ss hbcu part b

Assessment of Historically Black Colleges and Universities campus and college based activities to eliminate Health Disparities

Part B

Description of statistical methods

1. Respondent universe and sampling methods

The web-based HBCU Health Disparities Survey instrument will be sent to four administrators (including the Presidents, Deans of Faculty, and Deans of Students) at each of the 105 HBCUs located in the United States. Through web-based searches, we have also identified 100 Directors of Health Disparities Research Centers and Initiatives across these 105 HBCUs that will also be invited to participate in this study. Because we will engage the entire population of HBCUs, it will not be necessary to develop a detailed sampling plan or sample weights for this instrument. Tab F provides a list of the institutions to be surveyed.

2. Information collection procedures

Our data collection for the full HBCU community will begin following OMB review and approval and continue until all of the institutions’ data has been collected. We estimate it will take 2-3 months to complete this data collection. Respondents will be able to contact the survey administrator (an AIR employee) for assistance in accessing the survey and/or completing the survey. We plan to synthesize the information gathered into a comprehensive database of the institutions and their activities which can be used by Department staff.

3. Methods to maximize response rate

The expected response rate for the survey will be at least 85%. Several general procedures will be used to maximize response rates for the survey used in this study. First, an invitational/introductory email will go out to each respondent explaining the study and requesting respondents’ cooperation in completing the survey. Survey response and completion rates will be monitored closely, and individuals who have not completed their surveys within two (2) weeks after start-up will be reminded via e-mail and a follow-up phone call; a second reminder email and call will go out to respondents who have not completed the survey one month after startup.

The respondents will be asked if they received the survey; and any questions or problems in completing the survey can be discussed during our call to them, or another time that is convenient to the respondent. Staff will reiterate the importance of the respondent’s input.

Finally, if necessary, to achieve the desired response rate, AIR staff will conduct telephone interviews with any non-responding participants.

If response rates do not exceed 85%, project staff will schedule and conduct telephone interviews with the pertinent university administrators. The following protocol will be used to conduct follow-up telephone interviews:

• Project staff will contact the office of the HBCU Presidents and Deans to schedule a convenient time for the interview.

• Project staff will mail a copy of the survey to the respondents a week prior to the scheduled telephone interview.

• Telephone interviews will be limited to one hour per interview.

One issue of concern in survey research is non-response. Non-response is particularly important when survey results are used for making population estimates or when results are generalized. This survey; however, has a non-probability sample in which respondents are selected with intent due to the nature of the positions which they occupy at the HBCUs. The survey is not designed for the purpose of making estimates or generalizing to a larger population; but was designed for exploratory purposes in order to obtain descriptive information that has not been collected to date. In light of this design, a formal non-response analysis is not planned. The overall response rate will be calculated, and if there are known characteristics for those institutions that do not respond, these characteristics will be stated in the summary report as a potential bias and limitation of the data.

4. Test of procedures

A pilot test of the survey instrument was conducted in September-November 2005 for two purposes – to ensure that the web-based instrument and procedures work effectively, and to verify preliminary estimates of the respondent burden. For the survey pilot, researchers selected individuals that have experience reporting information of the type included in the pilot survey. A total of four (4) individuals participated in the pilot. The sample included directors of federally-funded health disparities centers and administrators at institutions of higher education that are not HBCUs. Pilot test participants were asked to complete a web-based version of the survey. At the end of the survey, respondents were asked to provide any comments or suggestions about the survey that they completed. The respondents were asked to identify:

• questions or response options that are difficult to understand;

• questions in which none of the response options is an accurate description of their circumstance;

• terms that are not defined that should be defined;

• and questions for which the information requested is unavailable.

The web-based survey also included underlying code that allowed the research team to document how the respondent navigated through the survey and the length of time required to complete survey questions. Pilot participants were also asked to comment on the functionality and ease of navigation of the web-based survey.

Pilot respondents’ ratings of the “ease” and “usability” of the survey, as well as analysis of the navigational methods used by respondents as they completed the survey, also suggests that they had no difficulties using the survey. For example, there were no instances where respondents had to “scroll back” to prior questions and no “time outs” while respondents were trying to complete the survey.

Revisions to clarify language within the survey were made based on information provided by pilot test participants.

5. Statistical Consultants

This project is being conducted under contract to the U.S. Department of Health & Human Services Health Resources & Services Administration by the McKenzie Group, which is now a wholly owned subsidiary of the American Institutes for Research (AIR). The study design was developed by Dr. Carmen Arroyo who will oversee the data collection and analysis.