ch. 9_102109

# ch. 9_102109 - Lab 227 Power Statistical decision table...

This preview shows pages 1–3. Sign up to view the full content.

Lab 227 10/21/09 Power Statistical decision table Actual Situation Decision Reject null Retain null Null is true – no effect Type I error (alpha) Correct decision (1-alpha) Null is false – effect exists Correct decision (power) Type II error Power is the ability to reject the null hypothesis when it is in fact false. It is also understood as the ability to detect a statistically significant effect when one exists. What affects power of a statistical test? 1. Sample size Increasing sample size will decrease the standard error, which will make the Z or t statistic larger – making it more likely that we can reject the hull hypothesis. 2. Effect size Effect size increase (numerator of Z or t statistic) will make it a correct rejection more likely. 3. Alpha level Setting a less stringent alpha will make it more likely that we would reject the null hypothesis. Alpha level can be controlled by performing one or two-tailed tests. A one- tailed test will have more power to detect an effect if one exists. To calculate power, we need to know all these components Example 1 µ = 50 σ = 10 n = 25 Expected effect of 4 points (treatment will add 4 points to each individual score) What is the probability of detecting an effect in this sample, if one exists? 1. Draw a sketch 2. Compute the standard error 3. Compute the critical Z value in the null distribution

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
4. Convert the Z value to a raw score. 5. Compute a Z value for the raw score using the alternative distribution 6. Look up the new Z value in the unit normal table to determine the probability of the shaded area corresponding to the power. σ
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

### Page1 / 5

ch. 9_102109 - Lab 227 Power Statistical decision table...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online