# type item number for each answer and a true or false answer only will be accepted and counted do not include item sentences below or any questions for this email examples 1 true 2 false and so on to include 20 item numbers with true or false

Type item number for each answer, and a â€œTrueâ€ OR â€œFalseâ€ answer only will be accepted and counted. Do not include item sentences below or any questions for this email. Examples: 1. True, 2. False, and so on to include 20 item numbers with True OR False answers.

Final Exam

Ch.13

1. When the null hypothesis is true, the F-ratio for ANOVA is expected, on average, to be zero.

2. The larger the differences between treatment means, the larger the F-ratio.

3. The critical region for an ANOVA is located entirely in one tail of the distribution.

4. Post hoc tests are necessary after an ANOVA when H0 is rejected and there are more than two treatments.

5. In general the distribution of F-ratios is positively skewed with all values greater than or equal to zero.

Ch. 14

6. For a repeated-measures design, the differences between treatments may be caused by a treatment effect or by chance/error, but they cannot be caused by individual differences.

7. For the repeated-measures analysis, the F-ratio, on average, is expected to be zero when H0 is true.

8. A two-factor ANOVA with 2 levels of factor A and 3 levels of factor B involves six separate hypothesis tests.

9. In a two-factor ANOVA, all of the F-ratios use the same denominator.

10. In the F-ratio for repeated-measures ANOVA, variability due to individual differences is automatically eliminated from the numerator but must be computed and subtracted out of the denominator.

Ch. 15

11. A negative correlation indicates that decreases in X tend to be accompanied by decreases in Y.

12. The Spearman correlation is used to measure the relationship when both variables have been measured on an ordinal scale (both are ranks).

13. The point-biserial correlation is used with data that also would be appropriate for an independent-measures t hypothesis test.

14. The numerical value for a correlation can never be greater than 1.00 and can never be less than -1.00.

15. In the general linear equation, Y = bX + a, the value of b is called the slope.

Ch. 16

16. In a chi-square test, the observed frequencies are always whole numbers.

17. The value obtained for the chi-square statistic will always be greater than or equal to zero.

18. If the null hypothesis for a goodness-of-fit test predicts no preference, then the expected frequencies will be the same for every category.

19. In general, a large value for the chi-square statistic indicates that the null hypothesis is correct.

20. The phi-coefficient can be used to measure effect size for a chi-square test for independence provided there are exactly two categories for each of the two variables.