Premium
Analysis of Dichotomous Response Variables in Teratology
Author(s) -
Marubini E.,
Leite M. L. Correa,
Milani S.
Publication year - 1988
Publication title -
biometrical journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.108
H-Index - 63
eISSN - 1521-4036
pISSN - 0323-3847
DOI - 10.1002/bimj.4710300813
Subject(s) - statistics , allowance (engineering) , analysis of variance , mathematics , variance (accounting) , teratology , binomial distribution , econometrics , negative binomial distribution , litter , biology , fetus , engineering , business , mechanical engineering , pregnancy , agronomy , accounting , poisson distribution , genetics
In the recent biometrical literature the methods suggested to analyze dichotomous data arising from experiments testing the teratogenic effect of chemicals, rely upon the Beta‐Binomial distribution of the relative frequencies of response. This paper reports the results of a simulation study aiming at investigating the performance of logistic‐linear models based on the previously mentioned distribution in comparison to that of a two‐steps procedure based on a weighted Analysis of Variance (ANOVA). The weights are computed to make allowance of both the intralitter correlation and the number of fetuses varying from litter to litter. The results attained in this study and considerations emerging from perusing the toxicological literature, show that it appears sensibile to suggest the use of weighted ANOVA to process dichotomous responses in teratology.