0704-xxxx_ssb_12.21.2020

SUPPORTING STATEMENT – PART B
B. COLLECTIONS OF INFORMATION EMPLOYING STATISTICAL METHODS
1. Description of the Activity
The survey will be conducted online. Survey variables will be used to provide a snapshot of the
current and emerging non-medical counseling needs for military children, to better understand
how the current CYB-MFLC program addresses those needs within the school context, and to
identify how the CYB-MFLC program operates within the broader landscape of non-medical
counseling services in the community. Data analysis will be descriptive and will not focus on
specific outcomes.
The Needs Assessment of Child and Youth Non-Medical Counseling will collect information
from the universe of school principals at civilian schools involved with the CYB-MFLC program
(N=900). We are expecting a lower than average response rate (~180 assuming a 20% response
rate). Thus, a census survey is proposed which will allow for an adequate final sample size to
look at important subgroups and thereby increasing the usefulness of the data.
By collecting information on the non-medical counseling needs for all CYB-MFLC programs,
the study will accurately identify the full range of non-medical counseling needs across all CYBMFLC programs. As no data currently exists on many of the potentially important grouping
variables (for example, presence of counseling staff or resources to address nonmedical
counseling needs), no sampling frame could be created that would adequately reflect these
variables.

2. Procedures for the Collection of Information
Describe any of the following if they are used in the collection of information:
a. Statistical methodologies for stratification and sample selection
The Needs Assessment of Child and Youth Non-Medical Counseling survey will be conducted
online. The target population includes school principals associated with the CYB-MFLC
program (N=900). As no data currently exists on many of the potentially important grouping
variables (for example, presence of counseling staff or resources to address nonmedical
counseling needs), no sampling frame could be created that would adequately reflect these
variables
b. Estimation procedures;
The Needs Assessment of Child and Youth Non-Medical Counseling will collect information
from the universe of school principals at civilian schools involved with the CYB-MFLC program
(N=900). We are expecting a lower than average response rate (~180 assuming a 20% response
rate). Thus, a census survey is proposed which will allow for an adequate final sample size to
look at any important subgroups and thereby increasing the usefulness of the data.
c. Degree of accuracy needed for the Purpose discussed in the justification;
1

Survey variables will be used to provide a snapshot of the current and emerging non-medical
counseling needs for military children, to better understand how the current CYB-MFLC
program addresses those needs within the school context, and to identify how the CYB-MFLC
program operates within the broader landscape of non-medical counseling services in the
community. Data analysis will be descriptive and will not focus on specific outcomes.
d. Unusual problems requiring specialized sampling procedures; and
N/A
e. Use of periodic or cyclical data collections to reduce respondent burden.
N/A. Stakeholder survey questions are unique to the CYB-MFLC program. Stakeholder
perceptions of military child needs, CYB-MFLC services, and gaps in non-medical counseling
support for military children have not been assessed in other data collection efforts.
3. Maximization of Response Rates, Non-response, and Reliability
This study will employ several approaches to maximize response rate and reliability.
Introductory email from Military Community and Family Policy (MC&FP). Leadership
from MC&FP will send an email to potential respondents the day before the survey launch to
provide information about the purpose of the survey, and summarize the value of their responses
to MC&FP leadership. This email also lets respondents know that MC&FP has approved of this
data collection effort. A draft of this email is included in Appendix A.
Reminder emails. Reminder emails will be sent 3 days and 2 weeks after survey launch. Drafts
of these email reminders are included in Appendix B.
Development of nonresponse weights. A number of strategies to maximize response rates to the
survey will be used. However, some nonresponse is expected so appropriate statistical
procedures to correct for any potential nonresponse bias will be used. To allow for survey results
to be generalizable to all schools that participate in the CYB-MFLC program, non-response
weights will be designed based on program and school characteristics (e.g., type of school—
elementary/middle/high, size of student body, setting) that will allow for reweighting the sample
of survey respondents to be similar to the population of schools that were invited to participate.
Logistic regression models will be used to predict the propensity of a school principal
participating in the survey and the inverse of the propensity will be used as non-response
weights. Extremely large weights will be trimmed to avoid outliers and influential observations.
These obtained weights will be used throughout the analyses for inference.
4. Tests of Procedures
The survey will be quality reviewed internally once programmed to ensure clarity and accuracy
and to ensure it does not take longer to complete than anticipated.
5. Statistical Consultation and Information Analysis
a. Provide names and telephone number of individual(s) consulted on statistical aspects of the
design.
Lane Burgette ([email protected]) RAND Corporation, 412-683-2300 ext. 4148

2

b. Provide name and organization of person(s) who will actually collect and analyze the
collected information.
Thomas Trail ([email protected]) RAND Corporation, 703-413-1100 ext. 5681
Laurie Martin ([email protected]) RAND Corporation, 703-413-1100 ext. 5083.
Lane Burgette ([email protected]) RAND Corporation, 412-683-2300 ext. 4148.

3