using hierarchical clustering on principal component analysis tocompare civil court branches as measured by justice performanceindicators

In both developing and developed countries, the performance of judicialunits in justice system is measured by empirical indicators which are correlated witheach other. the findings of existing indicator initiatives have traditionally been basedon expert surveys, document reviews, and administrative data, or public surveys. thisstudy combined principal component analysis (pca) and cluster analysis in order toresolve the problem of evaluating multiple correlated indicators. actually, hierarchicalclustering on principal components (hcpc) is used to classify civil branches of a trialcourt in iran to create a comprehensive evaluation. based on pca, the hierarchicalclustering algorithm is applied, which divides justice performance indicators into threeclusters based on the dissimilarity matrix. then, three groups of that court brancheswere identified on the basis of the three variables. it allows justice system policymakersand reformers to measure individual courts’ performance and track backlog reductionand delay reduction programs. further, this can help improve the operational levelof justice indicators on justice systems, efficiency, accountability and credibility. thepractical example illustrates the importance of the current approach to evaluating theperformance of judicial units.