What does the research say?

Awareness campaigns tend to emphasise helmet use and children’s safety. Pictured here is a busy junction in Gothenburg where cyclists share space with pedestrians, cars and trams.
Source: Sofia Sabel / imagebank.sweden.se

Data, communication, and political advocacy often work together. However, reviewing recent scholarship about these ideas presents challenges. Various methods for data collection are present in every study on cycling, and communication—whether in the form of public awareness-raising campaigns or political advocacy—is often mentioned as an implication of the research without further clarification on how to operationalize such work.

Despite these obstacles, some studies have been made in the past decade that evaluate the effectiveness of communication efforts or test new methods for data collection. Based on a literature review conducted in Autumn 2025

A literature review was conducted using the Scopus database in October 2025 using keywords such as bicycle, communication, political advocacy, campaign, and data. A total of 26 open-access articles were reviewed.

, the Nordic countries focus on several ideas at the intersection of data, communication, and advocacy. These include awareness campaigns for improving cycling safety; how and where cycling is visually portrayed through campaigns, media, and political communication; mixed methods approaches for collecting and analysing data; and factors that should be considered in data collection or indicator lists to account for various nuances.

Awareness campaigns for improving cycling safety

When it comes to developing campaigns, much of the research focuses on safety campaigns, especially around helmet use and/or targeting children’s safety. For example, a recent (2025) study in Odense suggests that campaigns should focus on children’s bicycle safety to reduce injuries or the severity of injuries since there is a high number of bicycle-related injuries among young children (Thorsdottir et al., 2025). In a study considering barriers and factors that contribute to helmet use among children and parents in Norway, Lajunen (2016) found that children reported using a helmet less frequently than adults. Low helmet use among children was correlated with the belief that a helmet was unnecessary or with a preference towards wearing a cap or hat instead. In contrast, habitual helmet use and an increase in perceived safety when riding were key factors correlated with children wearing helmets. Rather than using “scare campaigns”, which may inadvertently reduce cycling instead of improving safety, the researchers point to other safety awareness ideas. Campaigns for encouraging children to use helmets could, for example, showcase peers their own age since “among children, having friends who use a helmet was also an important correlate of helmet use” (Lajunen, 2016) or with school-based interventions. A study in Denmark also suggests that there may be a tendency for others to buy into the idea of wearing a helmet if they see role models wearing them, or if wearing a helmet becomes normalised (Olsson, 2025). These ideas somewhat contrast the message from Thorsdottir et al. (2025) who suggest that policy measures may also be required for combatting safety concerns among children cyclists. This could involve increasing the minimum age requirement for riding independently—in Denmark, the minimum age is 6 years old, while in Iceland, for example, the age is 9 years old.

In a later study made by Lajunen and several colleagues (2025), the researchers claim that adolescents (compared to younger children) are more influenced by their peers than by adults. This means interventions “should focus on … an internalized personal safety norm,” and encouraging them to “perceive helmet use as part of their identity—rather than something imposed by parents—may increase adherence in adolescence and adulthood” (2025, p. 133). Strategies such as “peer-led safety campaigns, integrating helmet use into youth cycling culture, and reinforcing positive social norms around helmet-wearing could help sustain these behaviours beyond childhood” (2025, p. 133). They also propose things like “helmet days”, workshops within schools or communities to raise awareness around benefits and safety risks, and behaviour nudges “such as providing free or subsidized helmets and encouraging parents to sign helmet pledge programs” (p. 133).

