Age-Friendly Cycling Infrastructure—Differences and Preferences among 50+ Cyclists

Abstract

In this paper, the needs, attitudes, and perceptions of older (50+) cyclists were examined with the aim of determining the level of comfort, safety, and the way of using different types of cycling infrastructure. Considering that by 2050, 1 in 6 people will be over the age of 65, and that this category of users (particularly cyclists) still receives insufficient attention, the authors believed that in this way, a significant contribution can be made to the existing literature. Data from 389 50+ cyclists were collected through a survey, including Canada, the United States (USA), and Serbia, and analyzed using visual preference testing (VPT), ANOVA, and Kruskal–Wallis test. The countries were chosen to include certain similarities (traffic characteristics), as well as differences (cultural characteristics), in order to enable an adequate exchange of knowledge, good practice, and experience. The results indicate the existence of differences between these countries, especially regarding the perception of safety and the way of using certain infrastructure in Serbia (e.g., major urban collectors and shared space). Based on the obtained results, a set of general guidelines was proposed for countries with similar traffic and cultural characteristics on how to treat and provide sustainable infrastructure for older cyclists.

1. Introduction

Cycling represents a healthy and sustainable form of mobility, which is often seen as an attractive alternative to driving for short distances. Cycling infrastructure is an important part of the transportation infrastructure since it is used to meet the mobility needs of individuals, as well as the industry. In recent years, the number of older adults who cycle has increased, with many 50+ individuals continuing to cycle or taking up cycling as a way to stay active and maintain their mobility. However, insufficient attention is given to the needs of older cyclists. Because the preferences of older cyclists for cycling infrastructure, safety, and comfort may differ between countries, this paper reports on a comparative study of older cyclists from three countries: the USA, Canada, and Serbia. The countries were chosen to include certain similarities, for example, similar modal split for bicycle users, with the USA at 0.5–0.6% [1], Canada at 1.5% [2], and Serbia, especially Belgrade as a capital city, at 0.7% [3], but also some cultural differences, in order to enable an adequate exchange of knowledge, good practice, and experience. The research group chosen for this paper is cyclists over the age of 50.

According to recent data, the number of older adults who cycle has been on the steady rise in the mentioned countries. In the USA, the number of older cyclists increased from 4% to 6% between 2001 and 2017 [4]. Similarly, in Canada, this rate increased from 23% to 42% between 1995 and 2020 [5]. There are still no such data for Serbia.

As the number of older cyclists grows, it is essential to understand not only their preferences but also their way of thinking. This is especially important considering the natural psycho-motor changes that occur with age [6]. For example, research has shown that older adults tend to have lower levels of perceived self-confidence in cycling, which may be due to physical limitations or fear of injury [7]. On the other hand, other studies suggest that older adults may overestimate their cycling abilities, leading to an increased risk of injury [8,9].

Overall, it is important to understand how confidence influences behavior and the risk of crashes for the different types of infrastructure, especially because research has shown that different countries perceive the risk of using a bicycle differently [10,11,12]. This can have various negative consequences for traffic safety, such as an increase in the number of accidents involving vulnerable categories, an increase in fatalities in accidents, etc. [13,14]. Also, the culture and environment can make a significant difference in cycling prevalence among older people [15,16,17]. In the most developed cycling countries like Denmark and the Netherlands, people are more likely to cycle across all age groups than in the US or Serbia. These countries create extra provisions for cyclists, such as dedicated cycle lanes and various strategic measures: privileges when buying a bicycle or using a parking space, etc.

Research has shown that older adults (and users in general) may prefer dedicated bike lanes or separated bike paths, which provide increased safety and comfort compared to cycling on shared roads with motorized traffic [18,19]. These preferences may also vary based on various environmental factors, such as the availability of safe and convenient cycling routes, accessibility, connectivity, weather, habits, etc. [20]. An important thing to note here is that users usually prefer infrastructure improvements the most. Research in the Dominican Republic confirmed the above, stating that citizens report more interest in the improvement of road infrastructures than in the implementation of any other set of measures [21]. Furthermore, cyclist safety and comfort are critical factors that affect the overall experience of older adults. According to recent data, older cyclists are more likely to be involved in cycling collisions than younger cyclists [22]. This can be explained by the fact that older adults have the highest rate of distractions (from text messages, billboards, weather, the behavior of other users, obstacles, etc.) while cycling compared to the other age groups [23]. For example, in the Republic of Serbia, for the period from 2018 to 2020, every second cyclist who was killed (48%) was an older (65+) person [24].

