Assignment 1_AB Testing

.docx

School

University of South Carolina *

*We aren’t endorsed by this school

Course

352

Subject

Industrial Engineering

Date

Apr 3, 2024

Type

docx

Pages

Uploaded by gnickerson19 on coursehero.com

Assignment 1: – A/B Testing Q1: Please review the Rocket Fuel case along with our discussion in the class and answer the following questions. (25 Points) Q1a: What is the independent variable in the experiment described in the Rocket Fuel case? (7 Points) Whether users are shown the PSA advertisement (control group) or the campaign advertisement (treatment group). Q1b: What type of data ( nominal, ordinal, interval, ratio) is the independent variable? (5 Points) Q1c: What is the dependent variable in the experiment? (7 Points) Whether users purchase the handbag or not. Q1d: What type of data ( nominal, ordinal, interval, ratio) is the dependent variable? (6 Points) Q2: Was the advertising campaign effective? (i.e., did additional consumers convert because of the advertising campaign?). Answer each of the following questions below will lead you to the answer of this question. (25 Points) Q2a: Based on the NOIR classification (or data type) of the independent variable and the dependent variable, which statistical test do you think fits the situation most appropriately? (5 Points) A. Chi-Square Test B. t-Test Q2b: What is the test formula of the correct statistical test? (5 Points) X 2 =  ¿¿ Q2c: Please report the test value based on your computation. (5 Points) Column1 Did not Purchase Purchased Total Control Group (Exposure to PSA) 23,104 420 23,524 Treatment Group (Exposure to Ad) 550,154 14,423 564,577 Total 573,258 14,843 588,101

54.32 Q2d: Is the test result statistically significant? Why? (5 Points) Yes, the test result is statistically significant. This is because the test value is greater than the chi-square critical value of 3.84. Q2e: Was the advertising campaign effective? (5 Points) Yes, the advertising campaign was effective as there is significant statistical evidence to show a relationship between ad type and number of purchases. Q3: In A/B testing and other field experiments, it is important to design the right sample size. (25 Points) Q3a: If Rocket Fuel had reduced the size of the existing control group by 50%, do you see any potential issues? Yes . (3 Points) If your answer is yes, describe the potential issue(s) in no more than 30 words. (5 Points) It most likely won’t be able to detect how many purchase the handbag/ the increase in sales due to the ad. Conversion rates are usually low in comparison to other experimental settings, so a larger control group is required to notice a lift and allow for a deeper analysis. Q3b: If Rocket Fuel had increased the size of the existing control group by 50%, do you see any potential issues? No . (3 Points) If your answer is yes, describe the potential issue(s) in no more than 30 words. (5 Points) Q3c: If Rocket Fuel had run the trial campaign from November 2015 to December 2015, do you see any potential issues? Yes . (3 Points) If your answer is yes, describe the potential issue(s) in no more than 50 words. (6 Points) Reducing the campaign from 4 to 2 months would most likely result in it getting less exposure, reducing the number of people who see either ad (control or treatment). Having both a smaller control and treatment group would make it more difficult to assess the effectiveness of the ad and it would also most likely result in less sales overall for TaskaBella. Q4: Assume that TaskaBella now would like to run another A/B testing. In this A/B testing, TaskaBella would like to figure out whether they should adjust the image of their advertisement campaign. Their current image is the following. (25 points)

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help