How doctors conceptualise 'P' values: A mixed methods study

Background and objectives: Researchers and clinicians have been criticised for frequently misinterpreting and misusing P values. This study sought to understand how general practitioners (GPs) in Australia and New Zealand conceptualise P values presented in the manner typically encountered in a medical publication. Methods: This mixed-methods study used quantitative and qualitative questions embedded in an online questionnaire and delivered through an Australian and New Zealand GP-specific Facebook group in 2017. It included questions that elaborated on the participant's conceptualisation of 'P = 0.05' within a scenario and tested their P value interpretation ability and confidence. Results There were 247 participants who completed the questionnaire. Participant conceptualisations of P values were described using six thematic categories. The most common (and erroneous) conceptualisation was that P values numerically indicated a 'real-world probability'. No demographic factor, including research experience, seemed associated with better interpretation ability. A confidence - ability gap was detected. Discussion P value misunderstanding is pervasive and might be influenced by a few central misconceptions. Statistics education for clinicians should explicitly address the most common misconceptions.