Young professionals and their risk perceptions of global climate
change: The influence of human values and gender
Master Thesis / Dissertation
Frank van Urk, 2510340 – 9018324
F.van.urk@student.rug.nl
/
F.van-urk2@newcastle.ac.uk
Supervisors: dr. Bartjan Pennink & dr. Alan McKinlay
DDM - Advanced International Business Management and Marketing
2
Abstract
Despite a global scientific consensus that climate change is happening and human activity is contributing, the public remains sceptical. The key objective of this study is to investigate the human values of both men and women and examine the predictive power of human values on climate change risk perceptions of both genders, and young professionals. Young
professionals are of specific interest because they are the future policymakers and are most likely to outlive older generations, while they will be facing various imminent climate change related issues in the future. These relationships are studied by exploring the effects of
Schwartz’ theory of high-order value types on climate change risk perceptions of respondents of the European Social Survey Round 8 (n=44,387) from 23 participating countries. It is expected that men hold more self-enhancement and conservation values and women holding more self-transcending and openness-to-change values, while young women have higher levels of climate change risk perceptions than young men. Moreover, the different value orientations are hypothesized to either have a negative or positive effect on climate change risk perceptions of young professionals. By conducting several t-tests, it is found that men hold more enhancement and openness-to-change values, while women hold more self-transcending and conservation values, and show higher levels of climate change risk perceptions than men. Furthermore, various regression analyses reveal that
self-transcendence values and the belief that climate change is caused by humans positively influence climate change risk perceptions of young professionals, whereas self-enhancement values negatively influence climate change risk perceptions of young professionals. The findings of this study thereby add to the growing importance of the climate change literature, international business landscape and risk perception theory and have several theoretical and managerial implications.
3
Acknowledgements
4
Table of Content
Abstract ... 2 Acknowledgements ... 3 List of Figures ... 6 List of Tables ... 6 List of Abbreviations... 6 1. Introduction on research ... 7 2. Literature review ... 112.1 Risk and risk perception ... 11
2.1.1 Gender differences of risk perceptions ... 12
2.2 Risk perceptions of global climate change... 14
2.2.1 Climate change doubts ... 15
2.3 Human values ... 17
2.3.1 Gender differences of human values ... 18
2.3.2 Human values and risk perceptions ... 18
2.3.3 Human values and risk perceptions of global climate change ... 19
2.4 Human values of young professionals ... 22
2.4.1 Young professionals and GCC risk perceptions ... 23
2.4.2 Gender differences of GCC risk perceptions of young professionals ... 24
2.5 Hypotheses ... 25
2.6 Conceptual model ... 28
2.7 Communication of GCC issues ... 29
2.8 Implications for business ... 29
3. Research Methodology ... 31
3.1 Data Collection and research ethics ... 31
3.2 Sample ... 32
3.3 Measurement of variables ... 32
3.3.1 Independent variables ... 32
3.3.2 Dependent variables ... 33
3.3.3 Control variables ... 33
3.4 Empirical data analysis ... 34
4. Results ... 35
4.1 Test of basic assumptions... 35
4.2 Descriptive statistics ... 35
5 4.4 T-test results ... 39 4.5 Regression results ... 41 5. Discussion... 45 6. Conclusion ... 49 6.1 Theoretical implications ... 49 6.2 Practical implications ... 50
6.3 Limitations and future research ... 51
References ... 53
Appendices ... 70
Appendix A: ESS questions human values ... 70
Appendix B: Preliminary analysis ... 71
Appendix C: Distribution per country... 73
Appendix D: Descriptive statistics random samples... 74
Appendix E: Correlation matrix, random sample all ages ... 75
Appendix F: Correlation matrix, random sample young professionals ... 76
6
List of Figures
Figure 1: Conceptual Model
Figure 2: Conceptual Model after results
List of Tables
Table 1: Descriptive statistics
Table 2: Correlation matrix, sample all ages
Table 3: Correlation matrix, sample young professionals Table 4: T-test results hypothesis 1
Table 5: T-test results hypothesis 2 Table 6: Regression analysis
List of Abbreviations
AC Awareness of consequences
CSR Corporate Social Responsibility
ESS European Social Survey
GCC Global climate change
IPCC Intergovernmental Panel on Climate Change ISSP International Social Survey Programme PVQ Portrait Values Questionnaire
SD Standard deviation
UK United Kingdom
USA United States of America
7
1. Introduction on research
Today, many business leaders recognize the dangers of Global Climate Change (GCC) and the value of environmental protection (Senge et al., 2008; Vandenbergh and Gilligan, 2015) as opposed to around 25 years ago when most business leaders were sceptical about environmental requirements (Glicksman, 2010). Companies cannot thrive in societies with public and ecological health problems as a result of GCC (Ameer and Othman, 2012; Porter and van der Linde, 1995), where a strategic focus on environmental issues can even pay off in the marketplace (Ameer and Othman, 2012). Because multinationals are expected to play a major role given their global activities and influence, companies need employees and managers that recognize the risks and dangers of GCC and take these in consideration when making important (managerial) decisions (Howard-Grenville et al., 2014). Therefore, any information on what influences perceptions of GCC is important information for corporates when finding the right people for the right job, especially when these jobs concern sustainability or environmental matters. This study aims to contribute to the international business landscape, risk perception theory and the climate change literature by focusing on the relationship between human values, gender and climate change risk perceptions of young professionals.
GCC is a major, multilevel societal process causing some of the most daunting political and ethical challenges of the twenty-first century. These complex intergenerational issues have experienced increased attention over the last decades (Arnold, 2011). GCC is impacting both human well-being (Levy and Patz, 2015; Watts et al., 2015; Beck, 2010) and the well-being of the earth (Letcher, 2015). GCC is causing significant negative effects on mental health and well-being (Fritze et al., 2008; Costello et al., 2009), mostly felt by those with pre-existing serious mental illness and vulnerable populations (Page and Howard, 2010). Moreover, extreme weather events and degraded landscapes resulting in localized or immediate consequences, such as stress or injuries, may be regarded as direct and personal impacts of climate change (Kolbert, 2015). Implications on the earth’s well-being are widespread and include negative effects on sea life, coral reef ecosystems, marine biodiversity, plant ecology, coastline degradation, sea ice and insect communities (Letcher, 2015). Present and future populations are put in jeopardy due to GCC, with the poor amongst the most adversely impacted (Arnold, 2011).
