Values, personal norms and speeding
intention
The moderating effect of injunctive norms and mediating
effect of environmental self-identity
By
Linda Jansen
Values, personal norms and speeding
intention
The moderating effect of injunctive norms and mediating
effect of environmental self-identity
Author
Linda Jansen (s2749734)
l.jansen.6@student.rug.nl
Bernoulliplein 17
9714 BV Groningen
+31611068461
Master Thesis
June 17, 2019
University of Groningen
MSc Marketing Management
Department of Marketing
Faculty of Economic and Business
First supervisor
Second supervisor
ABSTRACT
The present paper examined the relation between values, personal norms and speeding intention. Furthermore, it was tested if the injunctive norm and one’s environmental self-identity play a role. A survey with a 3 by 2 between-subjects factorial design was used to test if the hypotheses that seeing a biospheric or altruistic message resulted in a higher intention to follow the speed limit (compared to an egoistic message) and whether this was mediated by the personal norm towards speeding. Although only the altruistic message was effective, results showed that both a higher biospheric and altruistic value orientation lead to a higher intention to keep within the speed limit. Moreover, there was evidence that personal norms play an important role in this relationship. Hence, the present paper shows the relevance of using values and personal norms in the context of speeding behavior. Also, environmental self-identity shows a mediating role with regard to biospheric and altruistic value orientation and the personal norm. Injunctive norms, on the other side, do not have a moderating effect in the present context. Since also the practical implications are of much relevance, researchers are stimulated to further research this topic.
Key words: values, personal norm, speeding intention, injunctive norm, environmental
TABLE OF CONTENTS
INTRODUCTION ... 5
LITERATURE REVIEW ... 8
Intentions to reduce speeding ... 8
Values and intentions to reduce speeding ... 9
Personal norms and intentions to reduce speeding ... 12
Environmental self-identity as a mediator between altruistic and biospheric values and the personal norm towards speeding ... 13
Injunctive norm as a moderator between values and the personal norm towards speeding ... 14
Conceptual model ... 16 METHODOLOGY ... 17 Participants ... 17 Design ... 17 Materials ... 18 Procedure ... 19 Plan of analysis... 20 RESULTS ... 21 Descriptive statistics ... 21 Manipulation checks ... 24 Hypotheses testing ... 25 DISCUSSION ... 38 Theoretical implications ... 40 Practical implications ... 41
Limitations & further research ... 41
Conclusion ... 42
REFERENCES ... 43
APPENDIX A – Messages on posters ... 47
APPENDIX B – Questionnaire ... 49
APPENDIX C – Randomization check ... 52
APPENDIX D – Normal Q-Q plots ... 53
INTRODUCTION
It is estimated that between 2030 and 2052 global warming will reach 1,5C if it continues to increase at the current rate (IPCC, 2018). This will lead to an increase risk to, among others, health, water supply, and human security. An important contributor to global warming is the increase in carbon dioxide (CO2) (IPCC, 2014; Friedlingstein & Solomon, 2005). It is expected
that today’s carbon dioxide emissions contribute significantly to the climate of future generations. This is caused by an increase in the concentration of CO2 in the atmosphere as we
burn fossil fuels, which enhances the natural greenhouse effect and consequently warms the planet (Anderson, Hawkins & Jones, 2016). One industry that depends on fossil fuels is the transport sector and it can be seen as a major contributor of greenhouse gases (Chapman, 2007). According to the European Commission, 12% of the total EU emissions of CO2 are caused by
cars (EC, n.d.). Besides, this contributes to increased air pollution (Chapman, 2007).
One low cost opportunity for emission reductions in transport is support for eco-driving (ECMT, 2007). Eco-driving can reduce both fuel usage and thereby the emissions of CO2 and
other pollutants (Wengraf, 2012). Moreover, it can lead to improvements in road safety. Examples of eco-driving behaviours are vehicle maintenance, changing up gears as soon as possible, and smooth and reasonable speeds. Aggressive driving for example can raise fuel consumption by 37% (Wengraf, 2012). According to Barkenbus (2010), the alteration of the current driving styles is an overlooked action that could produce significant CO2 savings.
Eco-driving is described as the change from an aggressive Eco-driving style to a more refined style. Savings benefit both individuals (e.g. cost savings), as well as society at large (e.g. reduced CO2 emissions). Hence, a win-win.
Previous research on eco-driving focused, among others, on the goal of avoiding road-traffic noise, safety, time saving, and fuel saving (Lauper, Moser, Fischer, Matthies & Kaufmann-Hayoz, 2015; Dogan, Steg & Delhomme, 2011). However, recent research took a different perspective and studied the role of moral considerations in promoting eco-driving (Ünal, Steg & Gorsira, 2018). This paper will focus on one aspect of eco-driving, namely driving at the speed limit, and will take an ethically oriented consumer behaviour approach. The present paper will examine how moral considerations can be used in motivating people to following the speed limit.
(Wegman & Aarts, 2005). CO2 emissions are released during the combustion of petrol and
diesel and contribute to the greenhouse effect. Litres of fuel can be saved by keeping the speed limit and maintaining a sedate driving behaviour. Besides a positive effect to the environment, keeping the speed limit also enhances safety on the road (Barkenbus, 2010). Furthermore, it is estimated that one third of all fatal crashes is (partly) due to driving too fast (SWOV, 2016). In 2017 there were 613 road fatalities in the Netherlands, a little bit less than the year before (KiM, 2018). However, the aim is to bring this number down to zero in 2030. On a yearly basis approximately ten pedestrians and cyclists die due to speeding violations of 10-15 km above the limits in 30 and 50 km/h areas (Rijksoverheid, n.d.)
Nevertheless, it is not easy to convince people to keep the speed limit since individual interests are not in line with social interests (Wegman & Aarts, 2005). Individual drivers mostly experience the benefits of speeding, such as the fun and excitement. Moreover, time saving is an important reason for individuals to speed since this will result in an earlier arrival time at the destination (Wegman & Aarts, 2005). Also, drivers speed because they want to adjust to other traffic or because they are not aware that they are speeding (Duijm, Kraker, Schalwijk, Boekwijt & Zandvliet, 2012). The negative environmental effects are for the individual driver more abstract and further away than the individual advantages (Wegman & Aarts, 2005). The safety risks of driving too fast not only affect the driver but also other road users. Hence, the cons of speeding are in particular visible at the social level while the benefits of it are noticeable on the individual level.
