Week 4: Factorial Design Flashcards
Regarding a simple experiment with 1 independent variable, what’s required to make sure it’s a good design?
A:
B:
C: You must have a comparison group, and the independent variable must have 2 or more levels.
C: You must have a comparison group, and the independent variable must have 2 or more levels.
How do we test a simple experiment with only one independent variable and no more than two levels?
A: When there is only one independent variable with two levels, the appropriate statistical test to use would be ANOVA.
B: When there is only one independent variable with two levels, the appropriate statistical test to use would be a z-score.
C: When there is only one independent variable with no more than two levels, the appropriate statistical test to use would be a t-test.
C: When there is only one independent variable with no more than two levels, the appropriate statistical test to use would be a t-test.
NOTES:
> You CAN use ANOVA for a simple experiment too.
> ANOVA comes into play when there are 3 or more levels/scores.
> So if you have a simple experiment with 3 levels/scores you would use a one-way ANOVA
> “One-way, “two-way,” three-way,” refers to the number of IVs in the experiment
> So, since simple experiments only ever have 1 IV it will ALWAYS be a one-way ANOVA if the simple experiment has 3 or more levels/scores
> Otherwise, we would always use a t-test for a simple experiment because they can’t test anything more than 2 levels/scores which is most likely what we’ll be working with in this class
In a simple experiment with only one independent variable with two levels using a within-subjects or repeated measures design, where the same participants experience both levels of the independent variable, the appropriate t-test to use is a (an) _____________.
A: ANOVA
B: Dependent samples t-test
C: Independent samples t-test
B: Dependent samples t-test
For a within-subjects or repeated measures design, where the same participants experience both levels of the independent variable, the appropriate t-test to use is a dependent samples t-test, also sometimes called a paired samples t-test. This compares the differences between each participant’s scores on the two conditions.
In a simple experiment with only one independent variable with two levels using a between-subjects design, where different participants are randomly assigned to each level of the independent variable and the groups are independent of each other, the appropriate t-test to use is a(an) _________________.
A: ANOVA
B: Dependent samples t-test
C: Independent samples t-test
C: Independent samples t-test
For a between-subjects design, where different participants are randomly assigned to each level of the independent variable and the groups are independent of each other, the appropriate t-test to use is an independent samples t-test. This compares the mean scores of the two independent groups.
What’s a factorial design?
A:
B: A factorial design is an experimental design that has a minimum of 2 independent variables, with each independent variable having at least 2 levels. It allows researchers to study the effects of multiple independent variables and their interactions. Each independent variable must be combined with every level of the other independent variables to form all possible combinations.
C:
B: A factorial design is an experimental design that has a minimum of 2 independent variables, with each independent variable having at least 2 levels. It allows researchers to study the effects of multiple independent variables and their interactions. Each independent variable must be combined with every level of the other independent variables to form all possible combinations.
EXAMPLE:
Hypothesis: “Music benefits test performance:”
IV #1 - Music:
> Level 1 = music
> Level 2 = no music
IV #2 - Test type:
> Level 1 = multiple choice
> Level 2 = essay
DV: Test score out of 100 points
BENEFITS:
> It allows researchers to look at the effects of two or more independent variables rather than just one. This provides more information than studying variables individually.
> Another key benefit is that a factorial design lets researchers examine whether the independent variables influence each other, through analyzing interactions. This can reveal how the effect of one variable depends on the level of the other variable.
> Factorial designs are also more efficient because the same participants can be used to study multiple variables.
What is the purpose of “notation” in a factorial design?
A: Notation refers to the standardized way of writing out the number of independent variables and levels in a factorial study.
B:
C:
A: Notation refers to the standardized way of writing out the number of independent variables and levels in a factorial study.
MORE DETAILS:
> The Number of placeholders indicates the number of independent variables
> For example two place holders (___ x ___), means there are two independent variables
> The specific values you’ll see in here (___ x ___), indicate the number of levels for each IV
For example 2x2 means there are two IVs and that they both have two levels each.
> It doesn’t matter which variable is listed first or second - the order is arbitrary. The important thing is to be consistent when reporting results.
> Notation is used to communicate the study design. It allows readers to understand how many variables were manipulated and to what degree (number of levels).
