ICRA_AdCouncil_2011_Tracking_rev_6 23

SUPPORTING STATEMENT JUSTIFICATION FOR

FOOD SAFETY EDUCATION CAMPAIGN TRACKING RESEARCH

1. Circumstances Making Collection Of Information Necessary:

This is a request for approval of information collection addressing paperwork requirements related to evaluation research for the Food Safety and Inspection Service (FSIS)’s forthcoming public service advertising campaign on food safety.

The FSIS has been delegated the authority to exercise the functions of the Secretary as provided in the Federal Meat Inspection Act (FMIA) (21 U.S.C. 601 et seq.), the Poultry Products Inspection Act (PPIA) (21 U.S.C. 451 et seq.), and the Egg Products Inspection Act (EPIA) (21 U.S.C. 1031 et seq.). These statutes mandate that FSIS protect the public by verifying that meat and poultry products are safe, wholesome, unadulterated, and properly labeled and packaged.

FSIS, in partnership with the Ad Council, the Food and Drug Administration, and the Center for Disease Control, has developed a new national public service advertising campaign to educate the public about the importance of safe food handling and how to reduce the risks associated with foodborne illness. The Ad Council and FSIS are seeking approval of an information collection to help evaluate the impact of the campaign. The collection will take the form of a survey of members of the target audience, and will help gauge awareness of the advertising, attitudes regarding safe food preparation, and self-reported prevention behaviors. The survey will be fielded once prior to launch of materials in mid-June, and then again 9-12 months following launch to monitor any shifts over time.

Campaign Overview:

After receiving a briefing on foodborne illness and USDA priorities for the public education campaign, the Ad Council and JWT conducted an audit of existing research and statistics surrounding the issue and prevention behaviors. Following this review, the Ad Council and JWT conducted consumer research to better understand perceptions of foodborne illness and food safe behaviors held by the target audience. These research sessions were conducted with OMB approval in November 2010. Next, the Ad Council and JWT developed a communications strategy based on research findings that clearly articulates the proposed approach to communications.

JWT then developed creative concepts – scripts, graphical treatments, etc – that stem directly from the communications strategy. These concepts were qualitatively tested with members of the target audience in March. Finally, prior to the release of the advertising campaign in July, the Ad Council will conduct a tracking study to monitor awareness of the campaign as well as any changes in perceptions of foodborne illness and reported safe food handling behaviors.

The campaign targets parents, aged 20 to 40, who are caregivers for children between the ages of 4 and 12. Parents have been identified as the target audience because they are most likely to be preparing food for themselves and others, and they have an incentive to listen to food safety messages and adopt or change their behaviors as a result.

The survey will be administered using a national random digit dial (RDD) phone methodology in both English and Spanish. Each respondent will answer questions about their attitudes about food safety, their awareness of the risks of foodborne illness, their own efficacy with regard to preventing foodborne illness, and their own use of safe food-handling practices. The PSAs will also be described to respondents in order to gauge recognition of the ads in market.

Once the post-wave survey is fielded 9-12 months after the benchmark survey, the Ad Council will compare results to identify any shifts in attitudes, awareness, or behaviors that occurred while the PSAs were in market.

This collection is just one of the evaluation mechanisms that will be used by the Ad Council to assess the impact of the campaign. Other tools include monitoring exposure of the advertising in donated media placements, website analytics, social media monitoring, and orders of fulfillment and collateral materials. What makes the tracking survey unique is that it allows for an estimation of attitudes and perceptions that are otherwise impossible to gauge through instruments other than a survey. The Ad Council utilizes tracking surveys as a standard evaluative tool for most of its 50+ PSA campaigns. The approach allows for awareness and aided PSA recognition questions to be asked alongside attitudinal and behavioral questions and allows any correlations between variables to be observed. We have shared examples of prior tracking reports and campaign evaluations with OMB to illustrate the analysis and use of tracking data to further campaign goals.

2. How, By Whom and Purpose Information Is To Be Used:

This public education campaign is being conducted in support of the Dept. of Health and Human Services’ Healthy People 2020 goals. With regard to food safety, that initiative lays out specific targets to increase reported adoption of safe food handling behaviors by 2020. The targets are as follow:

• Clean – wash hands and surfaces often:

  • Goal: 10% improvement by 2020

• Separate: Don’t cross-contaminate

  • Goal: 3.4% improvement by 2020

• Cook: Cook to proper temperature

  • Goal: 35% improvement by 2020

• Chill: Refrigerate promptly

  • Goal: 3.4 % improvement by 2020

The Ad Council PSA campaign is only one component of several food safety education programs that will address these behaviors, but it is expected to be an important one. . FSIS expects that this tracking survey will serve several critical purposes functions: 1) To evaluate the impact of the PSA campaign, and determine if any ‘course corrections’ need to be made after year one of the PSA program; 2) To report back to internal FSIS leadership and others on the return of investment, as measured by shifts in awareness, attitudes and behaviors, in order to justify the funding of this program; 3) To evaluate how the campaign is or is not contributing to improvements in the four Healthy People 2020 food safety behaviors listed about.