8 Anderegg et al., 2010) and enounce (Cook et al., 2013) that it is occurring. Scientists unequivocally link GCC to human activity, with the climate changing because of the inefficient consumption of (too much) fossil fuels such as coal, oil and natural gas; and because of deforestation (Karl et al., 2009; Arnold, 2011). However, this view is not shared by the public. In the United States of America (USA), just above fifty percent of the population believes anthropogenic climate change is happening. Anthropogenic climate change, which is also known as global warming, is climate change caused by human activity (Weber and Stern, 2011; Leiserowitz et al., 2013). Public perceptions have shown to be critical components of the socio-political context that policy makers operate in. These public perceptions can fundamentally constrain or compel social, political and economic action to focus on certain risks (Leiserowitz, 2006). Existing literature often examined the willingness to pay for public goods, including risk reduction. However, contingent valuation studies rarely asked respondents how risky they assess certain environmental hazards to be (O’Connor et al., 1999). Studies that found how much people are willing to pay in order to reduce certain risks failed in identifying whether these respondents view these risks as highly probable, threatening, and resulting in severely negative consequences. Hence, we know much about GCC denial and scepticism, but not much about how risky the public perceives climate change to be. General environmental beliefs are found to be predictors of behavioural intentions, in other words, risk perceptions about GCC are predicting the willingness and perceived need by the public to deal with causes of GCC (O’Connor et al., 1999).
9 topic of this paper. Moreover, research proved that cultural, evolutionary and psychological factors affect the diverse human behaviours and perceptions towards nature (Kaplan and Kaplan, 1989; Gobster, 2011). Social scientists and humanists have found that attitudes are framed by risk perceptions and value systems. These components are part of a few core principles and cognitive structures that influence and help explain ultimate behaviour towards nature and invasive species (Churchill et al., 2002; Fischer and van der Wal, 2007; Norgaard, 2007).
Strazdins and Skeat (2011) argued that previous crucial policy decisions related to climate change have not adequately considered the views and interests of younger people or ‘young professionals’ -used in this study to describe people born 1982-1999-, despite the fact that these decisions will ultimately shape their and the world’s future. Besides the lack of involvement and consideration of this age group, the lower the age, the higher the disproportionately effect of environmental hazards because of a relatively longer period of exposure (McMichael et al., 2005). In general, people aged over 65 appear to be the most sceptical about climate change (Defra, 2007; Schubert and Soane, 2008; Whitmarsh, 2011), while younger people exhibit mostly higher levels of reported concern and interest with lower scepticism (Corner et al., 2015). 76% of the respondents of studies in Austria, Germany and Switzerland reported to perceive climate change as a fairly or very big problem (Schneekloth et al., 2010; Corner et al., 2015), with similar results found studies in the USA (Wachholz et al., 2014), Oman (Ambusaidi et al., 2012) and India (Chhokar et al., 2011), while all studies display elevated levels of concern of younger people compared to older age groups. Finally, younger people are generally outliving older people, making them of significant interest.
While different age groups perceive GCC in different ways, so do men and women. Men are found to deny GCC more as compared to women. Additionally, women are better informed about GCC related aspects and they are more modest about their knowledge (McCright, 2010; Swim et al., 2018), indicating the higher importance they place on this knowledge field (Swim and Geiger, 2017; Orlikowski, 2002). Moreover, in the USA, more women were found to believe climate is happening (59%) as compared to men (54%); and more women believe it is happing because of human activities (64%-56%) (McCright and Dunlap, 2011; Swim et al., 2018), but reasoning for this disparity remains (Swim et al., 2018).
10 combination of human values and its influence on risk perceptions of GCC of young professionals, which is the focus of this thesis. This research gap is aimed to be bridged by examining the following research question:
To what extent do human values influence climate change risk perceptions of young professionals and how does this differ between and men and women?
11
2. Literature review
Weather and climate are often used interchangeably, however, weather research concerns the prediction, generation and propagation of individual weather elements, such as hurricanes and other short-term occurrences in the atmosphere, whereas climate research covers the entirety of these elements and focuses more on longer-term changes. In other words, “climate is what you expect, weather is what you get” (Allen, 2003). Since the beginning of the industrialization, a strong increase in the concentration of greenhouse gasses in the atmosphere has been observed, resulting in global warming on the earth´s surface. An 0.8 °C increase in global temperature has been witnessed over the last century, of which 0.6 °C were added just in the last thirty years (Latif, 2009). The Intergovernmental Panel on Climate Change (IPCC) has a decisive role in both the evaluation and the communication of scientific results regarding GCC. In their latest report in 2018, the IPCC named human influence the dominant cause of GCC since the mid-20th century, among a multitude of factors (Allen et al., 2019). The transition towards socially and ecologically more sustainable ways of life has proven to be a major challenge for modern societies (Orru and Lilleoja, 2015). Individuals and societies all respond in their own way to GCC challenges, with the responses being contingent on public perceptions of implications, causes and consequences of GCC (Pidgeon and Fischhoff, 2011; Pietsch and McAllister, 2010; Höppner and Whitmarsh, 2012). Public risk perceptions of GCC are critical elements of the socio-political context within which policymakers operate. Political, social and economic action to address risks can be fundamentally compelled or constrained by the opinion of the public (Leiserowitz, 2005). Risk perception research, therefore, lies at the basis of this research.
2.1 Risk and risk perception
12 activities. However, this correlation is negative in people’s judgements and minds, where high risk is expected to yield lower benefits, and vice versa (Lerner et al., 2003; Slovic and Peters, 2006). Moreover, Alhakami and Slovic (1994) discovered that people rate activities on bipolar scales, such as good/bad or nice/awful, and that the strength of the negative or positive affect associated with such activities is linked to the inverse relationship between perceived benefit and perceived risk. This implies that not only knowledge, but also the way people feel about risks is included in judgements about risk. Favourable feelings towards certain activities tend to result in lower perceived risks and higher perceived benefits, whereas unfavourable feelings tend to result in opposite judgements (Finucane et al., 2000). Shalom Schwartz (1968) wrote about this decades ago and introduced the term awareness of consequences (AC) in combination with environmental concern. According to his research, for people with a strong social-altruistic orientation, believing that certain environmental conditions have adverse (negative) consequences for others will trigger pro-environmental behaviour. Here, risk perceptions or beliefs are influenced by human values and attributes. However, this relationship has seldom been tested, especially for uncertain, long-term environmental risks that are associated with GCC. Most risk perception studies focused on the role of information and knowledge in the formation of opinions on (environmental) risks. Being able to recognize a problem and its ameliorative options should motivate pro-environmental acts that are independent of risks perceptions (O’Connor et al., 1999).
2.1.1 Gender differences of risk perceptions
The research question of this paper consists of two questions, with the second question addressing gender differences. There is a strong reason to believe that gender differences exist in the relationship between human values of young professionals and risk perceptions of GCC (Perera and Hewege, 2013; Poortinga et al., 2019). Therefore, gender differences will serve as an independent variable in this study, which can later be seen in the conceptual model, and will be a recurring section in this paper. This section looks into gender differences of risk perceptions. Later sections will address gender differences of human values and gender differences of young professionals and climate change.