In Denmark, helmet usage has significantly increased among school children since 2004 (from 33% to 79% in 2022) as well as among adults (6% to 50%; Olsson, 2023). Olsson suggests that, while several factors may contribute to this behaviour change, one factor may have been the “nationwide focus on traffic safety education and behaviour change campaigns to encourage helmet wearing” (Olsson, 2023, p. 54). The author points out that legislation mandating bicycle helmets in Denmark is not necessary in order to see a change in bicycle helmet use; rather, “persistent behaviour change campaigning and education, stakeholder consensus, higher levels of road safety-oriented behaviours, and self-reinforcing processes could potentially be important factors” (Olsson, 2023, p. 54). The major waves of increased bicycle use between 2004-2010 and 2014-2020 align with the nationwide campaigns promoting wearing a helmet. Olsson concludes that communication is key: focusing on traffic safety education, national campaigns, and consensus among stakeholders who recommend wearing a helmet have a role to play for increasing helmet wearing. It is possible that similar results could be achieved through communication with regards to other cycling behavioural changes.

Lajunen’s 2016 Norwegian study found that the main barriers for helmet use among adults were related to “helmets being uncomfortable, helmets not fitting to hairstyle, or helmets being too cold or hot” (p. 300). Thus, safety campaigns targeting such barriers and emphasising benefits could be an effective path. In relation to adult helmet use as well, Lajunen et al. (2025) state that programmes related to both children and parents could be effective for targeting the household and making safety habits around cycling a “family-oriented activity” (p. 133). Indeed, Olsson (2023) shows how one of the major campaigns from 2014 in Denmark effectively targeted parents of preteen school children and contained humour and music in a memorable way for both children and parents. Reports from parents and children suggest that the videos were memorable and that families spoke about the campaign, which may have played a role in the rise of helmet use. However, it is important to recognise that such campaigns are only one of several factors at play. To be effective, campaigns must also sit alongside other road safety measures and improved cycling infrastructure (Thorsdottir et al., 2025), and interventions should be context specific, account for local perceptions, and use “culturally relevant messaging or leverage[e] community leaders as role models” (Lajunen et al., 2025, p. 133).

While many campaigns target children’s safety and helmet use, Ryan et al. (2016) turned their attention towards a different population group. Focussing on the Swedish city of Malmö, they made a survey of older persons’ perspectives on cycling. Through focus groups and a survey, the researchers conclude that “campaigns aimed at increasing the awareness and consideration of other road users towards older cyclists, as well as the introduction of clearer and more visible signage, could support older cyclists in prolonging their cycling, as well as improving the experience they have as they do cycle” (2016, p. 46). The researchers also suggest that campaigns which increase awareness of the health benefits of cycling could be another method to encourage people to continue cycling as they age.

Høye and colleagues (2020) studied the correlations between cyclist behaviours and use of safety equipment in Norway in order to understand whether it is sensible to promote cycling from a “safety gear” perspective. (In other words, developing campaigns that highlight cyclists wearing helmets, high-visibility clothing, and using lights). Through questionnaires about cyclists’ behaviour, crash involvement, and safety equipment use, they concluded that “using bicycle lights and high-visibility clothing are negatively related to collision involvement” and using some type of safety equipment is associated with a greater likelihood of using additional safety equipment. For example, if one uses a helmet, they are more likely to also use a bicycle light). The results suggested that campaigns using models who wear safety gear was appropriate since using such gear was correlated, to some degree, with safer cycling behaviour and reduced collisions. However, showcasing cyclists in safety gear has been a point of contention in public discourse as it may lead to undue perceptions of cycling as dangerous or that cyclists hold full responsibility for their safety, while factors like road design, infrastructure, and other modes of transport also contribute.

In addition to the wide range of studies looking into helmet safety campaigns, alcohol abuse was also mentioned due to injuries sustained while cycling while inebriated. In their study of injuries related to e-scooters and bicycles in Oslo, Stray et al. (2022) recommend developing awareness campaigns and implementing stricter policies as a way to reduce the number of injuries. Based on the demographic insights of injuries related to scooters and bicycles, such campaigns could target teenagers specifically. A study made by the Swedish National Road and Transport Research Institute (VTI) in 2021 investigated perceptions around cycling while mildly intoxicated from alcohol, given that there is no legally defined alcohol limit while cycling in Sweden (Warner, 2021). The study raises concerns around health and safety of cyclists based on high numbers of bicycle injuries and fatal crashes associated with alcohol intoxication, with suggestions to disseminate information or make campaigns to reduce alcohol intoxication in connection with cycling (2021). Somewhat in disagreement with Stray et al. (2022), Warner suggests using campaigns to influence behaviour change rather than focussing on policy.

