Chapter 8 - Between Subjects design

Question 1

The between and within groups designs are for which type of research study?

Answer 1

The experimental research strategy

Question 2

How many scores are associated with each participant for the between groups design?

Answer 2

1 score = 1 participant

1 level of ind v = 1 result on dep v

Question 3

Can multiple measures be produces with between subjects design?

Answer 3

Yes, but only one score will be produced (average)

Question 4

Are between subejcts experiments sensitive to carryover effects? Why?

Answer 4

No, because each participant only makes 1 score (1 level of ind v)

Question 5

What are 2 disadvantages of the between subjects design?

Answer 5

1- Can require a lot of participants

2- Individual differences

Question 6

What are individual differences?

Answer 6

The participants of a study do not all have the same individual characteristics (ex: age)

Question 7

What are the 3 main concerns regarding individual differences in studies?

Answer 7

1- Can become confounding variables
2- Can produce high variability in scores
3- Threatens internal validity

Question 8

How are individual differences also called?

Answer 8

Selection or assignment bias

Question 9

How can we avoid individual differences from influencing the results in between subjects experiments?

Answer 9

By making sure that the groups of participants are as similar as possible except for their level of the ind v

Question 10

What are the 2 major sources of concern for confounding variables in between subjects designs?

Answer 10

Individual differences and environmental variables

Question 11

What are environmental variables?

Answer 11

○ Any characteristics of the environment that may differ between groups
Ex: tested in the morning/night, in different rooms, etc

Question 12

What is the concept of equivalent groups?

Answer 12

To avoid individual differences and environmental variables separate groups must be:
• Created equally
• Treated equally (except for treatment conditions)
• Composed of equivalent individuals

Question 13

What are the 3 procedures used to limit confounding by individual differences in between subjects designs?

Answer 13

1- Random Assignment (randomization)
2- Matching groups (matched assignment)
3- Holding constant (restricting range of variability)

Question 14

What is random assignment (randomization)?

Answer 14

All individuals have = chances of being assigned into any group, makes it likely that confounding variables will also be distributed

Question 15

What is restricted random assignment?

Answer 15

Creating random groups of equal sizes

Question 16

What is the problem with random assignment?

Answer 16

Sometimes chance does not distribute variables equally; does not always create equal groups

Question 17

What is the procedure os matching groups (matched assignment)?

Answer 17

Use matching to ensure that possible confounding variables are equally distributed between groups
Match individuals in the groups on a specific individual difference

Question 18

What are the 3 steps required to do matching?

Answer 18

1. Identification of the variable to be matched across groups
2. Measurement of the matching variable for each participant
3. Assignment of participants to groups by restricted random assignment

Question 19

How can we hold variables constant or restrict their range of variability?

Answer 19

Eliminate the variable or restrict its range of values (this limits external validity)

Question 20

What is the difference between holding constant and matching?

Answer 20

Holding constant: same variable for all the participants

Matching: the individual differences on one variable in one group are matched into all other groups in the same manner

Question 21

What is variance?

Answer 21

Variance: statistical value that measures the size of the differences from one score to another

Question 22

Why do we want to decrease the variance of individual differences within the groups?

Answer 22

Because it makes it easier to see the impact of the treatment

Question 23

Do we prefer differences between treatments or variance within treatments?

Answer 23

• Big differences between treatments are good because it makes it easier to see the effect of the treatment on the groups
• Big variance of scores within conditions is bad because it obscures the effect of the treatment (less distinguishable)

Researchers try to increase diff between treatments to reduce variance within treatments

Question 24

How can we minimize the variance within treatments?

Answer 24

• Standardize procedures and treatment settings
• Limit individual differences
• Random assignment and matching (minimize between group differences due to individual differences)
• Sample size (takes a big increase in sample size to obtain an effect on variance)

Question 25

What is attrition?

Answer 25

Participant’s withdrawal before completion of the study

Question 26

What is differential attrition?

Answer 26

Differences in attrition rates from one group to another (threat to internal validity) - obscures the effect of the treatment

Question 27

What is diffusion?

Answer 27

spread of the treatment from the experimental group to the control group

Question 28

What is compensatory equalization?

Answer 28

when one group learns about the treatment of the other group and demands the same treatment or equal

Question 29

What is compensatory rivalry?

Answer 29

when one group learns about the treatment of the other group and works extra hard to show that they can perform just as well as the individuals with the other treatment

Question 30

What is resentful demoralization?

Answer 30

untreated group becomes less productive and less motivated because they resent the expected superiority of the treated group

Question 31

What statistical analysis we use when there are only 2 groups to compare the treatments?

Answer 31

Single-factor group design (two-group design)

Mean is computed for each group, then t-test

Question 32

What statistical analysis method can we use when we have more than 2 groups to compare?

Answer 32

Single-Factor multiple-group design

Mean is computed for each group and then a Single-factor analysis of variance (ANOVA) is performed

Question 33

How can we compare the results when we have ordinal/nominal measurements?

Answer 33

We cannot compute the means; we measure the frequency of scores for each category and then we use the Chi-Square test