Abstract

Do non-English speaking drivers pose a safety hazard to others? The argument that non-English speakers cause more accidents than natives has been used for a while to call for English only driver’s license tests.

“We welcome non-English speaking people, who are legally in the U.S., to Alabama. However, if you want to drive in our states, public safety concerns dictate that you need to speak English.” – Tim James, 2010 Alabama gubernatorial race

More recently comments on Facebook such as

“…please excuse me for my following comment: but this is probably WHY the reason the driver test should only been taken in ENGLISH…‼️‼️‼️‼️ if people don’t know the language, they become a danger to themselves and others…‼️‼️‼️‼️”

“The DMV is the one to blame that make them take the test in different Chinese dialects because they don’t speak English and all our street signs are in English. No common sense at all…. well if you are living in the USA there should be a law that only give the test in one language only English no discrimination.”

This argument has been disproved in the past by looking at correlation between immigration and accidents data.

Introduction

In this paper we used two datasets serving as surrogates for Non-English Speakers and Auto Accidents.

We obtained 2016 data for US New Permanent Residents from the US Department of Homeland Security, and Fatal Car Accidents by State from the Insurance Institute for Highway Safety (2016 is the last year available from the IIHS).

The data was standardized as Percent of each state’s population (also from IIHS), in order to be able to compare them between states. As a simple comparison, we calculated the correlation between the Fatal Car Accidents Percent and New Permanent Residents Percent. If in fact, non-English speakers cause more auto accidents, we would have found that states with high percent of New Permanent Residents will have high percent of Fatal Car Accidents.

For 2016, we found a very low negative correlation between between Fatal Car Accidents Percent and New Permanent Residents Percent. States with high Fatal Crashes Percent had low New Permanent Residents Percent and states with low Fatal Crashes Percent had high New Permanent Residents Percent. However, this relationship was weak (measured between -1 and 1).

Data

Fatal Crashes by State 2016

The number and types of motor vehicle crash deaths differ widely among the 50 states and the District of Columbia. A state’s population has an obvious effect on the number of motor vehicle deaths. Fatality rates per capita and per vehicle miles traveled provide a way of examining motor vehicle deaths relative to the population and amount of driving. However, many factors can affect these rates, including types of vehicles driven, travel speeds, rates of licensure, state traffic laws, emergency care capabilities, weather, and topography.

Insurance Institute for Highway Safety 2016 https://www.iihs.org/iihs/topics/t/general-statistics/fatalityfacts/state-by-state-overview.

Persons Obtaining Lawful Permanent Resident Status By State in 2016

The 2016 Yearbook of Immigration Statistics is a compendium of tables that provide data on foreign nationals who are granted lawful permanent residence (i.e., immigrants who receive a “green card”), admitted as temporary nonimmigrants, granted asylum or refugee status, or are naturalized. The Yearbook also presents data on immigration enforcement actions, including apprehensions and arrests, removals, and returns.

US Department of Homeland Security 2016 https://www.dhs.gov/immigration-statistics/yearbook/2016/table4

Standadizing Data for Each State

To be able to compare the data from different states, one needs to standardize the data by transforming both, New Permanent Residents and Fatal Car Crashes, to percentages of each State Population.

