Module 2: Elementary Arguments & the Role of Chance Flashcards
Researchers typically aim for a power criteria of…
.80 (80%)
Type I error
Concluding a mean difference exists in the populations (rejecting the null) where there is no difference (the null is true)
ALPHA
What are the statistical assumptions for an independent samples t-test (there are 3)
- Groups are independent of each other
- Sample means are normally distributed
- Homogeneity of variance between groups
p value > 0.05 =
Fail to reject the null hypothesis
p value < 0.05 =
Discredit the null hypothesis
The smaller the p-value the more ________ variation there is
systematic
Type II error
Concluding NO mean difference exists in the population (failing to reject the null) when there is in fact a difference in means (the null is false)
BETA
What is power’s relation to type II error?
Power is inversely related to type II error (if one increases the other decreases and vice versa)
The stricter the alpha, the lower the _______
power
Power
likelihood of a test finding an existing effect
The larger the sample size, the greater the ______
power
The larger the effect of the independent variable, the greater the _______
power
Determinants of power (there are 3)
- Alpha
- Sample size
- Effect size
What is the null hypothesis of an independent samples t-test?
The sample means of the populations are the same
What is the null hypothesis of the repeated (paired) samples t-test?
The sample means of each condition are the same