Evaluation of the Communities Putting Prevention to Work (CPPW)
National Prevention Media Initiative
New
Supporting Statement: Part B
August 3, 2012
Jeffrey McKenna
Associate Director of Communication Science
Phone:
770-488-8238
FAX:
770-488-5962
E-mail: [email protected]
U.S. Department of Health and Human Services
Centers for Disease Control and Prevention
Atlanta, Georgia
Section B
B. Collections of Information Employing Statistical Methods 1
B.1 Respondent Universe and Sampling Methods 1
B.2 Procedures for the Collection of Information 2
B.3 Methods to Maximize Response Rates and Deal with No Response 3
B.4 Tests of Procedures or Methods to be Undertaken 4
B.5 Individuals Consulted on Statistical Aspects and individuals Collecting and/or
Analyzing Data 4
LIST OF ATTACHMENTS
Attachment 1. Authorizing Legislation
1a. Public Health Service Act
1b. American Recovery and Reinvestment Act of 2009 (ARRA)
1c. Patient Protection and Affordable Care Act
Attachment 2. Federal Register Notice
2a. Notice Published May 23, 2012
2b. Summary of Public Comments and CDC Response
Attachment 3. (MAPPS) Interventions for Communities Putting Prevention to Work
Attachment 4. Screener for the Community Telephone Interview
Attachment 5. Community Telephone Interview
The study will consist of data collected, using a telephone-administered questionnaire, from respondents in each of the 39 CPPW communities where awardees were funded to address obesity. The study will include a sample of adults aged 25 or older identified from landline and cell phone lists. Two cycles of information collection will be conducted. Information from the first cycle of information collection will be analyzed to assess community initiative recall associated with media as well as mediating domains (e.g., attitudes, beliefs, behaviors, intentions) during the initiative but after media and messaging. The second data collection will be conducted to permit analysis of recall and mediating domain decay some 4-5 months after the initial data collection.
In each of the two data collections, samples of equal size will be obtained from each of the 39 communities; the size of the entire sample for each data collection cycle will be 6,000 with approximately 153 respondents coming from each of the communities. The goal of the two surveys is not to obtain a representative sample for each of the CPPW communities, but rather to have a representative sample for respondents in CPPW communities across the U.S. This collection of information will enable CDC to explore associations between selected sub-populations in CPPW communities. These include differences by race/ethnicity, parental status, and dealing with a personal or family risk of being overweight.
Because research (National Health Statistics Reports; Number 39, April 20, 2011; Wireless Substitution: State-level Estimates From the National Health Interview Survey, January 2007–June 2010) has shown that there are demographic and location differences between households that have landlines (alone or in combination with cell phones) and cell phone only households, the sample pool will include respondents from both types of households; approximately 75% of the sample will be from landline telephone lists and 25% will come from cell phone only lists. The advantage of this approach is that it includes both groups and the data will reflect any potential differences in associations between CPPW initiative exposure, other demographic factors, and the obesity-associated attitudes, intentions, and behaviors being measured. Because the sample size for the cell phone only portion of the sample will be sufficiently large (n≈1,500) a comparative analysis of these two groups will also be able to be conducted.
To reduce the effects of sampling and non-sampling error, the sample will be weighted to reflect probability of selection; post-stratification weighting adjustments will also be applied to the sample.
We conducted power analyses to determine the optimal sample size for detecting statistically significant differences between African-Americans, Hispanic, and White respondents who live in CPPW communities as well as between respondents by parental status with regard to recognition of CPPW initiative taglines and/or identifiers. We expect to find that 56% of respondents are able to identify the initiative. We assume, based on other studies that speak to decay in memory of media and communication efforts, many respondents living in CPPW communities will not be able to remember the initiatives tag line(s)/initiatives names and therefore would like to be able to declare significant a comparison group where 50% as compared to 56% were able to identify the initiative. Given a type I error rate of 0.05 (alpha = 0.05), we will achieve 90% power to detect this difference if we look at race/ethnicity and parental status with a sample size of 1,453 (or ≈1,500) per comparison for a total sample of 6,000.
Because of the interest in looking at minority population differences, African Americans and Hispanics will be oversampled. Across all CPPW communities, 25 percent of respondents will be African American, 25 percent will be Hispanic, approximately 35 percent will be white, and approximately 15 percent will be Other; the Other group will come from areas like the communities in Hawaii where there are few-to-no Black or Hispanic population groups. In the end, the sample will contain approximately 2,100 white respondents, 1,500 African American respondents, 1,500 Hispanic respondents, and 900 Other respondents. Quotas for parental status will also be set so that 50 percent (or 3,000) of the sample will be someone who is in a household that contains at least one child between the ages of 6 and 15.
We have developed a computer-assisted telephone interview instrument for this study to collect information from respondents (see Attachment 5, Community Telephone Interview). The questions on the instrument will be read to the respondents by professional telephone survey administrators, all of whom will be trained in this instrument in particular.
