Chapter 11

Question 1

How many levels does the independent variable have in the simplest experimental design?

Answer 1

only 2 levels (also called groups or conditions)

Question 2

When referring to an independent variable, what do we mean by level?

Answer 2

The operationalization of the independent variable in an experiment, often referred to as the conditions or groups.

Question 3

Why might researchers be interested in having three or more levels of the independent variable?

Answer 3

• researchers are frequently interested in comparing more than 2 groups. (sometimes an additional control condition can provide greater clarity)
• a design with only 2 levels of the IV may not provide enough info about the relaitonship between the independent and dependnt variables. (2 variables only allows us to identify linear relationships but the reality might be a non-linear relationship which would require additional levels to conclude)

Question 4

Can an experiment with only 2 conditions detect curvilinear relationships?

Answer 4

no.

Question 5

Can you add aditional independent variables to an experiment? Why might you choose to do this?

Answer 5

yes.

adding more independent variables to an experiment brings it closer to the real world conditions in which many influences (ie IVs) interact with one another to produce some behaviour or effect.

Question 5

How many levels must be used if a curvilinear relationship is predicted?

Answer 5

at least 3

Question 6

What does having more than one independent variable allow you to do? What are these designs known as?

Answer 6

examine how these 2 variables influence the dependent variable, and also how they might interact.

These designs are known as factorial designs.

Question 7

What is a factorial design?

Answer 7

An experiment with more than one independent variable (also called a factor), with each factor having at least two levels (i.e., two conditions).

Question 7

What is the simplest factorial design? What is the shorthand we use to describe a factorial design? What would be the shorthand for the simplest factorial design?

Answer 7

a design that has 2 independent variables, each with 2 levels.

Shorthand:

Number of levels of first IV x Number of levels of second IV

Shorthand for simplest:
2 x 2 factorial design

Question 8

How many conditions does a 2 x 2 factorial design result in?

Answer 8

4

Question 9

How many distinct kinds of information do factorial designs yield? What are they?

Answer 9

2 distinct kinds of info:

1. information about the affect of each independent variable, taken by itself: the main effect of each independent variable.
2. an interaction. If there is an interaction between two independent variables, the way that one independent variable affects the dependent variable depends on the level of the other variable. Interactions are a very valuable source of info tat does not exist in simple experiments with only one IV

Question 10

What is a main effect? In a design with 2 independent variables, how many main effects are there?

Answer 10

The direct effect of an independent variable on a dependent variable, ignoring any interaction with other variables.

in a design with 2 IVs there are 2 main effects.

Question 11

Do interactions exist in simple experiments with only one IV?

Answer 11

no

Question 12

What is an interaction?

Answer 12

When the effect of one independent variable on the dependent variable depends on the level of another independent variable.

Question 13

What is a cell of a table?

Answer 13

A single entry in a table, sometimes used to refer to one condition in an experiment, or to a combination of conditions in a factorial design.

Question 15

What does the main affect of Independent variable A capture? What about the main effect of IV B?

What does the main effect essentially pretend?

Answer 15

captures its overall effect on the dependent variable.

captures the effect of this independent variable on the dependent variable.

the main effect essentially pretends that the other IV doesn’t exist in the experiment.

Question 15

Describe the common method of presenting results for a factorial design.

Answer 15

In the candy experiment provided:

• the number in each cell of the table represents the mean number of candies people ate.
• the mean number of candies eaten for participants who were in the thin-confederate 30 candy condition can be found in the corresponding cell of the table (it is 9.82)

Question 15

What is mean a formal term for?

Answer 15

arithmetic average

Question 16

What is a marginal mean? Why are they called this? how is the marginal mean calculated?

Answer 16

In a factorial design, the average score of all participants in one condition of one independent variable, collapsing across all other variables.

the marginal means are found in the bottom row of the table which along with the rightmost column are called the margins of the table, hence marginal means.

the marginal mean is the average of the 2 averages form both of the independent variable’s conditions.

Question 17

If there is a possibility that an interaction exists, why is it most important to interpret it?