The survey questionnaire being submitted along with Statements A and B asks a range of questions related to preventing foodborne illness. Several questions are of central importance, and the success of the campaign will be largely judged on statistically significant increases on the key measures listed below.

1. Safe Food Behaviors – extent to which respondents practice key behaviors relating to cleaning, separating, cooking and chilling food during meal preparation. (Question 13)

   1. Top Box Answer Choice: “I take this step every time I prepare a meal”

   2. Top 2 Box Answer Choice: “I take this step every time I prepare a meal” OR “I take this step most of the time, but not always. “

2. Perceptions of importance – how important respondents consider key behaviors when they prepare food. (Question 14)

   1. Top Box Answer Choice: “Extremely important”

   2. Top 2 Box Answer Choice: “Extremely/Very important”

3. Self Efficacy – how much control do respondents feel they have over preventing their children from getting sick from food prepared in the home? (Question 17)

   1. Top Box Answer: “Complete control”

   2. Top Box Answer Choice: “Complete control” OR “A lot of control”

4. Awareness of advertising and issue – aided and unaided recognition of PSAs. (Questions 19-26)

   1. “Yes” [respondent reports having seen the ad described]

Once the benchmark research is fielded, the Ad Council and Cayenne Global will analyze the data to establish baseline – pre-campaign – levels and also note any key differences between subgroups within the sample. Upon fielding the post-wave survey in 9-12 months, the Ad Council will monitor any shifts between survey waves in the key metrics listed above. This information will be used together with data from other sources – website analytics, donated media reporting, etc – to prepare an overall assessment of the campaign’s performance in its first year. In addition, any insights or trends regarding perceptions of safe-food handling practices will be used to inform future rounds of PSA strategy and creative execution.

The threshold to defining the campaign a “success” will be based on whether pre- to post- increases are measured in this study, and if these increases pass statistical significance testing. It is difficult to predict the precise percentage increases we hope to observe in our key attitudinal measures. In our experience across our 50+ PSA campaigns, we have found that every campaign performs differently based on the issue, the target audience, the donated media support and the overall media environment surrounding the issue. In similar tracking surveys conducted on behalf of other Ad Council campaigns, several have shown no significant behavioral increases in the first year, while others have demonstrated behavioral increases of 10 points or more, sometimes over the course of several years of research. Acknowledging the limitations of this study – specifically questions of non-response bias detailed in Form B - the Ad Council considers success to be statistically significant growth (increases of 4-5 percentage points are generally the threshold of statistical significance of studies with this sample size) in the percentage of respondents who select top-box and/or top two box answers to the questions outlined above.

T the current PSA plan intends to produce another round of advertisements in 1-2 years time, which will be accompanied by additional waves of this survey with identical methodology. As such, the desired increases may be observed over a course of several survey waves, assuming funding and approval occurs to support this research. Along with providing a key piece of program evaluation for the campaign, this research will help FSIS assess progress toward the Healthy People 2020 goals outlined above.

The following is a discussion of the required information collection and recordkeeping activities.

The Ad Council, on behalf of FSIS, plans to work with Cayenne Research, a third-party vendor, to conduct a phone survey of 1200 parents of children 4-12 – 600 English-speakers and 600 Spanish-speakers. The survey will take approximately 15 minutes to complete. It is estimated that there will be about 6000 non-respondents to the interviews, i.e., parents who are contacted but chose not to participate or do not qualify.

There are 500 total burden hours for the information collection request relating to the Food Safety Education Campaign tracking research.

3. Use of Improved Information Technology:

All phones will be administered via a Computer Aided Telephoning Interview (CATI).

4. Efforts to Identify Duplication:

The Ad Council consulted with FSIS, FDA, CDC, and USDA during the planning of this tracking study with the intent of avoiding duplication of other previous studies. Indeed the FDA and FSIS jointly conducted the 2006 Food Safety Survey which offers data on safe-food handling behaviors. Several of the questions in Ad Council’s survey were derived from the FDA/FSIS questionnaire. What differentiates this survey from the earlier efforts is the ability to link PSA awareness with attitudinal and behavioral metrics. Our primary interest is attempting to assess any impact that the PSAs from this campaign may have had upon attitudes towards food safety, which is an area that previous research has not covered.

The Ad Council survey focuses on “Cook, Clean, Chill and Separate,” which are behaviors that Healthy People 2020 (an initiative from the Dept. of Health and Human Services) seeks to promote.

5. Methods to Minimize Burden on Small Business Entities:

Only consumers will be surveyed.

