05-10049. Frontal New Car Assessment Program

AGENCY:

National Highway Traffic Safety Administration (NHTSA), Department of Transportation (DOT).

ACTION:

Notice of final decision on the New Car Assessment Program (NCAP) pilot programs for child safety.

SUMMARY:

The Transportation Recall Enhancement, Accountability, and Documentation (TREAD) Act required that a safety rating for child restraints be established to provide practicable, readily understandable, and timely information to consumers. In addition, the TREAD Act directed the Secretary of Transportation to consider placing child restraints in the rear seat of vehicles crash-tested under NCAP. In response to this mandate, NHTSA established a consumer information program for add-on child restraints based on their ease of use, and announced that it would perform two pilot programs to gather additional information about other aspects of child passenger safety. One pilot program would subject child restraints to a 48 km/h (30 mph) sled test. This program focused on the protection provided by the child restraint. The second pilot program placed child restraints in the rear seat of vehicles in frontal NCAP tests. This program focused on the protection the vehicle provided to properly restrained children. Based on the data collected from both pilot programs, the agency has decided not to implement a dynamic child restraint system (CRS) rating based on sled tests, and to continue collecting data from NCAP frontal crash tests to resolve some outstanding technical issues and to gather additional data on the Hybrid III 6-year-old and 10-year-old child dummies. In addition, the agency will develop a better understanding of the real world data and its relationship to NCAP child results. The agency will make a decision on the merits of a vehicle child protection rating in conjunction with any possible revisions to the frontal testing program, which the agency is currently evaluating.

Privacy Act: Anyone is able to search the electronic form of all submissions received into any of our dockets by the name of the individual submitting the petition (or signing the petition, if submitted on behalf of an association, business, labor union, etc.). You may review DOT's complete Privacy Act Statement in the Federal Register published on April 11, 2000, (Volume 65, Number 70; Pages 19477-78) or you may visit http://dms.dot.gov.

FOR FURTHER INFORMATION CONTACT:

For technical issues concerning the dynamic sled pilot program, contact Mr. Sean Doyle of the New Car Assessment Program. Telephone: [phone]. Facsimile: [phone]. Electronic mail: [e-mail]. For technical issues concerning the vehicle pilot program, contact Mr. Brian Park of the New Car Assessment Program. Telephone: [phone]. Facsimile: [phone]. Electronic mail: [e-mail]. For legal issues, contact Ms. Deirdre Fujita of the Office of Chief Counsel. Telephone: [phone]. Facsimile: [phone]. Electronic mail: [e-mail]. You may send mail to these officials at: The National Highway Traffic Safety Administration, 400 Seventh St., SW., Washington, DC, 20590.

I. Introduction

II. CRS Dynamic Pilot Program

A. Sled Testing

B. Findings

C. NHTSA's Decision on a CRS Dynamic Rating Program

III. Vehicle Pilot Program

A. Vehicle Testing

B. Findings

C. NHTSA's Decision on a Vehicle Rating System for Child Protection

IV. Conclusions

Appendix A

I. Introduction

On November 1, 2000, Congress passed the Transportation Recall Enhancement, Accountability, and Documentation (TREAD) Act (Pub. L. 106-414, 114 Stat. 1800). Section 14(b) of this act directed the Secretary of Transportation to determine “whether to include child restraints in each vehicle crash tested under NCAP.” Additionally, section 14(g) directed NHTSA to “establish a child restraint safety rating consumer information program to provide practicable, readily understandable, and timely information to consumers for use in making informed decisions in the purchase of child restraints.”

NHTSA published a notice on November 6, 2001, which discussed existing programs throughout the world that rate the dynamic performance of child restraints, and addressed comments in response to the agency's child restraint system (CRS) safety plan.^[1] The notice also discussed possible methods of rating CRS, including, dynamic performance in sled tests and ease of use. The notice also discussed using child dummies in the rear seat of frontal NCAP crash tests to rate vehicles on child protection.