Finally, awareness campaigns were also suggested by one Danish study looking at users’ perceived safety of electrically assisted bicycles (Haustein & Møller, 2016). This could include, for example, communicating with the public about the growing presence of e-bikes and their higher speeds among conventional bicycle users or other mobility forms in shared spaces. The article also raises the point that e-bike education may be necessary before riding them in high traffic areas.

How and where cycling is visually portrayed through campaigns, media, and political communication

Academic studies also highlight ways in which communication methods—whether through campaigns, media, or political communication—portray cycling and how this influences people’s perceptions of it.

In a study of urban bicycle tourism from Greater Copenhagen, Nilsson (2021) makes a document analysis of marketing material and media, alongside other planning and strategic documents. The case study emphasises the importance of communicating the values of bicycle tourism to local taxpayers because it can lead to public investment in bicycle infrastructure that supports residents. The report also highlights cycling as a key element of the region’s brand, but with some distinctions between the Swedish and the Danish sides of the region (Nilsson, 2021, p. 1657). The Swedish material shows cycling as a rural, nature-based tourism activity, compared to the Danish material which focusses on urban contexts and modern lifestyles. In both cases, the images “reinforce the region’s profile of being sustainable and progressive” and “as a visitor, you are invited to take part in this particular form of localness” (Nilsson, 2021, p. 1658). In their study of older persons’ perceptions of cycling in Malmö, Ryan et al. (2016) also found that cycling is closely connected to identity. The researchers found that “cycling has reached the stage where it is ubiquitous in the city of Malmö, even for those in later life who have cycled for most of their lives and see cycling as an inherent element of their everyday mobility” (p. 46).

Another study from Sweden evaluated the national cycling strategy compared to cycling advocacy initiatives and online blogs to determine how different communication forms discuss cycling and influence mobility trends (Balkmar, 2020). The study shows that there are distinct visions of cycling (or “future imaginaries”) provided by top-down compared to bottom-up media. While the national strategy conveys a positive initiative for increasing cycling, it “fails to move beyond the existing mobility paradigm or to explicitly call for interventions to restrict car driving,” thereby emphasising change on the part of cyclists rather than car-drivers (Balkmar, 2020, p. 336). Advocacy organisations like Cycling Sweden (Svensk Cykling) have launched initiatives that emphasise other benefits of cycling for the wider public which may have the potential to “mobilize the political energy that is required to enact the necessary transformations to make a different future possible” (p. 337). The author highlights how online blogs make more critical and radical comments on the topic from the perspective of cyclists, enabling them to “define their own situation and use the forums to co-think alternative ways of practicing sustainable mobility” including making calls for justice and imagining alternative futures which remain limited in the national strategy (p. 337). Overall Balkmar’s article highlights the need for cycling politics to be a joint effort between top-down and bottom-up perspectives.

When it comes to where campaigns take place, one study suggested an urban-rural divide (Hiselius & Svensson, 2017). The researchers surveyed local transport planners in Swedish municipalities who worked with e-bike campaigns including demonstration projects, e-bike rentals, and other media and local press campaigns to promote e-bike use. About half of the 38 surveyed municipalities reported that they conducted campaigns related to increasing e-bike use, with most campaigns focusing on trips within the city or between cities and their surrounding towns (compared to trips in rural contexts). But, due to the sustainability potential of increased cycling in rural areas, the authors argue that it may be more effective for e-bike campaigns to target rural areas.