6. Consequences If Information Were Collected Less Frequently:

This study will be conducted twice – once in June, 2011 prior to the launch of the advertising campaign, and again 9-12 months following launch. Without a post-wave study, the benchmark data will lack context and any shifts while the ads are in market will be impossible to observe. Since the postwave will be fielded more than 6 months following OMB review of this request, the Ad Council will submit an approval package closer to the date of the postwave.

7. Circumstances That Would Cause the Information Collection to Be Conducted In a Manner:

• requiring respondents to report informa­tion to the agency more often than quarterly;

• requiring respondents to prepare a writ­ten response to a collection of infor­ma­tion in fewer than 30 days after receipt of it;

• requiring respondents to submit more than an original and two copies of any docu­ment;

• requiring respondents to retain re­cords, other than health, medical, governm­ent contract, grant-in-aid, or tax records for more than three years;

• in connection with a statisti­cal sur­vey, that is not de­signed to produce valid and reli­able results that can be general­ized to the uni­verse of study;

• requiring the use of a statis­tical data classi­fication that has not been re­vie­wed and approved by OMB;

• that includes a pledge of confiden­tiali­ty that is not supported by au­thority estab­lished in statute or regu­la­tion, that is not sup­ported by dis­closure and data security policies that are consistent with the pledge, or which unneces­sarily impedes shar­ing of data with other agencies for com­patible confiden­tial use; or

• requiring respondents to submit propri­etary trade secret, or other confidential information unless the agency can demon­strate that it has instituted procedures to protect the information's confidentiality to the extent permit­ted by law.

All information collection and recordkeeping activities in this submission are consistent with the guidelines listed above.

8. Consultation with Persons outside the Agency:

FSIS has consulted with the Ad Council and JWT in developing this research. FSIS published a 30 day notice in the Federal Register (76 FR 28727; May 18, 2011). The Agency received two public comments. One of the comments did not deal with the information collection per se. The other comment was submitted by the Center for Science in the Public Interest. This comment objected to the answer option on questions 13 and 14 because it mislead consumers. In response, we revised the answer option from “Cook meat, poultry, and fish to a safe temperature (e.g., 160 degrees formeat using a food thermometer” to “Use a food thermometer to make sure that meat, poultry, and fish have been cooked to a safe temperature (e.g., 160 degrees for ground beef).” CSPI also objected to the use of statements of “separate meat and dairy while cooking, boil milk before drinking, wash meat or chicken before cooking” as potentially misleading consumers. In response, we removed these three answer options from question 13. However, we kept them in question 14 but changed the question language to say, “There are some steps that are important to take when preparing food at home to help prevent food poisoning and others that are not necessary. Please indicate, in your opinion, how important or unimportant the following steps are with regard to preventing food poisoning in your home.”

9. Payment or Gifts to Respondents:

Respondents will not receive financial compensation for their participation in the survey.

10. Confidentiality Provided To Respondents:

The survey, while asking a range of personal questions – educational attainment, household income, food habits, etc. – will not be personally identifiable to the respondent. All identities will be kept anonymous, and respondents will be so notified during the survey.

11. Questions of a Sensitive Nature:

The applicants are not asked to furnish any information of a sensitive nature.

12. Estimate of Burden

The Ad Council estimates that in order to recruit 1200 respondents for English and Spanish language surveys, 6000 will choose not to respond or won’t qualify and will spend 2 minutes being administered screening questions prior to termination. The 1200 respondents (English and Spanish) who qualify will each spend approx. 15 minutes answering the survey. The total burden for the collection is 1200 survey respondents and a total of 500 burden hours.

BURDEN HOURS

Type of Respondent	No. of Respondents	No. of Res- ponses per Respondent	Total Annual Responses	Time for Response in Mins.	Total Annual Time in Hours
Survey Respondents - English	600	1	600	15	150
Survey Respondents – Spanish	600	1	600	15	150
Survey Non-respondents: English and Spanish	6000	1	6000	2	200
Total	7200	1	6200	15	500

There are no costs to the respondents other than their time. The Agency estimates that it may cost respondents $12 an hour in loss of potential salary by participating in the survey. Respondents will spend a total of 500 hours and $6,000.

13. Capital and Start-up Cost and Subsequent Maintenance

There are no capital and start-up costs and subsequent maintenance burdens.

14. Annual Cost to Federal Government:

The cost to the Federal Government for these information collection requirements is $70,592 for the tracking research. The costs arise from data collection, analysis, and reporting.

15. Reasons for Changes in Burden:

This is a new information collection totaling 500 burden hours.

16. Tabulation, Analyses and Publication Plans:

The data and analysis may be cited in press coverage surrounding campaign launch

17. OMB Approval Number Display:

FSIS will display the OMB approval number on the interview guide.

18. Exceptions to the Certification:

There are no exceptions to the certification. This information collection accords with the certification in item 19 of the OMB 83-I.

Page 8