded
a complete field enumeration of households or dwellings within the sampled areas, from which a
sample was selected. An alternative to area sampling is the application of random route sampling.
The basic idea of random route sampling is that within each sampled primary sampling unit one
address is selected by a random method to serve as a starting point and the interviewer then follows
rules that specify the route he or she should take from there, sampling systematically using a
prespecified interval (Häder and Gabler, 2003).
Response rates need to account for nonresponse, a problem a lot of recent literature on face-to-face,
telephone and online surveys taps into. It is deemed essential to plan for and implement strategies
which minimise nonresponse, including setting the response rate high, sending advance letters,
21
training of interviewers in response-maximisation techniques and in doorstep interactions, etc. In
terms of sample selection, strict probability sampling is a necessary prerequisite for crossnational
comparability. Calculating the effective sample size involves predicting, under certain simplifying
assumptions, the design effect due to unequal selection probabilities (DEFFp) and the design effect
due to clustering (DEFFc). In some nations, it is necessary to select the sample in stages, with the
penultimate stage being addresses or households. Another reason for unequal selection probabilities
is that minority groups are oversampled for substantive reasons.
Finally, as regards documentation, comprehensive and clear documentation of all relevant
methodological aspects of the survey was demanded. At the level of sampling units, this meant that
indicators of sampling stratum, primary sampling unit and the selection probability at each stage of
sampling should be included on a micro-level data file that carring the same identifiers as the
questionnaire and other data files. A detailed file specification was provided. Supply of these data
would allow the application of design weights and the use of appropriate methods for the analysis of
data from a complex survey.
Cross-national sampling frames are rare and the data collected are insufficient to ascertain sampling
effects.
All cases where sampling equivalence is a prerequisite.
Colleagues
Peter Lynn, Siegfried Gabler, and Seppo Laaksonen
References
Lynn, P., Haeder, S., Gabler, S., & Laaksonen, S. (2007) Methods for Achieving Equivalence of
Samples in Cross-National Surveys: The European Social Survey Experience. Journal of Official
Statistics, Vol. 23, No. 1, pp. 107–124
22
Susan Embretson
University of Minnesota, Minneapolis, Minnesota, USA
Item Response Theory – Tests without Items

Introduction
Susan Embretson has focused on modern psychometric methods (e.g., item response theory),
cognition and intelligence, and quantitative methods. Her research programme has been to integrate
cognitive theory into psychometric models and test design. Her current research directions include
developing item response theory models, the measurement of change, automatic item generation,
and validity issues in Internet testing. As part of this, she has been developing new item response
theory models and conducting empirical research on the cognitive basis of an individual’s
responses. Recently, this effort has led to the possibility of ‘tests without items’. The measurement
areas she has worked on have included fluid reasoning, spatial ability, mathematical reasoning and
verbal comprehension.
Susan Embretson’s work is included here for recently producing an extension of a method: more
targeted and efficient item development and little or no need for item tryout (i.e., automatic item
generation). Items are automatically generated by artificial intelligence to target levels and
cognitive sources of difficulty in order to measure each individual examinee during testing.
There are many practical advantages of automatic item generation for test development. In
automatic item generation, the development of appropriate psychometric models, the decomposition
underlying components, and the programming structure all lead to the realisation of valid test items.
Items are produced by an algorithm that yields items with predictable psychometric properties. The
algorithm can be based either on an item model (i.e., an existing item), such as used by Bejar, or on
underlying features of the item that characterize cognitive complexity, as in Embretson’s approach.
What was required was developing cognitive complexity models for existing items. In automatic
item generation, the algorithm is embedded in a computer programme that can produce many
variants from a single specific structure.
The primary goal was to develop a dynamic test to measure the susceptibility of reasoning to stress,
using matrix completion items. Embretson was seeking a new application of an IRT model for
learning and change that was published in Psychometrika (1991).
The IRT models may be applied to measure personality traits, moods, behavioural dispositions,
situational evaluations and attitudes as well as cognitive traits.
Colleagues
Isaac Bejar, also recognised for his similar research in this area, although from a somewhat different
perspective.
References
Embretson, S. E. (2004) Measuring human intelligence with artificial intelligence: Adaptive item
generation. In R. J. Sternberg & J. Pretz (Eds.). Cognition and Intelligence. New York: Cambridge
University Press
23
Stephen Guastello
Marquette University, Milwaukee, Wisconsin, USA
Orbital decomposition
Introduction
Dr. Guastello’s main research interests concern nonlinear dynamics (chaos, complexity, and
catastrophe theories, and related principles) and their applications to work motivation, work
performance and turnover, occupational safety and health, creativity, group and organisational
processes. As part of working in those areas he has contributed to the development of nonlinear
methods. He is included in this report for the development of a technique for the analysis and
quantification of qualitative data called orbital decomposition.
Orbital decomposition falls within Symbolic Dynamics, a group of methods that ‘map’ a system in
a ‘sequence space’. Orbital decomposition is used to analyse behaviour strings (sequences of
behaviour). It combines different kinds of calculations, including a statistical analysis that tests how
well the results of the procedures fit the original pattern. “The procedure requires three calculations
in parallel: Shannon entropy (HS), topological entropy (HT), and a likelihood χ2 test for strings of
responses of varying length C. The calculations provide measures of dimensional complexity, the
determination of an optimum behaviour string length, a set of behaviour strings with associated
probabilities, and a χ2 test that provides a measure of fitness for the string structures” (Guastello,
2005, p. 266).
The process involves coding data either in metric or nominal terms, i.e. assigning a numeric or other
symbolic value. One must code each possible piece of an interaction, and then sequence the
interaction pieces into strings. For this process a coding scheme is required. Then, each event in a
sequence needs to be assigned a letter code. The algorithm finds:
1. Topological entropy - an analysis of observed codes and how they group. When the
sequences cease to be followed directly by themselves, denoting an entropic state or
breakdown of the pattern, the trace reaches 0, topological entropy has in effect stabilised,
and the last string length that has a value greater than 0 is considered to be the optimal string
length.
2. Shannon entropy traces the decomposing pattern.
3. The Lyapunov Exponent is a standard test of chaos, and converts to the fractal dimension of
the pattern. The goodness of fit between the data and quantitative observations are evaluated
using χ2 and φ2 computations.
4. The χ2 allows one to check expected frequencies of strings’ occurrences against actual
observations.
5. Finally, a φ2 procedure for all χ2 tests offers a statistical indication of the amount of variance
in the observed string lengths that were accounted for by the patterns that were isolated in
the analysis.
In other words, the algorithm finds the optimal string length, the Entropy statistics associated with it
and Lyapunov exponent and fractal dimension. It also runs a χ2 test to check for whether string
patterns occurred by chance and a φ2 test to calculate the degree of fit between patterns and the
original data.
24
Orbital decomposition has been used to analyse sequences of behaviour within the context of
problem-solving groups (Guastello, 2000; Guastello, Hyde & Odak, 1998) and therapy groups
(Pincus, 2001; Pincus & Guastello, 2005). Recently it has been used in Spohn’s (2008) study of
violent societies. This study used orbital decomposition to analyse the patterns of how governments
lose their monopolies on violence, therefore allowing those societies to descend into violent states
from which it is difficult to recover. Orbital decomposition turned up insights not apparent in the
qualitative data or through linear statistical analysis, both about the nature of the descent into
violence and about the progression itself. The process involved each recorded event being coded for
5 variables with several categories per variable. Each combination of categories was given a letter
code to start the algorithm.
Psychology
References
Guastello, S. J. (2005) Nonlinear methods for the social sciences. In S. Wheelan (Ed.), The
handbook of group research and practice. Thousand Oaks, CA: SAGE, pp. 251-272
Guastello, S. J. & Liebovitch, L. S. (in press) Introduction to nonlinear dynamics and complexity.
In S. J. Guastello, M. Koopmans, & D. Pincus (Eds.). Chaos and complexity in Psychology: Theory
of nonlinear dynamics. NY: Cambridge University Press
Spohn, M. (2008) Violent Societies: An Application of Orbital Decomposition to the Problem of
Human Violence. Nonlinear Dynamics, Psychology, and Life Sciences. Vol. 12, No. 1, pp. 87-115
25
Crossing Boundaries
The second section presents innovative research practices driven by the need to transgress
disciplinary boundaries in order to study social problems. The areas and disciplines these developed
in are mainly ethnography, anthropology, politics and professional/consultancy work.
26
Norbert Ross
Vanderbilt University, Nashville, Tennessee, USA
Folkbiology
Introduction
Norbert Ross currently acts as the director of the Culture, Language and Cognition Laboratory at
Vanderbilt University. Norbert Ross has been engaged in bridging cognitive science and
anthropology on the assumption that cognition brings ‘the mind’ into anthropology while
anthropology situates cognition in ‘the social’. His research practices to this end are
multidisciplinary and make use of methods developed and used across disciplines which is the
reason his work is included in this report.
Folkbiology refers to a multidisciplinary approach to the workings of culture in which thinking is
treated as an emerging product (what we think cannot be separated from how we think) —
combining anthropology with cognitive sciences. In this respect the focus is on how representations
are formed, transmitted and transformed through social factors (social and expertise networks and
learning strategies) and cognitive factors (mental representations through categorisation and
knowledge acquisition processes e.g. framework theories).
The research process itself is both ethnographic and experimental and involves the following: GIS
Based Spatial Analyses, Social Network Analyses, Cultural Consensus Modelling, Analysis of
Residual Agreement, Experimental psychological laboratory research, Computer Simulation of
micro and macro-processes. The focus of this process is on identifying how information is passed
on, involving social network analysis, complexity theory, and the modelling of cultural processes in
a situated context, building on GIS and agent-based modelling.
Former research concentrated on folkbiological knowledge. Current research focuses on Conceptual
and Cultural Change in Folkmedicine. This evolves in: Mapping of Representations / Social
Factors: (1) Kinship; (2) Friendship; (3) Physical Distance; (4) Expertise Networks; (5) Distance to
Health Facilities, etc. The distribution of representations gives an idea of the role of social factors
for conceptual and cultural change. This is then followed by anthropological field experiments and
psychology lab research asking such questions as: How do conceptual structures change? How does
information change conceptual structure and gets transformed in the process of communication?
Finally, an Agent Based Model of information exchange and cultural change is built: i. Agents act
within simulated space (GIS analysis); ii. Agents act within a simulated social network; iii. Agents
are endowed with an algorithm for conceptual change.
The inspiration for formulating this research process was driven by the contention that employing
anthropology and cognitive sciences in isolation produce unsatisfactory ‘science’.
Cognitive Anthropology, Folkmedicine, Religion, Knowledge Acquisition
Colleagues
The members of the Culture, Language and Cognition Laboratory are:
Postdocs: Jonathan Maupin, Catherine Timura
27
Graduate Students: Mike Tidwell, Jeff Shenton
Field Researchers: Alejandro Gomez Gutierrez, Julia Gomez Perez, Mariano Ruiz Ortiz
Co-investigators have been: Atran, S., Medin, D., Palmeri, T., Noelle, D.
References
Ross, N. (2008) Asking the Right Questions: Outline for a study programme in culture, language
and cognition. Vanderbilt University: Culture, Language and Cognition Laboratory
28
Michael Agar
University of Maryland; the International Institute for Qualitative Methodology; University of Alberta, Edmonton,
Canada; the University of Buenos Aires, Argentina; Ethknoworks LLC, Eldorado, New Mexico, USA
Ethnography as a Non-linear Dynamic System

Introduction
Mike Agar is Emeritus Professor at the University of Maryland and currently works mainly
independently as Ethnoworks, in New Mexico. Ethnoworks focuses on research, writing and
consultation in ethnography, language, complexity theory, and organisational analysis from both
theoretical and practical standpoints. A proponent of a continuum between academic and
professional work and between theory and practice, Mike Agar advocates that the study of dynamic
dimensions and series of relationships has started liberating social research since the 1980s and
1990s when a new research framework grew on the basis of ‘complexity’. The growth was driven
by a problem shared by researchers of all kinds and enabled by advances in computer technology,
since simulation is the only way to model and explore such systems. That enabled Agar, who had
been arguing for a while that ethnography is a kind of logic rather than a unit of study, to develop a
general framework for ethnographic research as a complex adaptive system (CAS).
He argues that “It’s less about a study of qualitative versus quantitative and more a study about
‘fitness’ which presents an interesting methodological twist away from methodological recipes and
towards ‘the right way to do ethnography’ through a non-linear dynamic lens.”
Mike Agar has developed a general framework to do ethnography based on the premise there is
epistemological compatibility between complexity theory and ethnography. He used the model in
both short-term applied and long-term research applications. The framework or model for
organisational development with social services involves the following steps:
• Clarify indicator-based problem – The initial problem is always defined in quantitative terms
e.g. patient waiting time, performance etc. In other words, the definition of the problem is
based on one or more system-level quantitative indicators that need to be clarified before
locating the ‘tasks’ these indicators are meant to measure.
• Locate actual tasks that those indicators are meant to measure – task in this case refers to
what the indicator measures - real moments that involve real people doing real things.
• Do fieldwork in those tasks using iterative recursive abductive (IRA) logic and
context/meaning (C/M) questions. Iterative recursive abductive logic calls for taking
surprises seriously and creating new concepts to account for them rather than consider them
deviant or ‘errors’, and of returning and repeating the process as surprises never stop.
Context/meaning questions refer to being attentive to the fact that there are multiple
perspectives and interpretations in play in any human social research and it is, thus,
important to take into account the ways in which members of various groups ‘translate’
meanings and situations.
• Analyse the data for patterns in an ethnographic way: a systematic process of identifying
patterns that replicate and whose dynamics explain the movement of the indicator. These are
the leverage points: leverage points refer to the point at which actions and changes can lead
to improvements.
• Do more fieldwork looking at task variation - where leverage points have already been
discovered and used, or are discussed or speculated about, on the part of those who perform
it. These are positive deviant cases: positive deviance refers to certain individuals’ practices
which enable them to find better solutions to community problems. Do more fieldwork with
29
the positive deviant cases to come up with suggestions for how to use leverage points to
solve the problem.
• Loop back to the top of this list, probably with better indicators, and try the new strategy out.
But encourage experimentation, variations and modifications on the part of those involved in
the task. This approach is related to transformative participatory action research on the
premises of appreciative inquiry, as well as Edgar Schein's concept of "clinical ethnography."
In a recent seven year long term research project funded by the National Institute of Health, he
applied this framework to the study of drug epidemics. Previous research (both practical and
theoretical) had failed to explain adequately and account for the processes, factors, actors and
‘ecology’ of the issues. Nonlinearity, researcher influence and path dependence were to be
eliminated. Studies had few dynamic dimensions and a lack of co-ordination of materials.
In contrast, the framework proposed is of multilevel, multivariable interaction, a series of
relationships as a general framework to do ethnography. These relationships include the
ethnographer – since ethnography is a construction that shows how social action in the context of
one world can be understood as coherent from the point of view of another. For the purposes of this
framework ethnographic study has to be done first: involving issues such as the structure-agency
relationship with representations based in reality and attentive to constraints such as
intersubjectivity, scaling, dynamics, pragmatics and narrative construction. The approach may
include conventional methods for data collection and analysis. In the case of drug epidemics for
example methods included: ethnographic work, interviews, investigative journalism, policy,
hearings; the analysis was based on identifying patterns and contradictions. The result was a kind of
economic history that included ethnographic data, was conducted using ethnographic epistemology,
and was cast in a nonlinear dynamic theoretical framework. Agent-based models were constructed
to show how the theory could explain epidemic incidence curves.
Drug epidemics, organisational intervention: state court and cancer treatment centre, ongoing work
on sociocultural process in ecology, cultural dynamics in computer-based language instruction, and
clinical performance in hospitals.
Inspired by applying complexity theory to address the methodological problem of ethnography of
distinguishing between credible and non-credible research.
General framework for ethnographic research – social research projects in an applied way to model
the dynamics of structure and agency and incorporate both macro and micro levels in the same
model.
Colleagues on current projects include the Information Sciences Institute at the University of
Southern California, the Long Term Ecological Research program of the National Science
Foundation, the Health Sciences Center at the University of Texas in San Antonio, and the
Department of Epidemiology at George Washington University.
References
Agar, M. (2004) We Have Met the Other and We’re All Nonlinear: Ethnography as a Nonlinear
Dynamic System. Complexity. Vol. 10, No. 2, pp. 16-24
30
Agar, M. (2008) On the Ethnographic Part of the Mix: A Multi-Genre Tale of the Field. Paper
Presented at the Qualitative Research in Management Conference at the University of New Mexico,
March 13, 2008, pp. 1-26, now in press in Organizational Research Methods.
31
Dorothy E. Smith
University of Toronto, Toronto, Ontario, Canada
Institutional Ethnography
Introduction
Institutional ethnography has been developed and named by the Canadian sociologist Dorothy E.
Smith (1987) in the early1980s; since then it has matured into a method of inquiry designed to
allow researchers to explore the social organisation of knowledge and its consequences in
contemporary society.
Institutional Ethnography is the label that has come to be used for an approach to investigation of
the social that focuses on ‘textually-mediated social organisation’ (Smith, 1990b).
Institutional ethnographies are built from the examination of work processes and study of how they
are coordinated, typically through texts and discourses of various sorts. Work activities are taken as
the fundamental grounding of social life, and an institutional ethnography generally takes some
particular experience (and associated work processes) as a “point of entry.” The investigator attends
to all of the work that’s done in the setting, and also notes which activities are recognised and
accounted institutionally and which are not. Analysis proceeds by way of tracing the social relations
people are drawn into through their work (with the term “social relations” taken in its Marxist sense
to mean not relationships but connections among work processes). The point is to show how people
in one place are aligning their activities with relevances produced elsewhere, in order to illuminate
the forces that shape experience at the point of entry.
Institutional ethnography’s focus on texts comes from an empirical observation—that technologies
of social control are increasingly and pervasively textual and discursive (Smith, 1999). Texts such
as medical charts, enrolment reports, strategic plans, and so on are mechanisms for coordinating
activity across many different sites. Treating the text ethnographically departs from discourse
analytic work in that it does not focus on how the text enters into the organisation of sequences of
action in multiple different sites. There is a possibility of trying to develop a sociology that looks at
society from the point of the people and their experience of it. Institutional ethnography picks up
this idea, to explore the institutional order and the ruling relations from the point of view of people
who are in various ways implicated in and participating in it.
In the U.S., Ellen Pence, at Praxis International (www.praxisinternational.org) has worked
extensively with advocates to reform domestic violence case processing. She has developed a
version of institutional ethnography that serves as the basis for a community audit process, designed
to centralise concerns for women’s safety and to produce recommendations for community
responses and case processing that will be more responsive to women’s and children’s needs.
In Canada, Marie Campbell (2000) led a participatory research project which involved a group of
disability activists in the examination of the organisation of home support services.
Institutional ethnographies of organisational work often focus on specific texts such as policy
documents (Eastwood, 2005; Ng, 1995; Stooke, 2003), funding proposals and planning documents
(Grahame, 1998; Turner, 2001), the accounting records of bureaucratic workplaces (McCoy, 1998;
Mykhalovskiy, 2001), or the charts and records of professional-client relations in health care, social
work, and educational settings (André-Bechely, 2005; Parada, 2002; Rankin, 2001).
32
Finally, Institutional ethnography has provided the basis for efforts across a wide range of
substantive concerns, including areas such as access to AIDS treatment (Mykhalovskiy and
McCoy), policy and environmental planning (Turner, 2001; and see www.rwmc.uoguelph.ca/),
and the racialised organisation of garment-industry work (Ng, 2000).
The method is designed to shift the focus of sociological research away from questions generated
by administrative concerns and on objectified knowledge and towards the puzzles of peoples’
everyday lives. Its critique of objectified knowledge and its use in the management of institutional
life suggests that, frequently, and in systematic ways, the categories and conceptual frameworks of
administration are inattentive to the actual circumstances of the diverse lives people live in
contemporary societies. Institutional ethnographic studies contribute to a social justice agenda by
making knowledge from the standpoints of people’s everyday lives, seeking to demystify relations
of ruling, and pointing to possible interventions in ruling relations.
Nursing, education, social work, planning, not only within academia; but also taught as a
professional skill to be used in political activism.
Colleagues
Marie Campbell, Marjorie De Vault
References
DeVault, M. L. (2006) Introduction: What is Institutional Ethnography? Social Problems. Vol. 53,
Issue 3, pp. 294–298
Smith, D. (2001); Text and the Ontology of Organisations and Institutions. Studies in Cultures,
Organisations and Societies. Vol. 7, pp 159-198
Smith, D. (2005) Institutional Ethnography: A Sociology for People. Toronto: AltaMira Press
33
Helga Wild & Chris Darrouzet
Water Cooler Logic, Palo Alto, California, USA
Water Cooler Logic: Participatory Ethnography at Work

