Although I had learnt a lot about golf clubs randomly appearing out of nowhere and hitting you around the face, I still felt that there was much about the world that I didn't understand. For one thing, could I learn to predict the presence of these golf clubs that seemed inexplicably drawn towards my apparently magnetic head? A child's survival depends upon being able to predict reliably what will happen in certain situations; consequently they develop a model of the world based on the data they have (previous experience) and they then test this model by collecting new data/experiences. Based on how well the new experiences fit with their original model, a child might revise their model of the world.

According to my parents (conveniently I have no memory of this at all), while at nursery school the model of the world that I was most enthusiastic to try out was 'If I get my penis out, it will be really funny'. To my considerable disappointment, this model turned out to be a poor predictor of positive outcomes. Thankfully for all concerned, I soon revised this model of the world to be 'If I get my penis out at nursery school the teachers and mummy and daddy will be quite annoyed'. This revised model was a better 'fit' of the observed data. Fitting models that accurately reflect the observed data is important to establish whether a theory is true.

You'll be relieved to know that this chapter is not about my penis but is about fitting statistical models. We edge sneakily away from the frying pan of research methods and trip accidentally into the fires of statistics hell. We will start to see how we can use the properties of data to go beyond our observations and to draw inferences about the world at large. This chapter lays the foundation for the whole of the rest of the book.