MASTER THESIS
MSc Marketing Management
DYNAMIC NORMS AND PEOPLE’S LIKELIHOOD TO SWITCH
TO NON-FOSSIL RENEWABLE ENERGY:
THE MEDIATING EFFECT OF RESPONSE EFFICACY
By:
Larasati Sihkristantini
DYNAMIC NORMS AND PEOPLE’S LIKELIHOOD TO SWITCH
TO NON-FOSSIL RENEWABLE ENERGY:
THE MEDIATING EFFECT OF RESPONSE EFFICACY
University of Groningen Faculty of Economics and Business
MSc Marketing Management Master Thesis
First supervisor: dr. J.W. Bolderdijk Second supervisor: dr. M. Keizer Date of submission: 13th January 2020
Larasati Sihkristantini S3156346 Korreweg 103A
9714AE, Groningen, The Netherlands +31616150758
Abstract
Global warming as a collective problem has become a severe problem for the Earth and the whole ecosystem on it, including us, humans. However, people tend to avoid taking action because they believe that their actions will not make a difference. The author intended to address this problem by researching how to make people believe that their actions will make a difference/ are not in vain. Therefore, this paper examined that by exposing people to information that there is an increasing number of people who act pro-environmentally (dynamic norms), people no longer feel that their actions are in vain, which might encourage them to act. The author tested the potential solution in the context of people's likelihood to switch to non-fossil renewable energy using an online survey. The author proposed that dynamic norms increase people’s likelihood to switch to non-fossil renewable energy through increasing people’s sense of response-efficacy. In the end, the results offer evidence that response-efficacy does influence people's likelihood to switch to non-fossil renewable energy. However, the author did not find that exposing people to dynamic norms increases people’s sense of response-efficacy, nor increases people's likelihood to switch to non-fossil renewable energy. In the other hand, the author also finds interesting potential variable that could moderate the relationship of the dynamic norm on response efficacy.
Acknowledgement
I wrote this thesis in order to summarize my Master of Science program in Marketing Management offered by the University of Groningen. I am proud to say that this is the result of my hard work of four years of bachelors and one-and-a-half-year master education. I took great pleasure in writing my thesis in the past semester. It pushed me to develop my research and writing skills and broaden my marketing knowledge.
Table of Contents
Abstract ... 3
Acknowledgement ... 4
Introduction ... 6
Theoretical Background and Hypotheses ... 10
Conceptual Model ... 16 Methodology ... 17 Participants ... 17 Procedure ... 18 Independent Variable ... 20 Variable Measures ... 21 Response Efficacy ... 21
Likelihood to Switch to Non-Fossil Renewable Energy ... 22
Degree of Concerned to Climate Change... 22
Data Analysis ... 23
Results ... 25
Data Cleaning ... 25
Descriptive and Correlation Analysis ... 26
Dynamic Norm and Likelihood to Switch to Non-Fossil Renewable Energy ... 27
Hypothesis Testing ... 27
Additional Analysis ... 29
Analysis with only Indonesian Participants ... 29
Analysis with only non-Indonesian Participants ... 31
Discussion ... 33
Indonesian and non-Indonesian Participants ... 35
Conclusion ... 38
Practical Implications ... 38
Limitations and Future researches ... 39
References ... 41
Appendix A ... 46
Table 1. Descriptive Variables and Correlation Table of The Full Dataset ... 46
Table 2. Robustness Table of The Full Dataset ... 47
Introduction
An environmental issue, like climate change, has become a significant problem. More than 11,000 scientists from around the world evidently and undoubtedly declared that planet Earth had encountered a climate emergency (Ripple et al., 2019) due to the acceleration of climate crisis which is more severe than anticipated, threatening natural ecosystems and the fate of humanity (IPCC, 2018). Fortunately, there is an increase in concern relating to climate change. Governmental bodies also started to make climate emergency declarations. The United Kingdom (UK) is the first country to announce a ‘climate emergency’ situation where they pledged to dedicate more of its resources in combating climate change (Forbes, 2019). Besides the UK, various cities, states, and provinces from across the world have declared climate emergency where they made their own goals in fighting against climate change. This climate emergency declaration shows how critical climate change is that it attracted global attention and concern.