State New Permanent Residents Fatal Crashes Population New Permanent Residents % Fatal Crashes %
Alabama 4,736 937 4,863,300 0.097 % 0.019 %
Alaska 1,726 78 741,894 0.233 % 0.011 %
Arizona 20,694 865 6,931,071 0.299 % 0.012 %
Arkansas 3,158 488 2,988,248 0.106 % 0.016 %
California 223,141 3,357 39,250,017 0.569 % 0.009 %
Colorado 14,225 558 5,540,545 0.257 % 0.01 %
Connecticut 12,669 281 3,576,452 0.354 % 0.008 %
Delaware 2,204 116 952,065 0.231 % 0.012 %
District of Columbia 3,114 26 681,170 0.457 % 0.004 %
Florida 136,337 2,933 20,612,439 0.661 % 0.014 %
Georgia 29,572 1,422 10,310,371 0.287 % 0.014 %
Hawaii 6,285 109 1,428,557 0.44 % 0.008 %
Idaho 2,562 232 1,683,140 0.152 % 0.014 %
Illinois 43,207 1,003 12,801,539 0.338 % 0.008 %
Indiana 9,946 768 6,633,053 0.15 % 0.012 %
Iowa 5,299 356 3,134,693 0.169 % 0.011 %
Kansas 5,709 381 2,907,289 0.196 % 0.013 %
Kentucky 7,098 763 4,436,974 0.16 % 0.017 %
Louisiana 5,784 704 4,681,666 0.124 % 0.015 %
Maine 1,748 151 1,331,479 0.131 % 0.011 %
Maryland 26,077 472 6,016,447 0.433 % 0.008 %
Massachusetts 35,706 359 6,811,779 0.524 % 0.005 %
Michigan 22,569 980 9,928,300 0.227 % 0.01 %
Minnesota 15,603 357 5,519,952 0.283 % 0.006 %
Mississippi 2,149 628 2,988,726 0.072 % 0.021 %
Missouri 6,868 868 6,093,000 0.113 % 0.014 %
Montana 566 171 1,042,520 0.054 % 0.016 %
Nebraska 5,654 194 1,907,116 0.296 % 0.01 %
Nevada 11,555 303 2,940,058 0.393 % 0.01 %
New Hampshire 2,332 130 1,334,795 0.175 % 0.01 %
New Jersey 56,187 569 8,944,469 0.628 % 0.006 %
New Mexico 4,104 358 2,081,015 0.197 % 0.017 %
New York 159,878 965 19,745,289 0.81 % 0.005 %
North Carolina 20,811 1,348 10,146,788 0.205 % 0.013 %
North Dakota 1,595 102 757,952 0.21 % 0.013 %
Ohio 17,251 1,053 11,614,373 0.149 % 0.009 %
Oklahoma 5,960 624 3,923,561 0.152 % 0.016 %
Oregon 10,033 446 4,093,465 0.245 % 0.011 %
Pennsylvania 27,217 1,088 12,784,227 0.213 % 0.009 %
Rhode Island 4,194 48 1,056,426 0.397 % 0.005 %
South Carolina 5,104 936 4,961,119 0.103 % 0.019 %
South Dakota 1,229 103 865,454 0.142 % 0.012 %
Tennessee 10,032 966 6,651,194 0.151 % 0.015 %
Texas 110,651 3,407 27,862,596 0.397 % 0.012 %
Utah 7,271 259 3,051,217 0.238 % 0.008 %
Vermont 886 57 624,594 0.142 % 0.009 %
Virginia 29,242 722 8,411,808 0.348 % 0.009 %
Washington 27,304 504 7,288,000 0.375 % 0.007 %
West Virginia 928 250 1,831,102 0.051 % 0.014 %
Wisconsin 7,111 544 5,778,708 0.123 % 0.009 %
Wyoming 462 100 585,501 0.079 % 0.017 %

Relation between Fatal Accidents Percent vs. New Permanent Residents Percent

The plot, as well as the correlation, show weak negative relationships betwen Fatal Crashes Percent and New Permanent Residents Percent for 2016 in the US. In other words, states with high Fatal Crashes Percent have low New Permanent Residents Percent and states with low Fatal Crashes Percent have high New Permanent Residents Percent.

Correlation between Fatal Accidents Percent and New Permanent Residents Percent

Since the relationship is not linear, we looked at both the Spearman and Kendall correlations.

The Spearman correlation between two variables is equal to the Pearson correlation between the rank values of those two variables; while Pearson’s correlation assesses linear relationships, Spearman’s correlation assesses monotonic relationships (whether linear or not).


    Spearman's rank correlation rho

data:  New.Permanent.Residents.Percent and Fatal.Crashes.Percent
S = 37466, p-value = 6.054e-08
alternative hypothesis: true rho is not equal to 0
sample estimates:
       rho 
-0.6952941

The Kendall rank correlation coefficient is a statistic used to measure the ordinal association between two measured quantities


    Kendall's rank correlation tau

data:  New.Permanent.Residents.Percent and Fatal.Crashes.Percent
z = -5.3688, p-value = 7.927e-08
alternative hypothesis: true tau is not equal to 0
sample estimates:
       tau 
-0.5184314

Appendix

State Fatal Crashes per State Population

Below are the five tops states with Fatal Crashes as a percent of the population.
State Fatal Crashes Population Fatal Crashes % of Population
Mississippi 628 2,988,726 0.021 %
Alabama 937 4,863,300 0.019 %
South Carolina 936 4,961,119 0.019 %
New Mexico 358 2,081,015 0.017 %
Kentucky 763 4,436,974 0.017 %

State New Permanent Residents per State Population

Below are the five tops states with New Permanent Residents as a percent of the population.
State New Permanent Residents Population New Permanent Residents % of Population
New York 159,878 19,745,289 0.81 %
Florida 136,337 20,612,439 0.661 %
New Jersey 56,187 8,944,469 0.628 %
California 223,141 39,250,017 0.569 %
Massachusetts 35,706 6,811,779 0.524 %