T RAFFIC S AFETY F ACTS

The National Highway Traffic Safety Administration's (NHTSA) National Center for Statistics and Analysis (NCSA) conducted a 15-passenger van tire pressure study. The purpose of the study was to determine the extent of underinflation and observe the tire condition in 12-and 15-passenger van (and vans with similar body styles) tires. The following topics are covered in this research note: sampling, data collection methodology, analysis and results.


Summary
The percentage of passenger vehicle drivers talking on handheld phones decreased from 2.6 percent in 2020 to 2.5 percent in 2021 (Figure 1 and Table 1).The percentage of drivers speaking with visible headsets while driving remained unchanged from the previous year's level of 0.4 percent (Figure 1 and Table 2).Drivers' visible manipulation of handheld devices stood at 3.4 percent in 2021 (Figure 1 and Table 3).In 2021 data collected for drivers visibly manipulating handheld devices while traveling included manipulating infotainment systems.This was an errant departure from the usual data collection protocol.As such, the data point for visible manipulation of handheld devices in 2021 is not connected to the rest of the trend line.For 2022 the data collection procedure will revert to excluding manipulating infotainment systems.NHTSA does not know the extent to which manipulating infotainment systems contributed to the visibly manipulating handheld devices results.These results are from the National Occupant Protection Use Survey (NOPUS), which provides the only nationwide probability-based observed data on driver electronic device use in the United States.The National Center for Statistics and Analysis (NCSA) of the National Highway Traffic Safety Administration conducts the NOPUS.The percentages provided in this research note are interpreted as the percentage of drivers nationwide during an average daylight moment.

Results
NOPUS observes three types of driver electronic device use while driving: "holding phones to their ears," "speaking with visible headsets on," and "visibly manipulating handheld devices."The results of these observations follow.

Drivers Holding Phones to Their Ears While Driving
The percentage of drivers holding cellphones to their ears while driving decreased from 2.6 percent in 2020 to 2.5 per-cent in 2021 (Figure 1 and Table 1).This translates to an estimated 373,066 passenger vehicle drivers holding cell phones to their ears while driving at a typical daylight moment in 2021.An estimated 7.6 percent of drivers were using some type of phone, either handheld or hands-free, at a typical daylight moment in 2021 (refer to the section Estimating Overall Cellphone Use, Both Handheld and Hands-Free for more details on this estimate).
The 2021 NOPUS found that handheld cellphone use continued to be higher among female drivers than male drivers, and the difference has been increasing since closing the gap in 2019 (Figure 2 and Table 1).In previous years, handheld cellphone use was found to be highest among 16-to 24-yearold drivers and lowest among drivers 70 and older except year 2020.This remains the same in year 2021 (Figure 3 and Table 1).Handheld cell phone use increased among 16-to 24-year-old drivers after consecutive years of decreases.
The only subcategory that experienced a statistically significant change from the previous year is drivers traveling during the weekends which increased from 1.6 percent in 2020 to 2.2 percent in 2021.

Drivers Speaking With Visible Headsets on While Driving
Table 2 shows the percentages of drivers speaking with visible headsets on while driving in 2020 and 2021, by major characteristics.No characteristic had a statistically significant change from the previous year.
The percentage of drivers speaking with visible headsets remained unchanged at 0.4 percent, as shown in Figure 1 and Table 2. Figure 4 displays the trend of drivers speaking with visible headsets for the three age groups over a 10-year period.There was no change in headset use from 2020 to 2021 for the age group 25 -69, but the percentage dropped from 0.4 percent to 0.2 percent for the age group 16 -24.