On November 5, 2002, a Notice of Final Decision was published in response to comments received relating to the proposed rating systems.^[2] In response to the congressional mandate outlined in the TREAD Act, a final protocol for an ease of use rating for child restraints was established and immediately implemented. The agency also announced its intent to conduct two pilot programs. One would investigate the feasibility of rating child restraints on their ability to protect children, based upon a dynamic sled test. The other would investigate the possibility of rating vehicles on their ability to protect children in the rear seat, based upon frontal NCAP tests incorporating CRS.

The first pilot program was a 48 km/h (30 mph) dynamic sled test pilot program to assess the dynamic performance of child restraints using the test seat assembly, test dummies (Child Restraint Air Bag Interaction (CRABI), Hybrid III 3-year-old dummy and Hybrid III 6-year-old dummy), and Injury Assessment Reference Values (IARVs) of the then proposed upgrade to Federal Motor Vehicle Safety Standard (FMVSS) No. 213, “Child Restraint Systems.” ^[3]

The second pilot program placed CRS restrained child dummies in the rear seat of frontal NCAP vehicle crash tests. It was hoped that the data attained from this pilot program would allow the agency to determine the relative contributions of both the child restraint, and the vehicle in which the child restraint is installed, to child occupant protection. Also, the agency wished to use this pilot program to evaluate whether one dummy size could be used as a surrogate for other child dummy sizes, and whether various child or harness types of restraints affected performance in a frontal crash test.

At the conclusion of the pilot programs, NHTSA said it would evaluate all the test results and make that evaluation available to the public. If the information attained through the pilot programs allowed the agency to resolve the remaining technical issues discussed in the 2002 notice, and the test data indicated that NHTSA could develop meaningful rating programs, it was intended that both a child restraint rating, based on the sled testing, and a vehicle rating, based on the child occupant performance in frontal NCAP tests, would be implemented in model year 2005.^[4]

The test results and technical analyses of both the sled and vehicle pilot programs are addressed in separate reports and are located in the NHTSA docket (Docket No. 2004-18682). This notice will discuss the agency's decision as it pertains to each of these pilot programs. The agency has determined that it will not proceed with a child restraint rating based on sled tests conducted at 48 km/h (30 mph). The agency has also determined that a decision on a vehicle rating for child occupant protection is not possible at this time based on the current test data. Therefore, NHTSA will continue testing child restraints in frontal NCAP vehicle tests to gather additional test data as well as determine the usefulness to consumers that such a program would provide. If a vehicle rating program based on rear seat child occupant protection were to be implemented in the future, it would occur simultaneously with any revisions that will be made to the frontal NCAP.^[5] Changes to the frontal NCAP are being considered as a result of changes made to FMVSS No. 208.^[6]

II. CRS Dynamic Pilot Program

A. Sled Testing

The two main goals of the dynamic CRS sled test pilot program were to (1) statistically compare the dynamic performance between different CRS configurations, and (2) determine the range of dynamic performance for CRS models. The testing was conducted in accordance with the recent upgrade in the FMVSS No. 213 rulemaking. Sled tests were performed at 48 km/h (30 mph) with the 1-year-old CRABI dummy (restrained in infant and convertible restraints), the Hybrid III 3-year-old dummy (restrained in convertible, combination, and booster restraints), and the Hybrid III 6-year-old dummy (restrained in combination and booster restraints).^[7] The test matrix was designed to perform paired t-tests^[8] , which controlled for all differences within a test except the variable of interest. The same model of child restraint was tested in the two outboard positions on the sled bench in two different configurations. Given the large number of dummy-CRS combinations, it was important to test as many combinations as possible, yet have a sample size that would permit meaningful statistical comparisons. Each CRS was tested in more than one configuration by either varying orientation of the restraint (forward-facing or rear-facing), attachment to the sled (LATCH ^[9] , belt with tether, or belt only), CRS usage (with or without a base), or test dummy.

The test matrix resulted in a total of 40 different CRS models being tested. These 40 models represented a large majority of the restraints available in the market at the time of testing. Various child restraint types, models, and brands were tested in six different test series. These series included:

(1) One child restraint on the sled bench versus two child restraints.^[10]

(2) Infant seats with and without their optional base (same CRS model per comparison).