Answer 17

because it an interaction indicates that the main effects need to be qualified (qualified refers to the fact that the main effects are conditional or contingent on something else)

Question 18

What does an interaction between independent variable indicate?

Answer 18

that the effect of one independent variable varies at different levels of the other independent variable.

Question 19

If there is an interaction, how do we need to structure our conclusions?

Answer 19

you must consider both of the independent variables together when making claims about the impact on the DV

Question 21

in many studies, what are interactions discussed in terms of?

Answer 21

a moderator variable

Question 21

How can we describe a moderator variable in terms of the interaction in the candy consumption body size study?

Answer 21

“People eat more food when an eating companion takes more rather than less food to eat. “ BUT we have to make a qualification by stating

” People eat more food when an eating companion overindulges only when the companion is thin: when the companion takes little food, the amount people eat is not influenced by the companion’s body type.”

AKA the body type variable is a moderator variable.

Question 21

What is a moderator variable?

Answer 21

A third variable that influences the relationship between an independent variable and a dependent variable. In a factorial design, the effect of moderator variables are revealed as interactions.

Question 22

What can moderator variables be aspects of ?

Answer 22

the situation or of the participants

Question 23

What are the 3 possibilities when analyzing the results of a 2 x 2 factorial design?

Answer 23

1. there may not be a main effect for IV A
2. there may not be a main effect for IV B
3. there may or may not be an interaction between the IVs

Question 24

When looking at a graph of the results of a factorial design, what question should you ask to determine whether there is an interaction?

Answer 24

Are the 2 lines in the graph parallel to each other?

If the lines are NOT Parallel then there IS and interaction

AKA the effect of one variable is different at different levels of the other variable

Question 25

Is it easier to examine a table or graph when looking for main effects? Why?

Answer 25

easier to examine a table because the marginal means will stand out in a table.

Question 26

When there is an interaction, what do you need to carefully examine in te corresponding tables, to understand the specific pattern of interaction?

Answer 26

the means.

Question 27

Is it possible for there to be a strong relationship between the first IV and the DV at one level of the second IV but no relationship (or a weak one) at the other level of the second independent variable?

Answer 27

yes.

Question 28

Can an interaction indicate that an independent variable has opposite affects on the dependent variable depending on the level of the second IV? what would this pattern look like?

Answer 28

Yes.

lines would be in an x formation

Question 29

KNOW FIGURE 11.4

Question 30

What do we need to do to understand a signficant interaction? What is the next step?

Answer 30

whenever there is a significant interaction, we need to break it down further to understand it.

the next step is to look at the simple main effects.

Question 31

what is a simple main effect?

Answer 31

In a factorial design, the effect of one independent variable on the dependent variable, at one particular level of another independent variable.

the mean difference at each level of one independent variable.

Question 32

Are simple main effects and main effects the same thing? Why or why not?

What would we do to identify whether simple main effects are statistically significant or not?

Answer 32

no.

with simple main effects (in contrast) the results are within each level of the other independent variable.

would use statistical tests to identify if significant.

Question 34

Describe how we can look at the simple main effect of A in the food obesity study.

Answer 34

We can look at the simple main effect of A (food selection) within each level of B. Here, we compare the average food selections when the companion is either obese or thin. In this cas,e the stat. tests showed that the simple main effect of food selection is not statistically significant when the eating companion is obese (means of 6.25 vs 4.26), but the simple main effect of food selection is statistically significant when the eating companion is thin (means of 9.82 vs 3.20). We ignore the marginal means of fod selection and instead interpret the cell means within each level of the confederate body type.

Question 34

What does a common type of factorial design include? What are the 2 names for this design?

Answer 34

both experimental (manipulated), and non-experimental (measured or non-manipulated) variables.

an independent variable by participant variable design

OR

IV x PV design

Question 34

Should you analyze both of the simple main effects? Why or Why not? How do you decide which one?

Answer 34

the analyses for both overlap so we must choose to analyze only one of the simple main effects. Which analysis you are most interested in will depend on the predictions you made when you designed the study. The key point to remember here is that a significant interaction in a factorial design must be decomposed by examining cell means using a simple main effect analysis.

Question 34

Describe how we can look at the simple main effect of B in the food obesity study.

