Chapter 19: Significance Tests for Proportions

Question 1

Significance Test

Answer 1

Used to test a claim about the value of a population parameter and to decide whether the evidence supporting a claim is likely or unlikely to happen by chance alone

Question 2

Null Hypothesis (H0)

Answer 2

Default belief about parameter’s value; assumed to be true until evidence convinces us otherwise; often a claim of no difference or no change

Question 3

Alternative Hypothesis (Ha)

Answer 3

Claim that we hope to support with evidence from collected data

Question 4

1-PropZTest Conditions

Answer 4

1. Random sample
2. n≤10% of pop.
3. np0≥10 & n(1-p0)≥10; p0=proportion in null hypothesis

Question 5

1-PropZTest On Calculator

Answer 5

p0: null proportion
x: # of successes
n: sample size
prop[≠p0; <p0; >p0]

Question 6

Standardized Test Statistic (z or t)

Answer 6

standardized test statistic = (sample statistic - null value of the parameter)/standard deviation of the statistic

Question 7

p-value

Answer 7

Measures how likely it is to get evidence for Ha as strong or stronger (that differs for ≠) than the observed evidence by chance alone when H0 is true

Question 8

p-value Interpretation (1-Prop Z)

Answer 8

“Assuming [H0] is true, there is a [p-value] probability of getting a p̂ of [p̂] or stronger/weaker by chance alone.”

Question 9

Conclusions Of Significance Tests

Answer 9

Because the p-value of [p-value] is [≤ or >] [significance level], we [reject if ≤ or fail to reject if >] H0. There is [not if >] convincing statistical evidence that [Ha in context].”