Explaining modelling in simple terms

Question

One of the most challenging ideas to describe to non-technical/mathematical people is "modelling". I've used quite a few explanations so far but all of them were around specific modelling techniques. For instance, I usually draw some data points and lines in between or I sometimes draw some decision trees etc but I feel that there must be a better/simpler way to help them understand the concept.

You might want to check the thread: [What is the meaning of “All models are wrong, but some are useful”](https://stats.stackexchange.com/questions/57407). — usεr11852, Jan 22 '19 at 21:35

score 3 · Accepted Answer · answered Jan 05 '19 at 12:27

In the simplest terms: A model is a description of the data. Good models try to describe the key parts of the data and to separate signal from noise.

The thing is that any more specific description is going to leave out some types of model; indeed, even though you ask this on CrossValidated (and, so, are probably interested in statistical models) there are lots of models that aren't statistical at all - and some that don't even use any math.

I have recently bought The Model Thinker by Scott Page. I haven't started reading it but the table of contents and index look very interesting and the reviews were very positive. It might be a good long answer to this question.

Thanks a lot for the answer and the recommendation! – Vasilis Vasileiou Jan 05 '19 at 13:08 — Vasilis Vasileiou, Jan 05 '19 at 13:08

Explaining modelling in simple terms

1 Answers1