Normal probability plots to discern significant effects

Normal probability plots to discern significant effects 

There is a method by which this can be achieved without prior knowledge of the experimental error variance, viz. through using normal probability plots. This very 

If we have a set of nonnally distributed measured data, [pp y, ., yn], their 

The probability, P(y a), that a measured value randomly drawn from the set is less or equal to a given value a, is proportional to the surface under the curve in the interval [- , fl], see Fig. 6.6. 

6 The probability P(y s a) is proportional to the surface under the frequent distribution. 

---