DARKNESS AND DEATH IN THE U.S.: WALKING DISTANCES ACROSS THE

,


INTRODUCTION
Pedestrians make up 17% of US roadway deaths, and this number is rising (NHTSA, 2021).US pedestrian death rates are 2 to 3 times higher per capita and 5 to 10 times higher per walked-mile than those in Western European nations (Buehler and Pucher, 2021).While 10.5% of US persontrips are made by walking (this includes trips made for exercise or recreation, as well as those to workplaces, stores, schools, and other destinations), walking distances are just 0.85% of Americans' person-miles traveled (PMT) (FHWA, 2017).Those in the UK, France, Germany, and, the Netherlands, for example, walk 136%, 118%, 90%, and 45% more than the average American, respectively (Buehler, 2022).Pedestrians are very vulnerable to injury and death, especially in the face of high speeds and bigger vehicles.The US pedestrian death count rose 52% between 2009and 2019(NHTSA, 2021)).In the State of Texas, pedestrians were about 18% of all roadway deaths in 2021 (TxDOT, 2022), with a total count more than twice that of 2009 (TxDOT, 2014).While most Americans may be aware of the health benefits of walking, walktrips per household rose just 6.5% from 2001 to 2017, while pedestrian deaths as a share of total traffic crash deaths increased from 12% to 16% during the same period (NHTSA, 2019).
Roughly 75% of US pedestrian deaths occur after sunset and before sunrise (during "nighttime"), with the death count almost doubling between 2009and 2018(GHSA, 2020;;Tefft et al., 2021).Darkness and nighttime conditions come with a much higher frequency and severity of pedestrian injuries (Rahman et al. 2022;Zhao et.al., 2020).Arizona, California, Florida, Georgia and Texas (all relatively southern US states) accounted for 47% of all US pedestrian deaths while representing just 33% of the nation's population (GHSA 2020).The top 15 states with the highest pedestrian fatality rates are in the nation's southern latitudes (NHTSA 2019(NHTSA , 2022)), with New Mexico topping the list at almost 4 pedestrian deaths per year per 100,000 population.When dividing pedestrian fatalities by vehicle-miles traveled (VMT), all top ten US states are relatively southern, with Florida and Arizona taking the lead (at approximately 300 deaths per 100 billion VMT) (NHTSA, 2022).Pedestrians of lower income appear to be at greater risk, with a 1% rise in death rate for every $1000 drop in a US Census tract's median household income (Mansfield et. al, 2018).Those of American Indian origin and Alaska Natives are five times more likely to die as pedestrians, per mile-walked, than White persons; African Americans are twice as likely (Glassbrenner et al., 2022).
People wonder why pedestrians are at much greater risk of death in the US than in comparable nations.Also, why are so many US pedestrians dying at night?And what makes southern US states less safe than northern US states?This study examines the various demographic, location/position, time of day and year attributes that characterize Americans' walk-miles traveled (WMT), in order to offer daytime vs nighttime WMT values for crash rate comparisons.The 2016/17 National Household Travel Survey (NHTS) data are used in hurdle regression models to predict each respondent's WMT and "nighttime" WMT on his/her survey day (with "nighttime" hours being those between sunset and sunrise, and thus varying across the nation, by latitude, longitude, and day of year).