13 most studies that examine gender in relation to risk perceptions fail to explain, or even attempt to explain the gender differences found (Wachinger et al., 2010; Gustafson, 1998). An explanation for these differences may be that women assume they would be more emotionally harmed or upset by negative consequences if these should occur. Another option might originate in the assessment of outcomes by women, predicting more unfavourable outcomes without projecting stronger negative responses to these outcomes than do men (Harris, 2006). Hillier and Morrongiello (1998) studied physical risk-taking perceptions of children and found that girls appraised more risks than do boys. In addition, boys judged risks differently as they rated injury severity, while girls rated vulnerability of any type of injury. Therefore, boys tend to avoid risky situations in case of severe perceived injury, whereas girls tend to avoid risky circumstances with any probability of perceived injury. A similar study that focused on adults found that gender differences exist in the perceived importance of risks. Women are more focused on family and home, often regarding risks as threats to their homes and families, e.g. risk of death, health risks and accident risks. Men are more concerned about their working life and mentioned risks of unemployment and economic problems (Enander and Jakobsen, 1996). Johnson et al. (2004) found that men predicted to obtain greater benefits from engaging in certain types of risky behaviour in comparison to women. To summarize, men put more emphasis on the potential beneficial outcomes of a risky decision (Gardner and Steinberg, 2005) and are also more impulsive (Waldeck and Miller, 1997), whereas women are more sensitive to uncertainty, consequences and punishment (Lee et al., 2009; van den Bos et al., 2013), and are more risk averse (van Leijenhorst et al., 2008; Eckel and Grossman, 2008).
14 of genders that could lead to a different framing of risks: the same risk may not always mean the same thing to men and women. This could result in the rise of different levels of concern (Davidson and Freudenburg, 1996). Therefore, socio-political factors, such as status, power and trust might cause gender risk perception differences (Flynn et al., 1994; Slovic, 2016; Whitmarsh, 2011). This will be essential for the formation of the first hypothesis.
2.2 Risk perceptions of global climate change
O’Connor et al. (1999) conceptualized risk perceptions of GCC as: ‘’the perceived likelihood of negative consequences to oneself and society from one specific environmental phenomenon: global warming’’. Current research on this topic has mostly been dominated by studies from the USA and Australia, and to some degree from Europe (Lee et al., 2015). These findings have contributed significantly to the understanding of GCC risk perceptions and its complexity, however, the results may be culture- and country-specific, making it difficult to generalize across a culturally, economically and geographically diverse planet. Additionally, sociological research has investigated the influence of contextual processes and factors as powerful forces that shape individuals’ and communities’ engagement with the issue (Lee et al., 2015). These national differences in risk perceptions of GCC may help to explain the existing national differences in political support for climate action. Yet, currently we lack even a rudimentary understanding of all the elements that together shape people’s GCC risk perceptions and awareness, due to past data unavailability (Lee et al., 2015; Cook et al., 2013). Therefore, GCC risk perceptions will serve as the dependent variable in this study, as can later be seen in the conceptual model.
15 percentages of people that deny GCC and/or regard GCC as a risk and threat to the survival of the earth. For example, 9,7% of the Russian population believe that the climate is ‘’definitely not changing’’, while percentages are much lower in other European countries (Steg, 2018). In the USA, polling data showed that, in 2010, 25% of the sampled population said to worry a great deal about GCC, whereas this percentage increased to 34% in 2014 (Capstick et al., 2015). Personal concern for GCC is believed to be associated with extreme weather conditions and events. Donner and McDaniels (2013) looked at data from national weather station observations and focused on the extent to which these corresponded with public opinion data over the same period. This study covered twenty years of data (1990-2010) and found that personal concern for and belief in GCC were strongly related to temperature anomalies over the last year. Both attitudinal measures were lower when experiencing colder temperatures than usual, and higher for periods that were warmer. Moreover, religion and religious values can also influence the concern for environmental issues more generally, with certain religions promoting societies’ exploitation of nature, whereas other religions appeal to maintain the creations of God (Kvaløy et al., 2012). That is why religion will serve as a control variable in this study.
2.2.1 Climate change doubts
16 information that is provided (Lord et al., 1979). Public doubt in GCC can, for example, be caused by affective and cognitive factors (McCright and Dunlap, 2011; Gifford, 2011), media coverage that grants disproportionate weight to GCC controversy (Boykoff, 2007) and scientists, specialized in other disciplines, who are publicly sceptical about GCC (Anderegg et al., 2010).
There are several older and more recent studies that suggest that risk perceptions are not determined by expertise, but rather by ideological variation (Slovic and Peters, 1998; Sjöberg, 2003; Kahan et al., 2012). This can be explained by the lacking habits of mind of the public that result in the inability to assimilate and assess scientific knowledge in a dispassionate, analytical manner; and making them constrained to rely on fallible heuristic alternatives, which is an example of the ‘’bounded rationality’’ position (Sunstein, 2006; Kahan et al., 2012). Attitudinal heterogeneity is caused by differences in individual variation in information processing and impact, which means that information on its own can lead to completely different risk perceptions (Irwin and Wynne, 2003). When it comes to GCC, public uncertainty and scepticism may be influenced by probabilities, expert disagreement (although this is getting less) and model-based uncertainty (Patt, 2007). Whereas most risk study theories claim that knowledge and education are two of the most important predictors for how people perceive risks (Slovic, 2016), this seems to be different regarding GCC (Whitmarsh, 2011). While knowledge and education are meaningful predictors of climate change awareness, climate change sceptics are generally as knowledgeable as non-sceptics (Poortinga et al., 2019; Hornsey et al., 2016) and knowledge has proven to sometimes even have diverging effects depending on people’s political orientation (Kahan et al., 2012; Lee et al., 2015; Malka et al., 2009), due to a variety of possible reasons. Examples include individual factors, such as cognitive dissonance reduction between espoused beliefs and current actions, personal habits, the feeling of helplessness and perceived uncertainty; and social factors, such as a lack of visible governmental or political action, prevailing social norms and a lack of enabling initiatives (Gifford, 2011; Lorenzoni et al., 2007). Nonetheless, education will be included in this study as a control variable, to rule out any potential influences. In addition, Lee et al. (2015) argued that another important predictor of perceived risks of GCC is the understanding or belief that GCC is human-caused. Therefore, the belief that GCC is human-caused will also be included as a control variable.
17 attention (Pidgeon, 2012) and the so-called ´finite pool of worry´ with other pressing factors, such as a recession or war (Whitmarsh, 2011). In recent years, studies have emerged that found indications of abilities, values and beliefs as possible determinants of GCC risk perceptions (Kahan et al., 2011), which is the focus of this research. Corbett and Durfee (2004) found that people with strong pro-environmental values are likely not to change their minds after being exposed to information that presents GCC as controversial. In addition, GCC risk perceptions have been found to be associated by where people fit on cultural values scales (Kahan et al., 2012), which will be discussed hereafter and will be pivotal in constructing the hypotheses.
