EXAMINING THE RELATIONSHIP BETWEEN VALUE ORIENTATIONS AND SUBJECTIVE QUALITY OF LIFE
IN AN E-BIKE CONTEXT
By
Benjamin van Marle
University of Groningen
Faculty of Economics and Business
MSc Marketing June 2019
Rosmolenstraat 18 8061GV Hasselt
0630612227
benjamin.van.marle@student.rug.nl Student number: S3272206
EXAMINING THE RELATIONSHIP BETWEEN VALUE ORIENTATIONS AND SUBJECTIVE QUALITY OF LIFE IN AN E-BIKE CONTEXT
ABSTRACT
Present study integrates the concept of altruistic, biospheric and egoistic value orientations and subjective quality of life (QoL) and appraises the relationship between values and subjective QoL in an e-bike context. Also, this study examined to what extend seven QoL- dimensions (factored out of the initial 22 indicators explaining overall subjective QoL) mediate the relationship between value orientations and overall subjective QoL. Results of a questionnaire among 133 respondents could not establish a relationship between values and overall subjective QoL mediated through the seven QoL-dimensions. Further, no relationships were found between value orientations and the QoL-dimensions, except a relationship
between altruistic and egoistic values and the QoL-dimension Self-Direction, in which the egoistic value had the highest explanatory power. So, the stronger one’s altruistic and particularly one’s egoistic values, the stronger an individual will value the QoL-dimension Self-Direction in relation to e-bike adoption. This implicates that the stronger one’s altruistic or egoistic values, the stronger an individual values the reliance on and gratification from one's independent capacities for decision-making, creativity, and action. Results of this study can help marketers to develop marketing activities that align with the characteristics of the QoL-dimension Self-Direction among customers with stronger altruistic or egoistic value orientations to make these marketing activities more effective.
Key words: Values, bioshperic, altruistic and egoistic value orientations, subjective quality of life (QoL), QoL-dimensions
Research theme: E-bikes
Seminar supervisor: Judith de Groot
EXAMINING THE RELATIONSHIP BETWEEN VALUE ORIENTATIONS AND SUBJECTIVE QUALITY OF LIFE IN AN E-BIKE CONTEXT
1. INTRODUCTION
The lack of physical inactivity is a major health problem (World Health Organization, 2010).
Despite warnings about the potentially negative health consequences of a sedentary lifestyle, a large proportion of adults is physically inactive (Simons, Van Es & Hendriksen, 2009).
Moderate-intensity physical activity can provide important health benefits (Rjeski, Brawley &
Shumaker, 1996). One way to overcome the inactivity problem is active transport, which consists of walking and cycling. Active transport offers a great potential to improve health and has several other positive side-effects. Active transport is non-polluting, poses little danger to others and is socially inclusive (Woodcock, Banister, & Edwards, 2007).
This paper is mainly focused on the (e-)cycling part of active transport. Cycling is associated with health, environment, efficiency and life quality benefits (European Commission, 2015).
In the Netherlands solely, there are around 22,8 million bicycles located (Raivereniging, 2018). Lately, electric assisted bicycles (e-bikes) have become increasingly popular and represents one of the fastest growing segments of the transport market (Fishman & Cherry, 2015). Almost 300.000 e-bikes were sold in the Netherlands in 2016, which was 31% of total bike sales (Fietsplatform, 2016). Currently, there are nearly 1,9 million e-bikes in The
Netherlands (Raivereniging, 2018). Encouraging cycling with conventional bikes and e-bikes gives rise to achieve a number of goals, such as: reducing local pollution and greenhouse gas emissions, increasing physical activity and thereby addressing obesity and other health-related issues (Cairns et al., 2017). Although e-bikes are less environmentally friendly and require less physical activity than conventional bikes when riding the same journeys, the differences are small compared with using other forms of motorized transport, such as the car (Cairns et al., 2017). Moreover, the activity required to ride an e-bike is still sufficient to count as at least ‘moderate intensity’ physical activity (see Gojanovic, Welker, Iglesias, Daucourt, &
Gremion, 2011; Simons et al., 2008).
Having an e-bike might be promising for increased bicycle use. It has an integrated battery augmenting the pedal-power of the rider and is in Europe legally classified as a bicycle if it fulfils certain criteria, such as a maximum speed limit of 25 kilometers per hour. In Europe e- bikes are partly human powered, because it provides electrical assistance only when the rider is pedaling (European Commission, 2009). An e-bike weights noticeably heavier than a regular bike, because of its battery and it is harder to pedal when the battery is switched off or
flat (Fyhri & Fearnley, 2015). Also, Hendriksen (2008) outlined that using an e-bike instead of a car reduces the use of fossil fuels, motorized traffic and environmental impact. E-bike users have reported advantages such as higher speed with less effort, less sweat while riding, reduced travel time and easier to climb hills compared to conventional bicycles (see Berntsen, Malnes & Langaker, 2017; Ling, Cherry, MacArthur, & Weinert, 2017). E-bike usage can increase trip distance, reduce time to travel and diminish rider effort compared with an ordinary bike (Fishman & Cherry, 2016). Doing so may result in more bike trips and longer bike trips, and increase the diversity of people bicycling, including people with a disability or chronic injury (MacArthur, Dill & Person, 2014).
According to Berntsen et al. (2017), riding an e-bike is faster and less intensive than riding a regular bike, which makes it suited for busy modern lives. It will get someone quicker to work and a person might not need a shower afterwards. On the other hand, less time spent cycling at a lower intensity is not ideal as most people are inactive (Berntsen et al., 2017). Riding an e-bike leads to reduced activity and intensity over the same distance covered compared to riding an regular bike, but it offers a physically active alternative to the largely sedentary behavior associated with motorized commuting (Bourne et al., 2017). Consistent evidence indicates that time spent travelling by car has a negative impact on health and subjective wellbeing (Ding, Gebel, Phongsavan, Bauman, & Merom, 2014). This is important to acknowledge, since most individuals travel to work by car (Bassett, Pucher, Buehler, Thompson, & Crouter 2008). Previous research concludes that e-bike users ride longer distances than those riding a conventional bike (see Langford, Cherry, Yoon, Worley &
Smith, 2013; Fyhri & Fearnley, 2015; Cairns et al., 2017), but this effect seems primarily be explained for individuals that travel bigger distances as a commuter traveling from home to work and vice versa (see Edge, Dean, Cuomo, & Keshav, 2018; Kroesen, 2017;
Kennisinstituut voor Mobiliteitsbeleid, 2017; Fyhri & Fearnley, 2015). Therefore, e-bike usage and distance travelled seem to increase when it is used for commuting purposes.
There might be a decrease in physical activity for other user groups since riding an e-bike requires less physical energy (Gojanovic et al., 2011). This might also negatively affect one’s mental health (Raglin, 1990). Consequently, one can question whether all e-bike users
enhance their physical and subjective well-being after adopting an e-bike. Substituting an ordinary bike for an e-bike might have implications for one’s subjective quality of life (Brown et al., 2003). Subjective quality of life (QoL) refers to individual’s cognitive and affective evaluations of one’s life (Diener, 2000). Individuals that do not compensate their initial
cycling duration and intensity with more and longer e-cycling trips might perceive a backfire effect in terms of actual physical activity and, consequently one’s subjective QoL. Scholars argue that the lack of compensation of more frequent and longer rides to accrue comparable health benefits for e-bike owners might have health and well-being implications and can affect one’s subjective QoL (see Rjeski, Brawley & Shumaker, 1996; Brown et al., 2003). But to what extent does e-bike adoption actually affect subjective QoL? It is crucial to understand how individuals, with different life values affecting beliefs and behavior, evaluate their subjective QoL after a major change in bicycle mobility; exchanging a conventional bike for an e-bike.
