Topic 6 - Decision Errors Flashcards
What is a Type 1 Error?
Rejecting the null hypothesis when it is true
In a Type 1 Error what does p = ?
Alpha
What does p mean for decision errors?
The probability of that error occurring
What % of the time is a T1 error made?
5%
What is a Type 2 Error?
Retaining the null hypothesis when it is false
In a Type 2 Error what does p = ?
Beta
How can a Type 2 Error arise?
When H1 = true, some of the means may fall close to the mean of Ho
When we retain the Ho and the Ho is true, what does p = ?
p = 1 - a
When we reject the Ho and the Ho is false, what does p = ?
p = 1 - B (power)
What is the relationship between Type 1 and Type 2 errors?
They are mutually exclusive
TRUE or FALSE
When you change one error type, there is no effect on the other error type
FALSE
The changes in one error type has an effect on the other error type
What needs to happen to Beta to minimise Type 2 errors?
Reduce Beta and Increase power
How do we minimise Type 2 errors?
- Increase alpha
- Increase n
- Powerful stats tests
- Have good experimental design
What is the effect on a Type 2 error when increasing Alpha?
- This increases power and decreases the probailty for T2 Error
- Also produced higher probabiltiy of a T1 error
When consequences of T1 error are severe where would be set the Alpha?
Low alpha (e.g. 0.01)
