In many countries the introduction of competitive rankings and assessment frameworks have necessitated that universities continually monitor and strategically promote their strengths. This management objective also requires that universities be able to promote and encourage research behaviour that increases the probability of research success using research administrators and/or managers as facilitators. Research administrators are now regarded as key participants in research planning at the department, college, and university levels to attract and manage strategically desirable research and researcher behaviour. In order for organisations to structure their research management strategies more efficiently, as well as to inform practitioners as to the best way to deliver their services, an understanding of techniques and state of the research administration role is needed.
The research management/administration profession has sought to define itself in recent years. In the UK, in 2009 the Higher Education Funding Council for England and Medical Research Council funded a study entitled 'Professionalising Research Management'. The study's main objective was to identify whether there was a demand for the development of a professional framework for the training of Research Managers and, if so, how this demand could be addressed. However, as part of this study, the authors articulated a range of work activities and skill requirements associated with research work. It also identified the variety of research management structures within universities, the levels at which research managers operate, and their involvement at strategic levels within the university. Building on such understandings of 'research management', the UK's National Association of Research Administrators (ARMA) has recently implemented a 'Professional Development Framework' which outlines the 'activities, knowledge, skills and behaviours required across the full range of research management and administration roles' (https://www.arma.ac.uk/professional-development/PDF). This framework describes the key activities at the operational, management and leadership levels. As a result of this framework, the Association has developed professional certificates in research administration, management and leadership. It could be suggested, therefore, that there is now a detailed understanding of the constituent parts that broadly make up 'research management'. However, as noted by Green and Fangley (2009), the huge variety in how it is delivered across the sector, and the constant restructuring of research services within universities, suggests a lack of understanding regarding how it can most effectively be delivered. Indeed, recognition that 'research management' lacks the consistency and standardization of professions such as Finance and Human Resources means that it is more difficult for those outside of the profession to understand and value its function, and more complicated to define and situate in terms of its role within a university.
Hockey & Allen-Collinson (2009) state that formal research on administrative/management staff in higher education is lacking (Mclnnis, 1998; Whitchurch, 2006b; Allen-Collinson, 2006). Research management provides a balance between promoting the needs of institutions to meet their organisational objectives and the ability of academics to determine the best means of performing research. Despite the importance of research management as part of the modern university, there is little consensus within the literature available regarding what are the successful strategies for this profession. In particular, which management models and strategies specifically for the research management profession are the most effective? In addition to that, those outside of the profession are often unsure with regards to what constitutes 'research management', what value it adds, and how best it can be operationalised (Green & Fangley, 2009). What is required, therefore, is an evidence-informed understanding of best practice for research management.
The aim of this review is to draw from the literature an understanding of how the role of research management is considered, as well as to investigate the evidence base for successful strategies of research management. By addressing this, this review provides one of the first investigations of both the academic and professional literature of the role of research management. The objective is to review the state of research management/administration research, and to provide a description of the effectiveness of strategies and structures investigated in the literature.
The systematic literature review originated as an approach within medical science and healthcare as a way to ensure rigorous secondary research that could be used to inform practice. It is distinct from more narrative approaches to literature review as it adopts 'a replicable, scientific and transparent process, in other words a detailed technology, that aims to minimize bias ... and by providing an audit trail of the reviewers' decisions, procedures and conclusions' (Tranfield, Denyer & Smart 2003, p.209). This approach is now being used more widely by researchers as a means of assimilating "'best practice" to provide insights and guidance for intervention into the operational needs of practitioners and policy makers' (Tranfield, Denyer & Smart 2003, p.208). As this study seeks to identify evidence to inform research management practice, the methodology is appropriate.
This research involved the key components of a systematic review (Spencer et ah, 2003; Buchanan & Bryman, 2009), including:
Formulating a research question;
Locating studies with the aim of locating, selecting and appraising as many studies as possible that were relevant to the review;
Setting exclusion and inclusion criteria to inform study selection;
Critically evaluating and appraising the literature selected;
Drawing inferences from the literature's recommendations;
Making recommendations for future research.
The review model adopted prioritised a divergent/convergent approach which allowed the authors to remain open to the variety of research management literature sources available (initial diverging), but to also employ an empirically structured approach designed to identify the structures and strategies deemed successful by the academic evidence base (subsequent converging).
The diverse nature of the research management field, as well as a hypothesised separation between the academically- and professionally-based literatures, necessitated the adoption of such a semi-structured approach to the consideration of the literature.
Between the authors, a definition was developed in order to guide the identification of suitable papers and to aid the development of relevant inclusion and exclusion criteria (described below). In addition, the development of a definition for this review was an attempt to maintain some of the characteristics of a systematic review, while adhering to the convergent-divergent model necessary for analysing the social science literature. The following question guided the selection of relevant literature for this review;
"What are the successes of different models and structures of research management within research organisations?"
A list of journals was constructed by the authors that were considered as potential targets for research involving research management. These journals were drawn from the management, innovation, professional, and sociology literature in order to capture as many relevant articles as possible. Following the identification of potentially relevant journals, a series of key words were identified in order to develop suitable search strings. Articles included in the final sample id were:
Based on cases, policies or data generated in the US, UK, and Europe;
Published in English; and,
Published within 2003-2013.
Three unique search strings were employed independently to a representable sample of articles for the review. Each search string was run separately, with the results of each search string then combined and any repeated articles deleted. The number of articles resulting from each search string, with the total number of articles (minus repeats) identified, is shown below in Table 2.
The search string described above returned a total of 4211 articles. Articles were then manually checked by GD and AN to eliminate any irrelevant articles including those (i) not relevant to research management or administration, (ii) not focused on academic research either in universities or research organisations, and (iii) did not include a consideration of the structures and strategies of research management. This process successfully eliminated 3842 irrelevant articles.
At the same time, manual checks of the journals initially identified as potential targets, but not indexed by Web of Science, were conducted by GD. This process added a further 55 relevant articles to the sample. A final, more in-depth consideration of the articles was conducted where each article's relevance to the central research question, according to a detailed analysis of its abstract and full papers, was determined. At the conclusion of this process, articles were automatically discarded if a conflict in classifications between AN and GD still existed. This resulted in 98 articles being included in the final review. A diagram of the above process and the number of articles included at each stage of consideration are included in Figure 1.
Analysing Article Characteristics
An analysis of the journals of the articles was performed. The purpose of this analysis stage was to guide the overall review of the literature, especially for the development of themes described in the critical analysis of the literature. A secondary purpose analysing the publication characteristics of the sample articles was...