How do you calculate the total number of conditions (or cells) in a factorial design study?
A: To calculate the total number of conditions (or cells) in a factorial design study, you multiply the numbers written in the notation
B:
C:
A: To calculate the total number of conditions (or cells) in a factorial design study, you multiply the numbers written in the notation.
EXAMPLES:
> A 2 x 2 design would have 2 times 2, which equals 4 conditions.
> This is because there are two levels for each of the two independent variables, creating all possible combinations.
> As more independent variables are added, the number of conditions grows very quickly due to multiplication, which has practical implications for the study design and resources needed.
True or false: In factorial designs, each independent variable can be manipulated either as within-subjects or between-subjects.
A: True
B: False
A: True
In factorial designs, each independent variable can be manipulated either as within-subjects or between-subjects. There are three possible combinations:
1) Both IVs within-subjects
2) Both IVs between-subjects
3) One IV within, one between (these are called Mixed-Subjects)
NOTE:
Some variables may only be possible to study as between-subjects due to practical or ethical reasons. The choice depends on the research question and the feasibility of different approaches.
True or false: It matters which IV you call #1 or #2
A: True
B: False
B: False
Your hypotheses may lead to doing it one way or another, but there are no rules. It just needs to be consistent.
What are cell means?
A:
B:
C: Cell means are the individual data points or scores within each condition of the factorial design matrix.
C: Cell means are the individual data points or scores within each condition of the factorial design matrix.
MORE DETAILS:
> So in a 2x2 design with four conditions, there would be four cell means, one for each combination of the independent variable levels.
> These cell means represent the average score or dependent variable value for participants in that particular condition cell.
> Examining the cell means is important for interpreting main effects and interactions in the data.
What are marginal means?
A:
B: Marginal means refer to averaging or aggregating the scores across one independent variable while ignoring or not differentiating based on the levels of the other independent variable(s).
C:
B: Marginal means refer to averaging or aggregating the scores across one independent variable while ignoring or not differentiating based on the levels of the other independent variable(s).
MORE DETAILS:
> For example, in a 2x2 design, the marginal mean for the first independent variable is calculated by averaging the scores in the “music” condition cells and averaging the scores in the “no music” cells, regardless of the other variable
> Marginal means provide the overall averages to examine main effects in ANOVA analysis.
What is a boundary condition?
A: Boundary conditions refer to circumstances or factors that influence or limit the effect or applicability of a relationship or condition. In research, boundary conditions are variables that may strengthen or weaken the effect of another variable. Studying boundary conditions helps determine under what specific contexts or situations an effect is more or less likely to occur. It helps establish the limits and scope of a finding or theory.
B:
C:
A: Boundary conditions refer to circumstances or factors that influence or limit the effect or applicability of a relationship or condition. In research, boundary conditions are variables that may strengthen or weaken the effect of another variable. Studying boundary conditions helps determine under what specific contexts or situations an effect is more or less likely to occur. It helps establish the limits and scope of a finding or theory.
EXAMPLE:
> Studying the effectiveness of different types of masks (cotton vs. medical) in mitigating virus transmission.
> Researchers found medical masks were generally more effective, but facial hair acted as a boundary condition - when people had facial hair, there was no difference between the mask types.
> This showed facial hair weakened the effect of mask type.
> Boundary conditions help determine if the impact of a variable (mask type) depends on other conditions (facial hair).
> Studying boundary conditions can reveal stronger or weaker effects under certain circumstances.
What is an ANOVA test, and when should we use it?
A:
B:
C: ANOVA stands for analysis of variance. It is a statistical test used to analyze the differences between group means in experimental studies. ANOVA should be used when there is 1 independent variable with 3 or more levels (simple experiment with 3 or more levels), or when there are multiple independent variables in a factorial design. ANOVA compares whether the means of the different groups are all equal or if some are significantly different from others. It allows researchers to test for main effects and interactions between independent variables.
C: ANOVA stands for analysis of variance. It is a statistical test used to analyze the differences between group means in experimental studies. ANOVA should be used when there is 1 independent variable with 3 or more levels (simple experiment with 3 or more levels), or when there are multiple independent variables in a factorial design. ANOVA compares whether the means of the different groups are all equal or if some are significantly different from others. It allows researchers to test for main effects and interactions between independent variables.
