Understanding Power in Hypothesis Testing

These flashcards cover key concepts related to power, beta, and hypothesis testing as outlined in the lecture notes.

Type II Error

Occurs when the null hypothesis is retained when it should be rejected.

Power

The probability of correctly rejecting the null hypothesis when it is false.

Effect Size

A measure of the magnitude of a treatment effect; larger effect sizes increase power.

Alpha

The threshold probability for rejecting the null hypothesis; increasing alpha can increase power but also raises the risk of Type I error.

Sample Size

Increasing the number of trials, which increases the power of the test.

Variability

Less variability in data can increase power if group means are far apart.

Relationship between Power and Beta

Power + Type II Error Rate (Beta) = 1.00, indicating that they are complementary.

Visualizing Power

Power can be visualized by understanding distribution overlap between null and alternative hypotheses.

Effect of Increasing Sample Size on Power

Larger sample sizes reduce standard error and increase power by making distributions skinnier.

Relaxing Alpha

Increasing alpha shifts the rejection area right, allowing for a greater probability of rejecting the null hypothesis.