Therefore, this research will tackle important gaps in the existing literature through in-depth analysis of the various characteristics and the needs of older adults regarding their behavior and the use of cycling infrastructure. The motive for conducting this research is reflected in the need to perceive and understand the important characteristics, requirements, and preferences of older cyclists, bearing in mind that the traffic system needs to be designed and adapted for the “weakest” user. Therefore, in this paper, we will discuss the following research questions:

• How do older cyclists from three different countries perceive their level of cycling confidence?
• Are there any differences in the perceived level of safety and comfort of older cyclists for different types of cycling infrastructure?
• What are the preferences of older cyclists towards cycling infrastructure?

The novelty of this paper is reflected through the analysis of a specific group of cyclists (people over the age of 50), whose demands and needs in traffic still receive insufficient attention. In addition, the preferences towards different cycling infrastructures from several countries (Canada, the USA, and Serbia) were analyzed, which will represent a significant contribution to the scientific literature, particularly from the point of view of older adults.

To answer previously stated questions, we conducted a comparative study of older cyclists in the USA, Canada, and Serbia. For the analysis and data processing the VPT and standard statistical testing were used. The results of this study will provide valuable insights and understanding of the factors that influence older adults’ cycling behavior. Moreover, the study’s findings may have implications for the design of infrastructure and cycling policies that promote older adults’ cycling behavior, contributing to active and healthy aging, especially for countries developing a cycling culture for older adults.

The paper is divided into five main sections. Section 1 is an introduction that outlines the paper’s purpose, primary contributions, and main research questions. The research methodology, including data collection techniques and statistical analysis, is covered in Section 2. The findings of the study are presented in Section 3 and include information on the sample characteristics, general riding patterns, self-confidence perception, safety and comfort, and preferred user paths. The most significant findings and insights of this research are discussed in Section 4, while the concluding thoughts and recommendations for further research are included in final Section 5.

2. Research Methodology

2.1. Data Collection

For the purpose of this paper, data from the USA, Canada, and Serbia was used. The data for the USA and Canada was part of the 50+ Cycling Survey. This survey has been conducted since 2018 (every year) for the USA and Canada. For compatibility, data from 2023 was used, when the research was also conducted in Serbia.

The sample of respondents consisted of 136 older cyclists from Canada and 147 from the USA. One of the goals of this paper was to compare countries that already have a certain level of cycling culture with Serbia in order to exchange examples of good practice and potentially improve the level of cycling in Serbia as a developing country. The results and recommendations from this paper can be used as a starting point for the improvement of cycling in other developing countries as well.

Based on the already existing questionnaire, the research was performed on the territory of Serbia. The survey was conducted online, using the Google Forms platform, during May and June 2023. The sample of respondents from Serbia consists of 105 older cyclists, while the total number of respondents analyzed in this paper is 388.

The questionnaire is divided into three sections (

Table 1. Questionnaire structure.

Section 1: User’s socio-demographic characteristics

Age

(a) 50–59, (b) 60–69, (c) more than 70 years

Section 2: General characteristics of cycling

What type of cycling do you do?

(a) Road, trail (b) Touring, (c) Mountain, (d) Around town, urban, (e) Racing

For what purposes do you use the bicycle?

(a) Work, (b) Shopping, (c) Companionship, (d) Recreation (exercise), (e) Competitive cyclist

What type of cyclist are you?

(a) Interested but concerned, (b) Relaxed and somewhat confident, (c) Experienced and confident, (d) Combination of the above

Section 3: User preferences regarding infrastructure

How would you use the displayed infrastructure?

(a) Neighborhood street, (b) Edge lane road, (c) Major urban collector, (d) Rural road, (e) Single lane roundabout, (f) Shared space, (g) Two-way multiuse trail, (h) Suburban collector road, (i) Residential neighborhood street, (j) Separate two-way bicycle facility, (k) Two-lane neighborhood commercial street

What is your level of comfort and safety when using the displayed infrastructure?

(a) Very uncomfortable and unsafe, (b) Somewhat uncomfortable and unsafe, (c) Neutral, (d) Somewhat comfortable and safe, (e) Very comfortable and safe

). The first section includes general socio-demographic characteristics (of which only the age of the respondents was analyzed for the purposes of this paper), the second section includes general cycling characteristics, while the last section includes questions related to the safety and comfort of different types of infrastructure, as well as the desired movement paths of users on the defined infrastructure.

In the baseline survey conducted for the US and Canada, there was a significantly greater variety of infrastructure types. To compare the answers more precisely, certain types were excluded from the analysis because they do not exist in Serbia. A total of 11 types of infrastructure were considered in the final survey (as shown in Table 1). The infrastructure-type questions included several answers that depend on the infrastructure characteristics. The offered answers included different types of bicycle infrastructure, including roadway, sidewalk, moving in the middle of the street, along the edges of the street, moving in a group, independently, etc. All questions were of the closed-ended type, except for the third section of the questionnaire, where a five-step Likert scale was used, as shown in Table 1.