Communication measures can come in various forms. A Swedish study claims that “pull” measures for climate-related passenger transport had greater support than “push” measures and bans (Habla et al., 2025). This could have major influence on communication within local and regional strategies as well as greater policymaking. The study also highlights political tensions that make communication and political advocacy for cycling a challenge. For example, their results suggest “a strong rivalry between the different modes of transport, where citizens tend to disagree [over] subsidizing the mode of transport that mainly competes with the mode of transport they primarily use” (Habla et al., 2025, p. 18). The authors also note that, due to large numbers of undecided citizens with regards to climate-related transport policy measures, “appropriate information provision through targeted campaigns and communication strategies could be an important tool. Governments could increase transparency about the expected outcomes of the climate-related transport policies, which could enable undecided citizens to make informed decisions” (Habla et al., 2025, p. 18).

Methods for collecting and analysing data

While many articles focus on campaigns and communication, there are also several key insights in the literature describing which methods work best for collecting cycling data. The research also shows the importance of combining qualitative and quantitative datasets. For example, in a Gothenburg-based study, researchers used data from an accident database, but they also made observational studies and interviews of cyclists who self-reported unsafe episodes to understand “subjective risk perception” (Werneke et al., 2015). With regards to data collection, their work points to the significance of combining naturalistic data with other qualitative methods to provide a more complete picture of “safety-critical” situations. They also highlight the importance of data points around uncomfortable/unsafe situations which are unlikely to be included in accident databases. This is especially valuable data to collect about cyclists since a non-crash event can still have a significant impact on perceptions of safety and comfort. In other words, assessing instances which result in death or injury give an incomplete understanding of the safety conditions of a transport route.

Studies on perceptions of comfort and safety were explored in another Swedish-based study, in which the researchers compared responses based on where questionnaires were administered: out in traffic versus online (Kazemzadeh et al., 2021). The study concluded that, when answering questionnaires online, cyclists were more positive and optimistic compared to answering while in traffic. There were also distinctions regarding the types of questions they were asked—strategic (holistic view of cycling at the city-level), tactical (perceptions about their frequent routes), or operational (perceptions of a specific location). The study highlights how the context in which surveys are conducted can make a difference. This also has implications for how researchers and planners administer and interpret data from surveys with regards to cycling. “Planners and academics could coordinate their data collection methods (in-traffic or online) based on the chosen level of analysis. For instance, when what is needed is the evaluation of the strategic and tactical aspects of a given infrastructure, we have demonstrated that the survey method does not affect the results; hence it is expected that the responses would be quite similar” (Kazemzadeh et al., 2021, p. 537). However, when it comes to operational level questions, it may be useful for questionnaires to be administered in situ.

Mixed qualitative and quantitative data approaches were also encouraged by Argyros et al. (2024). who used trajectory data from Hövding airbag helmets alongside OpenStreetMap data and bicycle traffic flow data from Copenhagen’s COMPASS flow data. Another study looking at cycling perceptions from Malmö combined questionaries by post with focus groups to gain a more nuanced understanding of their target group (Ryan et al., 2016). When it comes to collecting data about perceptions, researchers in Finland found that participatory evaluation methods played an important role, not only in collecting information about children’s enthusiasm towards cycling, but also “as an activating measure in its own right” (Kunnasvirta et al., 2026, p. 574).

The academic literature also highlights other tools for data collection, such as bicycle counters for measuring bicycle traffic flows (Holmgren et al., 2017), Bluetooth and Wi-Fi sensors for collecting travel times from bicycle traffic (Ryeng et al., 2016), and video analysis through drones to understand cycling conditions and to evaluate microscopic movements (Yastremska-Kravchenko et al., 2024). In the latter case, researchers evaluating cyclists in Linköping and Lund argue that drone footage (rather than ground-based cameras) provides “near elimination of mounting and power supply issues … one camera view covering a large area of interest …, and the drone’s perspective and relatively large distance from the subjects observed” (Yastremska-Kravchenko et al., 2024, p. 2). But these technological methods also have some caveats. In a Norwegian study, Ryeng and colleagues (2016) found that, while Bluetooth and Wi-Fi sensors were successful data collection tools, they still needed to employ a filtering system to ensure collection of cycling traffic alone (distinct from car traffic) in the dataset.

