The quantitative data analysis effort focused on determining the extent of the involvement of Hispanics in both pedestrian and bicycle crashes. It was of course, necessary to restrict the analysis to those databases that included race/ethnicity as a data element. In an effort to understand some of the reasons for Hispanic involvement in crashes, a limited examination of the 2000 U.S. Census was done to identify possible crash risk factors. In order to see if There are any major differences in crash involvement that may be attributable to country of origin, some of the crash data was examined in specific geographical areas that tend to have concentrations of Hispanics from specific countries/areas.

The following areas were identified to focus on four specific Hispanic sub-groups:

Country of Origin	Area
Hispanics of Mexican Origin	California
Hispanics of Central/ South American Origin	Washington, DC, Maryland, Virginia
Hispanics of Cuban Origin	Florida
Hispanics of Puerto Rican Origin	New York, New Jersey

It is recognized that these area/county of origin matchings are not perfect. The Hispanic populations in all these areas include those from a number of different countries. However, it is still worthwhile examining the different geographical areas to determine if any of the crash factors are related to country of origin.

Exposure issues involving Hispanic pedestrians and bicyclists are discussed in the next section. While reading the exposure issue section, the reader is advised to keep in mind two Other issues that are very relevant to Hispanic pedestrian/bicyclist exposure statistics: illegal immigrants and unreported crashes.

Illegal immigrants. The possible influence of illegal immigrants is a difficult but potentially very important issue. Ethnicity crash rates that are presented throughout this report are based on the number of individuals of a given ethnicity that responded to the U.S. Census. There is an unknown but reasonably large proportion of the Hispanic population that is "undocumented" or "illegal aliens." Typically these individuals do not participate in U.S. Census reporting and therefore are not included in the population numbers published by the Census. This means that There are more Hispanics in the U.S. than are reported by the U.S. Census. Thus, the crash rate data that is computed by dividing the number of crashes by the population - i.e., deaths per 100,000 people - is an over-representation of the actual crash risk because the population of people at risk is actually larger than the "official" numbers we use.

Unreported crashes. Unreported crashes are anOther issue that is important in pedestrian and bicyclist safety, in general, and especially important relative to Hispanic pedestrian/bicyclist safety. Unreported pedestrian/bicyclist crashes involve collisions between motor vehicles and pedestrians/bicyclists that are not reported to the police. About three quarters of the hospitalized victims of pedestrian and bicyclist motor crashes are identified with an official motor vehicle crash report. Some of the crashes not reported to the police are because they occur on private property, e.g., a parking lot, and a police report is not required. Some crashes are not reported because the injured pedestrian or bicyclist chooses not to call the police. It is believed that Hispanics in general and undocumented aliens, in particular, are far less likely to call the police when they are involved in a pedestrian/bicyclist crash than documented residents. However, the degree to which this conjecture is true is not known. If, in fact, Hispanics are involved in more unreported pedestrian/bicyclist crashes than Other ethnicities, then the numbers reported in the following section are an under-representation of the actual Hispanic pedestrian and bicyclist crash problem.

A related issue involves the difficulty in determining ethnicity of those involved in a crash from the crash report. Most city and state police crash reports do not indicate the ethnicity of the participants - either pedestrians, bicyclists or drivers - on the ir crash report forms. The lack of this critical data in many data files makes it difficult to evaluate the role of ethnicity in crash involvement.

In summary, when considering the crash statistics that follow, the reader is advised to consider the role that illegal, undocumented aliens (i.e., population under-reporting) and unreported crashes (crash under-reporting) may have on both the crash rates and on the absolute crash numbers.

The following four data sets will be discussed:

• Exposure and Other Crash Risk Factors
• Fatal Accident Reporting System (FARS) Data -Fatal Pedestrians & Bicyclist Crashes by Ethnicity. FARS data examined include the following ethnicity categories:
  • Non-Hispanic White
  • Non-Hispanic Black
  • Non-Hispanic Other
  • Hispanic-Mexican Origin
  • Hispanic-Puerto Rican Origin
  • Hispanic-Cuban Origin
  • Hispanic-Central/South American Origin
  • Hispanic-European Spanish or Other Origin
• Center for Disease Control (CDC) Data - Pedestrian and Bicyclist Deaths and Rates per 100,000 Populations by Ethnicity. The analysis includes the following categories:
  • Non-Hispanic White
  • Non-Hispanic Black
  • Hispanic
• Miami/Dade County Data. Analysis of Pedestrian Crashes in Miami/Dade County 1996-2000 by Ethnicity. Crash files use the following ethnicity categories:
  • Not Hispanic White
  • Not Hispanic Black
  • Hispanic

A. Exposure and Other Crash Risk Factors.

While examining the pedestrian and bicycle crash experience of Hispanics it is important to remember a number of Other important group characteristics. Hispanics, as a group, have a lower per capita income and a lower educational level. This limits greatly the ir transportation and housing options. They, therefore, tend to live in more densely developed urban areas with more vehicular traffic. These are the kind of areas that historically have had a disproportionate number of pedestrian and bicycle crashes.

Table 3 (next page) shows several crash risk factors for the Hispanics and White populations of the U.S. It is based on the 2000 U.S. Census. Factors are shown for the entire U.S. as well as seven states. Hispanics are 2.4 times more likely to live in a home with no vehicle available than Whites. As shown, 17.2% of Hispanic households do not have a vehicle available. Nearly 9% of Hispanics take public transportation to work - three times that of the White population. Using public transit typically involves walking to and from transit stops and crossing major arterials. Hispanics are 1.5 times more likely to walk to work and 2.3 times more likely to bike to work. Although There is some variation between the seven sets of state data, no remarkable differences are apparent.

the national Survey of Pedestrian and Bicyclist Attitudes and Behaviors (2002) provides some insight into the walking behavior of Hispanics and non-Hispanics. The following data was presented on the number of walking trips on the last day that they walked:

Ethnicity	Average Number of Walking Trips
Non-Hispanic Black	1.9
Hispanic	1.8
Non-Hispanic White	1.6

Hispanics walk more than non-Hispanic Whites but not as much as non-Hispanic Blacks. This increased walking leads to increased exposure and may explain some of the Hispanic over-involvement in pedestrian crashes.

Table 3 – Crash Risk Factors for the Hispanic and White U.S. Population

Crash Risk Factors for the Hispanic and White U.S. Population
Geographical Area	% of Housing Units with no Vehicle		% of Workers:
			Use Public Transportation		Walk to Work		Bike to Work
	Hispanic	White	Hispanic	White	Hispanic	White	Hispanic	White
United States	17.2%	7.3%	8.9%	2.9%	4.0%	2.6%	0.7%	0.3%
California	13.5%	6.9%	7.6%	3.1%	3.6%	2.5%	1.0%	0.8%
District of Columbia	46.1%	25.9%	40.7%	29.1%	11.8%	17.5%	1.1%	2.2%
Florida	11.7%	6.0%	3.5%	0.8%	2.0%	1.5%	0.6%	0.5%
Maryland	12.8%	6.9%	13.6%	3.7%	3.7%	2.3%	0.3%	0.2%
New Jersey	27.0%	8.2%	15.5%	6.5%	6.8%	2.3%	0.7%	0.2%
New York	56.7%	19.4%	44.1%	15.0%	9.4%	5.3%	0.6%	0.3%
Virginia	10.0%	5.3%	8.1%	2.7%	4.3%	2.0%	0.3%	0.2%

When examining the crash data that follows, it is critical that the role of these kinds of risk factors be considered. In many instances the data indicates that Hispanics may be over-involved in certain types of crashes. It is important to remember that it is not known if this involvement is due to the environmental factors, socio-economic factors, or simply because of language or cultural factors associated with the crash victim being Hispanic.

