This section of the report identifies the research methodology used in meeting the project objectives. The methodology included: (1) conducting a workshop to identify the most important CRS misuse measures for study; (2) establishing criteria for site selection; (3) selecting States and State site coordinators (SSCs) to oversee the data collection; (4) conducting a train-the-trainer workshop with SSCs and their field site managers (FSMs); (5) developing data collection instruments; (6) having project methods reviewed and approved by an institutional review board (IRB); (7) recruiting qualified field personnel; (8) conducting training for field observers and greeters; (9) collecting field data over a 2 to 3 month period; (10) identifying socio-economic and demographic characteristics of field sites; (11) performing data analysis activities; and (12) providing documentation that summarizes results and makes recommendations for future research and programs.

2.1 WORKSHOP TO IDENTIFY CRS MISUSE MEASURES (BASED ON INJURY SEVERITY)

The first key task was to conduct a workshop with child passenger safety experts from the fields of biomechanics, injury prevention, public health, CRS manufacturing, and program implementation to prioritize CRS misuse characteristics according to their potential for resulting in injury to the child during a crash.

The workshop was held in Washington D.C. on March 12, 2002; and was attended by the contractor (TransAnalytics) and subcontractor (Children’s Hospital of Philadelphia), expert panelists, and NHTSA staff. Expert panelists in attendance were: Paul Butler (Ford Automobile Safety Office), David Campbell (David Campbell and Associates), Karen DiCapua (National Safe Kids Campaign), Dr. Susan Ferguson (Insurance Institute for Highway Safety), William Hall (Highway Safety Research Center, University of North Carolina), Lorrie Walker (Florida Traffic Safety Resource Center, Florida Atlantic University), Kathleen Weber (University of Michigan Medical School, retired), and Dr. Narayan Yoganandan (Medical College of Wisconsin). Kelly Orzechowski (Crash Injury Research and Engineering Network Project, Children’s National Medical Center) was also in attendance as a visitor. The Principal Investigator (L. Decina) was the moderator and the session was recorded and transcribed.

The workshop agenda included: opening remarks; discussions on types of child injury in crashes and their severity; identification of types of CRS misuse and their relationship to serious injury; identification of most important CRS measures to include in data collection; CRS misuse definitions; and summary and concluding remarks.

Not surprisingly, opinion was strong on any misuse that results in excursion of the child from the CRS. (Excursion defined as the distance traveled by an occupant or test dummy in the direction of impact during a crash.) Loose CRS installations in the vehicle and loose CRS harness straps on the child cause the greatest injuries, as does installing a seat in the wrong direction for infants, or placing infants in front of an air bag.

The workshop attendees recommended the following critical CRS misuse measures as important to study in the field observations:

• Age and weight appropriateness of CRS.
• Direction of CRS.
• Placement of CRS in relation to air bags.
• Installation and secureness of CRS to the vehicle seat (tight SB).
• Secureness/tightness of harness straps and crotch strap of the CRS.
• Use of locking clip for certain vehicle safety belts.
• Fit of vehicle SBs across child in belt-positioning booster seat.
• Defective or broken CRS elements.

Results of the workshop were used in the development of the first draft of the data collection instrument, and guided procedures presented to State site coordinators and their field managers at the train-the-trainer session. (See Appendix A for definitions of correct use.)

2.2 CRITERIA FOR SITE SELECTION

The next key task was to select six States to conduct the observations of CRS misuse among the public. The selection criteria included the following:

• Diverse geographic regions across the country.
  
• Coverage of urban, suburban, and rural areas.
  
• Diverse socio-demographic and economic characteristics of communities across geographic regions.
  
• State site coordinators (SSCs) with extensive child passenger safety knowledge and experience.
  
• SSCs with experience in managing similar studies.
  
• SSCs who represent child injury prevention organizations.
  
• SSCs with AAA CPS certification as instructors/technicians.
  
• SSCs with established contacts in their communities to secure observation sites
  
• SSCs experienced in training, recruiting, and managing field observers.

2.3 STATE AND GEOGRAPHIC REGION SITE SELECTION

A candidate list of States and study site areas which met the above criteria was developed and submitted to NHTSA. Upon NHTSA approval of the SSCs and their State sites, contractual agreements were made with the SSCs and their organizations. The SSCs, their affiliations, and the site regions for study are listed below.

