1. The standard error of the mean can be calculated by dividing μ by the square root of the number of values in the distribution.
2. According to the central limit theorem, a population which is skewed to begin with will still be skewed when it is re-formed as a distribution of sample means.
3. The mean of the distribution of sample means will have the same value as the mean of the distribution of individual scores upon which it is based.
4. What question does the z test answer?
- Is the individual characteristic of the group?
- Has there been a type I error?
- Does the sample represent the population?
- Are the data normal?
5. The desired sample size depends only the size of the population to be tested.
6. What is the probability of type II error when the null hypothesis is rejected? @Answer found in section 4.3 The One-sample t-Test, in Statistics for Managers
7. The test statistic follows the formula of difference divided by variability.
8. How do statistical tests like the one sample t adjust for the absence of parameter values? @Answer found in section 4.3 The One-sample t-Test, in Statistics for Managers
- The values are estimated from sample data.
- The values are assumed to have a constant value.
- The test is reconstructed so that the values aren't needed.
- The test is reformulated so that data are always normal.
9. How does variability in the distribution of sample means compare to variability in a population based on individual scores?
- Samples tend to vary less than individual scores.
- Samples exaggerate differences among scores.
- Individual scores tend to be more stable over time than samples.
- Sample means vary less than individual scores.
10. Which of the following is a provision of the central limit theorem?
- A skewed distribution will remain skewed however it is plotted.
- There are limits to the range of scores that can be fitted to a distribution.
- A distribution based on sample means will be normal.
- There will always be theoretical differences between distributions.
11. While rejecting the null hypothesis for the goodness of fit test means distributions differ, rejecting the null for the test of independence means the variables interact.
12. Point estimates provide less confidence in indicating a parameter's value than a confidence interval.
13. The Chi-square test is very sensitive to small differences in frequency differences.
14. For a two sample confidence interval, the interval shows the difference between the means.
15. For a one sample confidence interval, the interval is calculated around the estimated population mean or standard (μm ).
16. For a one sample confidence interval, if the interval contains the μm , the corresponding t-test will have a statistically significant result - rejecting the null hypothesis.
17. The Chi-square test results having expected values of less than 5 in a cell may produce a greater likelihood of having type I errors (wrongly rejecting the null hypothesis).
18. The percent confidence interval is the range having the percent probability of containing the actual population parameter.
19. A contingency table is a multiple row and multiple column table showing counts in each cell.
20. Confidence intervals provide an indication of how much variation exists in the data set.