LISU Tel: +44 (0) 1509 63 5680  Loughborough University

Project Outline - Publication and Dissemination Behaviour of Researchers

   

The following is an outline of the proposed project methodology. As the work progresses, alternative methods and strategies may be found to be more appropriate, so that some of the detail may change during the course of the research.

Desk research

An extensive review of existing academic and ‘grey’ literature will be undertaken, utilising the resources available through the libraries of the partner institutions, to identify and summarise previous research in this area, and to obtain a broad international perspective on the issues outlined above. In particular, evidence will be sought from countries with different models of research assessment as to how the development of these models influenced authors’ behaviour, and the infrastructure developed for scholarly communication.

There has been considerable research in the area of author, and researcher, attitudes towards open access in recent years, and research assessment has emerged as a key issue influencing publishing behaviour. Evidence from existing studies will be synthesised with a view to describing trends, and to provide a basis for subsequent data gathering.

Institutional policies and other external influences will also be investigated from the available literature and web sites.

Bibliometric analysis

In parallel with the literature review, a bibliometric analysis will be carried out as a mapping exercise to ascertain whether there are any trends in the citation behaviour of researchers and, if identified, how these trends have evolved over the last five years. The methodology will be fully documented to provide the basis for future longitudinal studies.

Our approach

We will take two stratified random samples, from Web of Knowledge, each of 400 UK academics with publications in 2003 and 2008. The samples will be stratified by discipline, according to the proportions of academic staff of all grades in UK universities in 2006‑07 (see Table 1). The final stratification will be informed by the distribution of academic staff returned in the 2001/2008 RAE. An additional 10% will be allowed for authors subsequently found to have no relevant publications.

Table 1   Suggested sampling design

 Broad discipline
Suggested sample

 Medical sciences

100

 Physical sciences

70

 Engineering

50

 Social studies

80

 Humanities

40

 Arts

30

 Education

30

 Total

400

We will undertake searches for all publications from these authors in the two years – publication date 2003, and publication date 2008. This master list will be obtained as follows:

  • For those academics in subject areas well or reasonably covered by WoS, i.e., STM and certain social sciences such as psychology, an author search will be run on WoS using the selected author’s name and (where the name is a common name) the name of the author’s institution as an address word. Using WoS, we will identify all the citations in each of the publications for that year.
  • For those academics in other subject areas, we will use a variety of methods to obtain publication details, including searching alternative databases, such as SCOPUS, IBSS, etc; obtaining a publication list from the person’s web site; or possibly contacting them directly. Items indexed by WoS will be treated as above; all other items, e.g., monographs, grey literature and the like, will be treated as follows: a member of the project team will source a physical copy of the items, from another academic institution or British Library, obtain access to copies of said items and record the entries in the bibliographies of said items. These may be scanned in, or photocopied for subsequent scanning, depending on permissions that are possible.

The cited items will be analysed as follows:

  • Number of journal articles
  • Number of conference papers
  • Number of grey literature items
  • Number of books
  • Number of items with URLs

We will also examine the level of co-authorship of each published item, and note whether the co-authors work in the same institution as the author we are studying, elsewhere in the UK, or abroad.

Pilot study

We will do a pilot run to assess the time taken to do such analyses, and will report back to the Expert Panel within two weeks to confirm the level of detail to which we will be carrying out the analysis.

Fixing authors to disciplines

We intend to assign authors to disciplines based upon their current academic department. In addition, we will ensure that there is a reasonable spread of authors in the classification of subjects reflected in the four quadrants of Table 2 below. In this way we will be able to link citation practices both to academic disciplines and also to the findings of the case-studies.

Focus groups

The primary data-gathering phase of the study will comprise a series of focus groups of research-active academics representing a cross section of institutions and disciplines. It is anticipated that 10-12 of these groups will be formed regionally, each with 8-10 participants. Staff at all stages of their research career will be invited to participate, and a small incentive (a £25 Amazon voucher is suggested) will be offered to encourage participation.

It is suggested that the focus groups be held at venues in London, Manchester and the East Midlands, with exact venues to be decided. A fourth venue, possibly in the north of England or in Scotland, may be included.

