Approximately 1.3 million people are killed annually as a result of road traffic crashes, and between 20 and 50 million people suffer non-fatal injuries, with many incurring a disability as a result of their injury (WHO, 2022). The numbers of people killed or severely injured in road crashes have gradually been reduced in recent years as a result of traditional safety strategies focusing on safety behaviors, technology, and infrastructure (Elvik et al., 2009). It has been argued that additional reductions are contingent on developing new approaches to prevention, for example the traffic safety culture approach (Edwards et al., 2014; Ward et al., 2010). Nævestad and Bjørnskau (2012) argue in favour of applying the traffic safety culture approach to the municipal level, but they conclude that this line of thought is poorly developed. In this study, we provide a way of applying the traffic safety culture concept to the municipal level.

The present paper focuses on Traffic Safe Municipality, which is a Norwegian approval scheme for systematic traffic safety work in municipalities. The background of this approval scheme is the recognition that Norwegian municipalities can influence traffic safety in a number of different ways. First, municipalities have a direct influence on traffic safety, through their employees who drive, cycle or walk etc. in their work. Norwegian municipalities are a significant employer, with over 500 000 people employed in total. We may assume that many of the municipal employees drive, cycle, walk etc. regularly at work, and that a certain proportion of them are involved in traffic accidents each year. Traveling in traffic is often the biggest risk you are exposed to during a normal working day. A total of 36% of fatal road accidents in Norway involve at least one driver who has been driving for work (Nævestad et al., 2015). An earlier study found that approximately 1 500 people are injured annually on Norwegian roads in work-related traffic accidents (Nævestad et al., 2015)6.

In addition, the municipalities have an indirect influence on traffic safety in a number of areas, including through their role as road owner, which includes tasks related to road design, lighting and speed limits on municipal roads. The municipality also influences traffic safety as a transport buyer, through establishing safety requirements to actors that they buy transport services from, for example safety equipment, safety routines, etc. Within adolescence and school, the municipalities also have a large indirect influence on traffic safety through information and training, aimed at parents and schoolchildren, which child health clinic, kindergartens and schools provide.

Previous studies show that measures aimed at safety culture and safety management can lead to a significant decrease in traffic accidents in organizations that employ drivers (Nævestad et al., 2018). Safety culture refers to shared and safety-relevant ways of thinking or acting that are (re)created through joint negotiation by people in social settings (Nævestad, 2010). The few robust studies that exist of this indicate that safety culture and safety management measures can reduce the incidence of traffic accidents by between 20% (Wouters & Bos, 2000) and 60% (Gregersen et al., 1996; Nævestad et al., 2018). There is reason to believe that these conclusions also apply to municipal agencies.

However, previous research indicates that organizations with employees who drive at work seem to have little focus on the importance of safety culture and safety management (Nævestad et al., 2015). It has been pointed out that this applies to both transport companies, public agencies and other organizations that have employees who drive in their work. In 2015, The Norwegian Road Safety Coincil (Trygg Trafikk) launched the concept ‘Traffic Safe Municipality’. This is an approval scheme that defines criteria for systematic traffic safety work, which the municipality's agencies must fulfil, for the municipality to be approved. As of December 2022, there were approximately 151 municipalities that had been approved as Traffic Safe Municipalities out of a total of about 350 municipalities in Norway. The first was approved in August 2015, and the approved municipalities are scheduled to be re-approved after three years. The re-approval involves a new evaluation of whether criteria are fulfilled and further measures aiming to improve the traffic safety work.

In a previous study, we concluded that Traffic Safe Municipality, and the work with traffic safety plans which are a fundamental part of this, can be defined as a measure to establish traffic safety culture through the implementation of a safety management system (SMS) in the municipality (Nævestad et al., 2017; Nævestad et al., 2017). There are two reasons for this. First, as Traffic Safe Municipality focuses on defining roles, responsibilities and procedures for traffic safety in the municipality (Thomas, 2012). Second, as it has a traffic safety plan as a key element, in line with the continuous improvement, which is fundamental in the SMS approach. In addition, safety culture change is one of the main goals of Traffic Safe Municipality.