This conflict between the social and individual consequences of speeding can be seen as a social dilemma, which concerns a situation in which a conflict is experienced between (short term) individual interests and (long-term) collective interest (Dawes & Messick, 2000). Hence, the collective interest in this case is to follow the speed limit. It can be argued that people can be motivated to comply to the speed limit based on moral considerations. These considerations include safety for others and a concern for the environment. Road safety is the most important reason for keeping within the speed limit (Duijm et al., 2012). Environmental concern on the other side is considered less often with regard to compliance with the speed limit. Thus, it seems that people are acting from an egoistic perspective when it comes to speeding, while there are clear moral reasons for keeping within the speed limit as well. A question that rises is then to what extent moral considerations are important for the intention to follow the speed limit.
obligation) and the associated behavior can be influenced by values. Values are defined as desirable goals that transcend situations and serve as guiding principles in the individual’s life (Schwartz, 1992). By focussing attention towards specific values and thereby increasing their saliency, the way one’s values direct attention to value-congruent behavior can be affected (De Groot & Thøgersen, 2019). The present study argues that making biospheric and altruistic values more salient by means of normative messages can reduce speeding behavior, because these values will activate one’s personal norm towards driving safety.
Within the path from values to personal norms to intention to keep within the speed limit, other variables might play an influencing role. Therefore, the present study also takes the concepts of environmental identity and injunctive norms into account. Environmental self-identity is defined as the extent to which we see ourselves as a pro-environmental person (Van der Werff, Steg & Keizer, 2013). Biospheric values influence one’s environmental self-identity, which in turn influences the personal norm (Van der Werff & Steg, 2016). Injunctive norms are the behaviours that are typically approved or disapproved (Cialdini, 2003). Injunctive norms have been found to influence personal norms (Doran & Larsen, 2016).
The research question therefore is: How do values, injunctive norms towards speeding
LITERATURE REVIEW
Intentions to reduce speeding
Previous actions aimed at reducing drivers’ speed, like road-safety communication campaigns are often ineffective (Delhomme, Chappé, Grenier, Pinto & Martha, 2010). These campaigns focus most of the time on the consequences of speeding in relation to crashes or sanctions for violations (Delhomme et al., 2010), which are typically regarded as ‘’egoistic’’ considerations. In information campaigns it is assumed that arguments based on egoistic considerations are more persuasive than altruistic or biospheric considerations (De Groot & Steg, 2009). For example, reducing speed violations with regard to young adults could be accomplished by offering financial incentives on car insurance (Bolderdijk, Knockaert, Steg & Verhoef, 2011). However, they reasoned that the fact that one is being monitored might deter drivers from speeding and that the fading awareness of this resulted again in increased speeding. Other approaches to make speeding less attractive also focused on targeting ‘’egoisitic’’ goals such as speeding cameras and speed bumps (Steg, Lindenberg & Keizer, 2016). This may promote requisite behavior but in general provide a fickle basis for reducing driving speed since these approaches are targeted at motivating egoistic considerations only.
People tend to act a priori more on egoistic considerations and less on altruistic and biospheric considerations (De Groot & Steg, 2009). However, since these egoistic considerations provide a fickle basis for reducing one’s driving speed in the long term as suggested above this seems to be problematic. Therefore, it is important to focus on altruistic and biospheric considerations as well. The intention to follow the speed limit can be treated as a pro-environmental behavior as these behaviors often do not benefit individual interests in the short term but mainly benefits other people or the environment (Thøgersen, 1996), which is the case with reducing one’s speed. Following the speed limit namely enhances safety on the road and benefits the environment.
With regard to biospheric considerations, research suggests that people may find environmental benefits more worthwhile than financial benefits concerning specific eco-driving behaviours (Dogan, Bolderdijk & Steg, 2014). Delhomme et al. (2010) for example thought of a new argument about a different risk of speeding, which is related to the environment, namely air-pollution. They reasoned that the pollution of car use could be important enough to drivers to abide by the speed limits. Results showed that the air-pollution argument was more effective compared to a crash-risk argument, but only for women. However, they argued that at the time the studies were carried out, environmental problems were not as present in the media and so available in the mind. Since CO2 emissions are still rising and global warming is an actual topic these days (IPCC, 2018), an intervention with an environment argument might be more effective.
As there seems to be a conflict between egoistic and social consequences when reducing one’s speed limit, it is interesting to look at the underlying mechanisms that explain the choices one makes. Therefore, egoistic and social values, and in particular biospheric values, seem to be relevant (e.g. Ünal et al., 2018). Values can be seen as motivational factors when it comes to intentions. There is a conflict between egoistic values on the one hand and altruistic and biospheric values on the other hand (De Groot & Steg, 2009), which typically include a dilemma between the self and the social such as the case with reducing one’s speed.
Values and intentions to reduce speeding
Values are desirable goals that transcend situations and serve as guiding principles in the individual’s life (Schwartz, 1992). Values vary in their relative importance and choices are based on the value that is considered most important (De Groot & Thøgersen, 2019). As Steg (2016) stated, people tend to favour and choose options that support the values they prioritize. Subsequently, options are evaluated in an overly positive or negative way based on how these options are expected to impact their values. Values reflect the overarching goals people find more important in life and are relatively stable over time (Steg, Bolderdijk, Keizer & Perlaviciute, 2014). There are four values which all individuals endorse to some extent: hedonic, egoistic, altruistic, and biospheric values.
Dietz & Kalof, 1993). Acting on egoistic values on the other hand does often not result in pro-environmental behaviour (De Groot & Steg, 2009). This can be explained by the personal costs of this behavior which are often higher than the personal benefits when it concerns egoistic values. Hence, a risk of relying on self-interest (i.e. economic appeals) as a tool to promote pro-environmental behaviour is that it is not even effective in changing the target behaviour (Bolderdijk, Steg, Geller, Lehman & Postmes, 2013). In case of altruistic and biospheric values, there are high societal and environmental benefits connected to behaving pro-environmentally (De Groot & Steg, 2009). Since behavior in favour of the environment is most likely when altruistic and/or biospheric values are strong, acting pro-environmentally is typically seen as acting morally right.
De Groot and Thøgersen (2019) stated that values mostly influence behavior indirectly (via beliefs, attitudes, and norms). This is because they reflect general goals people strive for (Steg, 2016). However, a direct relationship has also been found. Making the value more salient, by focussing attention towards the specific value, can influence the way one’s values directs attention to value-congruent information. Strengthening the saliency of altruistic and biospheric values in a certain situation, may enhance stable (i.e. long term) pro-environmental behavior (De Groot & Steg, 2009). This affects the values which people focus on and ultimately how they influence behavior.
Increasing the saliency of altruistic and biospheric values can be achieved via different ways. Verplanken and Holland (2002) primed environmental values and studied the effects of value activation. One way to activate values is by activating the self through enhanced self-focus. They argued that values only influence behavior if they are sufficiently central to a person’s self-concept. In the study of Bolderdijk et al. (2013) biospheric values were made more salient by means of focusing on environmental reasons. As a consequence, people had a more positive self-concept and it led to more pro-environmental behavior. Moreover, economic appeals (which are related to the egoistic aspects) triggered less compliance than environmental appeals. Thus, the response to an appeal may be based on the influence of compliance on the self-concept.
speeding. The second way concerns making values subject to a process of moralization, where they are connected to supporting emotions. These emotions can include, among others, expressions of disgust in case of deviant behaviour. For example, saying that it is a bad thing when someone is violating the speed limit.