Cycling indicators

Beyond methods, several academic articles identify unique indicators or factors that should be considered when collecting cycling data. One study notes the importance of monitoring things like surface quality within cost-benefit analyses to ensure effective investments lead to better comfort, safety, and accessibility (Argyros et al., 2024). Another considers the heterogeneity of cyclists themselves when evaluating bicycle traffic flows (Perez-Castro et al., 2025). When it comes to safety indicators, Utriainen et al. (2018) used data from Sweden and Finland (as well as the Netherlands) to understand how data on “seriously injured” individuals could be combined with “fatalities” as a valuable, combined indicator rather than looking at fatalities alone. In part, this is because the costs of accidents resulting in injuries is often higher than the cost of fatal accidents and includes the impact of sustained human suffering and other public health consequences from injuries. The idea of looking beyond the most dramatic result of traffic accidents to understand cycling safety resonates with Werneke et al. (2015).

The most in-depth study on cycling indicators comes from a Stockholm-based study which builds up a bikeability index (Paulusová & Sharmeen, 2025). In previous research, the authors developed a school bikeability index based on a literature review evaluating children’s possibilities to cycle to school (Paulusová & Sharmeen, 2024). The index considered factors such as cycling infrastructure, connectivity and urbanization, safety, and the surrounding environment (see Table 1). Since then, they refined the index to focus on only local, built environment factors, contextualised the index for Stockholm, and then measured the bikeability for children based on the indicators.

In their study, Paulusová & Sharmeen (2025) found that the “most important factors influencing children’s bikeability were the availability of bicycle parking at school, the degree of urbanisation, and the speed limit at the school entrance. Conversely, the least important factors were physically separated bicycle infrastructure, bicycle infrastructure connected to the school entrance, and intersections with bicycle infrastructure” (p. 11). Such indicators are valuable for future data collection by city officials and researchers alike to continue to understand the which factors provide support or hinder different population groups from cycling.

The revised version contains fourteen key categories and a list of detailed indicators (see Table 2) that can provide insights into strengths as well as remaining areas that require prioritization to improve bikeability for school children in the city (Paulusová & Sharmeen, 2025). Notably, the authors suggest future research to refine the index, validate index weights, and include more indicators based on local contexts.

INDICATOR	NECESSARY DATA
Cycling infrastructure
Presence of bicycle infrastructure	A dataset detailing the location of cycling infrastructure, not necessarily distinguished by type, and a road network dataset
Type of bicycle infrastructure	A dataset detailing the type of all segments of available bicycle infrastructure. The typology should be based on the built characteristics (shared street, bicycle lane, shared or exclusive bicycle path), rather than on the use type (main v. local network)
Width of bicycle paths	A dataset specifying the width of available bicycle paths
Pavement type and quality	A dataset describing the type of pavement of existing bicycle infrastructure. Additionally, a subjective quality assessment of the pavement might be used
Quality of bicycle infrastructure at intersections	A dataset describing the presence and type of bicycle infrastructure at intersections (e.g., bicycle passages and crossings, and traffic lights for cyclists
Signage/signalling	A dataset of signage presence, plus a potential readability and usefulness assessment
Bicycle parking	A dataset on the presence of bicycle parking. Ideally specifying its size and type
Obstacles on bicycle path	An objective dataset on the presence of obstacles such as bollards and gates, and construction work. Plus a subjective assessment on obstructions, such as illegally parked cars
Connectivity and urbanisation
Street connectivity/intersection density	A road network dataset
Degree of urbanisation	Population density data at a spatial grid
Safety
Traffic control devices and speed limiting measures	A dataset on locations of such decides and measures
Bicycle traffic volume	A dataset on bicycle traffic volumes, measured at various places across the city
Traffic volume	A dataset on traffic volumes with high-enough density of measuring points
Lighting	A dataset on the availability of public street lighting
Traffic speed	A dataset on the maximum allowed traffic speed for each road segment
Road type	A dataset on the classification of all road segments
Collisions	A spatial dataset on the occurrence of collisions that involved a bicyclist
School site entrance location and school zones signage	A dataset on the presence of school zones and position of school entrance
Criminality	A dataset on the rate of criminal offences occurring the in area, or a subjective data rating on the perception of crime-safety in the area
Pedestrian traffic flow	A dataset on pedestrian traffic volumes in each area
Surrounding environment
Slope	Elevation data, also ideally of the road network
Aesthetic and attractiveness	Design characteristics of the built environment. Ideally, a subjective judgment of area aesthetics and attractiveness to be cycled
Air quality	Air pollution data with detailed resolution
Natural elements	A dataset on the presence (and extension) of natural elements, such as parks and forests