Introduction
Helga Wild and Chris Darrouzet used to work for the Institute for Research on Learning (IRL) in
Menlo Park CA, which pioneered the use of ethnography in researching the social basis of learning
in schools and corporations. To this use of ethnography they joined techniques from design to bring
the lessons from research into praxis.
As Water Cooler Logic (WCL) Inc. they evolved the method further to involve the client
organization in a highly participatory form. When conducting their action research and learning
projects they train a cross-functional team of client representatives as para-ethnographers, explore
with them the organization's functional networks, and design with them new processes and
practices. Two steps are noteworthy in this innovative research method: the first refers to the use of
ethnographic methods for understanding the communities via exemplary situations; the second
refers to the participation of the studied communities, which activates residual social relations and
social capital.
Water Cooler Logic rejects the traditional experimental design in favour of an open systems/
complex systems approach to the field of study. Its other premise is that people already possess an
ethnographic consciousness provoked by the complexity of the social field that surrounds them, but
are lacking the stage and context to articulate their insights and are too often prevented from joining
individual insights into a meaningful whole. On this basis, the developers argue that for creating
solutions it is important to recruit frontline people as well as supervisors and managers close to the
work. Higher-up managers become more useful later when the organisation’s approval is required
to move forward. This feature of the complex organisation – emergence - explains why they
consider their kind of problem solving more a discovery of solutions already immanent to the work
practice of their clients.
There are three phases to the Water Cooler Logic after a project emerges; first it involves mapping
the organisational terrain of the subject of study, recruiting members of the organisation as
participants, teaching them a bit of ethnography, and second, it unravels into going into the field
with them and conducting the ‘ethnographic moments’ (Discovery Phase). It is in this sense that
WCL is participatory and leads participants to a reflective stance and vantage point. It is innovative
in terms of what the developers do with the team members. This involves the development of social
dynamics and macro culture. Participants bring insider knowledge – translated into ‘perspectives’,
views from somewhere to this work – which combine with the field knowledge ‘perspectives’ they
collect through observations and interviews into a more complete, though not necessarily more
harmonious, conceptual landscape. In this process people get transformed in their organisational
identity but also the implementation of organisational change of practice becomes easier because
people are part of the process. The transformation takes place as ethnographic authority is
transferred to the team, which becomes empowered to spearhead the change while the management
is in a supporting role. The team members develop a shared framework, a common/joint agenda to
implement changes and improve their practices. Corrective moves may be applied by consultants,
only when the development is in danger of being derailed. The framework provides a performative
context for staff in the local sites to engage in this ethnographically inspired work of puzzling out
their experiences and responsibilities in the organisation.
34
Overall, the method cannot be isolated from the process as it is the methodological framework
which is innovative. Conventional methods of data collection are used such as interviews,
observations and an analysis of stories and discourses that take place between group members at a
metalevel. This involves content-based analysis to package up this knowledge into a design. The
design is part of the process itself by analysing group discussions and presentations and playing
down the report; management attends these sessions and listens to the staff presentations (Co-
Design Phase). Raw data are not offered while anonymity and confidentiality are preserved. The
consultants may offer recommendations about what the organisation should do, but ideally the
recommendations take the form of presentations of team members to their own peers and managers,
which is already starting the implementation and change process. Proposed changes span the
breadth of organisational measures and resources ranging from – and at times combining – changes
to the physical environment, the organisational structure, technology and processes, and there is
usually a training aspect as well.
There is a third phase in some of these projects and called the implementation phase which involves
following up on what everyone was designed to do, keeping the local efforts in synchronicity with
one another and when needed cycling back to re-discovery, re-design etc.
To sum up, the framework involves ethnographic work plus a formal design workshop:
anthropological study of needs assessment plus state of the art design; both the discovery and the
design solutions presented to management refers back to the staff.
Prosthetics
Client-Provider communication
Bed control and patient discharge
Formal and informal learning at the workplace
Accepted principles of good experimentation in psychology and neuroscience frame their subject
matter so as to exclude many aspects from consideration (or keep them equal, which amounts to the
same) and thereby make artificial the situation they study. In contrast, this is not experimentation, it
does not exclude anything and deals with the multiplicity and variability of dimensions and factors
that are part of the natural situation. It sacrifices repeatability in order to stay close to reality and
relevant to practical applicability.
Anthropology/ethnographic consultancy
Colleagues
Sue Wilkinson
References
Darrouzet, C., Wild, H., Wilkinson, S., Kantor, C., Iveljic, S., Klute, G. K., & Creasey G. (2008)
Assessing Needs of a Network of Stakeholders for Bringing Innovation into Clinical Use. LLANA
methods paper- v9 ck. Palo Alto, CA: Water Cooler Logic
Darrouzet, C., Wild, H., & Wilkinson, S (2008) Participatory Ethnography at Work: Practicing in
the Puzzle Palaces of a Large, Complex Healthcare Organisation. Palo Alto, CA: Water Cooler
Logic, pp. 70-95
35
Carolyn Funk
Virginia Commonwealth University, Richmond, Virginia, USA
Genetic Transmission of Political Orientations

Introduction
This line of research tests the possibility that political attitudes and behaviours are the result of both
environmental and genetic factors. Far from claiming that genetic factors determine behaviour, the
claim is that they influence behavioural sensitivity. The research team developing this work was
Drs. John Alford, Carolyn Funk, and John Hibbing and now includes a mix of collaborators from
political science, psychology and behavioural genetics.
The method makes use of behavioural genetics but specifically concentrates on twin studies and
twin methodology on the assumption that hereditary traits are most influential in predicting political
orientations – though not necessarily party affiliation, which is deemed to result mainly from
parental socialisation – compared to environmental factors. Twins are chosen because the division
between monozygotic (MZ) - who share 100% of genetic materials – and dizygotic (DZ) – who
share 50% of genetic material – is instrumental in comparing correlations on the basis of genetic
characteristics. The assertion is that the effect of genetics is measurably distinct for MZ and DZ
twins, while the effect of the environment is either equivalent or at least randomly distributed
around equivalence.
In this respect, by predicting a large influence for genetic inheritance, the authors depart from
typical behaviouralist expectations anticipating that political attitudes will be predominantly
influenced by environmental factors, rendering genetic inheritance largely, if not completely,
inconsequential. Genetics have been widely applied to study ‘psychological traits’ rather than
political ideology.
The study makes use of the Wilson–Patterson (W–P) Index, commonly used to measure
conservatism. This index is administered by presenting subjects with a short stimulus phrase such as
death penalty or royalty and eliciting a simple agree, disagree, or uncertain response. The standard
techniques in behavioural genetics are based on correlation analysis. Using techniques from
behavioural genetics, correlations are computed separately for male/male and female/female twin
pairs to provide an appropriate comparison, since all MZ twins are same-sex pairs, while DZ twins
are a mix of same-sex and opposite-sex pairs. Heritability is estimated by subtracting the correlation
for DZ pairs from the correlation for MZ pairs and then doubling the resulting difference (2 x
(MZ−DZ)).
The results indicate that genetics plays an important role in shaping political attitudes and
ideologies but a more modest role in forming party identification; as such, they call for finer
distinctions in theorising about the sources of political attitudes. Nevertheless, predictably
dissimilar correlations of social and political attitudes among people with greater and lesser shared
genotypes suggest that behaviours are often shaped by forces of which the actors themselves are not
consciously aware, a point that is made with some force by Bargh and Chartrand (1999), Marcus
(2002), Marcus, Neuman, and MacKuen (2000), McDermott (2004), and Wegner (2002).
On these grounds, political scientists are urged to incorporate genetic influences, specifically
interactions between genetic heritability and social environment, into models of political attitude
formation.
36
Employing standard methodological approaches in behavioural genetics - specifically, comparisons
of the differential correlations of the attitudes of monozygotic twins and dizygotic twins - data are
drawn from a large sample of twins in the United States, supplemented with findings from twins in
Australia are analysed.
The researchers claim that twin studies have not been conducted by political scientists and that
properly refined measures of political variables have not been constructed and employed In other
words, the heritability of political behaviour has not been analysed at all.
Apart from psychology and political science, the study suggests that it could be applied in cases
dealing with behaviour.
Colleagues
Alford, J., Hibbing, J. R, Smith, K., Hatemi, P., Krueger, R., Eaves, L., Hibbing, M.
References
Alford, J., Funk, C. and Hibbing, J. R. (2005) Are Political Orientations Genetically Transmitted?
American Political Science Association. Vol. 99, No. 2, pp. 152-167
37
Technological Innovation/Crossing Boundaries
The third section presents innovative research practices that either cross disciplinary boundaries or
are driven by technological development. The areas and disciplines these have been developed in
are mainly anthropology and ethnography.
38
Dwight Read
University of California, Los Angeles, California, USA
Kinship Algebra Expert System (KAES)

