Reliable and accurate data are needed to raise awareness about the magnitude of road traffic injuries, and to convince policy makers of the need for action. Robust data are critical in correctly identifying problems, target populations, and prioritising risk factors. It is the foundation of monitoring and evaluation.  Road safety data will be used by a variety of stakeholders (including policy makers, the police, transport departments, civil society organisations and the media) and is paramount in persuading political leaders that road traffic injuries are a priority issue.

Often data are weak, particularly in low- and middle-income country settings. Visualising the journey to school requires localised data which is often not immediately available. The data tools presented here account for these weaknesses, and provide an evidence base for targeted interventions. Given constraints on resources, data gathering exercises must be affordable, a key premise of this toolkit.



Schools and communities will often have an awareness of the level of road traffic injuries and fatalities among their children. Teachers and parents will often know where children are facing risks from road traffic. Such community level testimony can help in identifying priority areas for interventions.

Official data on injury and fatality localized to a school area is often unavailable, weak or incomplete. This is where both school area assessments (such as the SARSAI methodology) and the ‘Star Ratings for Schools’ (SRfS) app prove effective. They identify the level of risk and the exposure of children to road traffic as they go to and from school. These tools provide a basis for protecting children on their journey to school. The data section in this toolkit provides step-by-step guidance for assessing risk, establishing a baseline before implementation, and for data-led monitoring and evaluation.

Identification of 'high risk' schools

Whilst crashes can occur anywhere and any given time, some locations have more crashes than others. Tackling road safety in these “hot spots” is one of the most traditional approaches to road safety and is known to reduce casualties. Identifying and focusing on these locations makes it possible to set priorities and maximise the impact of limited resources. On average, targeting high risk locations results in an 18% reduction in casualties, and in most cases is cost-effective.

Lists of all schools in the area can usually be obtained from local authorities ideally in spreadsheet format. By analysing secondary data and speaking with the head teacher of schools, identify schools where there is a high rate of road traffic injuries and deaths. Examples of secondary data that may be of use in identifying hot spots are police data, hospital registries, newspaper reports, and school registries.

School Area Road Safety Assessments and Improvements (SARSAI) projects have used the following steps to identify schools with high rates of injury:

  1. Obtain list of schools and contact details
  2. Ring all head teachers (or visit schools, if possible) and ask for number of road traffic injuries among pupils in the last year (anecdotal data) as well as the school population
  3. Estimate a value for road traffic injury rates by diving the number of injuries at the school by the population (Estimated RTI rate = no. of injuries/school population)
  4. Visit the schools with the highest estimated injury to see if they are suitable for SARSAI. In one Amend study, all the schools identified had estimated initial injury rates between 0.2% - 1%.

Appropriate schools for Amend-type interventions:

SARSAI focuses on schools where minor infrastructural improvements can lead to major reductions in road traffic injuries. If the school is near a major highway, it may not be a suitable candidate for SARSAI, because larger infrastructural changes such as overhead footbridges or major intersection alterations are not within the scope of SARSAI-based projects.

For more on Amend and SARSAI visit

Site visits

At each school site visit, make a note of surrounding roads, school entrances (whether they are on a main road or side road), footpaths, existing traffic-calming measures, etc to assess whether the school is an appropriate choice for the project.

During the site visit, observe student flow at drop-off and pick-up times. Where are the heavy foot traffic areas? What hazards are the children exposed to on their journeys to and from school? If necessary ask permission to follow students as they travel by foot and make a note of any areas where accidents are likely.

Key features to look out for during sites visits are:

  1. What are the most dangers to children? Speeding traffic? High traffic volume? Type of vehicles? E.g. Motorcycles weaving between vehicles or overtaking stopped vehicles? Large vehicles like trucks which limit children's visibility? Parked vehicles which limit visibility?
  2. Does there exist designated crossing areas e.g. zebra crossings, pedestrian traffic signals? If not, is there room to provide this?
  3. Are there safe places for children to walk away from traffic e.g. footpaths? Do the footpaths connect crossing points to the school entrance? If not, is there the opportunity to provide this?
  4. Are there traffic calming features in place? E.g. humps, rumble strips, chicanes etc to slow vehicles down? If not, is there the opportunity to provide this?
  5. What is the road alignment like? Are there lots or bends or slopes which might make it difficult to provide zebra crossings and speed humps at safe locations?

Interview and focus groups (pupils, teachers, parents, community around school)

These interviews and discussions build a picture of what a student’s journey to school looks like. What are the challenges faced by the students, parents and school staff? Details of how to carry out in-depth interviews and focus group discussions will be available soon.

