Week 8 Flashcards
Define a repeated-measures design and what is it sometimes called?
Sometimes called “within-subject design”, the dependent variable is measured two or more times for each individual in a single sample.
Explain how a repeated-measures design differs from an independent-measures design.
Single sample as opposed to two or more samples.
Define a matched-subjects design.
Each individual in one sample is matched on a particular variable with an individual in the other sample.
Explain how a matched-subjects design differs from a repeated-measures design.
Repeated measures designs have perfect matching, matched-subject have partial matching.
Explain how a matched-subjects design differs from an independent-measures design.
Each sample is matched on variables rather than being two or more randomly selected groups.
Describe the data that are used for the repeated-measures t statistic.
Uses D scores rather than X scores.
Describe the hypotheses for a repeated-measures t test.
H₀: μD = 0 and H₁: μD ≠ 0.
What are D scores?
The difference between X₁ and X₂ (before and after treatment).
What is MD? (D is subscript)
Mean of D scores.
What is μD?
The mean of difference scores for the whole population.
What is the t-statistic formula for a repeated measures design?
t = MD - μD/sMD
What is sMD? (M and D are subscript)
The estimated standard error of MD.
The related-samples t statistic requires two basic assumptions, what are they?
- The observations for treatment condition must be independent.
- The population distribution of difference scores (D values) must be normal (or large).
What is the estimated Cohen’s d for repeated measures designs?
d = MD/sD
How does the r² test (better known as ω²) differ for independent measures t tests and repeated measures t tests?
It doesn’t, it is always r² = t²/t² + df.
What is the confidence interval formula for repeated measures t tests?
μD = MD ± tsMD
Larger samples produce a larger value for the t statistic (farther from zero) and increases the likelihood of rejecting H₀, TRUE OR FALSE?
TRUE.
Larger variance produces a smaller value for the t statistic (closer to zero) and reduces the likelihood of rejecting H₀, TRUE OR FALSE?
TRUE.
How does variance in D scores indicate consistency of treatment?
Little variance in D scores indicates that the treatments have approximately the same effect on all subjects.
What are the advantages of repeated measures designs?
- Requires fewer subjects.
- Allows observation of change over time.
- Removes individual differences.
- Significantly less variance.
What are the disadvantages of repeated measures designs?
- Order/sequencing effects.
2. Time factors.
What is a nuisance variable?
A type of extraneous variable that increases the variability within participants in an experiment.
What are the two main types of nuisance variables?
- Participant.
2. Environment.
How do you control participant nuisance variables?
- Repeated measures designs with counter balancing.
- Matched-designs (variables can be overlooked and missed however).
- Independent group design (requires larger groups).
How do you control environmental nuisance variables?
Difficult, but one solution is to hold them constant.
How do you calculate sMD?
sMD = sD/square root of n.
