Safety Evaluation of On-Street Bicycle Facility Design Features (2025)
Chapter: 3 Selection of Contextual Factors in Bicycle Safety Treatments
CHAPTER 3. SELECTION OF CONTEXTUAL FACTORS IN BICYCLE SAFETY TREATMENTS
3.1 INTRODUCTION
In this chapter, the research team lists the facility types, treatment types, design elements, and contextual factors that should be considered for selecting midblock locations. The findings summarized in this chapter are based on the results of the literature review and the expertise of the research team , as well as input from the Project Panel.
Throughout this chapter, the research team uses the following terminology:
- Bikeway type refers to the bicycle lane type (e.g., conventional bicycle lane, separated bicycle lane).
- Treatment type refers to the application of a treatment to a roadway segment. For example, installing a bicycle lane is a treatment applied to a roadway to accommodate bicyclists. This can also be referred to as a countermeasure and the safety performance of a treatment can be represented as a CMF in safety literature.
- Bikeway design refers to individual design features of a bikeway type. For example, a separation type is a design feature of a separated bicycle lane.
- Contextual factor refers to the roadway, land use, bicyclist exposure, socioeconomic, and meteorological factors that affect a bicyclist’s operational or safety performance that can be used to inform the decision-making for installing or modifying bicycle lanes.
3.2 BIKEWAY TYPE
Previously in this study, the research team evaluated the safety effectiveness of the following on-street bikeway types that are installed at midblock locations: conventional bicycle lanes (BLs), buffered bicycle lanes (BBLs), contraflow (or counterflow) bicycle lanes (CBLs), and one-way and two-way separated bicycle lanes (SBLs). Table 10 provides the definition, location characteristics, and separation type of these facilities.
Table 10. On-Street Bikeway Designs.
| Facility Type* | Definition | Install Location | Separation Type | |
| Conventional bicycle lane |  |
A portion of the roadway designated by striping, signage, and pavement markings for the preferential or exclusive use of bicyclists | Right side of the street that flows in the same direction as traffic | None |
| Buffered bicycle lane |  |
A conventional bicycle lane paired with a designated buffer space separating the bicycle lane from the adjacent motor vehicle travel and/or parking lane | Streets with extra lanes or extra lane width | Painted buffer |
| Contraflow bicycle lane |  |
A conventional bicycle lane converting one-way traffic street to two-way, designed to allow bicyclists to ride in the opposite direction of motor vehicle traffic | Low-speed, low-volume streets, where large numbers of bicyclists are already riding | None |
| One-way separated bicycle |  |
A bikeway that provides an exclusive space for bicyclists along or within a roadway that is separated from motor vehicles and pedestrians using a variety of horizontal and vertical design elements | Constructed at street level, sidewalk level, or intermediate level | Vertical separation |
| Two-way separated bicycle |  |
A physically separated bicycle lane that allows bicycle movement in both directions on one side of the road | Same as one-way SBL | Same as one-way SBL |
3.3 TREATMENT TYPES FOR DEVELOPING CRASH MODIFICATION FACTORS
According to the literature review results, two categories of treatments or countermeasures have been considered when developing CMFs for bicycle facilities:
- Installation of a new facility.
- Modification or adjustment of an existing facility.
Table 11 shows the extended list of treatment types per bikeway type for developing CMFs. As observed, installation of a new bikeway on a roadway without an existing bicycle lane can be applied to all bikeway types. In some instances, this treatment may result in decreasing the roadway lane width, removal of shoulder, or decreasing the number of lanes (e.g., road diet). Increasing bicycle lane width can also be applied to most of the bikeway types. Converting one type of bikeway to another bikeway is applicable to buffered and separated bicycle lanes; a conventional bicycle lane can be modified by adding a buffer or a separation element. Finally, separation type is only applicable to separated bicycle lanes. As shown in the table, the list of separation elements used in the design includes flexposts (plastic bollards), curbs (plastic or concrete), parked vehicles, trees and shrubs, and concrete barriers. The research team considered as many treatment types as possible, based on the available data for developing the CMFs (see Chapter 4).
