In the early 1900s, roads had to be shared by horses, bicycles, pedestrians, and cars. Ideas for making roads safer were based on trial and error, anecdotes, and personal opinion.
If a city decides something works, such as widening a road, neighboring communities may also try to widen the road.
When a visiting state legislator in 1919 advocated traffic light enforcement at intersections in Phoenix, he cited his “personal experience” as evidence that such enforcement would reduce accidents.
We are now in the era of big data. There were an estimated 5.9 million police-reported crashes on U.S. roads in 2022, including more than 42,000 fatalities. Data about these crashes, not just on highways but also on neighborhood roads, is stored electronically.
As data is collected, it presents us with opportunities. We can transform data into information. We can learn from past mistakes and look at data from rigorous research to make the world safer.
Lead by example
A classic example of converting data into information is in urban road systems. Now, the Federal Highway Administration (FHWA) is helping city planners make more informed decisions.
Planners are often faced with difficult questions. Based on real data, what road changes are proven to save lives?
FHWA has created a tool to help. The Crash Modification Factors (CMF) Clearinghouse (cmfclearinghouse.org) is a searchable database of measures tried across the country to reduce crashes and loss of life.
The database contains statistics that estimate the expected reduction (and in some cases increase) in the rate of various types of crashes if this countermeasure is adopted.
For example, with recent attention to pedestrian fatalities in the Valley, planners can look to the Clearinghouse for possible crosswalk enhancements.
Among the 84 CMFs returned by a search for the word “crosswalk” are nine CMFs from a 2017 study by the National Academies of Sciences, Engineering, and Medicine on Pedestrian Crosswalk Enhancements on Multilane Urban Streets. Contains estimates.
For example, installing “pedestrian hybrid beacons” (traffic lights that are activated when a pedestrian presses a button) is estimated to reduce vehicle-pedestrian collisions by about 55 percent.
Readers can compare this statistic with other measures such as the installation of central evacuation islands (so pedestrians can focus on crossing in one direction at a time) or the installation of advanced YIELD or STOP markings to reduce different types of collisions. can be compared with the estimated decrease in These three crosswalk enhancements are also highlighted on FHWA's Proven Safety Measures website.
The CMF Clearinghouse does not give absolute answers regarding road safety measures. It's more of a teaching tool.
Each study has its own relevance and quality, and different studies may reach different conclusions. Some studies, such as the National Academies report, are carefully conducted and include numerous sites across multiple cities (this study includes sites in Phoenix, Scottsdale, and Tucson) .
Other studies may be less rigorous or include too few sites to support general conclusions. Analysts are encouraged to investigate the differences between the examples in the database and the actual implementation site.
Informed decision making
Clearinghouse data has another desirable characteristic. That is, anyone can use it for free.
Citizens with an interest in road safety can investigate and propose their own potential safety measures.
So when a city or neighborhood proposes new pavement markings, changing speed limits, converting unregulated intersections to four-way stops, or installing new “no left turn” signs, each of us can You can check to see if any research has been carried out. What kind of research has been done on these interventions, if any, and what effects have they had?
People in neighborhood associations often have strong opinions about what should be done to improve road safety, and using data can help focus the discussion on what works.
We have come a long way since the 1900s. We have the opportunity to make changes to our roads based on large amounts of data and high-quality statistical analysis. This information allows someone to spend their money on changes that are scientifically proven, rather than just what they think is a good idea based on personal experience.
Norma Hubele and Sharon Lohr are professors emeritus of statistics at Arizona State University. They write about statistical issues at theautoprofessor.com and sharonlohr.com.