  1. Pupil catchment area surveys

    This is used to understand and visualise where students live. Ask school administrative staff for a list of the catchment neighbourhoods for the school based on their school records. In a similar “hands-up” approach, ascertain class by class where students travel to and from school. Use this information to compile a visual catchment map e.g. on Google.

  2. Mode of transport surveys

    This survey is used to determine how students travel to and from school each day. Instructions on how to carry out a comprehensive mode of transport survey are Tool icon here. Time and resource availability may necessitate the survey being simplified to a quick hands up count, class-by-class. Try to ascertain the main mode of transports for the students, e.g. do they spend most of their journey to/from school walking, in a car, bus etc.

  3. Pedestrian counts

    Pedestrian counts are used to determine the number of pedestrians crossing the road at a particular stretch within the school zone. Previous site assessments will have identified heavy foot flow areas where students cross roads to get to and from school. Observations of how many pedestrians (students and general public) should be taken for a minimum of one hour before school starts and as school finishes on differing days of the week. Detailed instructions on how to carry out a pedestrian count will be available soon.

  4. Photovoice

    photovoice.jpgPhotovoice is a qualitative method of community-based participatory research which documents real life situations and commentaries surrounding a particular topic through the medium of photographs. Participants are asked to express their points of view surrounding a research topic, in this case road safety, by taking photographs, combining them with short narratives to give more context. Photographs are collaboratively interpreted and discussed by groups and can be used to highlight specific issues of concern by the participants. This then enables researchers as well as other stakeholders to identify the issues faced by the participants and promotes dialogue to develop effective solutions that address specific needs.

    Save the Children have developed a 10-step guidance document on how to involve children in needs assessments using Photovoice:

    Step 1: Sensitize the community and recruit participants

    Step 2: Conduct a workshop to introduce participants and facilitators in order to establish group dynamics

    Step 3: Set the project goals (remember objectives should be SMART)

    Step 4: Get the children to focus on a specific thematic area (in this case the route to school)

    Step 5: Introduce the use of cameras and photography to the children

    Step 6: Provide guidance to children on how to “speak out through photography”

    Step 7: Discuss strategies for taking pictures with the children

    Step 8: Discuss ways to capture stories and develop captions for pictures

    Step 9: Plan to exhibit or showcase the children’s pictures

    Step 10: Go public through meetings, workshops, media, etc.

    Tool icon The Save the Children Photovoice guide can be downloaded here.

    Please be aware of the ethical and safety issues surrounding the involvement of children in projects of this nature. Be sure to obtain the necessary permissions at all levels before embarking on such a project.

    Safe Kids Worldwide used Photovoice effectively to document the pedestrian-related issues facing children, parents and teachers through a project entitled: “PHOTOVOICE: Children’s Perspectives on Road Traffic Safety”. The project successfully captured on film the environments children faced while walking in Brazil, Canada, China, India, Korea, Mexico, Philippines, Thailand, the United States and Vietnam. A total of 5,743 photographs were taken by students aged 9-14 in the 10 countries. The subsequent discussions and accompanying narratives enabled participants to effectively advocate for permanent infrastructure changes which led to a safer journey to school for the students. Safe Kids Worldwide provided a number of grants to these communities permitting the implementation of permanent infrastructural changes to the risks identified through the photographs. A sample of before and after shots from this particular project can be found here.

Traffic counts

Traffic volume studies are conducted to determine the number, movements, and classifications of roadway vehicles at a given location. They can help identify critical flow time periods, determine the influence of large vehicles or pedestrians on vehicular traffic flow, or document traffic volume trends. Similar to pedestrian counts, these are conducted around peak school times at the beginning and end of the school day and at 15 minute intervals at various locations around the school. Detailed instructions how to carry out a traffic volume count can be found Tool icon here.

Speed surveys

Assessment of traffic speed around the school will be vital in determining high-risk areas and the effectiveness of traffic calming measures in reducing speed. Tool icon Detailed instructions how to carry out a speed survey can be found here.

  • International Road Assessment Programme

    The International Road Assessment Programme (iRAP) is a registered charity dedicated to preventing the more than 3,500 deaths and 100,000 injuries that occur on roads every day worldwide. iRAP works in partnership with government and non-government organisations to:

    1. Inspect high-risk roads and develop Star Ratings and Safer Roads Investment Plans
    2. Provide training, technology and support that will build and sustain national, Regional and local capability
    3. Track road safety performance so that funding agencies can assess the benefits of their investments.

  • Star ratings

    Star ratings are an internationally recognised measure of risk on roads which have been developed by iRAP. It is an "objective measure" of road safety level which is classified into five levels. Five to four stars represent the safest conditions produced from well-designed roads that represent the elements of forgiving roads, such as "grade separated intersections, good quality safety barrier, paved shoulders", while one to two stars are the worst safe conditions generated from poor roads elements, such as: undivided pavement, high operational and design speed, poor and inadequate delineation, and frequent curves and intersections.

    iRAP Star Rating assessments are playing an increasingly vital role in preventing road deaths and injuries and the United Nations has agreed Global Road Safety Performance Targets for 3-star or better roads worldwide. Research worldwide suggests that crash costs are approximately halved for each incremental improvement in star rating.