Health related issues seem to be an important subject within the e-bike area (Fishman &
Cherry, 2016). However, the perception of one’s healthiness is just one out of 22 indicators that comprise overall subjective QoL (Diener, 2000). This implies that the health domain explains a small part of one’s subjective QoL after adopting an e-bike. Thus, a lot of attention is given to the health aspect of e-bike usage (Fishman & Cherry, 2016), but it might be a small indicator in explaining overall subjective QoL after exchanging an ordinary bike for an e-bike. Also, negative changes in one QoL-aspect may be compensated by positive changes in other QoL-aspects (Sarch, 2012). Other benefits of e-bikes, such as increased speed, might outweigh a potential negative health evaluation in explaining overall subjective QoL.
Moreover, perceptions of one’s own health can be very different from one’s objective health evaluations (Arts, Frambach & Bijmolt, 2011). Even though activity levels might actually decrease after e-bike uptake, an individual might believe that he is still exercising and being outdoors frequently, which might give the perception that one is doing fine in terms of healthiness. Subjective QoL is about the affective evaluations of one’s cycling behavior and, therefore, does not measure actual cycling behavior. Individuals might not perceive a backfire effect even though it is likely that this occurs in their normal life, since a substantial part of e- bike owners seem to have troubles compensating their initial cycling activity after e-bike uptake (Gojanovic et al., 2011). Therefore, it is important to examine which QoL-aspects might compensate for a potential negative health evaluation and which QoL-indicators are important for bike users after adoption an e-bike.
In environmental literature, which covers sustainable transport (An, Chen, Xin, Lin & Wei, 2013), it is argued that three different value orientations are relevant for understanding environmental beliefs and intentions: egoistic, altruistic, and biospheric (De Groot & Steg, 2008).Value orientations are overarching principles of one’s life that guide beliefs and
behavior (Schwartz, 1992). Values are universal and culturally shared and give rise to engage in corresponding beliefs and behavior. All 22 indicators that comprise overall subjective QoL are based on extensive literature on needs and values as well (see Poortinga, Steg & Vlek, 2004; Gatersleben, 2000; Vlek, Skolnik & Gatersleben, 1998). A major difference between value orientations and subjective QoL is that the latter is more specific towards a particular aspect of one’s life, such as e-bike uptake. In contrary, value orientations involve guiding principles of one’s life (Rokeach, 1973).
The main aim of this study is to examine whether altruistic, biospheric and egoistic value orientations are related with the seven QoL-dimensions and overall subjective QoL related to an e-bike domain. This study is particularly interested in the general relation between these theorized constructs and their viability in an e-bike environment. Thereafter, the central question of this research paper is: How do biospheric, altruistic and egoistic values explain subjective QoL when people switch from a regular bike to an e-bike?
This study extends our current knowledge about the concept related to subjective QoL. Even though previous studies have focused on the effects of sustainable transport on subjective QoL (see De Groot & Steg, 2006; Steg & Gifford, 2005), yet there is no research conducted on anticipated changes in subjective QoL among individuals substituting a conventional bike for an e-bike. According to Diener and Suh (1997) value orientations seem to be an important starting point in subjective QoL research, since universal values suggest a systematic way of selecting indicators that reflect subjective QoL. This implicates that value orientations serve as a conceptual vehicle for overall subjective QoL and the QoL-dimensions derived from this construct. Therefore, present study integrates the concept of altruistic, biospheric and egoistic value orientations and subjective quality of life (QoL) and appraises the relationship between values and QoL in an e-bike domain. Future studies can build further on developing theories related to values and subjective QoL, since investigating the relationship between these constructs is still an unexplored area up to this moment.
Present study extends current literature by connecting the role of value orientations with the seven QoL-factors, extracted from the initial 22 QoL-indicators discovered by Poortinga et al.
(2004), in order to explain total changes in subjective QoL on cycling mobility.
Understanding QoL has tremendous potential implications, because improving subjective QoL is a major policy and lifestyle goal (Schuessler and Fisher, 1985). Investigating changes in subjective QoL is important for public health policy, bearing in mind that a substantial part of individuals does not sufficiently engage in physical activity (Simons et al., 2009). Scholars
might fear that e-bike uptake results in downward activity levels and, consequently, a decrease in subjective QoL (see Rjeski, Brawley & Shumaker, 1996; Brown et al., 2003).
Therefore, present study examines to what extent people’s overall subjective QoL actually will be affected, since individuals will use compensating strategies to adhere a sufficient overall subjective QoL (Sarch, 2012).
2. LITERATURE REVIEW
2.1. The importance of physical activity, active transport and the implications for quality of life. The lack of physical activity is a major health problem, whereas participating in moderate intensity physical activity on a regular basis can provide important health benefits (Bourne et al., 2018). Despite warnings about the potentially negative health consequences of a sedentary lifestyle, a large proportion of adults is physically inactive (Simons et al., 2009). According to the World Health Organization (2019) adults aged between 18 and 64 should do at least 150 minutes of moderate intensity physical activity throughout the week or do at least 75 minutes of vigorous intensity physical activity throughout the week or an equivalent combination of moderate and vigorous intensity activity. Moderate intensity physical activity is an equivalent to a brisk walk and noticeably accelerating the heart rate (Haskell et al., 2007). Adults can meet the moderate intensity activity recommendation by participating in a variety of activities throughout the day, such as gardening, walking, cycling and other transportation activities (Jones et al., 1998). Vigorous intensity activity is exemplified by jogging and causes rapid breathing and a substantial increase in hart rate (Haskell et al., 2007). Moderate and vigorous intensity activities are complementary to each other in the production of health benefits. Both forms of intensity activities can be combined to meet the general recommendation. The sum of those activities is based on the amount of activity (intensity and duration) performed during the week (Simons et al., 2009).
Increasing physical activity has several physical and subjective well-being advantages.
Blumenthal, Babyak and Moore (1999), for example, published a study about the effects of physical activity among three different groups. It was found that exercising was just as effective at treating depression as antidepressant medication or a combination of both. Yet exercise is a lot less expensive and has usually no negative side-effects, except soreness, injuries and psychological effort. Furthermore, the study described multiple benefits of physical activity for health and subjective well-being (Blumenthal, Babyak, & Moore, 1999), including a decrease of anxiety and stress, protections from dying in general (and from dying of heart disease or cancer, in particular), controlling weight issues and increases in subjective
quality of life (QoL), where subjective QoL is concerned with individuals’ subjective experience of their lives (Diener & Suh, 1997).
Next to the direct physical health related benefits of physical activity, moderate-intensity physical activity can be considered as a viable means for reducing stress, improving self- perception, life satisfaction, mood and QoL (Raglin, 1990). Relative increases in physical fitness and habitual physical activity are associated with greater emotional well-being and, therefore, might affect subjective QoL (Galper, Trivedi, Barlow, Dunn & Kampert, 2006).