Climate change can be caused by humans and nature (e.g. ice age). This paper focused on human-caused climate change, specifically global warming, which is one of the prominent symptoms of climate change.
Global warming is categorized as a collective problem. A collective problem is a situation when a group of people could benefit from cooperating by doing a particular action which prevents or avoid the harm of a problem, while in the other hand as an individual, people can never have a sufficient incentive to act alone (as cited in Nyborg et al., 2016). Consequently, in order to make noticeable progress in combating global warming, it requires individuals to
cooperate. Electricity is one of the notable contributors to global warming (EIA, 2019). Considering the contribution of electricity to global warming, in this paper, I was interested in analyzing pro-environmental behavior that could reduce the level of CO2 in the atmosphere, which caused by fossil-fueled electricity. Therefore, I decided to focus the current research on the context of people's likelihood to switch to non-fossil renewable energy (e.g., hydro, wind, biomass, and solar). I also chose this context because of the increase in popularity and the cost-competitiveness of renewable energy production and usage around the world (GSR, 2018). Furthermore, Scientists are also insisting on humanity to replace fossil-fuels with non-fossil renewables (Ripple et al., 2019). Thus, this context is indeed very relevant to today's society.
energy, while the others still using fossil-fueled energy, it would not make enough difference to the number of carbon releases to the atmosphere. Another example is when most people still eat meat despite climate change.
The knowledge that the majority is not contributing can be demotivating, and it makes people feel that their efforts are in vain. This was supported by a past study that found that external factors can make people undergo personal change (Scheffer, 2009). And switching to non-fossil renewable energy is an act of personal change. It is implying that what people do can be influenced by other people’s behavior, such as their electricity choice and consumption. Thus, the knowledge that the majority did not use non-fossil renewable energy might leads people to feel reluctant to switch to non-fossil renewable energy.
Furthermore, the process of undergoing personal behavior change is not an easy process to go through. When the cost of undergoing behavior change outweighs the benefits, it might lead people to avoid change, vice versa (Coates, 1988). Moreover, it was also found that people will undertake and pursue a behavior when they believe that they can perform that behavior and when they believe the behavior will produce the desired outcome (Goodwin, 1990). So, people might able to switch to non-fossil renewable energy, but if they feel that they are the only ones who switch, they do not feel that switching will produce the 'desired outcome'.
people are doing the same as them, which can give them the feeling that their actions will not produce the desired outcome.
Looking at the arising problem; therefore, the research question of this paper will be: Are people more likely to act more pro-environmental if they feel their actions are not in vain? And how to make people believe that their actions make a difference?
Considering evidence which found that people are more willing to cooperate when they see others also cooperate (Brekke, Kipperberg & Nyborg, 2010; Chaudhuri, 2011), this paper introduces a solution to, which is by exposing people to information that more and more people are doing the same, people will feel that their actions are not in vain, which in turn could also influence people's behavior.
Theoretical Background and Hypotheses
It was mentioned before that global warming is a collective problem that has become a global challenge. Furthermore, in order to make significant changes to address global warming, people need to take action. However, people possess specific barriers, which often make them feel reluctant to act upon global warming because they believe that their actions will not make any difference. This is supported by past studies which found that when the cost of pursuing behavior change outweighs the benefits, it might lead people to avoid change (Coates, 1988).
Additionally, it was also found that people are unlikely to attempt to change if they do not believe they can, even if they view the change as desirable or beneficial (Chambliss & Murray, 1979; Rotter, 1966; Strecher, McEvoy DeVellis, Becker, & Rosenstock, 1986; as cited in Sparkman & Walton, 2019). Furthermore, it was stated that people would undertake and pursue a behavior when they believe the behavior will produce the desired outcome (Goodwin, 1990). These imply that even when people, in fact, willing and know that a particular behavior change is advantageous, they can avoid change if they believe that the change is going to be in vain or will not produce the desired outcome. The degree of one's beliefs that specific goals or progress will be achieved or a specific threat will be avoided by doing the recommended action is called response efficacy (Swim, Fraser, & Geiger, 2014). So, in other words, people often feel reluctant to act because they have low response efficacy.
believing one’s action will make a difference leads to people taking actions considering that global warming is also a threat.