  • Star Ratings for Schools (S4RS)

    The Star Ratings for School (SR4S) Global App is the first ever systematic evidence-based approach for analysing the risk on roads around schools. The easy-to-use universal application is a low-cost way to support quick interventions that start saving lives and preventing serious injuries from day one.

  • The S4RS App

    The SR4S Global App was developed in response to the increasing rate of road traffic injuries and deaths among children as they get to and from their place of education. Sometimes they are killed just yards from the school entrance. Each death and each injury violates a child's right to an education. The United Nations has included road safety in its Sustainable Development Goal (SDG) 3.6, calling for road deaths and serious injuries to be halved by 2020. SR4S is key to meeting these global targets for child and adolescent pedestrians.

    With support from the Road Safety Fund and Fed Ex, iRAP, Safe Kids Worldwide, the University of North Carolina Highway Safety Research Center and Servicios Mexicanos de Ingeniería Civil (SEMIC) conducted a pilot project at the Benito Juarez School in Mexico City with the aim of researching, developing and applying a Star Rating for Schools methodology. As a result of this project, the iRAP methodology was adapted so that it can easily be used to both Star Rate roads around schools and generate a list of economically viable safety countermeasures. The App is sponsored by FedEx and supported by a number of partners including the Global NGO Alliance for Road Safety NGOs.

  • Objectives of SR4S

    SR4S combines an easy-to-use School Assessment Android tablet app and a Global Reporting for Schools web application. These applications work together to harness the power of the iRAP Star Ratings to measure the risk children are exposed to on their journey to school.

    SR4S is a tool for measuring and communicating the risk children are exposed to on their journey to and from school. SR4S harnesses the power of the iRAP Star Rating for Pedestrians, combing an easy to use assessment Android tablet app and a Global reporting for Schools web application. It can be used to show the risk to pedestrians at particular spot locations at, near, or on the way to or from schools

    SR4S aims to provide all communities worldwide with access to an evidence-base tool to assess and rate safety on the roads surrounding schools and inform interventions to save lives. It also enables global leaders, donors, NGOs and governments to measure success in improving safety for this most vulnerable group. It highlights the particular dangers that they face each day and provides a means of pinpointing safety deficiencies on these roads.

  • Who is SR4S for?

    The star ratings can be undertaken by any trained road safety professional. Government staff, development bank staff, consulting engineers, automobile associations, research agencies can all undertake their own star rating assessments. iRAP provide a range of free software, methodology documents, research and technical papers, presentations and specification documents to support partners worldwide. As a global standard it is essential that all partners adhere to the required quality specifications and the consistent application of the star rating methodology worldwide.

    It's easy to train people to use the School Assessment Android tablet app. The app can deliver a quick and easy measurement of safety at the site and can point towards potential treatment options.

For iRAP visit
Find out more on Star Ratings for Schools

  • Stakeholder engagement

    To ensure an assessment has the local support of key education, policing and road agencies, an important first step is to identify and engage with the local stakeholders. This may include, but is not limited to, the education department, the local school, road authority, police department and relevant NGOs.

    As with all Star Rating assessments the work is designed to help make a difference. Stakeholder support for the upgrade of a road, the reduction in speeds around the school and/or educating children to the risks they face are all potential positive outcomes of a Star Rating for Schools Project
  • Data for targeting

    Investigating crash data, where any exist, and/or the experience of local School Principals is important in order to identify where risk is already high and the potential to make a difference is greatest. This will help ensure that available resources of the NGO and the responsible government agencies can be well targeted to locations where the greatest difference is possible.
  • The Star Rating for Schools App

    Around schools, the safety of children walking along and across roads is the major concern and as such the focus of the Star Rating for Schools methodology is on measuring and reducing risk for pedestrians. By utilising a mobile app and focusing on one road user group, the risk assessment process is somewhat simplified and therefore easier for school communities around the world to apply. Applications must be low-cost and able to be used widely. Once the risk is measured, effective treatments to mitigate the risk can be considered by exploring the impact of changes in various road features and speed management actions. The implementation of the treatments can ultimately be tracked so that the partner and funder can see the benefits of their investment and the school teaching staff can educate the pupils on the correct use of the treatments.

    SR4S Flow Diagram:


  • Undertake the assessment

    The user will require a tablet computer. A health and safety plan will be needed for the team to make sure that everybody is safe at all times during the assessment. The larger the network to be assessed, the longer the assessment will take. It is important to make sure that this process is participative so that every member of the assessment team agrees with the attributes that are being registered.