2.2. Statistical Analysis

In this paper, two types of analysis were conducted. For the first type, the data were analyzed using IBM SPSS 22. We used standard methods of descriptive statistics, the ANOVA, and the Kruskal–Wallis test. These tests aimed to determine whether there are differences between the users’ responses considering the safety and comfort pertaining to analyzed countries.

The second type of analysis represents the implementation of VPT, which allows respondents to review and provide a variety of information based on visual input. In addition to this part, the network analysis method is applied in order to perceive the potential correlation of variables. These results are presented in the discussion chapter.

For the purposes of this paper, the images of eleven characteristic types of infrastructure were used as the visual input. The same examples are shown to respondents in every analyzed country. The goal of the second part of the research was to determine the preferences of users and the potential way of their movement on the presented infrastructure. Testing was also performed to determine users’ attitudes regarding the comfort and safety of different types of infrastructure. In this part of the research, a standard five-point Likert scale was used, where each user had to rate the level of comfort and safety for the visual representations, from 1—the worst—to 5—being the best grade.

3. Results

In this chapter, the results of several important analyses are shown. In the first three subchapters, the input data is presented, related to the age of the respondents, the characteristics of cycling, and the perception of self-confidence. As already mentioned, the goal of this paper was to determine potential differences between the analyzed countries in terms of attitudes about comfort and safety, but also in terms of the way of using the bicycle infrastructure, depending on certain factors, e.g., age and perceptions of self-confidence.

Accordingly, the fourth subchapter presents the results of ANOVA and the Kruskal–Wallis test in order to determine potential differences between countries in terms of comfort and safety, while the last, fifth subchapter gives the analysis of the use of cycling infrastructure, depending on the already mentioned factors.

All conducted analyses served to identify and define the general differences and preferences of 50+ cyclists regarding bicycle infrastructure.

3.1. Sample Characteristics

The total sample analyzed in this paper is 388 respondents, of which 136 (35%) are Canadians, 105 (27%) are Serbs, and 147 (38%) are Americans. For this paper, the social characteristics of the respondents related to age were analyzed by country, (

Table 2. Sample characteristics.

Respondents’ Characteristics	Canada		Serbia		United States		Total
	n	%	n	%	n	%	n	%
Age
50–59 years	51	37.5	41	39.0	44	35.1	136	35.0
60–69 years	54	39.7	34	32.4	63	38.9	151	38.8
More than 70 years	31	22.8	30	28.6	40	26.0	101	26.2
Total	136	100	105	100	147	100	388	100

), where n represents the number of respondents. If we look at the total sample, 35% of the participants were aged 50–59, 38.8% were aged 60–69, and 26.2% were aged over 70 years of age.

3.2. General Cycling Characteristics

In this part, the general characteristics of cycling are presented, including answers to questions about the type of cycling, the purpose, and the length of the trip.

Table 3. Cycling characteristics.

What Type of Cycling Do You Do?	Canada		Serbia		United States		Total
	n	%	n	%	n	%	n	%
Road, trail	127	35.1	101	28.0	133	36.8	361	100
Touring	63	47.3	33	24.8	37	27.8	133	100
Mountain	25	49.1	7	13.7	19	37.2	51	100
Around town, urban	123	40.2	93	30.4	90	29.4	306	100
Racing	9	60.0	1	6.7	5	33.3	15	100
For what purposes do you use the bicycle?	Canada		Serbia		United States		Total
	n	%	n	%	n	%	n	%
Work	51	38.1	24	17.9	59	44.0	134	100
Shopping	89	39.9	47	21.1	87	39.0	223	100
Companionship	106	33.4	94	29.6	117	37.0	317	100
Recreation (exercise)	124	35.1	90	25.5	139	39.4	353	100
Competitive cyclist	17	32.7	10	19.2	25	48.1	52	100

shows the answers to the first two questions.

From the given table, it can be seen that the bicycle is most often used for road, trail, and urban cycling, which is one of the observed similarities between the mentioned countries. When it comes to the purpose, it can be seen that the similarity between the analyzed countries indicates that the bicycle is most often used for recreation and leisure (socializing).

Considering the length of the trip for the mentioned purposes, only the most interesting results are briefly presented. Namely, the data indicate that respondents from Serbia travel much shorter distances for the same purposes compared to respondents from Canada and the USA. This is the situation for all trip purposes analyzed in this paper.

Figure 1 shows the results of the average trip length for companionship and recreation (exercise). From the diagram, it can be seen that the largest number of respondents in Serbia (43) travel shorter distances (3–16 km) for leisure (e.g., companionship, socializing) compared to respondents from Canada and the USA who, for the same purpose, travel on average between 16 and 40 km.

