3.6 Chapter summary

Inferential statistics refers to the techniques we use for inferring things about a population from a sample drawn from that population. We proceed by setting up a model of the situation of interest in the population. This model contains a number of unknown parameters. We use the data from the sample to obtain estimates of these parameters, and standard errors for those estimates, which reflect their precision. A General Linear model is a statistical model in which a continuous outcome variable is predicted by a weighted sum of one or more predictor variables. It is implemented with R function lm(). The predictor variables can be continuous or categorical. Care must be taken in thinking about which ones to include, and what the parameter estimates mean. The General Linear Model assumes that the errors (the departures of individual cases from the model’s predictions) are drawn from a Normal distribution, and that their variance is a constant.