2.3 Human values
18
2.3.1 Gender differences of human values
Two theoretical typologies of typical feminine and masculine value orientations appear repeatedly in literature on gender-based differences in value priorities (Lyons et al., 2005; Struch et al., 2002). The first distinction involves the expressive person-oriented role, typical of women, and the instrumental or task-oriented role, typical of men (Parsons et al., 2014). The second distinction encompasses the communal orientation, which is typical of women, and the agentic orientation, which is typical of men. The communal orientation points out openness, connections with others, nurturing and cooperation, emphasizing a more social and emotive orientation, whereas the agentic orientation includes isolation, self-protection, repression of emotion and self-assertion, emphasizing a more assertive orientation (Bakan, 1966). Several studies have attempted to discover gender-based differences in particular human values. These studies found women to be more concerned about meaning in life, compassion, personal happiness, absence of inner and interpersonal conflict, religious values, self-respect and love, whereas men were found to be more concerned about materialism and achievements while being more hedonistic and intellectually oriented (Beutel and Marini, 2006; Rokeach, 1973). These differences are key for the development of hypotheses later in this paper. Some studies found lower correlation between gender and values (Struch et al., 2002; Schwartz et al., 2001); however, the consistency of findings in other studies suggests that gender-based differences seemingly exist in human values (Beutel and Marini, 2006; Lyons et al., 2005).
2.3.2 Human values and risk perceptions
19 endorsed patterns of interpersonal relationships which results in different risk perceptions. For example, hierarchical cultures regard technological and industrial risks as opportunities, therefore, as low risk, whereas egalitarian cultures view these risks as threats to their social structure and, therefore, perceive them as riskier (Weber, 2010; Dake, 1991; Douglas and Wildavsky, 1983). Such differences in worldviews have proven to cause diverse and different risk perceptions of GCC (Hulme, 2009; Kahan et al., 2011, 2009; O’Connor et al., 1999).
2.3.3 Human values and risk perceptions of global climate change
People typically rely on their fundamental values when making decisions, which is especially the case for highly controversial and polarizing topics, such as human-induced GCC (McNeal et al., 2014; Lombardi and Sinatra, 2013). GCC related communication that challenges deeply held beliefs of groups or individuals may quickly be rejected. Moreover, the confirmation bias may be an explanation of why people ignore GCC related information that challenges alternative conceptions and why people only focus on evidence that support their own beliefs (Kahan et al., 2012, 2013). Perceptions of major risks, such as GCC, are assumed to be affected by where an individual scores or fits on two value scales: hierarchicalism versus egalitarianism and individualism versus communitarianism (Kahan et al., 2009, 2012). People who score high on the hierarchicalism scale are more inclined to believe that authority is based on characteristics such as gender, class and race and comes from a strict social order. People with more individualistic than communitarian values believe that they are responsible for their own well-being without inference from others or society. When applying this to global warming, hierarchical individuals are expected to be more sceptical about GCC risks because accepting these risks would undermine their trust and belief in commerce and industry. Contrary, egalitarians are assumed to perceive GCC risks more accurately because they view commerce and industry as drivers of several environmental risks that promote individuals over society (Carlton et al., 2015; Kahan et al., 2011, 2012).
self-20 transcending or altruistic values. The reverse is true for self-enhancement or egoistic values (Brown and Kasser, 2005; Corner et al., 2014; de Groot and Steg, 2007). Awareness and severity of consequences are important factors in both models for risk perceptions (de Groot and Steg, 2007; Karpudewan, 2019; Sjöberg, 2000). Especially Schwartz’ theory (1992) is of specific interest to this study and will later be essential in the formulation of hypotheses and serves as dependent variable for the first hypothesis, while scores on his dimensions will be used as independent variable in the final and most important hypothesis, as can later be seen in the conceptual model. The first dimension contrasts openness-to-change versus conservation, and captures conflict between independence of action, thought, readiness and feelings for change (stimulation and self-direction), and values that emphasize resistance to change, order, preservation of the past and self-restriction (tradition, conformity and security). The second dimension contrasts self-enhancement versus self-transcendence and captures conflict between values that emphasize the pursuit of one’s own interests and one’s own relative dominance and success over others (achievement and power), and values that accentuate concern for others’ interest and welfare (benevolence and universalism) (Schwartz, 1992). These two dimensions yield the ten basic values that were put in brackets. These values should not be regarded as good or bad, or better or worse. Rather, these dimensions must be seen as opposites or contrasts. For example, when pursuing achievement values, one might encounter conflicting situations when also pursuing benevolence values. Seeking personal success is contradictory to the aim to enhance the welfare of others. People are expected to differ in the importance they attribute to the ten basic values, but these values are all organized by the same structure of motivational oppositions and compatibilities (Schwartz, 2003; Stern et al., 1998).
21 pro-environmental behaviour and are more likely to believe that the world’s climate is changing (Orru and Lilleoja, 2015; Poortinga et al., 2011).
Since values are part of both theories discussed above, they have the ability to explain variation in risk perceptions across individuals. People are expected to reference to their values when responding to a novel and controversial phenomenon such as GCC (McNeal et al., 2014; Lombardi and Sinatra, 2013). Knowledge may be another factor that is able to predict risk perceptions. Several studies found contradicting evidence concerning the influence of knowledge on risk perceptions. Also, in other limited existing literature on the predictors of risk perceptions of GCC, conflicting results can be found. On the one hand, researchers argue that content knowledge about GCC is a significant predictor of GCC risk perceptions (Shi et al., 2016; Aksit et al., 2018). On the other hand, researches stated that personal values play the major role in shaping one’s risk perception (Corner et al., 2014; McCright and Dunlap, 2011; Kahan et al., 2012; Malka et al., 2009). Psychological research found that people tend to develop behaviours and attitudes that support their behaviour. This idiosyncratic construction of attitudes and behaviours leads to biased cognitive processing which could be a reason for differences in GCC risk perceptions (Irwin and Wynne, 2003). This study does not claim that either of them is right or wrong, rather it focuses on the extent of human values as a predictor for GCC risk perceptions.
22
2.4 Human values of young professionals
Over the past decades, changes in society have resulted in a shift in observed core human values. Both men and women have expressed different value priorities from one generational cohort to the next (Lyons et al., 2005). Generational shifts in work and personal values have increasingly been studied over the last few years. Twenge et al. (2010) segregate today’s workforce into Generation Me (or GenY, Millennials; born 1982-1999), Generation X (born 1965-1981), Baby Boomers (born 1946-1964) and the Silent Generation (born 1925-1945). Obviously, Generation Me is the largest group of young people in the job market. Recruitment of this generation is a top priority for HRM departments at almost all leading organisations in the world (Yeaton, 2008; Twenge et al., 2010; Tapscott, 1999). This study uses the term young professionals when speaking about Generation Me. This means that profession is of no interest for this paper, and age of main interest. Young professionals are found to be more individualistic and self-focused (Sirias et al., 2007; Twenge et al., 2010). Several leading companies have tried to target this new generation by adding amenities that focus on improving work-life balance. Examples include meditation rooms, onsite massages and free weeks off (Twenge et al., 2010; Kaplan, 2010).