The situation is similar for recreation (exercise). Respondents from Serbia travel the shortest distances (3–16 km) for this purpose compared to respondents from Canada (16–40 km) and the USA (40–65 km).

This is one of the mentioned cultural differences that exist between the countries. Also, this difference can be explained by the density of the objects of attractions and facilities, that is, urban planning and space/object distribution in the analyzed countries.

3.3. Perception of Self-Confidence

This part gives the results of the question related to the perception of self-confidence as a cyclist; i.e., which type of cyclist are you? The results are shown in Figure 2 and classified according to the analyzed countries.

From Figure 2, it can be seen that the perception of self-confidence, to a certain extent, is similar between countries, but with a slightly more pronounced number of experienced and confident cyclists in Canada and the United States than in the case of Serbia. Other types of users are in a similar ratio between the analyzed countries. Such results are suitable for further examination and analysis because otherwise, any comparison of different groups of users would be meaningless.

3.4. Safety and Comfort

This part presents average scores, i.e., ratings of different types of infrastructure in terms of safety and comfort, as well as the most important results of the ANOVA and Kruskal–Wallis test.

Table 4. Safety and comfort scores by country.

Infrastructure Characteristics				Traffic Safety and Comfort Score
Streets and Roads		Appearance		Canada		Serbia	United States
Neighborhood street with sidewalks and one-way parking				4.2		3.5	4.3
Edge lane road with bicycle priority				3.8		3.0	4.0
Major urban collector with no designated bike facility				2.2		2.5	2.1
Rural road with no shoulder, low traffic volumes, and widely spaced out housing and other development				3.8		3.2	3.9
Single-lane roundabout with crosswalks and sidewalks for cyclists and pedestrians				3.2		3.0	3.4
Shared space where people walking, cycling, scootering, driving, etc., can travel freely without designated pathways				3.5		3.3	3.6
Two-way multi-use trail in parkland and a forested area				4.7		4.2	4.7
Suburban collector road with a bike lane without a buffer and a sidewalk				3.9		3.2	3.9
Residential neighborhood street with parking and shared bicycle lane				3.7		3.1	3.9
Separated two-way bicycling facility along a one-way neighborhood collector street in an urban mixed-use area				4.6		4.4	4.6
Two-lane neighborhood commercial street with shared lane marking				3.6		2.9	3.7
Average score				3.7		3.3	3.8
	Worst score (grade 1)		Average score (grade 3)		Best score (grade 5)

shows safety and comfort scores for each type of infrastructure, classified by country. The colors shown in Table 4 have the following meaning: red depicts the worst score for safety and comfort (grade 1), while green depicts the best score (grade 5).

From Table 4, it can be seen that the lowest average scores for all countries were obtained for the major urban collector with no designated bicycle facility: Canada (2.2), Serbia (2.5), and the USA (2.1). On the other hand, the highest average scores for Canada (4.7) and the USA (4.7) were obtained for the two-way multiuse trail, while for Serbia (4.4), it was a separate two-way bicycling facility. The highest overall average score for all types of infrastructure was registered for the USA and the lowest for Serbia.

Table 5. Results of ANOVA and the Kruskal–Wallis tests.

Infrastructure Characteristics		One-Way ANOVA, Kruskal–Wallis	Post Hoc Analysis	Difference Effect
Streets and Roads	Appearance
Neighborhood street		FANOVA = 86.901 p = 0.000	Serbia-Canada Serbia-United States	η2 = 0.31 High effect
Edge lane road		FANOVA = 31.829 p = 0.000	Serbia-Canada Serbia-United States	η2 = 0.14 High effect
Major urban collector		Chi2K-W = 15.221 * p = 0.000	Serbia-Canada Serbia-United States	η2 = 0.17 High effect
Rural road		FANOVA = 127.521 p = 0.000	Serbia-Canada Serbia-United States	η2 = 0.40 High effect
Single lane roundabout		FANOVA = 31.196 p = 0.000	Serbia-Canada Serbia-United States	η2 = 0.14 High effect
Shared space		FANOVA = 3.479 p = 0.032	Serbia-United States	η2 = 0.18 High effect
Two-way multiuse trail		Chi2K-W = 26.575 * p = 0.000	Serbia-Canada Serbia-United States	η2 = 0.09 Medium effect
Suburban collector road		FANOVA = 25.434 p = 0.000	Serbia-Canada Serbia-United States	η2 = 0.12 High effect
Residential neighborhood street		FANOVA = 69.505 p = 0.000	Serbia-Canada Serbia-United States	η2 = 0.26 High effect
Separated two-way bicycling facility		FANOVA = 1.531 p = 0.218	–	–
Two-lane neighborhood commercial street		FANOVA = 24.935 p = 0.000	Serbia-Canada Serbia-United States	η2 = 0.12 High effect

