SOME FACTORS LIMITING THE USE OF GENERALIZED LINEAR MODELS IN AGRICULTURAL RESEARCH

The generalized linear model (GLM) is a hot topic in statistics. Numerous research articles on GLM\u27s appear in each edition of all major journals in statistics. GLM\u27s are the subject of substantial numbers of presentations at most statistics conferences. Despite the high level of interest and research activity within the statistics community, GLM\u27s are not widely used, with some exceptions, by biological scientists in the statistical analysis of their research data. Why? Reasons include 1) many statisticians are not comfortable with GLM\u27s, 2) the biological research community is not familiar with GLM\u27s, and 3) there is little in introductory statistics courses as currently taught to change (1) or (2). Whether or not this is a real problem is unclear. This paper looks at some of the factors underlying the current state of GLM\u27s in statistical practice in biology