Quantitative Methods

Activity No. 2

THE CIGARETTE ADVERTISEMENT CASE

In an article in the Journal of Marketing, Mazis, Ringold, Perry, and Denman discuss the
perceived ages of models in cigarette advertisements. To quote the authors:

Most relevant to our study is the Cigarette Advertiser’s Code, initiated by the tobacco industry in
1964. The code contains nine advertising principles related to young people, including the
following provision (Advertising Age 1964): “Natural persons depicted as smokers in cigarette
advertising shall be at least 25 years of age and shall not be dressed or otherwise made to appear
to be less than 25 years of age.”

Tobacco industry representatives have steadfastly maintained that code provisions are still
being observed. A 1988 Tobacco Institute publication, “Three Decades of Initiatives by a
Responsible Cigarette Industry,” refers to the industry code as prohibiting advertising and
promotion “directed at young people” and as “requiring that models in advertising must be, and
must appear to be, at least 25 years old.” John R. Nelson, Vice President of Corporate Affairs for
Philip Morris, wrote, “We employ only adult models in our advertising who not only are but look
over 25.” However, industry critics have charged that current cigarette advertising campaigns
use unusually young-looking models, thereby violating the voluntary industry code. Suppose
that a sample of 50 people is randomly selected at a shopping mall. Each person in the sample is
shown a typical cigarette advertisement and is asked to estimate the age of the model in the ad.
The 50 perceived age estimates so obtained are given below. Use these data to do the following
activity.

26 30 23 27 27 32 28 19 25 29
31 28 24 26 29 27 28 17 28 21
30 28 25 31 22 29 18 27 29 23
28 26 24 30 27 25 26 28 20 24
29 32 27 17 30 27 21 29 26 28

1. Consider constructing a frequency distribution and histogram for the perceived age
estimates.
a. Develop a frequency distribution, a relative frequency distribution, and a percent
frequency distribution for the perceived age estimates using eight classes each of
length 2. (Note that, while the procedure presented in our previous class would tell us
to use six classes, in this case we get a more informative frequency distribution by using
eight classes.)
b. Draw a frequency histogram for the perceived age estimates.
c. Describe the shape of the distribution of perceived age estimates.

2. Construct a frequency polygon of the perceived age estimates. Use the classes of No. 1.
3. Construct a dot plot of the perceived age estimates and describe the shape of the
distribution. What percentage of the perceived ages are below the industry’s code
provision of 25 years old? Do you think that this percentage is too high?

4. Using the frequency distribution you developed in No. 1, develop:

a. A cumulative frequency distribution.
b. A cumulative relative frequency distribution.
c. A cumulative percent frequency distribution.
d. A frequency ogive of the perceived age estimates.
e. How many perceived age estimates are 28 or less?
f. What percentage of perceived age estimates are 22 or less?