* Levene’s test is statistically significant, so the ANOVA test was not used. Therefore, we used the Kruskal–Wallis test, as well as post hoc analysis, to determine differences between pairs of countries.

shows the results of ANOVA and the Kruskal–Wallis test, post hoc analyses, as well as the difference effect by Eta squared metric. From Table 5, it can be seen that statistically significant results were not obtained only for separate two-way bicycling facilities. All the other results indicate a statistically significant difference in the responses between the analyzed countries. Such a result is expected considering the characteristics of a separate two-way bicycling facility. In fact, this is the only type of infrastructure that is exclusively used by cyclists. For this reason, all respondents generally gave similar positive ratings. What was partly unexpected was the result of a two-way multi-use trail and a lower score for Serbia. However, this result can be explained by the fact that both cyclists and pedestrians are allowed to use this infrastructure, which is why respondents from Serbia rate it as less safe, which can be also seen from the results in Table 4.

Based on the post hoc analysis, it can be concluded that the attitudes of respondents from Serbia differ from those of respondents from the USA and Canada. The highest difference effect was recorded for rural roads. What is also important to note is that cyclists from Serbia, generally, as a less safe infrastructure, rated those examples where cyclists and motor vehicles travel on the same level (road bicycle lanes, shared space, etc.). This result is also expected considering the nature and manner of travel on shared infrastructure. The subjective sense of safety of users contributes to this, conditioned by a significantly higher number of traffic accidents involving cyclists and motor vehicles in Serbia compared to the other two countries [25].

It is also interesting to consider and analyze the perception of safety and comfort for different infrastructures depending on the age of cyclists by country. The goal of this analysis was to determine how the respondents’ perceptions change by age. For the sake of clarity, only the infrastructure with the biggest differences in the desired travel paths was chosen (which will be further discussed in the next chapter).

When it comes to the characteristics of older cyclists by age, this paper fills the gaps, complements the existing literature, and points to potential differences among respondents. So far, it has just been indicated that older users may have a preference for separate infrastructures [26]. The results obtained in this paper show that the perception of infrastructure safety and comfort decreases with age (

Table 6. Infrastructure score for safety and comfort by age.

	Canada	Serbia	United States
Neighborhood street
50–59	4.2	4.2	4.3
60–69	4.2	2.8	4.2
more than 70	4.1	2.8	4.3
Major urban collector
50–59	2.8	2.8	2.7
60–69	2.3	2.3	2.4
more than 70	1.5	2.3	1.3
Single lane roundabout
50–59	3.3	3.3	3.4
60–69	3.3	3.1	3.4
more than 70	3.1	2.7	3.4
Shared space
50–59	3.6	3.7	3.6
60–69	3.7	3.2	3.7
more than 70	3.2	3.0	3.4

).

This conclusion is most prominent for the major urban collector. What is interesting to note is the fact that a clear gradation and an almost linear downward trend in the perception of comfort and safety can be observed among respondents in Serbia for all four types of infrastructure. The same trend for the USA and Canada can best be seen for the major urban collectors. These results support the fact that respondents from Serbia, even by age category, are conditionally much more “cautious” compared to respondents from the USA and Canada. Further elaboration of this part will be made in the discussion section.

3.5. Preferred User Paths

In this part, the results on the preferred user paths for different types of infrastructure are presented in

Table 7. Differences between countries with respect to the preferred users’ paths.

Infrastructure			Preferred Travel/Movement Path
Streets and Roads		Appearance	Canada		Serbia		United States
Neighborhood street			Street, Street far from parked cars		Street, Street far from parked cars, Sidewalk		Street, Street far from parked cars
Edge lane road			Bicycle lane		Bicycle lane		Bicycle lane
Major urban collector			Street		Street, Sidewalk		Street
Rural road			Road solo, Road with others		Road solo, Road with others		Road solo, Road with others
Single lane roundabout			Street, Crosswalk, Sidewalk		Crosswalk, Sidewalk		Street, Crosswalk, Sidewalk
Shared space			Street far from the edge lane, Street close to the edge lane		Street close to the edge lane		Street far from the edge lane, Street close to the edge lane
Two-way multiuse trail			Multi-use trail		Multi-use trail		Multi-use trail
Suburban collector road			Bicycle lane		Bicycle lane		Bicycle lane
Residential neighborhood street			Shared bicycle lane		Shared bicycle lane		Shared bicycle lane
Separated two-way bicycling facility			Bicycle facility		Bicycle facility		Bicycle facility
Two-lane neighborhood commercial street			Shared bicycle lane, Between the outside white lane and the curb		Shared bicycle lane, Between the outside white lane and the curb		Shared bicycle lane, Between the outside white lane and the curb
	completely different answer			partially different answer		completely the same answer

and briefly discussed. What is important to note is the color code, which has the following meaning: the green color means that the largest number of respondents from all countries gave the same answer, the yellow color means that the answers are partially different, and the red color means that the answers between countries are completely different.

