Hypothesis testing

Question 1

Null hypothesis

Answer 1

A general statement or default position that there is no relationship between two measured phenomena or no association among groups

Question 2

Type 1 error

Answer 2

When the null hypothesis is rejected but it is actually true

Also known as a false positive or alpha error

Question 3

Type 1 and Type 2 error chart

Answer 3

Question 4

Type 2 error

Answer 4

When a null hypothesis is NOT rejected when it is NOT true

Also known as a false negative or beta error

Question 5

p-value

Answer 5

The probability of obtaining the study results by chance

Also known as the type 1 error rate or alpha level

Question 6

Alpha level

Answer 6

The predetermined level of statistical significance

A standard alpha level is < 0.05

Question 7

When do we reject the null hypothesis?

Answer 7

When the p-value is < the alpha level

Question 8

Does statistically significant = clinically significant?

Answer 8

No, because:

1.) It may not have relevance (Ex. a medication that lowers blood pressure by 1 point)

or

2.) There may be other study design flaws (Ex. bias, confounding, etc.)

Question 9

Power

Answer 9

The probability of a hypothesis test of finding an effect if there is an effect to be found

Question 10

Effect size

Answer 10

The size of the association that we would like to be able to detect in the sample

Question 11

What factors increase power?

Answer 11

• A larger sample size
• A larger effect size
• A larger significance level (alpha)
• A smaller standard deviation

Question 12

Alternative hypothesis

Answer 12

Statement that there is a relationship between two measured phenomena or association among groups

Rejection of the null hypothesis