Answer 34

We can look at the simple main effect of B (body type) within each level of A. This will tell us whether the difference between the thin and obese eating companion is statistically significant when she eats 2 candies or 30 candies. In this case, the simple main effect of body type is not statistically significant when the eating companion eats 2 candies (mean of 3.2 vs 4.26), nor is the simple main effect statistically significant when the eating companion eats 30 candies (9.82 vs 6.25)

Question 35

What deos the IV x PV design allow? When we say different types of people what do we mean?

Answer 35

allows researchers to investigate how different types of people respond to the same manipulated variable.

By different types of people we mean individuals differing with respect to personal attributes such as age, ethnic group, personality characteristics, or clincial diagnostic category. (known as participant variables)

Question 36

Can participant variables be randomly assigned or controlled? Why? What does this mean for the IV x PV design?

Answer 36

no.

because they are things the participants bring with them.

this means the IV x PV design is not a true experiment becuase it contains a variable that is measured and not manipulated.

Question 36

What does the simplest IV x PV design include?

Answer 36

one manipulated IV that has two levels and one participant variable with 2 levels.

Question 37

What might the 2 levels of the participant variable in the IV x PV design be?

Answer 37

differnt age groups, groups that are low nad high on a personality trait, groups of short and tall individuals etx.

Question 38

Can techniques or assigning participants to conditions be generalized to factorial designs?

What techniques can be used?

Answer 38

yes.

a between subjects design, a within subjects design, or a mixed factorial design ( combination of the 2)

Question 39

What is a mixed factorial design?

Answer 39

A factorial experimental design that includes both between-subjects and within-subjects variables.

Question 40

How would a between subjects design work for a 2 x 2 factorial design? How many participants should be use?

Answer 40

different participants will be assigned to each of the 4 possible conditions.

for 10 observations in each condition, we should use 40 different participants.

Question 41

How would a within subjects design work for a 2 x 2 factorial design? How many participants should be use?

Answer 41

the same people will participate in all conditions.

if you want 10 participants in each condition, a total of 10 participants would be needed.

Question 41

How would a mixed factorial design work for a 2 x 2 factorial design? How many participants should be use?

Answer 41

Ex: psychopathy and head movements in truth and lie conditions

psychopathy is a between subjects design
truth or lie, is a within subjects design.

all participants told a turthful story and a lie but a participant was either a psychopath or not.

For 10 participants in each condition, 20 participants are necessary for the between subjects variable (10 in each of the 2 conditions), but only 10 participants are required to recieve both levels of the within subjects variable.

Due to this, a total of 20 participants would be needed. Each of the 10 participants in each conditions of the between subjects variable (aka all 20) would experience both of the conditions for the within subjects variable.

Question 42

What are 2 ways of increasing complexity of a factorial design?

Answer 42

increasing the number of levels of one or more of the independent variables

• increase the number of independent variables

Question 43

How do you describe more complex facotrial designs? (2 ways)

Answer 43

of levels of first IV x # of levels of second IV x # of levels of third IV and so forth.

Question 44

Explain what information you can get from this equation:

a 2 x 3 factorial design

Answer 44

contains 2 independent variables (A and B)

IV A has 2 conditions

IV B has 3 conditions

the 2 x 3 design has 6 conditions

Question 45

What do marginal means show?

Answer 45

the main effects of each of the IVs

Question 46

When are line graphs used?

Answer 46

when the independent varaible represented on the horizontal axis is quantitative (i.e the levels of the IV are increasing amounts of the variable)

Question 47

in an IV x IV x PV 2 x 2 x 2 design be seen as a 2 x 2 design?

Answer 47

yes. can be seen as a 2 x 2 design for each of the levels of the PV.

Question 48

What is the possible number of main effects in a factorial design dependent on?

Answer 48

the number of IVs (3 IVs, 3 possible main effects)

Question 49

What are three way interactions? Are these common?

Answer 49

an interaciton that involves all three variables. here, we want to determine whether the nature of the interaction between the 2 variables differs depending on the particular level of the other variable.

In behavioural sciences, three way interactions are complicated and less common than two way interactions.