2) See each part for the question to answer.

a) Paulette is looking to decide which of two online ad campaigns (let’s call them A and B) would generate the highest purchase conversion. Describe how she can figure out whether to go with campaign A or B. 

b) Acme Dance Company delivers a 30-week dance program to inner city students at their school once a week, culminating in a stage dance performance by students to a large audience at the end of the school year. Acme would like to measure the effectiveness of its program in impacting students’ attitude toward school. A market research firm suggests doing a pre-survey with students before the program begins, and then doing a post-survey after the program is completed to measure the changes in student attitudes towards school. Do you agree or disagree with this approach of measuring program effectiveness? Why or why not?

c) Suppose we find that phenomenon J and outcome K are highly correlated. Name two possible reasons why J may in fact not be causing changes in K.

d) Going back to the Bristol tea experiment in the class session, let us suppose Bristol is being tested on her ability to figure out if tea or milk was first added to the cup. If she was offered 6 cups to taste, with 3 containing tea poured first, and 3 containing milk poured first in the preparation, what is the probability that she would guess the right preparation for every cup, by luck alone (i.e., she has no special ability to detect the preparation)?

e) Give an example of how spurious correlation between two events can arise.