On the contrary, a low response efficacy or not believing that taking the recommended action will make a difference will cause a psychological disengagement of that individual (Abramson, Seligman, & Teasdale, 1978). A study conducted by Maier and Seligman (1976), implies that people faced by uncontrollable events experience more emotional disruption than those faced by relatively more controllable events. When faced with uncontrollable events, people learned that behavior and outcomes are independent, thus generate emotional effects of uncontrollability. Such events can disrupt one’s emotional balance, which might cause the individual to feel helpless, thus hinder the individual from taking actions.
Therefore, in the context of this research paper, I argue that the increase in response efficacy could make people as an individual to conform to the change in behavior of combating global warming by switching to non-fossil renewable energy because they perceive that their individual efforts could contribute to the collective effort of combating global warming in the society. In other words, people who believe that their actions will make a difference are more likely to switch to non-fossil renewable energy. On the contrary, people who believe that their actions will go in vain are less likely to switch to non-fossil renewable energy.
Hypothesis 1: People with high response efficacy are more likely to switch to non-fossil renewable energy.
consumption, are often still not yet become the norm – a typical behavioral arrangement within a group, backed by a mutual understanding of tolerable actions and maintained through social interactions within that group (Ostrom, 2000) – even though increasing in importance. This is supported by data provided by GSR (2019) that still a minority around the world used non-fossil renewable energy. However, indeed the percentage of people using non-fossil renewable energy is steadily increasing over time. This implies that there is an increase in prevalence in switching to non-fossil renewable energy. Still, as long as the majority does not use non-non-fossil renewable energy, people may feel their actions will go in vain. However, the fact that the majority is not acting does not have to be problematic for people's response efficacy. This is because there may be an increasing minority who use non-fossil renewable energy, which, if made known, can make people feel their actions do make a difference. Thus, exposing people to this information about more and more people use non-fossil renewable energy might counteract the problem of people's beliefs that their actions are in vain, which in turn will influence people's final decision to switch to non-fossil renewable energy.
examining dynamic norms when the existing static norm is negative (still growing minority conforming to the new behavior). Therefore, this paper pertains to the definition of dynamic norms as the information of a growing minority doing a collective change in their behavior.
There is evidence which supported that dynamic norms indeed positively affect people's behavior. Firstly, it was found that people are more likely to conform to the most salient aspect of the norm (Cialdini, Reno, & Kallgren, 1990). Including when the change in collective behavior is itself a salient aspect of the norm, even when the current static norm is in contradiction with the new behavior itself. For example, as mentioned before, when most people are still not using non-fossil renewable energy, there could be an increasing minority who use non-fossil renewable energy. When the information about this 'increasing minority' becomes salient, people will be more likely to switch to non-fossil renewable energy (new behavior), even when the majority are still not using non-fossil renewable energy. Secondly, it was found that participants' interest in limiting their meat consumption has increased after exposure to the dynamic norm (Sparkman & Walton, 2017). It was also proven that people were more engaged in water conservation in response to the dynamic norm (Sparkman & Walton, 2017; Mortensen et al., 2017). Lastly, a recent study also produces a significant effect of dynamic norms towards smoke cessation efforts (Sparkman & Walton, 2019).
Firstly, is the phenomenon of pre-conformity (Sparkman & Walton, 2017). Past studies found that people's perception of norms can be influenced by information beyond the here and now (Shrum, 2002). Make it possible that exposed by dynamic norm lead people to an imagined social world. Also, people's perception of reality can be significantly influenced by the work of fiction to the point that it can utilize behavior change (Paluck, 2009). In the context of this paper, the dynamic norm gives the notion that using non-fossil renewable energy has increased in popularity. Making people believe that more and more people are doing the same as them and they are part of a bigger movement, thus their individual efforts are not in vain anymore. This makes people anticipating that the current norm is going to change in the future, that using non-fossil renewable energy is going to be the new norm in the future. This leads to pre-conformity where people conform to the arising norm before it becomes a norm as if the new behavior, using non-fossil renewable energy, is the current norm.
dynamic norms can provide insight into the increasing number of people using non-fossil renewable energy. When people see that more others are also changing, this makes them reevaluate existing barriers (Sparkman & Walton, 2017) – the feeling of helplessness due to uncertainty, that leads to the belief that their action will go in vain. Also, people may also perceive that the change in behavior becomes more critical, more possible (more worth it, hence will not go in vain), or more appropriate than they had thought before.
