Study: interaction between mother & daughter determines level of aggression in g
ID: 2955567 • Letter: S
Question
Study: interaction between mother & daughter determines level of aggression in girls. 5 girls for each 4 interaction levels -measured their aggression scores.data are given below:No interaction: 4,5,6,6,5 Low Interaction: 4,6,5,4,5Moderate Interaction: 4,4,3,5,4
High Interaction: 3,4,4,3,3,
Summary of information Source Sum of Squares
df Mean of Squares
F Between Groups
9.75 3
3.250 5.91 Within Groups
8.80 16 0.550 Total 18.55 19
QUESTION:
*
What are the independent variable and the dependent variables?
*
Should we conduct a between-subjects ANOVA or a within-subjects ANOVA? Explain your answer.
*
How many factors are involved in Katie's study?
*
How many levels are involved? Name the level(s).
*
What are the null and alternate hypotheses?
*
Using a = 0.05, what is Fcrit?
*
Is the F-test significant?
*
Based on your answer about significance, what else should you do and why? Summary of information Source Sum of Squares
df Mean of Squares
F Between Groups
9.75 3
3.250 5.91 Within Groups
8.80 16 0.550 Total 18.55 19 F Source Sum of Squares
df Mean of Squares
F Between Groups
9.75 3
3.250 5.91 Within Groups
8.80 16 0.550 Total 18.55 19
Explanation / Answer
a) What are the independent variable and the dependent variables? The independent variable is the interaction status (categorical variable), and the dependent variable is the aggression index (continuous variable). b) Should we conduct a between-subjects ANOVA or a within-subjects ANOVA? I don’t under this part of the question. I’m interpreting this as a one-way ANOVA, The between group sum of squares is far greater than the within groups sum of squares, which would indicate that the between groups sum of squares matter. So I’m guessing we should conduct a between-subject ANOVA. c) How many factors are involved in Katie's study? This is a one-way ANOVA. There is one factor. d) How many levels are involved? Name the level(s). There are 4 levels of interaction: none, low, moderate and high e) What are the null and alternate hypotheses? Null: The aggression averages across the 4 groups are the same. Alternative: At least one average is different. f) Using a = 0.05, what is Fcrit? F(3,16,a = 0.05) = 3.238 g) Is the F-test significant? Since 5.91 > 3.238, the F is significant. h) Based on your answer about significance, what else should you do and why? Test the differences between the groups. There would be 4 choose 2 = 6 contrasts that you would examine. You may want to apply Bonferroni or a Tukey correction in conducting multiple tests. Hope this helps, Mike