  • Generate results

    Once the assessment has been completed, the user will be able to submit their data for review and generate draft Star Ratings.
  • Explore treatment options

    Once the Star Rating have been generated, it will be necessary to consider the different treatment options and discuss them amongst the assessment team. The treatments may include active speed management, infrastructure treatments at the site or the relocation of a school entrance to a safer location.

  • Assess the likely risk reduction

    Once the different treatment options have been considered and the assessment team has evaluated them, it will be necessary to assess the risk reduction associated with the proposed treatment options by generating a new Star Rating. Gauging the likely extent of the impacts for the specific treatment prior to implementation will benefit the assessment team.

  • Agree on improvements to be implemented

    Once the risk reduction has been calculated and there is clarity among the assessment team about the impact of the treatments, the assessment team will plan the implementation of the selected treatments.

  • Measure and celebrate results

    When the implementation process has been completed, the assessment team will be able to reassess the network and celebrate the safety improvements.

    Assessing and recording the improvement achieved will help demonstrate the impact of the partnership and enable all parties to positively report on the difference they have made.
  • What will success look like?

    Success will be measured by:
    • Number of schools assessed
    • Number of children made safer
    • Incremental star rating improvements and estimated reduction in risk
    • Actual before and after crash data if available
    • Aggregate results of conflict studies
    • Investment made and investment leveraged
    • Number of volunteer hours
    • School attendance rates

Monitoring and evaluation is an important aspect of any project and are often overlooked. Monitoring refers to the systematic collection of data regarding the performance of a road safety programme or intervention during or after its implementation. Evaluation involves the analysis of this data to determine the effectiveness of the program.

Monitoring and Evaluation are fundamental aspects of good programme management:

  • To provide data on programme progress and effectiveness
  • To improve program management and decision-making
  • To provide data to plan future resource needs
  • To provide data useful for policy-making and advocacy
  • To allow accountability to stakeholders, including donors, partners, and project users or beneficiaries
  • To provide opportunities to learn from experience of the current project
  • To provide evidence about what works to inform future program and scaling up
  1. Identify the purpose and scope of the M&E system
  2. Plan for data collection and management
  3. Plan for data analysis
  4. Plan for information reporting and utilization
  5. Plan for M&E human resources and capacity building
  6. Prepare the M&E budget

It is useful to build the monitoring and evaluation framework alongside planning the intervention and base it around the main objectives of the project. Early M&E planning allows for preparation of adequate time, resources and personnel before project implementation. It also informs the project design process itself as it requires people to realistically consider how practical it is to do everything they intend to measure.

An agreed M&E framework is essential in order to carry out monitoring and evaluation systematically.

This framework serves as a plan for monitoring and evaluation, and specifies:

  • What is to be monitored and evaluated?
  • What activities are needed to monitor and evaluate?
  • Who is responsible for monitoring and evaluation?
  • When monitoring and evaluation is planned?
  • How monitoring and evaluation are carried out?
  • What monitoring and evaluation resources are required?
  • What are the risks and assumptions in carrying out M&E activities?

Linking project objectives and activities with indicators that can be used to measure processes is a fundamental aspect of planning M&E. This exercise can be aided by using simple templates such as the logic framework (see below) which subdivides project objectives into: inputs, activities, outputs, short-term outcomes, long-term outcomes, impact. In reality the pathway is unlikely to be linear as there maybe multiple outcomes and feedback loops along the way but visualising the project in the form of a change pathway can be useful.


Key concepts:



The long-term change to which the project will contribute, e.g. reduction of child injuries / deaths / disabilities from road traffic crashes, in city X. The aim is usually a general statement about what the project hopes to achieve.


Objectives are the specific steps that need to be taken in order to achieve the aim of the project. In other words, the “how” of the project. Objectives, therefore, always start with a verb, e.g. to collect, to test, to produce, to analyze, etc. They should be as specific as possible (remember SMART objectives: Specific, Measureable, Achievable, Reachable, Timebound).


These are the key resources needed to support the project. They include human resources, funds, equipment, etc.


These are the interventions (or actions) that will be taken that will lead to the outputs. They can include workshops, trainings, infrastructural changes, enforcement, supervision, etc.


These are the tangible results of the project such as the number of people trained, the number of workshops conducted, the number of zebra crossings painted, the amount of material distributed, etc. They should lead to the outcomes.


The outcomes are the changes that result from the intervention. They may be short, medium or long-term. They should directly relate to the aim/objective of the programme. They can include changes in knowledge and attitudes (short-term), changes in behaviour (medium term) and finally reductions in injuries, deaths and disability.

