Nida-b OMB

Supporting Statement for Monitoring and Evaluation of the National Institute on Drug Abuse NIDA Goes Back to School National Dissemination Campaign

Section B

Introduction

B.1. Respondent Universe and Sampling Methods.

Random samples will be drawn from the two groups of requestors – school personnel, and community leaders, living/working in the United States. Regarding accordance with the Draft Provisional Guidance on the implementation of the 1997 Standards for Federal Data on Race and Ethnicity, the random samples will be representative of the requestor population. And since no demographic information of requestors will be available at the time of sampling, oversampling for representative race and ethnicity of the general population will not be possible. However, the requestors for the materials are from all over the U.S., and include a variety of socio-economic areas, so that race and ethnicity will be dependent on the geographic location of the survey respondent. Representation from the different socio-economic regions of the country is important for this study, because the purpose is to learn of administrative and environmental barriers to use of the materials, and these are more likely barriers to use than person characteristics of race and ethnicity per se. Representativeness of the sample on the basis of the demographic information from survey participants will be assessed and reported from the information collected. The anticipated response rate is approximately 80 % to 85% (complete responses and/or participation). This anticipated response rate is based on the high rate of email requests for the NGBTS materials that are requested voluntarily. In addition, requestors in the sample will be contacted up to two times after initial contact, to re-solicit their response . Specifically, requestors will initially be solicited for participation by email, using the contact information available from the requestor form. The request will be an email sent by the Project Officer or designee,. Subsequently up to two emails to non-respondents will be sent by the Contractor on behalf of the Project Officer.

Table 4. Target Enrollment Table.

 

		School	Community	Total
	2003	200	200	400
	2004	200	200	400
	2005	200	200	400
	Total	600	600	1,200

B.2. Procedures for the Collection of Information.

For the purpose of the proposed evaluation which is program improvement (as described in the justification), response rates of 40% to 60% should be sufficient. There are no unusual problems requiring specialized sampling procedures. Each requestor solicited for participation will be assigned an ID number. Each requestor will be provided with the password for the survey, and directed to the survey websites set up for school personnel and for community leaders. Only participants provided with the password needed for the relevant survey will have access to the instruments of the study. Quality control measures will include a brief review of the survey responses as they come in to obtain some sense of the instrument and how effective it is. Tracking of respondents with the list manager tool will be conducted, and with checks prior to and during the study to ensure there are not duplicate submissions from respondents. Survey participants will be informed of a toll-free “help line” to call for assistance with the survey site. The help line is monitored during normal business hours.

At the end of the last wave of the survey the percentage of the sample of requestors who have responded, will be calculated. This percentage will be calculated for each year and a comparison among the study years will be calculated for an idea of non-response due to possible memory recall. It is also anticipated that the non-response propensity would be greater for those requestors who do not use the materials (e.g. those choosing the last two categories in the use question), and the percentage of non-responses among non-users and low-users of the materials will be calculated. The association of patterns of non-responses among non-users and low-users with institutional or community characteristics obtained from the sampling frame may be conducted. It might also be possible to do a quick survey of 9 non-responders and interview them by telephone, for information on their use of the materials, and from determining the confidence limits of the true proportion of ‘non-users’ we could obtain a better estimate of the extent of non-users in the sample. This would be highly dependent on availability of requestors to participate, since resources from the study would not be available for incentives.

B.2.1. Statistical Methodology for Stratification and Sample Selection

There is no stratification in this collection. Two requestor population groups from three recent years were considered relevant for this collection. Sampling will be without stratification.

The population of requestors in each year of the study period will be stratified to obtain school personnel requestors and community leader requestors. Random sampling of these groups conducted, for each relevant year.

B.2.2. Estimation Procedure

There are no estimation procedures in this collection.

B.2.3. Degree of Accuracy Needed for the Purpose Described in the Justification

Random sampling of the requestor population will be used.  Participation in the survey is voluntary.  The anticipated response rate is 80-85%, and is planned to be achieved with re-requesting participation of respondents up to two times each.

B.2.4. Unusual Problems Requiring Specialized Sampling Procedures

No unusual problems are anticipated. No specialized sampling procedures will be used.

B.2.5. Use of Periodic (Less Frequent Than Annual) Data Collection Cycles

No periodic annual collection of information is planned for this collection.

B.3. Methods to Maximize Response Rates and Deal With Nonresponse.

This study is a cross-sectional observational study, and analyses will be descriptive. The study purpose is to obtain information about use of the NGBTS materials, and information on the barriers to using the materials. Response rates for the study will be calculated as the number of respondents who provide useable responses divided by the number of requestors selected in sampling. For the surveys, 200 school personnel requestors from each study year, and 200 community leader requestors from each study year, will be enrolled, with the expectation that a total of 1020 to 960 of them will yield usable questionnaires, for the preferred completion rate of 80%-85%.

It is anticipated that response rates for the information will be maximized without remunerating participants for their time, (see Section A.9). In the past they have availed of the free resources provided by NIDA, and some sense of social reciprocity is anticipated. If those who were sent information for participation have offered no response after an allowable time for receipt of the NIDA request, they will be e-mailed twice at one week intervals, and asked again to participate in the survey. Potential participants who refuse to participate or who leave the study at any point will be allowed to do so without penalty or jeopardy.

B.4. Test of Procedures or Methods to Be Undertaken

In the proposed evaluation, participants will complete the information collection questionnaire. The instrument will have been revised and finalized on results and recommendations from the pilot test of it. Testing of the methodology is not considered necessary at this time.

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

The NIDA Project Officer is Brian Marquis. The Project Officer, the NIDA Evaluator, one member of the pilot test participants, and the Contractor, will review the final report, including data analysis. The NIDA Evaluator and the Project Officer will co-direct the project. Table 5 below shows the individuals who will be consulted on conduct of the project, data collection, and data analysis, or who will collect and analyze data:

Table 5. Consultants

Name	Title	Affiliation	Statistical Role	Phone Number
Bohne Silber, Ph.D	Contractor, Principal	Silber and Associates	Data Collection & Analysis	410-531-2121, ext 1#
Genevieve deAlmeida-Morris, Ph.D, MPH	Evaluator, Planning & Evaluation Officer	NIDA	Data Collection & Analysis	301-594-6802
Cathrine Sasek, Ph.D	NIDA Science Education Officer	NIDA	Consulting on Data Collection & Analysis	301-594-6312

 

In addition, research and web technical assistants working on the Contractor’s staff will also assist with data collection.

References

Dillman, D.A. (2000) Mail and Internet Surveys: The Tailored Design Method. New York. Wiley

Dillman, D.A , Christian, L.M. (2005) Survey Mode as a Source of Instability in Responses across Surveys. Field Methods, 17(30) Retrieved on 02/06/2007, from http://fmx.sagepub.com/

Seltiz, C., Wrightsman, L.S., Cook, S.W. (1976) Research Methods in Social Relations, 3thEdition. New York. Holt, Rinehart, & Winston