Laserfiche WebLink
Outlier Test <br />Dean and Dixon <br />The data set containing N values is sorted either in an ascending or descending order, with x, being the suspect <br />value. Then the test statistic Q is calculated using the equation <br />i~,-~~~ <br />The decision whether x, is an outlier is performed by comparing the value Q to the critical values listed in the <br />following table: <br />Dixon's critical value table: <br />N 10% 5% 2% 1% 0.5% <br />3 0.886 0.941 0.976 0.988 0.994 <br />4 0.679 0.765 0.846 0.889 0.926 <br />5 0.557 0.642 0.729 0.780 0.821 <br />6 0.482 0.560 0.644 0.698 0.740 <br />7 0.434 0.507 0.586 0.637 0.680 <br />8 0.479 0.554 0.631 0.683 0.725 <br />9 0.441 0.512 0.587 0.635 0.677 <br />10 0.409 0.477 0.551 0.597 0.639 <br />11 0.517 0.576 0.638 0.679 0.713 <br />12 0.490 0.546 0.605 0.642 0.675 <br />13 0.467 0.521 0.578 0.615 0.649 <br />14 0.492 0.546 0.602 0.641 0.674 <br />15 0.472 0.525 0.579 0.616 0.647 <br />16 0.454 0.507 0.559 0.595 0.624 <br />17 0.438 0.490 0.542 0.577 0.605 <br />18 0.424 0.475 0.527 0.561 0.589 <br />19 0.412 0.462 0.514 0.547 0.575 <br />20 0.401 0.450 0.502 0.535 0.562 <br />21 0.391 0.440 0.491 0.524 0.551 <br />22 0.382 0.430 0.481 0.514 0.541 <br />23 0.374 0.421 0.472 0.505 0.532 <br />24 0.367 0.413 0.464 0.497 0.524 <br />25 0.360 0.406 0.457 0.489 0.516 <br />