Two other important factors that influence the way how people act in line with their values are the costliness of behavior and situational factors (Steg, 2016). The costliness of behavior means that people are less likely in acting pro-environmentally when this behavior is associated with high financial cost, effort, or inconvenience. In case with speeding it can for example be argued that people are less likely to keep the speed limit when they are in a hurry. However, compared to switching from car to public transport, individual cost of keeping the speed limit are not that high which makes it more likely people are intended to reduce speeding. Situational factors can act as a reminder for the values people find important, since they make a specific value more salient (Steg et al., 2014). Situational factors may consist of value-related symbols (e.g. environmental symbols or moral symbols), but also the extent to which others acted on particular values (Steg, 2016). A situational cue that signals that other people are not doing the right thing increases the probability that people will violate norms in general (Steg et al., 2014). On the other hand, a situational cue that signals that others respect the norm (i.e. doing the right thing) increases the probability that people are conforming to norms.
Personal norms and intentions to reduce speeding
The personal norm first has to be activated to become relevant as a determinant of behavior (Schwartz & Howard, 1981). According to the norm-activation theory, there are four conditions which have to be met to activate a norm. These conditions are awareness of need, awareness of consequences, ascription of responsibility, and perceived behavioural control. As Klöckner (2013) stated, many researchers have applied the norm activation theory in explaining environmental behaviour. Personal norms also contribute to the explanation of pro-environmental behaviours, even when these behaviours have an often invisible and objectively small contribution to reduce environmental damages (Bamberg, Hunecke & Blöbaum, 2007). Following the speed limit is such a behaviour. The value-belief-norm theory (Stern, 2000) in turn links the assumptions of the norm activation theory to the relation between general values, environmental beliefs and behavior. In accordance with the norm activation theory, value-belief-norm theory assumes that personal norms are the determinants of behavior and that these norms are activated by ascription of responsibility and awareness of consequences (Klöckner, 2013)
In general, a lot of people are motivated to do the right thing as people are motivated to see themselves as morally right, which is caused by the desire to maintain a positive self-concept (Bolderdijk et al., 2013). Considering this type of motivation may be beneficial for changing behavior. This includes pro-environmentally behavior, which can make people feel good about themselves (Steg, 2016). Hence, people can be intrinsically motivated in acting pro-environmentally, since pleasure is derived from doing the right thing and by benefiting the environment (Pelletier, Tuson, Green-Demers, Noëls & Beaton, 1998). So, pro-environmental behavior is stimulated by the feelings of moral obligation, also referred as personal norms (Steg, 2016). According to Stern (2000) personal moral norms are the main basis for people’s general tendency to act pro-environmentally.
limit. Moreover, keeping within the speed limit is not a very fun or pleasurable behaviour in itself.
Positive moral emotions are elicited when acting upon personal norms, such as feeling proud (Steg et al., 2014). Not acting in line with personal norms on the other hand leads to negative emotions, such as guilt. Bolderdijk et al. (2013) also argue that biospheric appeals allow people to perceive compliance to the appeal as a morally good conduct since the selfless and societal aspects of the pro-environmental behaviour are highlighted. This resulted in more positive affect. Conversely, economic appeals may stress the egoistic aspects of the behaviour which reduces the ability of perceiving compliance as selfless and morally good.
Worthwhile to mention is that it has been shown that personal norms only have an indirect effect on behavior, fully mediated by intentions when these are included in the model (Klöckner, 2013). A strong intention is an indicator for one’s willingness to invest physical and psychological effort in performing the behavior (Bamberg et al., 2007). In this study the focus will also be on the intention with regard to speeding behavior. Furthermore, the present study uses messages (in line with the study of Bolderdijk et al. (2013)) to make altruistic, biospheric and egoistic values more salient. The following hypotheses are proposed:
Hypothesis 1: Seeing a moral message (altruistic or biospheric) will result in a stronger intention to reduce speeding than seeing an egoistic message.
Hypothesis 2: The personal norm towards speeding mediates the relation between seeing a message (altruistic, biospheric and egoistic) and the intention to reduce speeding.
Hypothesis 2a: Seeing a moral message (altruistic or biospheric) will lead to a stronger personal norm towards speeding than seeing an egoistic message.
Hypothesis 2b: The stronger the personal norms towards speeding, the stronger the intention to reduce speeding.
Environmental self-identity as a mediator between altruistic and biospheric values and the personal norm towards speeding
Environmental self-identity can be described as the extent to which we see ourselves as a pro-environmental person (Van der Werff et al., 2013). Since central values are part of one’s self-definition, they contribute to one’s sense of identity (Verplanken & Holland, 2002). Values influences one’s environmental self-identity and are important in encouraging pro-environmental actions (Steg et al., 2014). Conforming to the value-identity-personal norm model, environmental behavior is influenced by feelings of moral obligation (i.e. personal norms) to engage in environmental behavior (Van der Werff & Steg, 2016). The personal norm is influenced by environmental self-identity, which is in turn influenced by biospheric values. Hence, stronger biospheric values lead to a stronger environmental self-identity.
In the study of Van der Werff et al. (2013) for example, environmental self-identity was strengthened which resulted in a stronger moral obligation to act in line with this identity. The influence of environmental self-identity on pro-environmental intentions was mediated by personal norms, since the personal norm to behave pro-environmentally was also strengthened. Environmental self-identity namely elicits feelings of a moral obligation (Van der Werff et al., 2013). According to Gatersleben, Murtagh and Abrahamse (2014), identities and personal norms could become more important when it concerns behaviors in which one feels relatively free to act. In case with speeding behavior there is a limit which should be followed, however, since these limits are often violated it can be seen that it is one’s own choice and one may feel unconstrained.
Since acting pro-environmentally is more likely when people strongly endorse biospheric and altruistic values because it will benefit the environment or others (Steg, 2016), the present study expects that besides biospheric values also altruistic values will influence environmental self-identity in a positive way.
Hypothesis 3: Environmental self-identity mediates the relation between seeing a moral message (altruistic or biospheric) and a stronger personal norm towards speeding
Injunctive norm as a moderator between values and the personal norm towards speeding
norms. Descriptive norms involve perceptions of which behaviors are typically performed, whereas injunctive norms refer to which behaviors are typically approved or disapproved (i.e. the moral rules of society) (Cialdini, 2003). An example of the influence of social norms can be found in the study of Goldstein, Cialdini and Griskevicius (2008). Participants were more likely to follow the descriptive norm of individuals with whom they shared the same setting in the context of reusing towels. This could be applied to the context of speeding since highlighting the behavior of other motorists might influence one’s driving speed.
