What makes a policy brief stick? Lessons from a pioneer experiment

Posted on 29 August 2012 in Making your research accessible by

By Christelle Chapoy*

How many times have we had to write a policy brief on a very complex piece of research that has contained no clear take away messages? And when writing the brief, how often have we wondered what the decision maker or government official that our organisation is targeting will do after reading it – if he does read it.

In an attempt to find what makes a policy brief effective, the International Initiative for Impact Evaluation (3ie) and the Institute of Development Studies (IDS) in collaboration with Norad recently conducted an experiment to test how ‘sticky’ a policy brief can be in influencing the mind of the reader and prompting them to act. The response may not come as a surprise for people like us who are in the business of producing policy briefs, but defies the recent policy brief mania that pushes organisations to spend a considerable amount of time and money in producing policy briefs each year.

A policy brief alone is only effective in creating ‘evidence accurate’ beliefs mainly amongst those with no prior opinions. Evidence from the social psychology literature shows that the tendency is to accept evidence that confirms ones prior opinion at face value while holding ones prior belief or ‘fundamental prior’ even more fiercely when confronted with evidence that contradicts those same beliefs.

In this experiment, we don’t suppose that any researcher or communications expert would propose to influence policy using a standalone policy brief. But what we have learned is that having an opinion piece from a sector expert in the brief increases the information sharing effect. So, messengers matter. Readers would even share the message with their network without having absorbed or fully comprehended the knowledge on the basis of the title of the policy brief, its subject or the author’s name. This has implications for our assumptions about the roles of ‘trust’ and ‘power’ in knowledge brokering activities.

In a recent blog discussion Enrique Mendizabal was describing a policy brief “as something one leaves behind after a meeting “. It is one important research communications tool, but it does not have influence on its own and as stressed by the head of communications at IDS, James Georgalakis, what may be more important in this case is “the process that allows individuals and institutions to develop positioning and policy”.

Another interesting finding is that readers who perceived themselves as having a high level of influence were more likely to pass on the brief and keep the messages in circulation. So targeting those ‘movers and shakers’ could provide good outcomes for research communications work.

This is the first time that a randomised control trial was conducted in the field of research communications. While the study has some limitations (including a sample size of over 800 self-selected, highly educated and interested in the topic so not the most representative people we would seek to influence), it is frontier work in trying to test the effectiveness of a knowledge product like a policy brief. 3ie is interested in funding more research in this area. So do share your ideas!

Download summary of study findings (254.2 KB)

Download full report (905.5 KB)

*3ie Senior Advocacy and Policy Officer