Evidence-based policy is a term often applied in multiple fields of public policy to refer to situations whereby policy decisions are informed by rigorously established objective evidence. Underlying many of the calls for 'evidence based policy' is often a (stated or unstated) concern with fidelity to scientific good practice, reflecting the belief that social goals are best served when scientific evidence is used rigorously and comprehensively to inform decisions, rather than in a piecemeal, manipulated, or cherry-picked manner.
Some have promoted particular types of evidence as 'best' for policy makers to consider, including scientifically rigorous evaluation studies such as randomized controlled trials to identify programs and practices capable of improving policy-relevant outcomes. However, some areas of policy-relevant knowledge are not well served by quantitative research, leading to debate about the methods and instruments that are considered critical for the collection of relevant evidence. For instance, policies that are concerned with human rights, public acceptability, or social justice may require other evidence than what randomized trials provide, or may require moral philosophical reasoning in addition to considerations of evidence of intervention effect (which randomised trials are principally designed to provide ). Good data, analytical skills and political support to the use of scientific information, as such, are typically seen as the important elements of an evidence-based approach.
Although evidence-based policy can be traced as far back as the fourteenth century, it was more recently popularized by the Blair Government in the United Kingdom. The Blair Government said they wanted to end the ideological led-based decision making for policy making. For example, a UK Government white paper published in 1999 ("Modernising Government") noted that Government must "produce policies that really deal with problems, that are forward-looking and shaped by evidence rather than a response to short-term pressures; that tackle causes not symptoms".
Evidence-based policy is associated with Adrian Smith because in his 1996 presidential address to the Royal Statistical Society, Smith questioned the current process of policy making and urged for a more "evidence-based approach" commenting that it has "valuable lessons to offer".
Some policy scholars now avoid using the term 'evidence based' policy, using others such as 'evidence informed'. This language shift allows continued thinking about the underlying desire to improve evidence use in terms of its rigor or quality, while avoiding some of the key limitations or reductionist ideas at times seen with the 'evidence based' language. Still, the language of 'evidence based' policy is widely used and, as such, can be interpreted to reflect a desire for evidence to be used well or appropriately in one way or another - such as by ensuring systematic consideration of rigorous and high quality policy relevant evidence, or by avoiding biased and erroneous applications of evidence for political ends.
The earliest form of evidence-based policy was tariff-making in Australia which was required under legislation to be educated by the public report issued by the Tariff Board. These reports were initially only reporting on the impacts but changed to also report on the effects of industries and the economy.
Many scholars see the term "evidence-based policy" as evolving from "evidence-based medicine", in which research findings are used as the support for clinical decisions and evidence is gathered by randomized controlled trials (RCTs), which is comparing a treatment group with a placebo group to measure results. In 1993, the Cochrane Collaboration was established in the UK, and works to keep all RCTs up-to-date and provides "Cochrane reviews" which provides primary research in human health and health policy. There was then an increase in research and policy activists pushing for more evidence-based policy-making which led to the formation of the sister organization to Cochrane Collaboration, the Campbell Collaboration in 1999. The Campbell Collaboration conducts reviews on the best evidence that analyzes the effects of social and educational policies and practices.
The Economic and Social Research Council (ESRC) became involved in the push for more evidence-based policy making with its 1.3 million pound grant to the Evidence Network in 1999. The Evidence Network is a center for evidence-based policy and practice and is similar to both the Campbell and Cochrane Collaboration.
More recently the Alliance for Useful Evidence has been established, with funding from ESRC, Big Lottery and Nesta, to champion the use of evidence in social policy and practice. The Alliance is a UK-wide network that promotes the use of high quality evidence to inform decisions on strategy, policy and practice through advocacy, publishing research, sharing ideas and advice, and holding events and training.
Recently questions have been raised about the conflicts-of-interest inherent to evidence-based decision-making used in public policy development. In a study of vocational education in prisons operated by the California Department of Corrections, Andrew J. Dick, William Rich, and Tony Waters found that political considerations inevitably intruded into "evidence-based decisions" which were ostensibly technocratic in nature. They point out that this is particularly where evidence is paid for by policy makers who have a vested interest in having past political judgments confirmed, evidence based research is likely to be corrupted.
There are many methodologies for evidence-based policy but they all share the following characteristics:
The form of methodology used with evidence-based policy fit under the category of a cost-benefit framework and are created to estimate a net payoff if the policy was to be implemented. Because there is a difficulty in quantifying some effects and outcomes of the policy, it is mostly focused broadly on whether or not benefits will outweigh costs, instead of using specific values.
Several critiques have emerged. Paul Cairney, professor of politics and public policy at the University of Stirling in Scotland, argues that supporters of the idea underestimate the complexity of policy-making and misconstrue the way that policy decisions are usually made. Moreover, evidence has emerged of front-line public servants, like hospital managers, making decisions that actually worsen patients' care in order to hit pre-ordained targets. This argument was put forward by Professor Jerry Muller of the Catholic University of America in a book called The Tyranny of Metrics. A comprehensive list of critiques, including the fact that policies shown to be successful in one place often fail in others, despite reaching a gold standard of evidence, has been compiled by the policy platform Apolitical.
The Overseas Development Institute has pioneered RAPID Outcome Mapping Approach (ROMA) over the past five years as a means to help aid donors and partners better transform research into policy initiatives. This approach is supported by other research within the development community. For instance, in a literature review focused on development, an integrated, participatory, structured and enpowering approach to using evidence and data in decision-making to inform development decisions was tied to improved results. 
Six key lessons have been developed, which are:
These lessons show that the relationship between research, policy and practice is complex, multi-factoral, non-linear, and highly context specific. What works in one situation may not work in another. Developing effective strategies in complex environments is not straightforward. Simple tools such as cost-benefit analysis, logical frameworks, traditional project management tools and others may not work on their own, as they fail to take into account the existing complexity.
Based on research conducted in six Asian and African countries, the Future Health Systems consortium has identified a set of key strategies for improving uptake of evidence in to policy, including: improving the technical capacity of policy-makers; better packaging of research findings; use of social networks; establishment of fora to assist in linking evidence with policy outcomes.
ROMA approach takes these lessons into account has been field tested through more than 40 workshops and training courses worldwide. It is an eight-step approach for each of which the ODI has developed resources and policy tools to ensure each step is comprehensively addressed:
An example of ROMA approach can be seen in the case of the Wildlife Enforcement Monitoring System (WEMS) Initiative  where a systematic approach of agreement has brought its implementation in Africa.
This has resulted in:
The Coalition is a nonprofit, nonpartisan organization, whose mission is to increase government effectiveness through the use of rigorous evidence about "what works." Since 2001, the Coalition's work with U.S. Congressional and Executive Branch officials has advanced evidence-based reforms in U.S. social programs, which have been enacted into law and policy. The Coalition has no affiliation with any programs or program models, and no financial interest in the policy ideas it supports, enabling it to serve as an independent, objective source of expertise to government officials on evidence-based policy.[unreliable source]