Choosing the right indicators to measure inputs, outputs and outcomes will depend on what questions are needing to be answered so having a clear aim fofr the evaluation is essential. The breadth of an evaluation will always be limited by the resources available, but a well-designed, simple evaluation can be as useful as a more complex and costly one.

An indicator is a variable that measures some aspect of a program that is directly related to that program’s objectives. Indicators can be either quantitative or qualitative and will measure magnitude and changes. They should be:

  • Clearly defined: valid, precise
  • Meaningful: can be linked to program objectives
  • Measurable: from available data sources or from easy-to-execute studies
  • Reliable: consistently measurable by different observers
  • Non-directional: does not assume positive or negative values

Avoid vague indicators such as “number of people who “know” about speed reduction in the area.” Better to use “proportion of parents who can correctly state the new speed limit surrounding the school” as measured through a parent survey.

Examples of indicators for each part of the framework for the evaluation of a road safety program in a school are given below:


Generally you are likely to need information to track and assess what has changed (both intended and unintended) and understand the reasons for changes - i.e. what factors/organisations/individuals have facilitated/constrained change (including your contribution). Secondary data sources such as police data, vital registries, school registers or hospital data may be available and often provide a good source of outcome data. However, as secondary data is not designed specifically for program needs, it is important to avoid the trap of using irrelevant secondary data just because it is available. Check the relevance of secondary data for:

  • Population – does it cover the population about which you need data?
  • Time period – does it cover the same time period during which you need data?
  • Data variables – are the characteristics measured relevant for what you are researching? For example, just because the data may be on road safety, if your program focuses on speed reduction, a road safety study on helmet wearing rates may not be relevant.

There are many different types of studies that will help in collecting the necessary data for your indicators – they can be both qualitative and quantitative. Qualitative data such as in-depth interviews and focus group discussions can be used in the formative evaluation to provide information on why an intervention may or may not have worked and whether the program went to plan. There is a hierarchy of quantitative methods for examining the effectiveness of a program ranging from randomised-controlled trials to before-after studies with no control group. Whilst the former is more rigorous and is the gold standard by which evidence is assessed, it is also the most expensive type of evaluation to carry out. Conversely, the before-after study is cheaper but it can only provide weak indicative evidence as to a program’s effectiveness. Using a quasi-experimental design such as a controlled before-after study is often more practical and involves measuring the outcome of interest e.g. vehicle speeds before and after the intervention for both the target population and an control population A control group allows trends that may have been occurring in the population separately from those happening as a result of the programme to be taken into account.

Types of studies that can be carried out are:



Written questionnaire designed to measure indicators Systematic, can be given to a large number of people Not possible to explore answers in greater detail


Ask specific information regarding indicators Flexible, can be done in-person or by phone Requires interviewing skills so as not to ask leading questions

Focus Group

Skilled facilitator mediates a conversation with 6-12 people Cheap, allows opinions from a larger group at one time People influence one another, feelings may not be shared honestly

Observational study

Roadside observations of risk factor behavior
e.g. speed, helmet wearing.
Relatively cheap. Allows estimation of risk factor behavior such as speed, seatbelt wearing rates, helmet wearing etc. Not always accurate and relies on observation skills of researcher leading to potential bias. Time consuming

Mobility survey

A self-administered questionnaire which collects data on usual mode of travel to school and road injury Reliable and valid method of data collection on mode of travel and attitudes surrounding road safety Time consuming and potentially costly. Need trained researchers. Large sample size.

The methods used for the evaluation will depend on the aim and the budget for the evaluation. However there may be ways to reduce the costs of your evaluation. Ask yourself the following questions:

  1. Is the information necessary and sufficient?
    • Collect only what is necessary for program management and evaluation. Limit information needs to the stated objectives and indicators.

  2. Are there reliable secondary sources of data?
    • As discussed above, secondary data can save considerable time and costs – as long as it is reliable.

  3. Is the sample size adequate but not excessive?
    • Determine the sample size that is necessary to determine change. While sample size calculators are available freely online it may be better to consult a statistician regarding estimating the sample size.

  4. Can the data collection instruments be simplified?
    • Eliminate unnecessary questions from questionnaires and checklists. In addition to saving time and cost, this has the added benefit of reducing survey fatigue among respondents.

  5. Is it possible to use competent local people for the collection of survey data?
    • This can include university students, health workers, teachers, government officials and community workers. There may be associated training costs, but considerable savings can be made by hiring a team of external data collectors, and there is the advantage that local helpers will be familiar with the population, language, etc.

To ensure M & E is relevant to your stakeholders it is important that you consider their information needs, as well as your own. You will therefore need to identify the key internal and external stakeholders, and decide how to involve them in the design, implementation, analysis and/or communication of findings. Set up regular project team meetings to discuss M & E results.

