This paper provides a detailed critique of an advertising pretest conducted as part of a youth drug prevention campaign in Boston. The analysis evaluates the pretest's design components—including control ads, experimental groups, testing instruments, and format—before drawing conclusions from the results and offering recommendations for the de-marketing campaign. The paper identifies significant methodological limitations, particularly regarding sample representativeness and the appropriateness of testing curtailment-oriented ads on non-drug-using youth. Key findings suggest that while preference data favors the research group's ads, the results largely confirm previously known insights and fail to definitively establish superiority over existing campaigns. Recommendations focus on expanding sample recruitment, targeting actual drug users for curtailment messaging, diversifying media channels beyond television, and implementing post-launch monitoring systems.
After conducting a research plan to control drug use among young people in the target area, the field study group designed an action plan by implementing their findings. Before presenting their results to the mayor, they pretested their advertising campaign on the target audience of 10- to 18-year-olds in the same area where they conducted their previous research.
The group divided their advertising into two categories: prevention-oriented and curtailment-oriented. For the prevention category, they presented two storyboards of their own design and one existing advertisement to 25 youngsters divided into five groups. For the curtailment category, they presented the same format to 23 people divided into three groups. This analysis evaluates each component of the pretest.
Control ads serve as an effective comparison tool between existing advertising and new propositions, helping assess whether the research produced findings that enable more efficient communication with young people. For the prevention side, the group selected the advertisement "Straight" because it used many of the themes they wanted to employ in their new campaign. However, this choice is problematic. Rather than selecting an ad that resembled their own work—which amounts to testing a variation of their designs—they should have chosen an advertisement using a completely different approach. This would have properly assessed the effectiveness of their new concepts.
For the curtailment side, the group chose an advertisement featuring a celebrity of the moment. This is a stronger choice because it represents a different approach and confirms findings from the previous research, where youth reported that celebrity endorsements conveyed a "fake" feeling. However, this ad targets prevention rather than curtailment, raising questions about its placement. It should perhaps have been tested in both categories to fully evaluate celebrity effectiveness across campaign types.
Testing curtailment and prevention-oriented ads separately is reasonable, as they address different goals and use different components. However, the sample size is very small and likely unrepresentative of the entire population. Since the protocol relied on questionnaires rather than expensive production costs, expanding the sample would not have been prohibitively expensive—only an additional $10 per respondent incentive. More critically, the paper faces the same sampling problems identified in the previous case: only community school students were surveyed, with no representation of out-of-school youth more likely to use drugs.
Compounding this issue, the sample composition remained identical for both categories. This raises a fundamental validity question: Is it appropriate to assess curtailment-oriented anti-drug advertisements to youth who do not use drugs? Evaluating such ads on non-users resembles testing alcohol-and-driving awareness campaigns on people who neither drink nor drive. For the curtailment section, the research group should have specifically targeted drug users to meaningfully evaluate message effectiveness and perceived solutions.
The testing instruments were chosen wisely to maintain reasonable costs. Storyboards, selected over full advertising clips, cost considerably less without meaningfully sacrificing differential effects. The questionnaire design proved effective and straightforward, focusing on themes revealed by previous research: reality, likability, overall impression, anti-drug message, and perceived effectiveness.
The testing format incorporated several bias-reduction measures. A consistent narrator across all interviews enabled result comparison and prevented narrator-induced bias. Storyboards were shown in varying orders to prevent any single advertisement from disproportionately influencing responses. The questionnaire and protocol remained simple, short, and efficient. The advertising recall process between commercial presentations proved particularly valuable, revealing whether messages were striking enough and which elements remained salient for potential revision. Open discussions, though brief, effectively captured immediate thoughts and criticism.
The pretest results present data on likability, effectiveness, reality, anti-drug messaging, and overall impressions of youth regarding the storyboards. The "Basketball" and "Grave" storyboards—designed by the research group itself—emerged as preferred and appear most effective. This suggests the sample is most receptive to realistic ads with messages directed at youth themselves (rather than parents) and ads that emphasize family connections.
The results confirm that young people reject celebrity endorsements for anti-drug campaigns and respond negatively to unrealistic depictions. However, these conclusions are not new. All these findings were already established in the previous research phase, and the pretest adds no novel insights to inform campaign development.
While these results may hold value for prevention campaigns, their reliability for curtailment messaging is highly questionable. The sample's drug use status remains unknown; researchers only established that respondents are likely to be exposed to drugs. This methodological flaw makes the results fundamentally unreliable for curtailment campaign decisions. Assessing curtailment messages on non-drug-using youth produces conclusions that cannot responsibly guide policy implementation.
A central research aim was demonstrating that the group's advertising concepts reach young people more effectively than current campaigns. However, the pretest design fails to establish this claim. The control advertisement "Straight" employs the same conceptual approach as the group's findings, meaning any performance difference could reflect inferior execution rather than superior strategy. Furthermore, only one control ad was tested. Results might differ substantially with alternative comparison advertisements, limiting the strength of any conclusions about relative effectiveness.
The only solid conclusion the pretest supports is the ineffectiveness of celebrity endorsements—a finding already established through prior research. Beyond this, no reliable conclusions can be drawn about the campaign's likely success.
Recommendations for the de-marketing campaign center on two core objectives: employing a more representative sample and targeting research to populations genuinely affected by the issue.
"Proposed improvements to sampling, targeting, media strategy, and measurement"
You’re 68% through this paper. Sign up to read the remaining 1 section.
Sign Up Now — Instant Access Already a member? Log inAlways verify citation format against your institution’s current style guide requirements.