B. Fatal Accident Reporting System (FARS) Data

the national Highway Traffic Safety Administration (NHTSA) maintains a database of all fatal traffic crashes in the U.S. This data, the Fatal Accident Reporting System (FARS) was examined to determine the nature and extent of the Hispanic pedestrian and bicycle safety problem. Since 1999, FARS has included a data element to describe the race/ethnicity of the crash victim. For this report, FARS bicycle and pedestrian data was examined from 1999 through 2003. The subsequent analyses combine the data to maximize the available sample. The following data will be presented:

Pedestrian Fatalities by Ethnicity and Year, 1999-2003

Male Pedestrian Fatalities by Ethnicity and Pedestrian Age

Female Pedestrian Fatalities by Ethnicity and Pedestrian Age

Alcohol-related Pedestrian Fatalities by Ethnicity and Gender

Bicycle-related Fatalities by Ethnicity and Year, 1999-2003

Bicycle-related Fatalities by Ethnicity and Bicyclist Age

Alcohol-related Bicycle Fatalities by Ethnicity and Bicyclist Age

The reader is cautioned that many of the percentages shown in the tables that follow are based on relatively small numbers. As such, they should be interpreted very carefully.

1. Pedestrian Fatalities (FARS)

Two kinds of FARS data will be presented in this section. The first describes characteristics of the pedestrians involved in the crashes - i.e., ethnicity, age, gender, alcohol involvement, etc. The second kind of data variables involve the crash itself, i.e., time of day, site characteristics, etc. These variables are called Accident Level Variables.

a. Pedestrian Characteristics

Table 4 shows the annual pedestrian fatalities for 1999-2003 for the different ethnic groups. The percentages show the percent of that year's fatalities in each ethnicity category. Overall, Hispanics we reinvolved in 16.3% of the pedestrian fatalities for 1999-2003. This percent has varied from a high of 19%in 2001 to a low of 11.4% in 2002. Within the Hispanic sub-groups those of Mexican origin account for the majority of the fatalities - 10.5% of the national total.

Table 4 – Pedestrian Fatalities by Ethnicity

Ethnicity	Year										Total
	1999		2000		2001		2002		2003
	N	%	N	%	N	%	N	%	N	%	N	%
Non-Hispanic White	2,045	56.3%	2,032	58.3%	2,244	56.6%	1,691	62.8%	1,808	60.7%	9,820	58.6%
Non-Hispanic Black	710	19.5%	668	19.2%	731	18.4%	567	21.1%	551	18.5%	3,227	19.3%
Non-Hispanic Other	226	6.2%	224	6.4%	235	5.9%	129	4.8%	169	5.7%	983	5.9%
Mexican Origin	444	12.2%	368	10.6%	521	13.1%	181	6.7%	250	8.4%	1,764	10.5%
Puerto Rican origin	35	1.0%	45	1.3%	41	1.0%	15	0.6%	21	0.7%	157	0.9%
Cuban origin	34	0.9%	33	0.9%	30	0.8%	22	0.8%	30	1.0%	149	0.9%
Central or South American origin	72	2.0%	66	1.9%	93	2.3%	31	1.2%	76	2.6%	338	2.0%
European Spanish or Other Hispanic origin	67	1.8%	49	1.4%	69	1.7%	57	2.1%	73	2.5%	315	1.9%
Total	3,633	100.0%	3,485	100.0%	3,964	100.0%	2,693	100.0%	2,978	100.0%	16,753	100.0%

Source: Fatal Accident Reporting System: 1999-2003

The next most frequently involved subgroups were those of Central and South American origin – 2.0%, and European/Spanish/Other – 1.9%. Although one might hypothesize that the recent increase in immigration of Hispanics from Mexico and South/Central America would result in an increase in pedestrian fatalities, such does not appear to be true. If anything, the involvement of those of Mexican origin may have declined in the past two years.

Figure 1 shows the pedestrian fatality data by ethnicity graphically. It is readily apparent that Hispanics of Mexican origin account for the majority of the Hispanic fatalities. The next largest sub-group, Hispanics of Central/South American origin, account for 2% of the pedestrian fatalities nationally. Together, Hispanics of Mexican and Central/South American origin account for 2,102 out of the 2,723 Hispanic fatalities – 77%.

Figure 1   Pedestrian Fatalities by Ethnicity

Source: Fatal Accident Reporting System: 1999-2003

Table 5 (next page) shows the ethnicity and gender of the 16,753 pedestrian fatalities occurring from 1999-2003. Some interesting patterns are evident. Of all pedestrian fatalities nationwide, 69.7% are male. By contrast, 77.2% of the Hispanic fatalities were male, while 67.3% of the non-Hispanic Whites were male. Differences are also apparent among the various subgroups of Hispanics. Only 66.2% of those of Puerto Rican origin and 70.5% of those of Cuban origin are male. Those of Central/South American origin (75.4%), Mexican origin (78.7%) and European Spanish/Other Hispanic origin (79.7%) tend to have more male pedestrian fatalities.

Table 5 – Pedestrian Fatalities by Ethnicity & Gender

Pedestrian Fatalities by Ethnicity & Gender
Ethnicity					Both Genders
	Males		Females
	N	%	N	%	N	%
Non-hispanic White	6,613	67.3%	3,206	32.7%	9,819	100.0%
Non-hispanic Black	2,330	72.2%	896	27.8%	3,226	100.0%
Non-hispanic Other	623	63.4%	360	36.6%	983	100.0%
Hispanic-All	2,103	77.2%	620	22.8%	2,723	100.0%
Total	11,669	69.7%	5,082	30.3%	16,751%	100.0%
Non-Hispanic White	6,613	67.3%	3,206	32.7%	9,819	100.0%
Non-Hispanic Black	2,330	72.2%	896	27.8%	3,226	100.0%
Non-Hispanic Other	623	63.4%	360	36.6%	983	100.0%
Mexican Origin	1,388	78.7%	376	21.3%	1,764	100.0%
Puerto Rican origin	104	66.2%	53	33.8%	157	100.0%
Cuban origin	105	70.5%	44	29.5%	149	100.0%
Central or South American origin	255	75.4%	83	24.6%	338	100.0%
European Spanish or Other Hispanic	251	79.7%	64	20.35	315	100.0%
Total	11,669	69.7%	5,082	30.3%	16,751	100.0%

Source: Fatal Accident Reporting System: 1999-2003

Since it appears that gender may play a role in Hispanic pedestrian fatalities, the examination of the variable, age, is being examined separately for males and females. Table 6 shows the distribution of male pedestrian fatalities by age and ethnicity. Overall, all Hispanics are somewhat over-represented (like the non-Hispanic Blacks) among those under 6 years of age (4.0% of Hispanic fatalities vs. 2.2% of non-Hispanic White fatalities). Hispanics from 6-15 years (4.5%) are under-represented relative to non-Hispanic White (5.7%). The age category showing the largest Hispanic over-representation is the 21-29 age group – 19.3% of the Hispanic fatalities are in that group vs. 11.6% of the non-Hispanic Whites. Further examination of the country of origin data shows that this over-involvement is due to the over-involvement of males of Mexican origin (20.0%) and of Central/South American origin (25.5%). Although the absolute number is relatively small (56 fatalities in 5 years), it is interesting that over one-half (53.3%) of the Hispanics of Cuban origin fatalities involve pedestrians 60 and older. This is far higher than the nationwide total (21.1%) or the 24.3% in the 60+ age group of non-Hispanic Whites.

