PSYCH 100A Lecture Notes - Lecture 7: Statistical Significance, Type I And Type Ii Errors, Standard Deviation
Document Summary
In most real-world research situations we do not know the population variance on variable x, so we cannot use it to calculate the standard error. Family of theoretical distribution that take on slightly different shapes depending on the sample size. More variable than z-distribution because you are estimating the population. A sample variance is both a descriptive statistics, and an estimator of the population variances. A statistic is considered unbiased if the expected value is equal to the population parameter. Over the long haul, the average of the sample variance will equal to the population parameter. A sample variance is neither an overestimate or underestimate from the population parameter. S = sx/ square root of n. But the price paid is that the sampling distribution for the means is no longer distributed as z, but as t with n-1 df. There is a different t distribution for each df. Smaller the sample size, the thicker the tails.
