An example of the impact of different analytical treatments of ordinal explanatory variables on the explanatory/response relationship

Thomas B Farver, Ruben N. Gonzalez

Research output: Contribution to journalArticle

1 Scopus citations


Research studies in veterinary epidemiology typically are directed toward an evaluation of the relationship between an outcome or response variable and one or more risk factors. Frequently, the relationship of interest is adjusted for other known risk factors or other nuisance information such as age, sex, breed, weight, lactation, level of milk production, etc. By the end of the study, a mathematical expression has emerged that summarizes the relationship observed between all explanatory variables and the response. This paper demonstrates how the analytical treatment of ordinal explanatory variables can alter the conclusion that is reached regarding the nature of their relationship with the response. This demonstration is made using data from a recent prospective cohort study to evaluate the effect of vaccination on the incidence of mastitis in dairy cattle.

Original languageEnglish (US)
Pages (from-to)209-218
Number of pages10
JournalPreventive Veterinary Medicine
Issue number3
StatePublished - 1989


ASJC Scopus subject areas

  • Animal Science and Zoology
  • veterinary(all)

Cite this