DATASETS ASSEMBLED
The 2016/17 NHTS dataset contains almost 924,000 person-trips (including over 80,000 walk trips) made by nearly 130,000 U.S. households with 264,000 persons from mid-April 2016 through April 2017.Walk trip distances are capped here at 3 miles to avoid very long hikes and runs that are not normally near roadways (or were mis-reported to the NHTS survey team).About 4% of walk trips were missing household income, education, race, age, gender, worker 1 status, trip miles, mode or origin/destination coordinates; so those records are not analyzed here.2 The NHTS data contain origin and destination block groups (within Census tracts) for every trip 3 sampled inside the US, and block group centroids were used for trip origin coordinates.There 4 are 239,780 distinct block groups in the US (US Census Bureau, 2022) and 106,592 within the 5 NHTS trips' origin and destination zones.In addition to a walk-trip's starting position, the trip's 6 timing is key, since 75% of US pedestrian deaths occur in darkness (GHSA, 2020).Python's 7 'Suntime' library was used to find sunset and sunrise in Coordinated Universal Time (UTC), 8 given trip starting coordinates and date (with NHTS sampling trips from April 2016 through 9 April 2017).UTC time was converted to the local time of trip location with the help of a Python 10 package called 'TimezoneFinder', to determine whether respondent-reported walk-trip start 11 times were before sunrise or after sunset (and labeled "night" or "nighttime").Given the high number of observations with a 0 value (making no walk trips for the first model and no walk trips at night for the second), hurdle regression model was considered appropriate as it offers flexibility in its interpretation.So, a hurdle specification splits the dependent variable into two parts: the logistic probability of a respondent not walking at all (and not walking at all at night) on the survey day and an exponential density function for all positive WMT (and positive nighttime WMT) possibilities (Cragg, 1971).Measure of each explanatory variable's practical significance was included for both models.Practical significance was found by increasing each covariate by 1 standard deviation (SD), in every person-day record, and taking the ratio of average WMT predictions after vs before the increase.It indicates the change in average WMT per 1 SD increase.A positive sign indicates increase in WMT, while negative indicates decrease.Population weights were applied to all models to ensure that parameter estimates better reflect the US population.

RESULTS AND ANALYSIS
Table 2 provides parameter estimates for the WMT/person/day model, along with measures of each explanatory variable's practical significance.Workers, those without college degrees, Caucasians, older persons, and those in higher-income households are less likely to walk on the survey day.Gender was not important for whether or not to take a trip, but males walk about 10% more per day with a 1 SD increase in the indicator variable (after capping each walk trip at 3 miles).Education level is practically very significant.Table 3 provides parameter estimates for the "Nighttime" WMT model, along with measures of practical significance.All covariates in Table 1 were initially included, and then systematically removed if they had a p-value greater than 0.10.There is a higher chance of making a "nighttime" walk trip on the survey day if the person has a college degree, is male, African American, worker and/or of lower household income.The second, exponential regression estimating those "nighttime" walk distances indicate similar trends as the previous model.A 1 SD increase in indicator variables for Bachelor's and Graduate degrees corresponds to a +36.6% and 46.6% increase in miles walked per day person, respectively.Similarly, increasing Texas and Wisconsin resident indicator variable by 1 SD decreases daily WMT predictions by -47.0% and -53.6% (per respondent), respectively.A 1 SD increase in indicator for trips originating in the central U.S (between -125º and -85º longitude) raises average WMT per person per day by +31.4%.Estimates also indicate that 1 SD increase in indicator variable for 45º -50º latitude band lowers average WMT per person at night by -17.8%.This could suggest that lower pedestrian fatality rates in northern states compared to southern states may be attributed to people in the northern region walking less during darkness.Results show how race, education, age and income status have statistically significant effects on walk-mode and walk distance choices in this large data set.Resident state and longitudes (locations) also have the great impacts, with residents from southern states walking shorter distances (even at night).Length of daylight also has statically significant effects on walk distances.Intriguingly, Americans in southern US locations (which enjoy more sunshine/better lighting and warm weather) do not walk as much, and face higher pedestrian crash risk per mile walked.Dividing pedestrian fatalities (average of 2014 to 2019) (FHWA, 2017;NHTSA, 2022) by WMT extrapolations for each state's resident population (and visitors, which are key in Washington DC's and Delaware's cases), places all southern states among the nation's 10 most deadly, with Alabama and Mississippi topping the list (with 1.01 and 0.98 pedestrian deaths per million WMT). Figure 5 displays a map ranking Pedestrian Fatality by WMT for all US states.Higher ranking states are all concentrated in the southern latitudes.Alongside all these insights, the mystery of why pedestrian death rates (vehicle and walk-mile traveled) are so high or more often at night in southern states remains.Model results do not suggest that more "nighttime" walking is responsible for higher pedestrian fatalities in southern regions.Neither traffic fatality rates due to alcohol consumption nor overall alcohol consumption per capita support this result (NHTSA, 2022; Wisevoter 2023), but hard-alcohol (ethanol) consumption per capita is highest in California, Florida, and Texas (with Georgia falling close behind) (Vinepair 2020).Differences in built environments and car-centric cultures may also play a role.Southern-latitude states are 7 of the top 10 in VMT per capita, while only three states (DC, Hawaii & California) in the southern latitude appears in the ten states with the lowest annual VMT per capita (U.S.Department of Energy, 2019) and regularly lead in crash deaths per capita (NHTSA, 2022).1.44 The model's findings suggest that in addition to the demographic and geographic factors mentioned earlier, other factors such as lack of enforcement, poor design of road infrastructure, weaker licensing laws, and driver culture may contribute to the higher pedestrian fatality rates observed in many southern states.These systemic issues can significantly impact the safety of pedestrians and all other travelers on the roads.
Addressing these challenges and improving road safety in the southern states, as well as across the entire United States, is crucial.It is imperative to prioritize and invest in comprehensive measures that enhance traffic law enforcement, promote safer road design and infrastructure, strengthen licensing regulations, and foster a positive driver culture that prioritizes pedestrian and traveler safety.
By taking proactive steps to address these issues, the United States, particularly its southern states, can work towards ending the streak of daily deaths among pedestrians and all other travelers.This collective effort would not only save lives but also create safer and more sustainable communities for everyone.It is a shared responsibility to make our roads safer and protect the lives of pedestrians and travelers across the country.

