Question

A scientist is investigating the hypothesis that taking more math classes in college directly leads to higher income later in life. He follows 100 students chosen randomly from the students at Valinor College, obtaining data on how many math courses they took and their annual income 10 years after graduation.

1. Briefly explain why it would not make sense to do a t test comparing the number of math classes taken and the income ten years later.

2. The experimenter notices after the experiment that students who took more math classes also took more units in general, compared to other students. Briefly explain why might that be a problem when analyzing the experiment.

Answer 1

A) Given, the sample size choosen for experiment is 100, it is not suitable that one uses t test for comparison. This is because of the reason that we uses t test only when sample size is less than 30 because for larger samples t distribution tends to normal distribution and it is preferable to use z test.

B)The problems while analysing the experiment:-

I) the constraints present in the experiment which in turn affects the degree of freedom of the test performed.

Ii) The independence of the samples choosen for the test.

Iii) Choosing of the level of significance i.e. alpha level for performing the test.