icandoityoucandoit

B. STATISTICAL METHODS

1. Respondent Universe and Sampling Methods

The sample will be a non-representative sample of children and youth with physical,developmental/cognitive, sensory, and/or behavioral (emotional and/or substance abuse) disabilities. Results from the longitudinal pre-post study will not be used to generalize to the universe of children and youth with disabilities in the United States. Therefore, sample size and power are of secondary importance. No prediction that necessitates a requisite amount of power is being made.

2. Procedures for the Collection of Information

Recruitment of participants/mentees will be a two-stage process. In the first stage, cooperating agencies that agree to implement the program will be recruited. Potential cooperating agencies will be informed of several criteria that they must meet in order to participate, including:

• Being able to recruit the required number of participants/mentees (120);

• Being able to recruit sufficient numbers of mentors; and

• Agreeing to not have participants enrolled in any other health promotion intervention during the project period.

Once cooperating agencies have been identified, agency coordinators will be responsible for recruiting participants.

A. Statistical Methodology For Stratification And Sample Selection

The sample will be a convenience sample. Stratification will not be used.

B. Estimation Procedure

The estimated number of respondents completing the registration form is 1,320. Attrition throughout the program is expected, with an estimated 1120 completing the pre-test survey, 1020 completing the post-test survey and 920 completing the second post-test survey. The estimated number of mentors completing the registration form and post assessment is 900. The estimated number of agency coordinators completing the survey is 13.

C. Degree of Accuracy Needed

The web-based survey was designed to meet the non-inferential goals of this study. The use of web-based surveys was selected to increase the accuracy of the data by: (a) providing increased privacy conducive to accurate and honest reporting by respondents, (b) eliminating the need for hardcopy records, (c) providing easy back-up to ensure no loss of data, and (d) reducing respondent error related to skip patterns.

Because the data will be entered directly by respondents, the only data entry errors will be on the part of the participants. Edit checks will be built into items as needed to ensure that respondents cannot give out-of-range or other inappropriate responses. Participants will be given identification numbers to enter in order to gain access to the survey, to ensure that participants do not complete the survey more than once.

D. Unusual Problems Requiring Specialized Sampling Procedures:

No unusual problems requiring specialized sampling procedures are anticipated.

E. Any use of less frequent than annual data collection to reduce burden

All questions have been limited to those considered essential to evaluate the efficacy of the program. This study is an ad hoc data collection (i.e., a one-time study), and therefore the data are collected less frequently than annually.

3. Methods to Maximize Response Rates and Deal with Nonresponse

The OD’s ability to gain the cooperation of agency coordinators is the key to the success of this endeavor. Through an orientation program for them, the use of small non-monetary incentives for participants/mentees, and the encouragement of mentors, the OD believes that the response rate will be high. Agency coordinators will be provided with the code numbers of participants who have not completed one of the three surveys, so they can be contacted and urged to complete the study. These contacts will be within the guidelines for the protection of human subjects to avoid even the appearance of coercion of respondents. Respondents will always be reminded that they are able to not respond to one or more items or an entire survey if they choose. Contacts with agency coordinators will be structured to simply remind them that they have not responded, and there is still time to do so.

4. Test of Procedures or Methods to be Undertaken

The OD plans to thoroughly test the web-based survey with a group of nine or fewer children and youth with disabilities prior to data collection. The complete study protocol, including the security of the transmission of data, the receipt system and other aspects of the program will be tested internally before the start of data collection.

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

JoAnn Thierry, Ph.D.

National Center on Birth Defects and Developmental Disabilities (NCBDDD)

Centers for Disease Control and Prevention

1600 Clifton Road, NE, MS: E-88

Atlanta, GA 30333

[email protected]

404-498-3022

3