Topic 6: Power Flashcards
(29 cards)
What is a Type 1 error?
Rejecting the null hypothesis when it is true
What is a Type 2 error?
Retaining the null hypothesis when it is false
What is power conceptually?
The sensitivity of an experiment to detect a real effect of the independent variable on participants’ behaviour
What is power statistically?
The probability of finding a significant difference - if the effect that you are looking for is real
What is 1 - beta?
Probability of correctly rejecting a false null hypothesis
What is 1 - alpha?
Probability of retaining the null hypothesis when it is true
What defines the region of beta?
Beginning of alternative hypothesis near the null that overlaps with alpha
What is beta?
A Type 2 error (falsely retaining the null)
What is alpha?
A Type 1 error (falsely retaining the null)
What happens if alpha shrinks - in regards to beta and to power?
Beta grows, therefore shrinking power
What happens if alpha grows - in regards to beta and to power?
Beta shrinks, and power grows
What are the four circumstances where you would test for power?
A priori: to find the desired sample size
A posteriori: if you retained the null but expected to reject it
If you retained the null and wanted to
If you rejected the null as expected
What factors influence power?
Three major, 3 minor
Effect size, meaning:
- treatment effect (mean diff)
- variability
N, or n
Alpha level, directional vs non-directional, research design
What is the problem with increasing sample size a posteriori?
It increases your chance of a Type 1 error (falsely rejecting the null)
How is power computed?
Find d, then delta, then the table
How is effect size calculated with an independent-samples design?
Cohen’s d, wherein the numerator is the difference between the two means, and the denominator is the pooled SD error (Sp)
How is the sample size needed found from power and effect size?
Make sure you have d, and then use the table to find delta from the power and the alpha
Then solve for N
What values of power are acceptable in behavioural sciences?
Excellent: 0.90
Acceptable: 0.50 - 0.70
What is the relationship between effect size and power, in relation to variance and treatment effect?
A larger treatment effect (mean difference) results in a larger power
As variability grows larger, power grows smaller
Why is related samples design more powerful than independent samples?
It reduces the variability caused by the bigger N in independent samples
Why is related samples design more powerful than independent samples?
It reduces the variability caused by the bigger N in independent samples
How does sample size influence power?
The larger the N = the smaller the variance = the larger the effect
How does directionality of a hypothesis influence power?
One-tailed test: higher power, as long as the hypothesized direction is correct
How is d found in a single samples design?
mean of the group - mean-null / SD
