
Key Concepts Study Tool: Chapter 09
Click on each concept below to check your understanding.
1. Sampling Distribution of Means
- Found by repeatedly re-sampling the population and calculating the average values from each sample.
- Distribution of means will be tightly clustered around the true population mean.
2. Central Limit Theorem
- Any sample statistic (such as the mean) generated from a known population will lie along a normal distribution.
- Any one random sample mean should be close enough to the population mean.
3. Standard Error of the Mean (σx̅)
- An estimate of how closely the mean of the sample approximates the population.
- Think of the standard error as a type of standard deviation, except that instead of measuring how far an observation is from a mean, it refers to the distance that a sample mean is from a population mean.
4. Confidence Intervals
- Confidence intervals give a range for the mean and the probability that the true score is within that range.
- Confidence limits give the upper and lower ranges of the “confidence interval.”
- If population standard deviation is known: Confidence Interval = (x̅) ± (zcritical * Xx̅)
- If population standard deviation is unknown: Confidence Interval = (x̅) ± (zcritical * sx̅)
- Where, zcritical,/em> is found in the z-table (Appendix A), σx̅ is the population standard error, and sx̅ is the sample standard error. ±
5. The t-Distribution
- An infinite number of curves, one for every sample size greater than or equal to two. As sample size increases, the t-distribution increasingly resembles the standard normal distribution.
- Values for the t-distribution can be derived by using one of the following equations (when the mean = 0 and the variance is > 1).
- Where:
- x̅ = sample mean
- μ = population mean
- sx = sample standard deviation
- sx̅ = sample standard deviation
- n = total sample size
6. Degrees of Freedom
- All but one of the values are uncertain, or free, because the final value is always known by subtracting the cumulative total of contributions from the total.
- When working with the mean: df = n – 1
- At df = 120 the values of t- and z-distribution are identical; t should only be used when the n < 120.
7. Estimating a Population Mean with a Known Confidence Interval
- Calculate the sample mean.
- Assuming that the population standard deviation (σx) is known, calculate the standard error of the sample mean using the following equation:
- Or, you’ll need to rely on the sample standard deviation and use this equation:
- Find the relevant value of zcritical in the z-table in Appendix A that corresponds with a 95 percent confidence interval (1.96 for 95 per cent, 2.58 for 99 per cent in a two-tailed test).
- Insert the relevant values into one of the following equations:
- Confidence Interval = X̄ ± (Zcritical * σx̅) OR
- Confidence Interval = X̄ ± (Zcritical * sx̅)
8. Estimating Population Proportions Using Only Sample Characteristics
- Estimate the standard error of the sample mean:
- Use df = n – 1 to calculate the degrees of freedom to find the critical value of z, or t, to find the desired confidence interval: CI = P ± zcritical * Sp
- It will be necessary to do this for both upper and lower bounds (which explains why ± appears in the equation), meaning that you will have to solve the equation for a positive and negative value of zcritical