Table 6 - Male Pedestrian Fatalities by Ethnicity and Age

Male Pedestrian Fatalities by Ethnicity & Age
Ethnicity	Age Category														Total
	< 6		6-15		16-20		21-29		30-59		60+		Unknown
	N	%	N	%	N	%	N	%	N	%	N	%	N	%	N	%
Non-hispanic White	144	2.2%	376	5.7%	402	6.1%	765	11.6%	3,287	49.7%	1,606	24.3%	33	0.5%	6,613	100.0%
Non-hispanic Black	137	5.9%	223	9.6%	124	5.3%	218	9.4%	1,286	55.2%	326	14.0%	16	0.7%	2,330	100.0%
Non-hispanic Other	18	2.9%	20	3.2%	52	8.3%	85	13.6%	270	43.3%	176	28.3%	2	0.3%	623	100.0%
Hispanic-All	84	4.0%	94	4.5%	136	6.5%	406	19.3%	991	47.1%	359	17.1%	33	1.6%	2,103	100.0%
Total	383	3.3%	713	6.1%	714	6.1%	1,474	12.6%	5,834	50.05%	2,467	21.1%	84	0.7%	11,669	100.0%

 

Male Pedestrian Fatalities by Ethnicity Sub-Group & Age
Ethnicity	Age Category														Total
	< 6		6-15		16-20		21-29		30-59		60+		Unknown
	N	%	N	%	N	%	N	%	N	%	N	%	N	%	N	%
Non-Hispanic White	144	2.2%	376	5.7%	402	6.1%	765	11.6%	3,287	49.7%	1,606	24.3%	33	0.5%	6,613	100.0%
Non-Hispanic Black	137	5.9%	223	9.6%	124	5.3%	218	9.4%	1,286	55.2%	326	14.0%	16	0.7%	2,330	100.0%
Non-Hispanic Other	18	2.9%	20	3.2%	52	8.3%	85	13.6%	270	43.3%	176	28.3%	2	0.3%	623	100.0%
Mexican Origin	65	4.7%	72	5.2%	100	7.2%	278	20.0%	654	47.1%	207	14.9%	12	0.9%	1,388	100.0%
Puerto Rican origin	4	3.8%	9	8.7%	4	3.8%	13	12.5%	50	48.1%	24	23.1%	0	0.0%	104	100.0%
Cuban origin	2	1.9%	1	1.0%	2	1.9%	6	5.7%	35	33.3%	56	53.3%	3	2.9%	105	100.0%
Central or South American origin	3	1.2%	4	1.6%	15	5.9%	65	25.5%	116	45.5%	42	16.5%	10	3.9%	255	100.0%
European Spanish or Other Hispanic	10	4.0%	8	3.2%	15	6.0%	44	17.5%	136	54.2%	30	12.0%	8	3.2%	251	100.0
Total	383	3.3%	713	6.1%	714	6.1%	1,474	12.6%	5,834	50.0%	2,467	21.1%	84	0.7%	11,669	100.0%

Source: Fatal Accident Reporting System: 1999-2003

The number of male pedestrian fatalities by ethnicity bandage category is shown in Figure 2. The relatively large numbers of Hispanics 21-29, 30-59 and 60+ involved in fatal crashes is apparent. The similarity in patterns in the various age groups is illustrated as is the relative over-involvement of Hispanics 21-29 years.

Figure 2 – Male Pedestrian Fatalities by Ethnicity and Age Category

Source: Fatal Accident Reporting System: 1999-2003

Table 7 shows the distribution of female pedestrian fatalities by age and ethnicity. Like the distribution for males, the female Hispanics are much higher (8.9%) for pedestrians under 6 years of age. This is due largely to the large number of Hispanics of Mexican origin in this group. Unlike the Hispanic males,Hispanic females do not show an over-involvement in the 21-29 age group. Otherwise, the involvement, by age category of female Hispanics is similar to that of the non-Hispanic Whites.

Table 7 – Female Pedestrian Fatalities by Ethnicity and Age

Female Pedestrian Fatalities by Ethnicity & Age
Ethnicity	Age Category														Total
	< 6		6-15		16-20		21-29		30-59		60+		Unknown
	N	%	N	%	N	%	N	%	N	%	N	%	N	%	N	%
Non-Hispanic White	85	2.7%	257	8.0%	174	5.4%	263	8.2%	1,274	39.7%	1,143	35.7%	10	0.3%	3,206	100.0%
Non-Hispanic Black	74	8.3%	127	14.2%	50	5.6%	89	9.9%	438	48.9%	114	12.7%	4	0.4%	896	100.0%
Non-Hispanic Other	17	4.7%	21	5.8%	9	2.5%	31	8.6%	129	35.8%	153	42.5%	0	0.05	360	100.0%
Hispanic - All	55	8.9%	50	8.1%	35	5.6%	52	8.4%	243	39.2%	184	29.7%	1	0.2%	620	100.0%
Total	231	4.5%	455	9.0%	268	5.3%	435	8.6%	2,084	41.0%	1,594	31.4%	15	0.3%	5,082	100.0%

 

Female Pedestrian Fatalities by Ethnicity Sub-Group & Age
Ethnicity	Age Category														Total
	< 6		6-15		16-20		21-29		30-59		60+		Unknown
	N	%	N	%	N	%	N	%	N	%	N	%	N	%	N	%
Non-Hispanic White	85	2.7%	257	8.0%	174	5.4%	263	8.2%	1,274	39.7%	1,143	35.7%	10	0.3%	3,206	100.0%
Non-Hispanic Black	74	8.3%	127	14.2%	50	5.6%	89	9.9%	438	48.9%	114	12.7%	4	0.4%	896	100.0%
Non-Hispanic Other	17	4.7%	21	5.8%	9	2.5%	31	8.6%	129	35.8%	153	42.5%	0	0.0%	360	100.0%
Mexican Origin	46	12.2%	34	9.0%	27	7.2%	41	10.9%	141	37.5%	86	22.9%	1	0.3%	376	100.0%
Puerto Rican origin	4	7.5%	2	3.8%	3	5.7%	6	11.3%	24	45.3%	14	26.4%	0	0.0%	53	100.0%
Cuban origin	0	0.0%	0	0.0%	1	2.3%	0	0 0%	9	20.5%	34	77.3%	0	0.0%	44	100.0%
Central or South American origin	3	3.6%	3	3.6%	3	3.6%	3	3.6%	38	45.8%	33	39.8%	0	0.0%	83	100.0%
European Spanish or Other Hispanic	2	3.1%	11	17.2%	1	1.6%	2	3.1%	31	48.4%	17	26.6%	0	0.0%	64	100.0%
Total	231	4.5%	455	9.0%	268	5.3%	435	8.6%	2,084	41.0%	1,594	31.4%	15	0.3%	5,082	100.0%

Source: Fatal Accident Reporting System: 1999-2003

The country of origin data underlines the fact that Hispanics of Mexican origin account for 61% of all female Hispanic fatalities and are over-represented in the 4 younger age categories. Interestingly, the over-representation of older Cuban males seen earlier is repeated here with 77.3% of the Cuban Hispanic female pedestrian fatalities 60 or older. Please note,however, that There are only 44 female Cuban pedestrian fatalities in this 5-year period.

Figure 3 shows graphically the data on female pedestrian fatalities by age and ethnicity. The patterns just discussed are apparent. Especially notable is the fact that the Hispanic females 30-59 and over 60 account for most (68.9%) of the Hispanic female pedestrian fatalities.

Figure 3 - Female Pedestrian Fatalities by Ethnicity and Age Categories

Alcohol involvement was either "not reported" or "unknown"in about one-half (57%) of the pedestrian fatalities from 1999 to 2003. In the remaining 43% of the crashes, alcohol involvement was determined to be a factor in about 16% and determined not to be a factor in about 27% of the crashes.

Table 8 shows the alcohol-related fatalities by ethnicity and gender. Of the 16,751 fatalities,2,725 (16.3%) were found to have alcohol involvement on the part of the pedestrian.