SMS relates to the formal aspects of safety in organizations (‘how things should be done’). SMS usually consists of formal routines and measures that enable the organization to work systematically with safety, for example by establishing formal safety policies and objectives, establishing important roles and responsibilities, systematically gathering information about incidents and hazards, developing countermeasures, monitoring their effect, and adjusting measures if necessary (Thomas, 2012). The principles in SMS are often in line with a continuous improvement approach and process: ‘Plan, Do, Check, Adjust’. Furthermore, the SMS defines the company's responsibility to prevent work-related accidents through the implementation of such formal measures. The informal aspects of safety management generally refer to safety culture (‘how things are actually done’), i.e. shared and safety-relevant ways of thinking or acting (Nævestad, 2010). The research into safety management and safety culture often shows that it does not help with extensive formal systems for safety, if the employees in practice do something other than what the systems say they should do.

The development of a positive safety culture is a formal goal and rationale for SMS in the maritime sector (IMO, n/d), in aviation (ICAO, 2016), in rail (ERA, n/d), and also in the voluntary ISO:39001 standard in the road sector (ISO, 2012). Thus, it is assumed that the development of formal procedures (‘how things should be done’) will influence informal practice (‘how things are actually done’) (Nævestad et al., 2018). Studies also indicate positive effects of SMS on safety culture in aviation (Adjekum & Tous, 2020), in the maritime sector (Lappalainen et al., 2012) and in railways (Zuschlag et al., 2016). Thus, although there are relatively few systematic studies in this area, research suggests that the implementation of SMS can lead to increased traffic safety.

In order for a Traffic Safe Municipality to be as effective a traffic safety measure as possible, we need more knowledge about relationships between Traffic Safe Municipality and safety outcomes (traffic safety culture, traffic safety behaviour, etc.)

The study has four aims:

We have developed the following hypotheses based on the previous research (cf. section 1.5):

The background of Traffic Safe Municipality is that previous municipal traffic safety plans generally focused on physical measures (e.g. walkways, cycleways, speed bumps, signage, design of intersections) to improve traffic safety. However, the physical aspect, which is typically included in traffic safety plans, comprises only a limited part of the total work on traffic safety in a municipality. Much of the important traffic safety work is given at child health clinics, in kindergartens, and at schools. Information and training are important parts of the traffic safety work that is carried out.

One of the main principles of the Traffic Safe Municipality approval scheme is that traffic safety is naturally integrated in all the municipality's sectors and that every employee has a contribution to the municipal and national traffic safety work. The municipality is a large workplace, and around 10% of the municipality's residents are employed by the municipality. Around 16%–18% of the working population is employed by the municipality.

Traffic Safe Municipality is an approval scheme that is based on laws and regulations that the municipality already is required to follow. There isthus no additional costs associated with the approval. The aim of the approval is to create cross-sectorial work on traffic safety. In order to be approved as a Traffic Safe Municipality, there are certain criteria that must be met. The criteria are based on routines and systems in the respective sector. The municipality must anchor responsibility for traffic safety work with the mayor and municipal director, while partial responsibility may lie with the individual head of department. Upon approval and regulatory approval, the municipal director must document in writing that:

There are also additional criteria for each sector in the municipality, e.g. kindergarten, school, technical department, health.

As indicated above, research suggests that the implementation of SMS can lead to increased traffic safety. An extensive literature review indicates that SMS implementation is related to positive safety outcomes (Thomas, 2012). Based on previous research, we can assume that the relationship between SMS and culture is explained by SMS leading to new (common) ways of acting in the organization (Reason, 1997). In addition, it is an explicit aim of many SMS interventions that the new ways of acting that are introduced (for example routines for reporting, risk analyses) should lead to new (common) ways of thinking, or new ways of understanding hazards in the work on (Phillips et al., 2018). These new ways of understanding hazards at work will motivate and reinforce the new ways of acting (Nævestad & Milch, 2021). Finally, the relationship between SMS and safety culture can be explained by employees experiencing an increased management focus on safety in the organisation, which makes it easier to address safety topics and prioritize safety in work (Nævestad, 2010; Schein, 2004). In line with these assumptions, several studies from different transport sectors find relationships between the implementation of SMS, or SMS elements, and positive safety culture in aviation (Hudson, 2003; Patankar, 2019), in railways (Zuschlag et al., 2016), in the maritime sector (Lappalainen et al., 2012) and in the road sector (Naveh & Katz-Navon, 2015).

Several of the above mentioned studies also find that improvements in safety culture also lead to improvements in safety behaviours (and subsequently reductions in accidents and safety incidents) (Naveh & Katz-Navon, 2015; Zuschlag et al., 2016).Naveh and Katz-Navon (2015) find an average reduction of 75% in traffic safety violations in the intervention units where safety climate was improved, while the control units had an increase in violations. The study of Zuschlag et al. (2016) from the railways finds similar relationships: improved safety culture, 80% reduction in risk-taking behavior and 81% reduction in accidents. The study has both before and after measurements and test and control groups.