(3) Hybrid III 3-year-old dummies versus 1-year-old CRABI dummies in rear-facing child restraints (same CRS model per comparison).^[11]

(4) Hybrid III 3-year-old dummies versus 1-year-old CRABI dummies in forward-facing child restraints (same CRS model per comparison).

(5) Child restraints with a lap belt and top tether versus LATCH.

(6) Hybrid III 3-year-old dummies in belt-positioning booster seats versus Hybrid III 6-year-old dummies in belt-positioning boosters.

The first goal was set to assist the agency in determining whether child restraints with multiple configurations would have to be tested in each of these configurations to get an accurate representation of the child restraints' dynamic performance. Subjecting each CRS to various testing configurations permitted a comparison between the dynamic performances of the same CRS in different configurations and allowed the agency to determine the feasibility of developing a rating program that would require testing of only one CRS configuration per restraint model, rather than potentially several configurations for some models. Such a rating program could then accurately reflect the child restraint's dynamic performance with one test, rather than needing to develop a more complicated rating scheme reflecting different performance for the various configurations.

To expand upon data acquired in previous testing, the agency also intended for the pilot program testing to help determine the range of dynamic performance between CRS models. This data would assist the agency in determining whether there are significant differences between the dynamic sled test performance of different child restraint models, and thus whether or not a rating program based on sled tests would provide meaningful information to consumers.

Further detail on the methodology and experimental design of this pilot program can be found in the technical report, “Child Restraint Dynamic Performance Evaluation in a 48km/h (30 mph) Sled Test,” located in docket NHTSA-04-18682.

B. Findings

Analysis of the sled test results was mainly based on two injury criteria: Head Injury Criterion (HIC) and chest acceleration ^[12] . These two injury criteria were chosen because HIC and chest acceleration are the two measurements that are most readily correlated to probability of injury. In addition, the agency felt that if a rating system were eventually developed, using HIC and chest acceleration would allow the agency to follow the same approach that is currently used for the adult dummies in frontal NCAP.

Statistical findings for the six series of tests are summarized in Table 1, and indicated the following:

▪ Similar performance was attained in forward facing (FF) CRSs irrespective of the dummy size.

▪ Securing the CRSs with LATCH or lap belt plus tether produced similar results for a given CRS.

▪ Higher HIC responses occurred when infant seat CRSs were tested with the removable base attached.

▪ For rear facing (RF) CRSs, higher HIC occurred when tested with the Hybrid III 3-year-old as compared to the 1-year-old CRABI dummy.

▪ For belt positioning booster (BPB) seats, higher HIC occurred when tested with the Hybrid III 3-year-old versus the Hybrid III 6-year-old dummy.

These results indicated that multiple tests would be necessary to establish a rating for infant, rear facing, and belt positioning booster CRSs, since testing in one configuration and/or with one particular dummy size would not assure that the results would apply to another configuration or dummy.

Table 1

CRS configuration	Statistical difference in HIC performance?	Statistical difference in chest acceleration performance?
One CRS on sled vs. two CRS on sled	NO	NO.
Hybrid III 3-year-old vs. 1-year-old CRABI FF	NO	NO.
LATCH vs. Lap Belt w/top tether	NO	NO.
Base vs. No base (infant seat)	YES, Base removed had lower HIC	NO.
Hybrid III 3-year-old vs. 1-year-old CRABI RF	YES, 1-year-old CRABI RF had lower HIC	NO.
Hybrid III 3-year-old vs. Hybrid III 6-year-old in BPB	YES, Hybrid III 6-year-old in BPB had lower HIC	NO.

Furthermore, the testing performed confirmed earlier studies conducted by the agency showing relatively little distinction in CRS performance based upon HIC and chest acceleration when tested on the sled in the same configuration.^[13] As shown in Appendix A, Figures A1-A3, tests using numerous make/model CRSs with the CRABI, Hybrid III 3-year-old, and Hybrid III 6-year-old dummies in rear facing, forward facing, and belt positioning booster CRS resulted in tightly clustered responses for both HIC and chest acceleration. The responses were also well within the established FMVSS No. 213 injury tolerance levels.