Forward (2009) looked at the role of descriptive norms in predicting drivers’ intention to violate and suggest to talk about those who obey the rules and thereby making them the norm. Hence, the focus would not be on accidents and violations but instead on making non-violations more normal and this would mainly be important in influencing young drivers. Another study of Cristea and Delhomme (2012) explored, among others, the role of social norms on behavioural intentions regarding speed limits and time headway. Results showed that social norms were important, with the injunctive norm to be stronger than the descriptive norm. Moreover, the influence of social norms was stronger with respect to the motivation of reducing pollution emissions.
When most people are showing the undesired behavior, research suggest a public message should focus on the injunctive norm instead of the descriptive norm (Cialdini, 2003). Since this study aims to reduce speeding behaviour it concerns an undesired behaviour and therefore injunctive norms will be applied. Previous research found that the relation between injunctive social norms and behavioural intention is mediated by personal norms (Doran & Larsen, 2016). It can be seen that expectations of important references influence behavioural intentions indirectly through their influence on personal norms. In the present study the approval of the reference group will be highlighted (i.e. other motorists that drive on the road) which makes the injunctive norm towards speeding more salient. Based on the above, we believe that these salient injunctive norms will moderate the relationship between values and personal norms:
Conceptual model
Based on the proposed hypotheses, the following conceptual model can be drawn.
Fig. 1: Conceptual model
This paper aims to contribute to the literature by providing more insights in the effect of moral considerations in the context of speeding. It becomes clear that different values often conflict which each other (i.e. egoistic values vs. altruistic and biospheric values) and there are various ways to make these values more salient in order to influence intentions and behavior. However, empirical evidence of the effectiveness of making these values more salient is often missing. Biospheric values for example are positively associated with pro-environmental behavior such as eco-driving but the effectiveness on one’s intention to reduce speeding cannot be confirmed yet. This paper also adds to our understanding via which underlying mechanisms one’s intention to following the speed limit is influenced as one’s environmental self-identity might play a role but needs to be tested first. Furthermore, current research points to the effectiveness of using injunctive norms in communicating a certain message. However more empirical evidence is needed in order to examine this in relation to value-based messages and in addition, apply this into practical implications for policy makers aiming to reduce speeding behavior. Knowledge from this study could be used in designing interventions and thereby improving road safety and reducing the negative impact of speeding for the environment.
Values - Biospheric - Altruistic - Egoistic Intention to reduce speeding Personal norm towards speeding Injunctive norm towards speeding Salient vs Not salient
METHODOLOGY
The study consisted of a self-report questionnaire with the main goal of testing the mediator effects of personal norm towards speeding on speeding intention and environmental self-identity on the personal norm towards speeding. Furthermore, the goal was also to test the moderator effect of the injunctive norm towards speeding on the personal norm towards speeding. In addition, this study tested the effectiveness of value-based messages (biospheric, altruistic and egoistic appeals) on changing personal norms and intentions to reduce speeding. In the present study a particular action was promoted by means of showing a message on a poster, namely ‘’follow the speed limit’’. This action involves on the one hand personal benefits (avoiding a fine) and on the other hand, there are societal benefits (contributing to the environment and improving road safety).
Participants
Only participants in possession of a driver’s license were invited to participate in this study. The survey was distributed via social media, using convenience sampling and snowball sampling. Convenience sampling was used since this technique is the least time consuming and it makes the participants easily accessible (Malhotra, 2009). With snowball sampling a larger amount of participants can be reached. Participants were randomly assigned to a condition. Based on the standard conditions of a significance level of p = .05, a medium effect size of f = .25, and desired power of .80, the sample size for 6 groups should be 35 per group, which means the total sample size should be 210 (Cohen, 1988). To reach a higher number of respondents, the survey is translated into Dutch. Hence, participants had the option to choose between the English and Dutch version.
Design
speeding behavior was personal norm towards speeding. Furthermore, the mediator variable between the biospheric and altruistic message and personal norm towards speeding was environmental self-identity. The moderator variable between the three messages and personal norm towards speeding was the injunctive norm towards speeding. Intention to speeding behavior was the dependent variable.
Materials
Values. The value scale developed by De Groot & Steg (2008) was used to measure
egoistic, altruistic, and biospheric value orientations. Participants rated the importance of these values as a guiding principle in life on a 7-point Likert scale ranging from 1 (not at all important) to 7 (very important). Egoistic values were measured with five items: social power, wealth, authority, influential, and ambitious. Altruistic values were measured with four items: equality, a world at peace, social justice, and helpful. Biospheric values were measured with four items: preventing pollution, respecting the earth, unity with nature, and protecting the environment.
Manipulation of value saliency. Participants were shown one of the six possible posters,
of either a biospheric message, an altruistic message, an egoistic message or a combination of these three messages with an injunctive norm. In the biospheric condition (n = 78), the poster included the following message: ‘Follow the speed limit of 30 km per hour and contribute to a better environment. In the altruistic condition (n = 78), the poster included the following message: ‘Follow the speed limit of 30 km per hour and contribute to a safe neighbourhood’’. In the egoistic condition (n = 76), the poster included the following message: ‘’Follow the speed limit of 30 km per hour and avoid a fine’’. When the injunctive norm was included, the message started with the following text: ‘’Motorists that drive here believe you should drive a max. of 30 km per hour … for a better environment/ for a safe neighbourhood/ to avoid fines. See Appendix A for all posters with the messages.
Manipulation checks. To examine the extent to which each of the appeals made the
corresponding value more salient, participants where asked about the extent to which reading the particular message made them think of the corresponding value. Hereby, the scale of Goldstein et al. (2008) was used and adjusted to the posters in the present study. The following question was used: ‘’To what extent would reading the poster make you focus on…the quality of nature and the environment/ the welfare of other human beings/ yourself? (1 = not at all to 7 = to a very great extent).
Three items were used: ‘’Most people around me believe that I should follow the speed limit’’; ‘’Other people expect me to keep within the speed limit’’; ‘Motorists approve of those who follow the speed limit’’ (1 = strongly disagree to 7 = strongly agree).
Environmental self-identity. Based on the scale of Van der Werff et all. (2013), the
following three items were used: ‘’Acting environmentally friendly is an important part of who I am’’; ‘’I am the type of person who acts environmentally friendly’’; ‘’I see myself as an environmentally friendly person. These items were rated on a 7-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree).
Personal norm towards speeding. The personal norm focused on the dependent variable
in the study, namely speeding behaviour. Items were adopted from the scale of Van der Werff et all. (2013) and adjusted to the current context. The following three items were used: ‘’I feel morally obligated to follow the maximum speed limit as much as possible’’; ‘’I would feel guilty if I did not follow the maximum speed limit’’; I would feel good if I would follow the maximum speed limit’’ (1 = strongly disagree to 7 = strongly agree).
