Resources:
Speed Limit Fact Sheet ( PDF )
Free to download, print & distribute.

The entire report is many pages long, therefore we’ll just post the summary and conclusions here.

Final Report
Report No. FHwA-RD-92-084
June 1996

U.S. Department of Transportation
Federal Highway Administration

Research, Development, and Technology
Turner-Fairbank Highway Research Center
6300 Georgetown Pike
McLean, Virginia 22101-2296

Performing Organization Name and Address:
Martin R. Parker & Associates, Inc.
38549 Laurenwood Drive
Wayne, Michigan 48184-1073

Sponsoring Agency Name and Address:
Office of Safety and Traffic Operations R&D
Federal Highway Administration
6300 Georgetown Pike
McLean, Virginia 22101-2296

Contracting Officer’s Technical Representative (COTR):
Howard H. Bissell, HSR-30 and Davey L. Warren, HSR-20.
Contract or Grant Number: DTFH61-85-C-00136.
Type of report and dates covered: Final, October 1985 – June 1996

Abstract

The objectives of this research was to determine the effects of raising and lowering posted speed limits on driver behavior for urban and rural nonlimited access highways. Sites selected for study were furnished by the participating States. The study was conducted during the period from October 1985 to September 1992, when the maximum speed limit was 55 mi/h (89 km/h) on nonlimited access highways.

During this period, the states and localities lowered and raised posted speed limits on short roadway segments, typically less than 2 mi (3.2 km) in length. The general types of sites included in the study were short sections, i.e. 0.5 mi (0.8 km) segments in rural communities, 1 mi (1.6 km) sections in urban and rural communities, and 2 to 12 mi (3 to 19 km) rural sections where the speed limits were raised.

The study included the collection of driver behavior and crash data in 22 states. The data were collected at 100 sites on nonlimited access highways, consisting of 172 mi (277 km) where speed limits were either raised or lowered, and at 83 comparison sites, consisting of 132 mi (213 km) where no changes in the posted speed limits were made. Changes in the posted speed limits ranged from lowering the speed limit by 5, 10, 15, or 20 mi/h (8, 16, 24, or 32 km/h) to raising the speed limit by 5, 10, or 15 mi/h (8, 16, or 24 km/h). Only one change in the posted speed limit was made at each site during the study.

There is statistically sufficient evidence in this dataset to reject the hypothesis that driver speeds do not change when posted speed limits are either raised or lowered. However, the differences in speeds, less than 1.5 mi/h (2.4 km/h) are not sufficiently large to be of practical significance, and are due primarily to large sample sizes.

Although the changes in vehicle speeds were small, driver violations of the speed limits increased when posted speed limits were lowered. Conversely, violations decreased when speed limits were raised. This does not reflect a change in driver behavior, but a change in how compliance is measured, i.e., from the posted speed limit. There is not sufficient evidence in this dataset to reject the hypothesis that crash experience changed when posted speed limits were either lowered or raised.

Summary of Findings

The pertinent findings of this study, conducted to examine the effects of lowing and raising posted speed limits on nonlimited access rural and urban highways, are listed below. NMA’s comments and clarifications are in [italics]

