Scores for individual relevant questions and individual charts were analyzed for patterns that might suggest ways to reduce errors and inconclusive results. Three experienced examiners scored 100 examinations that had verified results. Validity and reliability for the three examiners was determined to be high and in agreement with previous studies. Examiner results improved when at least two examiners agreed on a result in blind scoring. Instances of relevant questions and charts with zero scores or scores contrary to the total examination score were examined to determine if they influenced errors or inconclusive results. It was concluded that they did. Suggestions are offered for reducing errors and inconclusive results.