Abstract
For other domains that have adopted the evidence-based paradigm, the impact has included research outcomes having greater influence in terms of informing and influencing practitioners and policy-makers. We examine how evidence-based practices are being adapted for use in software engineering and discuss how decision-making in our own discipline can be liberated from over-reliance on expert judgement. To support our arguments we discuss some outcomes from recent studies and present an example in which performing a systematic literature review demonstrates the unreliability of depending only upon the outcomes of individual studies. Finally, we identify six challenges that need to be addressed in order to provide software engineers with standards and practices that are underpinned by evidence.