C. NHTSA's Decision on a CRS Dynamic Rating Program

Table A2 in Appendix A illustrates how there are often several different configurations for one specific child restraint type. The sled test data from this pilot program has shown that similar dynamic performance results cannot be assumed for different configurations of the same CRS model. As such, any CRS dynamic rating program would have to test many child restraints in multiple configurations, possibly with multiple dummies, to provide a rating for any one child restraint. Not doing so could consequently provide consumers with incomplete and inaccurate information. Moreover, if one restraint is tested in all applicable configurations, without a combined rating, the potential for multiple ratings for any one child restraint model could result in confusion for consumers.

The pilot program test results showed relatively small performance differences, particularly for chest acceleration, between the best and worst performer when tested under the same configuration. In effect, the agency found that for any given configuration, most makes and models produced results that were within an interval of 30 percent of the FMVSS No. 213 injury tolerance levels. This included the convertible and combination restraints as well as the infant restraints and belt-positioning booster seats. Given that all child restraints of the same type, when tested in the same configuration, perform very well and produce results that fall within a relatively tight response cluster, the agency believes that it is not feasible to develop a CRS dynamic rating that would provide meaningful consumer information over such a small range of dynamic performance, especially when multiple configurations are involved.

An additional agency concern is the frequent rotation of the CRS product line and short shelf life. Unlike vehicle models, which tend to have multiple year lifecycles before redesign, CRS model changes occur much more frequently. The shelf life for a typical CRS can be as short as six to eight months, and performing a dynamic sled test on each CRS model in its multiple configurations would make it unlikely that consumers would have relevant information available to them in making a purchasing decision. In consideration of the above, NHTSA has decided not to implement a dynamic CRS rating based on 48 km/h (30 mph) sled tests. The agency believes that when child restraints are used correctly, they are very effective in providing child passenger safety. Accordingly, the agency views the current ease-of-use consumer information program, which improves correct installation of child restraints by telling consumers which restraints are easier to use and by motivating manufacturers to make their child restraints easier to use, as sufficient in providing consumers with helpful and meaningful information when purchasing a child restraint.

III. Vehicle Pilot Program

A. Vehicle Testing

The three main goals of the vehicle pilot program were to investigate (1) whether or not the response performance for various dummy sizes and restraint configurations would indicate similar levels of occupant protection, (2) how different restraints affect performance, and (3) to separate the vehicle performance from the child seat performance. As such, the agency designed a test matrix to gather statistically comparable information as follows: Comparison of forward-facing vs. rear-facing child restraints, comparison of forward-facing child restraints vs. booster seats, comparison of one five-point harness model vs. another five-point harness model, and comparison of five-point harness vs. overhead shield restraints. The agency collected data from one hundred and eight frontal vehicle tests that used the 1-year-old CRABI dummy, the Hybrid III 3-year-old dummy, and the Hybrid III 6-year-old.

All convertible and forward-facing child restraints were installed using the LATCH system when used in the forward facing mode. For rear-facing child restraints, only the lower anchorages of the LATCH system were used to secure the child restraint. For every vehicle, the seating position behind the right front seat passenger had the same child restraint model (baseline CRS), which contained a forward facing Hybrid III 3-year-old child dummy. The CRS and child dummy used in the seating position behind the driver was varied in order to satisfy the program goals, and to serve as a comparison against the baseline CRS.

Further detail on the methodology, experimental design, and results of the pilot program can be found in the technical report, “Evaluation of Child Occupant Protection in a 56 km/h (35 mph) Frontal Barrier Crash”, located in docket NHTSA-04-18682.

B. Findings

This section discusses the findings in resolving the three main program goals. As with the analysis done for CRS Dynamic Pilot Program, HIC and chest acceleration were used for the analysis. The vehicle pilot program studied three CRS/dummy configurations. The first was the Hybrid III 3-year-old dummy positioned in a forward-facing convertible CRS compared to the 1-year-old CRABI dummy positioned in a rear-facing convertible CRS. The results of these paired tests showed no statistically significant difference in HIC values, but testing did show higher chest acceleration for the 1-year-old CRABI than the Hybrid III 3-year-old tested in the same vehicle. In addition, the testing also showed that some rear-facing restraints interacted with the front seatback during the crash event. However, due to the limited sample size, and the inability to quantify the interaction with video coverage and instrumentation, the agency feels that more research is needed to fully understand the importance of this interaction.