Table 8 – Alcohol-Related Pedestrian Fatalities by Ethnicity and Gender

Alcohol Related Pedestrian Fatalities by Ethnicity & Gender
Ethnicity	Males			Females			Both Genders
	Alcohol Involvement		Total Fatals	Alcohol Involvement		Total Fatals	Alcohol Involvement		Total Fatals
	N	%	N	N	%	N	N	%	N
Non-Hispanic White	1,197	18.1%	6,613	342	10.7%	3,206	1,539	15.7%	9,819
Non-Hispanic Black	354	15.2%	2,330	90	10.0%	896	444	13.8%	3,226
Non-Hispanic Other	183	29.4%	623	48	13.3%	360	231	23.5%	983
Hispanic - All	467	22.2%	2,103	44	7.1%	620	511	18.8%	2,723
Total	2,201	18.9%	11,669	524	10.3%	5,082	2,725	16.3%	16,751

 

Alcohol Related Pedestrian Fatalities by Ethnicity & Gender
Ethnicity	Males			Females			Both Genders
	Alcohol Involvement		Total Fatals	Alcohol Involvement		Total Fatals	Alcohol Involvement		Total Fatals
	N	%	N	N	%	N	N	%	N
Non-Hispanic White	1,197	18.1%	6,613	342	10.7%	3,206	1,539	15.7%	9,819
Non-Hispanic Black	354	15.2%	2,330	90	10.0%	896	444	13.8%	3,226
Non-Hispanic Other	183	29.4%	623	48	13.3%	360	231	23.5%	983
Mexican origin	325	23.4%	1,388	31	8.2%	376	356	20.2%	1,764
Puerto Rican origin	12	11.5%	104	1	1.9%	53	13	8.3%	157
Cuban origin	7	6.7%	105	2	4.5%	44	9	6.0%	149
Central or South American origin	39	15.3%	255	4	4.8%	83	43	12.7%	338
European Spanish or Other Hispanic origin	84	33.5%	251	6	9.4%	64	90	28.6%	315
Total	2,201	18.9%	11,669	524	10.3%	5,082	2,725	16.3%	16,751

Source: Fatal Accident Reporting System: 1999-2003

In addition, it is important to report that There were large differences in the percentage not reported/unknown by ethnicity. This varied from a low percentage unknown/unreported of 38.9% for Hispanics of Cuban origin to 67.5% for Hispanics of Puerto Rican origin. It is not known if the variability in unknown/unreported alcohol level is related to the ethnicity of the pedestrian or the geographic area where the pedestrian lives - e.g., many of the Cubans were in the Miami area. The coroners' offices in different areas have different policies when it comes to testing the alcohol level of fatally injured pedestrians.

The percentage unknown/not reported for the largest Hispanic sub-group, those of Mexican origins, is 58.5%. This is between the comparable percentage of non-Hispanic White (55.6%)and non-Hispanic Black (62.7%) pedestrians.

In spite of these potential issues with the alcohol involvement data, it nevertheless provides some insights into the nature of the Hispanic pedestrian problem. The numbers reported here are based on the 43% of the fatalities for which alcohol involvement data was available. Thus,the percentages shown are an underestimate of the alcohol involvement. If one assumes that the unreported/unknown cases have the proportion of alcohol involvement as the cases for which data is available, the alcohol involvement rates would be about twice the percentages reported here. In terms of ethnicity, the Hispanic pedestrians had higher (18.8%) alcohol involvement that either the non-Hispanic Whites (15.7%) or the non-Hispanic Blacks (13.8%). The percentages for the 'non-Hispanic: Other' is high (23.5%) because that group includes American Indians, a group with a long recognized alcohol/pedestrian safety problem.

The role of gender in the Hispanic alcohol problem is apparent. Hispanic females have lower(7.1%) alcohol involvement than any Other ethnicity. Hispanic men have The second highest involvement (22.2%)surpassed only by the male 'non-Hispanic: Other' group, which includes American Indians.

The alcohol involvement in the various ethnic subgroups is very interesting. Male Hispanics classified as European Spanish/Other have an alcohol involvement rate of 33.5%,although the number in the group (N=84) is small. The male Hispanic pedestrian of Mexican origin has an alcohol involvement rate of 23.4%; they are largely responsible for the high alcohol involvement rate of the entire group. Hispanics of Puerto Rican, Cuban, and Central/South American origin all have alcohol involvement rates below that of non-Hispanic Whites.

Female Hispanic pedestrians in all the Hispanic sub-groups have a very low alcohol rate, almost all less than one-half that of non-Hispanic White females and non-Hispanic Black females. Clearly, drinking by Mexican-Americans is an issue to be addressed in future pedestrian safety campaigns.

b. Pedestrian Crash Characteristics/Accident Level Variables.

The FARS database contains a number of variables that describe characteristics of the crash itself, i.e., weather, roadway characteristics, etc. This section includes cross-tabulations of these variables for the major ethnicity groups – Non-Hispanic Whites, Non-Hispanic Blacks, and Hispanics. Table 9 (next page) contains distributions of the following variables for each of the three major ethnic groups:

Roadway/Area Type

Surface condition

Light Condition

Roadway Classification

Number of Lanes

Relation to Junction

Posted Speed Limit

Traffic Controls Present

These variables were selected because they either provide insight into the nature of Hispanic pedestrian crashes or because they highlight potential differences between the pedestrian crash experience of Whites, Blacks and/or Hispanics. An umber of Other "accident level" variables in the FARS database were also examined but were found to provide no additional useful information or insights. These variables were:

Route Signing

Manner of Collision

Relation to Roadway

Traffic way Flow

Roadway Alignment

Roadway Profile

Roadway Surface Type

Construction/Maintenance Zone

Traffic Control Functioning

Atmospheric Conditions

Hit and Run

School Bus Related

Emergency Medical Services Information

The first pedestrian crash characteristic variable shown in Table 9 is Roadway/Area Type. About a third of the non-Hispanic White pedestrian fatalities occur in rural areas. By contrast, about a quarter of the non-Hispanic Black pedestrians and one-fifth of the Hispanic pedestrians are killed in rural areas. The Hispanic pedestrian problem has more of an urban focus than that of the Other ethnicities.

Table 9: Selected Accident Level Variables for Pedestrian Fatalities by Ethnicity

Selected Accident Level Variables for Pedestrian Fatalities by Ethnicity
	Percentage of Crashes
Variable	Non-Hispanic White	Non-Hispanic Black	Hispanic
	N=9820	N=3227	N=2723
Roadway/Area Type
Urban	64.4%	73.3%	79.9%
Rural	35.6%	26.7%	20.1%
Surface Condition
Dry	83.9%	85.7%	89.2%
Wet	13.6%	13.0%	9.2%
Snow or Ice	1.9%	0.7%	0.7%
Other/Unknown	0.6%	0.6%	0.9%
Light Condition
Daylight	31.9%	27.0%	25.4%
Dark – No Lighting	35.8%	34.1%	30.8%
Dark – Lighting	28.3%	35.6%	40.0%
Other/Unknown	3.9%	3.3%	3.9%
Roadway Classification
Arterials – Rural	17.8%	11.9%	10.0%
Collectors – Rural	9.7%	7.9%	4.7%
Local Road – Rural	7.3%	6.1%	5.0%
Major Arterials – Urban	34.5%	32.8%	40.9%
Minor Arterials – Urban	13.4%	14.2%	17.8%
Collectors – Urban	3.3%	4.4%	3.6%
Local Streets – Urban	12.0%	20.4%	16.9%
Other/Unknown	2.0%	2.3%	1.1%
Number of Lanes
Two or less	>66.7%	59.1%	58.4%
Three	10.3%	12.0%	14.0%
Four	15.9%	21.1%	19.2%
Five or more	4.8%	5.3%	6.2%
Unknown	2.2%	2.4%	2.3%
Posted Speed Limit
25 or less	9.1%	9.2%	8.9%
30-35	25.3%	34.9%	32.3%
40-45	26.1%	23.3%	26.6%
50-55	23.8%	20.0%	15.0%
Over 55	13.2%	10.0%	14.1%
Other/Unknown	2.4%	2.7%	3.1%
Relation to Junction
Non-intersection	74.8%	75.9%	74.0%
Intersection/Intersection-related	20.0%	19.3%	21.4%
Other/Unknown	5.2%	4.8%	4.6%
Traffic Controls Present
None	84.5%	86.6%	84.5%
Traffic Signal (RGA)	8.9%	8.4%	8.9%
Stop Sign	2.3%	2.3%	2.3%
Other Signs/Signals	4.3%	2.7%	4.3%

Source: FARS 1999-2003

Roadway surface conditions for the vast majority of all pedestrian crashes were "dry." Although the Hispanic fatalities were somewhat more likely (89.2) to occur during dry conditions than either Whites (83.9%) or Blacks (85.7%), this is probably an artifact of the ir state of residence, since 43% of the Hispanic fatalities occurred in "Sunbelt" states (AZ, CA, FL and TX). This is also shown by The relatively low incidence of Hispanic fatalities when it is snowy and/or icy.

