A parameter is a numerical descriptive measure of a population. For example, p – the probability of a succes in a binomial experiment, μ and σ - the mean and standard deviation of a normal distribution, are parameters.

A sample statistic is a numerical descriptive mearuse of a sample. It is calculated from the observations in the sample. For example, they are a sample mean x, sample varianse s₂, sample standart deviation s and other.

In statistical investigations a value of parameter is almost always unknown, beaucause it is a characterictic of a population to be explored. To make the decisions about it we use the sample data and compute sample statistics. However we must remember, that the sample measerements are observed values of random variables and statistics vary in a random manner from sample to sample. Therefore we need to take into consideration the sampling (probability) distribution of statistic.

You can select a distribution type (normal, exponential, or uniform) and the population size (from 5 to 35). The simulation graphs the distribution. You can compare it to the graph of normal distribution (select the corresponding checkbox) and make sure the Central Limit Theorem is true (for n > 30).