From Table 7, it can be seen that for the four types of infrastructure, there are differences in terms of desired travel paths, and this only applies to respondents in Serbia. It is interesting to mention the two types of infrastructure for which the biggest differences were observed. In the case of a single-lane roundabout, the largest percentage of respondents from Canada (60%) and the USA (61%) said that they would use the road, while for Serbia, that percentage is 18%. The largest percentage of respondents from Serbia (73%) stated that they would move on the sidewalk for this type of infrastructure.

When it comes to shared spaces, the highest percentage of respondents from Canada (51%) and the USA (54%) would move far from the street edge, whereas for Serbia, that percentage is 0%. The largest percentage of respondents from Serbia (81%), on this type of infrastructure, would move along the edges of the street.

It is also interesting to consider how users (depending on their level of confidence) would use a certain infrastructure.

Table 8. Preferred infrastructure movement in relation to user’s self-confidence.

Neighborhood Street	Canada	Serbia	United States
Interested but concerned	I would cycle anywhere in the street	I would cycle on the sidewalk	I would cycle anywhere in the street
Relaxed and somewhat confident	I would cycle in the street, far away from the parked cars	I would cycle in the street, far away from the parked cars	I would cycle in the street, far away from the parked cars
Experienced and confident	I would cycle in the street, far away from the parked cars	I would cycle in the street, far away from the parked cars	I would cycle in the street, far away from the parked cars
Major urban collector	Canada	Serbia	United States
Interested but concerned	I would cycle in the street	I would cycle on the sidewalk	I would cycle in the street
Relaxed and somewhat confident	I would cycle in the street	I would cycle on the sidewalk	I would cycle in the street
Experienced and confident	I would cycle in the street	I would cycle in the street	I would cycle in the street
Single lane roundabout	Canada	Serbia	United States
Interested but concerned	I would use the vehicle travel lane	I would use the crosswalks and sidewalks	I would use the crosswalks and sidewalks
Relaxed and somewhat confident	I would use the crosswalks and sidewalks	I would use the crosswalks and sidewalks	I would use the vehicle travel lane
Experienced and confident	I would use the vehicle travel lane	I would use the crosswalks and sidewalks	I would use the vehicle travel lane
Shared space	Canada	Serbia	United States
Interested but concerned	Street close to the edge lane	Street close to the edge lane	Street far from the edge lane
Relaxed and somewhat confident	Street close to the edge lane	Street close to the edge lane	Street close to the edge lane
Experienced and confident	Street far from the edge lane	Street close to the edge lane	Street far from the edge lane

illustrates the mentioned examples. For clarity, Table 8 shows only the infrastructure highlighted in red and yellow in Table 7.

It can be seen that the preferences of the experienced respondents coincide to the greatest extent, which is expected. Significant differences can be observed in the answers between other groups of respondents (interested but concerned and relaxed and somewhat confident). For example, respondents from Serbia choose to move on the sidewalk instead of the road, as is the case with Canada and the USA. That is, the cyclists chose the solution that will enable physical separation from the motor vehicles [27]. This can create a new problem, which is a potential conflict between pedestrians and cyclists. The infrastructure used for encouraging cycling must not discourage walking, which could be the case in this situation. Various studies have already confirmed that users, especially older cyclists, prefer sidewalks that are clearly separated from the cycling infrastructure [28,29]. Therefore, providing cycling infrastructure that is segregated from pedestrians can stimulate both cycling and walking among older adults.

4. Discussion

As people age, their health and wellness are enhanced when physical activity is part of their daily routine. Cycling is one way that older adults remain physically active, especially if they continue cycling as they age. This paper reviews how well the existing cycling network in three countries serves its older cyclists by examining the types of cycling they do, their self-confidence as cyclists, and their sense of safety and comfort in using various types of cycling infrastructure.

We will first consider certain similarities between the countries. Figure 3 and

Table 9. Correlation matrix.

