Psychology 3800F/G Lecture Notes - Lecture 3: Monte Carlo Method, Normal Distribution, Standard Deviation
Document Summary
Boneau study -- only study we actually need to know about for exam studying type 1 error as a function of violating assumptions studying power as a function of sample size. Imagine an empirical population of 2000 observations (numbers) We can calculate: the mean ( ) = 113. 5 the standard deviation ( ) = 7. 106 describe the shape of the frequency distribution (e. g. , normal, exponential, rectangular) Such a population might exist, or we can simulate one. In spss, you will see in your lab, that we can work with infinite populations with particular means, standard deviations, and shapes made up population based on sample data you collected. Probably not; we would expect some error especially with small samples. What would happen if we collected several samples of 10 observations from that population and: Calculated the mean of the sample means -- will be closer to the population mean than the sample means individually.