Background and Aims: Self-recorded, single-lead ECGs are increasingly used to diagnose arrhythmias. However, they can be of variable quality, which can affect the reliability of interpretation. In this analysis of ECGs collected in atrial fibrillation screening studies, our aims were to: (i) determine the quality of ECGs when recorded unsupervised (at home); and (ii) investigate whether telephone training improved ECG quality. Methods: Data was obtained from the Screening for Atrial Fibrillation to Reduce stroke (SAFER) programme, where participants recorded single-lead ECGs four times per day for three weeks using a handheld device. ECG quality was assessed using an automated algorithm, and participants who recorded >25% poor quality ECGs from days 4-10 of screening were identified for training. Telephone training to improve ECG recording technique was delivered when research team capacity permitted. Results: 14,727 participants recorded 1,206,972 ECGs, of which 43,513 (3.6%) were poor quality. Most participants (51.9%) did not record any poor-quality ECGs. 1,105 (7.5%) participants met the threshold for training. Of these, 165 participants received training and 896 did not. Comparing these groups, the mean (95% confidence interval) reduction in the proportion of poor-quality ECGs per participant from before training (days 1-3) to after training (days 11-21) was 21.1 (17.5-23.5) % with training and 15.7 (14.5-16.8) % without training (p<0.05). Conclusion: Most participants achieved adequate quality ECGs. For those that did not, ECG quality improved over time regardless, and training further improved ECG quality. Therefore, telephone training could be considered in atrial fibrillation screening programmes using single-lead ECG devices.