PSYC*2360 Chapter 8: Two-Group Design Flashcards
For an experiment to establish causation, what are the three specific criteria it must meet?
- Covariation
- Temporal precedence
- Internal validity
What is covariation?
When changes in one variable are associated systematically with changes in another
What does it mean to have temporal precedence?
When changes in the suspected cause (IV) occur before changes in the effect (DV)
T or F: Correlational designs may help establish temporal precedence, but they can’t establish covariation.
False. May help establish covariation, but can’t establish temporal precedence.
What term refers to the degree to which it is possible to rule out other possible causal explanations for an observed relationship between the independent and dependent variables?
Internal validity
What are extraneous variables?
Factors other than the intended treatment that might change the outcome variable
T or F: To establish causality, extraneous variables need to be eliminated or controlled.
True
Why is internal validity required for determining causality?
It must be shown that covariation is only due to the independent variable and not an extraneous variable
What is mundane realism?
The degree to which a study parallels everyday situations in the real world
What is experimental realism?
The degree to which participants becomes engrossed in a manipulation and are truly influenced by it
What is a two-group design also known as?
A simple experiment
What is a two-group design?
An experimental design that compares two groups or conditions
Which experimental design represents the most basic way to establish cause and effect?
A two-group design
In an experimental design, which group or condition gets the key treatment?
The experimental group
In an experimental design, which group or condition serves as a comparison for the experimental group?
The control group
In an experimental design, what is an all-or-nothing approach?
The experimental group treatment to its fullest extent and the control group receives no treatment
What is one drawback to an all-or-nothing approach?
It can lead to several differences between groups and make the exact effect of treatment unclear
Which term refers to the ability to keep everything between groups the same, except for the one element being tested?
Experimental control
In an experimental design, what is an a-little-more-versus-a-little-less approach?
The experimental group receives a bit more of the treatment and the control group receives a little less
Which term refers to a clear and specific prediction of how the independent variable will influence the dependent variable?
Experimental hypothesis
What is the assumption that each participant represents a unique and individual data point?
Independence
T or F: When people know they are part of a certain group, even if arbitrary, they tend to favour that group over others.
True
T or F: Even when people are aware that they are part of a group, they don’t often take cues from other members, and each individual continues to act naturally.
False. As soon as people know they are part of a group, they often take cues from other members in ways that make each individual act less naturally.
What is one step that can be taken to ensure independence in an experimental design?
Interactions between participants should be minimized, and they should each be tested individually
What is random assignment?
Any method of placing participants in groups that is non-systematic and non-biased
Which method of assignment to groups ensures that each participant has an equal chance of being in any group?
Random assignment
What is a matched-pair design?
A design in which one creates a set of two participants who are highly similar on a key trait, then randomly assigns individuals in a pair to different groups
Which type of experimental design ensures both groups have participants who are similar in the trait important to the study?
A matched-pair design
When does a matched-pair design work best?
When there are important characteristics that can be easily measured in a reliable and valid way
T or F: Matched-pair designs can sometimes give a false sense of security in terms of controlling for individual differences.
True
What are researcher notes used for?
To keep track of anything out of the ordinary that happens during a study
What is a manipulation check?
A measure that helps determine whether the manipulation effectively changed or varied the independent variable across conditions
What are the three tasks to accomplish during data analysis in a two-group design to help answer the research question?
- Make sure groups were similar on variables that may provide alternative explanations
- Make sure manipulation was effective
- Test hypothesis
In an experiment with a between-subjects design with two levels of the IV and a continuous DV, what is the appropriate statistical analysis to conduct?
A t-test for independent means (aka an independent samples t-test)
Which statistical test is used to compare groups’ means to see if they differ to a degree that could not have happened accidentally or by chance?
A t-test for independent means (aka an independent samples t-test)
How are degrees of freedom calculated in an independent samples t-test?
df= (n1-1)(n2-1)
In an independent samples t-test, what does p represent?
The probability that the results were due to chance (statistical significance)
In an independent samples t-test, what does it typically mean when p<0.05?
There can be confidence that the results represent a real difference between groups
In an independent samples t-test, what does it typically mean when p>0.05?
There is not enough evidence to support that the results were not merely due to chance
In an independent samples t-test, as n increases, does p increase or decrease?
Decreases
T or F: The statistical significance in a t-test is independent of sample size.
False
In an independent samples t-test, what does d represent?
Effect size
What is the statistical measure of the magnitude of the difference between groups?
Effect size (d)
T or F: The effect size is independent of sample size.
True
In an independent samples t-test, are significance level and effect size independent of one another?
Yes
Is Cohen’s effect size of 0.2 considered small, medium, or large?
Small
Is a Cohen’s effect size of 0.5 considered small, medium, or large?
Medium
Is a Cohen’s effect size of 0.8 considered small, medium, or large?
Large
What is the null hypothesis?
The hypothesis that there will be no difference between groups or conditions
Which hypothesis is the researcher typically trying to statistically reject?
The null hypothesis
