U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590

Skip to content
Facebook iconYouTube iconTwitter iconFlickr iconLinkedInInstagram

Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations

This report is an archived publication and may contain dated technical, contact, and link information
Publication Number: FHWA RD-03-081
Date: June 2003

Updated Minimum Retroreflectivity Levels

Final Report

PDF Version (837 KB)

PDF files can be viewed with the Acrobat® Reader®


Four workshops on MR were conducted in the summer of 2002.(14) Participation was by invitation only and included representatives from city, county, and State transportation agencies, professional organizations (including an industry group), agency attorneys, and FHWA staff. During the workshops, participants were presented with information on sign retroreflectivity, took part in a nighttime demonstration of sign retroreflectivity, and worked to develop recommendations regarding minimum levels of in-service retroreflectivity for signs

The nighttime demonstrations of sign retroreflectivity were conducted to familiarize participants with the visual appearance of signs at various levels of retroreflectivity. They were not conducted to identify acceptable values for minimum levels of in-service retroreflectivity. However, their results can be used to provide a general impression of how the MR levels correspond to the participants' subjective opinions of the signs' visibility at night


During the evening of the first day of the workshops, participants drove through a demonstration course and rated various signs as acceptable or unacceptable. After completing the evaluation, participants were provided with retroreflectivity information for each sign and drove the course a second time, but were asked not to change their evaluations. The evaluation results were collected and presented to the workshop participants the following day.

The widely varying access to research facilities or similar areas where controlled nighttime sign evaluations could be performed resulted in four considerably different set-ups for the nighttime sign evaluation portion of the workshops. The first evaluation (Lakewood, CO) was performed on gated Federal property that resembled a business park. The speeds were low (approximately 25 mph) and the lighting in the area would best be defined as urban. In Hudson, WI, evaluations were performed on a frontage road with very low volumes and little ambient light except the headlamps from the adjacent freeway. The speeds at this site were approximately 40 mph. In College Station, TX, the demonstration was performed on the Texas A&M University's Riverside Campus. The environment was dark and the speeds were approximately 40 mph. In Hanover, MD, the nighttime sign evaluation course was set up in a business park. There was a considerable amount of ambient lighting from the buildings. The speeds were approximately 25 mph and there was more traffic at this site than at the other three.


The same signs were not used at all the study locations, although 12 signs were used consistently at the last three workshops. These signs were specially made to give a rough estimate of the correlation between the workshop participants' subjective ratings and the recommended updated MR levels. The idea for this technique originated from a participant in the Colorado workshop, so it was only used at the last three workshops.

The signs were a STOP sign (R1-1), a CURVE sign (W1-2), and a DIVIDED HIGHWAY ENDS sign (W6-2a). The signs were chosen to represent a white-on-red iconic sign that conveys its message primarily through shape and color, a bold symbol sign that should be easy to recognize, and a fine symbol sign that requires legibility as opposed to recognition. For each sign type, there were four duplicate signs. One sign was fabricated with beaded high-intensity sheeting (ASTM Type III). The other three were fabricated with engineering-grade sheeting (ASTM Type I). The retroreflectivity of two of the engineering-grade signs was degraded by applying varying levels of polyurethane to the sign. This technique resulted in four levels of retroreflectivity for each sign type. The objective of having four levels of retroreflectivity for each sign type was to evaluate the overall acceptance at various levels. For instance, the new beaded high-intensity signs were thought to be adequate for all conditions and that most, if not all, workshop participants would approve of these signs. On the other end of the scale, it was thought that by severely degrading a set of engineering signs to a relatively low level, most workshop participants would fail these signs. The middle grouping of retroreflectivity levels was established to be near the proposed minimum levels. Table 33 shows the retroreflectivity levels of the signs used in this evaluation.

Table 33. Description of Signs Used in Nighttime Evaluations

Background Retroreflectivity
Retroreflective Sheeting Condition
246 250 45 Beaded High-Intensity (Type III) New
70 63 18 Engineering Grade (Type I) New
36 28 10 Engineering Grade (Type I) Degraded
21 15 5 Engineering Grade (Type I) Degraded


A total of 71 workshop participants rated the 12 signs listed in table 33, resulting in a total of 851 observations. For each sign of each sign type, the results were plotted as shown in figures 45 through 47.

Figure 46. Results for the CURVE Sign. Click here for more detail.

Figure 46. Results for the CURVE Sign (i.e., Bold Warning Sign)

Figure 47. Results for the DIVIDED HIGHWAY ENDS Sign. Click here for more detail.

Figure 47. Results for the DIVIDED HIGHWAY ENDS Sign (i.e., Fine Warning Sign)

Figure 48. Results for the STOP Sign. Click here for more details

Figure 48. Results for STOP Sign (i.e., White-on-Red Iconic Sign)

The points where the best fit lines intersect the recommended updated MR levels range from 58 percent passing for the CURVE sign, to 48 percent passing for the DIVIDED HIGHWAY ENDS sign, to 39 percent passing for the STOP sign. In other words, 39 percent of the participants would have passed the STOP sign at a retroreflectivity level of 7 cd/lx/m2 for the red background.


It is important to restate that this evaluation was not scientifically designed or controlled. At each of the three workshops where they were shown, the signs were placed so that they were seen in various sequences and the distances between them was not equal from workshop to workshop. Participants were generally younger than the criteria at which the updated MR values were based on (age of the participants was not recorded). Participants drove and viewed the signs from a large variety of vehicles; participants were not exactly free of biases: It is probably not unreasonable to suspect that some may have had a predisposition concerning the development and implementation of MR levels for traffic signs. Therefore, in an effort to make the final levels so low that their impacts would be negligible, workshop participants may have passed signs that would have been normally judged inadequate.

Despite these caveats, the researchers feel that the results help to provide some confidence in the proposed minimum levels. Overall, 49 percent of the subjects would have passed the three signs at the proposed minimum levels. One of the key issues that needs to be remembered is that the workshop participants were generally younger than the criteria used to establish the proposed minimum levels. Had all the workshop participants been 55 years and older, as is the case for the criteria used to establish the proposed minimum levels, the percent passing the three test signs at the proposed minimum levels may well have been much lower. A target value in terms of percent passing, however, has not been established. While the 49 percent passing may be appropriate, others may feel that something much lower in terms of percent passing should be reached before the minimum level should be set. However, subjective visual inspections such as those performed and reported here will produce relatively large levels of variability and therefore it is difficult, if not impossible, to speculate about the most appropriate percent-passing level.


Table of Contents

Federal Highway Administration | 1200 New Jersey Avenue, SE | Washington, DC 20590 | 202-366-4000
Turner-Fairbank Highway Research Center | 6300 Georgetown Pike | McLean, VA | 22101