Assessing dispersion

A simple way of assessing dispersion in models for count data

          

It has been some time since I have updated my blog, but I have plenty of good excuses for justifying my lack of activity (or so I like to tell myself). In my post on simple models for abundance, I introduced the Poisson-log GLM as the basic approach for modelling count data. The Poisson-log GLM has an important limitation when it comes to modelling ecological data: it assumes that the variance and the mean of the response (dependent) variable are the same, so it is rather inflexible for modelling response data with a variability exceeding the variation in the mean. When this condition is not met, we obtain over- or under-dispersion in the model. It is important to note that the terms over- and under-dispersion do not refer to the raw data values, but to the mean and variance expected with respect to a Poisson-log GLM. It is a frequent mistake to think that dispersion refers to the raw values of the response variable.

Simple models for distributions

Simple models for distributions

         

In my last post, I said that ecology is fundamentally the study of the distribution and abundance of living things. On that last post, I introduced a very simple model for abundance, so it was just logical to follow-up with an entry on simple models for distribution. I will define distribution as presence/absence of the target species on a given site. I am going to follow the structure of the previous entry to make any comparisons easy.

Simple models for abundance

Simple models for abundance

Ecology is fundamentally the study of the distribution and abundance of living things, and a core goal is to understand the factors that influence the spatial and temporal variation in distribution and abundance patterns. Over the last few months, I have had a few people asking me how to model abundance, so I decided to make a quick post on the topic.