The second configuration analyzed as part of this pilot program compared a Hybrid III 3-year-old in a forward-facing convertible CRS to a Hybrid III 6-year-old that utilized the vehicle seat belts and a belt-positioning booster. Again, no statistically significant difference was found between the HIC values for each of the paired tests. However, the chest acceleration values for the Hybrid III 6-year-old were significantly higher than those of the Hybrid III 3-year-old. Further testing and analyses are needed to better understand these results, since standard NCAP instrumentation and camera coverage do not provide sufficient information to fully assess potential causes for this result.

The third comparison evaluated by the agency examined two child restraints that were identical, with the exception of the harness type. One child restraint had a five-point harness, while the other had an overhead shield. Statistical analysis showed that there was no significant difference for HIC or chest acceleration for the dummy in the five-point harness compared to the dummy in the overhead shield.

The agency also evaluated whether the same pair of CRS models, tested in multiple vehicles, would display a similar spread in injury results between the two different child restraints in every vehicle tested. Eleven vehicles were tested with the same two forward facing child restraints, the Evenflo Vanguard V and the Britax Roundabout. Both restraints were chosen based on cost, popularity, and availability at the time of testing. The average cost of the Vanguard V was about one-third the cost of the Roundabout. All tests utilized the Hybrid III 3-year-old dummy and the child restraints were secured using LATCH. The results for these tests are shown in Appendix A, Table A3. For these eleven tests, the injury values were typically lower for the Vanguard V than for the Roundabout, suggesting that the cost of a child restraint may have little to do with the level of safety offered by a CRS. In addition, paired t-testing showed that the average difference between the two child restraints is small based upon the injury risk curves.^[14] The difference in average HIC response was 58, or a difference of less than 2 percent head injury risk. The difference in average chest acceleration response was 3 G, or about 2 percent difference in chest injury risk. Both t-tests did not achieve statistical significance.

Because the Hybrid III 3-year-old child dummy was positioned in the same child restraint in every vehicle crash, thus establishing a baseline, the agency was able to compare the vehicle crash pulse characteristics to the child dummy injury readings. While the HIC readings showed little to no correlation with the crash pulse for the Hybrid III 3-year-old dummy, the chest acceleration readings did. The chest acceleration readings had a correlation ^[15] (R² ≉ 0.7) with pulse duration, with higher chest acceleration associated with shorter pulse durations. Chest acceleration showed a weaker correlation (R² ≉ 0.4) with peak acceleration and static crush.

Determining the source of difference in child seat performance was of interest. We wanted to find out what percent of the total variation in the HIC and chest G values are due to the vehicles. Analysis of variance indicated that about 75% of the variation in chest acceleration and about 60% of the variation in HIC values could be attributed to the vehicle make and model. An additional finding was that the vehicle type, such as passenger car, sport utility vehicle, van, or truck, did not statistically correlate with the child dummy results, and that there was no statistical correlation between the driver or front passenger dummy readings and the rear seat Hybrid III 3-year-old dummy readings.

C. NHTSA's Decision on a Vehicle Rating System for Child Protection

The agency's test data indicates that for the Hybrid III 3-year-old dummy, chest acceleration has some correlation to a vehicle's crash performance and that both the HIC and chest acceleration readings are more influenced by the vehicle than by the child restraint. However, we have insufficient data for the other dummy sizes to make a determination. As such, the agency feels that additional testing is necessary before a final decision can be made.

The CRS in-vehicle testing pilot program gathered important data on the CRABI, Hybrid III 3-year-old, and the Hybrid III 6-year-old child dummies. However, the agency is concerned that the results may have been affected by the interaction of the child seat with the front seat. Therefore, the agency would like to collect additional data to better understand the effect of this interaction on performance. When comparing the Hybrid III 3-year-old to the Hybrid III 6-year-old, the agency found that the dummies had statistically different values for chest acceleration, but not HIC. Further testing and analyses are needed to better understand these results.