The main results of an intervention study of De Geus, Van Hoof, Aerts and Meeusen (2007) indicate that cycling to work at a moderate intensity has a positive influence on mental health and subjective QoL. Accordingly, participating in moderate physical activity by using an (e)- bike might enhance one’s overall subjective QoL. However, e-bike users spent less time cycling at a lower intensity when they cover the same distances when comparing with
conventional bike riders (Langford et al., 2013). It is not optimal when individuals spend less time cycling at a lower intensity, since most people are already inactive (Berntsen et al., 2017). Most adults are insufficiently active below the recommended 150 minutes physical activity of moderate intensity or 75 minutes of vigorous intensity per week (World Health Organization, 2010). This implies that it is essential to determine whether cycling with an e- bike can be health-enhancing in terms of subjective QoL.
2.2. Subjective quality of life. Quality of life refers to well-being, conceptualized as the objective conditions of an individual’s living or as the person’s experience of life (Diener et al., 1999). Subjective well-being or quality of life (QoL) is a multi-dimensional construct, and may be defined as the extent to which important values and needs of people are fulfilled (Diener, 2000). QoL is related but not equivalent to other constructs used in this type of research, such as self-esteem, adjustment and happiness (Edwards & Patrick, 2003). Present study focuses solely on subjective well-being or subjective QoL, which refers to individual’s cognitive and affective evaluations of one’s life (Diener, 2000). Thus, subjective QoL is about the way a person perceives his subjective living conditions.
2.2.1. Objective and subjective QoL and their implications. QoL in an e-bike context can be measured subjectively or objectively (Plaizier et al., 2017). The latter reflects people’s
objective circumstances in a given cultural or geographic unit. The hallmark of objective QoL is that it is based on objective, quantitative statistics rather than on one’s subjective
perceptions of their social environment (Diener & Suh, 1997). Plaizier and colleagues (2017) measured objective QoL after e-bike adoption by GPS-tracking. Subjective QoL was
measured in a transport context (see De Groot & Steg, 2006), but is not examined in an e-bike context yet. Therefore, present study focuses on subjective QoL in an e-bike domain, because objective measures in QoL fails to capture individuals’ anticipated changes in subjective QoL (Farquhar, 1995). Furthermore, Naes (1999) argued that subjective QoL correspondents more closely with values of individuals than an objective approach of measuring QoL. This is important to acknowledge, because examining underlying values in an e-bike context is an important aspect of present study.
A study of Sarch (2012) considered the complementary effects of subjective QoL-measures: a reduction in one QoL-indicator will result in the diminishment or enlargement of the effects of one other, and vice versa. Therefore, an individual might use a compensating strategy of one QoL-aspect with other aspects to keep a high overall subjective QoL.It is important to consider subjective QoL-implications after e-bike adoption, because it is important to get a profound understanding of anticipated subjective QoL-effects, which in turn is crucial for the feasibility and effectiveness of future bicycle behavior. When an individual believes the adoption of an e-bike significantly reduces or improves particular domains of subjective QoL, one should examine the basis and relative importance of these expectations that affects
subjective QoL after adopting an e-bike.
Then, when we understand the general concept of subjective QoL and its relevance, it is evident to consider which indicators comprise overall subjective QoL. Based on extensive literature review of needs, values and human well-being, a list of 22 QoL-indicators (see Appendix I for an overview of these indicators) has been developed and used in various research projects on sustainable and health related matters (see Poortinga et al., 2004;
Gatersleben, 2000; Vlek, Skolnik & Gatersleben, 1998). For example, Gatersleben (2000) studied possible QoL-changes people expected when they would develop a (imaginary) sustainable household consumption pattern. This briefly shows that various QoL-indicators have been used in several studies to determine the impacts of changes in societal and
environmental conditions and consumption (Poortinga et al., 2004). Also, the extent to which people evaluate the 22 QoL-indicators might depend on individual characteristics. This is demonstrated in previous research by Gatersleben (2000) showing differences in household type and income and a study of Poortinga and colleagues (2001) on different household types affecting evaluations of the QoL-indicators. These differences in individual characteristics imply to what extent e-bike uptake affect QoL-evaluations may vary between groups, such as gender, age, and general values.
The 22 QoL-indicators are considered to be very important to people’s lives, because these QoL-values refer to important needs and values of people’s lives (Steg & Gifford, 2005).
Also, this list of QoL-indicators represent a wide range of non-overlapping dimensions that are important to consumers and therefore, important for regular bike and e-bike travelers as well (Steg & Gifford, 2005). The results of a study of De Groot and Steg (2006) focusing on egoistic, altruistic and biospheric values in relation to subjective QoL suggest that biospheric, altruistic and egoistic values affect the QoL-indicators and their importance assigned to it, and therefore one’s overall subjective QoL. According to Rokeach (1973) values are usually conceptualized as important life goals or as normative standards that serve as a guiding principle in life. Importance judgments given to the various QoL-indicators may also be taken to reflect basic human values (Poortinga et al., 2004). Consequently, importance judgments based on values may be used to examine whether a set of QoL-indicators can help to explain overall subjective QoL on bicycling mobility (Poortinga et al., 2004). Therefore, present study argues that the three value orientations, consisting of altruistic, biospheric and egoistic values, will also affect overall subjective QoL related to e-bike adoption.
2.3. Introducing general values and the linkage with subjective QoL. Values are considered to be important, because they are general and, therefore, affect many different beliefs and
behaviors (De Groot & Steg, 2006). Schwartz (1992) refers to a value as a desirable trans- situational goal varying in importance, which serves as a guiding principle in the life of a person. This illustrates that a value reflects a belief on the desirability of a certain end-state.
Steg and colleagues (2012) postulate that values influence what people attend to, what knowledge becomes cognitively most accessible, how people evaluate various aspects of a situation, how much importance people ascribe to different consequences of actions and what alternatives are being considered. Values are culturally shared and people may endorse the same values. Be that as it may, each individual is likely to prioritize various values in a different manner (Steg et al., 2012). This suggests that individuals, when they face conflicting values, base their choice on the values they deem the most important to act on. This results in different value-related choices for individuals who prioritize their values differently.
The central elements of subjective QoL are derived from the context of one’s most important values and goals (Diener & Suh, 1997). To build further on this, improved subjective QoL is most likely to be experienced when individuals work for and make progress toward personal goals that are derived from their important values. If an individual values altruism, for example, it is likely that acting in line with this value brings him a feeling of improved
subjective QoL. Just like the value orientations identified by Schwartz (1994), most QoL- indicators are considered to be important to the lives of individuals, because these QoL- indicators refer to important needs and values (Steg & Gifford, 2005). The difference between value orientations and QoL-dimensions is that the latter one is more specific toward a specific aspect in life, such as e-bike usage. Conversely, value orientations are overarching goals in life and refer to important beliefs guiding life principles.
The nature of values portrays at least two reasons to show the importance of studying values.
First of all, there is ample theoretical and empirical evidence that values play a significant role in explaining specific beliefs and behavior (Stern, 2000; De Groot & Steg, 2008). This
implicates that value orientations can be used as predictors for examining attitudes, beliefs and behavioral intentions. Second, the total number of values that people may consider appears to be relatively small. According to Rokeach (1973), values provide an economically efficient instrument for describing and explaining differences and similarities between
persons, groups, cultures and nations. In addition, values are considered to be relatively stable in time, which makes it particularly relevant to study these values (Stern, 2000).
