-
QUESTION
Exercise 7
I only need a paragraph or less per questions and there are three questions with question 2 having an A and B part. I do not need an executive summary or a conclusion. It's just answering questions.
Please See attached instructions, questions, and data.Questions
Open the SPSS dataset “online service subscription.” The data represent about 3600 responses to an offer from an online retailer to join a discount club sent to previous customers. The dependent variable are customer responses to the solicitation. Look through the data and familiarize yourself with the variables and how they’re measured. The information should be self-explanatory.
Use SPSS to perform a logistic regression on the data. Use “response_subscrip” as dependent variable and the remaining variables as independents. Note any categorical (i.e., nominal) data and treat appropriately. Remove nonsignificant independent variables until you arrive at a final model.
To answer the following questions, refer to specific parts of your output that inform your responses.
1) Evaluate the statistical performance of the final model. How well did it perform according to the various metrics?
2) Look at the odds ratios reported in your results.
- A) How much greater are the odds of accepting the subscription service offer for a person earning $50,000 per year compared to a person earning $30,000 per year?
- B) Explain the odds ratio for the number of online purchases made in the previous month. What does an odds ratio of less than 1.0 mean? (Hint: look at the sign of the regression coefficient.)
3) Suppose the online subscription service offered by the retailer asked consumers to pay a monthly fee, and in exchange they received discounts on purchases that increased as the consumers purchased more from the retailer. Now look at the results of the individual hypotheses tests of the independent variables. Based on the albeit limited information provided about the subscription service, speculate about what the data tell a marketing manager about the offer. What target markets might be more attractive to the company? Do the data suggest to you any features for the service?
Removed Daily Hours
| Subject | Nursing | Pages | 7 | Style | APA |
|---|
Answer
Exercise 7
Model Performance
The model correctly predicts 71.9% of the dependent variable which is a good performance. Further, it is notable that those who respond negatively to online service subscription offers were more accurately predicted than their counterparts who would agree to the offers.
Odds Ratio Comparison
- Between Household Incomes
Odds of household income per thousand is 1.007.
The difference in odds is (50-30) × 1.007 = 20.14
A person earning $50,000 per year has a 20.14 odds of accepting the subscription service offer than a person earning $30,000 per year.
- Number of Online Purchases in the Previous month
Negative regression Coefficients mean that the odds ratio is less than 1, basically meaning that the odds ratio reduces by the factor when variable units rise. In this case, a unit increase in online purchases in the previous month lowers the odds of responding positively to online service subscription offers by 0.914.
Target market
The information given from the data is that men have higher odds of subscription than women, mode educated people are more likely to subscribe than lesser educated ones, older people have higher subscription odds, as well as higher earners. However, online purchases in the previous month lower the subscription odds. Therefore, the ideal target market is of the more educated older men with high incomes who did not shop in the previous month.