Taking into account all factors and supporting pieces of evidence regarding dynamic norms and response efficacy, this paper argues that people witnessing a growing minority conforming to particular behavior and attitude are more likely to have a higher level of response efficacy compared to those who only exposed by information about static norms, without any cues of growth. This is because dynamic norms provide information about the existence of a growing collective effort, which makes combating global warming, which is a collective problem, now seems possible to be done.
Hypothesis 2: People who are put under dynamic norms condition are more likely to have higher response efficacy.
Looking at people who believe that their actions will make a difference (high response efficacy) are more likely to switch to non-fossil renewable energy, and taking into account the positive relation between dynamic norms and response efficacy which argue that people put under dynamic norms condition are more likely to believe that their action will make a difference (higher response efficacy) compared to those who were put under static norms condition. Therefore, it is expected that there is also a positive indirect relationship between dynamic norms and the likelihood of people to switch to non-fossil renewable energy, which mediated by the person's level of response efficacy.
Conceptual Model
Figure 1. Conceptual model
H1: People with high response efficacy are more likely to switch to non-fossil renewable energy. H2: People who are put under dynamic norms condition are more likely to have higher response efficacy.
Methodology
I examined the psychological mechanism in which dynamic norm – the information of a growing minority changing their behavior – about non-fossil renewable energy consumption, compared to a static norm, could encourage personal behavior change via boosting response efficacy. We tested a mediation model where participants put under dynamic norm condition (vs those who were put under static norm condition) have higher response efficacy, which leads them to change their behavior.
Furthermore, due to the problem that this research wants to address – people avoid to take action because they feel that their action will go in vain – we limited participants to those who indicated concern (alertness level) about climate change using a screening question taken from Swim and Geiger (2017). This is because we believe that people who want to take action are those who at least possessed a certain degree of concern toward climate change. On the contrary, people who do not believe in nor care about climate change will not even consider taking any action in the first place. Thus, people who did not meet this requirement were excluded from the analysis.
Participants
Procedure
Participants received a link that directed them to the Qualtrics survey. The survey consisted of two seemingly independent studies. In the first study, participants were given a 'short text' to examined where the dynamic (vs static) norms were imposed. Then, participants were asked to evaluate and summarize the main idea of the text. After that, questionnaires measuring their response efficacy were given. After finishing the first study, participants were directed to the second study. Here, several questions about their daily electricity consumption were asked. Subsequently, a set of questionnaires measuring their likelihood to switch to non-fossil renewable energy was given. Finally, they were asked to indicate their concern level towards climate change.
In this research, a cover story was used to disguise the real purpose of the research to ensure the validity of research findings. Therefore, participants were blind to the hypotheses. I perceive it needed a cover story to avoid the problem that arises when participants are inclined to support the researcher by helping to uphold the noticeable hypotheses when the hypotheses examined are relatively straightforward.
fossil-fueled energy consumption or switching to cleaner non-fossil renewable energy. At the end of the text, the norm statements will be provided (see independent variable section). After reading the text, participants were asked to evaluate the text in terms of its length and clarity to reserve the cover story as well as to serve as indicators that participants indeed perceive the text as sufficiently clear, thus understand the text. Then, participants were asked to summarize the main idea of the text to guarantee an attentive reading to show further that they understand, and the purpose of the text was indeed reached. After that, participants were asked to fill the mediator's measures, which will indicate their response efficacy and being told that they have finished the first study.
Independent Variable
Participants were randomly assigned under the condition of either dynamic norm or static norm condition. In this experimental study, the condition manipulation was based on manipulation done by Mortensen et al. (2017), which also manipulate dynamic norm and static norm, with several adjustments to fit the context of this research. As explained in the previous subsection, participants, regardless of their condition, were given the same “short text” containing facts about climate change, its causes, and what individuals can do to fight climate change, then followed by statements containing the norm manipulation.