Table 11. Potential Bikeway Treatments for Safety Evaluation.
| Treatment Type | BL | BBL | CBL | SBL | |
|---|---|---|---|---|---|
| Install bikeway | X | X | X | X | |
| Increase bicycle lane/buffer width | X | X | X | X | |
| Convert bikeway type | Conventional bicycle lane to: | N/A | X | N/A | X |
| Buffered bicycle lane to: | N/A | N/A | N/A | X | |
| Change separation type | Flexposts | N/A | N/A | N/A | X |
| Curbs | N/A | N/A | N/A | X | |
| Parked vehicles | N/A | N/A | N/A | X | |
| Trees and shrubs | N/A | N/A | N/A | X | |
| Concrete barriers | N/A | N/A | N/A | X | |
3.4 DESIGN AND CONTEXTUAL FACTORS FOR DEVELOPING CRITERIA
Normally, on-street bike facilities are implemented based on design guidelines and/or engineering and planning judgment. In this project, the research team further accounted for the safety outcomes of contextual factors that affect bicyclist crashes and, based on this information, developed criteria for site selection. The research team assembled a list of bicycle design elements
and contextual factors based on the AASHTO’s Guide for the Development of Bicycle Facilities, the literature review results, and the research team’s expertise. Tables 13 and 14 show the comprehensive list of bikeway design and contextual factors, respectively, that were identified in the design guidelines and literature. Note that some of the bicycle design elements match with the treatment types listed above, since these design features can affect bicyclist safety. Therefore, they should be considered in developing the criteria. The variables were then divided into two groups: (1) primary factors that should be considered in site selection (Tier 1), and (2) secondary factors that should be considered to ensure geographic and demographic diversity among sites or if there is an opportunity to compare the variables among sites that are otherwise similar (Tier 2). Tier categorization is not applicable to those variables the research team recommended be excluded. The last column in each table provides a rationale for including or excluding the contextual variable in midblock site selection.
Table 12. Potential Bikeway Treatments for Safety Evaluation.
| Bikeway Design Elements | Include/Exclude | Tier (1, 2) | Rationale |
|---|---|---|---|
| Bicycle lane buffer width | Include | 1 | Not mentioned in literature, important design consideration |
| Bicycle lane width | Include | 1 | Supported in literature, important design consideration |
| Bicycle lane color | Exclude | 1 | Supported in literature, obtaining readily available data may be challenging |
| Type of separation | Include | 1 | Supported in literature, important design consideration |
| Direction | Include | 1 | Important design consideration |
| Bicycle lane length | Exclude | 2 | Not mentioned in literature, with the exception of contraflow bike lanes, facilities of a minimum length should be selected in an attempt to moderate anomalistic interactions |
| Bicycle network | Include | 2 | Not mentioned in literature, may be an important factor affecting overall safety |
Table 13. Contextual Factors Considered for Site Selection Criteria.
| Contextual Factor Category and Variable | Include/Exclude | Tier (1, 2) | Rationale | |
|---|---|---|---|---|
| Roadway design | Median type | Include | 1 | Supported in literature |
| Number of lanes | Include | 1 | Supported in literature | |
| Lane width | Include | 1 | Supported in literature | |
| Presence of a sidewalk | Exclude | N/A | Not within research scope | |
| Presence of street lighting | Include | 2 | High proportion of crashes occur at night | |
| Road curvature | Include | 2 | Supported in literature, site lines, encroachment, lane width on curves important design considerations | |
| Contextual Factor Category and Variable | Include/Exclude | Tier (1, 2) | Rationale | |
|---|---|---|---|---|
| Presence of curb and gutter | Include | 1 | Urban streets have curb and gutter present | |
| Presence and width of shoulder | Include | 1 | Supported in literature, important design consideration | |
| Shoulder color | Exclude | N/A | May be difficult finding U.S. sites | |
| Land use/access | Presence and number of bus (transit) stops | Include | 1 | Not mentioned in literature, but directly relates to ped/bike and bus/bike conflict density |
| Bus route length | Exclude | N/A | ||
| Transit headways | Include | 2 | ||
| Driveway access points (density) | Include | 1 | Supported in literature | |
| Driveway volume | Include | 1 | Not mentioned in literature, correlates with conflict potential, obtaining readily available data may be challenging, land use could be used as a proxy | |