23
2.4.1 Young professionals and GCC risk perceptions
Age differences can also be observed for risk perceptions. In general, younger people have significantly higher dread of hazards than older people, which means that they perceive risks as more threatening. This is most likely because of younger people´s feeling of heightened personal exposure to risks (Savage, 1993). This higher level of perceived risk also translates to climate change issues. European and Australian surveys show that younger people exhibit similar or elevated levels of reported interest or perceived risk about GCC as compared to older age groups (European Commission, 2014, 2015; Payne, 2005). Around three-quarter of all ‘young’ respondents in polls in Austria, Germany, Switzerland and the UK reported to regard GCC as a very or fairly big problem (Schneekloth et al., 2010; Corner et al., 2015; Stoll-Kleemann et al., 2001; Hibberd and Nguyen, 2013). Many polls and studies found that younger people or young professionals possess limited conceptual and sometimes even wrong factual knowledge about environmental issues and GCC (Loughland et al., 2003; Thielking and Moore, 2001). This could perhaps be explained by their more intensive use of online resources that more often and easier present exaggerated, inaccurate or even fake messages and information (Lazer et al., 2018). However, despite these low levels of carbon literacy, younger people are less sceptical about and display a greater acceptance of the scientific consensus on GCC (Feldman et al., 2012; Hibberd and Nguyen, 2013). Because younger people are growing up with scientific consensus and certainty about GCC, their perceptions and beliefs about GCC may be less influenced by ideology compared to a general public, including people from older generations (Aksit et al., 2018; Caprara et al., 2006).
24 increased by using self-transcending values while framing messages about GCC (Evans et al., 2013). Also, the willingness to think deeply about GCC and Schwartz’ value openness-to-change are positively linked to engagement with GCC (Sinatra et al., 2012). The relationship between Schwartz’ values, GCC risk perceptions and young professionals will be paramount to the formation of hypotheses as GCC risk perceptions of young professionals will be the dependent variable of this study as can later be seen in the conceptual model.
2.4.2 Gender differences of GCC risk perceptions of young professionals
Research that examined identifiable gender related patterns of GCC risk perceptions found inconsistent results. Whereas early studies (Feather, 1984; Rokeach, 1973) showed remarkably consistent results, recent research has not been able to detect consistent gender patterns over the years (Feather, 2004; Lyons et al., 2005; Smith and Mayer, 2018). A plausible explanation for these recently found inconsistencies is that, similar to human values in general, changes in societies also caused value priorities expressed by both genders to shift. The evolution of gender roles for both family and work, and the increased occupational and educational opportunities for younger women have made it possible for women to express value priorities that were traditionally considered as masculine values, such as power and achievement (Orenstein et al., 2000). Nonetheless, young women were found to be more understanding while also being more tolerant and caring more for the welfare of all people and nature (Schwartz et al., 2001; Lyons et al., 2005). Moreover, younger men and even boys show higher levels of agentic values (focusing more on self-advancement) and lower levels of communal values (focusing less on the maintenance of positive relationships) than do young women (Block et al., 2018). These findings link back to the higher family, as compared to career, oriented values that women possess, which were earlier discussed.
25 the curriculum (Whitmarsh, 2011). Gender and age effects in GCC risk perceptions can also be explained by looking at different motivations for maintaining social structures. Because older age groups are more integrated into existing social and societal orders, they are less likely to support changes that are needed to tackle GCC issues as they have more to lose. Moreover, values do change over the life course. Because people become more politically oriented and politically conservative as they age (Cornelis et al., 2009; Poortinga et al., 2019), political values, some of the strongest socio-political determinants of GCC perceptions, also change (Poortinga et al., 2019).
2.5 Hypotheses
Research on gender-based differences in value priorities yields several results. In general, women were found to be more concerned with feminine value orientations, such as love, personal happiness, meaning in life, absence of inner and interpersonal conflict, compassion and religious values. Men, on the other hand, focus more on masculine value orientations, such as materialism, status, power and achievement (Beutel and Marini, 2006; Prince-Gibson and Schwartz, 2006; Rokeach, 1973; Lyons et al., 2005). These value orientations seem to fit with Schwartz’ model of high-order value types: openness-to-change versus conservation; and self-enhancement versus self-transcendence (Poortinga et al., 2019). These two orthogonal dimensions are universally shared and cover ten lower level value types that encompass all human values. The feminine value orientations seem to suit the openness-to-change-, that includes the motivational basic values stimulation and self-direction, and self-transcendence dimensions, that includes the motivational basic values benevolence and universalism. The masculine value orientations seem to fit with conservation values, that includes the motivational basic values tradition, conformity and security, and self-enhancement dimensions, that includes the motivational basic values achievement and power (Poortinga et al., 2019; Lyons et al., 2005; Schwartz et al., 2001). This leads to the first hypothesis:
Hypothesis 1a: Women hold more self-transcending values Hypothesis 1b: Women hold more openness-to-change values
26 In general, research showed consistent results when it comes to risk perceptions of men and women. Men perceive less risk than women, and they are also involved in more risky activities and behaviours (Weber et al., 2002; Harris, 2006). Women were found to be more sensitive to consequences and uncertainty due to a variety of reasons (Lee et al., 2009; van den Bos et al., 2013). Therefore, women can be characterized as more risk averse than men (van Leijenhorst et al., 2008; Hartog et al., 2002; Holt et al., 2002). As GCC poses several serious and far-reaching threats, it can be expected that women perceive these threats as more serious than do men (Davidson and Freudenburg, 1996; Swim et al., 2018). Moreover, women have been argued to hold more self-transcending and openness-to-change values. Such values are negatively related to GCC scepticism (Poortinga et al., 2011) and positively associated with GCC concern and risk perceptions (Poortinga et al., 2019; Milfont et al., 2015), pro-environmental behaviour, beliefs that the world’s climate is changing and its human cause (Orru and Lilleoja, 2015; Poortinga et al., 2011). On the other hand, men are hypothesized to hold more self-enhancement and conservation values. Such values are negatively associated with GCC risk perceptions and pro-environmental orientations (Poortinga et al., 2019; Milfont et al., 2015). Furthermore, values are learned during the formative period (Lyons et al., 2007), which makes it likely that members of a generational cohort share different values as compared to members of other cohorts (Twenge et al., 2010; Lyons et al., 2005). It is interesting to see whether young professionals, the future policymakers, show the same gender differences when it comes to GCC perceptions because changes in societies possibly also cause value priorities expressed by both genders to shift. Because younger men were found to have higher levels of agentic values (focusing more on self-advancement) and lower levels of communal values (focusing less on the maintenance of positive relationships) than do young women (Block et al., 2018), the second hypothesis is:
Hypothesis 2: Young women show higher levels of GCC risk perceptions than young men
27 influenced this awareness and concern (Shi et al., 2016; Slovic, 2016; Aksit et al., 2018), other studies argued that value orientations are stronger predictors of environmental risk perceptions (Corner et al., 2014; McCright and Dunlap, 2011; Kahan et al., 2012; Malka et al., 2009). Most of the latter studies base their arguments on psychological and sociological qualitative research, while this study aims to obtain results by a quantitative approach. Because value orientations are expected to have a strong influence on environmental risk perceptions, this relationship is also expected for climate change as it is impacting people’s environment. Values are learned during the formative period (Lyons et al., 2007), which makes it likely that members of a generational cohort share different values as compared to members of other cohorts (Twenge et al., 2010; Lyons et al., 2005). It is therefore interesting to see whether young professionals, the future policymakers, show the same relationship between Schwartz’ theory of high-order value types and GCC risk perceptions. Self-transcendence and openness-to-change values are expected to be negatively related to GCC scepticism and positively associated with GCC concern and risk perceptions (Poortinga et al., 2011, 2019). Hence, it is hypothesized that:
Hypothesis 3a: Self-transcendence values will positively influence GCC risk perceptions of young professionals
Hypothesis 3b: Openness-to-change values will positively influence GCC risk perceptions of young professionals
On the other hand, self-enhancement and conservation values are expected to be negatively associated with GCC risk perceptions and pro-environmental orientations (Poortinga et al., 2019):
Hypothesis 3c: Self-enhancement values will negatively influence GCC risk perceptions of young professionals
28
2.6 Conceptual model
Based on the theoretical framework and hypotheses, a conceptual model can be derived, as shown in figure 1. This figure shows the relationship between the independent variable of the first hypothesis, gender, and the dependent variable human values split up into four value orientations as defined by Schwartz (1992). Additionally, the expected positive relationship between females and GCC risk perceptions (H2) is incorporated in this model as well as the final hypotheses (H3) that illustrates the expected relationship between the value orientations and GCC risk perceptions of young professionals.
