Maria Espinosa is a marketing analyst at Kimberly-Clark, a multinational personal care corporation. She is conducting research to better understand consumer behavior for new parents and the purchase decisions they make about products for their newborns. A sample of new parents received direct mail (a "mailer") from Kimberly-Clark, and then Maria's team conducted a survey of each household a week later. Maria wants to explore the relationship between whether a respondent opened the mailer and other variables. The relevant survey questions are:

"Did you open a mailer from Kimberly-Clark" (1=Yes, 0=No)

"How likely are you to use a coupon on your next purchase for diapers?" (a 1-5 scale, with 1= definitely would use a coupon, and 5=definitely would not use a coupon)

"Is this your first baby?" 0 for no and 1 for yes.

a)One hypothesis Maria wants to test is whether there is a relationship between noticing the mailer and use of coupons. Maria's co-worker, David, suggested that they run a linear regression with "notice mailer" as the dependent variable and "how likely are you to use a coupon" as an independent variable. Would you encourage this type of analysis or discourage it? Why?

b)A second hypothesis they want to test is the relationship between whether or not the respondent opened a mailer and compared to whether or not it was the first baby that the parents had. Maria conducted a bivariate chi-square test, and the SPSS results are shown below. State the hypothesis being tested below (H0 and Ha). Analyze the results of the test, using all 3 tables below. Be detailed in your answer, and include specific numbers from the results. If there is an association, describe the strength and the nature of the association.

First Baby? * Mailer Opened Crosstabulation Count Mailer Opened Yes No Total First Baby? Yes 170 12 182 No 110 8 118 Total 280 16 296 Chi-Square Tests Asymptotic Significance Exact Sig. Exact Sig. Value df (2-sided) (2-sided) (1-sided) Pearson Chi-Square 1.304a 253 Continuity Correction .771 .380 Likelihood Ratio 1.381 240 Fisher's Exact Test 301 192 Linear-by-Linear 1.300 .254 Association N of Valid Cases 296 a. 0 cells (0.0%) have expected count less than 5. The minimum expected count is 6.16. b. Computed only for a 2x2 table Symmetric Measures Approximate Value Significance Nominal by Nominal Phi -.066 253 Cramer's V .066 .253 N of Valid Cases 296