Subject recruitment will be conducted by randomly selecting telephone numbers from landline lists where the area code and exchanges have been link to zip codes that are within CPPW communities. Respondents (approximately 25%) will be recruited from cell phone lists where current respondent address can be identified – and is known to be in a relevant community.
The data collectors, using telephone lists with numbers randomly selected from master lists, will call the household and, using the scripted introduction statement at the beginning of the phone call (see Attachment 4, Screener for the Community Telephone Interview), explain the survey, its purpose and how long administration will take; screen the person on the telephone for eligibility; and as appropriate, administer the survey (Attachment 5, Community Telephone Interview).
The survey information will be keyed by the data collector in an electronic system that has been programmed with within range instructions, validation checks, and skip patterns, as appropriate. If no one answers the telephone the data collector will move on to the next number on the telephone list.
Telephone number information will not be included in the analysis data file; there will be no information in the response data base that can be directly tied to any individual or household. The data collection vendor will transmit an SPSS data file, which contains no personally identifying information, to FHI 360 (CDC’s evaluation contractor), for analysis.
Once the data are received, they will be logged and archived. Thorough data validation will be performed to assess the quality of the data available to perform the planned analysis. The data collection firm will send interim data every two weeks during data collection to assure that there will be no data quality problems.
The professionals conducting the telephone surveys have been instructed to stress early in their introduction of the survey that it is sponsored by the Centers for Disease Control and Prevention – “…sometimes known as the CDC.” Public opinion surveys have shown that the CDC is well regarded and trusted. Experience has shown that coming from a trusted source, coupled with being a short questionnaire (10 minutes), will enhance the survey acceptance rates – and minimize the number of individual who quit taking the questionnaire once they have started.
The current questionnaire was pre-tested by five individuals who had not seen it previously. During this pretest we assessed the ability of the respondents to understand the questions being asked, asked for language and question format modifications, and established estimated response times for persons who are and are not living with a child/children between ages 6 and 15.
The information learned from pretesting was used to finalize the screener and telephone interview instruments (see Attachments 4 and 5).
The data collection instrument contains questions in the following areas:
Attitudes and beliefs about obesity in their community
Characterization of the best approach to solving the problem of obesity in their community
Self-efficacy to take responsibility for healthy eating and being physically active as well as contributing to community obesity prevention efforts
Interest in obesity issues
Taking actions related to obesity prevention personally and at a community level
Intentions related to obesity prevention at the community level
Unaided campaign awareness
Aided campaign awareness
Demographics
Jeffrey McKenna, 770-488-8238. Associate Director of Communication Science is the Principal Investigator and Technical Monitor for the study. He has overall responsibility for overseeing the design and administration of the survey, and he will be responsible for analyzing the survey data.
FHI 360 is the project contractor responsible for development of the telephone version of this questionnaire, collection, and analysis of communication and media data from the CPPW initiative. John Strand, MA (202-884-8902), of FHI 360 serves as FHI 360’s Project Director. In this role, he is the primary contact with the Technical Monitor and oversees work on all project tasks.
The survey instrument, sampling and data collection procedures, and analysis plan were designed in collaboration with researchers at CDC, and FHI 360. The following personnel are involved in design of the protocol and data collection instrument:
(1) Suzanne Gates, MPH Communication Lead CDC / DCH Phone: (404) 625-7580 E-mail: [email protected] Specific contribution: Consultant on study and survey instrument design |
(2) Rita Brett Supervisory Health Communication Specialist CDC / NCCDPHP Phone: (770) 488-5304 E-mail: [email protected] Specific contribution: Consultant on study and survey instrument design
|
(3) Rachel Ciccarone MPH, CHES Health Communication Specialist CDC / NCCDPHP Phone: (770) 488-6542 E-mail: [email protected] Specific contribution: Consultant on study and survey instrument design |
(4) Amanda Aldridge Dudley, MPH Health Communication Specialist CDC / NCCDPHP Phone: (770) 488-5051 E-mail: [email protected] Specific contribution: Consultant on study and survey instrument design
|
(5) Thomas Lehman, MA Associate Director, Research FHI 360 / Social Marketing and Communication Center Phone: (202) 884-8863 E-mail: [email protected] Specific contribution: Evaluation Task Lead
|
(6) Rebecca Ledsky, MBA Senior Technical Director, Research FHI 360 / Social Marketing and Communication Center Phone: (202) 884-8814 E-mail: [email protected] Specific contribution: survey instrument development and sampling |
(7) Michelle Jones-Bell, MA Senior Technical Director, Research FHI 360 / Social Marketing and Communication Center Phone: (202) 464-3789 E-mail: [email protected] Specific contribution: survey instrument design |
|
| File Type | application/vnd.openxmlformats-officedocument.wordprocessingml.document |
| Author | DHHS |
| File Modified | 0000-00-00 |
| File Created | 2021-01-30 |