type 1 and type 2 errors

Question 1

What is meant by a type 1 error? (2m)

Answer 1

• researcher has used a lenient p value
• think the results are significant
• actually due to chance or error
• wrongly accept alternative hypothesis
• wrongly reject null hypothesis

Question 2

What is meant by a type 2 error? (2m)

Answer 2

• researcher has used a stringent p value
• think the results are not significant
• when they could be significant
• wrongly accept null hypothesis
• wrongly reject alternative hypothesis

Question 3

What is the difference between type 1 and 2 errors? (2-4m)

Answer 3

• in a type 1 error the null hypothesis is rejected when it is true
• whereas
• in a type 2 error the null hypothesis is accepted when it is false

Question 4

Example of a lenient p value…

Answer 4

• p < 0.1
• less than 10% probability results due to chance/error
• more than 90% probability results due to IV affecting DV/relationship co-variables

Question 5

Example of a stringent p value…

Answer 5

• p < 0.01
• less than 1% probability results due to chance/error
• more than 99% probability results due to IV affecting DV/relationship co-variables

Question 6

Example of a stringent p value…

Answer 6

• p < 0.01
• less than 1% probability results due to chance/error
• more than 99% probability results due to IV affecting DV/relationship co-variables

Question 7

Why do psychologist use the 5% significance level?

Answer 7

• p < 0.05
• is a conventional significance level
• strikes a balance between the risk of making a type 1 and 2 error

Question 8

How to check for a type 1 error…

Answer 8

• compare calculated value
• to critical value
• from a more stringent p value
• if results still significant NOT type 1
• if results now not significant IT IS type 1

Question 9

How to check for a type 2 error….

Answer 9

• compare calculated value
• to critical value
• from a more lenient p value
• if results still not significant NOT type 2
• if results now significant IT IS type 2

Question 10

Writing frame:

Answer 10

The researcher DID/DID NOT make a TYPE1/TYPE 2 error because when using a more STRINGENT/LENIENT p value of … for a ONE-TAILED/TWO-TAILED test where N=… the calculated value of … is GREATER/LESS than the critical value … and is STILL/NOW SIGNIFICANT/NOT SIGNIFICANT.

Therefore the researcher DID/DID NOT wrongfully ACCEPT/REJECT the alternative hypothesis and DID/DID NOT wrongfully ACCEPT/REJECT the null hypothesis.

They can be more than …% sure there is a significant difference/correlation between … and less than …% sure their results are due to chance/error.