	Neighborhood Street			Edge Lane Road			Urban Collector			Rural Road			Roundabout			Shared Space			Multiuse Trail			Suburban Collector			Residential Street			Bicycling Facility			Commercial Street
	US	C	S	US	C	S	US	C	S	US	C	S	US	C	S	US	C	S	US	C	S	US	C	S	US	C	S	US	C	S	US	C	S
Neighbor-hood street
Edge lane road
Urban collector
Rural road
Roundabout
Shared space
Multiuse trail
Suburban collector
Residential street
Bicycling facility
Commercial street
	Strong
	Moderate
	Weak
	Very weak

US: United States; C: Canada; S: Serbia.

show the network analysis by country, which considers the level of comfort and safety of users for the defined infrastructure. The color of the edge in Figure 3 indicates the type of correlation (blue—positive and red—negative), while the thickness of the edge indicates the strength of the correlation.

Most of the edges are positive (excluding two very weak connections for infrastructure in Serbia), meaning that types of infrastructure are similar in the comfort and safety attitudes towards them. Some of the edges between the types have larger weights, or show a greater degree of similarity. Summing the weights from a type of infrastructure (node) can help illustrate which types show some degree of similarity and which types are fairly unique. As shown by Figure 3, for the USA and Canada, commercial street, suburban collector, and residential street have the largest node strength, while for Serbia, those are commercial street, suburban collector, and roundabout. It is also important to note that in all three countries, a strong correlation was recorded between the types of infrastructure and commercial and residential streets in relation to attitudes about comfort and safety, and for the USA and Canada, a strong correlation was recorded between suburban collectors and commercial streets (Table 9). These results support the fact that the users evaluated the comfort and safety of the mentioned infrastructure in a comparable way, that is, they perceive the comfort and safety in a similar manner. The results presented in Table 9 indicate the similarities of certain types of cycling infrastructure, which users perceive through comfort and safety. On the other hand, urban collectors for the USA and Canada and edge lane roads for Serbia had the lowest node strength, meaning those are the infrastructure types with the most unique attitudes towards them. Practically, if we wanted to treat comfort and safety on a certain infrastructure in the USA, Canada, and Serbia, special emphasis should be placed on urban collectors and edge lane roads.

If we analyze the attitudes toward infrastructure comfort and safety by country, it is important to note that respondents from Serbia, on average, gave lower ratings compared to respondents from Canada and the USA. Based on this, it can be said that respondents from Canada and the USA are, to a certain extent, freer and more self-confident, despite a similar level of confidence reported among respondents. One of the reasons for this outcome may be the general feeling of insecurity experienced by cyclists in Serbia, which is related to the behavior of other traffic users, and consequently related to the number of traffic accidents and the probability of participating in traffic accidents. For example, the fatality rate for cyclists in Canada is 2.6 deaths per million, in the USA, it is 2.7 deaths per million, while in Serbia, this number is 5.7 [25].

If we look at the change in the perception of comfort and safety with age, similar but also very interesting results are obtained. Namely, the research shows that the perception of infrastructure safety and comfort decreases with age. Therefore, as the age of users changes, the level of safety and comfort should rise accordingly. This points to the need to design inclusive, sustainable, and age-friendly infrastructure. Also, this indicates that we should invest more in the application of modern urban strategies such as designing exclusive bicycle infrastructure for older adults, adapting the existing infrastructure to the needs of older cyclists (reduce inclines and sharp curves, if necessary, shorten the routes, apply stronger and contrasting colors for the surface of the path, and for traffic signalization, if necessary, increase the use of traffic equipment along the bicycle infrastructure, as an aid to moving, stopping and parking, etc.), applying dynamic-age-based guidance to the desired infrastructure in real-time and similar measures. In general, for low-cycling countries seeking to increase cycling, such as Serbia, this research suggests focusing on the preferences and inclusion of all age groups equally as a necessary element of sustainable and universal design for cycling.

Regarding the type of infrastructure, respondents from all three countries had the same preferences. Namely, physically separate cycling infrastructure was rated as the most desirable (safest and most comfortable), which was also confirmed by other researchers [30,31,32,33,34,35,36,37,38]. This means that investments in providing physically segregated cycling infrastructure would benefit most older adults. If there is no dedicated route available, then cyclists need a space where they feel safe to ride. This is the aspect of this research that was analyzed in more detail compared to other similar papers. For example, what would be the user’s preferences if physically separated infrastructure is not available? Such cases are common in Serbia, so the national law regulates the movement on the road no more than 1 m from the edge [39], which corresponds to Canada and the USA.

Another interesting aspect to consider here is the way of using different infrastructures. The paper showed that respondents from all three countries, for most of the types, move in the same way. It was expected that there would be greater differences, especially in the case of the respondents from Serbia, but with a more detailed analysis of the characteristics of the mentioned infrastructure, it was observed that it does not provide a large number of options where cyclists could travel. For example, for edge lane roads, cyclists can only travel on the street or cycle lane, while for rural roads, they can only use the road, etc. The only reported difference is for the types neighborhood street, major urban collector, single-lane roundabout, and shared space. These are also the types where there is no clearly indicated infrastructure for cyclists. In this sense, cyclists themselves choose the place and position in the street/road profile. Therefore, the most important thing is to provide cycling infrastructure wherever possible. Otherwise, there is a large dispersion in the movement/desired routes, which can result in a decrease in the level of traffic safety. This has already been shown in the literature, especially in the case of shared spaces [12,40,41].

