Error from taking only a sample of the population. It is only avoidable by taking a census rather than a sample.
A population that is normally distributed has two parameters: mean and standard deviation.
The mean (average) value of a normally distributed population.
Population standard deviation
A measure of how much deviation there is from the population mean. Used with normally distributed populations.
A sample has two parameters: mean and standard deviation.
The mean (average) value of the sample.
Sample standard deviation
A measure of how much deviation there is from the sample mean.
Central Limit Theorem
The theorem claims that if a sample is large enough (around 30) and from a normally distributed population, then the distribution of sample means will be normally distributed with a mean the same as the population mean.
Standard error of measurement (S.E.)
The standard deviation of the (normally distributed) distribution of sample means. It is half as big as the margin of error.
Margin of error (MOE)
Plus or minus two standard errors in a normal distribution. Combined with the mean of the sample means, it gives us an interval in which we are 95% confident the population mean lies.
95% confidence interval
An interval in which we are 95% confident the population mean lies. Requires the mean of the sample means and the margin of error.
Standard error of proportion (S.E.)
The standard error in a binomial distribution. It is half as big as the margin of error for proportion
Margin of error for proportion (MOE)
Plus or minus two standard errors of proportion (S.E.). Combined with the survey result, it gives us the 95% confidence interval for sample proportion.
When two probabilities depend on each other. The MOE of the difference is twice the MOE of each event.
When two probabilities are independent of each other. The MOE of the difference is 1.5 times the average of the two MOEs of each event.