L9 - Probability & Significance: Type 1 and 2 Errors

Question 1

Define ‘Level of Significance’

Answer 1

• The level at which the decision is made to reject the null hypothesis in favour of the experimental hypothesis.
• States how sure we are that the IV has an effect on the DV, not due to chance

Question 2

Define Chance

Answer 2

Something has no real cause you can identify, just happens.

Question 3

What are significance levels?

Answer 3

• If a real difference exists in our results.

- How certain are we that there is a real difference

Question 4

What is probability?

Answer 4

• Numerical measure that determines whether results are due to chance or if there is a real difference
• Real difference = significant results = reject null hypothesis

Question 5

What is the conventional level of significance?

Answer 5

p<0.05

Question 6

Why is p<0.05 used?

Answer 6

• Not too strict or too lenient

- Minimises Type 1 or 2 errors

Question 7

What does p<0.05 mean?

Answer 7

p = probability 
0.05 = 5% chance of results being a fluke but 95% certainty of significant data

Question 8

Why is a stricter significance level used? E.g 1%

Answer 8

• Findings are critical and important e.g drugs on humans

- Results should not be due to fluke

Question 9

What is a Type 1 Error?

Answer 9

• Reject null hypothesis
• Accept experimental hypothesis
• But actually, results are not significant and are due to chance

Question 10

What is a Type 2 Error?

Answer 10

• Accept null hypothesis
• Reject experimental hypothesis
• But actually, the results are significant + NOT due to chance

Question 11

Why might type errors be made?

Answer 11

Human error
Confounding variables
(use a stricter significance level)