Drivers Visibly Manipulating Handheld Devices While Driving
The percentage of drivers visibly manipulating handheld devices while driving increased from 2.8 percent in 2020 to 3.4 percent in 2021 (Figure 1 and Table 3); this was not a statistically significant increase.Data collected for drivers visibly manipulating handheld devices while traveling in 2021 included manipulating infotainment systems.This was an errant departure from the usual data collection protocol.As such, the data point for visible manipulation of handheld devices in 2021 is not connected to the rest of the trend line.For 2022 the data collection procedure will revert to excluding manipulating infotainment systems.NHTSA does not know the extent to which manipulating infotainment systems contributed to the visibly manipulating handheld devices results.Figure 5 shows that driver manipulation of handheld devices continued to be higher among 16-to 24-year-olds than other age groups.Table 3 presents the percentages of drivers visibly manipulating handheld devices in 2020 and 2021 by major characteristics.Several characteristics experienced statistically significant increases from the previous year as listed below.
■ Drivers who appear to be a race other than White or Black from 2.5 percent in 2020 to 4.4 percent in 2021.
■ Drivers on surface streets from 2.7 percent in 2020 to 3.3 percent in 2021.
■ Drivers traveling through clear weather conditions from 2.8 percent in 2020 to 3.5 percent in 2021.
■ Drivers of pickup trucks from 1.7 percent in 2020 to 2.5 percent in 2021.
■ Drivers in the west from 2.5 percent in 2020 to 4.6 percent in 2021.
■ Drivers in objectively characterized urban areas from 3.4 percent in 2020 to 4.3 percent in 2021.
■ Drivers traveling during weekdays from 2.9 percent in 2020 to 3.7 percent in 2021.
■ Drivers traveling during weekday non-rush hours from 3.1 percent in 2020 to 4.1 percent in 2021.
It is unknown whether these increases were due to an incorrect modification to the data collection procedure for 2021 whereby drivers manipulating the infotainment system were counted.The percentage of drivers speaking with visible headsets while driving, based on the subjective assessments of roadside observers. 3The Wilson confidence interval is used in the estimated percentages in the driver group (e.g., drivers in urban areas), which is in the form: {(2n EFF p + t 2 ) ± t√(t 2 + 4n EFF pq)}/ (2(n EFF + t 2 )), where p is the estimated percentage of drivers holding phones to ears, n EFF = n/D EFF is the effective sample size (where n is the sample size and D EFF is the design effect), t = t (1-α/2) (df), is a multiplier from the t-distribution with df degrees of freedom, and q = 1 -p.For percentages, these endpoints are multiplied by 100. 4 The regular symmetric interval was used for the estimated change in percentage point, which is in the form: p ± t (1-α/2) (df)√v(p), where p is the estimated change in percentage point, v(p) is its estimated variance, and t (1-α/2) (df) is a multiplier from the t-distribution with df degrees of freedom.The degrees of freedom used in 2021 is different from that used in 2020. 5A p-value of 0.05 or less indicates that there is a statistically significant difference (at the alpha=0.05level) between the 2020 and 2021 estimates for the group in question, indicated with bold type. 6Age, gender, and racial classifications are based on the subjective assessments of roadside observers. 7The "Change in Percentage Points" column was computed using unrounded estimates and may not equal the difference between the rounded estimates displayed in the table.NA: Data not sufficient to produce a reliable estimate.Sources: NOPUS, NCSA, 2020,2021

NOPUS Data Collection and Estimation
NOPUS is the only nationwide probability-based observational survey of driver electronic device use in the United States.The survey observes usage as it occurs at randomly selected roadway sites and thus provides the best tracking of the extent to which people in the United States use cellphones and other electronic devices while driving.
The survey data is collected by trained data collectors at probabilistically sampled intersections controlled by stop signs or stoplights, where data collectors observe, from the roadside, drivers and other occupants of passenger vehicles.Data is collected from about 7 a.m. to 6 p.m.Only stopped vehicles are observed to allow time to collect the variety of information required by the survey, including subjective assessments of occupants' age and race.Observers collect data on the driver, right-front passenger, and up to two passengers in the second row of seats.Observers do not interview occupants, so that NOPUS can capture the untainted behavior of occupants.The 2021 NOPUS data was collected from June 7 to June 25, 2021.The 2020 NOPUS data was collected from July 27 to August 16, 2020, which is 2 months later than the usual time frame due to the coronavirus pandemic.In a typical year, data is collected in early June, immediately following the Click It or Ticket campaign.
Statistically significant increases and decreases in the use of handheld phones, headset use, and manipulation of handheld devices from 2020 to 2021 are shown, respectively, in Table 1, Table 2, and Table 3 by having a p-value of 0.05 or less in the tables' far-right column.
The NOPUS uses a complex multistage probability sample, statistical data editing, imputation of unknown values, and complex estimation procedures.The sample sites for the 2021 NOPUS were entirely from the 2015 NOPUS sample redesign.Table 4 shows the observed sample sizes of the 2021 NOPUS.There were 50,913 vehicles observed at the 1,624 data collection sites.Due to ineligibility, construction, danger in the area, or road closure, the observations could not be completed at some of the sampled observation sites.

