For k independent samples from a continuous field, this tests:
H0: F1(x)=F2(x)=...=FK(x); that is, the k samples are from populations with the same distribution function
HA: HA:F1(x)≥F2(x)≥...≥FK(x) or H´A:F1(x)≤F2(x)≤...≤FK(x) with at least one strict inequality.
p<α will reject the null hypothesis in favor of an ordered alternative (either direction of ordering).