Do you want to… test the validity of a model?
Kevin’s geometry teacher has asked him to check if a NetLogo model demonstrating fish reproduction is calibrated correctly to produce an even number of male and female offspring. He beings by running the model for several trials and recording his data…
The Law of Large Numbers
The Law of Large Numbers states that the average obtained from a large number of trials should approach the expected value. This video demonstrates how this concept can be used as an introduction to testing the validity of a model.
Sampling variability is a challenging concept because one must always keep in mind that in science and math we rarely sample individuals, rather our samples often contain many individuals. The main idea behind sampling variability is that with some given sample size, our sample average will vary – meaning, the samples themselves will have a distribution. This video helps to illustrate this concept using several real world scenarios.
The Central Limit Theorem
The Central Limit Theorem is a powerful tool for Statistics – it states that given a large enough number of independent samples, the sampling distribution of the sample averages should approach normal regardless of the normalness of the parent population. This concept is explored further in this video.
Sampling distributions and the Central Limit Theorem allow for the creation of a set of tools to test the significance of data. This video demonstrates a few of those tests.