Module 8

Question 1

What are paired sample T-tests?

Answer 1

They build on single-sample-T-test’s to look at changes in the sampling unit.
They ask questions such as:
Does tutoring improve the grades of students?

Question 2

Paired sampling test are used for?

Answer 2

Looking at how sampling units change across a factor. To accomplish this, two measureing units are taken from each sampling unit–could be at different time or location.

Question 3

Case Stud Example of a Paired-Sampling Unit:

Answer 3

If we wanted to evaluate whether a particular medication (the factor) reduces blood pressure for a patient (the sampling unit), we could measure blood pressure before a patient takes the medication, and then collect a second measurement after.
OR
Location examples follow a similar idea, but involve taking measurements on two different parts of the sampling unit. For example, if you wanted to evaluate the effectiveness of using lemon juice (the factor) to prevent apples (the sampling unit) from turning brown, you could divide each apple in half and measure the degree of browning on each half after treating one half with lemon juice and the other with water. The data in these examples are considered paired because the two measurements come from the same sampling unit.

Question 4

What do paired sampling T test evaluate?

Answer 4

• Whether the factor causes a change in the two measurement value that differs from the reference value.
• The test uses the difference between the measurements to draw inference instead of the individual measurements.
• Most of the time the reference value is 0, which corresponds whether factor caused any difference.

Question 5

4 steps to defining a paired sample test–What’s step 1?

Answer 5

Define the Null and Alternative Hypothesis

• Statement about how the two measurements are different from the reference value.

Question 6

The research question of a paired sample test will ask?

Answer 6

whether the difference between the paired measurements in a sample is statisticaly different from the reference value

Question 7

If there is no directionality the null and alternative distribution are?

Answer 7

H0: d=μ and HA: d≠μ

Question 8

If there is directionality the null and alternative distributions are?

Answer 8

H0: d≤μ (or d≥μ) and HA: d>μ (or d<μ)