2.3.1. Biospheric values. In environmental literature, which for example focuses on sustainable transport, it is argued that three value orientations seem to be relevant for
understanding environmental intentions and beliefs, namely: biospheric, altruistic and egoistic (De Groot & Steg, 2008). An e-bike context is considered to pertain in environmental
literature, since it is a form of sustainable transport and is mostly associated with environmental friendly behavior (An et al., 2013). A biospheric value orientation is an orientation in which people will consider costs and benefits for the ecosystem and biosphere.
Biospheric values reflect a concern with the quality of nature and the environment for its own sake, without a clear link to the welfare of other human beings (De Groot & Steg, 2008).
2.3.2. Altruistic values. An altruistic value orientation refers to an orientation in which people will focus on perceived costs and benefits for other people (De Groot & Steg, 2008).
Altruistic values differ from biospheric values, because altruistic values particularly reflect a concern with the welfare of other human beings. Hence, both altruistic and biospheric values are positively related to pro-environmental attitudes and behavior (Stern, 2000; De Groot &
Steg, 2008).
2.3.3. Egoistic values. An egoistic value orientation is a transitional goal in which people will especially consider costs and benefits for them personally (De Groot & Steg, 2006).Egoistic
values typically correlate negatively with pro-environmental attitudes and behaviors (Steg et al., 2012). People with strong self-enhancement, egoistic values seems to act only pro-
environmentally when the perceived individual benefits of such actions exceed the perceived costs and vice versa (Steg et al., 2012). Therefore, differences between egoistic, biospheric and altruistic value orientations might appear, because environmental behavior often involves a conflict between immediate individual gains and long-term collective interests (Steg et al., 2012).
De Groot and Steg (2006) examined, for example, the role of value orientations in evaluating quality of life consequences of a transport pricing policy in which overall car prices were doubled. These researchers concluded that the three value orientations were significantly correlated with expected changes in the 22 QoL-indicators. De Groot and Steg (2006) found that egoistic value orientations and changes in perceived QoL-indicators are negative
correlated, suggesting that strong egoistic value orientations are associated with expecting more negative changes in specific QoL-indicators. There were significant relationships found for the QoL aspects freedom, comfort, privacy, environmental quality, nature/biodiversity, work, education, health, safety, social relations, aesthetic beauty and change/variation.
De Groot and Steg (2006) showed in their transport policy study that biospheric and altruistic value orientations were positively correlated with expected changes in QoL-indicators,
indicating that strong biospheric and altruistic value orientations are associated with expecting more positive changes in perceived QoL-indicators. It was found that biospheric value
orientations showed significant correlations with most QoL-indicators, except
status/recognition, privacy, education and health. QoL-aspects such as nature/biodiversity, environmental quality and change/variation appeared to have strong correlations with biospheric value orientations. It was also concluded that altruistic value orientations were correlated with the aspects social justice, environmental quality, aesthetic beauty,
nature/biodiversity and education.
2.4. Similar interpretations between subjective QoL-indicators and value orientations. The 22 QoL-indicators emphasize the aspects of subjective QoL that are important to consumers (De Groot & Steg, 2006). These QoL-indicators were developed to examine future consequences of environmental conditions and policies by assessing to what extend these changes would affect different QoL-indicators.These QoL-indicators reflect the three main dimensions of sustainability; consisting of economic, social and environmental dimensions (De Groot &
Steg, 2006). These dimensions can be linked to the egoistic, altruistic and biospheric value
orientations, which seem to have strikingly similar interpretations. Egoistic value orientations might have a link with an economic dimension, biospheric oriented values seems to have a similar meaning with the environmental dimension and altruistic value orientations can be associated with the social dimension.
2.5. Importance judgments given to different QoL-indicators. There might be differences in importance ratings between the 22 QoL-indicators among individuals after describing a specific environmental-related scenario. Since values are usually conceptualized as important life goals or as normative standards that serve as a guiding principle in life (Rokeach, 1973), importance judgments given to the various QoL aspects may also be taken to reflect basic human values (Poortinga et al., 2004). For example, a study of Poortinga and colleagues (2004) examined household energy use and concluded that impacts on health, partner and family, social justice, and freedom and safety were valued more important than impacts on material beauty, spirituality and religion, status and cognition and challenge and excitement. It is crucial to acknowledge perceived differences in importance ratings, since specific QoL- values, such as health, might outweigh other less significant values, such as material beauty or spirituality, and might not have an equal impact on overall subjective QoL in an e-bike
context.
2.6. Multiple dimensions of subjective Quality of Life. Poortinga and colleagues (2004) studied the role of values in the field of household energy use and investigated this by using the concept of subjective QoL. This researcher reduced the original 22 QoL-aspects, which represent various domains of what people may find important in life, and reduced them to seven clearly interpretable factors. The seven factors were identified as:
- Self-Enhancement (consisting of money/income, comfort, status/recognition, and material beauty);
- Environmental Quality (consisting of nature/biodiversity, aesthetic beauty, and environmental quality);
- Self-direction (consisting of privacy, freedom, and leisure time);
- Openness to Change (consisting of challenge/excitement, change/variation, and social relations);
- Maturity (consisting of spirituality/religion, identity/self-respect, and security);
- Family, Health, and Safety (consisting of partner and family, health, and safety);
- Achievement (consisting of work and education)
Poortinga et al. (2004) identified similar factors established by previous research of Schwartz and Bilsky (1990) who constructed a theory of universal types of values, comprising of:
achievement, enjoyment, maturity, prosocial, restrictive conformity, security, self-direction.
These universal types of values were based on a previous study of Schwartz and Bilsky (1987). This study proposed a theory of universal psychological structure of human values, identifying the facets necessary to define human values and specifying the motivational domains that people from different cultures are likely to distinguish. These human values seem to have similar interpretations as the QoL-dimensions described by Poortinga et al.
(2004) and give a more profound understanding of these QoL-domains.
2.6.1. Close linkages with biospheric value orientations. Poortinga (2004) identified the Environmental Quality factor. This dimension summarizes the QoL-indicators
nature/biodiversity, aesthetic beauty and environmental quality. Previous research concluded that this factor associates with a biospheric value dimension (Vlek et al., 1998).Hence, it is argued that biospheric oriented individuals have the closest association with the
Environmental Quality Factor. This assumption is supported by Steg and Gifford (2005) who posit that individuals with greater environmental concern evaluate environmental quality and personal freedom as more important, and material wealth as less important than individuals with less environmental concern.
The factor Maturity consists of the QoL-indicators spirituality/religion, identity/self-respect and security. According to Poortinga et al. (2004), this factor is more difficult to interpret and to assign to a specific value orientation. This factor is described as the appreciation,
understanding and acceptance of oneself, others and the surrounding world (Schwartz &
Bilsky, 1990). Since the Maturity dimension is closely linked with acceptance of the
surrounding world and the appreciation for the beauty of creation (Schwartz & Bilsky, 1987), it is argued that this factor has the closest associations with biospheric oriented individuals.
To that end, it is hypothesized that biospheric values have a direct relationship with the seven QoL-dimensions, especially with Environmental Quality and Maturity:
H1: Biospheric value orientations are related to the seven QoL-dimensions when adopting an e-bike.
