Supporting Statement B

Supporting Statement B

Initial Assessment and Evaluation of the Public Health Training Centers Program

OMB Control No. 0915-XXXX

2. Collection of Information Employing Statistical Methods

1. Respondent Universe and Sampling Methods

There are 37 Public Health Training Centers (PTHC) in the United States which comprise the universe of interest. This survey will focus on individual trainees who attend courses offered by these PTHCs. Due to the rolling enrollment of individuals into these training courses and the difficulty in obtaining enrollment information for all courses in these centers, we propose opening participation to all participants and using a quota sampling method to obtain the needed sample size. We assume equal participation across all training centers and across all course types. Since the survey will be implemented using a web-based forum, we estimate a response rate of 70%. Therefore, participation will be open to all registered trainees in courses offered by PHTC grantees. We will cease data collection when we have achieved 1,569 completed responses. This was selected as the targeted number of responses to achieve a statistical power of 80% with a discernable difference of 5% between these groups.

2. Procedures for the Collection of Information

RTI will program a self-administered web-based survey that will be distributed to PHTC trainees. RTI staff will send an introductory email to sampled respondents via the HRSA TRAIN LMS to explain the purpose of the study. Sample respondents will receive a separate email with a link to the survey questionnaire they will be asked to complete. A link to frequently asked questions regarding the survey and the study will also be provided, including contact information for project personnel who sample respondents can contact with questions.

This survey will not employ explicit stratification. We will, however, monitor the completed responses to ensure a balanced response from the various PHTC training centers, while trying to minimize respondent burden. For example, if the quota is reached but the majority of the responses come from a small cluster of training centers, we will hold open the data collection window to allow other trainees to have the opportunity to respond to the survey. Data will be analyzed using SAS® software. The analysis will be comprised of frequencies, cross-tabulations and graphs demonstrating percent change in perception between trainees at the beginning and end of their course.

RTI will conduct a thorough test of the questionnaire utilizing a formal testing plan to review all logic in the questionnaire and to ensure all questions are properly sequenced. RTI will then conduct a brief pilot test to ensure feasibility of collecting survey data, especially in regards to survey administration and the formatting of the questionnaire. After the first 9 completed responses we will review the data to ensure the quality and completeness. Throughout the data collection period we will continue to review the data that are collected. If deemed necessary, we will contact individual respondents to collect missing data items or to clarify any responses.

3. Methods to Maximize Response Rates and Deal with Nonresponse

We will utilize multiple methods to increase response rate and decrease non-response at the survey administration level as well as item non-response. We will conduct the survey utilizing a web survey. This mode of administration will minimize respondent burden as all PHTC trainees will be familiar with providing feedback via the Internet. Additionally, understanding that trainees are extremely busy, we will work closely with HRSA and PHTCs to provide regular survey reminders via email from RTI or via the HRSA TRAIN LMS.

Regarding item non-response, the questionnaire will be short, no more than 10 minutes, and will mainly consist of close-ended questions which will only require a point-and-click from the respondent to answer the question.

Additionally RTI will review the data collected throughout the data collection period and identify item non-response to determine how it may affect the analyses that will be completed. We will look for trends in item non-response to determine if certain questions are problematic in obtaining data. If deemed necessary, we will attempt to contact respondents to obtain any missing data.

4. Tests of Procedures or Methods to be Undertaken

RTI will conduct a small pretest of fewer than 10 respondents to test the feasibility of the questionnaire administration. No significant changes are anticipated to the questionnaire, only minor wording or formatting changes are anticipated.

5. Individuals Consulted on Statistical Aspects and Individuals Collecting and/or Analyzing Data

The sample design and power calculations were performed by Cynthia B. Augustine (caugustine@rti.org; 919-541-6154). Ms. Augustine will also analyze the data. The data will be collected via the web-tool by Brian Evans (evans@rti.org: 919-541-7366) who designed the data collection tool.

1