Hispanic pedestrian fatal crashes also are less likely to occur during daylight. While 25.4% of the Hispanic pedestrians were struck during daylight, 31.9% of the non-Hispanic Whites were involved during the daylight. Figure 4 graphically shows the percentage of fatalities by time of day. In general, the curves for non-Hispanic Whites, non-Hispanic Blacks and Hispanics are very similar. The non-Hispanic Whites have slightly more crashes from 6am to 6pm, times when it is usually light. The Hispanic crashes show small peaks at 8pm and 9pm and again at 1am and 2am. These peaks may be related to the previously described increased alcohol involvement in Hispanic pedestrian crashes.

Figure 4: Pedestrian Fatalities by Time of Day

Source: FARS 1999-2004

The roadway classification of the crashes shows that Hispanic pedestrians are more likely to be involved in crashes on major urban arterials (40.9%)and minor urban arterials (17.8%) than the Other ethnicities. The next variable listed, number of lanes, provides similar information. Hispanic pedestrians are less likely to be killed on two-lane roads (58.4%) than are non-Hispanic Whites (66.7%). Similarly, Hispanics are more likely to be killed on roads with 3 lanes(14.0%), 4 lanes (19.2%) or 5 or more lanes (6.2%). This over-involvement in crashes on multi-lane arterials is also shown under the "light condition" variables which show that 40% of the Hispanic fatalities occurred under dark conditions but where street lighting was present. This over-involvement on lighted, multi-lane, major arterials is probably related to the fact that Hispanic crashes more typically involve young adult males, many of whom have been drinking.

The data on posted speed limit also mirrors the trends noted relative to roadway classification and number of lanes. The Hispanic pedestrians are less involved(15.0%) on roadways with 50 or 55 mph speed limits and somewhat more involved on roadways with 30-35 mph speed limits, the speed limit information on the location of the crash relative to an intersection reveals no meaningful differences. Ethnicity does not appear to influence whether pedestrians are involved in crashes that are intersection or intersection-related or not near an intersection. It is interesting that only about 1/5 of all the fatal pedestrian crashes are intersection or intersection-related, regardless of ethnicity.

The last data element, Traffic Controls Present, reflects similar information. Only about 9% of all pedestrian fatalities occur at a signalized intersection with a Red-Green-Amber(RGA) signal. Unfortunately, it is not known if a pedestrian signal head is present at most of these locations. There are no differences between the ethnicities in terms of traffic controls at the crash location. Pedestrian safety programs that target crashes at signalized intersections and/or pedestrian crosswalk locations are not focusing on the location of most fatal pedestrian crashes, including those involving Hispanics.

2. Bicyclist Fatalities (FARS)

Two kinds of FARS data will be presented in this section. The first describes characteristics of the bicyclists involved in the crashes - i.e., ethnicity,age, gender, alcohol involvement, etc. The second kind of data variables involve the crash itself, i.e., time of day, site characteristics, etc. These variables are called Accident Level Variables.

a. Bicyclist Characteristics

Table 10 shows the annual bicyclist fatalities from 1999 through 2003. The top portion shows all the Hispanics group Together. The bottom portion of the table has the Hispanics broken out by country of origin. As was noted earlier, it is important to look at both the percentages and the number it represents. Some of the cells in the tables that follow have relatively small numbers and the computed percentage should be carefully considered.

Table 10 – Bicyclist Fatalities by Ethnicity

Bicyclist Fatalities By Ethnicity
Ethnicity	Year										Total
	1999		2000		2001		2002		2003
	N	%	N	%	N	%	N	%	N	%	N	%
Non-Hispanic White	377	64.4%	346	63.4%	377	61.0%	240	64.2%	248	63.6%	1,588	63.2%
Non-Hispanic Black	88	15.0%	95	17.4%	118	19.1%	72	19.3%	73	18.7%	446	17.7%
Non-Hispanic Other	27	4.6%	18	3.3%	17	2.8%	9	2.4%	15	3.8%	86	3.4%
Hispanic - All	93	15.9%	87	15.9%	106	17.2%	53	14.2%	54	13.8%	393	15.6%
Total	585	100.0%	546	100.0%	618	100.0%	374	100.0%	390	100.0%	2,513	100.0%

 

Bicyclist Fatalities By Ethnicity
Ethnicity	Year										Total
	1999		2000		2001		2002		2003
	N	%	N	%	N	%	N	%	N	%	N	%
Non-Hispanic White	377	64.4%	346	63.4%	377	61.0%	240	64.2%	248	63.6%	1,588	63.2%
Non-Hispanic Black	88	15.0%	95	17.4%	118	19.1%	72	19.3%	73	18.7%	446	17.7%
Non-Hispanic Other	27	4.6%	18	3.3%	17	2.8%	9	2.4%	15	3.8%	86	3.4%
Mexican Origin	73	12.5%	61	11.2%	73	11.8%	29	7.8%	30	7.7%	266	10.6%
Puerto Rican origin	5	0.9%	9	1.6%	8	1.3%	4	1.1%	2	0.5%	28	1.1%
Cuban origin	2	0.3%	0	0.0%	3	0.5%	4	1.1%	7	1.8%	16	0.6%
Central or South American origin	7	1.2%	11	2.0%	13	2.1%	6	1.6%	11	2.8%	48	1.9%
European Spanish or Other Hispanic origin	6	1.0%	6	1.1%	9	1.5%	10	2.7%	4	1.0%	35	1.4%
Total	585	100.0%	546	100.0%	618	100.0%	374	100.0%	390	100.0%	2,513	100.0%

Source: Fatal Accident Reporting System: 1999-2003

Overall, bicyclist fatalities have declined since 1999. The percentage of the fatalities that were Hispanic has also gone down slightly from 15.9% of the total in 1999 to 13.8%of the total in 2003. The year 2001 shows an increase in bicycle fatalities in all ethnic groups except for non-Hispanic Other. Overall, from 1999 through 2003, Hispanics have been involved in 15.6%of the bicyclist fatalities.

In terms of country of origin, Hispanics of Mexican origin account for the greatest number of crashes. Of the 393 Hispanic bicyclists killed from 1999 to 2003, a total of 266(67.7%) were of Mexican origin. Hispanics of Cuban origin (N=16) and those of Puerto Rican origin (N=28)are less frequently involved, although it should be remembered that they also represent a much smaller percentage of the Hispanics in the U.S.

Figure 5 shows the distribution of bicycle fatalities by ethnicity during the 1999 -2003 timeframe. Clearly shown is that Hispanics of Mexican origin make up 11% of the national bicycle fatalities. All Hispanic subgroups combine to represent 15.6% of all bicycle fatalities in the U.S. The next largest sub-group is the Hispanics of Central/South American origin; they account for 2% of the national total.