Under Anton's Law (Pub. L. 107-318, 116 Stat. 2772), Congress mandated that the agency develop a test dummy representing a 10-year-old dummy for use in testing child restraints used in passenger motor vehicles. The agency has completed development and evaluation of the dummy, and will soon propose rulemaking to incorporate it into the Code of Federal Regulations. In-vehicle testing with this dummy, along with the Hybrid III 6-year-old dummy, will allow the agency to gather additional data on booster seat performance and determine if either of these dummies should be used in any potential vehicle rating.

The agency continues to believe that child restraints are highly effective in reducing the likelihood of death and/or serious injuries to children in motor vehicle crashes. The agency notes that misuse and non-use of child restraints are the predominant cause for fatalities involving children, and that even in these very severe frontal NCAP tests, none of the forward-facing child restraints had a structural failure. Similarly, when examining the data, the agency also notes that many of the vehicles provided relatively good performance. However, the agency is concerned that some vehicles did show dummy measurements in excess of established child injury reference values. The agency is working to better understand the meaning of these measurements as they relate to a high-speed frontal collision. Thus far, the understanding of the injury mechanism and injury risks to properly restrained children in appropriate child restraints for full frontal crashes is limited. The agency is using comprehensive collection and assembly of all available data, and is working with other interested parties, to better understand the child injury measures for these NCAP tests and the corresponding real world injuries.

Lastly, recent amendments to FMVSS No. 208 require vehicles manufactured after September 1, 2007, to meet the injury criteria of that standard at the NCAP test speed for the belted 50th percentile male dummy. Because compliance tests will then be performed at frontal NCAP test speeds, the agency is considering possible changes to NCAP. The agency has proposed several alternative approaches to revise the frontal NCAP, including incorporation of rear seat child occupant measurements into the rating system [69 FR 61071].

Given these reasons, the agency feels that more information is needed in order to decide whether to begin rating vehicles for child occupant protection using the CRS restrained rear seat in the frontal NCAP tests. To resolve the technical issues discussed in this notice, the agency will continue to collect rear seat child protection data from NCAP frontal crash tests.

IV. Conclusions

The agency has concluded that a dynamic CRS rating program would not provide meaningful information for consumers, and has decided it will not implement a dynamic CRS rating based on sled tests. The agency believes that the current ease of use consumer information program is providing consumers with helpful and meaningful information when purchasing a child restraint.

In terms of a vehicle rating for child occupant protection, NHTSA has concluded that more testing and analysis is needed before a final determination can be made on the inclusion of child response information in NCAP. To gather the necessary information, NHTSA will continue to collect CRS restrained rear seat child occupant data using 1-year-old CRABI, Hybrid III 3-year-old, Hybrid III 6-year-old, and Hybrid III 10-year-old dummies in frontal NCAP crash tests. In addition, NHTSA plans to further examine and analyze the injury risks of children in real world frontal crashes. NHTSA plans to make a decision and publish a notice discussing the merits of a consumer information program that rates vehicles on their ability to protect child occupants in conjunction with any possible revisions to frontal NCAP.

Authority: 49 U.S.C. 322, 30111, 30115, 30117, and 30166; delegation of authority at 49 CFR 1.50.

Issued on: May 6, 2005.

Stephen R. Kratzke,

Associate Administrator for Rulemaking.

Appendix A

Table A1

Child restraint type	Description
Infant Seat	For infants from birth to about 27 inches who weigh up to 20 pounds.
Convertible Seat	When Used Rear Facing: ▪ All are recommended for use by infants less than 1 year and up to about 20 pounds. ▪ Some are recommended for rear facing use, for heavier infants (30-35 pounds), and less than 1 year.
	When Used Forward Facing: ▪ All are rated for children up to 40 pounds. ▪ Used forward facing by children who are between 20 and 40 pounds, and over 1 year.
Combination Seat	When Used Rear Facing: ▪ All are recommended for use by infants less than 1 year and up to about 20 pounds. ▪ Some can be used for children from birth in place of a infant seat. ▪ Some are recommended for rear facing use, for heavier infants (30-35 pounds), and less than 1 year.
	When Used Forward Facing: ▪ All are rated for children up to 40 pounds. ▪ Remove harness when child reaches 40 pounds and use the vehicle's adult lap and shoulder belt. ▪ Many can be used for children up to 8 years old in place of a booster seat.
Booster Seat	▪ Recommended for use by children approximately 20 to 40 pounds, when used with harness. ▪ Remove harness when child reaches 40 pounds and use the vehicle's adult lap and shoulder belt for children up to 8 years old.