Arizona - Nancy R. Avery, Inspector/Public Education Officer/Firefighter, City of Tucson Fire Department, Tucson SAFE KIDS, Children Always Ride Restrained (CARR) Program. The study area selected for Arizona was the city of Tucson.

Florida – Lorrie Walker, Program Coordinator/Administrator, Florida Atlantic University, Christine E. Lynn College of Nursing, Florida CPS Program and Resource Center. The study areas selected for Florida were Boca Raton, Fort Myers, Fort Pierce, and Miami.

Mississippi – Kay Brodbeck, Project Director for Mississippi Safety Services, and Cynthia Huff, Mississippi SAFE KIDS. The study area selected for Mississippi was the city and surrounding area of Jackson.

Missouri – Catherine Metzger, Co-coordinator SAFE KIDS St. Louis, Cardinal Glennon Children’s Hospital. The study area selected for Missouri was the city of St. Louis and the surrounding area.

Pennsylvania – Juli McGreevy, Consultant and Robert Mott, Coordinator, South Central PA Highway Safety. The study areas selected for Pennsylvania were the areas including and surrounding Carlisle and Harrisburg.

Washington – Kathy P Kruger, Executive Director, Washington State Safety Restraint Coalition. The study areas selected for Washington were Seattle, Tacoma, and their surrounding areas.

The SSCs and their staff were from three SAFE KIDS organizations, two Statewide CPS programs, one regional comprehensive highway safety program, and one private consulting firm. Six different regions of the United States were represented in the study.

Child restraint and safety belt laws for the six States at the time of data collection (September 2002) are shown in Tables 1 and 2, respectively.

Table 1. Child restraint laws of States in the study (as of September 2002).

Source: (IIHS, 2002 and Safety Restraint Coalition, 2002)

Table 2. Safety belt laws of States in the study (as of September 2002).

Source: (IIHS, 2002 and Safety Restraint Coalition, 2002)

2.4 TRAIN-THE-TRAINER WORKSHOP

The purpose of the workshop was to provide a forum with the SSCs and their field site managers (FSMs) to discuss the field operations. All participants in attendance were AAA CPS-certified instructors and technicians. A workshop notebook was given to each participant. This notebook included: an agenda; a summary report on previous CRS misuse observation studies; examples of data collection forms; guidelines for locating and recruiting sites; a training manual for data collection; the most recent information on CRS use and proper use guidelines; and a list of participants.

The workshop was conducted in August 2002 at the Philadelphia Airport Marriott Hotel in Philadelphia, PA. The first day of the workshop focused on establishing the CRS correct use definitions. Topics covered were proper age and weight requirements for each CRS type; proper CRS installation; and proper placement of child in the CRS. Guidelines were also established in the areas of harness strap tightness, harness connection, and harness retainer clip positioning. Appendix A, “CRS Correct Use Definitions” identifies the guidelines used in the study during field operations.

The second day of the workshop focused on creating the final version of the data collection forms. The Principal Investigator (PI) presented a draft of the greeter and observer forms to the participants. The drafts were based on previous CRS misuse observation forms used in the field by the PI, as well as forms used at CRS inspection station clinics and by other researchers. Input from newly established guidelines for correct CRS use was used to refine the data collection instruments.

The remainder of the workshop focused on the following: instructions for conducting the field observations and managing field crews, training techniques, recruitment of field observers (AAA CPS-certified instructors/technicians only) and greeters, suggested techniques in gaining site permission and community cooperation, and administrative issues (i.e., contractual agreements, time sheets, staff reimbursement, scheduling for data collection).

2.5 DATA COLLECTION INSTRUMENTS

One greeter (contact) form was completed for each target vehicle. The form included category boxes for recording the following data: form identification number; greeter and observer initials; date of contact/observation; State and site identification; vehicle make, model, and year; location of passive protection devices in the vehicle; driver safety belt use; and vehicle seating position and age and weight of target children.

One observation form was completed for each sampled child. The form included category boxes for recording the following data: form identification number; observer initials; date of contact/observation; State and site identification; restraint type (i.e., CRS, SB, or unrestrained); child seating position in vehicle and vehicle restraint type (including LATCH system); CRS misuse categories for each type of CRS and seat direction; and SB misuse categories for lap only, shoulder only, or lap/shoulder SB systems.

