BACKGROUND Little information is available on the intrarater reliability of parent ratings of medical student interview skills. The aim of this study was to compare maternal ratings of a videotaped medical student interview, with ratings by the same mother of the same interview seen a month later. METHODS Thirty mothers rated 1 of 2 'medical student' interview videotapes (random allocation). An actor played the role of the student and demonstrated different student skill levels in each interview. Mothers rated the same interview again a month later (mean +/- SD 34 +/- 13 days). Maternal satisfaction was measured using the Medical Interview Satisfaction Scale (MISS, maximum score 203) and the Interpersonal Skills Rating Scale (IPS, maximum score 91) with higher scores indicating higher satisfaction. RESULTS No significant differences in mean maternal ratings were observed following the first and second viewings of each interview. The average percentage difference in first and second viewing scores was MISS 4%, IPS 8% with correlation coefficients MISS 0.96, IPS 0.87 (P < 0.05). CONCLUSIONS The high levels of stability in maternal evaluations of simulated medical student interviews supports the use of maternal evaluations for longitudinal monitoring of medical student performance.