This will depend on your evaluation aims and your resources. At the very least you need to be collecting data on indicators at the very beginning of the project (to build a baseline) and at the end of the project. Collection of process and input indicators should be carried out throughout the implementation of the project. Ideally for observation studies it is useful to collect data at the beginning, part way through, at the end of the intervention.

Groups can learn from each other: what was useful, what worked, what tools can be shared, etc. Sharing of problems encountered and how these were overcome can be benefit others and avoid time wasting in future work. Dissemination is important in enabling the PI to connect with other individuals, groups or organizations conducting similar projects (create a network).
Results can be shared through:

  • Report writing
  • Social media (FB and Twitter)
  • Online resources (organizational website)
  • Conferences, meetings and workshops
  • Articles (scientific and newspaper) and reports

School registers can provide a fast, simple and cheap means of injury surveillance. They may already possess an electronic or paper-based database of student attendance and collect data pertaining to the absence.It is worth approaching the headmaster to see if there is an existing system of reporting child absences. If there is none, it is good to ask about the feasibility of implementing one (or augmenting the existing one) with the help of school administration staff. It could be as simple as a logbook or a simple excel spreadsheet with drop down menus to record the following variables for any children reported as absent:

  • Name
  • Gender
  • Age (or year at school)
  • Reason for missing school (sickness, sustained injury, holiday, family member unwell and unable to get to school, death etc)
  • If injury:
    • Date/time of injury
    • Cause of injury sustained (road traffic, fall, burn, assault etc)
    • Type of injury sustained (e.g. broken arm, head injury, stab wound etc)
    • Activity at time of injury (walking to/from school, playing etc)
    • Location of injury (home, school, sports field etc)
    • Outcome (minor injury – no medical treatment; moderate injury – seen in ER, severe injury – admitted to hospital, death)
    • Number of days missed from school

Speed data from specific sites like school zones is very hard to find. Often police data pertains to major highways, or if there is a speed camera within a school zone, harnessing that data from the relevant authorities may be problematic. Speed observation studies are a reliable way to obtain such information and can be used to help evaluate speed management to reduce child pedestrian injuries.

Site Selection

  • Readings must be taken at a sufficient number of stations to ensure an accurate representation of the traffic speeds throughout the study section.
  • Safety is a major aspect of all station selections. Ensure that the data collector is not located in a dangerous situation.
  • At least one site outside the school gate. 4-6 sites elsewhere in school zone at differing locations e.g. intersections, busy main road, side road, other entrances to school.
  • Vehicles need to be in free-flowing condition

Area Review

During the area review, the investigator should note the following elements:

  • The general environment of the area (e.g., rural, commercial, residential);
  • Pavement condition;
  • Driveway locations;
  • Geometric restrictions (e.g., crest vertical curves, horizontal curves);
  • Location of existing speed limit signs and other traffic control devices;
  • Location of no passing zones; and
  • Location of nearby traffic generators (e.g., factories, malls, schools, churches).

Timing of observations

  • Day selection: depends on what you are looking for.
    • Two week days (minimum), two weekend days e.g. Sun, Tues, Thurs, Sat (can change depending on study design).
  • Time selection: 60 minute slots from 7am to 7pm to catch rush hour, school drop off/pick up.
  • Ideally want to be recording for 2 weeks before intervention is put in place and for two weeks right after, then 3 months, 6 months, 1 year etc. Reality may be different – do the maximum amount your budget allows. (Minimum is 1 week pre and 1 week post at 3 months and 1 week at 6 months).


  • Researchers/data collectors need to be trained to correctly measure speed using a radar. (Rnsure radar is calibrated).
  • They should be familiar with the data collection form.
  • 4 researchers are usually needed per site, working in 1-2 hour shifts. (One to hold radar and shout out speed, the other to note vehicle type and record speed on form).
  • Ensure they are safe, visible, and have plenty of food and water.
  • Quality control (check forms every now and again / drive by sites to assess safety and ensure consistency)
  • Data collectors should be inconspicuous to make sure they don’t change driver behaviour.

Field Documentation

  • Time: Record the start time, end time and any down time. For any down times note when and why the study was interrupted, and when it was restarted.
  • Survey Data: During the speed study, survey and site data must be collected. Data should include information about existing speed zones, prevailing speeds, location of residence and business developments, locations of crash sites, roadway alignment, traffic volumes and general roadside physical characteristics. Also note any visibility restrictions and other deficiencies that may affect the overall prevailing speeds.
  • Weather Data and Road Surface Conditions: Note the type of weather and road surface conditions at the time of the study.
  • Miscellaneous: Note any additional information that may have affected speeds at a given point (e.g., special events, pedestrians, pavement condition, nearby maintenance or construction operation enforcement efforts).