Table 1. School bikeability index. (Source: Paulusová & Sharmeen, 2024)

INDICATOR	NECESSARY DATA
Cycling infrastructure
Presence of bicycle infrastructure	A dataset detailing the location of cycling infrastructure, not necessarily distinguished by type, and a road network dataset
Type of bicycle infrastructure	A dataset detailing the type of all segments of available bicycle infrastructure. The typology should be based on the built characteristics (shared street, bicycle lane, shared or exclusive bicycle path), rather than on the use type (main v. local network)
Width of bicycle paths	A dataset specifying the width of available bicycle paths
Pavement type and quality	A dataset describing the type of pavement of existing bicycle infrastructure. Additionally, a subjective quality assessment of the pavement might be used
Quality of bicycle infrastructure at intersections	A dataset describing the presence and type of bicycle infrastructure at intersections (e.g., bicycle passages and crossings, and traffic lights for cyclists
Signage/signalling	A dataset of signage presence, plus a potential readability and usefulness assessment
Bicycle parking	A dataset on the presence of bicycle parking. Ideally specifying its size and type
Obstacles on bicycle path	An objective dataset on the presence of obstacles such as bollards and gates, and construction work. Plus a subjective assessment on obstructions, such as illegally parked cars
Connectivity and urbanisation
Street connectivity/intersection density	A road network dataset
Degree of urbanisation	Population density data at a spatial grid
Safety
Traffic control devices and speed limiting measures	A dataset on locations of such decides and measures
Bicycle traffic volume	A dataset on bicycle traffic volumes, measured at various places across the city
Traffic volume	A dataset on traffic volumes with high-enough density of measuring points
Lighting	A dataset on the availability of public street lighting
Traffic speed	A dataset on the maximum allowed traffic speed for each road segment
Road type	A dataset on the classification of all road segments
Collisions	A spatial dataset on the occurrence of collisions that involved a bicyclist
School site entrance location and school zones signage	A dataset on the presence of school zones and position of school entrance
Criminality	A dataset on the rate of criminal offences occurring the in area, or a subjective data rating on the perception of crime-safety in the area
Pedestrian traffic flow	A dataset on pedestrian traffic volumes in each area
Surrounding environment
Slope	Elevation data, also ideally of the road network
Aesthetic and attractiveness	Design characteristics of the built environment. Ideally, a subjective judgment of area aesthetics and attractiveness to be cycled
Air quality	Air pollution data with detailed resolution
Natural elements	A dataset on the presence (and extension) of natural elements, such as parks and forests

Table 1. School bikeability index. (Source: Paulusová & Sharmeen, 2024)

