Method

The New Mexico Department of Health conducts an Annual Statewide Survey of Seat Belt Use, typically during the month of June.  This survey is based on a random proportional selection of observational survey sites on roadways that represent the State’s population demographics and roadway travel.  Observations at the survey sites (N = 108) are randomly distributed across all days of week and daylight hours of the day (7 a.m. –7 p.m.).

This study used the same New Mexico Statewide Survey of Seat Belt Use observation sites, since it is representative of travel patterns across New Mexico roadways, at all hours and on all days of the week. Observations were made at each of these 108 sites once during the day and again at night. 

Daytime observations followed New Mexico’s survey procedures for their statewide survey.  Observation start times for day observations were generally the same as in New Mexico’s official survey.  Observations were made for 20 minutes at each location.  Paired observers collected data.  Typically, one observer observed and one recorded data.  Survey teams observed and recorded information regarding occupant type (driver or passenger), vehicle type (passenger car, pickup truck, SUV, or van), gender (male or female), and seat belt usage (wearing or not wearing).  Day of week and hour of day were also recorded as was roadway type (Interstate, U.S./State highway, or local road).  Roadways were sub-classified as located inside or outside city limits. Only front seat outboard passengers were observed. 

Nighttime observations occurred between 9 p.m. and 3:59 a.m. at the same locations as daytime observations and usually on the night following the day observation.  Nighttime observation periods lasted 45, instead of 20, minutes in order to sample a sufficient number of vehicles given the lower traffic volume at night.  Observations at night were also made by pairs of observers.  Observers recorded the same information during the day and at night.  In locations with adequate overhead lighting, observations were made without vision-enhancing equipment.  At many locations, adequate lighting did not exist in the vicinity of an observation site.  In those situations, nighttime observations were aided by the use of night vision technology.  This was typical in rural locations and locations outside of city limits. Autogated night vision goggles (XR-5 image enhancing tube) were used in tandem with an infrared spotlight (see Chaudhary et al., 2005b).  This combination allowed for adequate illumination of the occupants of a vehicle even in the complete absence of lighting without affecting the occupants of the vehicle (as infrared light is not visible to the naked human eye but visible to the goggles).  Once an observation was made, the results were verbalized to the other observer who recorded the data.

One modification made to New Mexico’s statewide survey procedure was that the survey sites on Interstates were moved to the nearest on/off Interstate ramp. Both day and night observations were made at the new locations for comparability. This was done for the benefit of the nighttime survey where night-vision goggles were necessary.  Night vision technology worked well in dim- and no-light environments, but using the technology to observe occupants at speeds in excess of 55 mph was unworkable. Ramps provided an opportunity to observe Interstate travelers at slower speeds.  Consequently, this decreased the number of observed occupants by excluding a greater number of occupants passing through on Interstates that otherwise could have been observed.

Exact positioning of observers was decided at each observation site just before daytime data collection was conducted.  At that time, consideration was given to observer safety and ease of observation-data collection for both the day and nighttime observations.  Whenever possible, observers attempted to conduct nighttime observations under lighted conditions, for safety and ease of operation.

This study was conducted during the first two weeks of June 2005, immediately following the May – June Click It or Ticket (CIOT) program, a widely publicized mobilization of law enforcement activities focused on seat belt nonusers.  There were 9,707 observed occupants during the day and 5,791 observed occupants during the night. 

Analyses

Daytime surveys of statewide belt use are usually reported as weighted results.  Weighting is based on 24-hour Average Daily Density Traffic (ADDT) counts, for which the large majority of traffic occurs during daytime hours.  It is likely that the proportion of traffic on each functional class of roadway differs between day and night.  This suggests that the 24-hour ADDT used for daytime weighting of seat belt use would be inappropriate for nighttime data (See Chaudhary et al., 2005b).  Ideally, a separate set of weights based on traffic counts occurring only during nighttime hours would be used for the night observation data.  These counts, however, were unavailable, rendering weights for nighttime data impossible.  Therefore, because of possible sampling bias caused by unweighted nighttime belt counts, this study was unable to generalize daytime and nighttime differences in belt use to the entire State.  Instead, this study provides an exploratory and descriptive account of the daytime and nighttime belt use at a representative sample of sites in New Mexico.

Albuquerque

Greater Albuquerque, with approximately 43 percent of New Mexico’s population, differs from the remainder of the State in terms of population density. Also, observations in Albuquerque were done immediately after CIOT while observations in the remainder of the State were done in succeeding weeks.  

Survey Validity

The daytime observed belt use rate across all individual observations in this study (87.6%) was comparable to the official daytime statewide belt use survey (89.5%).  The data collection timing between the current observation and the official statewide survey were slightly temporally displaced.  Specifically, the New Mexico Annual Statewide Survey of Seat Belt Use had a higher percentage of observations occurring within the first week following CIOT.  This time period immediately after CIOT likely had a higher belt use rate than any later time period.  To compensate for the differences in timing, we compared raw data from the current survey with the official statewide survey for the subset of sites whose observations all occurred within the first week after CIOT.  For the subsets, both surveys produced belt use rates of 88 percent.  Therefore, the daytime observations conducted as part of this study were directly comparable to the daytime observations conducted by the State of New Mexico.