Name______

C. Mandell - Spring, 2008

47.375 - Research 3

Lab 2

One way Analysis of Variance

Problem 1 - Sample problem from Kirkpatrick, Chapter 10

A researcher conducted a study to examine the effects of different types of attachment (anxious, avoidant and secure) on the physiology of sleep. The investigator wanted to determine whether children with anxious or avoidant attachment styles experience more sleep disturbances than children with secure attachment. The percentage of deep sleep was monitored for three groups of children with different attachment styles.

Table 1. Percentage of deep-sleep in three groups of children with different attachment styles

Secure-attachment / Anxious-Attachment / Avoidant-Attachment
21
21
25
23
24
23
23
22
22
22 / 17
17
15
15
15
14
20
13
14
19 / 18
20
18
19
17
17
15
16
17
18

I. Looking for co-variation -

A. Use error bar plot to visually assess the amount of covariation between attachment-style and percent deep sleep (delta)

Summarize these findings:

B. Use explore to compare means, sd's, confidence levels for the different attachment styles.

Summarize these findings:

Secure / Anxious / Avoidant
Mean
Standard deviation
CI-upper limit
CI-lower limit

C. Use Analyze->Compare Means->One-way analysis to run the analysis of variance and Post-hoc tests?

A. What are the null and alternative hypotheses?

B. Do you have an overall significant F? What does this mean?

C. Based on the Tukey test (In Post hoc tests), where does the significance lie?

Problem 2 - Homework problem from Zechmeister, Chapter 12 - number 9

Tome tests three groupsof rats (n - 10) in a maze after they have been injected with one of three memory-enhancing drugs (A, B and C) Number of errors while learning the maze is the dependent variable. The results of his expiments for 30 rats are as follows;

Drug A: 4, 6, 8, 3, 8, 4, 5, 6, 7, 5

Durg B: 10, 11, 12, 6, 7, 9, 9, 9, 15, 12

Drug C: 2, 6, 5, 5, 5, 4, 7, 8, 9, 11

I. Looking for co-variation -

A. Use error bar plot to visually assess the amount of covariation between drug-type and number of errors.

Summarize these findings:

B. Use explore to compare means, sd's, confidence levels for the different drug types.

Drug A / Drug B / Drug C
Mean
Standard deviation
CI-upper limit
CI-lower limit

C. Use Analyze->Compare Means->One-way analysis to run the analysis of variance and Post-hoc tests?

A. What are the null and alternative hypotheses?

B. Do you have an overall significant F? What does this mean?

C. Based on the Tukey test (In Post hoc tests), where does the significance lie?