|
To use mathematical models in order to explain probabilistic phenomena have been essential in scientific research. However, in educational settings is common to work with variables which do not satisfy the demanded assumptions in the Linear Model (LM). The Generalized Linear Model (GLM) answers quite well to the problems originated by measurement variable questions. In this work some aspects of the GLM are commented in relation to the LM and this is done from the framework where both of them make sense: statistical modelling.
Besides, the use of statistical software R is emphasized because this one is not very much known in educational research. However, this software is very appropriate for the GLM and for the working style in statistical modelling. Emelina López-González Y Marcos Ruiz-Soler
Key Words: Generalized Linear Model, Linear Model, Poisson regression, statistical modelling, R.
To buy
|
Nº 248, enero-abril 2011 
