AimsThe aim of the study was to explore and compare junior doctors' perceptions of their self‐efficacy in prescribing, their prescribing errors and the possible causes of those errors.MethodsA cross‐sectional questionnaire study was distributed to foundation doctors throughout Scotland, based on Bandura's Social Cognitive Theory and Human Error Theory (HET).ResultsFive hundred and forty‐eight questionnaires were completed (35.0% of the national cohort). F1s estimated a higher daytime error rate [median 6.7 (IQR 2–12.4)] than F2s [4.0 IQR (0–10) (P = 0.002)], calculated based on the total number of medicines prescribed. The majority of self‐reported errors (250, 49.2%) resulted from unintentional actions. Interruptions and pressure from other staff were commonly cited causes of errors. F1s were more likely to report insufficient prescribing skills as a potential cause of error than F2s (P = 0.002). The prescribers did not believe that the outcomes of their errors were serious. F2s reported higher self‐efficacy scores than F1s in most aspects of prescribing (P < 0.001).ConclusionFoundation doctors were aware of their prescribing errors, yet were confident in their prescribing skills and apparently complacent about the potential consequences of prescribing errors. Error causation is multi‐factorial often due to environmental factors, but with lack of knowledge also contributing. Therefore interventions are needed at all levels, including environmental changes, improving knowledge, providing feedback and changing attitudes towards the role of prescribing as a major influence on patient outcome.