29
2.7 Communication of GCC issues
Because the next two sections are not building blocks for the conceptual model of this study, they are discussed after it. As GCC awareness depends on education and knowledge, appropriate communication of GCC related risks, implications and issues between policymakers, scientists and the public are a key requirement (Lorenzoni et al., 2005). However, GCC communication is of importance because of the better overview it delivers on how GCC risk perceptions are developed. Also, the implications for businesses are crucial to the worldwide debate concerning GCC.
Every individual values and judges information differently. Therefore, it is impossible to present information related to GCC and its risks in a neutral manner without any form of context Hulme (2009). The way of which information is being framed becomes paramount in the understanding of information by the public (Thaler and Sunstein, 2008). Public perceptions are constantly subject to contradictory information that are all covering separate parts of the truth. The complete set of information that one derives all comes down to intelligence, willingness, access to information and the environment, among other factors. Moreover, the rapid spread of communication channels and devices across the globe is likely to increase GCC awareness among individuals (Lee et al., 2015). Trust is an important issue when it comes to learning about GCC from external sources that are part of these communication channels and devices. When information comes from a trusted source, people are more inclined to discern and incorporate it into their actions and decisions (Slovic, 1999). All in all, it substantially matters how climate change communications are framed and where they come from on how they will be received (Moser, 2010).
2.8 Implications for business
30 employees. Such firms may realize this by making their commitments to “doing good” deeper and more transparent (Howard-Grenville et al., 2014). Millions of older workers are approaching their retirement, generating an enormous challenge for organizations in the coming years. A comparable number of younger people must replace these workers, and this new cohort of employees can be most effectively attracted and managed by having a clear understanding of their values and how these values may differ from the values of previous generations (Twenge et al., 2010). Behavioural intentions are found to be influenced by risk perceptions. If risk perceptions in turn are influenced by human values, these human values could predict actual behaviour and attitudes toward certain organizations (Twenge et al., 2010). The extent and existence of GCC is crucial for organizations because projected climate change, the climate itself and potential climate change mitigation measures will all affect long-term strategic decisions, especially for reinsurance companies, governments and governmental organizations at different levels, and the transportation and energy sector (Weber, 2010).
31
3. Research Methodology
3.1 Data Collection and research ethics
The previously described relationship between human values, gender and GCC risk perceptions of young professionals will be studied by a quantitative approach. The secondary quantitative data used comes from the European Social Survey (ESS) Round 8. The ESS is a biennial cross-European survey that examines interactions between Europe’s diverse and changing institutions and its population. This academically driven survey has been conducted since 2002, with each round containing two extra modules on key social themes on top of the core section that includes a comprehensive set of substantive issues and sociodemographic variables. The latest round, Round 8, was conducted in 2016 and included a module on ‘’public attitudes to climate change’’ as well as modules that include data about age, gender and human values. Round 8 was designed in English over a two-year period with extensive testing, piloting and translation (Schwartz, 1992). Each participating country with populations over two million needed to achieve a minimum effective sample size of 1,500 respondents that is representative of the population aged 15 years and older, while the threshold for countries with populations smaller than two million was 800 respondents. The data collection consisted of strict random probability sampling in order to obtain nationally representative samples. 44,387 interviews were conducted face-to-face in respondents’ own homes in 22 European countries and Israel1. All research was carried in accordance to the ESS Research Ethics committee guidelines and an informed consent form was signed by all participants. Here, ethical principles made sure there was no harm involved for participants as they were protected while their autonomy, dignity and decision-making were respected. Participants give full permission to ESS to make their answers publicly available for research and archiving purposes, in accordance with national laws and the General Data Protection Regulation. Therefore, because of the careful use of publicly available data that participants voluntarily provide, this research is not violating any ethical considerations. Post-stratification weight was used to correct for unequal probabilities of selection and sampling and nonresponse errors (European Social Survey, 2016).
1 Israel is included because of its inclusion in Round 8 of the ESS and because of its potential added value to the
32
3.2 Sample
A total number of 44,378 respondents took part in the ESS, 23,351 women and 21,027 men. This comprises the sample of this study used for hypothesis 1. The second and third hypothesis concern the young professionals, causing the sample size to decrease. Following Twenge et al. (2010), young professionals, or Generation Me, are defined in this study as people born between 1982 and 1999. Of the total sample size of 44,378, 10,823 respondents satisfy this criterion and are making up the sample size to test hypothesis 2 and 3, of which 5,401 are female and 5,422 are male. Because both samples are relatively large and therefore expected to yield significant results, every hypothesis will also be tested for a smaller subset of the sample. 2% of all respondents of the first sample are randomly selected, which yields a total sample of 893 respondents of which 471 are female and 422 are male. For the second sample, 3% of all respondents are randomly selected, yielding a total sample of 325 respondents with 148 females and 177 males.
3.3 Measurement of variables
3.3.1 Independent variables
33 of universalism (reversed) and benevolence (reversed) (α=0.75); self-enhancement consists of achievement and power (α=0.74); conservation consists of conformity, security and tradition (α=0.72); and openness-to-change consists of stimulation (reversed), self-direction (reversed) and hedonism (reversed) (α=0.76). Higher positive values indicate higher scores on self-transcendence and openness-to-change, relative to self-enhancement and conservation respectively.
