1.

A laboratory supervisor felt that a particular potency assay of a pharmaceutical material was capable of better precision and began to explore various ways of reducing assay variability. As part of this process she reviewed the most recent control chart (for control material) and picked out the measurements of the two analysts who had carried out the assay most frequently. Her thinking was that if differences between analysts do exist, then by identifying and eliminating the root causes, the assay variability would be reduced. The data are shown in Table 4.2 (units are % potency).

Do these results suggest that the two analysts differ in their average assay results? Assume that the precision achieved by the two analysts is the same and carry out a t-test for a relative bias (% potency) between the two analysts.

2.

Pendl et al.1 describe a fast, easy and reliable quantitative method (GC) for determining total fat in foods and animal feeds. The data in Table 4.3 (% fat) represent the results of replicate measurements on coconut bars, which were made on ten different days by two laboratories A and B.