In contrast to the assumed relationship between SMS and safety culture described above, several studies report discrepancies between formal and informal safety aspects in organizations (Antonsen, 2009; Bye & Aalberg, 2020; Størkersen et al., 2020). This suggests that it is not always the case that a positive safety culture automatically follows from SMS implementation. Several investigations of accidents often point to a gap between formal and informal aspects of safety culture, and that the organization 'did something different in practice than what was said in formal documents and procedures' (Nævestad, 2010). For example, this can be related to systems that are poorly adapted to the work the employees have to do, or to the organization in which they are implemented (Bye & Aalberg, 2020; Størkersen et al., 2020), it may also be related to workload and stress (Bye & Aalberg, 2020; Størkersen et al., 2020), and lacking senior managers' commitment (Edkins, 1998; Lappalainen et al., 2012; Thomas, 2012; Zuschlag et al., 2016). The mentioned studies also point to lacking employee motivation and support, as well as mistrust between managers and employees. Finally, support from a third party is also mentioned as an influencing factor by Zuschlag et al. (2016).

We have conducted in-depth interviews with 17 key people in 16 municipalities. The purpose of the interviews was to identify factors that promote or hinder implementation of the Traffic Safe Municipality initiative, to discuss experiences with implementation and the consequences of being approved as a Traffic Safe Municipality. In-depth interviews are particularly suitable when there is little previous research, and there is a need for in-depth knowledge of a topic (Brinkmann & Kvale, 2014).

The qualitative interviews were conducted in the summer and autumn of 2021. The majority of the interviews were conducted digitally via the Microsoft Teams platform. Three of the interviews were conducted over telephone. The interviewees received a request for participation via e-mail. Informed consent was obtained prior to the interviews. A semi-structured interview guide was used, which means that the interview is structured around some predefined themes. The guide is indicative in the sense that both the questions and the order of the questions are adapted to the interview situation, so that it is possible to follow up on new information or topics that emerge along the way. The interview guide contains questions about the process of introducing Traffic Safe Municipality, which people and roles have been particularly involved in the work, the traffic safety plan and other ways in which Traffic Safe Municipality is formalized in the municipality's activities. In addition, the guide contains questions about factors that have promoted or hindered the introduction of the measure.

Table 1 shows the age distribution of the respondents.

We see that 58% of respondents are between 36 and 55 years old. A total of 75% (n = 1685) of the respondents in the sample are women, while 25% are men (n = 570). The high proportion of women can be explained by showing which sectors in the municipality the respondents are recruited from. This is particularly the primary school, home care and kindergartens, where women are in a clear majority.

Table 2 shows the sectors in which the respondents are employed.

Table 3 shows at which administrative level in the municipality the respondents work.

Two thirds of the respondents define themselves as employees. The total number of respondents in the table is lower than in Table 2, because we lack information on administrative level for the respondents from municipal engineering.

The first aim of the study is to examine whether implementation of the criteria for Traffic Safe Municipality is related to perceptions of safety culture improvements among respondents in the approved municipalities.

The third aim is to map factors that promote or inhibit the implementation of Traffic Safe Municipality. In the following, results from the thematic analysis of the interviews are presented.

Involvement from top management. It is the first promoting factor that we have identified in the analysis. Several of those we interviewed mentioned that support from the municipal management was of great importance for the anchoring of Traffic Safe Municipality in the various agencies in the municipality. This indicates that it is very important that the municipal director and mayor have participated in the process and have shown their support for the initiative politically. First, this support is important to legitimize that this is something the municipality commits to and will work with in all the units in the municipality:

‘It is absolutely necessary that the mayor says that we should go for it. When the mayor and the management group are involved, the entire municipality becomes involved. Political decision is important for the result.’ (Interviewee, Municipality Q)

Second, involvement from top management is important for the extent to which the measure is formalized in the organization, through planning, objectives and traffic safety plans, and third, it seems to be important for creating commitment among managers and employees in the various agencies.

Continuity and dedicated personnel. In order for employees to be aware of what it means to be a Traffic Safe Municipality, it must be put regularly on the agenda and followed up in the organization. A dedicated promoter who is responsible for coordinating and following up the work towards the municipal units was mentioned by several as absolutely crucial for the successful implementation and follow-up in the organization.