3.3.2 Dependent variables
The dependent variables of the first hypothesis are self-transcending and openness-to-change values (hypothesis 1a and 1b) and self-enhancement and conservation values (hypothesis 1c and 1d) originating from Schwartz’ model of the relations between values (see figure 1). These values and dimensions are the same that are used as independent variables for the third hypothesis. The questions that were used to measure these values can be found in Appendix A. The dependent variable of the second and third hypothesis is GCC risk perceptions of young professionals. The ESS question that assessed this matter was: “How worried are you about climate change?” while respondents had to choose from 1 (not at all worried) to 5 (extremely worried). For testing these 2 hypotheses, the sample size includes only people born between 1982 and 1999, classified as young professionals in this study.
3.3.3 Control variables
34 of how people perceive risks as well as for GCC awareness, where more education would lead to more awareness and higher GCC risk perceptions (Slovic, 2016; Lee et al., 2015). Therefore, education is included as a control variable to rule out any potential influences between risk perception and education. The question ‘’About how many years of education have you completed, whether full-time or part-time? Please report these in full-time equivalents and include compulsory years of schooling’’ will be used. Thirdly, some studies argue that the more people belief that GCC is human-caused, the more likely they are to perceive GCC as a threat (Lee et al., 2015; Moser, 2010), which is why this variable is included as a control variable. The question ‘’Do you think that climate change is caused by natural processes, human activity, or both?’’ will be used and examined to control for this possible relationship. Respondents could choose from 1 (entirely by natural processes) to 5 (entirely by human activity).
3.4 Empirical data analysis
35
4. Results
4.1 Test of basic assumptions
Several assumptions are checked for in this study. First, the assumption of normality is assessed by examining the skewness and kurtosis as well as by making a P-P plot of the regression standardized residuals. This plot shows a linear relationship between the independent and dependent variables for both the primary and randomly selected samples. Moreover, according to the central limit theorem, a sufficiently large sample size (>200) and a small number of independent variables (<5) will cause the estimates of confidence intervals and p-values to be accurate in any case. Therefore, the assumption of normality is met. Second, the assumption of independent errors is tested by plotting the standardized residuals against the standardized predicted values. The Durbin-Watson scores of these plots are 1.78 (all ages) and 1.93 (young professionals) which fall between the commonly accepted range of 1.5-2.5. Therefore, the assumption of independent observations is met for both samples. Third, the assumption of multicollinearity is tested with the Variance Inflation Factor (VIF). As all tolerance scores are greater than 0.1 and the VIF values are below ten, multicollinearity is not an issue (Field, 2013). Last, the assumptions of homoscedasticity and linearity are assessed by making a scatterplot with standardized residuals which did not show any cone-shaped or odd pattern. For details see Appendix B.
4.2 Descriptive statistics
An overview of the descriptive statistics of the variables used in this study is presented in table 1. For the distribution of the sample per country, please see Appendix C. For the descriptive statistics of the randomly selected samples, please see Appendix D
36 and younger people’s values that accentuate concern for others’ interest and welfare. Scores on OC (M=4.08 SD=0.88; M=4.44 SD=0.78) have increased, indicating that, because these values are reversed, young professionals show higher levels of independence of action, thought, readiness and feelings for change compared to older people. Scores on SE (M=3.42 SD=1.03; M=3.08 SD=0.97) have decreased, indicating that young professionals have a higher average level of the pursuing of one’s own interests and one’s own relative dominance and success over others as compared to older people. Scores on CO (M=2.67 SD=0.82; M=2.82 SD=0.81) have slightly increased, meaning that young professionals display lower average levels of values that emphasize resistance to change and order.
Table 1: Descriptive statistics
Sample all ages Sample young professionals
Min Max Mean SD Min Max Mean SD
Gender 1 2 1.53 .50 1 2 1.50 .50 Year of birth 1916 2002 1967 18.62 1982 1999 1990 5.16 How worried about GCC 1 5 3.01 .935 1 5 3.01 .95 GCC human-caused 1 5 3.42 0.798 1 5 3.50 .80 How religious 0 10 4.50 3.12 0 10 3.78 3.06 Years of full-time education 0 54 13.04 3.85 0 34 13.92 3.07 Self-transcendence (ST) 1 6 4.83 0.73 1 6 4.82 .71 Openness-to-change (OC) 1 6 4.08 0.88 1 6 4.44 .78 Self-enhancement (SE) 1 6 3.42 1.03 1 6 3.08 .97 Conservation (CO) 1 6 2.67 0.82 1 6 2.82 .81
37
4.3 Correlations
Table 2 and table 3 present the correlation analyses that have been performed in order to check for the strength and direction of the linear relationship between the variables. The correlation tables for the randomly selected samples can be found in Appendix E and F.
As the variable gender consists of 1=male and 2=female, the positive and significant correlation between gender and ST for both samples implies that higher values of gender, in other words being female, leads to higher levels of ST, while the negative correlation between gender and OC indicates that the reverse is true for OC.
Table 2: Correlation matrix, sample all ages Gender Year of birth How worried about GCC GCC human-caused How religious Years of full-time educa tion Self-transcen dence (ST) Opennes s-to-change (OC) Self-enhanc ement (SE) Cons ervat ion (CO) Gendera 1 Year of birth -.036* 1 How worried about GCC .055* .043* 1 GCC human-caused -.006 .108* .308* 1 How religious .149* -.182* .034* -.041* 1 Years of full-time education .001 .223* .090* .078* -.106* 1 Self-transcendence (ST) .089* -.012* .227* .104* .081* .092* 1 Openness-to-change (OC) -.066* .305* .098* .055* -.097* .147* .368* 1 Self-enhancement (SE) .084* -.248* .009 -.003 -.011* -.044* -.128* -.451* 1 Conservation (CO) -.053* .157* -.021* .032* -.253* .146* -.442* -.083* .308* 1 Note: N=44,378; * p≤0.05 (2 tailed)
38 Table 3: Correlation matrix, sample young professionals
Gender Year of birth How worried about GCC GCC human-caused How religious Years of full-time educa tion Self-transcen dence (ST) Opennes s-to-change (OC) Self-enhanc ement (SE) Cons ervat ion (CO) Gendera 1 Year of birth .965 1 How worried about GCC .073* .033* 1 GCC human-caused .011 .024* .338* 1 How religious .121* -.040* .021* -.040* 1 Years of full-time education .070* -.280* .137* .090* .008 1 Self-transcendence (ST) .104* -.006 .269* .127* .096* .107* 1 Openness-to-change (OC) -.041* .151* .090* .037* -.039* .025* .401* 1 Self-enhancement (SE) .060* -.074* .059* .047* -.073* .018 -.136* -.402* 1 Conservation (CO) -.062* .055* .001 .049* -.276* .041* -.396* -.126* .380* 1 Note: N=10,824; * p≤0.05 (2 tailed)
a dummy variable: 1=male, 2=female
The positive and significant correlation between gender and SE implies that being female leads to lower levels of SE (because lower scores mean actually higher levels of self-enhancement values), while the negative correlation between gender and CO suggest that the reverse is true for CO for both datasets. Moreover, the same gender variables also suggest that the positive and significant correlation between gender and how worried about GCC means that being female leads to higher levels of GCC concern.