If we consider the differences in the way of moving through the infrastructure in relation to the level of confidence, we can see similarities between experienced and confident respondents in all three countries, which is expected. Larger differences appear in the case of other categories of users, especially in the case of Serbia, which is also expected, considering the nature of such users. However, more detailed conclusions about less confident users cannot be drawn from this analysis due to the small sample size. This analysis will be the subject of future research.

Overall, and apart from the infrastructure measures that include the provision of a denser bicycle network (side paths, trails, and protected bike lanes, as well as intuitive and safe intersections, roadway crossings, and signage) that can respond to the demands of both cyclists (and various categories of bicycles) and pedestrians, we propose a different approach to active mobility infrastructure planning. This approach consists of several important steps that focus on older cyclists and sustainable urban development:

Focus on the outcome for older adult mobility and wellness—indicating concrete benefits, for certain measures or activities globally will provide an adequate basis and guidance for future decisions.

Become interdisciplinary—the issue of active mobility, especially for the specific groups of users that were addressed in this work, requires a multidisciplinary approach and activism. An “activity-friendly environment” should include support from the physical environment as well as socio-cultural and community structures to enable and motivate active aging [42].

Be inclusive and adaptive—learn from older adults and their experiences what is needed. It is especially important to perform an in-depth analysis of traffic accidents. Determine what age-related changes in someone’s agility, strength, decision-making, awareness, etc., were at odds with the built environment that contributed to the accident, and what design and maintenance changes are needed to benefit older cyclists. This approach is important when working in a specific geographic area intended to serve the mobility needs of older adults specifically.

Design infrastructure for a specific user—as mentioned earlier in the paper, it is necessary to focus on specific groups of users and design according to their needs, instead of designing for pedestrians, cyclists, and motorized or non-motorized users as a group. User-focused design should point out some important infrastructure characteristics that would be excluded otherwise. For example, the potential usability of infrastructure may differ depending on its characteristics (number of lanes, width, etc., which is most pronounced for bicycle lanes [20,43]). That can be the impact of external factors, on which the type of infrastructure itself has no influence, such as the slope of the terrain, the flow of motor vehicles, etc., or direct design characteristics such as the width of the bicycle lane/path, the number of lanes, etc. [44,45]. In this sense, there are no design restrictions that would impose the use of any of the types, but only external factors, spatial limitations, preferences, and perceptions of users that influence the prevalence of certain infrastructure.

In addition to the aforementioned, it is important to identify the individual characteristics of the user, such as personality, and the potential correlation with the aspects already analyzed in this paper. This will be covered in more detail in the research to come.

5. Conclusions

The results of this paper can help in the process of understanding the important differences that exist between older cyclists in terms of cycling infrastructure preferences. Also, it can be beneficial to the decision-makers in the process of implementing inclusive, sustainable infrastructure and improving active mobility for older cyclists. It is important to note that the methodological procedure used in this paper is fully applicable to any other similar contexts (cities, countries, users, infrastructure) without major modifications.

Despite the existing cultural differences, the overall goal is to invest in age-friendly cycling infrastructure, especially within neighborhoods and other nearby destinations. This can be achieved only by understanding the various needs and requirements of older adults, through sectoral, multidisciplinary cooperation, and the inclusion of all interest groups in the decision-making process.

The limitations of this paper are primarily related to the sample size. With an increase in the number of respondents, the authors believe that the quality of the obtained results can be significantly improved. Also, the inclusion of the most developed cycling countries, such as the Netherlands, Denmark, etc., in research would significantly contribute to the quality of the work itself and the exchange of positive experiences. It should also be emphasized that despite the objectivity offered by the VPT method, concrete research in the field, i.e., the real use of the presented infrastructure, could provide deeper insight into analyzed characteristics. Also, the majority of the respondents from all three countries evaluated themselves as experienced and confident, which is why the research was focused on this group of users. Future research should focus primarily on the other two groups of users, namely interested but concerned and relaxed and somewhat confident, in order to offer a detailed insight into the preferences and requirements of all categories of cyclists equally. In this way, the entire existing state can be adequately comprehended and a broader palette of measures and activities can be proposed.

In addition to the above, directions of future research will be focused on the relationship between age, gender, and preferred cycling infrastructure among older adults. Also narrowing the limit of the older adults to 65+ could indicate certain interesting features of the oldest population and facilitate a potential comparison of the results with other countries.