Kinship Algebra Expert System (KAES) is a computer programme, a graphical user interface
application for investigating and modelling the structure of kinship terminologies and the ways in
which these are instantiated. The innovative aspect is twofold, the programme itself and also the
theory of kinship terminology structures on which it is based and which developed as a ‘paradigm
shift’ in the process of writing the KAES programme. In other words, the software became the
idiom and medium through which the theory was developed. The paradigm shift consists of
reversing the logic of the genealogical paradigm: it begins with kin terms viewed as cultural
constructs, including culturally determined computations for linking kin terms to one another
subject to structural constraints and thereby forming a structured system of kin terms. Kin terms can
then be linked to categories of kin types through the kin term structure, thereby producing a
structured system of kin type categories that provide genealogical definitions for kin terms. Read
calls the process through which the kin terms are linked to categories of kin types ‘cultural
instantiation’.
Based on this logic, the programme itself has three main components: map operations, algebra
operations and display operations. The first component (map operations) provides the machinery for
entering of kinship terms in the form of a kin term map that displays the manner in which kin terms
are linked to one another through a kin term product based on ethnographic observations about the
ways individuals calculate kinship relations. The data used are collected through ethnographic
research: eliciting people’s knowledge of their kinship system. What do people mean when they use
kin terms?
The algebraic modelling begins by structurally simplifying the kin term map through user-selected
options for simplification of a kin term map. The simplification is run in reverse during the
algebraic modelling as one elaborates on an initial algebraic model constructed in accordance with
the simplified kin term map. A modified kin term map is defined to be simplified when the
modified kin term map has a single, ascending generating term (and possibly a sibling generating
term) and only consists of ascending kin terms. The KAES programme allows for the user to model
a kin term map using user-selected options that activate an ensemble of individual steps composing
a stage in the algebraic modelling, such as constructing the descendant structure from the initial,
ascendant structure. Alternatively, the user may sequentially activate the individual steps making a
stage in the algebraic modelling. As the modelling proceeds, the KAES programme automatically
introduces structural equations that are part of the distinctions made regarding kin terms in the
kinship terminology. Each stage in the algebraic modelling of a kinship terminology structure is
successful if the algebraic modelling ends up with an algebraic structure isomorphic to the kin term
map from the initial simplification of the kin term map corresponding to the current stage in the
algebraic modelling. The algebraic construction stops when an algebraic structure that is isomorphic
with the kin term map for the complete kinship terminology has been generated.
While this is based on the groundbreaking assumption that people think in terms of kin terms rather
than genealogically and, thus, the correspondence between kin terms and genealogical relations is
not true of all terminologies, the third component (operations) constructs a genealogical
instantiation of the algebraic structure.
For all terminologies considered to date, the predicted genealogical definitions match with 100%
accuracy the genealogical definitions elicited by anthropologists. These genealogical definitions
39
have been assumed to be the primary and irreducible data for the analysis of kinship terminologies.
The KAES analysis demonstrates to the contrary that these definitions are predictable from the way
in which the terminology is generated as a structure. Instantiation is not limited to genealogical
instantiation. Instead, instantiation depends on cultural rules for mapping abstract symbols to
categories and categories to individuals and can be based on other criteria such as adoption or other
means by which individuals are incorporated within the conceptual structure of a kinship
terminology.
One possibility for the evolutionary origin of kinship terminology systems is that they provided a
computational device for calculating relations among individuals that was cognitively less
demanding than memorising extensive genealogical linkages. But even if it arose as a
computational system, the computational facility enabled the terminology to become the central
means by which individuals determine if they have a kinship connection or not. One can calculate
kin relations with the kinship terminological system without necessarily referring back to
genealogical linkages, and hence the relative importance of genealogical connections versus kin
terminology connections becomes society specific.
The idea for the KAES programme started with a suggestion from Dr. Clifford Behrens when he
was Read’s graduate student. He suggested an expert systems computer programme that would
embed the expertise of a mathematical anthropologist into a computer programme. (Hence the
name, Kinship Algebra Expert System.) They applied for, and received, an NSF grant for
developing KAES. At that time the most promising software languages for what they had in mind
were Prolog to do the algebraic part of the programme, Metawindows (which no longer exists) for
doing the graphics interface (at the time, Metawindows was the only PC software that had the
capability of developing a truly windowed and graphic interface using an object oriented
architecture), with Pascal as the software language for doing the nuts and bolts of the software
architecture. Later, when KAES was rewritten in Java by Dr. Michael Fischer, software
architectures had become far more sophisticated and with Java they had a fully object based
architecture within which the graphics user interface and its logic based on the idea of a kin term
map could be integrated with the architecture for the algebraic machinery originally developed in
Prolog.
See above for the ways in which the KAES programme has also been the basis for developing a
theory of kinship terminology systems and thereby also becomes a research tool for the exploration
of similarities and difference in the logic of kinship terminologies from different cultural contexts.
The programme can be used for representing other conceptual structures that have an algebraic
form, such as the idea of a binary opposition. Its application lies, perhaps, less in being directly
useful in other domains than as a way of making explicit the notion of culture as a ‘constructed
reality’ for the observation that a kinship terminology has an algebraic structure implies that the
structure is constructed and thereby generates a reality (namely how we conceptualise kin) that does
not exist without that construction. In this way it becomes a model for what is meant by the vague
idea of a constructed reality.
Other social relationships may be mapped using this programme as long as they have a similar
logic: a system that connects terms to each other e.g. Friend/enemy.
What kind of research questions could it be used for?
In general, it is useful for addressing the most fundamental questions in anthropology regarding
specifically what we mean by kinship and kinship systems and more generally, regarding the
40
relationships among cultural constructs and their instantiation in the form of social structure and
organisation and the behaviour of persons.
An extension of the application could be to use agent-based modelling (simulation) – embed
cultural concepts, things we elicit from people through ethnographic research and how do they
construct kin products. An agent-based model has been developed and was used to confirm the
ethnographic observations.
Colleagues
Dr. Clifford Behrens, Dr. Michael Fischer, Dr. Murray Leaf, Dr. Stephen Lyon, Dr. Giovanni
Bennardo, Dr. F.K. Lehman
References
Read, D. W. (2006) Kinship Algebra Expert System (KAES): A Software Implementation of a
Cultural Theory. Social Science Computer Review. Vol. 24, No. 1, pp. 43-76
Read, D. W. (2008) A Formal Explanation of Formal Explanation. Structure and Dynamics:
eJournal of Anthropological and Related Sciences. Vol. 3, No. 2, Article 4
Read, D. W. (2007) Kinship Theory: A Paradigm Shift. Ethnology. Vol. 46, No. 4, pp. 329-364
Bennardo, G., and D. W. Read. (2007) Cognition, algebra, and culture in the Tongan kinship
terminology. Journal of Cognition and Culture 7:49-88.
Leaf, M. (2006) Experimental analysis of kinship. Ethnology 45:305-330.
Read, D. W. and C. Behrens. (1990) KAES: An expert system for the algebraic analysis of kinship
terminologies. Journal of Quantitative Anthropology 2:353-393.
Read, D. W. (1998) Kinship based demographic simulation of societal processes. Journal of
Artificial Societies and Social Simulation. 1, http://www.soc.surrey.ac.uk/JASSS/1/1/1.html
Read, D. W. (2001) Formal analysis of kinship terminologies and its relationship to what constitutes
kinship. Anthropological Theory (Special Issue) 1:239-267.
Read, D. W. (2001) "What is kinship?," in The cultural analysis of kinship: The legacy of David M.
Schneider and its implications for cultural relativism. Edited by R. Feinberg and M. Ottenheimer,
pp. 78-117. Urbana: University of Illinois Press.
41
Robert Kozinets
York University, Toronto, Ontario, Canada
Netnography
Introduction
Robert Kozinets is an anthropologist who studies contemporary consumer cultures and
communities, their manifestations, and their implications. Netnography was introduced as a
methodology for consumer and marketing research in the late 20th century (Kozinets, 1998).
Nevertheless, in recent years, Netnography has transcended consumer and marketing research and is
currently widely used across disciplines.
Netnography is an interpretive method originally devised specifically to investigate the consumer
behaviour of cultures and communities present on the internet. More recently, it has been used to
investigate any community and culture present on the internet. It is a written account that results
from fieldwork studying on-line, computer-mediated or Internet-based communications. Based on
the traditions and techniques of cultural anthropology and ethnography the data collected consists of
the researcher’s field notes combined with the artefacts of the culture or community, and hence this
data is mainly textual such as downloaded files of newsgroup postings, transcripts of discussion
boards and chat sessions, and e-mail exchanges. As in traditional ethnography, once access is
gained into the online community studied methods other than participant observation may be
negotiated which include most of the conventional methods (e.g. interviews, focus groups etc.)
Netnography does not, therefore, coin conducting research online as it is not the mode (online
versus telephone or face-to-face) that differs but the field – on-line is not simply a way of gaining
access but the locus of social worlds, social life and social interaction for online communities and
cultures.
Kozinets (1998) points out that netnography is useful for three types of studies and in three general
ways: firstly, as a methodology to study ‘pure cybercultures and virtual communities’ that do not
exist off-line in real life, but are manifest exclusively through Computer-Mediated Communication
(CMC) or computer-mediated social interaction; secondly, as a methodological tool to study
‘derived’ cybercultures and virtual communities; and thirdly, as an exploratory tool to study general
topics related to online communities or cultures or including (derived) communities also available
online.
Online communication between consumers, for example, has been studied by using netnography
(Kozinets, 1998; Kozinets, 2002) for understanding their attitudes, perceptions, imagery, and
feelings. As Kozinets points out, the Internet offers increased opportunities for social group
participation and netnography provides access to the study of online communities where access
based on conventional methods is difficult. However, this is where ethical considerations and robust
procedures come into play in netnography. Kozinets claims that netnographic principles had not
been systematised for years, and netnographic research raised concerns about the ethical standards it
complied with. For this purpose, one of Kozinets’ recent contribution has been to systematise
netnographic procedures in a rigorous way.
Referring to common ethnographic procedures, Kozinets recommended the following
methodological stages and procedures for netnographic studies in general:
1. Planning;
42
2. Entrée: formulation of research questions, identification of appropriate online fora for study,
contacting the online community to be studied;
3. Data collection: online fieldwork - this may include – as in traditional ethnography – a
variety of methods for data collection and thus, various types of data; these are either
available and copied from the computer-mediated communications of online community
members or collected and recorded or produced by observations of the community and its
members, interactions and meanings by keeping diaries. Kozinets (forthcoming) has
specified and explicated in great detail the various options for retrieving, storing and
preparing netnographic data for analysis.
4. Analysis and interpretation: classification, coding analysis and contextualisation of
communicative acts; methods of analysis may include conventional methods such as
conversation and discourse analysis, documentary analysis, content analysis and the
application of qualitative analysis software.
5. Generally speaking, as with traditional research, analysis depends upon the type of data and
philosophical underpinnings of the study / researcher. Data collection and analysis should
not be confused with downloading data in textual form and analysing them. Engagement
with the online community studied is what netnography is about and one of the key
‘features’ which ultimately differentiates it from collecting data online.
Research ethics: “(1) The researcher should fully disclose his or her presence, affiliations, and
intentions to online community members during any research; (2) the researchers should ensure
confidentiality and anonymity of informants; and (3) the researchers should seek and incorporate
feedback from members of the online community being researched (4) The researcher should take a
cautious position on the private-versus-public medium issue. This procedure requires the researcher
to contact community members and to obtain their permission (informed consent) to use any
specific postings that are to be directly quoted in the research” (Kozinets, 2002, p. 65); 6. Member
checks and Research representation: presentations of some or all final research report’s findings to
the people who have been studied and to the scientific community.
Study of new mothers discourses and practices, study of skinhead culture and beliefs, study of fat
acceptance blogs, study of boycotters, study of online funs of Manchester United
Netnography is a specific set of procedures to guide ethical, accurate, and culturally attuned
ethnography of online communities and cultures. A number of scholars have used ethnographic
methods online, but none have systematised an organised method. The rationale and inspiration in
combining the bits together to develop netnography has been the need of scattered principles to be
formalised. This was identified as part of Kozinets’ study of Fan Communities for his PhD thesis.
Thus, Robert Kozinets has been developing a systematic and rigorous framework for doing
ethnographic research since 1995, presenting and publishing in this area since 1996.
Currently applied most widely in marketing research.
Colleagues and / or Relevant Research
Maclaran, Pauline, Miriam Catterall, Margaret Hogg, Steven Brown, Hope Schau, Andrea
Hemetsberger, Kristine de Valck
43
Other Information
Robert Kozinets and Russell Belk have also been developing Videography: “a form of visual
anthropology encompassing the collection, analysis, and presentation of visual data;” “an
audiovisually-based ethnography that is the product of a participant-observational research method
that records interviews and observations of particular peoples, groups, and their cultural artifacts,
utilises them as data, edits them into a format for presentation, and represents it in the form of a
film” (Kozinets & Belk, 2006, p. 318). This has resemblances to research-based films (see research
based drama below), systematising and developing the use of audiovisual methods in the social
sciences specifying particular ways in which ‘theory’ can be represented and presented in films.
References
Kozinets, R. V. (forthcoming), Netnography: Doing Ethnographic Research at the Age of the
Internet. London: SAGE
Kozinets, Robert V. and Richard Kedzior (forthcoming), “I, Avatar: Auto-netnographic Research in
Virtual Worlds,” in Virtual Social Identity and Social Behavior, ed. Michael Solomon and Natalie
Wood, Armonk, NY: M.E. Sharpe.
Kozinets, Robert V. (2006), “Netnography 2.0,” in Handbook of Qualitative Research Methods in
Marketing, ed. Russell W. Belk, Cheltenham, UN and Northampton, MA: Edward Elgar Publishing,
129-142.
Kozinets, Robert V. (2006), “Netnography,” The Sage Dictionary of Social Research, ed. Victor
Jupp, London: Sage, 193-195.
Kozinets, Robert V. and Russell W. Belk (2006), “Videography,” The Sage Dictionary of Social
Research, ed. Victor Jupp, London: Sage, 318-320.
Kozinets, Robert V. (2002), “The Field Behind the Screen: Using Netnography for Marketing
Research in Online Communities,” Journal of Marketing Research, 39 (February), 61-72.
Hine, C. (2004) Virtual ethnography. In Lewis-Beck, M.S., Bryman, A. and Liao, T.F. The Sage
Encyclopedia of Social Science Research Methods. SAGE
44
Intra-disciplinary & Intra-methodological
The fourth section presents innovative research practices driven by developments with disciplines
and/or methods. The areas and disciplines these developed in are mainly management, psychology,
sociology, and social work.
45
Chris Nash
Monash University, Melbourne, Victoria, Australia
Journalism as both Reporting and Research

Introduction
Chris Nash’s work has focused on Journalism in academia, treated in its conventional form as
reporting but incrementally as research as well. Spurred by the role and position of journalists
within the university, he has sought to explore a theory of practice that brings together journalism as
study and journalism as practice. This endeavour is treated as a research approach in this report.
The approach starts with identifying journalism’s domain: ‘Everything that comes under the
category of the present’. Therefore Journalism is interdisciplinary and critically defined by
temporality, yet involves the subjective element of intuition (which could nevertheless be claimed
for most qualitative research approaches). The development of this approach to journalism as
research and reporting mainly draws on the work of Tuchman (1978), the writings of Bourdieu and
the expositions on space and time of David Harvey, Henri Lefebvre and Alfred Gell.
Tuchman’s work on making news as constructing reality identified the constraints and agents of
journalistic ‘methods of enquiry’. Namely, of news as a ‘frame’, produced by and as a ‘web of
facticity’, cast in time and space, and presented as a narrative of knowledge and power. In this
process journalists are agents within the symbolic field as described above, operating intuitively in
fluid time and space, to research in and report accurately and selectively on other fields – those of
study - in order to represent them to publics (in yet another field/fields?). In view of that Chris
Nash’s work focused on building a theory of practice that engages with intuitive practice, relations
between fields of social activity, the empirical, time and space and different publics. On this he
draws on Bourdieu’s notions of field, capital and habitus. In particular, “all activity and practice
occurs within fields constituted by relations of power, where cultural capital is deployed using
habitus to exercise symbolic power/violence against competitors”. For the remaining dimensions
(including time, space, discourse, and publics) Chris Nash stresses four key dimensions of
journalistic practice: that it is empirical, socio-political, temporal and spatial. The first deals with
the who, what, when, where and how of inquiry; the second deals with the why; the third deals with
time as both phenomenological and chronometric (Gell, 2003); and the fourth treats space as
absolute, relative and relational (Harvey, 2006) and perceived, conceived, imagined (Lefebvre,
1991).
Overall, therefore, journalism as an exercise in research and reporting directly engages key issues in
ontology and epistemology (truth, power, time and space), does not require mediation by other
disciplines in this engagement, is fundamentally inter-disciplinary and requires theorisation as an
intellectual practice.
Media coverage of issues and contests in urban transformation under globalisation
Chris Nash wished to contest views that render journalism as a superficial, empirically ‘weak’ and
methodologically ‘void’ non-intellectual activity. As part of this endeavour he has aimed to
‘constitute’ journalism as a legitimate mode of intellectual inquiry, engaging with ontological and
epistemological questions.
46
Colleagues
Wendy Bacon (wendy.bacon@uts.edu.au) Professor of Journalism at the University of
Technology, Sydney
References
Gell, A. (1992) The anthropology of time: cultural constructions of temporal maps and images.
Oxford: OU.P.
Castree, N. & Gregory, D. (2006) David Harvey: a critical reader. Malden, MA: Blackwell
Harvey, D. (2006) Spaces of global capitalism: towards a theory of uneven geographical
development, Verso, London
Lefebvre, H. (1991) The production of space. Malden, MA: Blackwell
Tuchman, G. (1978) Making news: a study in the construction of reality. New York: Free Press
Bourdieu, P. & Wacquant L. (1992) An invitation to reflexive sociology. Chicago: University of
Chicago Press
Bourdieu , P (1990) The logic of practice, Cambridge: CU.P
Benson , R. and Neveu, E. (2005) Bourdieu and the journalistic field. Cambridge: Polity
47
Gervase Bushe and Michael Avital
Simon Fraser University, Vancouver, Canada
Discovery & Design, Vancouver, Canada
University of Amsterdam, Amsterdam, Netherlands
Appreciative Inquiry as a Research Method