NOPUS Categories and Definitions
NOPUS observes three types of driver electronic device use while driving: "holding phones to their ears," "speaking with visible headsets on," and "visibly manipulating handheld devices." Drivers are counted as "holding phones to their ears" if they are holding to their ears what appear to the data collectors to be phones.This would include behaviors such as drivers engaging in conversation, listening to messages, or conducting voice-activated dialing while holding phones to their ears.However, a data collector may not have knowledge of various types of wireless phones.Thus, the device that has been identified as a "phone" may only reflect his/her conception of what constitutes a "phone." Drivers are counted as "speaking with visible headsets on" if they appear to be speaking and wearing a headset with a microphone.This would include behaviors such as talking, engaging in conversation, or conducting voice-activated dialing via a wireless earpiece on the driver's right ear or via an ear bud connected by wire to a cellphone.Talking via a visible Bluetooth headset (usually on the driver's right ear) would also be included in this category.Note that the wireless earpieces that are obscured by hair or clothing or are on the driver's left ear would not be included because they would not be visible to the roadside observer.In addition, some wireless ear buds would not be included as they are too small to be observed from the roadside.The drivers with headsets who are not speaking at the time of observation are not included because they might have recently completed a call or be waiting for an expected call.Each driver in the survey is observed for about 10 seconds before the data collector decides whether or not the driver is speaking.Also, note that the drivers counted as speaking through a visible headset might have been talking to a passenger or using voiceactivated computer software rather than using a phone.
Drivers are counted as "visibly manipulating handheld devices" if they appear to be manipulating some type of electronic device such as a cellphone, a smart phone, tablet, video game, or some other device.This would include behaviors such as text messaging, using a web-capable smart phone (e.g., an iPhone) or a tablet (e.g., iPad) to view travel directions, check emails or calendar appointments, or surf the internet, manual dialing, playing handheld games, and holding phones in front of their faces to converse or check messages via speakerphone or use voice-activated dialing.Manipulation of the non-handheld devices (adjusting volume on stereos, pressing buttons on a dashboard GPS unit, etc.) is not included in this category.Also, note that a driver characterized by the survey as "manipulating handheld device" may or may not have been speaking.In 2021, data collected for drivers visibly manipulating handheld devices while traveling included manipulating infotainment systems.This was an errant departure from the usual data collection protocol.For 2022, the data collection procedure will revert to not including manipulating infotainment systems.NHTSA does not know the extent to which manipulating infotainment systems contributed to the visibly manipulating handheld devices results.
There are means by which the drivers can use cellphones that would neither be recorded as "holding phones to their ears" nor as "speaking with visible headsets on" nor as "visibly manipulating handheld devices" in the NOPUS.These would include: (1) a driver using a cellphone headset but is not speaking during the approximately 10-second period when he/she is being observed, and (2) a driver using technologies that cannot be observed from the roadside.The unobservable technologies would include: a wireless earpiece obscured by hair or clothing or on the left ear, a driver conversing via a speakerphone with the phone on the passenger seat or in a cellphone holder on the vehicle dashboard, a driver using a phone that is built into the vehicle (e.g., OnStar), and a driver using the cellphone hands-free via a Bluetooth car kit or via a Bluetooth system that is built into the vehicle (e.g., Sync).It is possible that at some point in the future, NOPUS may be able to capture such behaviors by directing a device that can detect cellphones in-use in the passing vehicles.
The racial categories "Black," "White," and "Members of Other Races" appearing in the tables reflect subjective characterizations by roadside observers regarding the race of occupants.Likewise, observers record the age group (8-15; 16-24; 25-69; and 70 or older) that best fits their visual assessment of each observed occupant.
"Expressway Exit Ramps" are defined as the access roads from roadways with limited access, while "Other Surface Streets" comprise all other roadways.
As of 2018, sites where light precipitation or light fog are present are collapsed into a single category, "Not Clear Weather Conditions." "Weekday Rush Hours" are defined as 7 a.m. to 9:30 a.m. and 3:30 p.m. to 6 p.m. on weekdays, while "Weekday Non-Rush Hours" comprise all other weekday hours (9:30 a.m. to 3:30 p.m.).
Since NOPUS is not a census and is based on a probability sample, it is impossible to produce State-by-State driver electronic device use results.However, NOPUS produces regional estimates of the use rates based on the following categories.They are defined as: • Urban: A site which intersects an urbanized area or urban cluster (i.e., an area of at least 2,500 people); and • Rural: All other sites.