Intention to speeding behavior. Intention to speeding behavior was used as a proxy for
speeding behavior. As Ajzen (1991) argued, intentions capture the motivational factors that influence a behavior and the stronger the intention to engage in the behavior, the more likely its performance. Intention must be assessed in relation to the particular behavior of interest (Ajzen, 1991). Hence, the following three items were used: ‘’I intend to keep within the speed limit in a 30 km/h zone as much as possible’’; ‘’I want to keep within the speed limit in a 30 km/h zone’’; ‘’In the near future I will keep within the speed limit in a 30 km/h zone’’ (1 = strongly disagree to 7 = strongly agree).
Procedure
behavior. Finally, participants were thanked for participating where they had the option to leave their email address behind to participate in the lottery to win the voucher (see Appendix B for the complete questionnaire).
Plan of analysis
RESULTS
Descriptive statistics Demographics
A total of 236 respondents completed the full survey. Four respondents were not in possession of a driver’s license and were therefore removed from the dataset. There was one respondent that did not noted his age. However, since this respondent answered all remaining questions, this respondent was not removed from the dataset. Furthermore, the data was checked for outliers with the use of boxplots. Some outliers were detected but there were all in the predefined range of answering possibilities and were therefore not deleted. This leads to a final dataset of 232 respondents.
The demographic characteristics of the sample are shown in Table 1. It is notable that the majority of the participants are female (72% vs. 28%). There is a low distribution in age where particularly participants between 20-25 are well presented. This, in combination with the high level of education, makes it a quite homogeneous sample.
Table 1: Demographic characteristics
Gender Age Education level
Male 65 28% Mean 31.20 Primary school 0% Female 167 72% SD 12.87 Lower General Secondary Education 0,9%
Min. 18 Higher General Secondary Education 6,0% Max. 63 Pre-University Education 10,3%
Intermediate Vocational Education 13.8% University of Applied Sciences 37,1%
University 31,9%
Randomization check
Table 2: Distribution of the respondents over the experimental conditions
Experimental condition Number of respondents Percentage Biospheric message + injunctive norm not salient 38 16,4% Altruistic message + injunctive norm not salient 39 16,8% Egoistic message + injunctive norm not salient 37 15,9% Biospheric message + injunctive norm salient 40 17,2% Altruistic message + injunctive norm salient 39 16,8% Egoistic message + injunctive norm salient 39 16,8%
Reliability analysis
The theoretical constructs were all measured on a scale with multiple items (except for the manipulation check of the corresponding value), to check the internal consistency between these items, a reliability analysis was conducted. The number of items, Cronbach’s alpha, mean, standard deviation and range of the items are shown in Table 3. All of the scales showed a Cronbach’s alpha above the minimum of 0,6. For the scale measuring the manipulation check of the injunctive norm, deleting one item resulted in a higher Cronbach’s alpha (α = .855 vs. α = .819). The third item (‘’Motorists approve of those who follow the speed limit’’) was therefore removed from the scale and a new sum variable was computed only using the first two items. For the other constructs, all items were kept and new average variables were computed.
Table 3: Number of items, Cronbach’s alpha, mean, SD and range
Variable Number of
items
α Mean SD Range
Correlations
To see how the variables are related to each other, a correlation analysis has been conducted. The results are shown in Table 4. There is a positive moderate relationship between the biospheric value orientation, and the variables altruistic value orientation, personal norm towards speeding, and intention to speeding behavior. A positive strong relationship can be found between the biospheric value orientation and environmental self-identity (r = .67, p = < .01). This is as expected, since people who are more oriented to biospheric values, and thus care more about the quality of nature, also see themselves more as a pro-environmental person. For the altruistic value orientation there is a positive moderate relation with environmental self-identity and intention to speeding behavior. There is also a positive moderate relationship between environmental self-identity and personal norm towards speeding and intention to speeding behavior. The other strong positive relation can be found between personal norm towards speeding and intention to speeding behavior (r = .61, p < .01). This is also not very surprising since someone who feels a higher moral obligation to follow the speed limit, is likely to have a higher intention to keep within the speed limit. Age also had some low to moderate correlations with some of the variables indicating that older participants might be more oriented to biospheric values, less oriented to egoistic values oriented and have a higher intention to keep within the speed limit. Therefore, age will be used as a control variable in further analyses.
Table 4: Correlations Variable 1 2 3 4 5 6 1 Age 2 Biospheric value orientation .23* 3 Altruistic value orientation .10 .44* 4 Egoistic value orientation -.29** -.06 -.09 5 Environmental self-identity .33* .67** .39** -.05 6 Personal norm towards speeding .12 .37** .28** -.05 .40** 7 Intention to speeding behavior .18** .36** .39** -.08 .41** .61**
Manipulation checks
Value-based message framing
Biospheric message-framing. To make sure the biospheric message on the poster also
made the corresponding value more salient, and thus examine if the manipulation was successful, a one-way ANOVA was conducted. ‘Message type’ (0=biospheric, 1=altruistic, 2=egoistic) was entered as the factor and ‘manipulation check biospheric value’ as the dependent variable. The analysis showed there was a significant effect of the message type on the manipulation check of the biospheric value, F(2,231) = 26.20, p < .001. A post hoc test revealed that the average biospheric value saliency for people in the biospheric message condition (M=3.99, SD=1.66) is higher than the average biospheric value saliency for people in the altruistic message condition (M=2.54, SD=1.66, p < .001), and in the egoistic message condition (M=2.16, SD=1.66, p < .001). Furthermore, the post hoc test showed that there was no difference between biospheric value saliency between people in the altruistic message condition and in the egoistic message condition (p = .16). Hence, it can be concluded that the manipulation of the biospheric value was successful.
Altruistic message framing. A second one-way ANOVA was conducted to examine the
extent to which the altruistic message on the poster also made the altruistic value more salient. ‘Message type’ (0=biospheric, 1=altruistic, 2=egoistic) was entered as the factor and ‘manipulation check altruistic value’ as the dependent variable. There was a significant effect of the type of message on the manipulation check of the altruistic value, F(2,231) = 40.13, p <.001. The post hoc test made clear that the average altruistic value saliency is higher for people in the altruistic message condition (M=5.67, SD=1.37) than the average altruistic value saliency for people in the biospheric message condition (M=3.51, SD=1.84, p < .001), and in the egoistic message condition (M=3.47, SD=1.98, p < .001). No difference was found between altruistic value saliency between people in the biospheric message condition and in the egoistic message condition (p = .89). Hence, also the manipulation of the altruistic value was successful.
