Question

Multiple Choice
The students in John's class are complaining about how hard word problems are to solve. They insist that they perform better when tested on numerical equations than the applied word problems. To test their claim, John uses the results from his latest math test to compare the number of points each student earned for the word problems to the number of points students earned for solving the numeric equations. Which t-test should she use to test for this difference?

○ Related samples t-test
○ Independent samples t-test
○ or Single sample t-test

Answer 1

As we want to see the effect of a treatment (word problems in this case), so we will control the sample elements.

We would like to compare the score of same student over the numerical problem and word problems so that there is no bias due to sample elements.

Thus a paired sample t-test or related samples t-test would be the best way to test the effect of word problems.

Because we will be using same student to record both the numerical and word problem scores to compare.

Hence, option A is correct.