Figure 5 - Bicycle Fatalities by Ethnicity

Source: Fatal Accident Reporting System: 1999-2003

Table 11 shows the ethnicity and gender of the 2513 bicycle fatalities from the FARS database for 1999-2003. The vast majority of all bicyclist fatalities (88.8%) are male. Hispanics have an even higher percentage of males (94.1%) than do the non-Hispanic Whites (87.0%) or the non-Hispanic Blacks (91.3%). This very high involvement of males in bicyclist accidents is true regardless of the country of origin. Hispanics of Mexican origin have the highest percentage of males (94.7%) of all the Hispanic sub-groups, not including the very small number (N=16) of Cuban bicycle fatalities.

Table 11: Bicyclist Fatalities by Ethnicity & Gender

Alcohol Related Pedestrian Fatalities by Ethnicity & Gender
Ethnicity	Males		Females		Both Genders
	N	%	N	%	N	%
Non-Hispanic White	1,381	87.0%	207	13.0%	1,588	100.0%
Non-Hispanic Black	407	91.3%	39	8.7%	446	100.0%
Non-Hispanic Other	370	94.1%	23	5.9%	393	100.0%
Hispanic - All	370	94.1%	23	5.9%	393	100.0%
Total	2,232	88.8%	281	11.2%	2,513	100.0%

 

Bicyclist Fatalities by Ethnicity Sub-Group & Gender
Ethnicity	Males		Females		Both Genders
	N	%	N	%	N	%
Non-Hispanic White	1,381	87.0%	207	13.0%	1,588	100.0%
Non-Hispanic Black	407	91.3%	39	8.7%	446	100.0%
Non-Hispanic Other	74	86.0%	12	14.0%	86	100.0%
Mexican origin	252	94.7%	14	5.3%	266	100.0%
Puerto Rican origin	26	92.9%	2	7.1%	28	100.0%
Cuban origin	16	100.0%	0	0.0%	16	100.0%
Central or South American origin	45	93.8%	3	6.3%	48	100.0%
European Spanish or Other Hispanic origin	31	88.6%	4	11.4%	35	100.0%
Total	2,232	88.8%	281	11.2%	2,513	100.0%

Source: Fatal Accident Reporting System: 1999-2003

Table 12 (next page) shows male bicyclist fatalities by Ethnicity and Age for 1999 to 2003 combined. A total of 2,232 fatalities are represented. Overall, the Hispanics show a different age distribution than the Other ethnicities. They are less involved as children 6-15 year of age. While 21.9% of the non-Hispanic Whites and 25.3% of the non-Hispanic Blacks were 6-15 years of age, only 11.6% of the Hispanics were in that age group. Conversely, they are far more involved as young adults. A total of 18.1% of the Hispanic male fatalities were 21-29 years of age while only 7.3%of the non-Hispanic Whites and 7.1% of the non-Hispanic Blacks were 21-29.

The country of origin data shown indicates that most (47 out of 67) of the male Hispanics aged 21-29 that were killed were of Mexican origin. Overall, Hispanics of Mexican origin account for 252 out of 370 (68.1%) of the male bicycle Hispanic fatalities.

Since There were only 23 female Hispanic bicycle fatalities in 5 years of FARS data, a separate table showing female bicycle fatalities by age and ethnicity is not being presented.

Table 12: Male Bicyclist Fatalities by Ethnicity & Age

Male Bicyclist Fatalities by Ethnicity & Age
Ethnicity	Age Category														Total
	< 6		6-15		16-20		21-29		30-59		60+		Unknown
	N	%	N	%	N	%	N	%	N	%	N	%	N	%	N	%
Non-Hispanic White	19	1.4%	303	21.9%	107	7.7%	101	7.3%	668	48.4%	181	13.1%	2	0.1%	1,381	100.0%
Non-Hispanic Black	9	202%	103	25.3%	24	5.9%	29	7.1%	215	52.8%	26	6.4%	1	0.2%	407	100.0%
Non-Hispanic Other	2	2.7%	17	23.0%	6	8.1%	8	10.8%	22	29.7%	18	24.3%	1	1.4%	74	100.0
Hispanic - All	7	1.9%	43	11.6%	37	10.0%	67	18.1%	172	46.5%	42	11.4%	2	0.5%	370	100.0%
Total	37	1.7%	466	20.9%	174	7.8%	205	9.2%	1,077	48.3%	267	12.0%	6	0.3%	2,232	100.0%

Source: Fatal Accident Reporting System: 1999-2003

 

Male Bicyclist Fatalities by Ethnicity Sub-Group & Age
Ethnicity	Age Category														Total
	< 6		6-15		16-20		21-29		30-59		60+		Unknown
	N	%	N	%	N	%	N	%	N	%	N	%	N	%	N	%
Non-Hispanic White	19	1.4%	303	21.9%	107	7.7%	101	7.3%	668	48.4%	181	13.1%	2	0.1%	1,381	100.0%
Non-Hispanic Black	9	2.2%	103	25.3%	24	5.9%	29	7.1%	215	52.8%	26	6.4%	1	0.2%	407	100.0%
Non-Hispanic Other	2	2.7%	17	23.0%	6	8.1%	8	10.8%	22	29.7%	18	24.3%	1	1.4%	74	100.0%
Mexican Origin	4	1.6%	33	13.1%	23	9.1%	47	18.7%	114	45.2%	30	11.9%	1	0.4%	252	100.0%
Puerto Rican origin	0	0.0%	5	19.2%	1	3.8%	2	7.7%	13	15.0%	5	19.2%	0	0.0%	26	100.0%
Cuban origin	0	0.0%	0	0.0%	1	6.3%	2	12.5%	9	56.3%	4	25.0%	0	0.0%	16	100.0%
Central or South American origin	0	0.0%	3	6.7%	9	20.0%	8	17.8%	23	51.1%	1	2.2%	1	2.2%	45	100.0%
European Spanish or Other Hispanic	3	9.7%	2	6.5%	3	9.7%	8	25.8%	13	41.9%	2	6.5%	0	0.0%	31	100.0%
Total	37	1.7%	466	20.9%	174	7.8%	205	9.2%	1,077	48.3%	267	12.0%	6	0.3%	2,232	100.0%

Source: Fatal Accident Reporting System: 1999-2003

Figure 6 graphically depicts the number of bicycle fatalities by age group and ethnicity. The increased involvement of Hispanic bicyclist in the 21-29 age group is evident. The figure also clearly shows that, in terms of number of fatalities, the 30-59 age group has the largest number of fatalities for each of the ethnicities.

Figure 6 - Bicyclist Fatalities by Ethnicity & Age

Source: Fatal Accident Reporting System: 1999-2003

Table 13 shows the number of alcohol-related bicycle fatalities by ethnicity and gender. As shown, 13.2% (49 out of 370) of the Hispanic male fatalities involved alcohol. This is higher than the 10.0% of non-Hispanic Whites and 9.8% of non-Hispanic Blacks that had alcohol involvement. Alcohol involvement among all female bicyclists is very low. When the various Hispanic sub-groups are examined it is apparent that male Hispanics of Mexican origin (15.5%) and European Spanish or Other origin( 19.4%) have the highest level of alcohol involvement. Some of the percentages shown in this table are based on very small numbers and should be very carefully interpreted. It is nevertheless quite clear that Hispanic bicycle safety programs should address the issue of drinking and riding by Hispanics of Mexican origin.