For participants who were put under dynamic norm condition ( dummy variable coded as ‘1’ in the dataset), the text contains information which shows a growing number of people in the society who successfully undergo personal change by conforming to a more mindful electricity consumption behavior, specifically using non-fossil renewable energy. Participants’ who put under this condition read,
“Interestingly, the Renewables Global Status Report (GSR) from REN21 (2019) has found that 26% of people globally have switched to renewable energy sources. This number has increased from 22% in 2013. Showing an 18.2% growth rate in the course of five years.”
While for those who put under a static norm (dummy variable coded as ‘0’ in the dataset), will not be given any indication of growth in the number of people in the society who use non-fossil renewable energy. Participants’ who put under this condition read,
“Interestingly, the Renewables Global Status Report (GSR) from REN21 (2019) has found that 26% of people globally are currently using renewable energy sources.”
Variable Measures Response Efficacy
To assess participants’ response efficacy, we used measurements used by Hornsey et al. (2015), which measure the perception of individual control. Therefore, a three-item Likert scale was designed to assess participants' response efficacy towards addressing climate change. All three of the items will be measured on a seven-point Likert scale ranging from 1=" Strongly disagree" to 7=" Strongly agree". (1) "I believe my actions have an influence on climate change." (2) "It is hard to imagine that my individual action can make a difference with respect to addressing climate change" (reverse coded). (3) “There is little point in me taking action against climate change because so many others will not” (reverse coded)
Likelihood to Switch to Non-Fossil Renewable Energy
To assess the likelihood to experience a personal change in the context of non-fossil renewable energy consumption, we used measurements used by Ferguson and Branscombe (2010), which measure willingness to conserve energy with several adjustments to fit the context of this research. Therefore, a three-items Likert scale was designed to assess participants' likelihood to switch to non-fossil renewable energy. All three of the items will be measured on a seven-point Likert scale ranging from 1=" Strongly disagree" to 7=" Strongly agree". (1) “I intend to switch to a utility that only uses renewable energy sources for energy production.” (2) “I intend to install solar/wind power at home.” (3) “I intend to purchase renewable energy through 'Renewable Energy Certificates' (RECs), which are a market-based instrument that represents the property rights to the environmental, social, and other non-power attributes of renewable electricity generation.”
Correlation analysis showed that all items significantly correlate with each other; The first and second items (r= .450, p<001), the first and third items (r= .512, p<001), and the second and third items (r= .574, p<001). Furthermore, Reliability analysis on all three items was high with Cronbach’s Alpha of .756 (M=4.489, SD=1.157). This implies that the items used to measure people's likelihood to switch to non-fossil renewable energy were reliable and measure the same construct. Thus, all three items can be sum up to measure respondents’ likelihood to switch to non-fossil renewable energy.
Degree of Concerned to Climate Change
Alarmed: I am very concerned about climate change and think the government and individuals need to act now. (2) Concerned: I am concerned and think we need to take action, but we have time to decide what the appropriate responses should be. (3) Cautious: I suspect that climate change is happening, but I am not certain. We have time to make careful decisions about when and whether to respond. (4) Disengaged: I have not really thought much about climate change. (5) Doubtful: I suspect that climate change is NOT happening, but I am not certain. I am concerned more about overreacting to climate change. (6) Dismissive: “I do not believe climate change is occurring and certainly do not think humans have caused it. So, I’m not motivated to take or support action to address it.”
Out of 174 participants (M=1.59, SD=0.75), 92 (52.9%) people indicated themselves as ‘Alarmed’, 66 (37,9%) people indicated themselves as ‘Concerned’, 14 (8.0%) people indicated themselves as ‘Cautious’, 1(0.6%) people indicated themselves as ‘Disengaged’, and 1 (0.6%) people indicated themselves as ‘Dismissive’.