Introduction
Appreciative Inquiry as a theory was developed by David Cooperrider and Suresh Srivastva in 1986
and studies ‘the best of what is in order to identify what could be’ (Cooperrider and Srivastva,
1987). Bushe argues that Appreciative Inquiry (Cooperrider & Srivastva, 1987; Cooperrider,
Sorensen, Yaeger &Whitney, 2005) was created explicitly as a response to Gergen’s 1978 paper,
“Toward Generative Theory,” in which he argued that many assumptions of positivism could not be
successfully applied to studying human systems. He proposed that, instead, researchers should aim
to create a social science focused on its generative capacity, the “…capacity to challenge the
guiding assumptions of the culture, to raise fundamental questions regarding contemporary social
life, to foster reconsideration of that which is ‘taken for granted’ and thereby furnish new
alternatives for social actions”(1978, p.1346).
Since then it has been taken up and expanded and commonly refers to an action research process
that studies something from the positive side – thus belonging to a group of approaches to research
termed ‘positive lens’ – to create a new kind of conversation among people as they work together to
improve a group or organisation.
The development of appreciative inquiry style interview guides, which focus on gathering ‘best of
stories’. On the basis of the narratives produced, theories and models of ‘best practice’ may be
developed. The assessment of the utility of appreciative inquiry as a methodology for model
building research is pending. Bushe has proposed using appreciative inquiry, a relatively unknown
(outside the field of organisation development) qualitative methodology. Appreciative inquiry has
proven a successful form of action research when applied to changing organisational cultures
(Bushe & Kassam, 2005; Powley, Fry, Barrett & Bright, 2004). Its utility and efficacy as a research
method, however, remain unclear. There have been recent calls encouraging greater use of
appreciative inquiry as a method of basic, as opposed to applied, research (e.g. Reed, 2007).
Responding to such calls Bushe embarks on an innovative approach to the use of appreciative
inquiry for model building research. This involves the application of ‘appreciative questions’
(Bushe, in press, Cooperrider, Whitney & Stavros, 2008) for developing questions that may have a
higher than normal capacity to be generative – that is to produce information that aids in generating
new social theory. In particular, an appreciative inquiry interviews participants about their ‘peak
experiences’ or ‘best of stories’ regarding the phenomena of interest, as well as asking participants
for their ideas and opinions about the best of what could be. When used in organisational change
initiatives, these stories catalyse dialogue among members of the organisation about their hopes and
aspirations, which in turn leads to the collective design of alternative and preferred futures.
Analysis may then follow depending on the desired outcome.
Generally speaking, studying ‘what is’ but also ‘what could be’.
48
A number of studies show that appreciative inquiry can help organisational stakeholders identify
and act on inspiring futures in ways that create collective will to work toward implementing those
visions, yet it is claimed that very little research has been conducted using appreciative inquiry as a
research methodology. Recent calls encouraging its use in research highlight its potential for theory
and model building. Because appreciative inquiry focuses on the best of what is in order to inspire
new theories and models of what could be, it seems particularly appropriate to address the questions
that motivate researchers.
It has been applied in organisation and leadership studies and management with success. As long as
there is an element of studying ‘what is’ in order to theorise and model ‘what could be’
Colleagues
References
Avital, M., Boland, R. J. & Cooperrider, D. (Eds.) (2008) Designing Information and Organisations
with a Positive Lens: Advances in Appreciative Inquiry. Vol. 2, Oxford: Elsevier Science
Bushe, G.R. (in press) Generativity and the Transformational Potential of Appreciative Inquiry. In
Zandee, D., Cooperrider, D.L. & Avital, M (Eds.). Organisational Generativity: Advances in
Appreciative Inquiry. Vol.3, Amsterdam: Elsevier
Bushe, G. (2007) Why Appreciative Inquiry is Unlikely to Generate Research Publications.
Appreciative Inquiry Practitioner. November 2007, pp. 8-11
Bushe, G.R. & Kassam, A. (2005) When is Appreciative Inquiry Transformational? A Meta-case
Analysis. Journal of Applied Behavioural Science. Vol. 41, No. 2, 161-181
49
Peter Sommerfeld
University of Applied Sciences Northwestern Switzerland, Brugg, Switzerland
Rasmussen, N. H., Furst, J. W., Swenson-Dravis, D. M., Agerter, D. A., Smith,

A. J., Baird, M. A. & Cha, S. S.
Mayo Clinic, Rochester, Minnesota, USA
University Minnesota, Minneapolis, Minnesota, USA
Reflecting Interview
Introduction
Innovative reflecting interview is an intervention driven approach to social work and psychology.
Reflecting interviews are teller-centred focus clinical interviews, which address participants’
progress, goals, and reflections on their personal progress as documented in previous monitoring
data. The inclusion of these monitoring data in the clinical interviews reveals a strong and direct
link between the psychic dynamics of individuals’ progress and social events.
It usually involves collecting longitudinal data about the social side of the reintegration processes
(complementing the Real Time Monitoring data on the psychic side). The process takes place in
over a period of one of intervals: In these interviews, participants are usually asked to reflect on an
aspect of their lives over between the periods sampled in a narrative, open way. In addition, they are
asked to review and analyse data they have produced during the preceding period by reflecting on
the significant peaks in that data – what happened at that time, how could this be understood, and
what might this mean. In the third and final part of the reflecting interviews, participants may set
personal goals for the next sampled period and/or reflect on progress made towards the personal
goals established during the preceding session.
Patient health care utilisation, physical health, mental function, and health care satisfaction.
Intervention is less time intensive and requires less professional staff training than other forms of
intervention.
It could be applied in any area.
References
Greene, J. C., Sommerfeld, P. and Haight, W. L. (in press) Mixing Methods in Social Work
Research. In Katharine Briar-Lawson, Joan Orme, Roy Ruckdeschel and Ian Shaw (eds.): The
SAGE Handbook of Social Work Research. London: SAGE
Rasmussen, N. H., Furst, J. W., Swenson-Dravis, D. M., Agerter, D. A., Smith, A. J., Baird, M. A.
& Cha, S. S. (2006) Innovative Reflecting Interview: Effect on High-Utilising Patients with
Medically Unexplained Symptoms. Disease Management. Vol. 9, No. 6, pp. 349-359
50
Rosemarie Anderson
Institute of Transpersonal Psychology, Palo Alto, California, USA
Intuitive Inquiry
Introduction
Rosemarie Anderson is Professor of Transpersonal Psychology in the Global PhD Programme of
the Institute of Transpersonal Psychology. In collaboration with students and faculty, Dr Anderson
developed a transpersonal research method in 1996 (published 1998) known as Intuitive Inquiry
that invites intuition into the research process, and a research practice known as Embodied Writing
that brings the finely textured experience of the body to the art of writing and data gathering. Both
Intuitive Inquiry and Embodied Writing blend the rigors of mindful spiritual practice with the
rigours of scientific inquiry.
In the mid-2000s, she developed the Body Intelligence Scale that provides quantitative measures of
three types of bodily awareness applicable to health and well-being. This scale is now available to
researchers who are interested in the relevance of the scale to stress-related diseases and personal
transformation. Her current interests include the further development of Intuitive Inquiry into a
multi-method approach to research, the study of the multi-cultural and global expressions of
intuition, and continuing studies on the applications of the Body Intelligence Scale. In 2007, Dr.
Anderson launched her own website, www.wellknowingconsulting.org offering intuition and
transpersonal research consultation to the public.
At the outset, the intuitive researcher initially identifies her or his values and assumptions through
active and connected engagement with the experience studied and then uses these values and
assumptions as hermeneutical lenses to explore and analyse similar experiences in others. This is
called the hermeneutical circle. Rather than bracketing the researcher’s values and assumptions (as
in established phenomenological approaches to research), the intuitive researcher employs her or his
values and assumptions as lenses to begin the interpretative cycles of analyses. Intuitive Inquiry is
composed of five iterative cycles of interpretation.
First, the researcher selects a text or image that repeatedly attracts or claims the intuitive
researcher’s attention and relates to his or her area of interest in a general and often obscure and
non-obvious way. The researcher then enters the circle of interpretation by engaging with the “text”
daily and recording both objective and subjective impressions. Thoughts, ideas, daydreams,
conversations, impressions, visions, intuitions, etc. occurring during sessions are recorded in a non-
invasive manner, so to least disrupt the stream of consciousness typically accompanying intuitive
insight. Intuitions manifest as visual, auditory, proprioceptive (a felt sense), or kinesthetic.
Second, the researcher reflects upon her or his own understanding of the topic in light of a set of
selected texts produced by others and found in extant literature about the topic and prepares a list of
preliminary interpretative lenses that express the researcher’s understanding of the topic as
unambiguously as possible. the researcher engages with the selected text(s) on a daily basis so to
keep the topic elevated in awareness and begins to note consistent patterns or clusters of ideas in her
or his understanding of the topic. Through a sometimes arduous process of combining,
reorganising, and identifying emerging patterns, the list shortens to between several clusters. These
lenses describe the researcher’s pre-understanding of the research topic prior to the collection of
original data. Typically, these lenses are prepared simultaneous with writing the Literature Review
as a means of presenting the researcher’s pre-understanding of the topic in light of the theoretical
and empirical literature relevant to the topic studied.
51
Third, the researcher (a) identifies the best source of data for the research topic; (b) develops criteria
for the selection of research informants or selection of extant historical, empirical, or literary
records; (c) collects the data; and (d) then prepares summary reports in as descriptive a manner as
possible. Since Intuitive Inquiry invites an in-depth process, one should choose the data sources that
satisfy their passion as researchers. After data collection, depending on the type of data collected,
the researcher organises and summarises research data using conventional thematic content
analysis, descriptive summaries, portraits, or conventional statistical analysis.
Fourth, the researcher then interprets data in order to modify, refute, remove, reorganise, and
expand his or her understanding of the research topic. This cycle invites researchers to expand and
refine their pre-understandings by incorporating the experiences of others and represents the
researcher’s summary of findings based on his or her interpretation of the original data collected.
Throughout Intuitive Inquiry, the most important feature of interpreting data is intuitive
breakthroughs, those illuminating moments when the data begin to shape themselves into view.
Patterns seem to reveal themselves with each fresh set of information.
Finally, the intuitive researcher stands back from the entire research process to date and takes into
consideration all aspects of the study anew, as though drawing a larger hermeneutical circle around
the hermeneutical circle prescribed by the forward and return arcs of the study. Attention to
recording both objective and subjective data accompanies each iterative cycle.
In terms of methods, various techniques and research strategies can be blended. Both observational
data and intuition are encouraged as sources of amplification and refinement of one another
throughout the course of the research endeavour. For the purposes of intuitive inquiry, intuition is
defined to include the more commonplace forms of intuitive insight as novel thoughts and ideas;
together with insights derived from non-rational processes such as dream images, visions,
kinesthetic impressions, a felt (or proprioceptive) sense, an inner sense or taste accompanying
contemplative practices and prayer, and spontaneous creative expressions in dance, sound,
improvisation, writing, and visual art. Intuitive inquiry openly invites the researcher to structure the
research method, procedures, setting and context to maximise rather than minimise the very
gateway through which the researcher understands or is inspired by the experience studied.
In intuitive inquiry participants are asked to detail physical descriptions of experience and
researchers are asked to contextualise and communicate findings by writing in an embodied manner
themselves. To communicate powerfully and resonately, intuitive inquiry invites research
participants to speak from their own unique and personal perspective born of their own experience.
Accordingly, researchers are urged to quote, often extensively, the actual words of participants so to
retain and portray the fullness of the participants’ unique voices and phenomenon studied in
communicating results. Communicating research findings close to the bone of the experiences
themselves, enables readers to resonate with the stories from within their own bodies by connecting
to the felt experience of the tellers. Researchers are enjoined to use creative media such as pictures,
photographs, recordings, drawings, videos, poetry, etc. so to resonantly portray the phenomenon
studied.
In terms of scientific validation, the immediate apprehension or recognition of a researcher’s insight
analyses could be verified using conventional experimental, quantitative procedures. Validity of
findings is thus formed through consensus building, noting consonance, dissonance, or neutrality
within a culture and across cultures. Subgroup by subgroup, a kind of mapping of the validity of a
research finding is created. A modified sociogram, constructed with concentric circles of resonance,
designates subgroups wherein the research findings are immediately apprehended and recognised or
reacted to with dissonance or neutrality.
52
Research topics such as reclaiming identity after abuse, the inward movement of beauty, the
qualities of serenity and contentment in everyday life, the experience of addiction and its impact on
long-term relationships, betrayal by a spiritual teacher, mutuality in relationship, and the experience
of long-term chronic pain have all emerged from personal experience and a desire to share and
amplify the experience through study of the experiences of others.
Intuitive inquiry was first inspired by the challenges of conducting research in the field of
transpersonal psychology. In particular, Intuitive Inquiry originally developed in response to
Anderson’s doctoral students’ needs in the field of transpersonal psychology. While transpersonal
psychologists and transpersonalists (a term more indicative of the field’s inter-disciplinary nature)
have been studying the rich dimensions of being human for roughly thirty years, research in the
field has often seemed stymied by a reliance on the experimental methods it inherited from the
dominant psychologies of the 1960s and 1970s, the context for its impetus. Intuitive inquiry
encourages a more inclusive and connected manner of conceptualising research topics, collecting
and analysing data, and presenting research findings in the study of human experience. The claim is
that one is connected to what they know through the very core of their body, mind, and self.
As a research method, it follows in the philosophic lineage of Western hermeneutics originating
with Friedrich Schleiermacher, Wilhelm Dilthey, and Hans-Georg Gadamer and the embodied
phenomenology of Maurice Merleau-Ponty and contemporary feminist theorists and researchers.
The research of Becky Coleman (1999) on obesity and right-body size for women provides a salient
example. Because the most integrative experiences of transforming her own experience of obesity
and right-body size had occurred in groups of women in intimate dialogue, Coleman chose to study
obese women meeting in small groups. Dialoguing with one another over time, Coleman took the
women on retreat during the final phase of her data collection. She was able to professionally
photograph and video-tape their exchanges and interactions to the fullest possible extent. There are
lots of other ways to study this phenomenon which would have been less costly and time-
consuming. However, Coleman’s own experience and understanding of the intimate demands and
integrity of understanding obesity committed her to a research strategy which focused on groups of
women sharing their stories about obesity and right-body size.
Rosemary Anderson claims that this is a research method which may constitute a general
framework for doing research beyond transpersonal psychology. However, intuitive inquiry may be
particularly suited to studying transformative experiences.
Colleagues
William Braud and doctoral students at the Institute of Transpersonal Psychology who have
encouraged and expanded upon intuitive inquiry in their own research, especially Becky Coleman,
Jay Dufrechou, Caryl Gopfert, Kelly Lynch, and Cortney Phelon.
References
Anderson, R. (1998). Intuitive inquiry: A transpersonal approach. In Braud, W. & Anderson, R.,
Transpersonal research methods for the social sciences: Honouring human experience. Thousand
Oaks, CA: SAGE
53
Anderson, R. (2000). Intuitive inquiry: Interpreting objective and subjective data. Journal of
Consciousness and Transformation. Vol. 22, No. 4, pp. 31-39
Braud, W., & Anderson, R. (1998). Transpersonal research methods for the social sciences:
Honoring human experience. Thousand Oaks, CA: SAGE
Coleman, B. (1999). Women, Weight & Embodiment: The experiences of women improving their
relationship with body through issues with weight. Unpublished doctoral dissertation proposal,
Institute of Transpersonal Psychology. Palo Alto, CA.
Dufrechou, J. (1999). The experience of grief in response to nature: A somatic-hermeneutical
exploration. Unpublished doctoral dissertation proposal, Institute of Transpersonal Psychology.
Palo Alto, CA.
Gopfert, C. R. (1999). Student experiences of betrayal in the Zen Buddhist teacher/student
relationship. Unpublished doctoral dissertation, Institute of Transpersonal Psychology. Palo Alto,
CA.
Lynch, K. S. Each age a lens: A transpersonal perspective of Emily Dickinson’s creative process.
Unpublished doctoral dissertation proposal, Institute of Transpersonal Psychology. Palo Alto, CA.
Phelon, C. R. (1998). The body-mind hermeneutic. Unpublished manuscript.
Phelon, C. R. (1999). A hermeneutical analysis of presence. Unpublished doctoral dissertation
proposal, Institute of Transpersonal Psychology. Palo Alto, CA.
54
Adele Clarke
University of California, San Francisco & Oakland, California, USA
Situational Analysis
Introduction
Adele Clarke’s primary research areas have been historical and contemporary sociology of
biomedical sciences and technologies, the development of qualitative research methodologies, and
women’s health. The development of qualitative research methodologies in particular led Adele
Clarke to a new approach to grounded theory in the postmodern turn: Situational Analysis.
Situational analysis arose in and through Adele Clarke’s own work since she entered the University
of California, San Francisco as a graduate student in sociology in 1980, especially through her
teaching of grounded theory and other qualitative research methods at UCSF since 1989. It also
emerged from her reading and teaching feminist theory, interactionist theory, Foucault, cultural
studies and science, technology, and medicine studies. As a student, she learned grounded theory
from Anselm Strauss who was at the time also engaged in elaborating his social
worlds/arenas/discourse/negotiations framework.
The report focuses on this development rather than situational analysis as used in management and
marketing.
Situational Analysis offers three main cartographic approaches as maps which are intended as
supplemental approaches to traditional grounded theory analyses that focus on the action-centred
‘basic social process’. Situational analysis offers these maps in order to shift attention to the
situation of inquiry. Through mapping the data, the analyst constructs the situation of inquiry
empirically. The situation becomes the ultimate unit of analysis and understanding its elements and
their relations are the primary goals. Situational analysis allows researchers to draw together studies
of discourse and agency, action and structure, image, text and context, history and the present
moment to analyse complex situations of inquiry. Thus it can situate research projects individually,
collectively, organisationally, institutionally, temporally, geographically, materially, discursively,
culturally, symbolically, visually, and historically and can support researchers from heterogeneous
backgrounds pursuing a wide array of projects.
Situational Analysis builds upon Anselm Strauss’s social worlds / arenas / negotiations / discourse
framework to offer these maps which are described as follows:
1. situational maps: that lay out the major human, nonhuman, discursive, historical,
symbolic, cultural, political and other elements in the research situation of inquiry and
provoke analysis of relations among them; these maps are intended to capture and
discuss the messy complexities of the situation in their dense relations and
permutations. They intentionally work against the usual simplifications so characteristic
of scientific work in particularly postmodern ways.
2. social worlds/arenas maps: that lay out the collective actors, key nonhuman elements,
and the arena(s) of commitment and discourse within which they are engaged in
ongoing discourse and negotiations. Such maps offer meso-level interpretations of the
situation, explicitly taking up its social organisational, institutional, and discursive
dimensions. They are postmodern in their assumptions: do not assume directionalities of
influence; boundaries are open and porous; negotiations are fluid; discourses are
multiple and potentially contradictory. Negotiations construct and constantly destabilise
the social worlds/arenas maps. They are open to the possibility of things always being
55
otherwise—not only individually but also collectively/organisationally/
institutionally/discursively—and these maps portray such postmodern possibilities; and
3. positional maps: that lay out the major positions taken, and not taken, in the data vis-à-
vis particular axes of difference, concern, and controversy around issues in the situation
of inquiry. Positional maps are not articulated with persons or groups but rather seek to
represent the full range of discursive positions on particular issues fully allowing
multiple positions and even contradictions within both individuals and collectivities to
be articulated. Complexities are themselves heterogeneous, and improved means of
representing them are required.
All three kinds of maps are keyed to taking seriously the nonhuman — including discourses — in
the situation of inquiry. The three maps are intended as analytic exercises. Situational analysis
offers an alternative approach to both data gathering and analysis/interpretation as it guides the
research design and the analysis of interview data, ethnographic data, narrative, visual, and
historical discourse materials taking into account all the elements in a situation and their
interrelationships. In doing initial situational maps, the analyst is asked to specify the nonhuman
elements in the situation, thus making pertinent materialities and discourses visible from the outset.
The flip side of the second kind of map, the social worlds/arenas map, is a discourse/arenas map.
Social worlds are ‘universes of discourse’ routinely producing discourses about elements of concern
in the situation. Such discourses can be mapped and analysed. Finally, positional maps seek to open
up the discourses by analysing positions taken on key analytic axes. Discourses can thereby be
disarticulated from their sites of production, decentring them and allowing further analytic bite.
The approach retains elements considered strong in grounded theory but aims to equip them with
constructivist principles. One such key element is theoretical sampling as a heuristic value of
developmental research designs. The proposition is instead to supplement basic grounded theory
with a situation-centred approach that in addition to studying action also explicitly includes the
analysis of the full situation, including discourses - narrative, visual, and historical. Such work can
enrich research by addressing and engaging the important complexities of postmodern theoretical
and methodological concerns. Adele Clarke argues on this basis that grounded theory was already
in many ways around the postmodern turn while in other ways it was not particularly so, and/or not
clearly so. Situational maps are proposed to make it so.
To regenerate grounded theory as part of the need to shift from action-based processes to the
situation of inquiry as the key unit of analysis in the postmodern turn. Clarke seeks to contribute to
the development of innovative methodologies that can specify the ‘complexities of situatedness’
and take into account the ecologies of actions, hoping to support research efforts toward greater
social justice and more democratic participation.
The three maps can be used in a wide array of research projects drawing on interview, ethnographic,
historical, visual, and/or other discursive materials, including multisite research.
Colleagues and/or Related Research
Adele Clarke cites Kathy Charmaz’s work as relevant to her work in aiming to open grounded
theory to more constructivist principles.
References
Clarke, A. E. (2005) Situational Analysis: Grounded Theory After the Postmodern Turn. Thousand
Oaks, CA: SAGE.
56
Ahmet Süerdem
Bilgi University, Istanbul, Turkey
Towards a New Approach in Social Simulations: Meta-Language

