Laserfiche WebLink
E <br />An example computer output is as follows: <br />Trend Analysis MW -3A <br />Kendall Correlation Coefficient x1: event; y1: 1, 1 -dichloroethane <br />N <br />Score <br />Tau <br />Z <br />Tau corrected for ties <br />Z corrected forties <br /># x tied groups: 0 <br />The parameters given in the output are <br />N = number of samples <br />Score = Mann -Kendall statistic (S) <br />10 <br />—19 <br />—.422 <br />—1.699 p = 0.0892 <br />—.509 <br />—2.048 p = 0.0406 <br /># y tied groups: 3 <br />x = independent variable, which in this case is the sampling <br />event, or time <br />y = dependent variables, which are VOC or inorganic con- <br />centrations <br />The Kendall -tau method assigns a positive or negative <br />score based on the differences between the data points. <br />The first step is to list the data in the order in which they <br />were collected over time, and then determine the sign of <br />all possible differences. <br />The Mann -Kendall statistic (S) is then computed, which is <br />the number of positive differences minus the number of <br />negative differences. If S is a large positive number, mea- <br />surements taken later in time tend to be larger than those <br />taken earlier, i.e., an upward trend. Similarly, if S is a <br />large negative number, measurements taken later in time <br />tend to be smaller, i.e., a downward trend. <br />pjb\b02\b020104i A - 2 Rev. 0 September 25, 1991 <br />