Survey Tool

Speed observation forms are relatively easy to create as a word document and printed on A4 paper for data collectors to use on-site. An example from AMEND is shown below:


  • Measure speed of oncoming vehicles on one side of the road. Swap to other side every other hour.
  • Make a note of type of vehicle AND speed.
  • If amount of traffic makes it difficult to measure every car passing checkpoint then decide on a fixed sampling frame e.g. every other car, every 4th car.
  • Multiple lanes: try and gain an average speed across all lanes i.e. measure speeds of cars on all lanes traveling in same direction. 1st car in nearest lane, 2nd car in middle lane, 3rd car in farthest lane, 4th car in nearest lane etc.

Amend, in partnership with the United States Centers for Disease Control and Prevention, conducted a multi-year population-based control study impact evaluation of SARSAI and found that the SARSAI programme results in a statistically significant reduction in the number of children injured in road traffic. For every 286 children whose schools are part of the SARSAI programme, one RTI is prevented per year. This is the first known road safety programme of any type proven to reduce RTI among children in sub-Saharan Africa. The methodology consisted of before and after household surveys in catchments areas of both intervention and control schools allowing for a more rigorous evaluation. Details of this evaluation can be found here.

  • Filling the data gap

    Data is necessary to make informed, evidence-based decisions about which areas are most in need of road safety interventions. For example, it may be clear that school areas are dangerous, but there may not be data showing which specific intersections present the most risk. Without the data, it is hard to know where very limited budgets would be best applied. Moreover, serious injuries or deaths may not have yet occurred at the school. Nonetheless, children may not feel safe going to school, and may have to run across streets or dodge speeding vehicles, particularly in low-income areas. Conflict analysis helps fill the evidence gap where official statistics may not be available. It provides data which indicates the level of risk that child pedestrians face on the journey to school.

  • The Conflict Analysis tool

    This observational tool measures the rate of conflicts between vehicles and pedestrians. The data can be used as an indicator for risk to child pedestrians. Traffic Conflict Analysis is used to observe and record hazardous interactions between road users (pedestrians and vehicles), which risk ending in a crash. It is a surrogate measure of safety which recognizes that conflicts and crashes are correlated52. A conflict, by definition, is as an observed imminent collision, or near miss, which occurs if the movements of two or more road users stays the same. A significant advantage of focusing on conflicts rather than, or in addition to crashes is that they occur more frequently and therefore provide the data necessary to accurately predict where a crash is likely to occur.

    There are several different methods of conducting Traffic Conflict Analysis with various levels of rigor, from human observation to video analytics. At its most basic level, traffic volume and pedestrian volume are counted, and conflicts are identified and categorized. This must be performed during peak hours, generally when students are traveling to and from school. Traffic Conflict Analysis can be performed by a range of individuals, from students to engineers.

Traffic Conflict Analysis requires carrying out an observation using video device or manually to detect and record interactions and conflicts between pedestrians and vehicles on roads. A conflict is an instance when two or more road users approach each other in space and time to such an extent that a collision is imminent if their movements remain unchanged.

Objective: To assess the conflicts/interactions between pedestrians and vehicles in a given road segment.

Note on the method: The methodology given below is simplified and non-technical and borrows from various methods that are deployed to assess traffic pedestrian interactions on roads.

The method presented here should be understood as a general method that can be tailored to suit local context. Further, different local authorities in different countries, have different versions of Traffic Conflict Analysis. Thus speaking in advance with local and national partners would help harmonize conflict analysis methods.

Mode: Video recorded observation

Frequency of measure:

Pre- intervention: 2-4 weeks before intervention
Post-intervention: 2-4 weeks after intervention

Duration: Duration refers to the time observed on the video recording. The recording should capture times when the traffic flow and pedestrian flow is high. For areas around schools, times when children come to school or leave from school are likely to be times with the highest pedestrian volume.

  1. Selection of sites which need to be investigated:
    • High pedestrian volume
    • High traffic volume from both sides
    • The selected signalized intersections should have reported high Road Traffic Injury (RTI) rate: Anecdotal evidence from community members/schools might be required in absence of data on RTIs at the local level

  2. Select time of observation
    • To ensure that the right time is being selected, a manual or video observation can be done, prior to the conflict assessment, to measure traffic count as well as pedestrian numbers the selected signalized intersections should have reported high number of pedestrian crashes.

