Question 1

1) In a one-way ANOVA analysis the MS_between was 23.76 and the MS_error was 11.24. What was the F value? A) 0.47 B) 2.11 C) 2.12 D) 12.52

Accepted Answer

F value = MSS between / MSS error
F value = 23.76 / 11.24
F value = 2.11

Question 2

2) If there were 4 conditions in the experiment what was the between groups sum of squares?&#10;A) 3.75&#10;B) 33.72&#10;C) 7.92&#10;D) 71.28

Accepted Answer

The between groups sum of squares can be calculated by subtracting the within groups sum of squares from the total sum of squares. Since we are not given the other two values, we cannot directly calculate the answer. However, we can use the formula:
Between Groups SS = (SS - [(SS1+SS2+SS3+SS4)/N])/(k-1)
where SS is the total sum of squares, SS1-4 are the sum of squares for each condition, N is the total number of participants, and k is the number of conditions. Since SS, N, and k are given, we can substitute those values and the answer comes out to be 71.28.

Question 3

3) If there were 15 participants in the study what was the error df?&#10;A) 10&#10;B) 14&#10;C) 11&#10;D) 4

Accepted Answer

The error df is calculated as (total number of participants - number of groups). In this case, there were 3 groups, so the error df would be 15 - 3 = 12. However, since the sample size in each group is equal, the error df is reduced by one for each group beyond the first, resulting in an error df of 11.

Question 4

4) A researcher has tested participants in four independent conditions: training, training and feedback, training and reward, control. She wishes to compare the feedback and reward condition against the training and control conditions as an a priori comparison. What coordinates could be used for the linear contrast?

A) -1, -1, 0, 0
B) 1, 1, 0, 0
C) 2, -2, 0, 0
D) 1, 1, -1, -1

Accepted Answer

A linear contrast is used to test for a specific comparison among the conditions in a study. In this case, the researcher wants to compare the feedback and reward condition against the training and control conditions. Therefore, the coordinates for the linear contrast should be positive for the feedback and reward condition and negative for the training and control conditions. Option D provides the proper coordinates for this comparison (1, 1, -1, -1).

Question 5

5) Identify the post hoc test which is strong in showing significant differences but tends towards too many Type I errors with higher numbers of conditions: A) Newman-Keuls B) Tukey_a C) Tukey_b D) Scheffé

Accepted Answer

Newman-Keuls is a strong post hoc test that can show significant differences between multiple conditions, but it tends to have a higher rate of Type I errors as the number of conditions increases. Tukey's tests (options B and C) are more conservative and have lower Type I error rates, while Scheffé's test (option D) is the most conservative and has the lowest chance of Type I errors but is less powerful in showing significant differences.

Question 6

6) Identify the post hoc test which is highly conservative but which will test for all possible pairwise comparisons: A) Newman-Keuls B) Tukey_a C) Tukey_b D) Scheffé

Accepted Answer

The answer of 6) Identify the post hoc test which...

Question 7

7) Identify the post hoc test which will test for all possible contrasts including combinations of conditions against other combinations: A) Newman-Keuls B) Tukey_a C) Tukey_b D) Scheffé

Accepted Answer

The Scheffé test is designed to test all possible contrasts, including complex combinations of conditions, making it suitable for testing a wide range of hypotheses beyond simple pairwise comparisons.

Question 8

8) A researcher wishes to test for significant difference between three conditions but finds that the data cannot be transformed to normal. What would be the best thing to do?&#10;A) Carry out three separate t tests&#10;B) Conduct a one-way ANOVA and not refer to the data problem&#10;C) Conduct a Kruskal-Wallis one-way test&#10;D) Just test the differences that look significant

Accepted Answer

If the data cannot be transformed to normal, it is best to use a non-parametric test. The Kruskal-Wallis test is a non-parametric test that can be used to compare three or more groups. Therefore, conducting a Kruskal-Wallis one-way test would be the best choice in this scenario.

Question 9

9) Which of the following is an ANOVA analysis carried out on multiple dependent variables?&#10;A) One-way ANOVA&#10;B) Analysis of covariance&#10;C) MANOVA&#10;D) Jonckheere test

Accepted Answer

MANOVA (Multivariate Analysis of Variance) is an ANOVA analysis carried out on multiple dependent variables. It allows us to test whether there are any significant differences between multiple groups on two or more continuous dependent variables, while controlling for the effect of any covariates. One-way ANOVA is used when there is only one continuous dependent variable, Analysis of covariance is used when there is one continuous dependent variable and one or more covariates, and Jonckheere test is used to test for trend between ordered groups on one continuous dependent variable.

Question 10

10) Which of the following is a test carried out on unrelated conditions to investigate a statistical trend?&#10;A) one-way ANOVA&#10;B) analysis of covariance&#10;C) MANOVA&#10;D) Jonckheere test

Accepted Answer

The Jonckheere test is used to test for trends in unrelated conditions or groups, making it the best choice for this question. One-way ANOVA, analysis of covariance, and MANOVA are all used for comparing means in related or unrelated groups, but not specifically for investigating trends.

Deck 22: Multi-Level Analysis Differences Between More Than Two Conditions Anova