Uitschieters kwantificeren en verwerpen: de Grubbs-test

Grubbs' test, like Dixon's Q-test, is a statistical test to identify the outliers in data with a normal distribution. Here, the number of observations should equal or exceed seven.

For a given data set, this test involves calculating the Grubbs' statistic G, which is the ratio of the absolute difference between the questionable value and the mean value to the standard deviation of the sample.

The calculated G value is then compared with the tabulated critical value of G for a given confidence level and the number of observations.

If the calculated G value exceeds the critical value of G, the questionable observation is considered an outlier of the data and rejected. 

Alternatively, if the calculated G value is smaller than the critical value of G, the data retains the questionable observation.