Why E-Bike and E-Scooter Crashes Are Undercounted—and How States Can Fix the Data

Why E-Bike and E-Scooter Crashes Are Undercounted—and How States Can Fix the Data begins with a practical question: what information or operating condition would let a public agency make a defensible decision? A crash form can capture a collision yet still fail to identify whether an e-bike or e-scooter was involved. Missing mode detail changes the question from how many crashes occurred to what the record can actually tell us. This guide treats e-bike scooter crash data undercount as a governance task rather than a slogan. It connects a policy objective to field evidence, clear responsibilities, a repeatable review routine, and plain explanations of what the evidence cannot establish. The assigned federal sources are useful starting points because they address micromobility, safety, and product or transportation context; they do not replace local legal, engineering, procurement, or accessibility review. [1]

Start with the decision, not a dashboard

Before collecting another number, name the decision at stake. For crash data completeness, staff may need to decide where to focus an inspection, what permit condition to change, whether a procurement submittal is complete, or which question requires more evidence. A dashboard without that decision can create the appearance of precision while hiding different reporting methods, missing fields, and uneven coverage. The National Highway Traffic Safety Administration discusses emerging issues in bicycle safety, including changes in devices and reporting context, while federal micromobility materials frame these devices within broader transportation systems. [1] A local protocol should therefore record the source, date, location or product context, and known limitations before it turns a report into a conclusion.

Define an observable operating rule

A useful rule says what staff will observe, who owns the response, and what evidence closes the loop. For this topic, agencies should set a mode-coding protocol, connect records where lawful, publish limitations, and revise forms when new devices appear. Avoid language that asks people to use “best efforts” without describing the result that will be checked. A written definition also helps distinguish an isolated event from a recurring condition. It creates a common reference for operators, vendors, inspectors, riders, residents, and partner agencies. The purpose is not to force every situation into one answer; it is to make exceptions visible, documented, and reviewable rather than accidental.

Policy evidence scorecard

Respect the boundary of each source

Administrative records, field observations, injury estimates, incident reports, and vendor documents each have a different unit of analysis. An injury estimate may indicate a national burden but cannot map a particular intersection. A police record may document a reported crash but may not capture every injury or correctly identify every device. A product certification document may support a procurement decision but cannot by itself demonstrate safe installation, charging, maintenance, or user behavior. CPSC materials on micromobility deaths, injuries, and hazard patterns offer a federal safety evidence base. [2] The responsible interpretation is to state what a source measures, what it misses, and why that scope fits the decision.

Build a record that can be audited

For crash data completeness, retain a small set of fields consistently: the date, the context, the report source, the condition observed, the person or entity assigned to respond, and the disposition. Keep sensitive details out of public summaries when they are unnecessary. Create version control for definitions so that a changed form, equipment specification, or operating boundary does not silently break a trend line. A short data dictionary is often more valuable than a long narrative because it tells later staff what a count actually means. This approach permits honest comparisons over time and makes corrections possible when new information arrives.

Use a local evidence protocol

Use a repeatable decision process

The following process is designed for ordinary operations. It does not substitute for applicable law or specialized professional judgment. It provides an accountable path from a reported condition to a documented response. Federal transportation resources describe micromobility as part of a wider active-transportation environment, which is why data and product decisions should be considered alongside street access, transit connections, and public communication. [3]

Process flow: define record, train reporter, reconcile sources, publish limitations

1. Observe: record the condition, location or equipment context, source, and time.
2. Decide: compare the evidence to the stated policy objective and identify uncertainty.
3. Act: correct, request documentation, adjust an operating rule, or escalate to the responsible authority.
4. Learn: review recurring patterns and update the rule, form, contract, or street treatment.

Make implementation fair and understandable

People should be able to understand what is expected before they experience a penalty, denied service, or avoidable hazard. Explain the rule in direct language, state who to contact, and make the response pathway accessible to people who do not have a particular app, payment method, technical vocabulary, or familiarity with agency procedures. If a vendor or operator is involved, align its customer messages with the public rule. If a rule affects a neighborhood or corridor, test the explanation with people who use that place at different times. A clear communication plan does not eliminate tradeoffs, but it reduces arbitrary outcomes and gives decision-makers feedback about where the policy is confusing.

Check distribution as well as totals

Aggregate counts can conceal uneven burdens. Review mode known at report; records linked across systems; late corrections; unclassified device rate. Then examine whether conditions differ by time, location, service availability, trip purpose, or type of user. Do not treat a low complaint count as proof that a problem is small: people may not know where to report, may have stopped using the service, or may have altered their route. Pair records with scheduled field checks and voluntary feedback. Publish a brief methods note that describes collection limits and the action that a result may trigger. That transparency lets the public distinguish a measured result from an unsupported assurance.

Test changes before treating them as settled

A time-limited pilot can expose details that a memo misses. Select contrasting sites or product configurations, document the baseline, state the evaluation question, and set a date for review. Watch for displacement and substitution: a speed control may change where riders travel; a certification clause may expose a documentation gap; a new charging location may shift demand to another building area. The goal is not to guarantee a perfect pilot. It is to learn in a bounded setting and preserve a written trail of observations, decisions, and revisions. This is especially important where technology, product standards, or rules are evolving.

Assign owners and escalation paths

Every measure in a scorecard needs an owner. Identify who enters the record, who checks completeness, who can require a correction, and who decides when a pattern merits a policy change. Separate quick operational fixes from decisions that require a contract amendment, capital project, legal review, or coordination with a transit, fire, or building authority. Establish a realistic response expectation and report whether it was met. For crash data completeness, accountability is strongest when the agency can show not only that it received a report, but also what it did with it and what it learned.

Keep public reporting useful

A quarterly update can be brief: restate the objective, describe the evidence reviewed, identify the most common conditions, summarize response performance, and explain any adjustments. Avoid publishing private complaint details or implying that a narrow sample represents every trip. When evidence is incomplete, say so. When a result worsens, name the next operational question rather than relabeling the measure. This practice gives residents, operators, and oversight bodies a shared record and helps prevent a dashboard from becoming a substitute for actual stewardship.

Implementation checklist

• Write the decision that the policy must support.
• Define the source, coverage, and limitations of each measure.
• Assign a response owner and an escalation route.
• Test communications and field procedures before scaling.
• Review patterns on a schedule and publish a concise methods note.

Ultimately, a sound crash data completeness policy makes assumptions testable. It uses assigned sources responsibly, distinguishes evidence from inference, and changes course when the lived condition does not match the intended public benefit. That is how an agency can improve safety and access without overstating what its records prove.

Review evidence with proportionate caution

• [1] Assigned source: https://www.nhtsa.gov/book/countermeasures-that-work/bicycle-safety/emerging-issues
• [2] Assigned source: https://www.cpsc.gov/s3fs-public/Micromobility_Products-Related_Deaths_Injuries_and_Hazard_Patterns_2017-2024.pdf
• [3] Assigned source: https://www.fhwa.dot.gov/environment/micromobility/