This paper evaluates the research methodology and design choices used in Boston's anti-drug campaign targeting youth ages 10–18. The author examines three key dimensions: the appropriateness of focus groups versus alternative qualitative methods, the effectiveness of pre-screening questionnaires and interview protocols, and the adequacy of the stratified sampling plan. The analysis identifies both strengths—such as exploring sensitive topics in a permissive group setting—and significant limitations, including inadequate drug differentiation, age-inappropriate questioning, sampling bias toward school-enrolled youth, and exclusion of high-risk populations. The paper concludes that the research design ultimately captured views from a non-representative population already aware of drug dangers, limiting the campaign's ability to refine messaging for vulnerable youth.
While no perfect research method existed for Boston's anti-drug campaign study, focus groups emerged as a defensible choice despite inherent drawbacks. The research objective required understanding why previous anti-drug communications—including celebrity-driven commercials—had failed to resonate with youth. Because the underlying causes of this communication failure were unknown, the study called for exploratory research designed to uncover young people's ideas, insights, and feelings toward drugs and anti-drug messaging. This exploratory purpose ruled out survey-based quantitative methods, necessitating qualitative data collection through either focus groups or one-on-one interviews.
The age range of the sample (10 to 18 years old) presented a practical constraint favoring focus groups. Children in this range often struggle to complete surveys independently and may provide unreliable responses to complex attitudinal questions. Focus groups, by contrast, do not require children to work through written instruments alone.
One-on-one interviews, while potentially valuable for other topics, presented a distinct disadvantage when the subject matter involved drug use. A sensitive topic like drug consumption would likely elicit defensive or dishonest responses in a one-to-one setting, as children feared repercussions from revealing honest opinions or experiences. Focus groups, paradoxically, create what might be called a "protective anonymity within community"—when one young person begins to speak candidly, others follow, and a context of psychological freedom emerges. This group dynamic encourages more honest exploration of attitudes than individual interviews allow.
Cost considerations further supported focus groups. With a limited budget, conducting one-on-one interviews would have reduced the achievable sample size to dangerously small levels, further undermining representativeness. Focus groups offered a more efficient use of scarce resources.
That said, focus groups carry real limitations in this context. The sample remained small, results could not be reliably generalized to the broader youth population, and the presence and influence of moderators introduced potential bias. Yet given the exploratory nature of the research question, the sensitive topic, the age of respondents, and financial constraints, focus groups represented a reasonable methodological choice.
The use of pre-screening questionnaires served a practical purpose: to increase the likelihood of sample representativeness. With focus groups inherently limited in sample size, pre-screening helped stratify participants across meaningful demographic segments, partially compensating for the method's generalizability weakness. However, the research team could not guarantee that pre-screened participants would actually attend the sessions, and several confirmed participants failed to appear, disrupting the intended balance across strata.
The overall questionnaire design demonstrated some smart choices. Offering small, age-appropriate incentives and concealing the true focus group topic until participation were both sound practices to minimize advance bias. Yet the question wording itself contained significant flaws. Question 4—"I know people/acquaintance who:"—lacked clarity. Did respondents indicate "never use drugs" only if all their acquaintances abstained, or if any did? The question should have been split: "Do you know people who take drugs?" followed by "How often?" This structure would have yielded clearer, more interpretable data. Similarly, Questions 3 and 6 lacked weighted or scaled response options. Question 6, which invited open-ended responses about where youth learned about drugs, generated data difficult to analyze and compare systematically.
The focus group interview protocol showed both strengths and critical oversights. The team correctly conducted brainstorming sessions, organized themes into five categories, and wisely sequenced questions to begin with non-controversial topics and gradually escalate toward sensitive subjects. This funneling approach typically improves response depth and honesty as participants become more comfortable.
However, the protocol contained a fundamental conceptual error: it treated all drugs as a single category. The research used the umbrella term "drugs" throughout sessions without distinguishing between cannabis (a milder substance) and heroin (a hard drug). This conflation prevented the team from understanding which substances youth knew about, what attitudes they held toward each, or which consumption patterns concerned them most. Without this differentiation, any resulting marketing messages could not be calibrated to address specific drug types—a critical flaw for a de-marketing campaign.
A second protocol problem emerged from the use of identical questioning across the two age strata (10–13 and 14–18 year-olds). While these groups may face similar drug pressures, developmentally appropriate communication differs substantially between a 10-year-old and an 18-year-old. The protocol should have included age-differentiated questions or at minimum separate interview guides tailored to cognitive and social maturity levels.
Finally, the protocol's decision to revisit unanswered questions in later sessions introduced contamination. If certain questions went unanswered, there was likely a reason—sensitivity, lack of knowledge, or discomfort—and forcing responses late in the session risked generating data of questionable validity. Including these late-session responses in the final analysis without weighting them differently than early answers inflated their apparent reliability.
"How school sampling and TV screening exclude high-risk youth populations"
"Synthesis of design failures and their impact on campaign messaging"
You’re 58% through this paper. Sign up to read the remaining 2 sections.
Sign Up Now — Instant Access Already a member? Log inAlways verify citation format against your institution’s current style guide requirements.