Behaviors and Attitudes


NHTSA studies behaviors and attitudes in highway safety, focusing on drivers, passengers, pedestrians, and motorcyclists. We identify and measure behaviors involved in crashes or associated with injuries, and develop and refine countermeasures to deter unsafe behaviors and promote safe alternatives.

Our recently published reports are listed chronologically below. To the right are additional resources including Behavioral Research Notes and Traffic Techs.

124 Results
Distracted Driving Enforcement Demonstrations: Lessons Learned

At any given daylight moment in 2015, approximately 542,000 drivers were using cell phones or manipulating electronic devices while driving; bout 156.7 billion text messages were sent every month. There were 3,477 people killed in crashes involving distracted drivers, and another 391,000 people injured. Of these 456 fatal crashes reported involving drivers with cell phones. This guide describes three NHTSA-funded enforcement demonstration programs conducted at six sites in Syracuse, New York, California, Delaware, Connecticut, and Massachusetts and presents key lessons learned, along with insights and ideas for law enforcement agencies and State highway safety offices to consider as tactics for combating distracted driving using enforcement.

Buckle Up Owensboro: Implementation and Evaluation of the 2015-2016 Seat Belt Campaign in Owensboro, Kentucky

This study evaluates the process, outcome, impact, and sustainability of the Buckle Up Owensboro program. NHTSA and the Owensboro, Kentucky, Police Department designed the program to include sustained routine seat belt enforcement and Seat Belt Coalition activities to spread awareness of the importance of seat belt use. The program ran from mid-October 2015 to mid-October 2016. The evaluation found increased observed seat belt use associated with the 12-month program. To determine if the program could be sustained, the research team collected process, outcome, and impact data for the 6-to-12-month period following the program. Data showed a decline in observed seat belt use as program activity levels decreased. While the program was successful at increasing observed seat belt use, the evaluation suggests higher levels of program activity are needed to sustain the effects on observed seat belt use than were implemented in the post-program period.

Alcohol Ignition Interlock Use Rates Following Changes in Interlock Legislation

Ignition interlocks are effective at preventing driving under the influence (DUI) of alcohol while installed on vehicles. However, the rate of interlock use is low relative to the numbers of DUI arrests and convictions, even in States that mandate interlock use for all DUI offenders. Accordingly, there is interest in identifying ways to increase interlock use, including by expanding the types of offenses for which an offender is eligible or required to use an interlock. This study examined how changes to interlock law affected interlock use in Florida and West Virginia. The study compared the number of new interlock installations, interlocks-in-place (interlocks currently installed), installations as a proportion of those eligible to use interlocks, and lack of use or low use of the vehicles with interlocks, before and after the States modified their laws. Florida data showed increases of 21.8%  percent in all DUI offenders mandated to install; 69% percent in first offenders mandated to install; 122.3% percent more installations overall; and 27.8% percent more installations for first offenders mandated to install. West Virginia data showed an increase of 242% percent in interlock installations after a 2010 change in the interlock law, and an increase of 60% percent after a 2014 change to interlock law.

Alcohol Ignition Interlock Use Rates Following Changes in Interlock Legislation [Traffic Tech]

This Traffic Tech presents a summary of the final report of the same name under DOT HS 812 989.

Quantifying Drowsy Driving – Traffic Safety Facts - Research Note

According to NHTSA’s National Center for Statistics and Analysis, drowsy driving was a contributing factor in 775 deaths in 2018, or 2.1% of total fatalities involving motor vehicle crashes on U.S. roadways. However, statistics, likely underestimate the extent of the drowsy driving problem. The American Automobile Association Foundation for Traffic Safety estimated the percentage of fatal crashes that involved a drowsy driver at 21%. This Research Note explores the feasibility of using machine learning algorithms to identify drowsy driving episodes in large-scale naturalistic driving datasets. The Strategic Highway Research Program 2 (SHRP2) Naturalistic Driving Study (NDS) collected data from more than 3,400 participating drivers in six States, yielding time series data for more than 5 million trips. The SHRP2 NDS database contains vehicle variables and video data, including a view of the driver’s face.

