Objective To assess whether utilization of a validated communication tool corresponds with faculty assessment and resident self-assessment on the pediatric communication milestone continuum. Methods Pediatric residents were recruited to participate in the communication skills assessment. Continuity clinic faculty completed an assessment of each residents communication skills utilizing the 6 pediatric milestones that address interpersonal and communication skills. Each participating resident completed a self-assessment of their own communication skills utilizing the same milestones. After being placed on the milestones, the residents participated in a standardized patient interview that was recorded and subsequently evaluated by a faculty observer utilizing the Common Ground Instrument. Results 16/16 of pediatric residents participated in the study. The milestones and common ground instrument were scored on a scale from 1 to 5 with 5 representing an expert rating. For PGY-1 residents, the average faculty score on the milestones was 3.17, self-assessed average score was 2.92 and common ground average score was 3.67. For PGY-2 residents, the average faculty score on the milestones was 4.40, self-assessed score average was 4.10 and common ground average score was 3.20. For PGY-3 residents, the average faculty score on the milestones was 4.70, self-assessed score average was 4.10 and common ground average score was 3.60. PGY-1s had significantly lower self and faculty assessments than PGY-2s or -3s. There were no significant differences among PGYs on the Common Ground Interview score. Faculty rated residents significantly higher than they rated themselves. Previous clinical skills training, standardized patient training, and English as a first language had no significant effect on the self-assessment, faculty assessment or Common Ground Instrument score. Conclusion Faculty and residents observe an improvement in communication skills as residents progress through training; however, scores on a validated communication tool do not reflect this improvement.