CRS types for which misuse data were collected were categorized as follows:

• Rear-facing seats (infant, convertible, other).
  
• Forward-facing seats (convertible, forward-facing only, integrated, other).
  
• Booster seats (belt-positioning, integrated, other).
  
• Other (shield booster, Laptop car seat, other).

Appendices B and C provide a copy of the greeter (contact) and observation forms, respectively.

2.6 INSTITUTIONAL REVIEW BOARD

An institutional review board (IRB) panel was used to review and formally approve the data collection plan of the project. Chesapeake Research Review Inc. (CRRI) from Columbia, MD was contracted to manage the IRB. They have in place a U.S. Department of Health and Human Services-approved and registered IRB panel to review selected types of research. In August of 2002, the IRB panel reviewed the data collection plan and field observation protocols. The panel approved the protocol for the site plan and determined that the research study met the criteria found in the pediatric risk category described in 45 CFR 46.404: “Research/clinical investigations not involving greater than minimal risk.” CRRI sent the confirmation letter describing the IRB acceptance of the data collection plan to NHTSA. Based on this letter, NHTSA then approved the plan to proceed with the fieldwork.

2.7 FIELD PERSONNEL

Each State site coordinator (SSC) had overall responsibility for recruiting field site managers (FSMs), field observers and greeters. The FSMs were selected during the initial negotiation activities with the SSCs and the selection of their States. SSCs recruited FSMs who were AAA certified CPS instructors/technicians and were already working with their organization on current CPS projects and programs, including local inspection station events.

Field observers were selected and recruited from the National and State lists of AAA certified CPS instructors and technicians who lived in the geographic areas near the SSCs and their FSMs. The SSCs and FSMs contacted people from these lists to inquire about their interest to participate in the study as paid field observers/data collectors.

SSCs recruited greeters by placing ads in local newspapers. In many cases, the candidates for field observers and greeters were active in the CPS field and were already participating in local CPS events. Efforts were also made to hire multi-lingual greeters to accommodate Spanish speaking drivers and to assist in gaining permission to make observations in their vehicles.

2.8 TRAINING

Training for field observers and greeters was conducted at each State site. SSCs followed guidelines established at the train-the-trainer workshop. All data collectors were given a training manual and were given classroom and field instruction. Classroom sessions covered the following topics: CRS misuse measures; observation and recording techniques; and protocols for greeters and observers when interacting with target drivers.

Classroom sessions were followed up with practice trials in parking lots using drivers and young children in mock situations. Various restraint system configurations were included in the practice. After a day of field practice, data collectors were taken to shopping centers to conduct real fieldwork. They were closely supervised by the training staff for at least two days or until staff was comfortable with their data collection activity.

2.9 DATA COLLECTION

Data were collected by teams of two consisting of a greeter and an observer. The greeter’s responsibilities included identifying the target vehicle entering the designated area; stopping the target driver; requesting permission to conduct the “child safety” observations; and assisting with data collection. The observer’s responsibilities included entering the vehicle and conducting the observational tasks necessary to record the type of restraint use and CRS misuse.

Field procedures used to collect CRS misuse data were:

• Select a target vehicle entering the site and approach the driver.
  
• Identify oneself, briefly explain the purpose of the study (including informing the driver that the children would not be removed from their CRSs), and request permission to conduct observation.
  
• Upon receipt of permission, direct driver to designated safety zone.
  
• Ask driver about ages and weights of target children; make CRS misuse observations; and record findings on form.
  
• Upon completion of observation, thank driver.
  
• Review what was observed and recorded.
  
• Move back into position to wait for next vehicle.

Each site had a field site manager (FSM) responsible for overseeing the field operation. Duties included: observing techniques used by greeters and observers; supplying pre-numbered forms; collecting the data forms; managing staff scheduling; collecting and checking timesheets; and reporting to the SSCs. In many cases, FSMs also participated as observers collecting data. SSCs and FSMs also checked for consistency, missing data, incorrect coding patterns, and other miscellaneous items. Questions about data were brought to the attention of the data collectors. Data were sent to the Principal Investigator on a regular basis.

