The Comparison of the Sensitivities of Similar Experiments: Applications

Abstract

effects of experimental treatments is a difficult one in certain areas of research. We are concerned with means of comparing the sensitivities of two available experimental techniques, where we use the term in a very general sense, that might be used to yield data for specified treatment comparisons. It is often not possible to determine on a priori grounds which of two techniques should be used and it may be desirable to use both techniques in separate experiments with a view to deciding on the technique to be used in future experimentation. When the two experiments are indeed independent of each other in probability, the methods of this paper may be used. Let us consider several typical situations indicated by subj ectmatter headings. Taste Testing: Hopkins [1950, 1953] investigated the use of discrete scoring scales in taste testing experiments as they are commonly used and Baten [1946] suggested the use of a continuous line scoring system. The latter obtained results which he interpreted as showing an increase in sensitivity for his method over the use of discrete scores. Procedures discussed in this paper provide means of comparing the sensitivities of the two scoring systems if data were available from two identically designed but statistically independent experiments, one for each scoring system, in a form suitable for analysis of variance. Many other comparable situations arise in taste testing. The sensitivities of two taste panels, two methods of preparation of test items, two environmental situations during tasting, or two methods of storage may be compared.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 50 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.