V. Statistical Evaluation and Interpretation of Experimental Data

Publisher Website
Google Scholar
Abstract
In evaluating responses in animal productivity resulting from administration of various levels of a drug, drug levels should be treated as a continuous variable. That is, the response should be analyzed as a continuous function of the level of treatment application. Frequently, effects of continuous treatments have been evaluated by point-to-point comparisons. As an example, four levels of an antibiotic are fed to pigs and weight gains are measured. Response of level 1 is compared to level 2; response of level 2 is compared to level 3; response of level 3 is compared to level 4. Alternatively, some other scheme of level-to-level comparison may be used to evaluate weight gain responses. Point-to-point comparisons of responses to continuous treatments have four main disadvantages: (1) there is no allowance for interpolation between levels; (2) a stepwise response is implied, which is not biologically consistent with the nature of the treatment response being evaluated; (3) statistical significance between different treatment levels is usually required for interpretation of the treatment response; and (4) the degree of statistical significance of the response between adjacent treatment levels decreases as the number of treatment levels increase over the same treatment range. Copyright © 1973. American Society of Animal Science. Copyright 1973 by American Society of Animal Science.

This publication has 0 references indexed in Scilit: