Qnt 565 Case Study Review: Consumer Direct

Case Study Review
Deon Edwards
QNT/565
February 9, 2015
Professor John Halstead
Case Study Review
Explain the various measurements collected in Consumer Direct
Consumer Direct is a metric developed between Yahoo! And ACNielsen designed to test for exposure to Internet banner ads. To determine the effectiveness of this particular advertising, a control group and test group were established with the control group consisting of households that did not see the tracked advertising, and the test group consisting of households exposed to tracked ads of consumer packaged goods when visiting Yahoo! Multiple measurements were collected including household size and previous purchase behavior as well as pre-ad exposure, during exposure, and post-exposure
After a year, the eight week ad exposure period began, and purchases were tracked. Data tracking continued for another six weeks following ad exposure. The ads were considered to increase sales under two conditions: (1) purchases increased after ad exposure compared to data collecting during the 52 week pre-ad data collection, or (2) purchases increased after exposure compared to the control group’s purchases. In addition to the collected measurements, an analysis of covariance (ANCOVA) was used to compare the ad-exposed group to the control group. ANCOVA is simply a multiple regression analysis with at least one quantitative and one categorical variable. In the Consumer Direct study, the control group is considered the quantitative variable and the test group is the categorical variable also known as the primary interest of the study. A matched subjects design was also used as it matches each subject in the control group with an equivalent in the test group effectively matching households based on size and previous purchase history. More than one variable can be matched which in this study includes both household size and prior purchase behavior. Matched subjects design decreases the possibility of differences between individuals that can skew results (Shuttleworth,
The multivariate analysis (MANOVA) would be appropriate because it provides multiple levels of analysis, is typical for consumer and market research, and one of its primary purposes is to predict (Cooper and Schindler, 2011). An important goal in most research situations is to be able to predict outcomes based on prior information, and a multivariate analysis would assist in achieving this goal. Through the analysis, the researcher could predict the chance of a consumer making a purchase after seeing an Internet ad. A prediction could also be made for the increase in sales a company can expect after the consumer has been exposed to the company’s brand of products or services. This will aid companies in tracking inventory more efficiently, ultimately increasing the company’s bottom