Specifically stated, when evaluating QoL-dimensions related to e-bike use:
H1a: The stronger one’s biospheric value orientation, the more positive the QoL-dimension Environmental Quality will be evaluated.
H1b: The stronger one’s biospheric value orientation, the more positive the QoL-dimension Maturity will be evaluated.
2.6.2. Close linkages with altruistic value orientations. Poortinga and colleagues (2004) distinguished the Openness to Change factor, which comprises the QoL-variables
challenge/excitement, change/variation and social relations. Based on the social relations variable, it is argued that this dimension fits best with an altruistic value orientation. This is supported by Schwartz (1994) in which the same-named factor was found after distinguishing different types of values by one’s motivational goals. This factor resembles with intergroup contact, which provides exposure to new and different ways of life and opportunities to learn about and exploring them. Therefore, it is argued that altruistic values have a relatively higher explanatory power than egoistic and altruistic values in explaining Openness to Change.
In addition, Poortinga et al. (2004) showed that the QoL-variables health, safety partner and family, loaded highly on the factor Family, Health and Safety and portrays traditional values.
It is assumed that this dimension has the closest association with an altruistic value
orientation, since family and safety variables are tightly linked with to the welfare of other human beings (De Groot & Steg, 2008). This factor was found as well in a study by Schwartz (1994), labeling this value dimension as Security.Security values include inner harmony, family security, national security, and a world at peace, whereas one’s physical and mental health is important. Further, concepts of Security, such as interaction and groups are tightly linked with altruistic value orientations (Schwartz, 1994). Interaction refers to universal requisites of coordinated social interaction, whereas importance of groups exemplifies smooth functioning and survival of groups. Bringing this altogether, it is hypothesized that altruistic values have a relationship with the seven QoL-dimensions and especially with the factors Openness to Change and Family, Health and Safety:
H2: Altruistic value orientations are related to the seven QoL-dimensions when adopting an e- bike.
Specifically stated, when evaluating QoL-dimensions related to e-bike use:
H2a: The stronger one’s altruistic value orientation, the more positive the QoL-dimension Openness to Change will be evaluated.
H2b: The stronger one’s altruistic value orientation, the more positive the QoL-dimension Family, Health and Safety will be evaluated.
2.6.3. Close linkages with egoistic value orientations. Poortinga et al. (2004) identified three factors out of seven that seem to be closely related to an egoistic value orientation. The QoL- indicators money/income, comfort, status/recognition and material beauty correlated highly with one factor that was labeled as Self-Enhancement. This dimension could be characterized as materialistic oriented. It may be closely related to what Schwartz and Bilsky (1990) call enjoyment, which is correlated with emphasizing one’s own pleasure and comfort.
The QoL-indicators privacy, freedom and leisure time were labeled under the name Self- Direction. This factor is perceived as a self-centered dimension and is merely about autonomy and independence. It may be closely related to what Schwartz and Bilsky (1990) mentioned Self-Direction as well. Values in the self-direction domain refer to reliance on and
gratification from one's independent capacities for decision-making, creativity, and action.
Another factor identified by Poortinga et al. (2004), labeled as Achievement, comprised the QoL-variables work and education. Just like the Self-Enhancement and Self-Direction QoL- factors, it is argued that Achievement most closely relates with an egoistic value orientation, since egoistic oriented consumers especially consider costs and benefits for personal gain (De Groot & Steg, 2006). This factor was also found by Schwartz (1994) and referred to
Achievement as well.
Achievement is about the development and usage of skills to obtain those resources required to thrive. It closely associates with the need for competent performance. The expression of Achievement lies in values such as achievement, competence and success to serve one’s own interests. Schwartz and Bilsky (1990) reasoned that achievement and enjoyment are
connected, because both are concerned with self-enhancement. In sum, achievement,
enjoyment, and self-direction values serve individualistic interests and seem to affiliate with egoistic value orientations.All of this considered, the following hypotheses are stated, assuming that egoistic values have a relationship with the seven QoL-dimensions and especially with the dimensions Self-Enhancement, Self-Direction and Achievement:
H3: Egoistic value orientations are related to the seven QoL-dimensions when adopting an e- bike.
Specifically when evaluating QoL-dimensions related to e-bike use:
H3a: The stronger one’s egoistic value orientation, the more positive the QoL-dimension Self- Enhancement will be evaluated.
H3b: The stronger one’s egoistic value orientation, the more positive the QoL-dimension Self- Direction will be evaluated.
H3c: The stronger one’s egoistic value orientation, the more positive the QoL-dimension Achievement will be evaluated.
2.6. Mediating effect of the seven QoL-dimensions between values and overall subjective QoL.
Figure 1 displays the conceptual framework of this study. It is argued that the relationship between value orientations and overall subjective QoL on bicycle mobility is mediated by the seven QoL-dimensions identified by Poortinga et al. (2004). This is based on previous
research of Diener and Suh (1997) suggesting that value orientations can be used as a
conceptual vehicle for constructing subjective QoL-variables, and therefore QoL-dimensions.
Hence, value orientations seem to be important in subjective quality of life research, since universal values suggest a systematic way of selecting indicators that reflect distinctive QoL- dimensions (Diener & Suh, 1997). In addition, multiple studies assessed the expected effects in subjective QoL of future scenarios in various domains, including an environmental and transport context (see Poortinga et al., 2004; Gatersleben, 2000; Vlek, Skolnik & Gatersleben, 1998). So, it is implied that the seven QoL-dimensions are useful in predicting overall
subjective QoL in a bicycle mobility context, such as the transition from owning an ordinary bike to owning an e-bike. Therefore it is hypothesized that the seven QoL-dimensions mediate the relationship between value orientations and overall subjective QoL, since the seven QoL- dimensions originate from the initial scale explaining overall subjective QoL.
To test a potential indirect effect, the following hypotheses are stated, assuming that the seven QoL-dimensions mediate the relationship between values and overall subjective QoL:
2.6.1. Biospheric value orientations
H4: Biospheric value orientations are indirectly related to overall subjective QoL through the 7 QoL-dimensions when adopting an e-bike.
Specifically, when evaluating value orientations and QoL-dimensions related to e-bike use:
H4a: The QoL-dimension Environmental Quality will mediate the relationship between biospheric values and overall subjective QoL.
H4b: The QoL-dimension Maturity will mediate the relationship between biospheric values and overall subjective QoL.
2.6.2. Altruistic value orientations
H5: Altruistic value orientations are indirectly related to overall subjective QoL through the 7 QoL-dimensions when adopting an e-bike.
Specifically, when evaluating value orientations and QoL-dimensions related to e-bike use:
H5a: The QoL-dimension Openness to Change will mediate the relationship between altruistic values and overall subjective QoL.
H5b: The QoL-dimension Family, Health and Safety will mediate the relationship between altruistic values and overall subjective QoL.
2.6.3. Egoistic value orientations
H6: Egoistic value orientations are indirectly related to overall subjective QoL through the 7 QoL-dimensions when adopting an e-bike.
Specifically, when evaluating value orientations and QoL-dimensions related to e-bike use:
H6a: The QoL-dimension Self-Enhancement will mediate the relationship between egoistic values and overall subjective QoL.
H6b: The QoL-dimension Self-Direction will mediate the relationship between egoistic values and overall subjective QoL.
