Second Test Flashcards
One factor
One Level
Single Sample T-Test
One Factor
Two Levels
Between Subjects
Independent Samples T-Test
One Factor
Two Levels
Within-subjects
Paired Samples T-Test
One Factor
Three + Levels
Between Subjects
Between subjects one way ANOVA
One Factor
Three + Levels
Within Subjects
Repeated Measures One Way ANOVA
Two + Factors
All Between Subjects
Between Subjects Factorial ANOVA
Two + Factors
All Within Subjects
Repeated Measures Factorial ANOVA
Two + Factors
Mixed (Some Between, Some Within)
Mixed Measures Factorial ANOVA
Conditional Hypothesis
Propositional hypotheses that assert a state of being of something or someone else
Relational Hypothesis
Specifies the relationship between two or more variables, as one changes so does the other
Causal Hypothesis
Not only are the variables related, but one or more of them is responsible for causing the other
Evidence for causality
X and Y occur together
X precedes Y
Steps to critically evaluate research
- Why was the research done? What are the researchers trying to test?
- What is the link between the theory (or applied problem) and the research hypotheses? Are the researchers testing what they think they are testing?
- Is the literature review adequate?
Main things a good intro does
Places your research within a larger framework and Offers a clear thesis as well as your response to this thesis
Relationships between Data, Hypothesis and Theory
- Data (dis)confirms hypothesis
- Hypothesis predicts data
- Hypothesis supports/refutes theory
- Theory generates hypothesis
Questions to ask of the independent variable
Within subjects or between subjects? Is comparison possible? Each independent variable is a factor.
Advantages and Disadvantages and Solution: Within subjects
A. Experimental control A. Guaranteed equivalence A. Fewer subjects D. Carryover effect S. Counterbalancing
Advantages and Disadvantages and Solution: Between subjects
A. Experimental control A. No testing treatment interaction A. No carry over effects D. More subjects D. Increased variance s. Random Assignment
Common threats to Internal Validity
History: Events in the environment not controlled
Maturation: psychological or physical changes
Testing Effects: Changes from being measured previously
Instrumentation: Unwanted changes in procedure
Regression: extreme scores move closer to the mean
Selection: Non random assignment
Attrition: Loss of subjects
Assumptions of the T-test
- The sample was selected randomly
- The groups are independent
- Variables are continuous and interval or ratio level
- The dependent variable is normally distributed
- Equal variance between the groups
What does the Single Sample T-test do?
Is the mean of a sample different from a specific known value?
What does the Paired Samples T-test do? Assume?
- Compares the means of two variables
- Calculates case by case difference scores
- Tests to see whether average difference is greater than zero
A. Both variables are normally distributed
What does the Independent Samples T-test do? Assume?
- Compares the means of two groups on one variables
- Estimated standard error is based on the pooled variance between the groups
A. Variable is normally distributed (V okay)
Groups are independent (V not okay)
A. Groups have approximately equal variance
What does Levene’s test do?
T-test. If it is not significant then the variances are not significantly different (good). If it is significant then go to the next row for significance and degrees of freedom.
What is ANOVA
Hypothesis-testing procedure used to evaluate mean differences between two or more treatments/populations.
Meaning of H0 and H1 in ANOVA
H0 = No differences between populations, the observed differences are due to chance. H1 = The populations do have different means, the different means are part of the reason the samples have different means.
The F-statistic conceptually
Variance between sample means divided by variance expected by chance. Treatment effect.
Total variance conceptually
How much variance is from treatment/individual differences vs random chance
Assumptions of ANOVA
- The sample was selected randomly
- The groups are independent (Exception repeated measures/mixed ANOVA)
- Variables are continuous and interval or ratio level (otherwise, use non-parametric)
- The dependent variable is normally distributed (robust if n is large enough
- Equal variance between the groups (df correction)
Post-hoc tests
Additional hypothesis tests done after an ANOVA to determine which means are significantly different from one another. Compares all conditions, controls for you doing multiple comparisons. Only done when F-ratio is significant. Can only use Tukey if you have even ns. If cell sizes are unequal then use Scheffe’ test.
Mauchly’s test
Repeated measures ANOVA. If Mauchly’s test is signifcant that means that your variances are unequal and the assumptions of the test have been violated. If it is significant use Greenhouse-Geisser correction.
Interaction
When the effect of one factor depends on the different levels of a second factor, there is an interaction between the factors. Represented by non-parallel lines that cross or converge
Statistical analysis of interactions
H0 = there is no interaction between Factors A and B. All mean differences are explained by the main effects of the two factors
There is an interaction between Factors A and B. The mean differences are not what would be predicted by the overall main effects of the two factors