Estimating Overall Cellphone Use, Both Handheld And Hands-Free
NHTSA's 2016 Motor Vehicle Occupant Safety Survey (MVOSS) estimated that for drivers using cellphones while driving, 33 percent tended to use handheld cellphones and 67 percent tended to use hands-free phones (Diecker, & Block,  2020).Applying the proportion 2.0303 (= 67/33) of these percentages to the 2.5 percent estimate of drivers using handheld cellphones in 2021 from NOPUS shows an estimated 5.1 (= 2.0303×2.5)percent of drivers using hands-free cellphones.Thus, a total of 7.6 (= 2.5+5.1)percent of drivers are estimated to be using either a handheld or a hands-free cellphone while driving at a typical daylight moment in the United States in 2021.Please note that the MVOSS estimates are based on selfreported use from both day and nighttime driving while the NOPUS estimates are observed during daylight hours.

State Laws on Driver Electronic Device Use (Enacted as of June 7, 2021)
Many States restrict cellphone use by drivers.As of June 7, 2021, no State completely bans all forms of cellphone use by drivers.However, Table 5 shows that a ban on driving while talking on a handheld cellphone was in place in 24 States, the District of Columbia, Puerto Rico, Guam, the U.S. Virgin Islands, and the Northern Mariana Islands (Governors Highway Safety Association, 2022; Highway Loss Data Institute, 2022).All these laws are primary enforcement-an officer may cite a driver for using a handheld cellphone without any other traffic offense taking place.
According to the Governors Highway Safety Association and Highway Loss Data Institute, 48 States, the District of Columbia, Puerto Rico, Guam, the U.S. Virgin Islands, and the Northern Mariana Islands ban text messaging for all drivers (

Figure 1 Driver
Figure 1 Driver Use of Electronic Devices, 2012-2021

Table 4 Sites and Vehicles Observed in the 2021 NOPUS
Please note that since 2015, we use an objective area type classification to replace the subjective area type based on trained data collectors' best judgement.Each NOPUS site is assigned an objective area type characterization based on the Census Bureau definitions of urbanized areas and urban clusters (see www.census.gov/programs-surveys/geography/guidance/geo-areas/urban-rural.html).
• South: WV, MD, DE, VA, KY, TN, NC, SC, GA, FL, AL, MS, AR, LA, OK, TX, DC • West: AK, WA, OR, CA, NV, ID, UT, AZ, NM, CO, WY, MT, HI Table 6).The only exceptions are Missouri and Montana.Missouri only bans text messaging for drivers 21 or younger.In 44 States, the District of Columbia, Guam, Puerto Rico, the U.S. Virgin Islands, and the Northern Mariana Islands, texting laws are primary enforcement, and 4 States have secondary enforcement of texting for drivers (Governors Highway Safety Association, 2022; Highway Loss Data Institute, 2022).