Johnson - OMB Section B

Table of Contents

B. COLLECTION OF INFORMATION EMPLOYING STATISTICAL METHODS

1. Describe the potential respondent universe and any sampling or other respondent

selection method to be used................................................................................................3

2. Describe the procedures for the collection of information.................................................4

3. Describe methods to maximize response rates...................................................................5

4. Describe any tests of procedures or methods to be undertaken..........................................5

5. Provide the name and telephone number of individuals consulted on statistical

aspects of the design...........................................................................................................6

SECTION B

B) Collections of Information Employing Statistical Methods

The National Highway Traffic Safety Administration (NHTSA) is seeking approval for information collection activities as part of its initial evaluation of the implementation and impact of its newly developed Child Pedestrian Safety Curriculum. Information collection activities will take place in Wilson County, NC at two test schools receiving the curriculum and two similar comparison schools where no pedestrian curriculum will be implemented. The two schools receiving the curriculum scheduled to begin on or about April 1, 2014 are Wells Elementary School and Margaret Hearne Elementary School. The two comparison schools are John Jones Elementary School and Winstead Elementary School.

NHTSA is requesting approval to conduct in-class oral surveys (NHTSA Form 1217) to assess knowledge and self-reported behaviors of K-5 students in the test schools pre- and post-curriculum implementation and at the comparison schools during the same time periods. All K-5 students present at the selected schools on the data collection days will be surveyed.

NHTSA is requesting approval of a paper-and-pencil survey of student caregivers (NHTSA Forms 1216 – the English version and 1216-A – the Spanish version) for both the test and comparison schools. These surveys will only take place after curriculum implementation. Every student attending one of the test schools in a household will bring home a survey to be completed by whichever caregiver in the household works on that child’s homework. Surveys will focus on knowledge, attitudes, and opinions relative to child pedestrian safety.

NHTSA is requesting approval of an Internet-based survey of instructional staff at the test and comparison schools (NHTSA Form 1215). These surveys will only take place after curriculum implementation. Surveys will focus on knowledge, attitudes, and opinions relative to child pedestrian safety.

All surveys and survey questions were developed specifically for this information collection to capture information relevant to the curriculum’s evaluation.

B.1 Describe the potential respondent universe and any sampling or other respondent selection method to be used.

The basic sampling frame for the study was limited to four schools (two experimental and two comparison) by the level of resources allocated to the study and the extent of logistics that would be required of the school system and contractor if a larger sample were employed. The cooperating school system selected the two experimental and two comparison schools that they believed were best matched and also were sufficiently dispersed geographically that spillover effects would not be an issue. Demographics for each school are provided in Table 1.

1. Demographics of Test and Control Sites.

School	Condition	Students	% White	% Black	% Asian	% American Indian / Alaska Native	% Hawaiian Native / Pacific Islander	% Two or more races	% Hispanic
Wells	Experimental	489	44%	46%	1%	1%	0%	4%	4%
Margaret Hearne	Experimental	450	19%	50%	0%	0%	0%	2%	30%
John Jones	Comparison	643	36%	48%	4%	0%	0%	3%	8%
Winstead	Comparison	367	6%	68%	0%	0%	0%	2%	23%

Source. www.greatschools.org accessed January 10, 2014.

For the K-5 student survey (NHTSA Form 1217), the respondents will include all K-5 students at both the test and comparison schools who are present on the surveying days (no attempt will be made to recover students absent from school on the data collection days). Based on average absentee rates at the schools, the response rate is expected to be 95%. Before implementation of the curriculum, schools will send home a customary parent notification sheet that will give caregivers the option of having their student opt out of the program. The participating school system indicates that very few, if any, parents opt out of this type of educational programming.

The paper-and-pencil survey of caregivers in English (NHTSA Form 1216) and Spanish (NHTSA Form 1216-A) at the test and comparison schools will be sent home in the daily homework folders of all K-5 students and will be self-administered by one parent/caregiver per student. If more than one caregiver is present, the caregivers will self-select who completes the survey guided by criteria in the survey preamble/instructions. If there is more than one student per household, a survey will be sent in each student’s folder for completion by whoever reviews that folder’s contents. This is the preferred approach since the most salient research questions relate to the individual child bringing home the folder. The survey will be returned in the students’ homework folders. The school district employs this procedure for surveys and other two-way communications throughout the school year.

A link to the Internet-based survey of all instructional staffs at the test and comparison schools (NHTSA Form 1215) will be sent out using the schools’ internal email system. One week after the initial email, a follow-up email will be sent out to remind any non-responders about the survey.

B.1.1 Total Sampling Needs

Participants will be asked to complete surveys as detailed in Table 2 below. Estimated sample sizes are provided based on the number of interviews to be attempted or surveys distributed. Actual response rates will likely vary with student response rate being the highest because of the method of delivery. In order to reduce burden on parents/caregivers and general teaching staff, the self-administered questionnaires will only be administered after the curriculum implementation has taken place. Each student’s household will receive a single questionnaire to be completed by one primary caregiver selected at the discretion of the members of the household based on the survey instructions. All surveys and observations of students will take place before and after implementation. Data collection at the comparison sites will follow the same schedule.

