3. Null Hypothesis Significance Testing (NHST)

Question 1

Parameters represent what two different effects?

Answer 1

1. Relationships between variables
2. Differences between means

Question 2

All parameters have an associated…

Answer 2

Sampling distribution

Question 3

For any parameter we can work out the probability…

Answer 3

of getting at least the value we have if the null hypothesis is true

Question 4

What p value is typically used as a threshold for ‘significance’?

Answer 4

p < 0.05

Question 5

The p-value is…

Answer 5

The probability of getting a test statistic at least as big as the one you have observed given that the null hypothesis is true

Question 6

What is a type 1 error?

Answer 6

Rejecting the null when it is true
Believing in effects that don’t exist

Question 7

What is a type 2 error?

Answer 7

Accepting the null when its false
Not believing in effects that do exist

Question 8

What is meant by statistical power?

Answer 8

The probability of a test avoiding a Type 2 error
The probability that a test detects an effect that is, in fact, true
The probability of rejecting H0 when H1 is true

Question 9

What are four problems with Null Hypothesis Significance Testing?

Answer 9

1. Tells us nothing about importance because p depends upon sample size
2. Provides little evidence about the null (or alternative) hypothesis
3. Encourages all-or-nothing thinking
4. Based on long-run probabilities

Question 10

What are three examples of standardised effect sizes?

Answer 10

1. Cohen’s d
2. Pearson’s r
3. Standardised beta