| Driveway frontage | Include | 2 | Lack of defined entries increases exposure | |
| Driveway-bikeway interaction | Include | 2 | Not mentioned in literature, important design consideration | |
| Number of upstream/downstream intersections | Include | 2 | Not mentioned in literature, could potentially affect the number and severity of bicyclist crashes | |
| Road signals and street parking signs | Exclude | N/A | Redundant to traffic volumes and parking presence factors | |
| Land use type | Include | 1 | Supported in literature, correlates to exposure, speed, safety in numbers | |
| Presence and type of parking | Include | 1 | Supported in literature, site lines, dooring risk important design considerations | |
| Parking width | Include | 2 | May be redundant to bike lane buffer width, minimum-width scenario of travel lane, bike lane, and parking lane is important design consideration | |
| Parking—time limited/metered | Include | 2 | Short-term, managed parking results in more turnover and potential for conflicts with adjacent bikeway | |
| Employment density | Include | 2 | Supported in literature, complements commercial land use type | |
| Parking buffer width | Include | 1 | Not mentioned in literature, important design consideration | |
| Loading/transit zones | Include | 1 | Supported in literature, important design consideration | |
| Bicyclist exposure and speed limits | Bicyclist counts and count ranges | Include | 1 | Supported in literature |
| Distance traveled | Exclude | N/A | Data availability, not typically a design consideration | |
| Effect of bicyclist exposure on conflict areas | Exclude | N/A | Redundant to bicyclist counts (volumes) | |
| Number of trips | Exclude | N/A | Data availability, not typically a design consideration | |
| Traffic volumes | Include | 1 | Supported in literature, important design consideration | |
| Speed limit | Include | 1 | Supported in literature, important design consideration | |
| Contextual Factor Category and Variable | Include/Exclude | Tier (1, 2) | Rationale | |
|---|---|---|---|---|
| Time traveled | Exclude | N/A | Data availability, not typically a design consideration | |
| Demographic and socioeconomic | Population | Include | 2 | Supported in literature, complements residential/mixed-use land use type |
| Average age per capita | Include | 2 | Sites that have significant older or younger populations could provide insights on all ages and abilities facilities | |
| Disability | Exclude | N/A | Equity consideration for making facilities accessible to people with disabilities, which may contribute to crashes | |
| Gender | Exclude | N/A | Bicycle use may vary based on this factor | |
| Income | Include | 2 | Equity consideration, may affect bicyclist volumes and presence of infrastructure | |
| Race/Ethnicity | Exclude | N/A | Equity consideration, may correlate with exposure and crash risk | |
| Vehicles per household | Include | 2 | May correlate with bicycle use | |
| Meteorological and temporal | Bikeway installation date | Include | 2 | Important factor for determining study design, obtaining readily available data may be challenging |
| Weather conditions | Include | 2 | Supported in literature, not a site characteristic, but will be considered to account for climate differences between northern and southern states | |
The research team then prioritized the list of contextual variables based on the rationale and feasibility of obtaining data. Tables 15 and 16 show the list of proposed design and contextual variables, respectively, used for selecting the midblock locations per facility type. Most variables are applicable to each bikeway type, with a few exceptions. For example, vehicle lane width may only be relevant for conventional or buffered bicycle lanes (BLs), while median type is relevant to all bikeway types.
Based on the availability of the data and sample sizes, the research team used the land use and roadway design elements to determine the base conditions of the roadway where the bikeway was to be installed. If enough midblock sites were identified (i.e., sample size), the selected midblock sites were classified into land use or facility types based on the land use type, number of lanes, and median type. The CMF was then developed for these base conditions pending available crash and volume information. For example, a CMF could be developed for separated bicycle lanes installed at residential streets. In this example, the installation of a separated bicycle lane is the treatment type, while the residential street is a contextual factor indicating the land use class.