Introduction
Ahmet Süerdem has been experimenting bridging qualitative and quantitative methods. His active
research areas involve culture and social cognition / language, social simulations, and the analysis
of policy discourse. His work is included in the report for extending the ways in which social
simulation may be applied.
Ahmet Süerdem’s approach is a mixture of network analysis and semiotic analysis. The aim is to
produce network measures and apply them to semiotic indicators like: knowledge domains,
modality, paradigms, paradigm shift etc. For this process, conventional data collection methods are
used such as qualitative interviews, and the aim is to eventually analyse textual data. Analysis
follows then a more complex and staged path which offers an alternative to conventional methods
and also to the ways of doing social simulations. It starts with thematic analysis to identify concepts
and proceeds with semiotic analysis to build a matrix in the form of cognitive semiotic maps; the
matrix is then used it for mathematical network analysis (and/or simulations). The cognitive
semiotic maps may be built in agents to run social simulations.
The semiotic mapping approach developed here is based on two ways of doing semiotic analysis: 1.
analysis of concepts related to one another through dyadic relations; 2. analysis on the basis of
concepts opposing one another (opposition maps). Depending on which way is followed, the
semiotic structure of a text is ‘revealed’/built (e.g. interview transcripts) which is then turned into a
map format – the mental map ‘behind’ a textual presentation.
This semiotic mapping approach aims to overcome the shortcomings of semantic and semiotic
approaches by combining quantitative map analysis techniques with qualitative thematic analysis.
Thematic analysis is an interpretive coding approach. During thematic analysis, a researcher
searches for themes that emerge as being important to the description of the phenomenon. The
process basically involves reading a piece of text and identifying its meaning. It is a form of pattern
recognition within the data, where emerging themes become the categories for analysis. The
interpretive meaning identification is kept at the paragraph level and leaves the pattern recognition
to the network analysis. The basic idea is reading the paragraph and identifying its connotation as a
statement which reflects its main theme.
In particular, the analytic process evolves in reading the interview transcripts (or textual data
available) and recognising an important moment to encode it prior to a process of further
categorisation. Encoding the information organises the data to identify and develop themes from
them. These themes are then refined by merging low-frequency codes into high-frequency ones. A
codebook is formed for deductive coding which involves a template codebook to be applied as a
means of organising text for subsequent interpretation. In the next stage a deductive coding process
is followed by using the codebook. Text segments to be coded are paragraphs and each paragraph is
coded according to the themes in their content. Paragraphs are chosen as units of analysis on the
assumption that they represent the basic units of thought within an argument. Each paragraph is
considered as a statement connecting two concepts according to a ‘refers to’ relation including
predicates such as ‘is’, ‘signifies’, ‘makes’, ‘suggests’ and so on if the first way of doing semiotic
analysis is followed or ‘is in opposition to’, ‘opposes’ etc if the second way of doing semiotic
analysis is followed (see above). Since semiotic theory assumes that meaning is generated in terms
57
of reference (positive or negative) of signs to each other within a formal structure, reference is the
ultimate relation for semiotic analysis. At the next stage, the cross-tabulation of codes and agents is
obtained to compare the content distribution of the discourses. These cross-tabulations are then used
to present the distribution of agents and codes on a two-dimensional graph with the help of the
correspondence analysis. Correspondence analysis helps to examine the discursive proximities and
the positioning of the agents according to the frequency of the concepts in their discourses.
Discourses are then grouped in terms of their articulation by agents and matrices representing the
semiotic communities and summarising how the codes refer to each other within the paragraphs are
obtained. The matrices are asymmetric since reference relation is asymmetric. Finally, the semiotic
maps and the associated network measures are extracted, and basic network measures are obtained
with the help of a network analysis programme (UCINET 6 for Windows in this case). On the basis
of the semiotic maps and network measures a meta-language approach allowing dynamic meaning
generation during the interactions of the agents may be developed by running social simulations.
Political discourse: compare semiotic structures: how people say – reference map, web of concepts
– two statements coming together form an argument
Cultural anthropology: people construct reality in different manners
Artifact analysis
Colleagues
Raif Serkan Albayrak
References
Albayrak, R. S. & Suerdem, A. K. (forthcoming) Towards a New Approach in Social Simulations:
Meta-Language. In Jaime Simão Sichman & David Nuno. Multi-Agent-Based Simulation IX, 9th
International Workshop, Lisbon, Portugal, May 2008, Revised Papers. Heidelberg: Springer-Verlag
Süerdem, A. (2009) A Semiotic Network Comparison of Technocratic and Populist Discourses in
Turkey: A Sociological Perspective. In Imbeau, Louis M. (Ed.) “Do They Walk Like They Talk?
Speech and Action in Policy Processes Series: Studies in Public Choice, Vol. 15., New York:
Springer
Süerdem A. (2008) (with R. S. Albayrak) A Method Proposal for Elicitation of Intensional
Ontological Classes: Opposition Map Approach, 7th International Conference on Social Science
Methodology, Campus di Monte Sant’Angelo, Naples, Italy, September 1-5 2008
Süerdem A. (2008) (with G. Çoban) La Construction Discursive de l’Espace Social Européen: les
paradigmes, les discours et les communautés épistémiques, Colloque 2008 de la Société
Québécoise de Science Politique, Université de Montréal, Quebec, Canada, May 8-9 mai 2008
Süerdem A. (2008) (with R. S. Albayrak) Formalising Epistemological Constituents of Emergence,
AISB 2008 Convention Communication, Interaction and Social Intelligence, Aberdeen, UK, April
1-4 2008
Süerdem A. (2008) (with R. S. Albayrak) Coding and Decoding Emergent Social Structure: A Meta
Language Approach, presented at 1st ICC Workshop on Complexity in Social Systems, Lisbon,
Portugal, January 11-12 2008
58
Süerdem A. (2007) Talking neo-liberal while walking populist: Analysis of social security system
reform discourse in Turkey, Colloque 2007 de la Société québécoise de science politique: Le
politique entre le discours et l’action, workshop : Do They Walk Like They Talk, Université Laval,
Québec City, Quebec, Canada, May 24-25 2007
59
Rich Furman
University of North Carolina, Charlotte, North Carolina USA;
Cynthia Lietz
Arizona State University at the West Campus, Arizona, USA;
Carol L. Langer
University of Wisconsin-Eau Claire, Eau Claire, Wisconsin, USA
The Research Poem (in Social Work)