Table A2

		Child restraint type
		Infant	Convert.	Combo 2-in-1	Combo 3-in-1	BPB
Dummy	CRABI	X	X		X
	3YO		X	X	X	X
	6YO			X	X	X
Orientation	Rear Facing	X	X		X
	Forward Facing		X	X	X	X
Attachment	LATCH	X	X	X	X
	Belt w/ Tether		X	X	X
	Belt Only	X	X	X	X	X
Usage	Base	X
	No Base	X
CRABI-Child Restraint Air Bag Interaction.
Infant Seat—Rear-facing seat for use by infants from birth until at least one year.
Forward-Facing Only Seat—CRS with internal harness used for toddlers age 1 to age 4.
Convertible Seat—Hybrid of infant seat and forward-facing only seat.
Belt Positioning Booster (BPB)—Forward-facing seat with no harness. Used to properly position vehicle 3-point belts on children age until at least age 8.
2-in-1 Combo—Hybrid of forward-facing only seat and belt positioning booster.
3-in-1 Combo—Hybrid of infant seat, forward-facing only seat, and belt positioning booster.

Table A3

Model	Evenflo Vanguard 5		Britax Roundabout
	HIC 36	Chest acceleration	HIC 36	Chest acceleration
Acura TL	646	47	710	48
Chevrolet Malibu	1027	53	830	52
Dodge Intrepid	694	40	791	51
Hyundai XG350	970	55	976	64
Lincoln LS	641	39	816	50
Mitsubishi Endeavor	694	54	889	47
Suzuki Aerio	793	56	729	68
Toyota Camry	765	50	906	52
Toyota Highlander	1000	64	1107	57
Toyota Sienna	676	41	705	40
Toyota Solara	625	47	716	51

Footnotes

1.  65 FR 70687, Docket No. NHTSA-2000-7938

2.  67 FR 67448, Docket No. NHTSA-2001-10053

3.  June 24, 2003 Final Rule (68 FR 37620 Docket No, NHTSA-03-15351)

4.  By model year 2005, we mean October 2004 to coincide with the commencement of the fiscal year 2005 New Car Assessment Program

5.  69 FR 61071

6.  Effective beginning with the 2008 model year, FMVSS No. 208 will be upgraded to the current NCAP test speed of 35 mph for a belted 50th percentile HIII dummy. May 12, 2000 Final Rule (65 FR 30680).

7.  Appendix A, Table A1, contains a description of various types of child restraints.

8.  For paired t-tests, the data is dependent, i.e. there is a one-to-one correspondence between the values in the two samples and it determines whether the two values differ from each other in a statistically significant way.

9.  Lower Anchors and Tether for Children.

10.  Current FMVSS No. 213 compliance test procedure only permits testing of one CRS at a time on the sled bench.

11.  FMVSS No. 213 does not currently have a standard test procedure for testing the Hybrid III three-year-old dummy in rear-facing child restraints. Therefore NCAP relied on an installation procedure used by test facilities and other organizations that have experience testing Hybrid III three-year old dummies in rear-facing CRS.

12.  Head and knee excursions, along with other injury measures were also collected and are available in the technical report, “Child Restraint Dynamic Performance Evaluation in a 48km/h (30 mph) Sled Test” located in docket NHTSA-04-18682.

13.  In the November 5, 2002 notice, the agency had concerns that a rating program based on 30 mph sled tests would not provide meaningful information to consumers, as all child restraints subjected to this test would have received either a four- or five-star rating.

14.  Viano, DC., and Arepally, S., Assessing the Safety Performance of Occupant Restraint Systems,” Proceedings of the 34th Stapp Car Crash Conference, SAE Paper 902328, Warrendale, PA, November 1990.

15.  Linear Correlation; R2. = 1.0 is perfect linear correlation, R2. = 0.0 has no linear correlation.