Data Analysis
change (Swim & Geiger, 2007). After that, the duration in which each participant had in filling out the survey was also considered. Outliers in participants’ duration were checked to discover if there were any participants who took too long (or too fast) in filling out the survey and were excluded from the analysis. It was done to assume the normality of the data distribution. Outliers in participants’ duration were checked using a boxplot method. Lastly, participants who failed at least two out of three indicators that determine their sufficiency (see result section) were deemed insufficient to be included in the analysis, thus removed.
Results
Data CleaningBesides degree of concern level to climate change and the duration of the survey participation, I adopted three indicators to determine whether participants will be excluded or included in the analysis; The first indicator was participants should not indicate the text clearness as extremely unclear nor somewhat unclear because when they perceived the text as unclear, I believe the purpose of the text did not reach the participants. The second indicator was the presence of adequate/appropriate summary of the text, the summary of all participants regardless of what the condition they put under should contain at least the idea that the text is about climate change, the cause of climate change, and what people can do to address climate change to show that they really read and understand the text. In addition, participants put under dynamic norms should contain an indication of people switching to non-fossil renewable energy. The last indicator was the attention check question, which instructs participants to choose 'somewhat disagree'. The participant will be excluded when they failed at least two out of three indicators. I use this practice to avoid the unnecessary exclusion of participants.
least two out of the three indicators that determined whether participants will be excluded or included in the analysis. Hence, 118 participants were included in the analysis.
Descriptive and Correlation Analysis
The final sample consists of 45 (38.1%) male and 73 (61.9%) female, with an average age of 25 (M= 25.4, SD=7.36). These participants were between the age of 17 and 57 years old. The distribution of participants who were exposed to dynamic norms condition is 50 (42,4%) participants, and the static norms condition is 68 (57.6%) participants. The majority of participants were Indonesian (82.2%, N=97), while the others are a combination of a wide range of nationalities with 21 participants altogether (17.8%). Correlation and reliability analysis was done and reported in the methodology section (see measure subsection in the previous section). Moreover, the VIF value of each variable in the model was less than three, which indicates that there was no indication of a multicollinearity problem.
Dynamic Norm and Likelihood to Switch to Non-Fossil Renewable Energy
Figure 2. The conceptual model without mediator with effect presented
Before testing the hypotheses, I ran an independent sample t-test to test whether there were differences in people’s likelihood to switch to non-fossil renewable energy between dynamic norms and static norms conditions’ means, ignoring the mediator variable – response efficacy. The result shows an insignificant difference in the scores between dynamic norms (N=50, M=4.44, SD=1.15) and static norms (N=50, M=4.55, SD=1.17) conditions; t(116)=-.519, p=.605. This result suggests that there is no significant difference between the two conditions.
Hypothesis Testing
Figure 3. The conceptual mediation model with effect presented
Regardless of the insignificant result produced in the previous subsection, I proceed to test the hypotheses. This is because it is still possible for mediation to still exist in the absence of a significant total effect (Kenny, 2018). In order to analyze all hypotheses, the PROCESS procedure by Hayes (2013) was used. The first hypothesis examined whether or not participants’ response
efficacy influences their likelihood to switch to non-fossil renewable energy. To test this effect, I controlled for dynamic norm condition (independent variable). This mean, both dynamic norm and response efficacy were used as predictors while likelihood to switch to non-fossil renewable energy was used as the criterion variable. As illustrated in Table 2, controlling for dynamic norm variable (B=.098, t=.459, p=.647), response efficacy produced a positive (B=.188, t=1.992, p=.0488) and significant result (R2=.036, F(2,115)=2.121, p=.0488). This shows that the more people perceive that their individual efforts could (in any way) influence global warming, the higher their likelihood to switch to non-fossil renewable energy. This implies that a unit increase in participants’ response efficacy will lead to an increase of .188 in their likelihood to switch to non-fossil renewable energy. Thus, hypothesis 1 is supported.
For hypothesis 2, I argued that participants put under dynamic norm condition (M=.42, SD=.496) would be more likely to have higher response efficacy relative to those who were put under static norm condition. The result shows that the dynamic norm condition had a positive (B=.075, t=.359, p=.721) but insignificant coefficient (R2=.001, F(1,116)=.129, p=.721). This implies that participants who were put under dynamic norm condition did not generate different answer about their response efficacy’s level compared to those who were put under static norm condition. Therefore, hypothesis 2 is not supported.