Introduction
Poetry has only relatively recently been used as a tool of social research and has been used in
various ways throughout the research process. The most popular application of poetry includes its
use as a tool of data representation and presentation. The use of research poems in social work has
been taken up by the authors in the past five years. This refers to a postmodern perspective on
research that uses various forms of poetry to represent the lived experience of social work clients.
Rich Furman’s work in particular has focused on the use poetic forms as qualitative data citing such
terms as research, interpretative and autoethnographic poems. In contrast to previous uses, in the
work of Furman (2004b), one finds poetry embedded in nearly all phases of the research process.
Based on practices from expressive arts research and more traditional qualitative methods, the
research poem can present evocative, powerful insights that can both express and interpret the lived
experience of the research participants or social work clients in this case.
The term research poem connotes the use of poetry less for expressive and literary means, and more
for the purposes of generating or presenting data. The difference between a literary poem and a
research poem can be seen in the position of the author to the data. Whereas research poems might
borrow methods from literary poems, as is the case in the present study, they are written with the
expressed purpose of presenting data that remain faithful to the essence of the text, experience, or
phenomena being represented. Literary poems do not necessarily follow this mandate, as they might
incorporate fantasy, shift reality to incorporate alternative perspectives, and even alter the original
experience in fundamental ways.
In this line of work though, poetry is used as the data from which narrative reflections are written.
These reflections serve as data as well as analysis of the original ethnographic verse. The various
types of data provide different lenses through which to understand various issues. In addition,
poetry is used as a tool of data analysis, where the authors create research poems.
In particular, the process suggested and followed by the authors evolves in the following way: Two
sets of three poems are presented here. Poems are written about clients/research participants aspects
of life studied. The author has three objectives for writing these original poems: (a) to represent
faithfully the salient affective and psychosocial issues that he/she encounters in his/her interactions,
(b) to create aesthetically satisfying poems, and (c) as a means for the author of exploring his
perceptions and feelings about the complex, personal issues implicated in cross-cultural and
international helping in this case. These poems can be viewed in much the same way as qualitative
data, as an exploration of the lived experience of the research subject/participant. The poems are
data about both subject and author, and, about the relationship between them. These poems are texts
that are subsequently analysed. The methods of data employed can used for analysing other types of
text, for instance original ethnographic transcripts, newspapers, and other media text, as well as any
60
other data that can be transcribed into written form. The subsequent research poems are derived
using the original poem as data. For example, treating the poems itself as text, Furman analyses
them for themes using traditional open and axial coding methods. During the first round of coding,
the researcher reads the poems without the intent to develop codes or themes. The goal is to
familiarise himself with the text. During the second round, general impressions are noted and
written in the margins of the text. The third round of coding consists of a line-by-line analysis of the
text. During the fourth round of coding, general themes are induced from the identified codes. The
researcher conducts a fifth round of coding using the identical method with the intent to discover
any uncovered themes and to ensure accurate and adequate representation of the themes.
Once these themes are identified, the researcher undertakes the process of representing some of the
observed themes in different poetic forms. Furman posits that different poetic forms would produce
different effects. For example, the use of the ‘pantoum’ and its repetitive lines allows for the
repetition of salient or emotionally evocative themes. The use of the ‘tanka’, which is characterised
by an extreme economy of words, forces the researcher to make decisions about what data should
be included and what may be left out. In many qualitative studies, the method of data presentation is
taken for granted and is viewed as a natural by-product of data collection and analysis. By
experimenting with re-presenting these data into different forms, the authors hope to stimulate
similar experimentation and attention to strategies of data re-presentation.
Overall therefore, the methods proposed can be viewed in two ways: first, as a process that can be
used to help social workers develop empathy and understanding of their clients; and second, as a
method of data reduction that re-represents data for the purposes of research, which is in line with
previous research work incorporating poetry.
Examples of the Method
Langer and Furman (2004) have used formal poetry as a tool of data analysis. The authors wrote
Japanese tankas as a means of representing interview data about the experience of ethnic identity of
a biracial Native woman. In the creation of their research poems, the authors began their process by
engaging in traditional thematic analysis, whereby they mined the data for codes and themes. Once
these themes and codes were identified, the authors found key passages within the text that
represented each theme. These words were than arranged into the form of a Japanese tanka.
The rationale is explained in the distinction between academic and social work. The authors claim
that social workers need research that helps them understand the emotional and contextual realities
of people whose life experiences they work with and which may often be very different from theirs.
Research poems are proposed as a method that can communicate contextual and affective realities
of thick qualitative descriptions yet are condensed enough to be easily consumed by community-
based social workers and human service practitioners, who might not be familiarised in research.
This refers to the development of practice wisdom as the transaction between scientific information
and that of the phenomenological experience that is derived from interactions with clients.
Furman was inspired to conduct his autoethnographic study of his father’s cancer through the use of
poetic forms.
Social Work, Psychology, Sociology, Ethnography, Anthropology, Socio-linguistics
Colleagues
Furman, R., Lietz, C. & Carol L. Langer
61
References
Furman, R., Lietz, C. & C. L. Langer (2006) The Research Poem in International Social Work:
Innovations in Qualitative Methodology. International Journal of Qualitative Methods. Vol. 5, No.
3, pp. 1-8
Furman, R. (2004a) Poetry as qualitative data for exploring social development and human
experience in Latin America. Journal of Latino-Latin American Studies, Vol. 1, No. 3, 81-103
Furman, R. (2004b) Using poetry and narrative as qualitative data: Exploring a father’s cancer
through poetry. Family, Systems & Health, Vol. 22, No. 2, pp.162-170
Furman, R. (2006a) Autoethnographic poems and narrative reflections: A qualitative study on the
death of a companion animal. Journal of Family Social Work, Vol. 9, No. 4, pp. 23-38
Furman, R. (2006b). Poetic forms and structures in qualitative health research. Qualitative Health
Research, 16, pp. 560-566
Langer, C., & Furman, R. (2004) Exploring identity and assimilation. Research and interpretative
poems. Forum: Qualitative Social Research, 5(2). Retrieved September 11, 2006, from http://
www.qualitative-research.net/fqs-texte/2-04/2-04langerfurmane. htm
Lietz, C., Langer, C. & Furman, R. (in press) Establishing trustworthiness in social work research:
Implications from a study regarding spirituality. Qualitative Social Work
62
Christina Sinding
McMaster University, Hamilton, Ontario, Canada
Ross Gray
University of Toronto, Toronto, Canada
Vrenia Ivonoffski
Ryerson University, Toronto, Canada
Research-based drama
Introduction
Research-based drama as a methodological approach is based on the premise that one never leaves
the stage. Performance is the way lives are lived and it should then be integrated into how social
knowledge is acquired and fed back into the community. In other words, performance is treated
both as a form of publishing research and as hermeneutics.
Research-based drama/theatre refers to a drama or theatre that is based in research. This may take
the form of:
i. A drama/theatre based on research conducted previously in order to write the script;
ii. A drama/theatre developing as part of research;
iii. A drama/theatre scripted to act as action/transformative or intervention research.
In Sinding and Gray’s case all three forms are present and are conceptually based on the
hermeneutic sense of performance as generative of meaning. In other words research-based drama
refers to the merge of dramaturgical and research techniques that work best with some types of
research rather than others. It also refers to merging participants in different categories and
positionings such as social actors, drama actors and researchers.
The process evolves as follows: primary research takes place using conventional methods of data
collection and analysis such as interviews or focus groups, thematic or discourse analysis. Themes
are then reflected in script writing which also incorporates features for creating and performing a
dramatic production – in other words, translating research into drama. The actors are called to
perform a research-based text. This refers to professional actors but also may include (and often
does) lay social actors or research participants whose data constitute the primary material on which
he script is developed. Importantly, one of the features of the drama is that roles may alternate,
especially if the understanding and embodiment of different subjectivities is sought. In addition,
versions may be modified to fit the needs of different audiences. In the performative setting,
perhaps even more clearly than in most qualitative research undertakings, participants are expected
not only to speak about but also to witness, enter into, and take on roles. As actors, they are required
to study for and enact a part. As ‘understudies’ they are asked to understand their roles and to
anticipate the possibility that they may someday be required to live it.
Finally, the processes of research, scripting and performance are documented (journal records and
follow-up interviews with different categories of participants involved).
The extent to which this is used for psychotherapy raises ethical concerns with reference to issues of
empowerment or disempowerment of participants. However, the researchers claim that the
therapeutic (or not) aspect of this type of work misses the point. Therefore, the major contribution is
the different kind of insights this method brings to the study of complex or sensitive phenomena as
63
it blends expert discourse, everyday discourse and art (discourse) on the basis of ‘performance’,
lives acted out.
Staging Breast Cancer, Rehearsing Metastatic Disease: linking research to drama. This took the
form of a drama titled: Handle With Care? Living With Metastatic Breast Cancer which was
developed from qualitative research findings and designed to share these findings with a broad
audience. The final script of Handle With Care? is comprised in no small part of dialogue recorded
during the focus group meetings. The drama is a form of scholarly representation.
Inspiration is derived from they ways in which performance is understood – as a way of knowing –
that also constitutes intervention involving the social group studied in performative ways. As an
epistemological and theatrical undertaking this type of work plays on multiple and sometimes
contrary cultural meanings of closeness. Closeness may contain risk, danger, and vulnerability
whereas distance contains meanings around both professionalism and safety. These meanings relate
to discussions about ethical research practice mentioned above.
Public Health Research, Social Work, Psychology, Sociology, Ethnography
Colleagues
Sinding, C., Gray, R., Fitch, M., & Greenberg, M., Vrenia Ivonoffski
Actors from Act II Studio at Ryerson University; and cancer survivors and activists who were
consultants and test audiences for the dramas.
References
References related to the research-based dramas:
Handle with Care? Women living with metastatic breast cancer
No Big Deal? (about the experiences of men with prostate cancer and their wives)
Ladies in Waiting: A Play About Life After Breast Cancer
The overview:
Gray, R., Sinding, C., Ivonoffski, V., Fitch, M., Hampson, A., & Greenberg, M. (2000). The Use of
Research-Based Theatre in a Project Related to Metastatic Breast Cancer. Health Expectations, 3,
137-144.
The whole story:
Gray, R., & Sinding, C. (2002). Standing Ovation: Performing Social Science Research about
Cancer. Walnut Creek, CA: AltaMira Press.
** The book comes with a video of Handle with Care? Women living with metastatic breast cancer
and No Big Deal? (about the experiences of men with prostate cancer and their wives). Available
through Mills library.
The documentary: http://www.fanlight.com/catalog/films/354_hcwlu.php
How Can We Love You? Behind the Scenes With the Play Handle With Care By Laura Sky Sky
Works Charitable Foundation Available at the Health Sciences library.
64
The scripts:
Ivonoffski, V., & Gray, R. (1998). Handle with Care? Women living with metastatic breast cancer.
Toronto: Toronto Sunnybrook Regional Cancer Centre.
Ivonoffski, V., & Gray, R. (1999). No Big Deal? Toronto: Toronto Sunnybrook Regional Cancer
Centre.
The process of creating the drama:
Gray, R., Ivonoffski, V., & Sinding, C. (2002). Making a mess and spreading it around: Articulation
of an approach to research-based drama. In A. Bochner & C. Ellis (Eds.), Ethnographically
Speaking: Autoethnography, literature and aesthetics (pp. 57-75). Walnut Creek, CA: AltaMira
Press.
Sinding, C., & Gray, R. (2002). Coming to knowledge about metastatic breast cancer: A drama.
Resources for Feminist Research, 29(3/4).
Sinding, C., Gray, R., Fitch, M., & Greenberg, M. (2002). Staging Breast Cancer, Rehearsing
Metastatic Disease. Qualitative Health Research, 12(1), 61-73.
Gray, R. (2000). Graduate school never prepared me for this: reflections on the challenges of
research-based theatre. Reflective Practice, 1(3), 377-390.
The ethical issues:
Sinding, C., Gray, R., & Nisker, J. (accepted). Ethical issues and issues of ethics. In J. G. Knowles
& A. L. Cole (Eds.), Handbook of the Arts in Qualitative Social Science Research. Thousand Oaks:
Sage.
What we learned about illness experience, and cancer in culture, from the dramas:
Gray, R., Sinding, C., & Fitch, M. (2001). Navigating the social context of metastatic breast cancer:
Reflections on a project linking research to drama. Health, 5(2), 233-248.
Sinding, C., & Gray, R. (2005). Active Aging - Spunky Survivorship? Discourses and experiences
of the years beyond breast cancer. Journal of Aging Studies, 19, 147-161.
How the dramas made a difference (and problematizing 'making a difference!')
Gray, R., Fitch, M., Labrecque, M., & Greenberg, M. (2003). Reactions of Health Professionals to a
Research-Based Theatre Production. Journal of Cancer Education, 18(4), 223 - 229.
Sinding, C., Gray, R., Grassau, P., Damianakis, F., & Hampson, A. (2006). Audience Responses to
a Research-Based Drama about Life after Breast Cancer. Psycho-Oncology, 15, 694-700.
Sinding, C., Gould, J., & Gray, R. (forthcoming 2009). Making a Difference with Research. In J.
Gould, J. Nelson & S. Keller-Olaman (Eds.), Method and Meaning: Participatory Research with
Marginalized Cancer Communities. Toronto: University of Toronto Press.
65
Transformative
The fifth section presents transformative research practices. The inclusion of a sample of work
below is to stress the role of transformative work on methodological developments. The areas and
disciplines these are observed in are mainly ethnography, social anthropology and social work.
66
Julie Park
The University of Auckland, New Zealand
The impact of multi-sited ethnography on researcher involvement

Introduction
Julie Park has been involved in experimenting with ethnographic approaches to social
anthropological concerns, and health in particular.
Description and Explanation of Method
The work presents novelty in how the project is put together, vis a vis the kind of involvement by
academics and students. In particular, both the extent of researcher involvement with the
community studied and writing up about departs from conventional ethnographic research. Namely,
the researchers (PhD students) were involved in all sorts of jobs for the well being of these
communities.
The projects are participatory, transformative and involve mixed methods such as life history
interviews, participant observation, clinics, interviews, note taking, group discussions, community
interviews, visual prompts with sensitive topics, media and policy analysis. The analysis included
thematic analysis and Geographic Information System (GIS) of networks. In this sense it is the
work is multi-cited and interdisciplinary with a ‘syndemic’ orientation. Using a syndemic
framework enables the examination of the clustering of health conditions in the context of history
and culture. By selecting these conditions the study takes a comprehensive approach to health
ecology incorporating historical political economy, cultural and biological interactions.
Ethnographic work on tuberculosis
Colleagues
The research team is based in anthropology, and draws in colleagues, students and perspectives
from population health, history, geography, development studies and policy disciplines
References
Park, J. & Littleton, J. (2007) Ethnography Plus in Tuberculosis Research. SITES Vol. 4, No. 1, pp.
3-23
67
Elizabeth Whitmore
Carleton University, Ottawa, Canada
Participatory Learning and Action

Introduction
Elizabeth Whitmore has been involved in participatory action research, participatory evaluation and
community-based research for the past 25 years during which time she has had to develop various
tools and techniques to ‘involve’ the community researched.
Her work involves hiring members of the community studied to do the research on the principles of
Participatory Learning and Action (PLA). This work has involved conventional and non-
conventional methods. In data collection non-conventional included the use of sticky notes while
during the research process and analysis user-friendly language is used for the research terms
involved. In the presentation stage then these terms are substituted with formal ones. On one of the
cases researched (street youth dropping centre) the participants compiled a kit to teach others to do
this kind of work.
Street youth drop in centre
Governments and universities are picking up this type of research very slowly, too slowly in
Whitmore’s opinion to incur any serious transformations.
References
Whitmore, E. (ed) (1998) Understanding and practicing participatory evaluation. San Francisco:
Jossey-Bass Publishers
See also www.manifestationjournal.org
68
Margareta Hyden
Linkoping University, Linkoping, Sweden
Narrating Sensitive Topics: Applied and Basic Research

Introduction
Margareta Hydén is a professor of social work and a licensed psychotherapist. Her primary research
areas have been battered women’s, battering men’s, and their children’s experiences and responses
to interpersonal violence. Recently her interests have expanded to include the social networks of the
involved, and their responses to the violence. Her awareness of the fact that a research interview
concerning sensitive issues forms a complex power relation has been a driving force for the
development of the method. Her primary aim has been to develop research that could be used in the
efforts of taking action against interpersonal violence.
Her contribution in conducting transformative research on interpersonal violence has been both to
meet the ethical concerns and methodological challenges concerning the study of sensitive topics, as
well as open up the empirical base of narrative research. In the first case, she offers a suitable
physical space and an open discursive space by conducting a teller- focused way of interviewing
which assists the interviewee in his/her way of telling things rather than pushes interviewer’s
agenda. In such an interview, the concepts of “question” and “answer” are to be thought of as a part
of a circular process, with the interviewer and the interviewee trying to make continuing sense of
what they are talking about. Experiences of sensitive events essentially are “untold stories”. Stories
of that type are not well suited to a format of pre-prepared questions followed by brief answers. The
method consists of a series of interviews, in order for the interviewer to follow participants of the
process of telling. Hydén has conducted the method in three variations, one for individual
interviews with adults, one focus-group interview with troubled adolescents, and one group-
interview to be used in therapeutic settings for interviewing children. In the second case, she
explores the boundaries of narrative analysis by using it as an approach for understanding
experiences of interpersonal violence. The results are fed back to the community studied for
reflection - yet in this case Hydén observes that this process also empowers the community studied.
Interpersonal violence, Children exposed to violence, Troubled young people
The development of the method was informed by narrative theory in sociology and psychology as
well as of psychotherapeutic practises.
Linking academia and practice-based work, this method could be used for basic research on
sensitive topics, as well as it could be applied for the development of practices. It could be used in a
wide array of research projects, for example in social work, health and human rights research.
References
Hydén, M. (2009) Listening to children’s experiences of being participant witnesses to domestic
violence. In H. Forsberg & Teppo Kröger (eds.) Social Work and Child Welfare Politics, pp. 129-
146 Bristol, UK: The Polity Press
69
Hydén, M. (2008) Narrating Sensitive Topics. In M. Andrews, C. Squire & M. Tamboukou (eds.)
Doing Narrative Research, pp. 121-136. London: SAGE
Hydén, M (2008) “Break-up narratives”. In Hydén, L-C & Brockmeier, J. (eds.) Health, Illness and
Culture: Broken Narratives, pp. 154-168. New York: Routledge
Hydén, M. & Överlien, C. (2005) “Applying narrative analysis to the process of confirming or
disregarding cases of suspected sexual abuse” Child and Family Social Work, 10 (1): 57-65
Hydén, M. (2005) ‘I Must Have Been an Idiot to Let it Go On’: Agency and Positioning in Battered
Women’s Narratives of Leaving. Feminism & Psychology, 15(2): 171-190.
Hydén, M. (2004) ”Doing” Narrative Analysis. I D. K. Padgett (ed.) The Qualitative Research
Experience, pp. 254-272. Belmont, CA: Wadsworth.
Överlien, C. & Hydén M. (2004) “ ‘You want to have done your living if you know what I mean’ -
incarcerated young women speaking about motherhood”. Feminism & Psychology 14 (2): 226-230
70
Software
The sixth section presents software development. It was considered relevant for the purposes of this
report to draw attention to experts outside the UK whose software development activity provides
the professional and academic social science community with research tools. The sample areas and
disciplines presented here are criminology, geography, methodology, social network analysis,
statistics and government.
71
Jerry Ratcliffe
Temple University, Philadelphia, Pennsylvania, USA
Intelligence-Led Policing Software

Description
Jerry Ratcliffe has designed software that corresponds to the Intelligence-Led Policing Model – a
business model and managerial philosophy where data analysis and crime intelligence are pivotal to
an objective, decision-making framework that facilitates crime and problem reduction, disruption
and prevention through strategic management and effective enforcement.
References
Ratcliffe, J. H. (in press) Intelligence‐led policing. In Wortley, R., Mazerolle, L., and Rombouts, S.
(Eds) Environmental Criminology and Crime Analysis. Cullompton, Devon: Willan Publishing:
72
Peter Kleiweg
Rijksuniversiteit, Groningen, Netherlands
Dialectometry: Measures Structure in Geographical Dialect Networks:

Rug/Lo4 Software
Description
Peter Kleiweg’s work focuses on applying computational methods to derive dialect areas without
expert intervention. This work is part of research into dialectology which is conducted as part of
LAMSAS: Linguistic Atlas of the Middle and South Atlantic States, at Rijksuniversiteit,
Groningen. Along with a group of colleagues, his work is based on the assumption that dialectology
must reveal geographic coherence in language variation in order to propose a yardstick with which
to compare measurements made using various parameter settings. Dialectometry then measures the
differences between dialects in ways that may involve many independently varying parameters. The
existence of many parameters of measurement and possible interaction introduces the problem of
how to choose parameter values and combinations. The software tool Rug/Lo4 is developed to:
derive a difference matrix for a set of locations based on sets of strings for each location, derive a
difference matrix for a set of locations based on sets of labels for each location, permutate linguistic
data, perform hierarchical clustering of a difference matrix, perform multidimensional scaling on a
difference matrix, validate the difference matrix and draw various maps.
73
Sharlene Nagy Hesse-Biber
Boston College, Boston, Massachusetts, USA
HyperResearch: Computer-assisted programme for analysing qualitative

data
HyperTranscribe: transcription tool
Description
Sharlene Hesse-Biber is renowned for the work on qualitative methodologies including grounded
theory and the use of technology in qualitative research. She is involved in identifying emergent
research methods for the social sciences. As part of this activity Hesse-Biber has published in the
field of research methodology. She has co-edited Approaches to Qualitative Research (Oxford,
2004) and the Handbook of Emergent Methods (Guilford, 2008), has co-authored The Practice of
Qualitative Research (SAGE, 2005) and Emergent Methods in Social Research (SAGE, 2005) and has
edited the forthcoming Handbook of Emergent Technologies for Social Research (Oxford, 2009).
She co-developed the software programme HyperRESEARCH, a programme for analysing
qualitative data, and the transcription tool HyperTRANSCRIBE (see www.researchware.com).
HyperRESEARCH allows you to work with text, graphics, audio, and video sources and
HyperTRANSCRIBE enables the transcription of most audio or video files with keyboard shortcuts.
Customisable shortcuts allow the text entry in a single keystroke while the media can be played in
segments of definable length to pace transcribing.
74
Christopher McCarty
Director of the Bureau of Economic and Business Research, University of Florida, Gainesville, Florida, USA
Personal Network Analysis & EgoNet Software