  1. A review of the before and after speed data at each site revealed that differences in mean speeds, standard deviations of speeds, 85th percentile speeds, and other percentile speeds were generally less than 2 mi/h (3.2 km/h) and were not related to the amount the posted speed limit was changed.
  2. When sites were grouped by the amount of speed limit change, the differences in percentile speeds for each group were less than 1.5 mi/h (2.4 km/h), irrespective of whether the speed limit was lowered or raised or the amount that the limit was changed. The average change in percentile speeds at sites where limits were lowered and at sites where speed limits were raised was less than 1 mi/h (1.6 km/h).
  3. The small differences in before and after speeds were statistically significant due primarily to the large sample size collected.
  4. At 34 locations, existing speed limits were posted within 5 mi/h (8 km/h) of the 85th percentile speeds. When speed limits at these sites were lowered more than 5 mi/h (8 km/h) below the 85th percentile speed, the mean difference in percentile speeds was less than 1 mi/h (1.6 km/h). [In other words, lowering the speed limit does not mean traffic will slow down.]
  5. At 21 other locations, existing speed limits were posted more than 5 mi/h (8 km/h) below the 85th percentile speeds. When the agencies raised the limits to within 5 mi/h (8 km/h) of the 85th percentile speeds at these sites, the mean difference in percentile speeds was less than 1 mi/h (1.6 km/h) [In other words, raising the speed limit does not mean traffic will speed up.]
  6. By defining driver compliance as the number or percentage of drivers that travel at or below the posted speed limit, major changes in compliance occurred when speed limits were raised or lowered. However, as reflected in small changes in vehicle speeds, driver behavior did not change, but the standard for measuring compliance, i.e., posted speed limit, changed. [Higher limits mean a greater number of drivers traveling at the limit instead of over it.]
  7. Based on the free-flow speed date collected for a 24-h period at the experimental and comparison sites in 22 states, posted speed limits were set, on average, at the 45th percentile speed or below the average speed of traffic. [This means that, on average, only 45 percent of us abide by present speed limits.]
  8. Only minor changes in vehicles following at headways of less than 2 s were found at the experimental sites with similar before and after traffic volumes. [People don’t follow any closer when the limits are raised.]
  9. The indirect effects of speed limit changes on a sample of five contiguous and adjacent roadways were found to be small and insignificant. [Changing a speed limit on one road does not mean that speeds on all roads nearby will change as well.]
  10. There is not sufficient evidence, in this dataset, to reject the hypothesis that total crashes changed when posted speed limits were lowered more than 5 mi/h (8 km/h) below the 85th percentile speeds. [Lowering the speed limit does not reduce the number of crashes.]
  11. There is not sufficient evidence, in this dataset, to reject the hypothesis that total crashes changed when posted speed limits were raised more than 5 mi/h (8 km/h) below the 85th percentile speeds. [Raising the speed limit does not increase the number of crashes.]

Conclusions

There is statistically sufficient evidence in this dataset to reject the hypothesis that driver speeds do not change when posted speed limits are either raised or lowered. However, the differences in speeds are not sufficiently large to be of practical significance, and are due primarily to large sample sizes.

Although changes in vehicle speeds were small, driver violations of the speed limits increased when the posted speed limits were lowered. Conversely, violations decreased when limits were raised. This does not reflect a change in driver behavior, but a change in how compliance is measured, i.e., from the posted speed limit.

The majority of motorists did not drive 5 to 10 mi/h (8 to 16 km/h) over the posted speed limit when speed limits were raised, nor did they reduce their speed by 5 to 10 mi/h (8 to 16 km/h) when speed limits were lowered.

Based on the sites selected for this study, it appears that highway agencies have a tendancy to set speed limits slightly below the average speed of traffic.

Changing posted speed limits alone, without additional enforcement, educational programs, or other engineering measures, has only a minor effect on driver behavior.

There is not sufficient evidence in this dataset to reject the hypothesis that crash experience changed when posted speed limits were either raised or lowered. Suggestions for Future Research

There is an immediate need to examine the state policies and practices used to set posted speed limits on nonlimited access facilities. In particular, attention should be given to identifying factors or a method that leads to establishing uniform speed limits for similar roadway and traffic conditions.

The use of automated equipment and other alternative economical means of collecting unbiased speed data used to set speed limits should be explored as an alternative to the conventional use of radar.

Actual prevailing speed data for a variety of roadway geometrics and highway systems should be summarized and provided to design engineers for use as a guideline when setting the design speed on a proposed roadway project.

The implications of setting speed limits based on samples obtained by using an hourly or minimum vehicle requirement should be reexamined. Based on the 2-h collection increments from the current study, as well as recent research conducted in Michigan, wide variations in the 85th percentile speeds occurred throughout the 24-h recording periods. In addition, the hourly variations were not consistent from site to site. This suggests that speed samples should be taken throughout the day to obtain a representative sample of the 85th percentile speed, as opposed to collecting a sample over a short time period as 2-h.

You can download the study here.