Data were collected from September to November 2002 in four of the States (AZ, MO, PA, and WA). Data were collected in October to December 2002 in Mississippi; and data were collected in November 2002 to January 2003 in Florida. Data collectors thanked the parents for their participation and provided information on child passenger safety.

2.10 SITE CHARACTERISTICS

Each SSC used knowledge of the local area and personal contact with community and business representatives to identify potential sites for the study. Sites were selected based on the following criteria:

• Large volume of target vehicles (drivers with young children) visiting the site.
  
• Limited number of entrances and exits (if possible) to the site.
  
• Adequate visibility and space for safely conducting the initial interaction with the driver and subsequent observations and data collection in the parked vehicle.
  
• Permission from site proprietors to use site.

Urban, suburban, and rural sites spanning diverse socio-economic and demographic characteristics were used. A variety of community, health, retail and other commercial locations were included, such as: child care centers; discount stores (e.g., Sam’s Club, Super Kmart, and WalMart); fast-food restaurants (e.g., McDonalds); hospital and pediatric centers; shopping centers and malls; stores specializing in infant and children’s merchandise (e.g., Babies R Us, Toys R Us); libraries and churches; entertainment complexes; and commuter parking lots (e.g., ferry docks). Special events, safety festivals, and holiday gatherings for young children were also used in the study.

Each State used a broad selection of sites, generally located in one geographical area of the State (i.e., Southcentral Arizona; Southcentral Florida; Central Mississippi; Eastcentral Missouri; Central Pennsylvania; and Westcentral Washington.)

In Arizona, 11 sites were used in the study. These sites were all located in Pima County and within the city limits of Tucson. The sites were located across several areas of the city. A diverse group of socio-economic and ethnic communities was included in the field observations. Many sites were set up with multilingual (English/Spanish) greeters. Sites with a large Mexican population were included.

In Florida, seven sites were used in the data collection effort. These sites were all located in the southern part of the State in four cities (Boca Raton, Fort Myers, Fort Pierce, and Miami). These cities are located in four counties (Lee, Miami, Palm Beach, and Saint Lucie). A wide range of socio-economic and ethnic communities was included in the field observations. Many sites were set up with multilingual (English/Spanish) greeters. Sites with large Caribbean and African-American populations were included.

In Mississippi, four sites were used in field observations. These sites were all located in central Mississippi in four municipalities located in two counties (Hinds and Rankin). Jackson was the largest city among the sites. Diverse socio-economic and ethnic communities were included in field observations. Sites with a large African-American population were included.

In Missouri, 15 sites were used in data collection. These sites were located in the east central section of the State. Sites were in nine cities/municipalities, located in three counties (Jefferson, Saint Charles, and Saint Louis). Saint Louis was the largest of the cities. Similar to the other States, a wide range of socio-economic and ethnic communities was included in the field observations. Sites with large Latino and African-American populations were included.

In Pennsylvania, eight sites were used in data collection. These sites were located in central Pennsylvania, across several cities (Carlisle, Harrisburg, and Mechanicsburg), municipalities, and townships. All of the sites were in Cumberland and Dauphin counties. A diverse range of socio-economic communities was included in the data collection effort. Sites with a large African-American population were included.

In Washington, 27 sites were used in data collection. Most of the data were collected at 8 sites. These sites were located in the central western part of the State, across several cities (Bellevue, Seattle, Tacoma) and municipalities. The sites were in four counties (King, Kitsap, Pierce, and Snohomish). A diverse range of socio-economic and ethnic communities was included in the field observations. Sites with large Asian and Pacific-Island populations were included.

2.11 DATA ANALYSIS

Data were checked in the field by the field observers, as well as by the project management team at the site. Inconsistencies and errors in recording information were resolved with data collectors. Data were then keyed into a Microsoft Access 2002 database by the data entry staff and provided to the research analyst.

Descriptive summaries of the data were prepared. Summary tables were developed for project briefing purposes and inclusion in the final report.

Data from the U.S. Bureau of Labor Statistics and the U.S. Census Bureau were used to identify county-level socio-economic and demographic characteristics of sites. Characteristics of interest included population, race (percent white), age (percent children less than age 5), household size (persons per household), household median income, and unemployment rate of each State’s study areas at the county level. The data were used to show characteristics of study sites across the country.

Appendix D identifies socio-economic and demographic characteristics of State sites at the county levels.