‘If you don't have a person who takes care of it and who coordinates it, then no one will take care of it.’ (Interviewee, Municipality X)

‘It requires a committed person. It requires that a person with responsibility has been added, and that it is anchored in governing documents, so that it is not perceived as something that I, as a consultant, have made up.’ (Interviewee, Municipality S)

An important aspect of this is to have a person with the resources and time to work continuously with the initiative. An equally important aspect that was highlighted is that the person has the interest and willingness to take on such a role. Many municipalities generally experienced a high turnover and replacement of personnel. In the interviews, this was mentioned by several as one of the most important challenges for coordinating and following up the work with Traffic Safe Municipality.

‘That [replacement of personnel] is the big challenge here. In a municipality, employees will always be replaced. In order for it [Traffic Safe Municipality] to function optimally, you need someone who has dedicated time to update and ensure that things are done.’ (Interviewee, Municipality H)

Good internal communication. To anchor the initiative in the various units, the importance of having established contact points with relevant managers and people was highlighted as a central success criterion by several interviewees.

Competing priorities in the municipality and high workload. In many municipalities, traffic safety is not regarded as the biggest challenge, and there may thus be other priority areas or objectives that are emphasized more. Several interviewees believed that anchoring the measure formally in planning and routines was very important, to achieve a good implementation. Without formal anchoring, it is easy for the work to be de-prioritized at the expense of other prioritized tasks. Compliance with the municipality's other priority areas/objectives is also important. The municipal employees generally have a high workload, which makes it challenging to set aside time to work with Traffic Safe Municipality in a busy working day.

External factors. All interviewees mentioned assistance from The Norwegian Road Safety Council as an important support in the work to introduce the measure. Both the administrative aids in the form of checklists and routines, but also professional assistance in adapting the measure to the municipality were mentioned. Other external factors mentioned by the interviewees was the support from county safety councils allocating funding, and that Traffic Safe Municipality is mentioned in the National Transport Plan. The interviews also indicate the importance of two negative external factors. The first is reorganization followed by county and municipal mergers, which occurred in 2020. This has required large administrative resources for some municipalities. The second factor was the COVID-19 pandemic in 2020, which created major upheavals and many additional tasks for the municipalities, leaving less time to work with Traffic Safe Municipality.

‘There is a state of emergency in the municipality during the day. Many things are higher on the priority list than traffic safety. Infection control and outbreaks and everything else.’ (Interviewee, Municipality H)

The first aim of the study was to examine whether implementation of the criteria for Traffic Safe Municipality is related to perceptions of safety culture improvements among respondents in the approved municipalities. Our first hypothesis was that implementation of the criteria for Traffic safe municipality has led to perceptions of safety culture improvement in the approved municipalities. Our results largely support Hypothesis 1, as respondents report positive changes in safety culture. The background for Hypothesis 1 is, first, that, based on previous research, we assumed that the implementation of the criteria for Traffic Safe Municipality has led to new ways of thinking and acting in relation to traffic safety. The assumed main explanation for the relationship between SMS and safety culture is that it is assumed that the development of formal procedures (‘how things should be done’) will influence informal practice (‘how things are actually done’) and facilitate new (joint) ways of acting in the organizations (Nævestad et al., 2018). This is the explicit motivation and justification for most of the sector legislation that requires SMS (ERA, n/d; ICAO, 2016; IMO, n/d). The relationships we have seen between the implementation of Traffic Safe Municipality and reported traffic safety culture improvements in the quantitative data support this assumption. So does the qualitative data. We received several good examples of new routines being introduced in the municipalities that took place in the study, for example relating to risk analyses, travel policies etc. In addition, the respondents in the survey report that they behave in a safer way when traveling in traffic at work.

The second aim of the study was to examine whether potential changes in safety culture are related to respondents’ self-reported increased focus on traffic safety in their work activities. SMS interventions are often based on the assumption that when the new ways of acting are introduced, these will lead to new (common) ways of thinking, or new ways of understanding hazards in the work (Kongsvik & Gjøsund, 2016; Lappalainen et al., 2012; Naveh & Katz-Navon, 2015; Phillips et al., 2018). The results of the survey support this assumption, because the respondents state that they think more about traffic safety, and because they state that they have learned more about how all the municipality's departments can contribute to increased traffic safety. We also saw this in the qualitative data, where a new ‘awareness’ was a word that was repeated as an example of an important consequence of being approved as a Traffic Safe Municipality. It was mentioned, for example, that approval as a Traffic Safe Municipality leads to a greater awareness of how different agencies in the municipality can have an indirect and direct influence on traffic safety. Thus, our results support Hypothesis 2: ‘Perceptions of traffic safety culture improvements are related to respondents' self-reported increased focus on traffic safety in their work’. Previous research also indicates that new ways of understanding hazards in work may motivate and reinforce the new ways of acting (Nævestad et al., 2021; Nævestad, 2010). Our results also resonate with this, because we find a relationship between thinking more about traffic safety and behaving in a safer way in traffic during working hours.

