Study Notes on Sampling Distributions and Central Limit Theorem
Sampling Distribution
Definition: Probability distribution for a statistic calculated from repeated samples of size n.
Sample Mean as a Random Variable
Sample mean () functions as a random variable.
For a random sample from population with mean (µ) and standard deviation (σ):
Mean of sampling distribution: = µ
Standard deviation (SEM): = σ/√n
Independence Assumption
Assume independence between observations for sampling characteristics to hold.
Random samples; if sampling without replacement: n < 0.05(N).
Question: Wouldn’t selecting only < 5% cause bias or scew the data? If you look at a chart of data, and you only focus in on a certain section with results that do not accurately represent your population, wouldn’t that lead you to a misleading conclusion about your data set?
Question: Can you show me how you pulled up the “1000 times” things on SC?
Shape of Sampling Distribution
Shape, center, and spread must be discussed:
Center =
Spread = σ/√n
Shape influenced by parent distribution.
Normal Distribution Sampling
If X is normally distributed, then is also normally distributed with:
Mean = µ
Standard deviation = σ/√n
Application of Sampling Distribution
Use probabilities for means:
P( > k) = P(Z > (k - µ)/(σ/√n)).
Central Limit Theorem (CLT)
CLT states sampling distribution of approaches normality as sample size increases (n > 30).
Non-Normal Distributions
Even with non-normal population, mean and standard deviation of sampling distribution () held true regardless of n.
Practical Examples
SAT example with 100 students (mean=1500, SD=250).
Caloric intake examples illustrating P( > k) with given means and standard deviations.