Table 13: Alcohol-Related Bicyclist Fatalities by Ethnicity & Gender

Alcohol Related Pedestrian Fatalities by Ethnicity & Gender
Ethnicity	Males			Females			Both Genders
	Alcohol Involvement		Total Fatals	Alcohol Involvement		Total Fatals	Alcohol Involvement		Total Fatals
	N	%	N	N	%	N	N	%	N
Non-Hispanic White	138	10.0%	1,381	5	2.4%	207	143	9.0%	1,588
Non-Hispanic Black	40	9.8%	407	2	5.1%	39	42	9.4%	446
Non-Hispanic Other	8	10.8%	74	0	0.0%	12	8	9.3%	86
Hispanic - All	49	13.2%	370	1	4.3%	23	50	12.7%	393

 

Alcohol Related Pedestrian Fatalities by Ethnicity & Gender
Ethnicity	Males			Females			Both Genders
	Alcohol Involvement		Total Fatals	Alcohol Involvement		Total Fatals	Alcohol Involvement		Total Fatals
	N	%	N	N	%	N	N	%	N
Non-Hispanic White	138	10.0%	1,381	5	2.4%	207	143	9.0%	1,588
Non-Hispanic Black	40	9.8%	407	2	5.1%	39	42	9.4%	446
Non-Hispanic Other	8	10.8%	74	0	0.0%	12	8	9.3%	86
Mexican origin	39	15.5%	252	0	0.0%	14	39	14.7%	266
Puerto Rican origin	2	7.7%	26	0	0.0%	2	2	7.1%	28
Cuban origin	0	0.0%	16	0	0.0%	0	0	0.0%	16
Central or South American origin	2	4.4%	45	0	0.0%	3	2	4.2%	48
European Spanish or Other Hispanic origin	6	19.4%	31	1	25.0%	4	7	20.0%	35

Source: Fatal Accident Reporting System: 1999-2003

b. Bicyclist Crash Characteristics

The FARS database contains a number of variables that describe characteristics of the crash itself, i.e., weather, roadway characteristics, etc. This section includes cross tabulations of these variables for the major ethnicity groups -Non-Hispanic Whites, Non-Hispanic Blacks, and Hispanics. Table 14 (next page) contains distributions of the following variables for each of the three major ethnic groups:

Roadway/Area Type

Surface condition

Light Condition

Roadway Classification

Number of Lanes

Relation to Junction

Posted Speed Limit

Traffic Controls Present

These variables were selected because they either provide insight into the nature of Hispanic bicyclist crashes or because they highlight potential differences between the bicyclist crash experience of Whites, Blacks and/or Hispanics. A number of Other "accident level" variables in the FARS database were also examined but were found to provide no additional useful information or insights. These variables were:

Route Signing

Manner of Collision

Relation to Roadway

Traffic way Flow

Roadway Alignment

Roadway Profile

Roadway Surface Type

Construction/Maintenance Zone

Traffic Control Functioning

Atmospheric Conditions

Hit and Run

School Bus Related

Emergency Medical Services Information

As shown in Table 14, Hispanic bicyclists are much more involved in fatal crashes in urban areas (78.6%) than are non-Hispanic Whites (60.3%). Non-Hispanic Blacks are less involved in urban crashes (68.8%) than Hispanics, but more so than non-Hispanic Whites.

Roadway surface conditions were dry in 9 out of 10 of the fatal bicyclist crashes, regardless of the bicyclist's ethnicity.

Table 14: Selected Accident Level Variables for Bicyclist Fatalities by Ethnicity

Selected Accident Level Variables for Bicyclist Fatalities by Ethnicity
Variable	Non-Hispanic White	Non-Hispanic Black	Hispanic
	N=1588	N=446	N=393
Roadway/Area Type
Urban	60.3%	68.8%	78.6%
Rural	39.7%	31.2%	21.4%
Surface Condition
Dry	90.9%	89.9%	91.3%
Wet	8.4%	9.4%	8.7%
Snow or Ice	0.4%	0.0%	0.0%
Other/Unknown	0.3%	0.7%	0.0%
Light Condition
Daylight	58.2%	43.5%	45.3%
Dark - No Lighting	19.6%	25.1%	17.8%
Dark - Lighting	17.3%	27.6%	30.0%
Other/Unknown	4.9%	3.8%	6.9%
Roadway Classification
Arterials - Rural	13.1%	11.0%	5.9
Collectors - Rural	14.4%	10.3%	5.9%
Local Road - Rural	11.4%	8.5%	8.7%
Major Arterials - Urban	26.2%	22.4%	32.1%
Minor Arterials - Urban	14.9%	15.0%	22.6%
Collectors - Urban	3.9%	6.3%	7.1%
Local Streets - Urban	13.8%	22.9%	16.3%
Other/Unknown	2.3%	3.6%	1.5%
Number of Lanes
Two or less	74.0%	67.3%	66.2%
Three	7.2%	8.7%	12.5%
Four	13.4%	19.7%	17.0%
Five or more	3.0%	2.5%	3.1%
Unknown	2.4%	1.8%	13%
Posted Speed Limit
25 or less	8.1%	13.2%	10.4%
30-35	27.5%	35.2%	29.5%
40-45	29.9%	27.4%	36.6%
50-55	27.6%	19.5%	17.0%
Over 55	4.7%	3.6%	2.3%
Other/Unknown	2.3%	1.1%	4.1%
Relation to Junction
Non-intersection	66.7%	62.1%	57.8%
Intersection/Intersection-related	27.8%	32.5%	37.4%
Other/Unknown	5.5%	5.4%	4.8%
Traffic Controls Present
None	76.5%	73.5%	71.0%
Traffic Signal (RGA)	10.5%	9.9%	12.7%
Stop Sign	9.7%	13.0%	13.7%
Other Signs/Signals	3.3%	3.6%	2.6%

Source: FARS 1999-2003

The light conditions at the crash scene do show some interesting differences. Non-Hispanic Whites are more involved in daytime crashes (58.2%) than are Hispanics(45.3%) Hispanics are more involved in nighttime crashes where There is street lighting (30.0%) than are either non-Hispanic Whites (17.3%) or non-Hispanic Blacks (27.6%).

Figure 7 shows the distribution of bicyclist fatalities by time of day for non-Hispanic Whites, non-Hispanic Blacks and Hispanics. Except for a brief period in the early morning (5, am, 6am, 7am and 8am), where There are more Hispanics involved, the curves for all ethnicities are relatively close together. The non-Hispanic Blacks have a similar period of over-involvement in the evening at 7pm and 8pm.

Figure 7: Bicyclist Fatalities by Time of Day

Source: FARS 1999-2004

The roadway classification data shows that most of the Hispanic crashes occur on either major urban arterials (32.1%) or minor urban arterials (22.6%). As previously discussed, the Hispanic bicyclists tend to be older than the non-Hispanic White bicyclists. Apparently, they also tend to ride on more major roadways. The data on number of traffic lanes supports this trend. Two thirds (66.2%) of the Hispanic crashes occur on roadways with 2 or fewer lanes versus 74.0% of the non-Hispanic White crashes. This leaves one-third of the Hispanic crashes on roadways with 3 or more lanes. Seventeen percent occur on 4-lane roadways.

Apparently, the urban/suburban minor/major arterials where these crashes are occurring are not the stereotypical urban/suburban arterials with 3 or more lanes. The data on posted speed limit shows that most of the Hispanic bicyclist crashes occur in areas posted at 30-35 mph (29.5%) or 40-45 mph (36.6%). Although the crashes are typically occurring on 2-lane roadways, these two-lane roadways have relatively high posted speed limits.

Hispanic bicyclists are more likely to be killed at or near an intersection (37.4%) than are non-Hispanic Whites (27.8%). This is probably related to the previous described over-involvement on multi-lane arterials. This supports the belief that intersections on multi-lane roadways are especially hazardous for bicyclists,and is a potential topic for safety educational programs targeting Hispanic bicyclists.

The data on traffic controls at the crash scene supports the over-involvement of Hispanic bicyclists at intersections. There was a traffic sign at 29.0% of all Hispanic fatalities, versus 23.5% for non-Hispanic Whites and 26.5% for non-Hispanic Blacks.

