Skip to main content
You can also sort pages by filters.
Table of Contents
Download the Full Book

Effectiveness: 3 Star Cost: $
Use: Unknown
Time: Medium

The purpose of elementary school pedestrian training is to equip school-age children with knowledge and practice to enable them to walk safely in environments with traffic and other safety hazards. A consensus from reviews is that practical training—that is, learning by doing with reinforcement of correct behaviors—is the most effective way for children to learn traffic safety skills (Bruce & McGrath, 2005; Dragutinovic & Twisk, 2006; Percer, 2009; Schwebel, Barton et al., 2014). The need for experiential learning is especially key for younger children who lack the capacity to generalize concepts and need to practice in environments with real objects that are as close as possible to those they will experience (Dragutinovic & Twisk, 2006). And for any age to move beyond the stage of developing knowledge, children need lots of opportunities to practice so that they can move beyond the associative stage of cognitive development and devote their mental resources to problem-solving, which is when a skill becomes autonomous.

Classroom education may be enhanced by using outdoor simulation, three-dimensional models, games, or other interactive learning methods such as with computer games and models, particularly in adult-led and small-group activities. These methods do not replace real-world practice but evidence from a few studies suggests that interactive training with opportunities for feedback, correction, and practice (more than one session) may lead to more lasting behavior improvements (Tolmie et al., 2005; Albert & Dolgin, 2010). Increasingly, researchers are exploring virtual reality (ranging from a 3-D environment or headsets to smartphones and tablets), as a tool for pedestrian safety training (Schwebel & McClure, 2014; Schwebel, Combs, et al., 2016; Morrongiello et al., 2018; Schwebel et al., 2018).

A couple of studies have tried to identify the demographic, age, and developmental factors most associated with risky decisions related to crossing the road (Barton & Schwebel, 2007; Congiu et al., 2008). As expected, younger ages were associated with unsafe crossing decisions while children who had some experience with independent walking were less likely to make incorrect decisions. However, the research is less clear about the age at which educational and training interventions may be most effective. More information about effectiveness is detailed below.

Common Resources:


Unknown. Education and training material are widely available, but not necessarily implemented as part of a systematic or national program. With schools being called on for a wider variety of services and narrower set of teaching requirements, finding time to add child traffic safety modules may be difficult. Newer technologies and information formats may help expand the reach of training information but does not eliminate the need for knowledgeable trainers or educators with an interest in teaching road safety topics.  


Child pedestrian training programs have been shown to increase knowledge, but long-lasting behavior improvements may be harder to achieve, and no studies have been able to use crash or injury data to demonstrate the effectiveness of an education or training program (Dragutinovic & Twisk, 2006; Schwebel, Barton, et al., 2014). There are several ways to consider the effectiveness of a program:  Evaluators may focus on the impact by age group, the method of lesson delivery/practice, or the behaviors the program is trying to influence. 

The research does not provide a clear answer on the question of which age groups are most receptive to education and training interventions. This is likely due to differences in the design and implementation of programs. For example, an examination of the effectiveness of NCDOT’s Let’s Go NC! program showed an increase in students’ self-reported pedestrian knowledge and improved behaviors during supervised crossing situations on simulated streets that was greater for older students (Grade 3-5) compared to younger students (Grade K-2) (Thomas et al., 2017). Earlier evaluations of 5-day and 3-day WalkSafe programs in the Miami school district that used videos, formal curricula, workbooks, and outside simulation activities on an imaginary road on school grounds showed improvements in safety knowledge compared to before (Hotz et al., 2004, 2009). Improvements were more consistent for grades K-3 than for 4 and 5. Actual in-traffic behaviors were also reportedly improved in the short term but did not hold up at 3 months after the program. The Let’s Go NC! evaluation used a control group while the WalkSafe evaluations did not include controls.

To examine longer-term retention and replicability, researchers in New Jersey implemented a slightly modified version of the WalkSafe program and evaluated more than 1,500 students receiving a one-time per year WalkSafe instruction over 2 years (Livingston et al., 2011). Short- and intermediate-term knowledge retention was observed among all grades, but long-term (i.e., more than a year) knowledge retention was observed only among children moving from 3rd to 4th grade. Knowledge change did not appear to result in improved pedestrian behaviors. The authors concluded that repetition and reinforcement may be needed for long-term knowledge and behavior change, as well as engagement by caregivers. Further demonstrating the importance of repetition, researchers in Detroit studied the effect of retraining (Gates et al., 2010). In the study of 930 students in grades 2 to 7, pedestrian safety training was provided once and then again 7 to 12 months later. Researchers observed street-crossing behaviors before and after the trainings and assessed knowledge change based on pre/post-tests. After the initial training, both test scores and observed behaviors improved, but were only partially sustained. Once retraining occurred, there were increases in test scores, and the cumulative difference (after initial training and retraining) was consistently larger than the impact of initial training alone for both test scores and observational behavioral measures.

