14_final_exam

.pdf

School

Southern Methodist University *

*We aren’t endorsed by this school

Course

6474

Subject

Industrial Engineering

Date

Apr 3, 2024

Type

pdf

Pages

4

Uploaded by AmbassadorSwan4128 on coursehero.com

1 Final Exam MAST 6478 Data Analytics MAST 6474 Introduction to Data Analysis I Final Exam Instructions: There are 4 questions in this exam, each with multiple parts. Before you begin, ensure that you have all 4 questions listed in this document. You have 3 hours and 15 minutes to complete the exam. You may not discuss this exam with anyone. By printing your name on the exam, you reaffirm your pledge to uphold the SMU honor code. In addition to this document, you will use the Final Exam Workbook to complete your calculations and record your answers. As soon as you open the workbook, save it to your computer and save it frequently throughout the exam period. After reading each question, locate any relevant data in the associated tab in the Final Exam Workbook. Complete your calculations on the associated Question tab, and record your answers on the “Answer” tab. For example, for Question 2, you will read each part of the question below, review the relevant data on the Question 2 tab in the Final Exam Workbook, and complete your calculations in the same worksheet next to the data. You will then record your answer in the indicated cell(s) of the “Answer” worksheet tab. Submit the Final Exam Workbook for grading. Good luck!
2 Final Exam Question 1 (6 parts): LinkedIn quizzes are typically evaluated using measures of viewer engagement. One of the most common engagement measures is the number of LinkedIn user ‘likes.’ A random sample of 100 LinkedIn quizzes can be found in the Question 1 tab of the Final Exam Workbook. It includes the following variables: Quiz ID Total Likes – Number of LinkedIn user ‘likes’ for the quiz Total Views – Number of LinkedIn user views of the quiz Total Responses – Number of LinkedIn users who took the quiz Response Rate – Percentage of quiz viewers who took the quiz Total Right Answers – Number of quiz takers who answered the quiz correctly % Right Answers – Percentage of quiz takers who answered the quiz correctly a. Determine the sample correlation coefficient, r , between Total Likes and Total Views . Test the alternative hypothesis that Total Likes has a linear relationship to Total Views . Specifically, what are the test statistic and the p- value for that test statistic? For α = .05 , what do you conclude about the relationship between the variables? b. Estimate a model with Total Likes as the dependent variable and the other variables as independent variables (but do not include Quiz ID ). Write the estimated regression equation. c. How well does the model estimated in part (b) fit the data? Report the relevant value. Given the other independent variables in the model, which independent variables (if any) are related to the dependent variable Total Likes at the α = .05 level? d. Is the model estimated in part (b) statistically significant at the α = .05 level? Write the null and alternative hypotheses, the appropriate test statistic, the p -value for that test statistic and your conclusion, based on that p -value. e. You might suspect that more right answers for a quiz will result in more user ‘likes.’ Using the model estimated in part (b) and focusing on the Total Right Answers variable, determine whether the sample data supports this suspicion. Report the appropriate test statistic, the p - value for that test statistic, and your conclusion based on that p -value at the α = .05 level. f. Now use the backward elimination approach to select a model that is well suited for predicting Total Likes for a LinkedIn quiz. What is the selected model’s adjusted R 2 ? List the independent variables included in the selected model. Question 2 (5 parts): A/B testing is often used to compare consumer response to similar versions of a digital “call to action.” The ‘A’ and ‘B’ versions of the call to action may differ in their information presentation— layout, colors, graphics, etc. The objective of A/B testing is to determine how the differences affect consumers’ propensity to click-through to a webpage where they can then purchase. In one A/B test, an online seller offered randomly-selected customers a product discount, presented either as a percentage or in dollars (the actual discount amount was the same). The sample data are shown in a contingency table in the Final Exam Workbook with discount (percentage vs. dollars) by row and click-through (yes vs. no) by column.
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