H6c: The QoL-dimension Achievement will mediate the relationship between egoistic values and overall subjective QoL.
Figure 1 Conceptual model
3. METHOD
3.1. Procedure and respondents. In May 2018 a survey in Qualtrics was created and
distributed among respondents in the Netherlands. Convenience sampling was used to get a substantial number of respondents. The questionnaires were distributed via e-mail, Facebook and WhatsApp among students, blue and white collar workers and the unemployed. Only those that did not already own an e-bike, but did own a regular bike were used in this survey.
3.2. Measures. The survey started with assessing the value orientations of each respondent, since value orientations showed to be important for understanding environmental beliefs and behavior (Steg et al., 2012). Starting the questionnaire with determining the value orientations assured that no respondent ws already primed with the e-bike scenario.
Measures of value orientations were based on an instrument conceived by De Groot and Steg (2005) that includes thirteen values that reflect these three value orientations. This scale aims to distinguish between egoistic, altruistic and biospheric value orientations. The following values were included: social power, wealth, authority, influential, ambitious (egoistic value
orientations), equality, a world of peace, social justice, helpful (altruistic value orientations), preventing pollution, respecting the earth, unity with nature, and protecting the environment (biospheric value orientations). Respondents indicated to what extent these values were important “as a guiding principle in their lives” on a 9-point scale ranging from –1 = opposed to my values, 0 = not important, to 7 = extremely important.
3.2.1. Expected changes in QoL by presenting the e-bike scenario. Expected quality of life effects of e-bike adoption were measured by asking respondents which consequences the following e-bike policy would have for their quality of life: ‘Imagine that you exchange your regular bike for an e-bike. Imagine how this transition fits in your life and to what extent this affects your current lifestyle’. This policy measure was chosen, because this hypothetical situation is easy to understand and easy to imagine for respondents. The change in subjective QoL of the e-bike adoption scenario were assessed by asking respondents to indicate to what extent the exchange from a conventional bike to an e-bike would affect the 22 relevant QoL- indicators (see Appendix 1). The list of subjective QoL-aspects resulted from extensive reviews of relevant literature (see Steg & Gifford, 2005; Gatersleben, 2000; De Groot & Steg, 2006).
The 22 QoL-indicators were introduced to assess the changes people expect after e-bike adoption. The participants rated the expected changes of the 22 QoL-aspects on a seven-point scale ranging from -3 (‘would decrease dramatically’) to 3 (‘would increase dramatically’).
Next to this scale, the respondents were asked to indicate on a seven-point Likert scale (-3 very unimportant, 3 very important) how important they found these 22 QoL-aspects for their own lives by having this e-bike scenario in mind.
Expected changes in subjective QoL were weighted based on importance judgements of the relevant subjective QoL-indicators, since changes in relatively more important QoL-indicators will be more significant for individuals than changes in QoL-indicators that are considered to be less important. Overall expected changes in subjective QoL on bicycle mobility were calculated by summing the expected changes on the QoL-indicators, after multiplying the importance assigned to each QoL-indicator (Steg & Gifford, 2005). These QoL-indicators were standardized before multiplying with each other, This was done, because different scales were used and this new variable would otherwise have created a different measurement scale.
3.2.3. Overall subjective QoL. At the end of the survey an overall judgement about expected quality of life on bicycle mobility (‘All things considered, to what extent would the exchange
from a regular bike to an e-bike influence your overall quality of life?’) was asked (De Groot
& Steg, 2006). The same seven-point Likert scale was used as with the 22 individual QoL- indicators (-3 ‘would decrease dramatically’ to 3 ‘would increase dramatically’).
4. ANALYSIS
4.1. Determining minimum sample size for factor analysis. There is not a single, definitive way to determine the minimum number of respondents for answering the hypotheses using multiple regression (Field, 2013). The minimum number of participants is mainly dependent on the required level of the alpha error, the number of independent variables and their
contribution to the R² (how well the independent variables explain the outcome variable), the number of covariates and their independent contribution to the R² (Field, 2013). Hence, the sample size required to test the overall regression model depends on the number of predictors and the size of the expected effect, which is portrayed in the R². A relative large effect size (.26) with up to 20 predictors suffices a sample size of 77 participants, whereas a relative small R² (.02) with six or less predictors should suffice a sample size of 100 (Field, 2013).
According to Van Voorhis and Morgan (2007) the general rule of thumb for regression analysis is to acquire no less than 50 participants with the number increasing with larger numbers of independent variables. Green (1991) provides two rules of thumb to determine the minimum acceptable sample size. First, testing the R² to test the overall fit of the regression model. A minimum sample size of 50 + 8k is recommended, where k is the number of
predictors. So, in this research, a minimum sample size of 50 + 80 (three values + seven QoL dimensions) = 130 is needed. Second, it is recommended to test the beta values of the model.
This means that the individual predictors within the model are tested. In order to test the individual independent variables a minimum sample size of 104 + k is advised, where k is the number of predictors. This suggests that a minimum sample size of 104 + 10 (three values + seven QoL dimensions) = 114 would suffice.
4.2. Determining the minimum sample size for factor analysis. There is not a single, definitive calculation to determine a minimum sample size for doing an appropriate factor analysis as well (Field, 2013). A rule of thumb involves a minimum of five or ten respondents per item (Van Voorhis & Morgan 2007). Yong and Pearce (2013) advice to obtain a minimum of 300 participants. Another study of Williams, Onsman and Brown (1996) ranked a sample size of 100 as poor, 200 as fair, 300 as good, 500 as very good, and 1000 or more as excellent.
However, when communality values are relatively high (above .60) even a small sample size (less than 100) might be perfectly adequate (MacCallum et al., 1999).
4.3. Reflecting of actual sample size on theorized minimum. In total, 168 respondents participated in this questionnaire. After cleaning the dataset by eliminating respondents that did own an e-bike or did not fully completed the questionnaire, a total of 133 respondents remained. According to Williams, Onsman and Brown (1996), this is a poor sample in order to conduct an appropriate factor analysis. However, the sample size was large enough to suffice the common rule of thumb for factor analysis (Van Voorhis & Morgan 2007), since the minimum number of 110 respondents was met. Moreover, this sample size was large enough to meet the requirements to conduct an appropriate regression analysis (Green, 1991).
To emphasize, the main aim of present study was to connect theorized constructs with each other, not to capture a representative sample. In addition, this study used confirmatory rather than exploratory factor analysis, because confirmatory factor analyses seems more robust for smaller sample sizes (Sun, 2005).
The total sample of present study (N = 133), consisted of 68 males. The age of the
respondents ranged from 18 to 59 years (M = 30,40. SD = 10,67). Forty-seven percent of the participants completed a bachelor or master’s degree. Sixty-four percent belonged to the working class, while 30,1% of the respondents were students. The ratio between males and females and between students and those focusing on work is quite balanced. However, the mean age and the educational level seem not to be quite representative to the population.
4.4. Factor analysis. Confirmatory factor analysis was conducted to factor the initial 13 value items in biospheric, altruistic and egoistic dimensions. Confirmatory factor analysis was used, because these value dimensions were extensively used in previous research (see De Groot &
Steg, 2006; De Groot & Steg, 2010). Confirmatory factor analysis was also conducted to reduce the initial 22 QoL-indicators into seven QoL-dimensions. Poortinga et al. (2004) found that the variable social justice had no high loading on any of the seven QoL-dimensions and omitted this variable from factor analysis and further analyses. This structure was followed in this research as well.
