A Beverage Distributor Wanted to Determine the Combination of Advertising

A beverage distributor wanted to determine the combination of advertising agency (two
levels)and advertising medium (three levels)that would produce the largest increase in
sales per advertising dollar. Each of the advertising agencies prepared ads as required for
each of the media-- newspaper, radio, and television. Twelve small towns of roughly the
same size were selected for the experiment, and two each were randomly assigned to
receive an advertisement prepared and transmitted by each of the six agency-medium
combinations. The dollar increases in sales per advertising dollar, based on a 1-month
sales period, are shown in the table. $\begin{array}{l}\text { Advertising Medium }\\\begin{array}{c|c|c|c} & \text { Newspaper } & \text { Radio } & \text { Television } \\\hline \text { Agency 1 } & 15.3,12.7 & 17.4,20.1 & 16.2,12.7 \\\text { Agency 2 } & 22.4,18.9 & 24.3,28.8 & 12.5,9.4\end{array}\end{array}$

The SPSS analysis is shown below.

(Note: SPSS uses ʺExplainedʺ instead of ʺTreatmentʺ in the factorial analysis. Also, SPSS
uses ʺResidualʺ instead of ʺError.ʺ)Would you test the main effects factors, agency and
medium, in this example? Explain why or why not.