.0996) (see Table 3). This means that there is no mediating effect on the tested model. Therefore, Hypothesis 3 is also not supported.
Additional Analysis
I conducted an additional analysis considering that the majority of participants are Indonesian. I split the dataset into two separated groups; One that consists of Indonesian participants (82.2%, N=97) and the other that consist of non-Indonesian participants (17.8%, N=21) to analyze whether there are differences between the two datasets. I followed the same steps to test all hypotheses from both datasets.
Analysis with only Indonesian Participants
Figure 4. The conceptual mediation model of Indonesian Participants with effect presented
First, I ran an independent sample t-test to test whether there were differences in people’s likelihood to switch to non-fossil renewable energy between dynamic norms and static norms conditions’ means, without the mediator variable – response efficacy. This time, it produces a
different outcome than the one that consists of the full dataset. The result shows a marginally significant difference in the scores between dynamic norms (N=43, M=4.48, SD=.96) and static norms (N=54, M=4.44, SD=1.16) conditions; t(95)=-1.777, p=.079. This result suggests that there could still be a significant difference in people’s likelihood to switch to non-fossil renewable energy between those who were put under dynamic norm condition and those who were put under static norm condition for Indonesian participants.
Second, the PROCESS procedure by Hayes (2013) was used to examine the three hypotheses. For the first hypothesis, both dynamic norm and response efficacy were used as predictors while likelihood to switch to non-fossil renewable energy was used as the criterion variable. Controlling for dynamic norm variable (B=.366, t=1.656, p=.101), response efficacy was indeed had a positive (B=.110, t=1.066, p=.289), but insignificant (R2=.044, F(2,94)=2.149, p=.289) towards likelihood to switch to non-fossil renewable energy. This shows that Indonesian with high response efficacy did not report a different answer than those with low response efficacy. Thus, hypothesis 1 is rejected.
For hypothesis 2, the result was aligned with the result from the dataset containing all participants. It shows that dynamic norm condition (M=.44, SD=.499) had a positive (B=.225, t=1.029, p=.306) but insignificant coefficient (R2=.011, F(1,95)=1.059, p=.306). This implies that Indonesians put under dynamic norm condition did not generate different answers about their response efficacy’s level compared to those who were put under static norm condition. Therefore, hypothesis 2 is not supported.
(B=.029 with indirect effect interval of -.0494 until .1242). This means that there is no mediating effect on the tested model.
Analysis with only non-Indonesian Participants
Figure 5. The conceptual mediation model of non-Indonesian participants with effect presented
Second, the PROCESS procedure by Hayes (2013) was used to examine the three hypotheses. For the first hypothesis, both dynamic norm and response efficacy were used as predictors while likelihood to switch to non-fossil renewable energy was used as the criterion variable. Controlling for dynamic norm variable (B=-1.460, t=-2.780, p=.012), response efficacy was indeed had a positive (B=.174, t=.951, p=.354), but insignificant (R2=.376, F(2,18)=5.430, p=.354) towards likelihood to switch to non-fossil renewable energy. This shows that non-Indonesian with high response efficacy did not report a different answer than those with low response efficacy. Thus, hypothesis 1 is rejected.
For hypothesis 2, the result shows that dynamic norm condition (M=.33, SD=.483) had a negative (B=-.762, t=-1.225, p=.236) but insignificant coefficient (R2=.073, F(1,19)=1.501, p=.236). This implies that non-Indonesians put under dynamic norm condition did not generate different answer about their response efficacy’s level compared to those who were put under static norm condition. Therefore, hypothesis 2 is also not supported.
Discussion
This paper discussed how being exposed to the dynamic norms could be used to encourage people to act more sustainably and more pro-environmentally. The main goal of this research was to examine the effect of dynamic norms on people’s likelihood to switch to non-fossil renewable energy which was mediated by people’s response efficacy.