Both initial items measuring value orientations and subjective QoL were assessed with a Likert scale. This is an ordinal scale, but is also assumed as a continuous interval scale, which is suitable for conducting factor analysis (Norman, 2010). To do a factor analysis, it is
necessary to follow a specific procedure. Before factors of value orientations were extracted a
Kaiser-Meyer-Olkin (KMO) test was conducted to measure sampling adequacy. Sampling adequacy predicts if data are likely to factor well, based on correlation and partial correlation.
A KMO-value greater than .50 is acceptable to proceed with factor analysis (Ferguson & Cox, 1993). Further, a Bartlett’s test of sphericity was performed to investigate correlations
between the variables. Then, communalities (the percent of variance in a given variable explained by all the extracted factors) were checked. Communalities higher than .40 are sufficient to proceed with factor analysis (Ferguson & Cox, 1993). After rotating the factors, which prevents that all variables load on a single factor and minimizes the number of
variables which have high loadings on each given factor, a reliability analysis was executed to test the internal consistency of the theoretically assumed factors. Since the underlying QoL- dimensions were already known from previous research, the Cronbach’s Alpha of each scale was measured. A minimum measure of .60 is an acceptable level to proceed with a factor (Field, 2013).
4.5. Regression analysis. Multiple regression analysis between value orientations and the seven QoL-dimensions were measured to examine to what extent biospheric, altruistic and egoistic value orientations have a relationship with the seven QoL-dimensions. To examine these relationships it is critical to assess the standardized beta’s of the predictors, since these standardized beta coefficients compares the relative strength of the effect of each
individual independent variable to the dependent variable.
4.5.1. Assumptions in multiple regression. In order to analyze the data using multiple regression, it was checked whether the data met the assumptions for conducting regression (Field, 2009). First, multiple regression requires at least two independent variables. This requirement was met, since this study used three value orientations and seven QoL- dimensions as independent variables at an interval (assumed continuous) scale (Norman, 2010).
Second, the dependent variable should be measured on a continuous scale, which is either interval or ratio scaled. Since the dependent variables, consisting of the QoL-dimensions and overall subjective QoL, were measured on an seven-point ordinal Likert scale, these scales can be interpreted as continuous variables (Norman, 2010). Therefore, it is assumed that this assumption was met.
Third, there must be a linear relationship between the dependent variable and the independent variables. Scatterplots (see Appendix II) showed that there was a linear relationship between
the independent variables and the outcome variable. After inspecting the histograms, there were some obvious outliers per QoL-dimension. Those outliers were substituted with the highest value that was not an outlier (Field, 2013).
Fourth, it is important that multicollinearity between the independent variables does not occur. Multicollinearity occurs when the independent variables are too highly correlated with each other. This assumption was tested using Variance Inflation Factor (VIF) values (see Appendix II). It can be concluded that the VIF-scores did not exceed the threshold of 10, which assumes highly correlated independent variables. Therefore, this assumption was met.
Fifth, the residuals need to be normally distributed. The histogram (after running a regression analysis) showed that the residuals were not normally distributed. Therefore, this assumption was violated.
Lastly, the variance of the error terms need to be similar across the values of the independent variables (i.e. homoscedasticity). A plot of standardized residuals versus predicted values can show whether points are equally distributed across all values of the predicting variable and shows if heteroscedasticity occurs (Field, 2013). Standardized residuals as Y-variable and standardized independent variables as X-variable showed that heteroscedasticity occurred (see Appendix II). Therefore, this assumption was violated. The data did not fully meet the
assumptions, since the data was not normal and was not homoscedastic. We used a
bootstrapping procedure to overcome these limitations, because this method gives an accurate estimate of the true population value of the beta for each predictor (Field, 2013).
4.6. Mediating role of QoL-dimensions on the relationship between values and overall subjective QoL. In order to test the relationship between value orientations and overall subjective QoL mediated through the seven QoL-dimensions, a mediation analysis was conducted. The prediction that the mediator is caused by the value orientations, and is in itself a cause for explaining overall subjective QoL, was tested. The mediation role of these seven QoL-dimension was tested by following specific steps in SPSS using a special PROCESS regression developed by Hayes (2004).
5. RESULTS
5.1. Validity and reliability of values and QoL-dimensions. In order to test if it is appropriate to do a factor analysis for reducing the value variables to three dimensions, a KMO measure of sampling adequacy was conducted. A measure of .68 was found, meaning that the data are likely to factor well (see Appendix II). A Bartlett’s test of sphericity was used to investigate if
the variables were correlated. The overall model was significant (p < .001) and, therefore, the variables were correlated. All communalities were higher than .40, since the lowest
communality showed a measure of .46 (see Appendix II). As predicted, factor analysis found three different dimensions. After rotation all altruistic, biospheric and egoistic variables loaded highly on a distinct factor (see Appendix II). Next, a reliability analysis was conducted to investigate whether the theoretical scale of value orientations is reliably measuring one and the same underlying construct. This proved to be the situation. The altruistic values showed a Cronbach’s alpha of .74. The biospheric values showed a value of .81 and egoistic values proved to be reliable with a Cronbach’s alpha of .75, which means that all of them exceeds the minimum threshold of .60 (see Appendix II).
Confirmatory factor analysis was also conducted to identify the seven QoL-dimensions out of the initial 22 QoL-indicators. To test if the theorized dimensions identified by Poortinga et al.
(2004) were found in this research as well, the internal consistency was measured. Table 2 shows that every QoL-dimension reached the minimum Cronbach’s Alpha of .60 except the factors Maturity (.51) and Health, Family and Safety (.57). Even if the lowest scoring item in this factor was deleted, the internal consistency of this factor would not exceed a Cronbach’s Alpha of .60.
QoL-dimension Expected changes of factor in QoL after e-bike uptake
Importance rating of factor after e-bike uptake Health, Family and Safety .57 (.55 if deleted) .88
Openness To Change .76 .90
Maturity .51 (.55 if deleted) .80
Environmental Quality .82 .77
Achievement .72 .84
Self-Enhancement .70 .86
Self-Direction .63 .82
Table 2 Cronbach's Alpha of QoL-dimensions
4.2. Descriptive statistics: mean and standard deviations of values, QoL-dimensions and overall subjective QoL. Table 3 shows that respondents expected different anticipated changes per QoL-dimension after e-bike uptake. Total expected changes in a particular QoL-
dimension were found by combining the indicated effects of e-bike uptake multiplied by importance ratings of dimension. Z-scores were portrayed, because both scales were
measured differently before combining them into a new factor. This study also assessed how
people’s overall subjective QoL is affected by the scenario of e-bike uptake. Unstandardized mean scores of this outcome variable are displayed in this figure as well.
The mean scores show that in general the respondents expected an increase in every QoL- dimension after e-bike adoption. This seems to be in line with the overall subjective QoL measure, which also slightly improves (M = 4.41, SD = 1,03) when individuals exchange their ordinary bike for an e-bike. Furthermore, the QoL-dimensions Openness to Change (M
= .45, SD = 1,12), Self-Direction (M = .44, SD = 1,11), Environmental Quality (M = .41, SD
= 1,06), and Achievement (M = .40, SD = 1,03) showed the highest mean score in terms of anticipated changes in subjective QoL when changing from an ordinary bike to an e-bike. The dimensions Maturity (M = .15, SD = .91) and Health, Family and Safety (M = .16, SD = .84) appeared to have the lowest mean score after potential e-bike uptake.