Figure 1 .
Figure 1.Average WMT per Person per Day by Day of Week and Trip Origin 14

Figure 2 .
Figure 2. Average WMT per Person per Day by Month, for Northern vs Southern Trip Origins (divided by 40° latitude as shown on USA map)

Figure 3 .
Figure 3. Average WMT per Person per Day by Month, for Trip Origin by Latitude

Figure 4 .
Figure 4. Portion of Walk-Miles by Latitude Trip Origin and Season

Figure 5 .
Figure 5. Rankings of Pedestrian Deaths per Walk-Mile Travelled (WMT) across US States

Table 1 . Summary Statistics of 2016/17 NHTS Person Records (n = 254,295 respondents)
*Represents indicator variables and they have a value of 1 if true/yes and 0 if false/no (E.g.-For 1 the variable Male, 1 = Male and 0 = Female).2 DATA ANALYSIS 3 Figures 1, 2, 3 and 4 highlight different WMT choices by day of week, latitude, month of year 4 and darkness.Weekdays, particularly Mondays and Thursdays, are more popular for walking, 5 with approximately 20% more walking distance per person compared to weekends.Walking is 6 highest in Hawaii (per person-day), and this is not due to non-resident visitors (who make up 7 45% of the Washington DC WMT, and just a few percent of Hawaii's).Unexpectedly, those 8 living in northern US locations (above 40° latitude) walk more than those in southern 9 (continental) settings, even though those in the northern locations experience up to 12% less 10 daylight during the non-summer months (and up to 7% more daylight during summer months) 11 when comparing daylight hours between 35° N and 45° N latitudes (United States Naval 12 Observatory, 2019).13

Table 2 . Model Estimates for WMT Per Person Per Day Using 2016/17 NHTS Data Logistic Selection Model for Pr(WMT > 0)
Increasing the Bachelor's and Graduate degree indicator variables by 1 SD increases daily WMT predictions by +37.6% and +49.16% (per respondent), respectively.Increasing Age indicator variable by 1 SD decreases WMT predictions per day by almost 40%.Location is also practically very significant.Interestingly, a person's resident state, where most of their walk trips occur, is practically very significant for some states.Georgia, California, South Carolina, Wisconsin, and Texas residents walk -32.9%, -35.1%, -39.0%, -51.6 and -58.5% less miles per day with 1 SD increase in indicator variables for those respective resident states.A 1 SD increase in indicator variable for trips originating at longitude greater -85º (east coast) raises average WMT by ~30% per day per person, while keeping all other variables constant.Length of daylight also indicates 22.7% more walking per day (per person) with a 1 SD increase in that variable.

Table 3 . Model Estimates for WMT Per Person Per Day at "Night" using 2016/17 NHTS Data Logistic Selection Model for Pr(Nighttime WMT > 0) Coef. T-stat P- value Pract. Sign.
Asterisked variables have their practical significance values shown in the exponential regression model for the same variable.All covariates with a p-value less than 0.10 were removed.
Table 4 displays pedestrian death rates by WMT, VMT, per capita and by alcohol consumption for all US states.