The third aim of the study was to examine whether potential changes in safety culture are related to self-reported improvements in respondents’ traffic safety behaviours. We find that self-reported improvements in traffic safety culture has an isolated effect on traffic safety behaviours in the multivariate analyses. This result is in line with Hypothesis 3: ‘Perceptions of traffic safety culture improvements are related to self-reported traffic safety behaviour improvements among the respondents‘. The result is also in accordance with previous research (Naveh & Katz-Navon, 2015; Zuschlag et al., 2016). Moreover, in line with the mentioned previous research, we find that SMS influences safety behaviour as it leads to safety culture change. In the multivariate analysis of factors influencing traffic safety behaviours, we found that the variable measuring SMS implementation ceased to contribute significantly when the traffic safety culture variable was introduced.

However, the most important predictor of changes in respondents’ traffic safety behaviours in the multivariate analyses was the variable 'Think more about traffic safety' (as a result of being approved as a Traffic Safe Municipality). This result indicates that the most important mechanism between safety culture change and safety behaviour change goes through respondents’ self-reported thoughts about traffic safety. It seems that safety culture change makes respondents think more about traffic safety and subsequently act safer in traffic. The relationship between thought and actions has been investigated in a number of different psychological theories and studies. According to the 'expectancy value model', the attitude towards a certain behavior is influenced by perceptions of the consequences, which are weighed by the subjective value of the consequences (Fishbein & Ajzen, 1975). Theory of Planned Behavior (TPB) again describes how attitudes are shaped into behavioral intentions and behaviour. The TPB assumes that our behavior is a result of our intention to perform the behavior, and that our intention to perform a specific behavior is influenced by our attitudes to the behavior, subjective norms and our perceived control over our own behavior (Ajzen, 1991).

The above-mentioned isolated effect of traffic safety culture change on behaviour change seems to indicate mechanisms additional to the relationship between thoughts and actions. This mechanism might be related to what colleagues do (‘descriptive norms‘) and, for example, the management focus on traffic safety. The contribution of the variable 'Top management focus on traffic safety' is reduced when the variable traffic safety culture change is introduced, indicating that the direct contribution of traffic safety culture change is related to this. The importance of management focus as a facilitating variable, was also highlighted in the qualitative data and in previous research (Flin et al., 2000; Thomas, 2012). Thus, we have identified at least two mechanisms between safety culture change and safety behavior change, that should be examined further in future research. The third potential mechanism ‘descriptive norms’, or ‘peer pressure’ in the workplace, should also be examined in future research. Qualitative interview data indicates that the safety culture changes involved new shared points of reference, legitimizing and justifying safer traffic safety behaviours (and also sanctioning of unsafe driving among inhabitants).

The fourth aim of the study was to map factors that promote or inhibit implementation of a Traffic Safe Municipality. Although we have seen consistently positive results of being approved as a Traffic Safe Municipality, both the implementation and the results varied in the various municipalities. The interviews indicate that this is related to the quality of the implementation of the criteria for Traffic Safe Municipality, and different factors influencing this.

When it comes to factors that promote implementation, quantitative and qualitative results indicate that top managers' commitment throughout the intervention period is one of the most important prerequisites for a successful implementation. This result, which is in accordance with Hypothesis 4, is also emphasized in several previous studies from many different sectors (Edkins, 1998; Flin et al., 2000; Lappalainen et al., 2012; Thomas, 2012; Zuschlag et al., 2016). The fact that Traffic Safe Municipality is something that top management is concerned with and puts on the agenda helps to legitimize the measure and to highlight it as an important priority area in the municipality.

