 |
 |
|
|
National Youth Anti-Drug Media Campaign
Phase 1 (Report No. 1)
2.4 Media Tracking
In Phase I of the Media Campaign, paid and unpaid anti-drug television advertisements and drug-related newspaper articles were monitored. The evaluation contractor tracked these media in target and comparison sites during the 3 months (October-December 1997) preceding the Media Campaign -- the baseline period -- and during the first 3 months (January-March 1998) following the Media Campaign -- the intervention period.
For the television monitoring component of the analysis, a media monitoring and analysis company was subcontracted to track anti-drug ads aired on affiliates of the three major national television networks (ABC, CBS, and NBC), national cable WBN (Time-Warner cable), CNN, FOX, Univision, TBS, UPN, IND, and Telemundo (Spanish-language cable). The following five communities were not able to be electronically monitored: Boise, Sioux City, Tucson, Eugene, and Duluth. Attempts to collect advertising information from the stations manually, through monthly telephone interviews, did not yield reliable or complete data and, therefore, the data were not included.
Data were collected on the sponsor, frequency, gross rating point (GRP), day part, volume, content, market share, and estimated cost of anti-drug advertising appearing in target and comparison sites before and during the Media Campaign. Of these variables, only sponsor, frequency, and GRP were analyzed for this report. Day part, volume, content, market share, and estimated cost data require further verification and analysis. These data, along with some refinements in the sponsor and GRP data, will be presented in the final report of Phase I. The variables are described below.
Sponsor data -- Sponsor data were separated into Campaign/PDFA and other drug ads in the monitoring of media activity. Other drug ad data were not divided by specific sponsors for Phase I because of the unreliability of sponsor identification. The Campaign/PDFA category includes any PDFA advertisements that were part of the paid Media Campaign as well as any other PDFA ads (e.g., PSAs that were part of PDFA's own rollout campaign) airing in the target and comparison sites. Because imprinting of the ads designated as "Campaign" ads was not done, it was not possible to distinguish the Campaign ads from other existing PDFA ads. Verification of sponsor data will be completed in subsequent phases of the Media Campaign. For the Phase I final report, the analysis will attempt to further divide the Campaign/PDFA designation into Media Campaign paid ads, pro bono ads resulting from the Media Campaign, and naturally occurring PDFA ads placed as a result of ongoing organization activities. The verification of these PDFA ads must be completed using placement verification reports provided by television stations, which are not yet available.
Frequency -- Frequency, or the number of anti-drug ads aired, was broken out by sponsor and reported in the aggregate form for all sites. Total frequency and changes in total frequency, before and during the Media Campaign, were analyzed within and across target and comparison sites.
Gross rating point -- GRP is a standard industry estimate based on the A.C. Nielson television index and records of viewer diaries. Each rating point estimate reflects 1 percent of the target audience exposed to an ad. In this case the target audience is the viewing demographic group aged 12-24. The ads oriented to parents, typically ages 35-45, aired in the same prime time slots (day parts) as ads for the target audience of 12- to 24-year-olds. Therefore, monitoring of a separate audience including only parents was not required to determine their exposure to ads targeted to them. GRPs were broken out by sponsor and aggregated for each site before and during the Media Campaign. Changes in total GRPs from the baseline to intervention period are represented through a "rate." Rate is the number of times an ad's GRP increased or decreased during the Media Campaign period. It is calculated by dividing the intervention total GRPs by the baseline total GRPs. The total GRP and change in total GRP data were examined within and across target and comparison sites. Average GRPs could not be obtained for this analysis because inaccuracies in measuring individual ad GRPs were detected and are in the process of being corrected. However, in the aggregate form, GRP is a reliable estimate of audience exposure to the anti-drug message.
Day part -- Day part indicates the time of day an advertisement aired. There are 10 time slots or "parts" of the day: early morning, daytime, early fringe, early news, kids, late fringe, late news, prime access, prime time, and weekend daytime. These viewing periods provide a critical measure of exposure because certain time slots reach a greater share of the target audience than others.
Volume -- Volume is the total amount of airtime an ad received. It is calculated by multiplying the number of times an ad aired by the length of the ad. Ad length ranges from 10-60 seconds.
Content -- Content or "type of drug" refers to the focus of the advertisement (i.e., drug-free marijuana, drug-free inhalant, drug-free heroin, drug-free cocaine, drug-free crack-cocaine, or drug-free unspecified.)