To mitigate the risks of conducting pedestrian safety training in a real-world environment, some researchers are turning to virtual reality (VR). To explore the effectiveness of different delivery methods, Schwebel, McClure, and Severson (2014) conducted a randomized control trial to examine three strategies designed to train 7- and 8-year-old children:  video/websites, non-immersive virtual reality (VR), and actual street-side training. Pre-test and post-test (immediately after completion and 6 months later) measures were collected for the 231 subjects via assessment of knowledge, behaviors in virtual reality lab, and behaviors in the field. Results showed that VR training resulted in improved pedestrian behavior post-intervention and at follow-up, but the street-side training resulted in even more impressive gains. Children receiving the video/website training showed minimal learning. A different trial suggested that video-based training may be an effective method for conveying knowledge and appropriate behaviors that would allow for a form of experiential learning while still in the classroom (Arbogast et al., 2014). However, neither before (baseline) nor long-term behavioral observations were conducted. Schwebel and McClure (2014) also used their clinical trial data to test additional hypotheses about knowledge change in relation to behavior change and found that children who received training via videos/websites gained safety knowledge, but their behavior measured in the lab and in the field did not improve, while children who received street-side training showed improvement in knowledge and behavior that was retained over a 6-month period. The authors conclude that perhaps pedestrian safety knowledge and behavior are independent constructs that should be considered separately for research and training purposes or that their measures were too disparate, so they encourage replication in future research.

A review and meta-analysis of prior behavioral interventions identified that crossing safely at midblock locations was a behavior that may have been particularly resistant to educational interventions (Schwebel, Barton, et al., 2014). In addition to examining the behaviors that interventions have targeted, researchers also analyzed the type of intervention. They confirmed that individual or small group trainings are generally effective (Tolmie et al., 2005; Albert & Dolgin, 2010). They also reported that 4- and 5-year-olds trained by adults in groups of 3 or 4 using a playmat model retained real-world behavioral (street crossing choices) improvement 6 months later compared to peers trained using two other less interactive methods or who received no training.

Since previous research identified mid-block crossings for middle-childhood kids as an especially risky behavior, Schwebel, Shen, and McClure (2016) conducted a within-subjects study of a semi-mobile, semi-immersive virtual pedestrian environment with a small sample of 44 seven- and eight-year-olds over a 3-week period. Students were assessed in a lab setting before they completed six 15-minute lessons in the virtual setting and then they were assessed again in the lab setting. In general, students made more efficient crossings in the after-period, but there were no significant changes in the rate of unsafe crossings. The researchers found that community settings (e.g., school or community center) may be effective for VR, but that more research is needed to determine the ideal duration of training and the effect of repeated feedback.

Thus, numerous studies suggest that knowledge and behaviors of young children may be improved through education and training programs, but that behavior in real-world traffic situations is more likely to be modified if the program incorporates interactive training with opportunities for practice and positive reinforcement (Percer, 2009). In addition to corrective feedback from adult trainers, opportunities for peer collaboration may also contribute to program success (Albert & Dolgin, 2010). Effectiveness of school-based child pedestrian training would also likely be enhanced if it combined child training with emphasis to teachers, parents, and other caregivers on the limits of children and the need for careful supervision, particularly for those younger than 10.


Printed material and other supplies are relatively inexpensive and easy to procure. The cost of a program will be higher if it requires a technology that the school district or community group does not already have access to. Often the most important resource is the time of knowledgeable staff members, community partners, etc.

Time to implement:

Short, once a decision is made by a local agency or organization to offer such a program. Time is needed to review the recommended material, work it into the school’s existing curriculum, and train instructors. As indicated by the above research, the training needs to be repeatedly implemented to sustain effectiveness.

Other considerations:

Hammond et al. (2014) found that trainers often modified the training from recommended best practices in a program (“Kerbcraft”) developed to provide roadside training for 5- to 7-year-olds in the United Kingdom. This deviation seems to have been towards conserving resources by conducting shorter trainings and introducing more classroom elements than the program recommended. It isn’t clear, however, if the adaptations diminish effectiveness, but that is certainly a risk since the modifications have not been evaluated. The other possible implication is that the longer, all-roadside training may not be practical for consistent implementation (Hammond et al., 2014). It is important that whenever programs are modified, however, that the changed program is also evaluated to ensure continued effectiveness.