39 worry about GCC and higher GCC risk perceptions. For both samples, the correlations of SE on how worried about GCC are positive, but only for young professionals these are significant. Because lower scores mean actually higher levels of self-enhancement values for the variable SE, higher levels of self-enhancement values lead to less worry about GCC for young professionals. This is different for the variable CO because this variable correlates negatively significantly with worry about GCC in the first sample. Similar to scores on SE, lower scores mean actually higher levels of CO, which implies that this negative correlation means that people that report higher levels of conservation have higher levels of GCC concern.
4.4 T-test results
40 more self-transcending and conservation (rather than the expected openness-to-change) values, while men hold more self-enhancement and openness-to-change (rather than the expected conservation) values.
Table 4: T-test results hypothesis 1
Equal variances assumed; * p≤0.05
In order to analyse whether the average levels of GCC risk perceptions of young women are different from the average levels of GCC risk perceptions of young men, another independent samples t-test was performed. Table 5 shows that the independent samples t-test was significant t(10,368)=-7.50, p=0.000. This means that the average GCC risk perception of young women (M=3.08, SD=0.94) differs from the average GCC risk perception of young men (M=2.94, SD=0.94). Also the independent samples t-test of the randomly selected sample was significant t(323)=-2.04, p=0.042 leading to the same conclusion that the average GCC risk perception of young women (M=3.10, SD=0.82) differs from the average GCC risk perception of young men (M=2.90, SD=0.90). Because the mean of GCC risk perception of women is (slightly) higher than that of men, hypothesis 2 can be confirmed and it can be said that young women show higher levels of GCC risk perceptions than young men.
Male Female
N Mean SD N Mean SD df t-test
41 Table 5: T-test results hypothesis 2
Equal variances assumed; *p≤0.05
4.5 Regression results
The results of the regression analysis can be found in table 6. Seven different models are used for this regression analysis. The results of the randomly selected sample of young professionals can be found in Appendix G. The first model depicts the effects of the control variables on GCC risk perceptions of young professionals. Here, all three variables have the ability to explain the dependent variable on a significant level.
In model 2 the independent variable ST is added in a separate manner to the control variables to see its effect on GCC risk perception and in order to test hypothesis 3a. The results of both samples show that this variable has a significant positive effect (B=.221; p=.000) and (B=.224; p=.000), also when looking at model 6 (B=.253; p=.000), which means that self-transcendence values positively influence GCC risk perceptions of young professionals, thereby accepting hypothesis 3a.
Then, the effect of OC on GCC risk perceptions of young professionals (hypothesis 3b) is tested in model 3. OC has a significant positive effect (B=.077; p=.000) for the larger sample and an insignificant negative effect for the randomly selected sample (B=-.011; p=.827), while the effect is also not significant in model 6 (B=.009; p=.411). These inconclusive results lead to the conclusion that openness-to-change values do not positively influence GCC risk perceptions of young professionals, thereby rejecting hypothesis 3b.
Model 4 tests the effect of SE on the dependent variable. Table 6 and Appendix G show that SE also has a positive significant effect (B=.045; p=.000) and (B=.136; p=.009), which is confirmed in model 6 (B=.057; p=.000). Because higher positive scores mean higher levels of ST and OC, and lower levels of SE and CO, this positive and significant effect indicates a negative relationship between SE and the dependent variable. Therefore, it can be concluded
Male Female
N Mean SD N Mean SD df t-test
Large sample
How worried about GCC 5,171 2.94 .94 5,199 3.08 .94 10,368 -7.50* Randomly selected
sample
42 that self-enhancement values negatively influence GCC risk perceptions of young professionals, thereby accepting hypothesis 3c.
Model 5 tests the effect of CO on the dependent variable and shows that the relationship is not significant and negative for the larger sample (B=-.012; p=.228), while being positive for the randomly selected sample (B=.003; p=.960). Therefore, it cannot be concluded that conservation values negatively influence GCC risk perceptions of young professionals, which means that hypothesis 3d is rejected.
Model 7 finds the same significant independent variables as model 6, however, the adjusted R square drops below .10. This indicates the relatively high explanatory power of (one of) the control variables. The adjusted R square is the highest for model 2 (explaining 17.4% of the variance in GCC risk perceptions of young professionals) where ST is tested, while model 1 shows that the control variables together almost have the same explanatory power as model 3, 4 and 5. This means that ST has the highest ability to explain variances in the dependent variable. Moreover, the control variable GCC human-caused has a remarkably higher unstandardized coefficient in all models as compared to the other control variables. Also the smaller, randomly selected sample shows these results, with the highest coefficients for ST and GCC human-caused, while also a higher coefficient for SE and related adjusted R square can be observed. Because of the surprising finding of the high explanatory power of GCC-human caused and ST, two other regression analyses were performed that included the interaction effect between these two variables. By this, it can be checked whether ST has a significant predictive ability on its own, or whether it is dependent on the ‘GCC human-caused’ variable. The results can be found in table 6 and Appendix G.
Table 6: regression analysis
GCC risk perceptions of young professionals
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Model 8 Model 9
Independent variables ST .221 (.012)* .253 (.015)* .318 (.015)* .233 (012)* .254 (.015)* OC .077 (.011)* .009 (.013) .001 (.014) .009 (.013) SE .045 (.009)* .057 (.011)* .064 (.011)* .056 (.011)* CO -.012 (.011) .071 (.013)* .103 (.013)* .069 (.013)* Control variables GCC human-caused .330 (.011)* .303 (.011)* .327 (.011)* .328 (.011)* .330 (.011)* .294 (.011)* .330 (.011)* .310 (.011)* .295 (.011)* How religious .033 (.003)* .011 (.003) .036 (.003)* .036 (.003)* .030 (.003)* .032 (.003)* .033 (.003)* .031 (.003)* Years of full-time education .107 (.003)* .086 (.003)* .106 (.003)* .107 (.003)* .108 (.003)* .079 (.003)* .107 (.003)* .079 (.003)* Interactions ST X GCC human-caused .042 (.015)* .032 (.015)* Adjusted R square a .127 .174 .133 .129 .127 .183 .090 .168 .184 F-statistic 491.077* 526.239* 383.387* 369.566* 360.828* 317.687* 247.484* 671.906* 277.513*
Below, the conceptual model after the discussed results is displayed. Men showed higher levels of openness-to-change and enhancement, whereas women showed higher levels of self-transcendence and conservation, and also GCC risk perceptions for the sample with young professionals. Religion and especially the belief that GCC is human-caused have an influence of GCC risk perceptions of young professionals, just as transcendence and self-enhancement values. Due to the use of a randomly selected smaller subset of the original larger sample, conservation and openness-to-change did not display significant results or compelling evidence that they influence these GCC risk perceptions of young professionals.
