HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Willett, E. L. Articles by Torrie, J. H. Search for Related Content PubMed Articles by Willett, E. L. Articles by Torrie, J. H.

Factors Influencing Experimental Error in Field Trials in Artificial Insemination

American Foundation for the Study of Genetics, Madison, Wis.

J. H. Torrie

University of Wisconsin, Madison

ABSTRACT

In Experiment I, involving 96,565 services from 1,080 collections from 90 bulls, error variances with four antibacterial-agent treatments were compared. The results were as follows: (a) With an average of 60 or more services per collection (not split) there were significant differences among variances for collections within bulls within treatments (error) for the four antibacterial-agent treatments studied, with the lowest being with the sulfa-streptomycin treatment and the highest with no antibacterial agents, (b) With smaller numbers of services per collection differences in heterogeneity of the error variances could not be detected, (c) Error variance decreased rapidly as numbers of services were increased to 80. Above this number the decrease in error variance was slight in proportion to the increase in numbers, (d) There was a little evidence suggesting that there was more variation among collections within bulls known to be infected with pathogenic vibrio than other bulls.

In Experiments II and III, comprising a total of 34,341 services, a study of the effect of experimental design upon relative efficiency indicated that, if collections average more than 160 services, they can be split into as many units as desired as long as each unit represents an average of 80 or more services. If collections average much less than 160 services, they should not be split and incomplete block or randomized complete block designs should be employed.