Description
Christopher McCarty is survey director of the Bureau of Economic and Business Research at the
University of Florida. Among his areas of research is the development of new methods and tools for
studying personal networks in a transcultural framework and authoring the software EgoNet.
EgoNet is used in the collection, analysis and visualisation of personal network data.
References
McCarty, C., Molina, J. L., Aguilar, C. & Rota, L. (2007) A Comparison of Social Network
Mapping and Personal Network Visualisation. Field Methods. Vol. 19, pp. 145-162
75
Mark Handcock
University of Washington, Seattle, Washington, USA
Description
Dr Handcock’s research involves methodological development based on motivation from questions
in the social sciences. He develops statistical models for the analysis of social network data, spatial
processes and longitudinal data arising in labour economics. Some of the software for modelling
social network data that he has developed are:
• statnet: Software tools for the representation, visualisation, analysis and simulation of social
network data.
• latentnet: Software to fit and evaluate latent position and cluster models for statistical
networks.
• networkis: A Package to Simulate Bipartite Graphs with Fixed Marginals through
Sequential Importance Sampling.
• ergm: A Package to Fit, Simulate and Diagnose Exponential-Family Models for Networks.
• glmc: Fitting Generalised Linear Models Subject to Constraints.
76
Gary King
Harvard University, Cambridge, Massachusetts, USA
Verbal Autopsy Software

Introduction
Professor Gary King and his research group develop and apply statistical methods in social
research, focusing on innovations that span the range from statistical theory to practical application.
His research areas are: Causal Inference, Content Analysis, Data Sharing and Informatics,
Ecological Inference, Event Counts and Duration Models, Incumbency Advantage, International
Conflict, Legislative Redistricting, Missing Data, Mortality Studies, Presidency Research and
Voting Behaviour, Qualitative Research, Rare Events, Survey Research, Unifying Statistical
Analysis. King’s work is widely read across scholarly fields and beyond academia. The statistical
methods and software he developed are used extensively in academia, government, consulting, and
private industry. An example of King’s work presented here is on verbal autopsy methods,
including an open source software that implements these methods.
Description and Explanation of Method
Verbal autopsy is a technique for estimating the cause-of death distribution in populations without
vital registration or other medical death certification. It involves collecting information about
symptoms (including signs and other indicators) from the caretakers of each of a randomly selected
set of deceased in some population of interest, and inferring the cause of death. Inferences in these
data are extrapolated either by physicians from their prior experiences or by statistical analysis of a
second data set from a nearby hospital where information on symptoms from caretakers as well as
validated causes of death are available.
The structure of the inferential problem studied can also be found in application areas fairly distant
from verbal autopsy applications. Some version of the methods may be of use in these areas as well.
For example, a goal of paleodemography is to estimate the age distribution in a large sample of
skeletons from measurements of their physical features by using a small independent reference
group where validated ages are available and skeletal features are also measured (Hoppa and
Vaupel, 2002). The verbal autopsy methods seem to have already proven useful for estimating the
proportion of text documents in each of a set of given categories, using a smaller reference set of
text documents hand coded into the same categories (Hopkins and King, 2007). The methods
introduced here imply that individual level classifiers can greatly reduce the assumptions necessary
for accurate generalisation to test sets with different distributional characteristics.
Other Statistical Methods for Social Research
• CEM: Coarsened Exact Matching
• MatchIt: Nonparametric Preprocessing for Parametric Causal Inference
• JudgeIt II: A Programme for Evaluating Electoral Systems and Redistricting Plans
• ReadMe: Software for Automated Content Analysis
• Amelia II: A Programme for Missing Data
• YourCast: Time Series Cross-Sectional Forecasting
• WhatIf: Software for Evaluating Counterfactuals
• ReLogit: Rare Events Logistic Regression
• CLARIFY: Software for Interpreting and Presenting Statistical Results
• EI: A Programme for Ecological Inference
77
• COUNT: A Programme for Estimating Event Count and Duration Regressions
References
King, G. & Ying, L. (2008) Verbal Autopsy Methods with Multiple Causes of Death. Statistical
Science. Vol. 23, No. 1, pp. 78–91
78
Others
The final section presents other sources of potential innovation. The sample areas and disciplines
cited here are survey methodology and demography, while a case for disciplinary integration is also
described.
79
Marcel Das
CentERdata, Tilburg University, Tilburg, The Netherlands
Facility for online data collection: Advanced Multi-disciplinary Facility

for Measurement and Experimentation in the Social Sciences (MESS)
Description
In 2006, CentERdata Tilburg University, Tilburg, The Netherlands received major funding from the
National Science Foundation (NSF) for the project entitled “An Advanced Multi-disciplinary
Facility for Measurement and Experimentation in the Social Sciences (MESS).
The grant was instituted by the Dutch Cabinet with a view to boosting the Dutch knowledge
economy and the climate for innovation in the country. The proposal was submitted by Arie
Kapteyn, Arthur van Soest, and Marcel Das. The funds have been used to establish a new online
panel of 5,000 Dutch households: the LISS panel (Longitudinal Internet Studies for the Social
sciences). The panel is the core component of the MESS project and is based on a true probability
sample of households drawn from the population register by Statistics Netherlands. Households that
could otherwise not participate are provided with a computer and Internet connection.
The recruitment strategy for the online household panel was based on an experiment in order to
assess five factors: contact mode, incentive amount, timing of the incentive, content of the
information letter and timing of the panel participation request. To meet scientific demands, the
panel is based on a traditional probability sample. It was found that participants for an online panel
based on a probability sample could be recruited quite effectively using telephone contacts
supplemented with face-to-face contacts for households without a known telephone number.
Satisfactory response rates were obtained when a monetary incentive of 10 euros was sent in
advance of participation. These rates were independent of information provided in an
accompanying letter.
The LISS panel focuses on fundamental longitudinal research and provides a laboratory for the
development and testing of new, innovative research techniques. The first questionnaires presented
to the LISS panel were fielded in the autumn of 2007. CentERdata invites universities, research
schools, and individual academics to submit research proposals, which they can carry out in this
panel free of charge. A Board of Overseers protects the project’s quality and will assess and select
the submitted research proposals. The Board of Overseers is composed of professors from a variety
of disciplines and universities.
The MESS project is strongly geared to integrating different academic disciplines. Economics and
the social sciences often collaborate in survey-based research, and now it means to branch out into
the medical and biomedical sciences and the behavioural sciences. It also aims to investigate themes
that are current in other national household panels in Europe and the US, which will facilitate
international comparison.
Colleagues
Annette Scherpenzeel, project leader of MESS
References
Scherpenzeel, A. (2008) Recruiting a probability sample for an online panel: effects of contact
mode, incentives and information. CentERdata, Tilburg: Tilburg University
http://lissdata.nl
80
Gabriele Bammer
The Australian National University, Canberra, Australia
Integration and Implementation Sciences (I2S)

Description
Gabriele Bammer has been developing a new cross-cutting discipline called Integration and
Implementation Sciences (I2S) based on a set of conceptual and methods skills. I2S covers four
domains, namely concepts and methods to enhance:
1. fresh thinking on intractable problems;
2. integration of disciplinary and stakeholder knowledge;
3. understanding and management of ignorance and uncertainty; and
4. the provision of research support for decision making and practice change.
I2S provides (a) the hub around which research institutions can organise teams to investigate real
world problems, (b) a baseline level of quality for such work, (c) an avenue for transmitting new
theory and methods between groups focusing on different real-world problems, and (d) a home for
methodologies addressing recurrent concerns in tackling complex problems that are not the
province of any discipline or practice area. This more about a framework of research than a method
and it is based on six key questions:
1. What is the integration aiming to achieve and who is intended to benefit?
2. What is being integrated?
3. Who is doing the integration?
4. How is the integration being undertaken?
5. What is the context for the integration?
6. What is the outcome of the integration?
Research integration has three intertwined dimensions:
1. the synthesis of a range of knowledge, information and perspectives to improve
understanding of an issue or problem;
2. using research knowledge to support decision makers in various domains, including public
policy, business, professional practice and community activism; and
3. using research knowledge to underpin the implementation of decisions in effective social
change.
It involves integrative methods (in the making). This approach may involve five classes of methods:
(i) dialogue-based; for example, there are a wide range of dialogue approaches such as
Delphi Technique, Nominal Group Technique and principled negotiation. For
example, the Delphi Technique is useful for integrating judgements, Nominal Group
Technique for synthesising different experiences and principled negotiation for
integrating different interests.
(ii) model-based; for example, there are many types of modelling from conceptual
mapping to formal system dynamics or agent-based models.
(iii) product-based; for example, designing and developing a new product can provide an
effective way of integrating different perspectives. Building the atomic bomb
illustrates this integrative method.
(iv) vision-based, for example, having an ideal to work towards. This is illustrated by the
World Commission on Dams, which was guided by the idea of “development
effectiveness”, in other words equitable and sustainable human development (World
81
Commission on Dams, 2000, p. xxxiii). Working towards the vision can then involve
one or more of the other methods outlined here. And
(v) common metric based, for example converting everything to a dollar value, which
then allows integration through a range of methods from simple arithmetic to
modelling. Other common metrics include disability-adjusted life years (DALYs)
which are used in public health
(http://www.who.int/healthinfo/boddaly/en/index.html) and the ecological footprint
(http://www.footprintnetwork.org/).
Effectively tackling real-world problems requires a new type of researcher, who can enhance
collaboration between discipline and practice experts. Such researchers need a solid foundation in a
set of conceptual and methods skills. In addition to research that advances understanding through a
single discipline, there is growing appreciation of the importance of cross-disciplinary research that
focuses on real-world problems.
Colleagues
Integration and Implementation Sciences Network (www.anu.edu.au/iisn):
References
Bammer, G. (2006) A systematic approach to integration in research. Integration Insights #1,
September. Available at www.anu.edu.au/iisn.
Bammer, G. (2006) Principled negotiation – a method for integrating interests. Integration Insights
#3, November. Available at www.anu.edu.au/iisn.
Bammer, G., McDonald, D., Deane, P. (2007) Dialogue methods for research integration.
Integration Insights #4, May. Available at www.anu.edu.au/iisn.
Bammer, G. (2008) The case for a new discipline of Integration and Implementation Sciences (I2S).
Integration Insights #6, May. Available at www.anu.edu.au/iisn.
82
Hayward Derrick Horton
Critical Demography
Introduction
Critical Demography was founded by Dr. Hayward Derrick Horton, Associate Professor of
Sociology at the State University of New York at Albany in 1996. In 2000, he received a grant by
the National Science Foundation to develop it further.
Description
Critical Demography is not quite a method but a paradigm which facilitates the development of
theories, methods and concepts that do not neatly fit within the boundaries of the prevailing
paradigm, conventional demography. Critical Demography makes explicit the manner in which the
social structure differentiates dominant and subordinate populations. Thus, critical demography
necessitates discussions of population control and population power. For example, one cannot
speak of race and sex without likewise articulating the impact of racism and sexism. In sum, critical
demography reintroduces and articulates the nature of the social structure and how it impacts upon
population phenomena.
The inspiration came when Professor Horton submitted a paper to be included as a chapter in a
proposed volume on racial and ethnic demography: “Toward a Critical Demography of Race and
Ethnicity: Introduction of the ‘R’ Word”. The reaction to the paper by the editors was so explosive
that it became clear that the issue at hand was greater than the inclusion of racism in demographic
analysis. There need for a new paradigm that would promote the development and nurturing of a
broad range of ideas that “don’t fit” within the dominant paradigm of the field was identified.
Professor Horton came to the realisation that he had emphasized the wrong issue from the “R
Word” paper’s title. It was in fact Critical Demography that was being introduced and not racism.
Colleagues
http://www.albany.edu/~hdh/criticaldemography/index.html
References
Horton, D. H. (1999) Critical Demography: The Paradigm of the Future? Sociological Forum, Vol.
14, No. 3, pp. 363-367
Horton, H. D. & Sykes, L. (2003) Critical Demography and the Measurement of Racism: A
Reconsideration of Wealth, Status and Power. Race & Society. Vol. 4, pp. 207-217
83
5. DISCUSSION AND REFLECTION
In this section some general observations about the innovative methods summarised in the previous
section will be attempted by clustering, grouping and reflecting on the types of method,
innovations, areas/disciplines, geographies and structures that have been found.
First, a schematic representation of the methods is presented including their approaches,
methodologies, methods, techniques, tools, software, panel structures and disciplinary
developments (see Table 1). This is arranged on the basis of the vertical axis of research methods
classification categories, presented hierarchically (see Figure 1).
Second, a map depicting geographical concentration of Innovations in Social Science Research
Methods (ISSRM) is presented (see Map 1).
Third, a timeline is attempted to locate the inception of the innovative methods summarised in the
report chronologically (see Figure 2).
Finally, a view of the key networks identified in the search for Innovations in Social Science
Research Methods outside the UK is examined (see Figure 3).
84
TYPE NAME DISCIPLINE(S) QUANTITATIVE/QUALITATIVE/MIXED GEOGRAPHY INSTITUTION INNOVATION
Genetically Transmitted Political Politics QUANTITATIVE VA, USA Academic CB
& Value Orientations
Approaches Netnography Management MIXED CANADA Academic CB/TI
Research-based Drama Psychology QUALITATIVE CANADA Academic/Professional INTRA
Journalism as Research & Journalism MIXED AUSTRALIA Academic/Professional CB
Reporting
Folkbiology MIXED TN, USA Academic CB
Kinship Algebra Expert System Anthropology QUANTITATIVE Academic CB/TI
(KAES)
Methodologies Intuitive Inquiry Psychology CA, USA Academic/Professional
Situational Analysis Sociology QUALITATIVE Academic
Water Cooler Logic Professional CB
Ethnography as a Non-linear Ethnography MIXED NM, USA Academic/Professional
Dynamic System
Institutional Ethnography QUALITATIVE CANADA Academic/Professional
Appreciative Inquiry as a Research Management MIXED CANADA INTRA
Method
Methods Research Poem Social Work QUALITATIVE AZ, NC, WI, USA Academic/Professional CB
Reflecting Interview Psychology QUALITATIVE SWISS, USA INTRA
Semiotic Analysis and Mapping Multidisciplinary MIXED TURKEY Academic INTRA
for Agent-Based Modelling
Orbital Decomposition WI, USA Academic CB
New Item Response Theory Psychology MN, USA Academic INTRA
Techniques Models: Tests without Items
Internet Survey Software Tools QUANTITATIVE Research CB
Methods for Achieving Survey Methodology GERMANY Research CB
Equivalence of Samples in Cross-
National Surveys
Intermediate Data Structure (IDS) History QUANTITATIVE Academic
Tools Survey Interviewing using Survey Methodology MI, USA Academic TI
Animated Agents (Avatars) MIXED
WeSearch Communication Studies ITALY Academic
Statistical Software for Social Government MA, USA Academic CB/TI
Research
Software Statistical Software for the Statistics QUANTITATIVE WA, USA Academic
Analysis of Social Network Data TI
Personal Network Analysis: Business & Economics FL, USA Academic
EgoNet Software
TYPE NAME DISCIPLINE(S) GEOGRAPHY INSTITUTION INNOVATION
Intelligence-led Policing Software Criminology PA, USA Academic
Dialectometrics: Rug/Lo4 Geography QUANTITATIVE NETHERLANDS Academic
Software Software TI
HyperResearch: Computer-assisted Sociology QUALITATIVE MA, USA Academic
programme for analysing
Qualitative Data
Advanced Multi-disciplinary Survey Methodology MIXED NETHERLANDS Research CB/TI
Panel Facility for Measurement and
Structure Experimentation in the Social
Sciences (MESS)
Critical Demography Demography MIXED NY, USA Academic
Disciplines Integration & Implementation Interdisciplinary MIXED AUSTRALIA Academic CB
Sciences (I2S)
Note: CB is the acronym for Crossing Boundaries; TI is the acronym for Technological Innovation; INTRA is the acronym for Intra-disciplinary
and Intra-methodological. Acronyms in the geography column represent the state initials in the US. Colouring is used to highlight clusters within
categories.
Table 1. ISSRM outside the UK
86
Table 1 indicates that the disciplines and areas primarily affected are ethnography, anthropology,
psychology and survey methodology. In particular, developments in research methodologies figure
higher in ethnography4 followed by anthropology while developments in research techniques are
shared between psychology and survey methodology. Overall, across the research methods
categories, psychology figures prominently in methodological developments followed by survey
methodology, ethnography, sociology and management. With the exception of software, panel
structure and discipline development, the types of research that the methodological innovations and
developments may be classified into are mixed (N=8), qualitative (N=7) and quantitative (N=7),
though these are treated as a continuum rather than discreet categories in this report.
Developments in quantitative methods figure higher in survey methodology while developments in
qualitative methods figure higher in psychology followed by sociology and then social work and
ethnography5. If one takes mixed methods into account on the basis of including qualitative
research, psychology retains its prominence in fostering methodological developments in qualitative
research, ethnography is strengthened, while management and survey methodology are added into
the mix. Developments is mixed methods are mainly shared between ethnography and survey
methodology.
Overall, the institutional structures identified as ‘hosting’ developments are primarily Academic
(N=19) followed by both Academic and Professional (N=7), then Research Centres and finally
Professional and Consultancy institutions. The institutional structures identified as ‘hosting’
developments in qualitative research are primarily both Academic and Professional (N=5), and then
Academic (N=1) and Professional/Consultancy institutions (N=1). This marks a clear difference
between the developments identified in qualitative research and overall methodological
developments as in the former hosting institutions are primarily both Academic and Professional
(N=5) while in the latter hosting institutions are primarily Academic (N=19). On the basis of the
results of the project therefore, it could be argued that developments in qualitative research and
methods derive from an interventionist, participatory and consultancy-type of approach to research.
The institutional structures identified as ‘hosting’ developments in quantitative and mixed methods
research are primarily Academic.
Finally, with the exception of software, panel structure and discipline development, the majority of
the innovations are a consequence of working across disciplinary boundaries, followed by
developments within methods and disciplines and then by developments in technology. However, if
one includes software, panel structure and discipline development in the count, there is then an
(nearly) equal distribution between innovations from working across disciplinary boundaries and
from developments in technology
The geographic concentration is depicted and discussed below.
4
It should be noted here that sociology may also be included in the disciplines from which developments derive by taking the
developers specialisation into account which is true in this case for both situational analysis and institutional ethnography.
5
It should be noted here that sociology may figure equally high as psychology in fostering methodological innovations in
qualitative methods if the developers’ specialisation is taken into account (institutional ethnography).
Map 1.Concentrations of ISSRM outside the UK
Map 1 depicts the geographical concentrations of developments in Social Science Research

Methods (SSRM). The map indicates a concentration of these developments – as defined for the
purposes of this project (see Section 1) – in North America, the United States in particular. The East
Coast features the highest concentration of innovations and developments in SSRM, represented by
such tools such as Virtual Anthropomorphic Interviewers (Avatars) and Intermediate Data Structure
(IDS), methods such as the Research Poem, methodologies such as Folkbiology and Genetically
Transmitted Political and Value Orientations. In the Mid West, methods and techniques in
Psychology have been developed, including New Item Response Models, Orbital Decomposition
and Reflecting Interview. The West Coast features methodologies practised in academic and non-
academic institutions (consultancy services and centres); in the first case this refers to Situational
Analysis and Kinship Algebra Expert System (KAES); in the second case this refers to the Water
Cooler Logic and Ethnography as a Non-linear Dynamic System, while Intuitive Inquiry is
practised in both types of institutional contexts. Overall, the distribution between East and West
coast is weighted towards the former.
Canada features as the geographical locus for methodologies and methods derived from Marketing
and Consumer Research – Netnography – and Management and Organisation Studies –
Appreciative Inquiry as a research method. More purely qualitative and institutionally transgressive
methodologies and approaches such as Institutional Ethnography, Research-based Drama and
transformative research in general are also located there.
In Europe, noting the exclusion of the UK for the purposes of the project, there seems to be a
concentration of technique development in survey methodology at the Gesis-Liebniz Institute in
Germany for both online surveys and large cross-national surveys. Web 2.0 tool development is
spotted in Italy – WeSearch. The Netherlands is depicted for the development of Appreciative
Inquiry as a research method, software and panel structure development – Advanced Multi-
disciplinary Facility for Measurement and Experimentation in the Social Sciences (MESS).
Switzerland has Reflecting Interviews; Turkey, the practice of an innovative approach to Agent-
based modelling and Sweden, Transformative research.
88
In Australasia, Australia features for the practice of Journalism as both Research and Reporting and
for the development of Integration and Implementation Sciences (I2S) while New Zealand is
included here for the practice of transformative research.
It should be noted that the map is inclusive as apart from methodological developments and
innovations it also depicts spots active in software, panel structure and discipline development, and
transformative research.
Moving from geography to time of development, Figure 2 shows the chronological inception of the
methodological developments identified as part of this project.
89
1984 KAES
Institutional Ethnography
1987
1996 Intuitive Inquiry
Netnography
1998
Water Cooler
2000
Research-based Drama
Figure 2. ISSRM Timeline
Folkbiology
IDS
2003 Sampling Equivalence
Research Poem Dynamic Ethnography
2004 New IRT
Internet Survey Tools
Avatars Situational Analysis
2005
Genetic Political Values
Orbital Decomp.
2006 Reflecting Interview
WeSearch
2007
Semiotics & ABM
2008 Appreciative Inquiry

The four methodologies distributed between 1984 and 1998 are included in this report for being
reformulated and adapted since their original development and/or for not having penetrated as such
the UK SSRM. This may tell a tale of marginal diffusion (KAES primarily and Intuitive Inquiry
secondarily), or a tale of development amidst resistance (scholarly and/or related to parallel
developments in research – Institutional Ethnography and Netnography). The rest of the
developments identified here are concentrated in the period between 2003-2006, with most
appearing in 2004 (N=5), followed by 2003 (N=3) and 2005 (N=3), then 2000 (N=2) and 2006
(N=2) and finally 2007 (N=1) and 2008 (N=1). Taking this into account, diffusion for some –
Avatars, IDS, WeSearch, Appreciative Inquiry as a Research Method etc. – remains to be seen. For
methodologies such as Water Cooler Logic, Folkbiology and Ethnography as a Non-linear Dynamic
System, the process of diffusion appears to be more long term and ongoing, while the impact of
techniques and tools is only detectable in more micro applications. This report itself contributes to
raising awareness for methodological innovations and developments outside the UK and may, thus,
ultimately have an impact on their penetration.
Finally and with a view to awareness, penetration, diffusion and impact, a network view of the key
networks identified in the search for methodological innovations and developments is shown in
Figure 3.
Michael Schober
Fred Conrad
Wolfgang Bandilla Jastine Cassell Eleanor Singer
Lars Kazcmirek Wil Dijkstra Mick Couper

Norbert Schwarz Jon Krosnick
Bob Belli Nora Schaeffer Peter Miller
Marcel Das
Johannes van der Zouwen Hanneke Houtkoop-Steenstra

Douglas Maynard George Alter
Juliet Corbin Sabine Hader Seppo Laaksonen Myron Gutmann Kees Mandemakers
Kathy Charmaz
Janice Morse Adele Clarke Peter Lynn Siegfried Gabler
Anselm Strauss Wendy Haight
Libby Bishop Peter Sommerfeld

Ian Shaw
Helga Wild Chris Darrouzet Jennifer Greene

Steven Brown
Sue Wilkinson Russell Belk
Mary Louise Codwell
Robert Kozinets
Miriam Caterrall
Gervase Bushe Jastine Cassell
Pauline Maclaran
Paul Henry
Michael Avital
Dian Marie Hosking John Shotter
David Cooperider
Kenneth Gergen
Jan Reed Sheila McNamee
Richard Bolland
Figure 3. ISSRM Key Networks
Note: The names highlighted indicate a shared link, key figure or point of departure in the network; Arrows indicate the direction of the
relationship. Overall, the Institute for Social Research, University of Michigan figures prominently in concentration of experts and
methodological developments; the National Science Foundation (NSF) appears in one quarter of the methodological innovations in the US.
Reflecting on the developments in Social Science Research Methods identified as part of this
project, there are a few points to be made. First, it is acknowledged that results occur within the
context in which they are framed. Thus, the methodological innovations and developments
presented here reflect the way in which methodological innovations were defined in the first place.
Second, the methodology employed also impacts on the types of results identified. Other
methodologies (such as the use of link analysis, citation analysis or bibliometrics) may not fully
replicate the results presented in this report. This project sought deep understanding of
methodological developments and processes involved in them, which is not achieved through the
use of link analysis or bibliometrics. Nevertheless, link analysis or bibliometrics would provide
interesting additional insights for further research. Fourth, access restrictions due to language
barriers and academic / research cultures particularities are acknowledged. In particular, it is
acknowledged that the methodological innovations and developments presented here are located in
the so-called ‘developed’ world. Fifth, while the decision to present the particular methodological
innovations in this report was based on a meticulous and critical background reading of material on
them, it was not the purpose of this project to reflect on the methods themselves in terms of their
ontological and epistemological assumptions and methodological implications. Finally, the
methodological innovations are presented here for not having penetrated the UK. The penetration
and diffusion of the ones still in the development process remains to be seen, while the penetration
and diffusion of existing ones appears to be more marginal.
6. CONCLUSION
To sum up, we can make four main generalisations about innovative research methodologies.
Firstly, innovative methodologies usually entail the use of one or more technological innovations
(visual, digital or online). This could be new software (e.g. animated agents, unitising datasets,
statistical analysis or CAQDAS) or new online methods to conduct research (e.g. Netnography,
analysing RSS feed and paradata collected online). Secondly, innovative methodologies entail
crossing boundaries, either those between disciplines or between existing methods or both. This is
observed in the combination of disciplines and methods such as in ethnography and anthropology
(where complexity theory seems to drive a significant amount of development in research practice),
psychology, social medicine and politics among others. Thirdly, innovative methodologies entail
the use of existing theoretical approaches and methods in reformed or mixed and applied ways. This
is observed in participatory methods, action research, professional work, social and consultancy
work (e.g. appreciative inquiry, intuitive inquiry and the water cooler logic). Finally, innovative
methodologies reside both inside traditional academic institutions (universities) and outside
(research centres, institutes, consultancy agencies and organisations), but even in the latter case
methods developers and experts have academic backgrounds and close academic ties.
Overall, psychology figured prominently in methodological innovations and developments followed
by survey methodology, ethnography, sociology and management. These developments were
classified into mixed, qualitative and quantitative types of research, treating these classifications as
part of a continuum rather than as discreet categories. The institutional structures identified as
‘hosting’ these developments are primarily Academic followed by both Academic and Professional,
then Research Centres and finally Professional and Consultancy institutions. The majority of the
innovations are a consequence of working across disciplinary boundaries, followed by
developments within methods and disciplines and then by developments in technology. Innovations
were mainly spotted in North America – the USA and Canada – Italy, Germany and the
Netherlands.
The observations derived from this project about ISSRM will be further reflected on and
disseminated as part of conferences in the UK and Europe. A workshop to bring together the
methods developers and experts identified in this report is being arranged, which will emphasise
talking about the experience and processes involved in developing methods as well as the
mechanisms, networks and structures involved in their diffusion. The workshop is planned to be
adjacent to the NCRM Research Methods Festival to be held in July 2010.
94
7. REFERENCES
Bardsley, N. & Wiles, R. (2006) A Consultation to Identify the Research Needs in Research
Methods in the UK Social Sciences. Southampton: ESRC National Centre for Research Methods
(NCRM)
Beissel-Durrant, G. & Lang, I. (2004) Consultation Exercise Report. Southampton: ESRC National
Centre for Research Methods (NCRM), pp. 1-44, Available from
http://www.ncrm.ac.uk/reports/ncrm_consult_report_web.pdf.
Beissel-Durrant, G. (2004) A Typology of Research Methods Within the Social Sciences.
Southampton: ESRC National Centre for Research Methods (NCRM) & Southampton Statistical
Sciences Research Institute (S3RI)
Blaxter, L., Hughes, C. & Tight, M. (2001) How To Research. 2nd ed. Buckingham: Open
University Press.
Creswell, J. W. (2009) Research Design: Qualitative, Quantitative and Mixed Methods
Approaches. 3rd ed. Los Angeles: SAGE
Das, M., Ester, P., Kaczmirek, L. & Mohler, P. (Eds.) (in preparation). Social Research and the
Internet: Advances in Applied Methods and New Research Strategies. New York: Taylor & Francis
Group
Dillman, D. A. (2007) Mail and Internet Surveys: A Tailored Design Method. 2nd ed. New Jersey:
John Wiley & Sons, Inc.
Forbes, I. (2003) Perceptions of cutting edge research in UK social science. Innovation: The
European Journal of Social Sciences. Vol. 16, No. 3, pp. 271-91
Gilbert, N. (Ed.) (2008) Researching Social Life. 3rd ed. London: SAGE
Hesse-Biber, S. & Leavy, P. (2008) (Eds.) Handbook of Emergent Methods. New York: The
Guildford Press
Hussey, J. and Hussey, R. (2003) Business Research: A Practical Guide for Undergraduate and
Postgraduate Students. 2nd ed. London: Macmillan Press
Morse, J .M. (2006) The politics of developing research methods. Qualitative Health Research.
Vol. 16, No. 3, pp. 3-4
Saunders, M., Lewis, P. & Tight, A. (2007) Research Methods for Business Students. 4th ed. Essex:
Prentice Hall
Taylor, C. & Coffey, A. (2008) Innovation in Qualitative Research Methods: Possibilities and
Challenges. Working Paper: 121. Cardiff: Cardiff School of Social Sciences
Wiles, R., Bardsley, N. & Powell, J. (2008) Assessment of the Training Needs in Research Methods
in the UK Professional Social Research Community. Southampton: ESRC National Centre for
Research Methods (NCRM)
Wiles, R., Durrant, G., De Broe, S. & Powell, J. (2005) Assessment of Needs for Training in
Research Methods in the UK Social Science Community. Southampton: ESRC National Centre for
Research Methods (NCRM)
95

ISSRM_Report_Public Innovations in Method of Research

Uploaded by

Copyright:

Available Formats

ISSRM_Report_Public Innovations in Method of Research

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

ISSRM_Report_Public Innovations in Method of Research

Uploaded by

Copyright:

Available Formats

ESRC National Centre for Research Methods

Innovations in Social Science Research Methods1

Maria Xenitidou & Nigel Gilbert

1 DO NOT CITE WITHOUT PRIOR PERMISSION FROM THE AUTHORS

Figure 1. Hierarchical classification of Research Methods Categories2

Managing Historical Longitudinal Data: Intermediate Data Structure

Virtual Agents (Avatars)

Human-Survey Interaction Framework and Internet Survey Software Tools

Application and Applicability

Methods for Achieving Equivalence of Samples in Cross-National Surveys:

The goal is functional rather than numeric or stylistic equivalence.

Item Response Theory – Tests without Items

Ethnography as a Non-linear Dynamic System

Water Cooler Logic: Participatory Ethnography at Work

Genetic Transmission of Political Orientations

Kinship Algebra Expert System (KAES)

Journalism as both Reporting and Research

Appreciative Inquiry as a Research Method

Rasmussen, N. H., Furst, J. W., Swenson-Dravis, D. M., Agerter, D. A., Smith,

Towards a New Approach in Social Simulations: Meta-Language

The Research Poem (in Social Work)

The impact of multi-sited ethnography on researcher involvement

Participatory Learning and Action

Narrating Sensitive Topics: Applied and Basic Research

Intelligence-Led Policing Software

Dialectometry: Measures Structure in Geographical Dialect Networks:

HyperResearch: Computer-assisted programme for analysing qualitative

Personal Network Analysis & EgoNet Software

Verbal Autopsy Software

Facility for online data collection: Advanced Multi-disciplinary Facility

Integration and Implementation Sciences (I2S)

Map 1 depicts the geographical concentrations of developments in Social Science Research

1996 Intuitive Inquiry

2008 Appreciative Inquiry

Lars Kazcmirek Wil Dijkstra Mick Couper

Johannes van der Zouwen Hanneke Houtkoop-Steenstra

Janice Morse Adele Clarke Peter Lynn Siegfried Gabler

Anselm Strauss Wendy Haight

Libby Bishop Peter Sommerfeld

Helga Wild Chris Darrouzet Jennifer Greene

Figure 3. ISSRM Key Networks

You might also like