CDC Fellowship Programs Assessments
OMB Control No. 0920-16ARO
Supporting Statement B
Fátima Coronado, MD, MPH
Deputy Associate Director for Science
Division of Scientific Education and Professional Development
Center for Surveillance, Epidemiology, and Laboratory Services
Office of Public Health Scientific Services
Centers for Disease Control and Prevention
Phone: 404-498-6551
Email: [email protected]
Submitted August 30, 2016
Table of Contents
PART B. COLLECTIONS OF INFORMATION EMPLOYING STATISTICAL METHODS 1
B.1. Respondent Universe and Sampling Methods 1
B.2. Procedures for the Collection of Information 1
B.3. Methods to Maximize Response Rates and Deal with No Response 2
B.4. Test of Procedures or Methods to be Undertaken 3
B.5. Individuals Consulted on Statistical Aspects and Individuals Collecting and/or Analyzing Data 3
Respondents to data collections under this generic clearance will include potential and current applicants, current fellows, alumni, mentors or supervisors, and employers who hire alumni. Both purposive and probabilistic samples will be employed for these information collections, depending on the purpose of the specific collection.
In some cases, purposive samples of respondents will be selected to maximize diversity across demographic subgroups (e.g., age, education level, gender). In other instances, samples will be drawn to include specific characteristics (e.g., discipline or current employment settings). Sampling frames will be created by a variety of methods, including solicitation through the use of the Fellowship Management System (FMS; OMB Control No. 0920-0765), FedScope (https://www.fedscope.opm.gov/), or other means. Potential participants will be randomly selected from the sampling frame and screened for appropriate demographic or other characteristics as indicated by the purpose of the specific project.
Probabilistic procedures will be used when necessary to constitute a representative sample of a population. As an example, a systematic random sample of fellowship program alumni might be contacted to assess their perceptions about the utility of the specific training they received.
Each genIC will describe the respondent universe and sampling methods. GenICs under this proposed generic clearance will use both qualitative and quantitative information collection methods. Qualitative methods are often used for exploratory and explanatory purposes. These methods rely on small samples and collect information in open-ended formats to generate saturated themes, which provide context and generate information to aid development of structured quantitative instruments. Quantitative methods will employ larger samples and collect information by using closed-ended response formats to describe the population.
The sampling and data collection will depend on the design of the proposed information collection. Participants will be identified, recruited, and stratified in accordance with the specific purpose of the genIC. For example, in some instances individuals will be stratified based on association with a certain fellowship program or duration since graduating from the fellowship program. Each genIC project team will describe its planned design, sampling methods, and data collection methods. CDC does not anticipate that these collections will yield generalizable data.
Each genIC submission will include the following:
Details about who will collect the information (e.g., trained interviewers, facilitators).
A description of analytical techniques to be employed and tools used for data collection (e.g., surveys, interviews, focus groups), including letters of invitation, screenshots of web-based surveys, and other visual aids.
An in-depth analysis plan and estimation procedures aligned with the project’s purpose, collection questions, and design. Depending on the type of collection and number of participants, multiple tools may be used for data analysis. For a limited number of respondents, the analysis might exclude use of qualitative data analysis software. For a large number of respondents, software such as NVivo, ATLAS.Ti, or MaxQDA will be used for qualitative data. For quantitative data, SPSS, SAS, EpiInfo, R, or other appropriate programs will be used.
Description of quality control procedures.
When necessary, the genIC will describe periodic data collection methods and unusual problems requiring specialized sampling.
Before collecting information, investigators will inform respondents that participation is voluntary, that respondents will not be personally identified in any published reports, and that their privacy will be protected to the extent allowed by federal law. Nonresponse bias may be expected (e.g., alumni distant from graduation year). Study designs and methods will be selected to minimize the effect of nonresponse bias and the project team will acknowledge when and how it might impact their results.
GenICs will describe the procedures planned to maximize response rates. The following are examples of procedures that may be used to maximize response rates:
Inform potential respondents of what the project is asking, why it is being asked, who will see the results, how the results will be used, and how the findings will be put into action.
Inform potential respondents about the importance of these studies and encourage participation through a variety of methods, including professional associations’ or community organizations’ newsletters, and letters of support from key individuals.
Address data security and anonymity.
Establish a dedicated email address or phone number for potential respondents to confirm legitimacy of the data collection, ask questions, voice concerns, or seek technical assistance.
Train interviewers and focus group moderators on strategies for engaging respondents, role playing, and techniques for fostering respondent cooperation and completion of data collection.
Provide respondents options for completing data collections (e.g., online, in person, fax, email).
Send follow-up reminders to nonrespondents, encouraging participation.
Personalize contacts with respondents.
Design electronic instruments for easy access for respondents.
If, during data analysis, systematic issues occur with nonresponse among certain subsets of the respondent universe, CDC will describe any trends in nonresponse and determine whether changes in methodology are warranted for future information collections.
All data collection instruments for each genIC will be pilot tested, but, when applicable, with no more than nine nonfederal respondents. GenICs will describe efforts to improve or refine the instruments based on the pilot test findings and feedback. The pilot test will provide the average time to complete the assessment, including time for reviewing instructions. In addition, pilot test feedback will be used to refine protocols for respondent recruitment or follow-up.
Each genIC will identify persons consulted, within or external to their CIO, regarding statistical aspects and staff responsible for collecting and analyzing data. DSEPD offers the following persons, including contractors, as available to provide advice to DSEPD fellowship program representatives about the design of statistical and sampling procedures for genICs under this generic clearance:
Fátima
Coronado, MD, MPH
Deputy Associate Director for Science
[email protected]
Office:
(404) 498-6551
M.
Kathleen Glynn, DVM, MPVM
Associate Director for
Science
[email protected]
Office:
(404) 498-6169
Byron
Robinson, PhD
Statistician
[email protected]
Office:
(404) 498-6508
| File Type | application/vnd.openxmlformats-officedocument.wordprocessingml.document |
| Author | Paek, Margaret (CDC/OPHSS/CSELS) (CTR) |
| File Modified | 0000-00-00 |
| File Created | 2021-01-23 |