Paired Samples T-Test And Type 1 / 2 Errors - Week 7

Question 1

When is the paired-samples T-Test used?

Answer 1

To compare the means of two measurements taken from the same individual / object / related mean
(Eg. Comparing exam before revision and after revision)

Question 2

Advantages of a within-subjects design (repeated measures)

Answer 2

More power in study
Changes in significance levels and confidence intervals

Question 3

How can you estimate significance from confidence intervals

Answer 3

CI don’t include 0 = statistically significant
CI contains a 0 = statistically non-significant

Question 4

Define Type I error

Answer 4

False positive
Reject null hypothesis when null hypothesis is true

Question 5

Define Type II Error

Answer 5

False Negative
Accept null hypothesis when it should be rejected

Question 6

Explain Type I Error Frequency

Answer 6

Rate is independent to sample size - always same regardless of population

Increased alpha level = increased type I errors