What is the difference between a t-test and ANOVA?
A:
B:
C: A t-test is used when there is one independent variable with two levels/groups (it cannot handle more than 2 levels), while ANOVA is used when there is one independent variable with more than two levels or multiple independent variables.
C: A t-test is used when there is one independent variable with two levels (groups), while ANOVA is used when there is one independent variable with more than two levels or multiple independent variables.
MORE DETAILS:
> A t-test compares the means of two groups, while ANOVA compares the means of three or more groups simultaneously.
> ANOVA can test for main effects and interactions between variables, while a t-test only looks at differences between two specific groups.
> ANOVA is more flexible and powerful than a t-test as it can analyze more complex experimental designs with multiple levels or variables.
What is a main effect?
A:
B: A main effect refers to the influence or effect of one independent variable on the dependent variable, assessed by averaging across or ignoring the levels of any other independent variables in the experiment. In ANOVA, a main effect shows whether the mean scores for one independent variable (e.g. music) differ significantly regardless of the other variables (e.g. test type). It indicates if that single variable makes a meaningful impact on the dependent variable.
C:
B: A main effect refers to the influence or effect of one independent variable on the dependent variable, assessed by averaging across or ignoring the levels of any other independent variables in the experiment.
In ANOVA, a main effect shows whether the mean scores for one independent variable (e.g. music) differ significantly regardless of the other variables (e.g. test type)
It indicates if that single variable makes a meaningful impact on the dependent variable.
For example, if the question is, “Is there an effect of music?” we should know that we’re looking for the MAIN EFFECT of IV #1, which is music (music, no music). Likewise, if the question is “Is there an effect of test type?” we should know that we’re looking for the MAIN EFFECT of IV #2, which is test type (MC, essay).
What is an interaction?
A: An interaction occurs when the effect of one independent variable on the dependent variable depends on the level of a second independent variable. In other words, the relationship between the first independent and dependent variables is different at different levels of the second independent variable. Interactions are important because they indicate that two variables influence each other jointly, rather than having just independent or additive effects.
B:
C:
A: An interaction occurs when the effect of one independent variable on the dependent variable depends on the level of a second independent variable.
In other words, the relationship between the first independent and dependent variables is different at different levels of the second independent variable.
Interactions are important because they indicate that two variables influence each other jointly, rather than having just independent or additive effects.
What is a simple effect?
A:
B: A simple effect refers to the effect of one independent variable on the dependent variable at a specific level of the other independent variable. It provides a more detailed look at the relationships between variables within a significant interaction. Examining simple effects helps determine the nature or pattern of an interaction effect.
C:
B: A simple effect refers to the effect of one independent variable on the dependent variable at a specific level of the other independent variable. It provides a more detailed look at the relationships between variables within a significant interaction. Examining simple effects helps determine the nature or pattern of an interaction effect.
EXAMPLE:
In a factorial design studying music and test type, the simple effect of music on scores for the multiple choice test condition alone could be analyzed.
A simple effect could be analyzing the effect of music (listening vs not) on scores specifically for the multiple-choice test condition.
Another simple effect would be looking at the effect of music on scores just within the essay test condition.
This breaks down the interaction to examine how one variable influences the outcome at each level of the other variable, helping to interpret what the interaction means in a more detailed way.
What is a post hoc test?
A: A post hoc test refers to additional statistical tests that are conducted after an ANOVA if a significant main effect or interaction is found. Post hoc tests are used to make specific comparisons between groups/conditions to determine which ones differ significantly. This helps interpret where the overall significant effect is coming from. Common post hoc tests mentioned include Tukey HSD and Bonferroni tests. Post hoc tests are needed when there are 3 or more groups being compared, but not for designs with only 2 groups.
B:
C:
A: A post hoc test refers to additional statistical tests that are conducted after an ANOVA if a significant main effect or interaction is found. Post hoc tests are used to make specific comparisons between groups/conditions to determine which ones differ significantly. This helps interpret where the overall significant effect is coming from. Common post hoc tests mentioned include Tukey HSD and Bonferroni tests. Post hoc tests are needed when there are 3 or more groups being compared, but not for designs with only 2 groups.
