Type 1 / Type 2 Errors

Question 1

● What is a Type I error?

Answer 1

When the researcher wrongly accepts the alternative hypothesis and rejects the null hypothesis

Question 2

● What is a Type II error?

Answer 2

When the researcher wrongly accepts the null hypothesis and rejects the alternative hypothesis

Question 3

● Why do psychologists use a 5% significance level?

Answer 3

To balance the risk of making Type I and II errors

Question 4

● What does p < 0.05 indicate?

Answer 4

Less than a 5% probability results are due to chance

Question 5

● What is meant by a lenient p value?

Answer 5

A higher probability value, e.g. p<0.10

Question 6

● What is meant by a stringent p value?

Answer 6

A lower probability value, e.g. p<0.01

Question 7

● How do you check for a Type I error?

Answer 7

Compare calculated value to a more stringent p value

Question 8

● How do you check for a Type II error?

Answer 8

Compare calculated value to a more lenient p value

Question 9

● What is another name for a Type I error?

Answer 9

False positive

Question 10

● What is another name for a Type II error?

Answer 10

False negative

Question 11

● What is the probability that something is due to chance with p<0.10?

Answer 11

Less than 10%

Question 12

● What is the probability that something is due to chance with p<0.01?

Answer 12

Less than 1%

Question 13

● What should you do if results are still significant at p<0.01?

Answer 13

More than 99% sure results are significant

Question 14

● What happens if results are not significant at p<0.01?

Answer 14

There is a risk of a Type I error

Question 15

● What happens if results are significant at a more lenient p value?

Answer 15

There is a risk of a Type II error

Question 16

▲ Why is it important to use the critical value table for checking errors?

Answer 16

To see if significance changes with different p values

Question 17

▲ How does using a more stringent p value affect confidence?

Answer 17

Increases confidence if results are still significant and not due to chance

Question 18

▲ What is the risk if the results are not significant at a stringent level?

Answer 18

Possible Type I error

Question 19

▲ What is the risk if the results are significant at a more lenient level?

Answer 19

Possible Type II error

Question 20

▲ What should you state when writing up an error check?

Answer 20

Which error you checked for and your level of confidence

Question 21

▲ How does the choice of p value affect the risk of errors?

Answer 21

Lenient p increases Type I error, stringent p increases Type II error

Question 22

▲ When do you have more than 99% confidence in results?

Answer 22

When results are still significant at p<0.01

Question 23

▲ Why might a drugs trial use a p value of 0.01?

Answer 23

Because people’s lives are at risk

Question 24

▲ Why does p<0.05 strike a balance?

Answer 24

Balances risks of Type I and II errors

Question 25

✪ Discuss what a Type I error.

Answer 25

A Type I error is when the researcher has used a lenient P value. The researcher thinks the results are significant when they are actually due to chance/error. So they wrongly accept ALTERNATIVE/EXPERIMENTAL hypothesis and wrongly reject the null.

Question 26

✪ Discuss what a Type II error is.

Answer 26

A Type II error is when the researcher has used a stringent p value. They think that their results are not significant (due to chance/error) when they could be significant. So they wrongly accept the NULL hypothesis and wrongly reject the alternate/experimental.

Question 27

✪ Explain the difference between Type I and II errors.

Answer 27

Type I is where a lenient P value is used whereas a Type II error is where a stringent P value is used.

Question 28

✪ Explain how to check for a Type I error.

Answer 28

To check for a type 1 error: you need to compare the calculated/observed value to a critical value from a more stringent p value. If the results are still significant then it is NOT a type 1 error. If the results are now not significant, then there is a chance of a type 1 error.

Question 29

✪ Explain how to check for a Type II error.

Answer 29

To check for a type 2 error: you need to compare the calculated/observed value to a critical value from a more lenient p value (p0.05). If the results are still not significant then there is not a type 2 error. If the results are NOW significant at the more lenient level then there is a risk that a type 2 error has been made.