Wednesday, 14 March 2012

Allan Best and Bev Holmes
evidence and policy journal cover
Many thanks to the Annette Boaz, managing editor of Evidence & Policy and to Policy Press for agreeing to make this article open access until April 16 2012. 

Best and Holmes (2010) paper has been an important publication in informing recent approaches to knowledge exchange at CRFR. The KE reading group met to consider how thinking in different ways about the processes of knowledge exchange might inform our work.

The authors set out three ‘generations’ of thinking:
  1. Linear models
  2. Relationship models
  3. System models
·        One-way process, ie research passed from researcher to user
·        Knowledge = product
·        Generalised across difference contexts and settings
·        Good when:
    • Well defined criteria is met
    • Good support for behaviour change
    • Well-resourced
·          Characterised by relationships between people (networks)
·          Knowledge = from multiple sources
·          Collaboration in both creating research and using research
·          Good when
    • Needs to be adapted to local setting
    • Organisational culture favours evidence-informed planning
·        Knowledge cycle tightly woven within priorities, culture and context
·        Circular model with emphasis on the importance of relationships, linkages and exchange
·        Explicit and tacit knowledge need to be integrated to inform decision-making and policy
·        Feedback loops essential
·        Good when:
    • All stakeholders are active collaborators
    • Partnering organisations willing to invest time and resources
    • KE = business strategy

Members of the reading group easily identified examples of linear and relationship models of knowledge exchange in their own work and the work of others, and it was clear to see when each approach might work in relation to different kinds of problems or projects.  It was apparent that most of the work we were now involved with followed a relationship model and acknowledged the value of working collaboratively with a range of stakeholders in the both the planning and implementation of research.

Good working examples of a systems approach to knowledge exchange were less evident, although the About Families project based at CRFR was established within this type of approach.

We reflected on our experiences of trying to positively influence relationships through knowledge exchange activities in terms of how we set up and conduct  meetings, conferences and online forums. Allowing for engaged dialogue, especially between people from different sectors, and avoiding approaches where academics are seen as experts with all the answers, can help shape more productive exchanges.. There is though, still a difficulty in getting people from different sectors and organisations together to address issues, especially when their core work focuses on different areas, and their incentives and rewards push them in different directions.

Often the most effective people in bringing about change can be people on the periphery of the system rather than leaders, partly because people on the sidelines can afford to be more innovative and have less to lose than people in the centre. This might inform our choice of partners for KE, especially in more entrenched or controversial areas of work

We also questioned how easy it would be to define the parameters of working relationships within a system model, as such close working relationships and continuous feedback loops can lead to a greater dependence on key workers if not properly managed.

But using this framework helps us to think through the ways we communicate research, the relationships through which KE occurs, and the systems and contexts where change might happen. This is helpful for planning for more effective approaches to KE. It challenges us to move beyond some of the basic and tired approaches, and to think more deeply about what we are doing and why.


  1. Thanks for providing this information from Best and Holmes, an article that I missed in 2010. in a recent book chapter Sandra Nutley and Sarah Morton echoed Sandra's 2007 book where institutional and systems examples of research use are only starting to emerge. A systems or institutional capacity to support KE and research use has many embedded characteristics but at the end of the day, isn't a system KE capacity the culmination of all the KE activities including linear and relationship models. Even a liner KE activity is appropriate in some circumstances.

    I guess this begs us to wonder where is the threshold of KE projects that moves one from practicing a portfolio of projects to a system of KE?

    I don't have an answer but if an organisation is practicing, resourcing and rewarding KE over time then it is probably a KE system.

    Next time I am in Edinburgh please invite me to a CRFR Reading Group. Sounds fun!


  2. You are right - we need the linear products, embedded within relationships. Best and colleagues argue that each generation of KTA is embedded within the next. Translation still matters, but on its own cannot achieve much - we need relationships through which our KE products are used, and we need to see these within the context of the systems where they operate.
    Hope you can join a reading group next time you are here!