C. Center for Disease Control (CDC) Data

The Center for Disease Control (CDC) maintains a database that includes fatal pedestrian and bicyclist crashes. The national Center for Health Statistics' Vital Statistics System provided information on the number of deaths and the Bureau of Census provided population data. This information was combined to determine the Average Crash Rate or the number of deaths per 100,000 population. The following three ethnicity categories were examined: Hispanic, Non-Hispanic White, and Black. To determine if country of origin may play a role in crash involvement, the data from seven selected states is presented in addition to the national statistics. As explained earlier, these seven states were selected because of the ir high proportion of Hispanics from Mexico, Central/South America, Puerto Rico and Cuba.

Table 15 presents the Pedestrian and Bicyclist Deaths and Death Rates (per 100,000) for the Hispanic, White and Black populations. The national death rate for Hispanics in pedestrian crashes is 2.88 per 100,000 population. For Whites, it is 1.78 per 100,000; 62% less. For Blacks it is 3.01 per 100,000 - very slightly (4%) more than the Hispanics rate. Clearly, in the national perspective, Hispanics and Blacks are over-represented in pedestrian crashes.

Table 15 – Pedestrian & Bicyclist Deaths & Death Rates by Ethnicity

Pedestrian and Pedal Cyclist Deaths and Death Rates (per 100,000) for the Hispanic, White and Black U.S. Population
Location	Pedestrian						Bicyclists
	Hispanic: All Races		Non-Hispanic: White		Non-Hispanic: Black		Hispanic: All Races		Non-Hispanic: White		Non-Hispanic: Black
	Number Deaths	Death Rate*	Number Deaths	Death Rate*	Number Deaths	Death Rate*	Number Deaths	Death Rate*	Number Deaths	Death Rate*	Number Deaths	Death Rate*
All 50 States	1,071	2.88	3,527	1.78	1,073	3.01	121	0.32	519	0.26	120	0.34
Selected States:
California	333	2.89	335	2.05	61	2.64	41	0.36	57	0.35	7	*0.29
District of Columbia	0	–	1	*0.62	8	*2.31	0	–	0	–	2	*0.57
Florida	97	3.38	327	3.06	86	3.49	24	0.84	75	0.70	20	0.81
Maryland	7	*2.87	58	1.72	48	3.16	0	–	10	*0.29	4	*0.26
New Jersey	23	1.96	100	1.78	37	3.20	5	*0.42	15	*0.26	6	*0.51
New York	60	2.01	264	2.22	75	2.52	7	*0.23	28	0.24	5	*0.16
Virginia	11	*3.09	75	1.48	28	1.95	1	*0.28	14	*0.27	5	*0.34
*Rates based on 20 or fewer deaths may be unstable; use with caution.

Source: Center of Disease Control - 2001 data

When the death rates in some of the selected states are examined, the differences are far less dramatic. Although the death rates for Hispanics in California are 40% higher than the rates for Whites, they are only 10% higher in Florida and New Jersey. One could hypothesize that the higher rate in California is because the more recent immigrants from Mexico in that state have more problems than in the more established Hispanic populations in Florida and New Jersey. The relatively high rates in both Maryland and Virginia, with the ir populations of recent immigrants from Central and South America, support this hypothesis. The rates, however, are based on only seven fatalities in Maryland and eleven in Virginia and may be unstable. The pedestrian death rate for Blacks is about as high or higher than that of Hispanics.

The right-hand side of Table 15 shows the bicyclist deaths and death rates for the same locations. The death rate for Hispanic cyclists is about 25% higher. As was the case with the pedestrian data, the death rate for Blacks is slightly higher than that of Hispanics. The bicyclist rate data for the individual states should be used with caution. Most of the frequencies are less than 20 and the rates may be unstable. However, in Florida, Hispanics are about 20% over-involved while There is no real difference in California. Overall, the bicyclists' death rate data is very similar to that of the pedestrians. Hispanics have somewhat higher death rates than non-Hispanic Whites while Blacks have rates comparable to those of Hispanics.

D. Miami/Dade County Data

Data files containing pedestrian crash data from 1996-2000 were obtained and analyzed. The data files identified the race/ethnicity of the pedestrians as White, Black, Hispanic or Other. Data on a total of 7,988 pedestrian crashes was analyzed. In order to calculate crash involvement rates, population figures for Miami/Dade County were obtained from the U.S. Census. This information is presented in Table 16.

Table 16 – Pedestrian Crash Rate by Ethnicity in Miami/Dade County, FL

Pedestrian Crash Rate by Ethnicity in Miami/Dad County Florida
Population	Total Population	Percent	Pedestrian Crashes 1996-2000	Average Yearly Crash Rate per 100,000	Pedestrian Fatalities 1996-2000	Average Yearly Death Rate per 100,000
Hispanic or Latino	1,291,737	57.3	1,987	30.8	126	2.0
Non-Hispanic, White	465,772	20.7	3,234	138.9	147	6.3
Non-Hispanic, Black	427,140	19.0	2,498	117.0	99	4.6
Other & Unknown	68,713	3.0	269	78.3	0	-
Total	2,253,362	100.0	7,988	70.9	372	3.3

Miami/Dade County has a total population of 2.25 million with 7,988 pedestrian crashes over the 5-year period from 1996-2000. The Average Annual Crash Rate per 100,000 is 70.9. The crash rates by race/ethnicity are very interesting. According to the Census data, over half of the population is considered Hispanic. Yet, they are involved in only 1,987 cashes - only 25% of the total crashes. This produces an Average Annual Crash Rate of only 30.8 per 100,000 - about one-quarter that of non-Hispanic Whites and non-Hispanic Blacks. The average yearly death rate per 100,000 for the Hispanics is only 1.95, less than one-third of the White death rate and less than one-half of the Black death rate. Overall, the death rate in Miami/Dade County is 3.30 per 100,000 - very close to the death rates previously reported in the CDC data.

It could be hypothesized that this low death rate for Hispanics is due to the large Cuban-American population in Miami - individuals who have been in this county for many years. Unfortunately, we do not know if the criteria used by the police department to classify a pedestrian as Hispanic are the same as the criteria used by the U.S. Census. Unless very similar definitions are used to define both the population of Hispanics and the Hispanic crashes, these rate computations may be suspect.

Even if the rate data is suspect, the Miami/Dade County crash data is still a very useful resource. It provides the opportunity to compare the crash characteristics of pedestrian crashes involving 1,987 Hispanics, 3,234 Whites and 2,498 Blacks. Cross tabs were run on all of the crash variables for the three race/ethnicity categories – Hispanic, White, Black. The following data elements were examined:

Year of Crash	Impact Speed
Day of Week	Posted Speed Limit
Time of Day	Vehicle Movement
Lighting Condition	Driver Alcohol/Drug Use
weather	Driver Race
Traffic Control Present	Driver Gender
Number of Lanes	Driver Age
Divided/Undivided Roadway	Contributing Causes - Driver
Roadway System Identifier	Number of Pedestrians
Crash Location	Pedestrian Alcohol/Drug Use
Vehicle Type	Pedestrian Gender

Overall There were no major differences in the crash characteristics of Hispanics, Whites and Blacks. Only the following variables were notable:

• Hispanics slightly more involved at locations with no traffic control present (67.8% of Hispanic pedestrian crashes vs. 63.9% of White pedestrian crashes).
• Hispanics slightly less involved in crashes at or influenced by an intersection (25.9% of Hispanic pedestrian crashes vs. 30.1% of White pedestrian crashes).
• Hispanics slight less involved in crashes on state roads (11.4% of Hispanic crashes vs. 20.4% of White pedestrian crashes) and more involved in crashes on county roads (26.1% of Hispanic pedestrian crashes vs. 17.3% of White pedestrian crashes).

Although these differences are small, the consistency between all three of these data elements does suggest that Hispanics may be crossing more at mid-block locations (jaywalking) on local streets. Educational messages could target this unsafe behavior.