Chapter_7_The Internal Validity of Research

Question 1

Internal Validity

Answer 1

The extent that we can have confidence that the results of a study are due only to the effects of the independent variable

Question 2

Extraneous Variables

Answer 2

Provide alternative explanations for the observed effect
- if they are present, the researcher cannot conclude that the independent variable caused the dependent variable

Question 3

2 Ways to Deal with Alternative Explanations

Answer 3

1. Logic
2. Control measures or research designs that estimate their effects as well as the effect of the independent variable

Question 4

2 Types of Alternative Explanations

Answer 4

1. Confound
2. Artifact

Question 5

Confound

Answer 5

When two VARIABLES overlapped to the extent that the effect of one cannot be separated from the effect of the other

Question 6

Artifact

Answer 6

When some aspect of the RESEARCH situation other than the independent variable affects the dependent variable

Question 7

Natural Confounds

Answer 7

In nature, some variables tend to be associated with certain other variables

Question 8

Treatment Confounds

Answer 8

When the manipulated independent variable (or treatment) in an experiment is confounded with another variable or treatment
- they actually received two combined treatments

Question 9

Measurement Confounds

Answer 9

sometimes a dependent variable measures more than one hypothetical construct
- e.g. anxiety and depression

Question 10

Threats to Internal Validity

Answer 10

1. History
2. Maturation
3. Testing
4. Instrumentation Change
5. Statistical regression

Question 11

Threats to Internal Validity: History

Answer 11

Events outside lab
- Solution: check to see if any such an effect is present

Question 12

Threats to Internal Validity: Maturation

Answer 12

Natural change over time
- age, experiences
- Solution: distributing experimental and control sessions of a study evenly across the time period

Question 13

Threats to Internal Validity: Testing

Answer 13

pretest affects posttest
- Solution: not to give a pretest
- Solomon Four-Group Experimental Design

Question 14

Threats to Internal Validity: Instrumentation Change

Answer 14

Artificial differences in scores at different points in time
- Mechanical and digital measuring
- observer drift: coders becoming
less reliable over time
- Solution: periodically test the equipment
- construction of classification systems
- training observers
- have different observers

Question 15

Statistical Regression

Answer 15

When extreme scorers are measured a second time, random error can have little influence in raising extremely high scores or lowering extremely low scores
- Solution: not to select research participants on the basis of extreme scores on the dependent variable

Question 16

Theoretical Validity

Answer 16

Some theories specify conditions that must be met for the predictions derived from the theories to be borne out
- if the conditions the theory sets for its effectiveness are met -> validity
- e.g. ELM replications differed from
the original study in an important way
- need for cognition should have an effect only for low relevance issues

Question 17

Selection Bias

Answer 17

participants in the control condition differ in some way from those in the experimental condition

Question 18

Volunteer Bias

Answer 18

• better educated
• higher SES
• more sociable than non-­ volunteers

Question 19

Pre-existing Groups

Answer 19

people in pre-existing groups are likely to have common characteristics
- e.g. students chose which section to enroll in

Question 20

Attrition

Answer 20

• characteristics match those of the “survivors”

Question 21

Reactivity

Answer 21

Whenever the PROCESS of measuring a variable rather than the CONTENT of the measure affects scores on the dependent variable

Question 22

2 Sources of Reactivity

Answer 22

1. Evaluation apprehension
2. Novelty effects

Question 23

Evaluation Apprehension

Answer 23

The nervousness people feel when they believe someone is judging their behavior
-> social desirability

Question 24

Novelty Effects

Answer 24

Any aspects of the research situation that are new (or novel) to the participants can induce reactivity
- paying attention to the novel features
- especially in children

Question 25

Controlling Reactivity

Answer 25

1. General Control Measures 2. Controls With Behavioral Measures 3. Controls With Self-Report Measures 4. Noting Instances of Reactivity

Question 26

General Control Measures

Answer 26

1. Avoid LABELING experimenters as psychologists 2. Avoid giving negative FEEDBACK 3. greeting them in a FRIENDLY manner 4. EXPLAINING the experiment to them 5. giving participants time to ADJUST to the new situation 6. DECEIVING about the nature of the research

Question 27

Controls With Behavioral Measures

Answer 27

- keep the observers HIDDEN or have them play a role that allows them to blend into the environment - researcher PRETENDS to be just another person in the situation that he or she observes

Question 28

Controls With Self-Report Measures

Answer 28

- respond anonymously -> reduce evaluation apprehension and so can lessen reactivity - convince people that you know when they are not being completely truthful

Question 29

Noting Instances of Reactivity

Answer 29

1. Knowledge of the circumstances that lead to reactivity 2. Situations that raise sensitive issues 3. Noting reactivity from natural responses during data analysis

Question 30

Demand Characteristics

Answer 30

the information present in the research situation that allows participants to form their own hypotheses

Question 31

Sources of Demand Characteristics

Answer 31

1. informed consent procedure 2. research procedures

Question 32

Participant Roles

Answer 32

after participants can deduce research hypothesis they react in 3 different ways

Question 33

The Good Participant

Answer 33

Social desirability response bias

Question 34

The Negative Participant

Answer 34

Feel they are being coerced into doing something they don’t want to do - PSYCHOLOGICAL REACTANCE

Question 35

The Apathetic Participant

Answer 35

Feel no motivation - pay little attention - responding in a random fashion - rushing through

Question 36

Controlling Demand Characteristics

Answer 36

1. Cue Reduction 2. Motivation 3. Role-Play Control Groups 4. Measurement of the Dependent Variable

Question 37

Controlling Demand Characteristics: Cue Reduction

Answer 37

Demand characteristics - review during proposal, pilot studies and post-experimental interviews - deceiving with false hypothesis

Question 38

Controlling Demand Characteristics: Motivation

Answer 38

- Freedom to decline participation and to withdraw from research - record the amount of time a participant spent completing the study -> rushing -> low motivation

Question 39

Role-Play Control Groups

Answer 39

describe your manipulation to a group of participants and ask them to role play—that is, to respond as if they were actually taking part in the study - reflect demand charateristics

Question 40

Measurement of the Dependent Variable

Answer 40

- People believe their participation in a study has ended - a dependent variable is measured outside the context of a study

Question 41

Experimenter Expectancy

Answer 41

Expectation about how participants should perform affect the way they behaved toward the participants - their behavior could then affect the participants’ responses to the research task, biasing the data

Question 42

2 Sources of Experimenter Expectancy

Answer 42

1. Research hypothesis 2. Data collected early in a research project

Question 43

Biased Observation

Answer 43

Expectation -> interpretation of what they observe

Question 44

Influencing Participants’ Responses

Answer 44

1. treating experimental and control participants differently 2. give nonverbal feedback 3. intentionally bias their data

Question 45

Techniques of Control

Answer 45

1. Rehearsal and Monitoring 2. Minimizing the Experimenter’s Role 3. Condition Masking 4. Avoidance of Data Snooping