Egoistic message framing. Again, a one-way ANOVA was conducted to make sure the
the biospheric message condition (M=3.31, SD=1.67, p < .001). Furthermore, the average egoistic value saliency was also higher in the altruistic message condition (M=4.03, SD=1.66) compared to the biospheric message condition (p < .05). Other than expected, there was no significant difference between people in the egoistic message condition and people in the altruistic message condition (p = .10). Thus, the manipulation of the egoistic value was only partially successful since it was only more salient in the egoistic message compared to the biospheric message and not compared to the altruistic message (although the mean was still higher but this was not significant).
Manipulation check of injunctive norm saliency
An independent samples t-test was used to examine the extent to which the injunctive norm message made the injunctive norm more salient. ‘Injunctive norm (0=not salient, 1=salient) was entered as grouping variable and ‘manipulation check injunctive norm’ as the test variable. The test was not significant, t(230) = .63, p = .53. Average injunctive norm saliency of people in the condition where the injunctive norm is not salient (M=5.36, SD=1.21) does not differ from people in the condition where the injunctive norm is salient (M=5.25, SD=1.32). Hence, the manipulation was not successful in making the injunctive norm towards speeding more salient. Surprisingly, the mean of the perception of the injunctive norm towards speeding was even a bit higher is the condition where the injunctive norm was not salient.
Hypotheses testing
Influence of altruistic, biospheric and egoistic messages on intention towards speeding
Assumptions
which showed some clear deviations (see Appendix D). However, since the one-way ANOVA is considered a robust test against the normality assumption, we continue with the analysis.
To conduct an ANCOVA, the independent variable and the covariate should also be independent of each other. This was tested with a one-way ANOVA with message type as the factor and age as the dependent variable. The analysis showed no significant effect, p = .70, so the assumption holds. Moreover, there should be homogeneity of regression slopes. There was no significant interaction effect between the message condition and age, p = .08, so the assumption was met.
Analysis
To examine the effect of the three message types on the intention to keep within the speed limit, a one-way ANCOVA was conducted with message type (biospheric vs. altruistic vs. egoistic) as fixed factor and intention to speeding behavior as the dependent variable. Age was added as covariate. There was a significant effect of the message type on intention to speeding behaviour after controlling for the effect of age, F(2,230) = 4.13, p < .05. Pairwise comparisons showed that the average intention to keep within the speed limit for people in the altruistic message condition (M=5.80, SD=1.14) is higher than the average intention to keep within the speed limit for people in the biospheric message condition (M=5.22, SD=1.32, p < .05). There was no difference in average intention to keep within the speed limit between people in the altruistic message condition and people in the egoistic message condition (p = .42). Furthermore, also no difference was found between the biospheric message condition and in the egoistic message condition (p = .52). See also Table 5 for the results.
Table 5: Average intention to keep within the speed limit per message type
Message type M (SD) Biospheric 5.22 (1.32) Altruistic 5.80 (1.14) Egoistic 5.52 (1.26)
Conclusion
are all between ‘somewhat agree’ and ‘agree’ in relation to the question if people intended to keep within the speed limit.
Influence of value orientations on intention to keep within the speed limit
Since the manipulation checks were not all as successful as intended, the influence of the three values on intention to speeding behaviour was tested again using the scores of the measurement on the three values orientations instead of the actual manipulation.
Assumptions
Before the regression can be performed, assumptions have to be checked. First, the sample should be based on independent observations, which holds. Second, there should be a linear relationship between the dependent and the independent variables. To test this, a scatterplot was made with the residuals plotted against the predicted values, which showed a roughly linear relationship around zero. Third, the residuals should be normally distributed. The P-P plot was inspected and showed some deviation from the line (see Appendix E). However, since this were no drastic deviations, normality is still assumed. In addition, the absence of multicollinearity was checked using VIF values. Each value was below 4, indicating that the assumption holds.
Analysis
In order to analyse whether or not the three value orientations of people influences their intention to keep within the speed limit, a multiple regression analysis was performed with value orientation regressed on intention to speeding behavior. The overall model was significant, F(3,228) = 18.10, p < .05. The results of the regression indicated that the model explained 19,2% of the variance. There is a positive relationship between the biospheric value orientation and intention to keep within the speed limit, B = .286, t(231) = 3.47, p < .05. Furthermore, there is a positive relationship between the altruistic value orientation and intention to keep within the speed limit, B = .448, t(231) = 4.18, p < .01. The influence of an egoistic value orientation was not significant, p = .52. See also Table 6 for the results.
Table 6: influence of value orientation on intention to speeding behaviour
Variable Intention to keep within the speed limit Biospheric value orientation .286*
Altruistic value orientation .448**
(Standard error) (.107)
Egoistic value orientation -.050
(Standard error) (.078) Constant 1.715* (Standard error) (.668) Observations 232 R² .192 Adjusted R² .182 *= p < .05, **= p < .01 Conclusion
It can be concluded that both a higher biospheric and altruistic value orientation lead to a higher intention to keep within the speed limit. So, people who are more oriented towards the quality of nature and the environment and towards the welfare of other human being, are also more intended to keep within the speed limit. As expected, the egoistic value orientation was negatively related but this was not significant.
Influence of altruistic, biospheric and egoistic messages on personal norms towards speeding
Assumptions
To see if the moral obligation to follow the speed limit differs between the message conditions, a one-way ANCOVA was conducted. However, first the assumptions have to be checked. First, observations should be independent which holds with the between-subjects design. Second, the independent variable should be categorical and the dependent variable should be continuous which both holds. Third, Levene’s test for homogeneity did not reject the assumption of equal variances (p = .61), so there is equality of variances. Fourth, the population should be normally distributed. Shapiro-Wilk test showed a significant result for each condition (p < .05), so the assumption was violated. The normal Q-Q plots were inspected and also showed some deviations from the line (see Appendix D). As with the previous analysis, we continue due to the robustness of the test.
homogeneity of regression slopes since there was no significant interaction effect between the message condition and age (p = .59).
Analysis
To examine the effect of the three message types on the personal norm towards speeding, a one-way ANCOVA was performed with message type (biospheric vs. altruistic vs. egoistic) as fixed factor and personal norm towards speeding as the dependent variable. Age was added as covariate. The analysis showed that there was a significant effect of the message type on the personal norm towards speeding after controlling for the effect of age, F(2,230) = 4.70, p < .05. Pairwise comparisons revealed that the average personal norm towards speeding for people in the altruistic message condition (M=5.03, SD=1.31) is higher than the average personal norm towards speeding for people in the biospheric message condition (M=4.34, SD=1.39, p < .05). There was no difference between people in the altruistic message condition and people in the egoistic message condition (p = .21). Furthermore, also no difference was found between the average personal norm towards speeding in the biospheric message condition and in the egoistic message condition (p = .68). See also Table 7 for the results.
