Unit 4: Experimental Research Methods

Question 1

Independent variable

Answer 1

The variable under the researcher’s control. Can be manipulated to see how it affects the dependent variable. The different levels are called conditions.

Question 2

Dependent variable

Answer 2

The variable that is measured. It is not manipulated at all.

In an experiment, observed changes to this variable can be attributed to manipulations made to the independent variable.

Question 3

Random assignment

Answer 3

When an experimenter uses a random method (flip of a coin, roll of the dice) to assign participants to a condition.

Ensures that groups on average are similar on all characteristics except what the experimenter manipulates.

Allows inference to be made that changes to independent variable caused changes to dependent variables.

Question 4

Confounds

Answer 4

Unintended variables that vary across values of the independent variable.

When you compare groups that systematically differ in some way, other than the intended manipulation.

Undermine the ability to draw causal inferences because you can’t differentiate between it’s effects and that of the independent variable.

Question 5

Control condition

Answer 5

Designed to eliminate possible confounds from the experiment, by removing possible differences between variables

Ex: How does yoga affect stress… all participants attend group classes 3 times a week (so experience of getting out of the house, interacting with class members and instructor is similar for each participant)

Question 6

Placebo effect

Answer 6

Participants’ awareness of receiving special treatment affects their expectation of how they should feel, and also their behaviour.

Experimenters’ expectations can also be affected if they know which participants receive a special treatment, and could influence their observations.

Question 7

Double-blind method

Answer 7

Both participants and experimenters are unaware as to who has been assigned to the control/experimental conditions.

Question 8

Control variables

Answer 8

Variables that are applied equally to all conditions of an experiment.

Ex: age

Question 9

Complex multi-factorial design

Answer 9

Experimental designs that involve at least 2 factors.

Ex: 2x3 study has 2 factors and 3 levels. So participants are divided into 2 groups, and each group has 3 different conditions, for a total of 6 groups.

Question 10

Benefits of complex design

Answer 10

1. More closely approximates the complexity of the real world. Increases validity of findings.
2. Allow for the study of interactions; when the effect of one factor is influenced by another factor.
3. More efficient use of researchers’ resources

Question 11

4 kinds of validities

Answer 11

1. Internal validity
2. External validity
3. Construct validity
4. Statistical validity

Question 12

Internal validity

Answer 12

When the way a study is conducted provides strong support for causal conclusions.

Ex: When two or more conditions are highly similar and manipulations to the independent variable produce only one difference in the dependent variable.

Question 13

External validity

Answer 13

When the way a study is conducted supports generalizing the results to people and situations beyond what was actually studied.

Question 14

Mundane realism

Answer 14

When the participants and the situation studied are similar to those that the researchers want to generalize to and participants encounter everyday.

Question 15

Psychological realism

Answer 15

When the same mental process is used in both the laboratory and in the real world.

Question 16

Construct validity

Answer 16

The quality of the experimenter’s manipulations and how well they relate to the research question.

Adding conditions may not increase it.

Question 17

Statistical validity

Answer 17

Whether the statistics conducted in the study support the conclusions that are made.

The number of conditions and the number of total participants will determine the overall size of the effect.

Question 18

Extraneous variables

Answer 18

Anything that varies in the context of a study other than the independent and dependent variables.

Ex: IQ

Question 19

Confounding variable

Answer 19

An extraneous variable that differs on average across levels of the independent variable.

Ex: Average IQ of condition 1 is considerably higher than the average IQ of condition 2. If variation of IQ is similar in both conditions, then the extraneous variable is not a confounding variable.

Question 20

Between-subjects experiments

Answer 20

When each participant experiences one level of the independent variable.

Researcher must assign participants to conditions so that the different groups are, on average, highly similar to each other to avoid confounding variables.

Question 21

Within-subjects experiments

Answer 21

When each participant experiences all levels of the independent variable.

Provides maximum control of extraneous participant variables because each group is made up of the same participants.

Disadvantage is the carryover effect.

Question 22

Block randomization

Answer 22

A modified random assignment that ensures equal number of participants per group.

Question 23

Carryover effect

Answer 23

When participation in one condition impacts performance in another condition later on. The order of the conditions becomes a confounding variable.

Types:

Practice effect; a task is performed better in later conditions because participants had a chance to practice it

Fatigue effect; a task is performed worse in later conditions because participants become bored or tired

Context effect; participants’ perception of stimuli or task is changed by previous condition.

Question 24

Counterbalancing

Answer 24

Testing different participants in different orders, to avoid carryover effect.

Controls the order of conditions so that carryover effect is no longer a confounding variable.

Or it allows carryover effect to be detected.

Question 25

Between-Subjects vs Within-Subjects (when is it best to use one or the other?)

Answer 25

Within-subjects is best if:

1. Proper counterbalancing in place
2. There is enough time per participant
3. No concerns about carryover effects

Between-subjects is best if:
1. It is too difficult or impossible to carry out a within-subjects experiment (i.e. participants won’t commit to time required to do all conditions)

Question 26

Single-subject research

Answer 26

Involves studying a small number of participants and focusing intensively on the behaviour of each one.

Ex: Applied behavioural analysis

Underlying assumptions:

1. It is important to focus intensely on the behaviour of individual participants because treatment can affect individuals differently (which would cancel out the effect in group research)
2. It is important to use experimental methods to discover causal relationships; has good internal validity
3. It is important to study strong and consistent effects that have biological or social importance; social validity