ABSTRACT
Before considering problems that can complicate statistical procedures, a few well-known simple techniques and terms will be reviewed in this section. Only the null case, where Δ actually equals 0, was discussed in the previous chapter. When Δ ^ 0 neither
are standard Normal variates. To see what these variates are, the relation ship
is informative because, in the alternative, as opposed to the null case, what can be expected of the difference between two ^-statistics calculated from size 4 and size 3 samples, is a positive multiple of the composite parameter
Based on the assumption that the expected values of the two z-statistics respectively equal some corresponding models’ location parameters, it fol lows that Δ is the standardized difference between the null and the true location parameters.