Table 7: Average personal norm towards speeding per message type
Message type M (SD) Biospheric 4.34 (1.39) Altruistic 5.03 (1.31) Egoistic 4.63 (1.49)
Conclusion
Influence of personal norms towards speeding on intention towards speeding
Assumptions
Before we continue with a linear regression analysis, assumptions have to be checked. First, the sample should be based on independent observations, which holds with the current research design. Second, there should be a linear relationship between the dependent and independent variable. Therefore, a scatterplot was made with the residuals plotted against the predicted value (see Appendix E). It appears that the relationship is roughly linear around zero. Third, the residuals should be normally distributed which was checked by inspecting the P-P plot (see Appendix E). The dots on the P-P plot are not all on a straight line. However, since there were no drastic deviations, normality is still assumed
Analysis
To examine whether or not a higher personal norm towards speeding leads to a higher intention to keep within the speed limit, a regression analysis of personal norm towards speeding on intention to speeding behavior was performed. The results of the regression, R² = .37, F(1,231) = 136.71, p < .001, reveal a significant effect. There is a positive relationship between the personal norm towards speeding and intention to speeding behaviour, B = .54, t(231) = 11.69, p = .00.
Conclusion
Hypothesis 2b can be confirmed, the stronger the personal norms towards speeding, the stronger the intention to reduce speeding. A higher moral obligation to follow the speed limit leads to a higher intention to keep within the speed limit (i.e. less speeding).
Mediating role of personal norm towards speeding
Although the hypotheses 2a cannot be confirmed, we still can look whether the effect of seeing a particular message on the intention to keep within the speed limit is mediated by one’s moral obligation to follow the speed limit. Since there was only a significant difference between the altruistic and biospheric message, only these two conditions will be included in the analysis.
Analysis
conducted using the PROCESS macro (Model 4) developed by Hayes (2013). Intention to speeding behavior was entered as the dependent variable, message type (biospheric vs. altruistic) as the independent variable, and personal norm towards speeding as the proposed mediator. Results show that message type was a significant predictor of intention to speeding behaviour (path c), B = .56, t(229) = 2.84, p = .005. Message type also significantly predicts the personal norm towards speeding (path a), B = .71, t(229) = 3.24, p = .001. Personal norm towards speeding also predicts intention to speeding behaviour (path b), B = .53, t(228) = 8.40,
p < .001. When personal norms are included in the model, message type does not predict
intention to speeding behaviour (path c’), B = .19, t(228) = 1.15, p = .25. Moreover, the bias-corrected 95% confidence interval (1000 bootstrap samples) for the indirect effect indicated that mediation occurred, indirect = .37, 95% CI[0.15, 0.64]. See Figure 2 for a visual presentation of the mediated relationship.
Conclusion
Hypothesis 2 can only be confirmed for the difference between the biospheric and altruistic message. It can be concluded that there was a full mediation effect and the relation between message type and intention to speeding behaviour is mediated by personal norm towards speeding. The personal norm towards speeding mediates the difference in intention to keep within the speed limit between the biospheric and altruistic message condition. Hence, the increase in one’s intention to keep within the speed limit caused by the altruistic message (as compared to the biospheric message) stemmed from the moral obligation to follow the speed limit which participants felt while seeing the altruistic message.
Figure 2: Mediated relationship between message type and intention to speeding behaviour with personal norm towards speeding as the mediator
Mediating role of environmental self-identity
Analysis
To test whether environmental self-identity acts as a mediator between the altruistic and biospheric message condition and personal norm towards speeding, a mediation analysis was conducted using the PROCESS macro (Model 4) developed by Hayes (2013). Personal norm towards speeding was entered as the dependent variable, message type (biospheric vs. altruistic) as the independent variable, and environmental self-identity as the proposed mediator. Results show that message type was a significant predictor of personal norm towards speeding (path c), B = .71, t(229) = 3.15, p = .002. Message type does not significantly predict environmental self-identity (path a), B = .34, t(229) = 1.73, p = .08. Environmental self-self-identity does predict the personal norm towards speeding (path b), B = .44, t(228) = 6.38, p < . 001. When environmental self-identity is included in the model, message type still predicts the personal norm towards speeding (path c’), B = .55, t(228) = 2.67, p = .008. No mediation effect has occurred, as indicated by the bias-corrected 95% confidence interval (1000 bootstrap samples) for the indirect effect, indirect = .15, 95% CI[-0.03, 0.39]. See Figure 3 for a visual presentation.
Conclusion
It can be concluded that there was no mediation effect of environmental self-identity between message type and personal norm towards speeding and therefore hypothesis 3 cannot be confirmed. Hence, the difference in the moral obligation to follow the speed limit is not mediated by one’s environmental self-identity. There was however a significant effect of environmental self-identity on the personal norm towards speeding which means that a stronger environmental self-identity leads to a higher moral obligation to follow the speed limit.
Figure 3: Mediated relationship between message type and personal norm towards speeding with environmental self-identity as the mediator
Value orientations and mediating role of environmental self-identity
Again, the mediating role of environmental self-identity will be tested using the scores of the biospheric and altruistic value orientations instead of the actual manipulation.
Analysis
To test if environmental self-identity acts as a mediator between the biospheric and altruistic value orientation and personal norm towards speeding, a mediation analysis was conducted using the PROCESS macro (Model 4, bias-corrected, 1000 bootstrap samples) of Hayes (2013). This analysis was conducted two times, first for the biospheric value orientation and subsequently for the altruistic value orientation. Personal norm towards speeding was entered as the dependent variable, value orientation as the independent variable, environmental self-identity as the proposed mediator. Age was added as covariate.
Results for the biospheric value orientation show that biospheric values were a significant predictor of personal norm towards speeding (path c), B = .46, t(228) = 5.11, p < .001. Biospheric values also significantly predict environmental self-identity (path a), B = .76, t(228) = 12.79, p < .001. Environmental self-identity also predict the personal norm towards speeding (path b), B = .36, t(227) = 3.71, p < .001. When environmental self-identity is included in the model, biospheric values do not predict the personal norm towards speeding (path c’), B = .18, t(227) = 1.61, p = .11. Mediation has occurred, indirect = .28, 95% CI[0.11, 0.45].
Results for the altruistic value orientation show that altruistic values significantly predict the personal norm towards speeding (path c), B = .50, t(228) = 4.32, p < .001. Altruistic values are also a significant predictor of environmental self-identity (path a), B = .57, t(228) = 6.29, p < .001. As with the biospheric values, path b was also significant, B = .39, t(227) = 4.90,
p < .001. When environmental self-identity is included in the model, altruistic values still
predict the personal norm towards speeding (path c’), B = .27, t(227) = 2.30, p < .05. Mediation has occurred, indirect = .22, 95% CI[0.11, 0.38].
Conclusion
obligation to follow the speed limit caused by a higher altruistic value orientation is partially caused by a stronger environmental self-identity.