2. Interview/Survey Approach, Time to Complete, and Sample Size by Group

Group	Pre-implementation	Post-implementation	Total Hour Burden Estimate
Parents/Caregivers (NHTSA Forms 1216 and 1216-A)	N/A	Self-administered questionnaire; 5 minutes to complete; Sample = Primary caregivers of all enrolled K-5 students at 4 schools (N = 2,000)	166.67 hours
All Instructional Staff at Schools (NHTSA Form 1215)	N/A	Self-administered questionnaire; 5 minutes to complete; Sample = All K-5 instructors in 4 schools (N = 200)	16.67 hours
Participants (students) (NHTSA Form 1217)	Fully structured oral survey; 5 minutes to complete; Sample is all students K-5 in 4 schools (N = 2,000)	Fully structured oral survey repeated; 5 minutes to complete; Sample is all students K-5 in 4 schools (N = 2,000)	333.33 hours

In sum, NHTSA proposes to interview or survey up to 4,200 unique participants (2,000 students included in this total will be surveyed twice) over the life of the project and estimates a burden of 516.67 total hours.

B.2. Describe the procedures for the collection of information.

For the K-5 student survey, Parent Teacher Organization (PTO) members from the chosen schools will be trained by the researchers and serve as surveyors. Classroom teachers/instructors will ask the students to talk to the surveyor and answer their questions for the 5-minute survey as if it was part of class activities. Only students present on the surveying day will be assessed. No make-ups will occur. Students will be surveyed in a place (e.g., corner of classroom, hallway) where other students or staff cannot hear their responses. Surveyors will read each question aloud and ask for a verbal response to a question or a verbal or display of actions (e.g., display of left hand) for questions concerning proper pedestrian behaviors. Surveyors will record the student responses along with the student’s grade, sex, race, and whether of Hispanic origin using a pen-and-paper recording form. The same procedure will be used before and after curriculum implementation for both the test and comparison schools.

The paper-and-pencil survey of caregivers will be sent home in all K-5 students’ homework folders and will be self-administered by one parent/caregiver per student. If more than one caregiver is present, the caregivers will self-select who completes the survey. The survey will be returned in the students’ homework folders. Researchers will collect the completed surveys from each classroom. The procedure will only be conducted after curriculum implementation.

A link (“click through”) to the Internet-based survey of all instructors and administrators will be sent out using the school district’s internal email system. Clicking on the link will take the respondent to a secure website where items will be presented. One week after the initial email, a follow-up email will be sent out to the entire population to remind any non-responders about the survey. No record will be maintained of who does and does not respond. The procedure will only be conducted after curriculum implementation.

B.3) Describe methods to maximize response rates and to deal with issues of non-response.

The use of PTO as surveyors ensures they are familiar with dealing with children in grades K-5, and they will be known to many of the children. In addition, the classroom instructor will be asking the students to talk to the surveyors as if the activity is part of class which should lead to high participation rates. If a student refuses to participate, the surveyor will record the grade and sex of the student and note on the survey that the individual was not responsive.

For caregiver surveys of this type, the school system provides rewards to classes that achieve certain response rates (e.g., 80% of caregivers respond). These rewards may include, but are not limited to, ice-cream parties, pizza parties, or extra recess play time for the class. The study will reimburse the school system for any expenditure for rewards associated with the conduct of the study. Researchers will know the total number of students enrolled in each grade and will be able to determine how many people did not respond to the survey.

For instructional staff surveys such as the one proposed here, the school system provides the chance for participants to win a prize for participation. The participant voluntarily provides his/her name when completing the survey and is entered into a drawing for the prize which is usually some form of school supplies or a gift card for a local establishment. The participant names are not linked to the survey responses, and the prizes are delivered by the school district through their internal mail systems. The study will reimburse the school system for any expenditure for prizes associated with the conduct of the study. Researchers will know the total number of staff members at each school and will be able to determine how many people did not respond to the survey.

B.4) Describe any tests of procedures or methods to be undertaken.

The oral survey of students will be pilot tested after surveyor training to ensure the surveyors are following protocols and to determine if any modifications to the materials are needed to acquire the desired information. Researchers will first serve as the participants. Once researchers are satisfied that the surveyors understand the process, each surveyor will practice with a small sample of two to three students. The protocols will be revised as needed and the pilot study process conducted again if warranted.

The instructional staff survey will undergo testing to ensure the Internet-based approach is working properly. The link to be provided in the blast email will be tested to make sure it is active and functioning as desired.

The parent survey will be pilot tested with a small number of adults and revised as necessary based on inputs from the participants and observations of researchers.

B.5) Provide the name and telephone number of individuals consulted on statistical aspects of the design.

The following individuals have reviewed technical and statistical aspects of procedures that will be used to conduct the information collection activities:

Richard Blomberg Dunlap and Associates, Inc.

(203) 323-8464 110 Lenox Avenue

Stamford, CT 06906

Dennis Thomas, Ph.D. Dunlap and Associates, Inc.

(203) 215-2948 110 Lenox Avenue

Stamford, CT 06906

Kristie Johnson, Ph.D. Office of Behavioral Safety Research (NTI-131)

(202) 366-2755 National Highway Traffic Safety Administration

1200 New Jersey Ave SE, W46-496

Washington DC, 20590

Richard Compton, Ph.D. Office of Behavioral Safety Research (NTI-130)

(202) 366-2699 National Highway Traffic Safety Administration

1200 New Jersey Ave SE, W44-304

Washington DC, 20590

1