      Record Traffic Volumes
      Example tables

      Peak HourSouthboundNorthboundTotalPeak time
      Wednesday AM       7:00-8:00am
      Wednesday Midday        
      Wednesday PM        

      Record Pedestrian volume

      Peak HourSouthboundNorthboundTotalPeak time
      Wednesday AM       7:45-9:00am
      Wednesday Midday        
      Wednesday PM        

      If a camera is used:

  3. Setting the camera in position for the video recording

    The camera must be positioned so that all pedestrians and vehicles on both sides can be captured. A digital overview camera covering the area should be set on top of the roadside to achieve a view of all parts of the road, and includes all road users. In large intersections, where one camera might not be sufficient, one might need to set up two cameras on opposite sides.

    Given the variation of traffic and pedestrian flow everyday, recording for at least 3-5 days during peak times is essential to determine a good average figure.

  4. Based on the site and time selection, start the recording

    The recording should capture as many vehicles and pedestrians using the information obtained through steps 1 and 2. If one is unable to capture an appropriate number of vehicles and pedestrians in the day of the recording, repeat it the next day.

  5. Analysing the video recording
    • Classify traffic objects: bicyclists, vehicle, crowd (and sometimes animals!)

    • Detect interactions and conflicts between vehicle and pedestrian in the video recording
      It is pertinent to note that all interactions between pedestrians and vehicles are not necessarily conflicts. The definition of conflict, given above, as well as the severity scale presented below, is a good guide to ascertain conflicts.

    • For each conflict the following information needs to be collected:
      • Student gender: Male/ Female
      • Vehicle type: Bicycle/ Motorcycle or motorcycle related vehicle
      • Road Description: Type of road (arterial road, collector road) Infrastructure (availability of pavements, traffic lights, zebra crossing, street lighting)
      • Factor leading to the conflict: Example, pedestrian not following traffic signals, vehicle speeding

    • Classification of pedestrian-vehicle encounters

      Specify the type of conflict between the vehicle and the pedestrian

      Situation 1 Vehicle passes directly in front of a pedestrian who is still crossing
      Situation 2 Vehicle clearly slows down or stops on the approach to the pedestrian crossing
      Situation 3 Vehicle passes immediately behind a pedestrian who is still on the zebra crossing
      Situation 4 Vehicle changes of speed or course sharply to avoid pedestrian
      Situation 5 Pedestrians running through the road due to small safe gaps
      Situation 6 Pedestrian waiting in the middle of the roadway

      The above following classification inspired from the classification by Olszewski (Olszewski et al., 2016), can be updated and tailored to situation on ground. A grounded approach, where classification is based on empirical evidence from the video recording is essential.

    • Classification on the seriousness of conflict situation pedestrians (Kaparias et al., 2013; Krivda, 2011)

      This can be determined with vehicle speed, proximity,
      Level I Potential conflict situations where there is mere breaking of road traffic rules by a single participant)
      Level II Conflict situations when one or more participants are restricted and the traffic flow is affected. It does not result into sudden change of speed or course but certain change is evident in response to the other road user
      Level III Conflict situations when one or more participants are endangered i.e. the situation when a collision can be avoided only by a prompt manoeuvring (sharp braking or sudden turning)
      Level IV Traffic crash
  6. Documentation

    To secure consistency between observations key factors, such as details of the road segment, location and observed time period must be documented. Videos need to be stored and the video recording must be analysed. If evaluation is centralised by certain departments, then video/other data needs to be shared and stored effectively.

    A key part of the analysis of Traffic Conflict Analysis is tallying numbers of conflicts, types of conflicts, and generating evidence that helps correlate different road and traffic conditions and its impact on conflicts. Basic tallying up of figures can be done through everyday software like Excel.

  7. Ethical consideration

    The local authority, the local school and community needs to be contacted to secure their permission for recordings to proceed. The times/date of recordings, and the purpose of the recording must be communicated. If required, MoUs would need to be signed.

  8. Limitations

    Even with the most stringent guidelines, traffic pedestrian conflict analysis is bound to have some of inter-observer variability which impacts the consistency of results.


Schools in Lusaka benefit from road safety improvements

In September 2018, Amend, with a local NGO partner and local engineer, worked in collaboration with the Mayor of Lusaka's office to implement road safety infrastructure improvements around Northmead Primary and Northmead Secondary Schools in Lusaka, as part of the Partnership for Healthy Cities programme. Average and 85th percentile speeds of 38km/h and 45km/h respectively were recorded outside the schools before any interventions, and thirteen road traffic injuries with two fatalities had been reported by head teachers in the previous year.

The infrastructure improvements comprised a safe school zone with speed reducing measures (speed humps and rumble strips), two raised zebra crossings, footpaths, bollards, pedestrian gates and signage. After the improvements, average and 85th percentile speeds dropped to 12km/h and 14km/hr respectively. The total school populations of over 3,500 at Northmead Primary School and 2,300 at Northmead Secondary Schools, as well as the wider community have benefited from the road safety improvements.