Each will be structured around a specific set of disciplines, to allow issues particularly relevant to those disciplines to be fully explored. The proposed broad discipline groups to be covered are natural sciences, life sciences, medicine, social sciences, arts & humanities, engineering and computer science; others may be added to this list following consultation with the RIN. The rationale for selecting the discipline-based focus groups will build on Whitley’s (2000) typology of disciplines as operationalised in terms of scholarly communication by Fry and Talja (2007).

Table 2 below demonstrates how disciplines might be coupled to Whitley’s (2000) theory of the intellectual and social organisation of the disciplines. The disciplines in italics indicate suggested case studies based on existing knowledge of the practices of those fields (either first hand or through the literature on bibliometrics) and contacts within them, and including some not studied by Fry and Talja e.g. economics, computer science, psychology and engineering.

Table 2   Whitley’s typology for selecting focus groups

 
Functional Dependence
Low
High
Strategic Dependence Low

Weakly bounded groups pursuing a variety of goals with a variety of procedures. Little coordination of results or problems. Low extent of division of labour across research sites.

Social/cultural geography
History
Literature and cultural studies
Psychology

Specialist groups pursuing differentiated goals with specific, standardized procedures. Considerable coordination of results and specialized topics, but little overall concern with hierarchy of goals.

Corpus-based linguistics
Nursing science
High Strongly bounded research schools pursuing distinct goals with separate procedures. High degree of coordination within schools, but little between them. Strong competition for domination of field.

Economics

As above, but strong hierarchy of specialist goals. Competition over centrality of subfields to discipline.

 

High-energy physics
Environmental biology
Computer Science
Engineering

(Based on Fry and Talja, 2007, p. 122)

Institutions invited to participate will represent a cross section of types, from the research intensive to those which are more teaching-led, but which may aspire to a higher research profile. A range of departments covering a spread of RAE scores will also be invited to take part. The institutions approached will be selected from those included in the HEFCE REF pilot and those not included.

The focus groups are expected to last for approximately two hours, and will be recorded. The recordings will be transcribed in full, and software such as Atlas-ti used in their analysis. The key issues described above will provide the basis for a coding scheme to support top-down coding. This will be combined with open coding to capture aspects of behaviour and motivation which might emerge naturally from participants.

Survey of researchers

A small scale survey of authors will be undertaken in a sample of institutions and departments, in order to triangulate the findings from the desk research and the focus groups. Those institutions and departments taking part in the focus groups will be included, with additional respondents sought from the wider academic community if necessary. The aim will be to attract a minimum of 400 responses; the initial sample will be considerably larger as contingency for a potentially low response rate. We also propose running an incentive prize draw, with a suggested total prize value of £100, to encourage response. Institutional policy will be investigated, and correlated with reported researcher behaviours. Participants may be followed up with telephone interviews to further investigate specific issues if appropriate. The survey data will be analysed using the SPSS statistical package; the depth of analysis possible will depend on the number of responses received.

As well as providing a further perspective on the issues from the qualitative stages of the research, an additional purpose of this survey will be to develop and test a survey instrument to provide sector-wide estimates describing attitudes and behaviours in relation to the key issues, as a basis for future longitudinal studies. We feel that the development of a survey instrument which can be used with a new sample of researchers at each iteration of a longitudinal study will provide the most reliable means of measuring trends in attitudes and behaviour over the longer term. The survey carried out will provide baseline figures from which future trends can be measured.

Synthesis and reporting

The evidence gathered at each stage will be analysed using appropriate software and synthesised into a report for the RIN, setting out the key findings, with recommendations and proposals for their implementation. Particular attention will be paid to recommendations for the development of a longitudinal study to monitor changes in researcher behaviour. The research teams at each institution will draft sections relating to their areas of expertise, which will be amalgamated into a draft final report by the project manager and management team.

An initial report will be submitted by the 5th June, for comment by RIN and the expert panel. Provided comments are received in a timely manner, the final report will be submitted by the 26th June.

 

Top
 
Website Maintained by: lisu@lboro.ac.uk
LU Home - Accessibility - Legal Information - © Loughborough University