QoL-dimension N Minimum Maximum Mean Std. deviation
Environmental Quality
133 -2.13 4.31 .41 1.06
Self-Direction 133 -1.10 5.04 .44 1.11
Openness to Change
133 -1.09 4.49 .45 1.12
Self-
Enhancement
133 -5.10 3.63 .35 1.02
Achievement 133 -1.42 4.81 .40 1.06
Health, Family and Safety
133 -2.40 3.15 .16 0.84
Maturity 133 -2,30 3,58 ,15 ,91
Overall
subjective QoL
133 1 7 4,41 1,03
Table 3 Anticipated changes in QoL after e-bike uptake
4.3. Relationships between value orientations and QoL dimensions. In order to test the relationship between value orientations and the seven QoL-dimensions, several multiple regressions were conducted. A bootstrap method with standardized variables was used since a few assumptions to conduct multiple regression were violated.
4.3.1. Regressing Environmental Quality and Maturity. The value orientations were regressed with the QoL-dimensions Environmental Quality and Maturity. Table 4 shows that even though altruistic values significantly explain both dependent variables Environmental Quality and Maturity (p < .05), the general models did not proved to be statistically significant (p = .12). This indicates that the overall regression model is not a good fit of the data gathered.
Thereafter, there is no relationship between value orientations and Environmental Quality and Maturity.
In sum, the three value orientations did not affect the QoL-dimensions Environmental Quality and Maturity in this e-bike adoption scenario. This is not in line with the initial hypothesis, because it was hypothesized that altruistic, egoistic and especially strong biospheric values result in a more positive evaluation of Environmental Quality and Maturity. Therefore, hypotheses h1a and h1b were rejected.
Standardized β p R² Adjusted R²
Model fit (F)
Sig ANOVA Dependent variable:
Environmental Quality
.04 .02 2.00 .12
Altruistic .22 .04*
Biospheric .04 .64
Egoistic .01 .97
Dependent variable:
Maturity
.04 .02 1.95 .12
Altruistic .13 .04*
Biospheric .11 .14
Egoistic .08 .37
Table 4 Regression of values on Environmental Quality and Maturity
4.3.2. Regressing Openness to Change and Health, Family and Safety. Another multiple regression was run to test to what extent the QoL-dimensions Openness to Change and Health, Family and Safety were explained from altruistic, biospheric and egoistic value orientations. It was hypothesized that stronger altruistic values (compared with biospheric and egoistic values) result in a stronger evaluation of the QoL-dimensions Openness to Change and Health, Family and Safety. Table 5 shows that the general model of the factor Health, Family and Safety did not proved to be statistically significant (p = .41). The overall
regression model of Openness to Change exemplified that the factor is a good fit of the data (p
< .05), implying that the independent variables explain Openness to Change. However, the model estimates of altruistic (p = .051), biospheric (p = .13) and egoistic values (p = .19) were not significantly different from zero and, therefore, do not have a relationship with Openness to Change. All in all, the three value orientations were no significant predictors for explaining the QoL-dimensions Openness to Change and Health, Family and Safety in this e-bike
scenario. Therefore, hypotheses h2a and h2b were rejected.
Standardized β p R² Adjusted R²
Model fit (F)
Sig ANOVA Dependent variable:
Openness to Change
.06 .04 2.67 .05*
Altruistic .19 .051
Biospheric .14 .13
Egoistic -.13 .19
Dependent variable:
Health, Family and Safety
.02 .00 .96 .41
Altruistic .00 .99
Biospheric .08 .30
Egoistic -.01 .30
Table 5 Regression of values on Openess to Change and Health, Family and Safety
4.3.3. Regressing Self-Direction, Self-Enhancement and Achievement. Next, another multiple regression was run to explain the relationship between altruistic, biospheric and egoistic value orientations and the QoL-dimensions Self-Enhancement, Self-Direction and Achievement.
Table 6 shows that the overall effects of the value orientations did statistically significantly affect these specific QoL-dimensions, except for the Self-Enhancement factor. Even though the overall model fit of Achievement was significant (p < .05), the relative contributions from the altruistic (p = .07), biospheric (p = .10) and egoistic values (p = .07) did not prove to be significant. Thus, there is no relationship between the predicting variables altruistic,
biospheric and egoistic values and the outcome variables Self-Enhancement and Achievement. Therefore hypotheses h3a and h3c were rejected.
The overall model explaining Self-Direction showed a significant effect. The F-test appeared to be highly significant (F = 5,22, p < .001), thus it is proven that the value orientations
explained a significant amount of the variance in the factor Self-Direction. Both biospheric (p
< .05) and egoistic values (p < .05) were significantly related to Self-Direction. Egoistic values proved to have a little stronger effect (.27) compared with biospheric vales (.22). So, the stronger one’s altruistic and in particular one’s egoistic values, the stronger this person will value the QoL-dimension Environmental Quality. This is in line with the initial proposition that strong egoistic values result in a more positive evaluation of the QoL- dimension Self-Direction. Therefore, hypothesis h3b was not rejected.
Standardized β
p R² Adjusted
R²
Model fit (F)
Sig ANOVA Dependent variable:
Self-Enhancement
.05 .03 2.4 .07
Altruistic .17 .05*
Biospheric .14 .10
Egoistic .07 .43
Dependent variable:
Self-Direction
.10 .08 5.22 .00*
Altruistic .12 .12
Biospheric .22 .04*
Egoistic -.27 .02*
Dependent variable:
Achievement
0.09 0.07 4.15 .01*
Altruistic .16 .07
Biospheric .15 .10
Egoistic -.22 .07
Table 6 Regression of values on Self-Enhancement, Self-Direction and Achievement
4.4. Mediation analysis. Mediation implies a situation where the effect of the independent variable on the dependent variable can best be explained using a third mediator variable, which is caused by the independent variable and is in itself a cause for the dependent variable.
This mediation analysis was conducted to test whether the QoL-dimensions mediate the relationship between the value orientations and overall subjective QoL. To start; biospheric, altruistic and egoistic values must explain overall subjective QoL in the first place (called effect c). Second, the value orientations must explain the mediators, which are specific QoL- dimensions (called effect a). Third, the mediators must explain overall subjective QoL (called
effect b). Fourth, the relationship between value orientations and overall subjective QoL should be smaller when the mediators are included in the model than when those mediators are not included (called effect c’).
Biospheric values and overall subjective QoL mediated through QoL-dimensions. Table 7 reports that biospheric values affect the mediators Maturity, Achievement, Self-Enhancement and Self-Direction. Also, it was found that Self-Enhancement and Environmental Quality predict overall subjective QoL. Nevertheless, a significant indirect could not be established.
Therefore, the QoL-dimensions Maturity and Environmental Quality are no mediators of the relationship between dominant biospheric value orientations and overall subjective QoL. The initial hypotheses proposed that the QoL-dimensions act as a mediator between value
orientations and overall subjective QoL. To that extent, the results of this particular analysis are not in line with the hypotheses h4a and h4b and were hereby rejected.