Table 14. Proposed Design Characteristics for Site Selection.
| Bikeway Design Elements | Measure | Bikeway Type | |||
|---|---|---|---|---|---|
| BL | BBL | CBL | SBL | ||
| Bicycle lane buffer width | <2 ft, 2–4 ft, >4 ft | X | X | X | X |
| Bicycle lane width | ≤4 ft, 5–7 ft, >7 ft | X | X | X | X |
| Type of separation | Paint, flexposts, concrete curb, parking, planter, other | N/A | N/A | X | X |
| Direction | One-way/two-way | X | X | ||
| Bicycle network | Density or number of bicycle lanes in a block group or census tract | X | X | X | X |
Table 15. Proposed Contextual Factor Characteristics for Site Selection.
| Contextual Factor Category and Variable | Measure | Bikeway Type | ||||
|---|---|---|---|---|---|---|
| BL | BBL | CBL | SBL | |||
| Roadway design | Median type | Two-way left-turn, curb | X | X | X | X |
| Number of lanes | ≤ 3, ≥4 | X | X | X | X | |
| Lane width | <10 ft, 10 ft, ≥11 ft | X | X | |||
| Presence of street lighting | Yes/no | X | X | X | X | |
| Road alignment | Horizontal and vertical curvature | X | X | X | ||
| Presence of curb and gutter | Yes/no | X | X | X | X | |
| Presence and width of shoulder | <4 ft, 4–6 ft, ≥7 ft | X | ||||
| Land use/access | Presence and number of bus (transit) stops | Yes/no | X | X | X | X |
| Transit headways | ≤20 min | X | X | X | X | |
| Driveway access points (density) | Per mile, weighted per segment length or other appropriate measure | X | X | X | X | |
| Driveway/land use type | Type (e.g., residential, commercial) | X | X | X | X | |
| Driveway frontage | Percent block frontage driveway | X | X | X | X | |
| Number of upstream/downstream intersections | Number of intersections, traffic control type | X | X | X | X | |
| Driveway marking treatments | X | X | ||||
| Land use type | Industrial, commercial (e.g., retail, office), residential (e.g., single-family, multifamily) | X | X | X | X | |
| Contextual Factor Category and Variable | Measure | Bikeway Type | ||||
|---|---|---|---|---|---|---|
| BL | BBL | CBL | SBL | |||
| Presence and type of parking | Yes/no | X | X | X | X | |
| Parking width | 7 ft, 8 ft, ≥9 ft | X | X | X | X | |
| Parking—time limited/metered | Yes/no | X | X | X | X | |
| Parking buffer width | 1–2 ft, 3 ft, ≥4 ft | X | X | X | X | |
| Loading/transit zones | Yes/no | X | X | X | X | |
| Bicyclist exposure and | Bicyclist counts and count ranges | AADB; AADB range (e.g., 0–5, 5–10); or low, medium, or high usage | X | X | X | X |
| Traffic volumes | AADT or AADT range (e.g., <6,000, 6,000–10,000, >10,000) | X | X | X | X | |
| Speed limit | 20–30 mph, ≥35 mph, etc. | X | X | X | X | |
| Demographic and socioeconomic | Population | Population number or density per spatial area considered | X | X | X | X |
| Average age per capita | Population size in age groups | X | X | X | X | |
| Income | Number of households in income group or median income | X | X | X | X | |
| Vehicles per household | Locations with high proportion of households without vehicles | X | X | X | X | |
| Meteorological and temporal | Bikeway installation date | Month, day, and year (mm/dd/yyyy) | X | X | X | X |
| Weather conditions | Average temperature, precipitation amount | X | X | X | X | |
3.5 CHAPTER SUMMARY
This chapter summarized the key findings used to develop a list of bikeway types, treatment types, and bikeway design and context-specific variables for selecting the midblock sites for safety evaluation. The research team first presented the list of facilities and their key characteristics. Next, the team provided a potential list of treatments applied to bicycle facilities. The research team considered these treatment types when developing the CMFs. Finally, the research team developed a comprehensive list of design and contextual factors that were considered for selecting the midblock sites with selected bikeway designs. These variables were subsequently used to determine the base conditions of the roadway where a bikeway was installed and to evaluate the effect of bikeway design on bicyclist safety.