Influence of the injunctive norm towards speeding on the relationship between seeing an altruistic, biospheric and egoistic message and the personal norm towards speeding
Assumptions
Before a two-way ANCOVA was conducted to test the interaction effect, assumptions have to be checked. First, there should be sample independence, which holds with the current research design. Second, the independent variables should be categorical and the dependent variable should be continuous. Both message type (i.e. biospheric vs. altruistic vs. egoistic) and the injunctive norm (i.e. not salient vs. salient) are categorical and the personal norm towards speeding is indeed continuous. Next, there should be variance equality in each group. To test this, a Levene’s test for homogeneity is conducted which did not reject the assumption of equal variances (p = .44), so this assumption also holds. Furthermore, the population should be normally distributed. Shapiro-Wilk test for normality was conducted and only showed a normal distribution for the condition with the biospheric message in combination with the injunctive norm being salient (p = .29) and no normal distributions over the remaining experimental conditions (p < .05). This was further analyzed by inspecting the normal Q-Q plots for each experimental condition (see Appendix D) which showed some deviations. However, due to the robustness of the test, we will continue the analysis. Next, independence between the groups and the covariate was proved with a one-way ANOVA which was not significant (p = .73). Finally, there should be homogeneity of regression slopes. Since there was no significant interaction effect between the experimental conditions and age, p = .74, the assumption holds.
Analysis
one-way ANCOVA. For the injunctive norm saliency no main effect was found, F(1, 230) = .07, p = .79. The scores are illustrated in Figure 4.
Figure 4: Personal norms towards speeding per experimental condition
Conclusion
It can be concluded that hypothesis 4 cannot be confirmed. Seeing a message (biospheric, altruistic, or egoistic) with the injunctive norm being made salient does not strengthen the relationship between these values and the personal norm towards speeding. For the biospheric and altruistic message, the average moral obligation to follow the speed limit was somewhat higher when the injunctive norm was made salient but this was not significant. A logical explanation for this has to do with the manipulation of the injunctive norm, which was not significant.
Value orientation and the mediating role of personal norms towards speeding and moderation effect of the injunctive norm
Again, the mediating role of personal norms will be tested using the scores of the biospheric and altruistic value orientations (as only these two were significantly related to one’s intention to keep within the speed limit) and thereby the influence of the injunctive norm will be examined using the measurement of the injunctive norm instead of the actual manipulation.
1 2 3 4 5 6
Biospheric Altruistic Egoistic
Personal norm towards speeding
Analysis
To test if the personal norm towards speeding acts as a mediator between the biospheric and altruistic value orientation and whether this effect is moderated by the injunctive norm towards speeding, a moderated mediation analysis was conducted using the PROCESS macro (Model 7, bias-corrected, 5000 bootstrap samples) of Hayes (2013). This analysis was conducted two times, first for the biospheric value orientation and subsequently for the altruistic value orientation. Intention to speeding behaviour was entered as the dependent variable, value orientation as the independent variable, personal norm towards speeding as the proposed mediator and the injunctive norm towards speeding as the proposed moderator. Age was added as covariate.
Results for the biospheric value orientation show that biospheric values were a significant predictor of the personal norm towards speeding (path a), B = .32, t(226) = 3.02, p = .003. The moderator (i.e. the injunctive norm) was also a significant predictor of personal norms, B = .37, t(226) = 3.62, p < .001. However, the interaction term between biospheric values and the injunctive norm was not significant, p = .91. The personal norm towards speeding was a significant predictor of intention to speeding behaviour (path b), B = .49, t(227) = 8.09, p < .001. Further, when personal norms are included in the model, biospheric values still predicts intention to speeding behaviour (path c’), B = .19, t(227) = 2.56, p = .01. Nevertheless, since there was no significant interaction effect, there was no moderated mediation, 95% CI[-0.08, 0.07].
Results for the altruistic value orientation show the same results. Altruistic values significantly predict the personal norm towards speeding (path a), B = .36, t(226) = 3.17, p = .002. Again, the moderator was a significant predictor of the personal norm towards speeding, B = .40, t(226) = 4.23, p < .001. The interaction effect could however not be demonstrated, p = .94. As with the biospheric values, path b was also significant, B = .48, t(227) = 7.97, p < .001. Also, when personal norms are included in the model, altruistic values predict intention to speeding behaviour (path c’), B = .36, t(227) = 3.99, p < .001. However, there was no moderated mediation, 95% CI[-0.08, 0.09].
Conclusion
DISCUSSION
The aim of this research was to examine how moral considerations can be used in motivating people to follow the speed limit. These moral considerations include safety for others and a concern for the environment. Although road safety is considered the most important reason for keeping within the speed limit (Duijm et al., 2012), it can be seen that people are acting from an egoistic perspective when it comes to speeding. Research has made it clear that many people are motivated to see themselves as morally right, also referred as personal norms (Steg, 2016). Both the personal norm and pro-environmental behavior can be influenced by values as a result of making them more salient (De Groot & Thørgersen, 2019). Therefore, the present study proposed that making biospheric and altruistic values more salient by means of showing a particular message could reduce speeding intention, because these values will activate one’s personal norm towards speeding. Moreover, it was argued that seeing a message when the injunctive norm is made salient would strengthen this relationship. Also, it was expected that seeing a moral message would resulted in a stronger environmental self-identity, which in turn would lead to stronger feelings of a moral obligation. Based on the results, it can be concluded that an altruistic message was indeed effective but no support was found for the biospheric message. However, both a biospheric and altruistic value orientation do have a positive effect. The proposed hypotheses will be discussed below.
to keep within the speed limit with again the altruistic value orientation being the most influential. Thus, this indeed proves the relevance of taking altruistic and biospheric values into in the context of speeding.
Further, it was expected that seeing a moral message (biospheric or altruistic) will lead to a stronger personal norm towards speeding than seeing an egoistic message. However, results showed no support for this. The altruistic message did lead to the highest moral obligation to follow the speed limit but other than expected the biospheric message seemed even less effective in making the personal norm to follow the speed limit stronger than the egoistic message. It could be that participants did not felt responsible for contributing to a better environment by following the speed limit, literature on personal norms namely stated that the activation of the personal norm also depends on the saliency of personal responsibility (Schwartz, 1973). Another explanation for this could be that the manipulations of the three values were not successful enough in making the corresponding value salient, although participants reported they focused more on the quality of nature and the environment in the biospheric message condition and more on the welfare of other human being in the altruistic message condition. In the egoistic message condition participants focused more on themselves but only a significant difference with the biospheric message condition was found.
It was also expected that a stronger personal norm towards speeding resulted in a stronger intention to reduce speeding, which was confirmed by the results. This is in line with the literature which states that pro-environmental behavior is stimulated by a feeling of a moral obligation (Steg, 2016). Also, this is in support of the value-belief-norm theory and norm-activation theory that focus on the central role of personal norms in explaining intentions. Moreover, evidence was found for the mediating role of personal norms between seeing a message which made biospheric and altruistic values more salient and the intention to keep within the speed limit.
