This report, prepared by the National Center for Safe Routes to School, describes the feasibility of conducting a crash-based evaluation of the safety effects of the Safe Routes to School (SRTS) program.
This report, prepared by the National Center for Safe Routes to School, describes the feasibility of conducting a crash-based evaluation of the safety effects of the Safe Routes to School (SRTS) program. The findings of the report are based on what has been learned in prior attempts to evaluate SRTS safety effects, on good evaluation practices from other safety areas, on calculations concerning needed sample sizes, and on evaluating crashes using actual jurisdictions in three states.
While this report provides a detailed description of the information and efforts needed for a national evaluation, the major conclusions are listed below.
Key Findings include:
There are insufficient data to conduct a meaningful national pedestrian and bicycle crash-based evaluation of the Federal Safe Routes to School Program. Currently, there are not enough after period (i.e., with-treatment) data since a relatively small number of SRTS projects were actually implemented before 2008.
An accurate program delivery inventory, which documents what SRTS treatments are conducted at each school and when they were implemented, is needed in order to conduct a crash-based evaluation. Currently, this type of national inventory does not exist.
To insure validity, it is recommended that the study be a before/after study using either an empirical Bayes or a matched-comparison-group methodology.
The sample of schools that need to be part of the evaluation will be large in order to identify the needed number of pedestrian crashes. Assuming a 20 percent crash reduction in crashes resulting from a SRTS program, the number of pedestrian crashes involving 5-13 year olds needed in both the before and after period is 150-300. To obtain this amount of crashes, it is estimated that this could require 600-1200 SRTS schools.
Collection of the needed SRTS program-delivery data, crash data and school enrollment data will require effort which will involve evaluators, State SRTS Coordinators, local SRTS program managers and perhaps local police agencies.
This report provides useful information about previous work in child pedestrian safety and the data that need to be considered to conduct a meaningful crash-based evaluation of the Federal SRTS program. The details described in the report can help guide future SRTS crash-based research by highlighting those crucial data that are available and those that need to be collected in the future.
Furthermore, it provides a potential evaluation method that can be used and built upon once sufficient data are available.