INDICATOR	NECESSARY DATA
Cycling infrastructure
Cycling network	Length of bicycle network compared to the total length of the road network within the buffer zone
Recommended cycling infrastructure for children	Length of bike lanes along roads with a maximum speed limit of 30 km/h and physically separated bicycle paths within the buffer zone, compared to the length of the road network
Physically separated infrastructure	Length of physically separated bicycle paths (both shared with pedestrians and exclusive for cyclists) within the buffer zone, compared to the length of the road network
Intersections with bicycle infrastructure	Number of road intersections with dedicated bicycle infrastructure (e.g., bicycle passage or crossing) within the buffer zone, compared to the total number of road intersections. Data on other types of bicycle infrastructure (e.g., bicycle boxes, traffic lights, etc.)
Connectivity of school with cycling infrastructure	Presence of a bicycle path or lane close to school entrance
Bicycle parking at school	Presence of bicycle parking at school (type not distinguished)
Connectivity and urbanisation
Intersection density	Number of intersections within the buffer zone, compared to the total area of the buffer zone
Degree of urbanisation	Number of residents within the buffer zone, compared to the total area of the buffer zone
Safety
Speed limit at school entrance	30 km/h speed limit at school entrance (within 100 m by road from the school)
Average maximum traffic speed limit	Average maximum speed limit of roads within the school buffer zone. Roads with walking speed limit were assigned speed of 7 km/h
Speed limiting measures	Density of speedbumps within the buffer zone, compared to the total area of the buffer zone. Presence of signage announcing school zone
Preferred road type	Length of roads of preferred functional class within the buffer zone, compared to the total length of the road network
Environment
Slope	Slope of school access roads (average slope within 100 m service area). Average slope of roads within school buffer area
Nature	Density of green areas within the school buffer zone

Table 2. Refined school bikeability index with built environment focus. (Source: Paulusová & Sharmeen, 2025)

INDICATOR	NECESSARY DATA
Cycling infrastructure
Cycling network	Length of bicycle network compared to the total length of the road network within the buffer zone
Recommended cycling infrastructure for children	Length of bike lanes along roads with a maximum speed limit of 30 km/h and physically separated bicycle paths within the buffer zone, compared to the length of the road network
Physically separated infrastructure	Length of physically separated bicycle paths (both shared with pedestrians and exclusive for cyclists) within the buffer zone, compared to the length of the road network
Intersections with bicycle infrastructure	Number of road intersections with dedicated bicycle infrastructure (e.g., bicycle passage or crossing) within the buffer zone, compared to the total number of road intersections. Data on other types of bicycle infrastructure (e.g., bicycle boxes, traffic lights, etc.)
Connectivity of school with cycling infrastructure	Presence of a bicycle path or lane close to school entrance
Bicycle parking at school	Presence of bicycle parking at school (type not distinguished)
Connectivity and urbanisation
Intersection density	Number of intersections within the buffer zone, compared to the total area of the buffer zone
Degree of urbanisation	Number of residents within the buffer zone, compared to the total area of the buffer zone
Safety
Speed limit at school entrance	30 km/h speed limit at school entrance (within 100 m by road from the school)
Average maximum traffic speed limit	Average maximum speed limit of roads within the school buffer zone. Roads with walking speed limit were assigned speed of 7 km/h
Speed limiting measures	Density of speedbumps within the buffer zone, compared to the total area of the buffer zone. Presence of signage announcing school zone
Preferred road type	Length of roads of preferred functional class within the buffer zone, compared to the total length of the road network
Environment
Slope	Slope of school access roads (average slope within 100 m service area). Average slope of roads within school buffer area
Nature	Density of green areas within the school buffer zone

Table 2. Refined school bikeability index with built environment focus. (Source: Paulusová & Sharmeen, 2025)

Based on the reviewed academic literature, Nordic research considers communication and cycling awareness campaigns, mainly in relation to safety, especially safety for children. There is some discrepancy between the role of campaigns versus policy/legislative actions for influencing transport behaviour. Many studies emphasise mixed methods approaches for collecting data on cycling, with some also testing new technologies to gather quantitative as well as qualitative insights. While some studies have looked at perceptions of cyclists, no specific research was found that considered drivers’ perspectives in relation to cyclists in the Nordic countries, or campaigns communicating awareness about cyclists targeted towards drivers.