The paper explores the relationship between diversity measures and ensemble accuracy in classifier ensembles. It introduces ten diversity measures, including four pairwise measures (Q statistic, correlation, disagreement, and double fault) and six non-pairwise measures (entropy of votes, difficulty index, Kohavi-Wolpert variance, interrater agreement, generalized diversity, and coincident failure diversity). Four experiments are conducted to examine the relationship between these measures and ensemble accuracy. The results show that while there are strong correlations among the diversity measures, the relationship between diversity and ensemble accuracy is less clear. The paper concludes that while diversity is important, it may not be the best measure for improving ensemble accuracy in real-life pattern recognition problems.The paper explores the relationship between diversity measures and ensemble accuracy in classifier ensembles. It introduces ten diversity measures, including four pairwise measures (Q statistic, correlation, disagreement, and double fault) and six non-pairwise measures (entropy of votes, difficulty index, Kohavi-Wolpert variance, interrater agreement, generalized diversity, and coincident failure diversity). Four experiments are conducted to examine the relationship between these measures and ensemble accuracy. The results show that while there are strong correlations among the diversity measures, the relationship between diversity and ensemble accuracy is less clear. The paper concludes that while diversity is important, it may not be the best measure for improving ensemble accuracy in real-life pattern recognition problems.