A survey of online consumers asked respondents to indicate how much money they spent for online purchases during the most recent Christmas holiday season. These data were analyzed to see if gender (1= Female; 2 = Male) and income level (1 = Low, 2 = Middle, 3 = High) influence the amount consumers spend online for holiday shopping. The two-way ANOVA results appear below. a. Is this an experiment or observational study?
b. Are the conditions for two-way ANOVA met?
c. If so, perform the hypothesis tests and state your conclusions in terms of amount spent online, gender and income level.
d. Is it appropriate to interpret the main effects in this case separately? Explain.

Question

A survey of online consumers asked respondents to indicate how much money they spent for online purchases during the most recent Christmas holiday season. These data were analyzed to see if gender (1= Female; 2 = Male) and income level (1 = Low, 2 = Middle, 3 = High) influence the amount consumers spend online for holiday shopping. The two-way ANOVA results appear below.   a. Is this an experiment or observational study?
b. Are the conditions for two-way ANOVA met?
c. If so, perform the hypothesis tests and state your conclusions in terms of amount spent online, gender and income level.
d. Is it appropriate to interpret the main effects in this case separately? Explain.

Quizplus · Accepted Answer

The Answer is a. Observational study.
b. Boxplots of online expenditures ($) during the last holiday shopping season by both income level and gender appear to satisfy the equal variance assumption. Also, the boxplots appear fairly symmetric. We can only assume that the survey participants were selected at random.
c. All P-values are very small. This indicates a significant interaction effect as well as significant main effects.
d. Since the interaction effect of gender and income level on online expenditure is significant, it is not appropriate to interpret the main effects due to gender and income level separately.

A Survey of Online Consumers Asked Respondents to Indicate How