Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
A-1 APPENDIX A: Description of Hoaglin et al. Outlier Method The method of identifying invalid data and outliers is a slightly modified version of a method for determining extreme data values described by Hoaglin et al.1 This method uses the range of the two inner quartiles of a data set to determine the cut off values for outlying data and extreme outlying data: )( LUUU FFkFIF â±= (Equation 12) )( LULL FFkFIF â±= (Equation 13) Where: IFU = Upper cutoff point for extreme value determination IFL = Lower cutoff point for extreme value determination FU = Upper quartile FL = Lower quartile k = constant k where k = 1.5 for outlying data and k = 3 for extreme outlying data The analysis technique in this study uses cut off limits at the same locations by using the range of the inner 75% of the data rather than use the inner quartiles (i.e. inner 50%). This way, the cut offs are based on a larger number of laboratories and the technique is more robust. Since the inner range of data is increased from 50% to 75%, the k values are decreased accordingly from k = 1.5 and k = 3 to k = 0.674 and k = 1.555, respectively. 1 Hoaglin, D. C., Iglewicz, B., Tukey, J. W., âPerformance of Some Resistant Rules for Outlier Labeling,â Journal of the American Statistical Association, Vol. 81, No. 396 (Dec., 1986), pp. 991-999.