Drug and Alcohol Prevalence in Seriously and Fatally Injured Road Users Before and During the COVID-19 Public Health Emergency

A gap in knowledge exists regarding drug use among drivers and other road users seriously or fatally injured in crashes in the United States. This study examines the prevalence of alcohol and over-the-counter, prescription, and illegal drugs in the blood of seriously or fatally injured crash victims near the time of their crashes before and during the COVID-19 public health emergency. Data from 3,003 participants was collected at Level 1 trauma centers and medical examiner offices. The trauma centers and medical examiners made available blood for toxicological analyses from that collected during normal clinical procedures. The results indicate drug prevalence was high among seriously and fatally injured roadway users before the emergency began and was even higher during, especially for alcohol, cannabinoids (active THC), and opioids. Drivers showed significantly higher overall drug prevalence, with 64.7 percent testing positive for at least one active drug, compared to 50.8 percent before. Drivers also showed an increase in testing positive for two or more categories of drugs going from 17.6 percent before to 25.3 percent during the public health emergency. Of particular note, active THC was more prevalent among drivers during the public health emergency than alcohol (32.7% versus 28.3%), and opioid use among drivers almost doubled from 7.5 percent to 13.9 percent. The results suggest the highway safety community should be concerned about the impact of other drugs as well as alcohol. In particular, the observed cannabis and opioid prevalence rates before and during the public health emergency could be indicative of a growing problem.

Examination of the Traffic Safety Environment During the Second Quarter of 2020: Special Report

The COVID-19 public health emergency has affected most people in America. NHTSA is taking a close look at changes in roadway travel and driver behavior since the start of the emergency, with an emphasis on the second quarter (Q2) of 2020. We are learning about the impact on motor vehicle crashes and fatalities, and this report draws on an array of sources to understand our current traffic safety environment and better address our changing traffic safety needs. This report explores changes in countermeasure use, including traffic enforcement and public communications and outreach. This report also examines whether some people who continued driving even when many communities had stay-at-home guidelines may be inherently higher-risk drivers. The report draws on EMS and hospital trauma center data and renews focus on strategies to address these new concerns.

Evaluation of Correct Child Restraint System Installation

This research project used an experimental design, called an “incomplete factorial” type, with a convenience sample of 75 novice and 75 experienced child restraint system (CRS) users to test whether user experience, child’s age/weight/height, vehicle characteristics, and CRS characteristics are associated with installation errors. This study identified conditions related to correct and incorrect CRS use to inform programming and education with the goal of increasing correct use. The results help frame the target population for programming and education as not only novice users, but also experienced users, as the study did not find a significant difference in errors by experience.

The Role of Law Enforcement in Supporting Pedestrian and Bicyclist Safety: An Idea Book

This “idea book” provides ideas and suggestions based on shared evidence around effective roles that law enforcement personnel can take in pedestrian and bicyclist safety programs and practices. It also serves as a framework for integrating a pedestrian and bicyclist safety program with law enforcement agencies by focusing on training, crash data reporting, identifying goals with partners, collaborating with traffic engineers, engaging the community, and measuring programs and plans. It can help States, local communities, law enforcement, and political support enforcement in addressing pedestrian safety.  Research studies show that law enforcement efforts are effective when properly implemented to affect the reduction of traffic related fatalities.

Older Drivers’ Self-Regulation and Exposure

This project examined relationships among older drivers’ functional abilities as assessed by a clinical test battery, behind-the-wheel (BTW) driving performance during an on-road evaluation, and naturalistic driving behaviors captured by video and tracking devices over a month of driving. The goal was to gain insights into the extent to which older adults modify their driving behaviors as their functional skills decline. Many of the functional assessment measures predicted performance on the BTW evaluation. Analyses of the exposure data showed that many participants with poorer functional and BTW scores limited their driving. The findings suggest that many people who experience functional declines regulate when and where they drive; however, some of those with the poorest functional and/or BTW scores did not appear to limit their driving.

For Access to older content please go to our archived Research page.