Share of market -- Share of market provides a general estimate of the proportion of airtime drug prevention ads receive relative to other major social issues. These other topics include heart health, lung disease prevention, general health, tobacco control and prevention, gun violence, environmental protection, and drinking and driving prevention. Frequency, GRPs, day part, and cost of placing ads were captured for these issues.
Estimated cost -- Estimated cost is the value of buying particular advertising slots. It is based on GRP, cost-per-point, spot length, day part, program the ad appeared in, and other market and historical data. Average cost will be used to estimate the "importance" of the anti-drug issue to television stations as the Media Campaign progresses. A station's willingness to place a drug prevention public service announcement (PSA) in a time slot that could otherwise be sold to a commercial advertiser (rather than placing the PSA in a commercially undesirable time slot), may reflect changing attitudes about the importance of the issue, either to the media organization or the community.
On a monthly basis, data were compiled on the variables cited above, in hard copy and electronic form. These data were reviewed for completeness, reliability, and validation. Quality control checks were also performed. Data were then reconfigured where applicable so that analyses of movement or change in media variables (e.g., frequency and GRP) could be conducted. This process is essential because much of television tracking data has been used only by industry analysts for purposes of monitoring political media campaigns and is untested against commonly accepted standards of social science research. The hard copy also included storyboards, which capture frames (in 4-second intervals) of advertisements that aired. These storyboards were used to verify the sponsor and content data.
Site-specific findings were tabulated and are presented in the target site profiles included in Chapter 4 (see Exhibit 4-2, Exhibit 4-3, Exhibit 4-4, Exhibit 4-5, Exhibit 4-6, Exhibit 4-7 ,Exhibit 4-8, Exhibit 4-9, Exhibit 4-10, Exhibit 4-11, Exhibit 4-12, and Exhibit 4-13). Each target site is presented next to its designated comparison site, for a total of nine pairs. Although data are available on Austin, they are not included in the tables. Austin's matching target city, Tucson, is among the five sites mentioned earlier that could not be electronically monitored. Because there is no basis for comparison, data on Austin are not being presented.
The newspaper monitoring component of Phase I involved identifying articles relating to drugs in each of the target and comparison sites, using a newspaper clipping service (Burrells), and two online services (Dialog and Dow Jones). The monitoring captured news reporting on illicit drugs exclusively; alcohol and tobacco were not included.
In each of the target and comparison sites, except Eugene, between one and six newspapers were monitored. Eugene newspapers were not indexed by the major news clipping services and, therefore, could not be monitored cost-effectively. Section 3.4.2 contains a list of newspapers selected for monitoring and their circulation.
Newspaper data were coded by type of article (i.e., feature, news brief, editorial, or announcement). These classifications were used to determine the "importance" of a drug related issue (i.e., a feature article requires more resources to develop than a news brief or announcement, and therefore indicates that its topic is more important than topics covered by news briefs or announcements). Articles are being further coded by subject matter (e.g., law enforcement and crime, treatment, prevention, drug policy, etc.) and will be presented in the final report of Phase I. The content topics will be used as an indicator of the nature of community interest in drug issues.
Preliminary analysis of Phase I newspaper tracking data covers only baseline statistics on the frequency of newspaper reporting in all forms, except advertisements. The lag time required to accumulate the newspaper data has prevented development of pre-Campaign versus intervention period trends in this phase of the analysis. Intervention period trends will be available in the final report of Phase I. Exhibit 2-3 shows the range of newspapers tracked across the sites.
The chart presented in Exhibit 2-4 provides an overview of the specific PDFA ads that were intended to be aired in each of the 12 target sites for the Phase I Media Campaign. ONDCP used existing PDFA advertisements for the Phase I Media Campaign because they were already available. New ads will be developed purposefully for subsequent phases of the Media Campaign. The ads shown in Exhibit 2-4 refer to television ads. These were targeted for specific sites and at different exposure levels depending on the nature of the particular drug problem. The tracking data that are currently being analyzed will verify whether the 12 target sites actually ran the scheduled ads as planned by the Media Campaign implementers -- this information will be available in the final report. A final note: some of these same ads may have been appearing in comparison sites but would have been aired as PSAs rather than paid ads.