When it comes to factors inhibiting implementation that are identified in our analysis, these are largely about organization and available resources, rather than factors related to the SMS measure that is implemented (Traffic Safe Municipality). In our analysis, we find that a lack of resources in the municipality, time pressure and high workload among employees seem to be the most important factors that may make it challenging to implement Traffic Safe Municipality in daily work. This matches with previous research on the implementation of SMS and safety culture, which shows that if managers and employees experience strong work pressure, they may have little time to work with SMS, and this can affect the quality of the process (Bye & Aalberg, 2020; Størkersen et al., 2020). Municipal employees generally have many tasks and often have to prioritize important tasks over others. The COVID-19 pandemic is an example of an acute and unforeseen situation that has required a lot of resources. Reorganisation and replacement of personnel are also important challenges. This also matches with findings from several studies which show that reorganization and replacements of people who have important roles in the introduction of a measure can have negative effects on the implementation of the measure (Amtrak, 2015; Zuschlag et al., 2016).

We illustrate the links that we observed in the multivariate analyzes in Figure 1. The figure indicates relationships between: (1) Influencing factors, (2) system implementation, (3) development of safety culture, and (4) safety behaviour. The most important variable related to perceived traffic safety culture change was the variable measuring implementation of Traffic Safe Municipality criteria. The most important variable related to whether respondents report that they think more about traffic safety was perceived changes in traffic safety culture. The most important variable related to respondents’ reports that they behave safer in traffic was that they think more about traffic safety. Finally, it is also important to note that the model assumes both direct and indirect effects on traffic safety. The first occurs through the influence of employees' safety behavior in traffic. The second occurs as employees in the municipality indirectly influences traffic safety. We have unfortunately not measured the indirect influence quantitatively.

Unknown response rate. We have not calculated response rates for the various municipalities and sectors, because we lack information on the number of people who have received the link to the survey. However, it is a strength that we have a large sample of respondents and a relatively large number of municipalities in the survey.

Representativity of municipalities. When we assess the results, it is important to consider how representative the municipalities that have responded to the survey are for the population of municipalities that are Traffic Safe Municipalities. This is especially true when we do not know the response rate. Hypothetically, there may be reason to assume that the municipalities that have participated are particularly dedicated. However, it can be noted that the variation in the results suggests that this is not necessarily the case. It must also be mentioned that we have recruited the municipalities that were about to be re-approved over almost two years. There is therefore reason to believe that the participating municipalities are representative of Traffic Safe Municipalities in general.

Representativity of respondents. Another relevant question is how representative the respondents who have answered are. We have seen that 75% of the respondents are women. At the same time, 76% of the respondents were recruited from kindergartens, schools and the municipality's home care. Comparable information from Statistics Norway shows that 90% of kindergarten staff are women, while 75% of primary school teachers are women (SSB, 2018). The high proportion of women in our sample thus appears to be in line with the proportion of women in the groups from which we have recruited the most respondents.

Our conclusions are largely based on self-reported numbers. A possible limitation is that our conclusions are based on self-reported numbers, with the possible biases that this may entail. Respondents can, for example, remember incorrectly, they can potentially over-report positive things and under-report negative things, etc. These are known issues when using self-reported data. However, we have no indications of substantial effects of this in our data.

We have not conducted a before and after study with a control group. A major weakness of the present study is that it is a cross-sectional study. If we had carried out a before and after studies with both test and control groups, we could potentially have gained better control over cause and effect in the study (cf. section 4.7.1). Moreover, it is also more likely that we could have gained better control over the effect of implementation of the Traffic Safe Municipality criteria compared, for example, with the general development in society over time such as an increased focus on traffic safety. Nevertheless, it is important to note that we also see variation among the municipalities in our relatively large sample, and we get important information about relationships between key variables, that can be examined further in future research.

We do not estimate the effect on accidents. Actual traffic accidents are the most important measure of the effect of traffic safety measures, because this is what you want to influence with traffic safety measures. However, we do not estimate the effect on accidents in our study. The reason is that we are focusing on Traffic Safe Municipality as a traffic safety culture initiative, that the number of accidents in the participating municipalities is small, and because we have reason to believe from previous studies that there is a relationship between safety culture and traffic accidents (Bjørnskau & Nævestad, 2013). It is particularly important to focus on traffic safety culture as an outcome measure for traffic safety, since the number of accidents and injuries in traffic has decreased dramatically over the past 20 years. With a small number of people killed and injured in traffic, the numerical basis for assessing the effects of measures is smaller. Then alternative indicators, such as traffic safety culture, are a good solution because these provide a broad and multifaceted measure of traffic safety, which we know from previous studies is related to behaviour, incidents and accidents (Bjørnskau & Nævestad, 2013).