Maddux & Rogers (1983) found that high response efficacy leads to active responses to threats, while a low response efficacy may cause a psychological disengagement of that individual (Abramson, Seligman, & Teasdale, 1978). In addition, there are also these past researches which found that people will pursue a behavior when their level of response efficacy towards that behavior is high (Goodwin, 1990; Geiger, Swim, & Fraser, 2017). This paper wanted to examine whether this phenomenon can be applied in the context of switching to non-fossil renewable energy (hypothesis 1). In the end, this paper managed to replicate these findings in the context of switching to non-fossil renewable energy. This paper suggests that people who believe that their individual actions could influence global warming are more likely to switch to non-fossil renewable energy.
A possible explanation for the insignificant result could be because of the items used to measure variable response efficacy because a mediation analysis strongly depends on the manipulation and the measurement of the mediating variable (Cohen, West, & Aiken, 2014). The items were taken from Hornsey et al. (2015) measure the perception of individual control. They measure whether people believe their actions, in a rather general context, can influence climate change. The 'actions' asked do not specifically refer to the act of switching to non-fossil renewable energy. This could explain why this paper, despite numerous evidence from existing past papers, failed to find evidence that the manipulation of the dynamic norm (vs static norm) lead to higher response efficacy. Therefore, I believe that changing the items measuring response efficacy to fit better to the context of switching to non-fossil renewable energy could generate more accurate findings, for both the effect of dynamic norms on response efficacy and the effect of response efficacy on people’s likelihood to switch to non-fossil renewable energy.
Furthermore, I intended to examine the different effect of the dynamic norm compared to the static norm. Hence in this paper, I did not include a control condition – where participants were not given any normative information regarding non-fossil renewable energy consumption – as one of the independent variable manipulations. However, because of the absence of the control variable, this paper cannot clearly identify the effect of the tested model. This is because by including a control condition, this could minimize the effects generated from variables/factors other than the independent variable, in this context is dynamic norms (vs static norms) condition. In other words, in this paper, there is still a possible effect generated from other variables/factors other than dynamic (vs static) norm that should be eliminated (controlled) in order to see the true effect of the tested model. Therefore, including a control condition in the research could produce more accurate and reliable results.
Because of the second hypothesis was not supported, this research failed to find any mediating effect of response efficacy between dynamic norms and people’s likelihood to switch to non-fossil renewable energy (hypothesis 3). However, as mentioned before, this is probably because of the inaccuracy in measuring the intended construct of response efficacy and the possibility of the manipulation of dynamic and static norms being not salient enough. Therefore, by incorporating the mentioned recommendations could produce a more accurate result.
Indonesian and non-Indonesian Participants
non-Indonesian sample size seems to be problematic because the sample size used in this dataset is not enough to detect a medium-size effect. Insufficient sample size may lead to an inaccurate representation of the population, lower chance of significance, and cause difficulties in determining whether the result comes from real effect or just random variation. A post hoc power analysis using the G*power (Faul, Erfelde, Bucnhner, & Lang, 2009) (f2=.15, a=.05, N=21) was performed, the result indicated that this sample size possessed only 0.28 effect size. However, since the Indonesian participants are sufficient, I still split the dataset to see whether there could be additional useful information that can be generated from the new separated datasets.
Despite of past studies that give evidence of the influence of dynamic norms to people's behavior in various contexts, such as, meat consumption (Sparkman & Walton, 2017), water conservation (Sparkman & Walton, 2017; Mortensen et al., 2017), and smoke cessation effort (Sparkman & Walton, 2019), with the full dataset, this paper failed to produce significant evidence that in line with those papers. However, interestingly, when the dataset was split into two new separated datasets, it shows two significant, but conflicting results. The dataset containing Indonesian participants indicated a positive effect while the other dataset with non-Indonesian participants indicated a negative effect.
the indication of growth over time in number of people switching to non-fossil renewable energy formed a pattern of possible future trend, and Indonesian with holistic way of thinking catch this ‘possible future trend’ that lead them to pre-conformity, taking into account that Indonesians’ behavior is very easily influenced by trend. In contrary, due to their analytic way of thinking, non-Indonesian participants tend to think thoroughly as to why an element or an event happened, this could even make them sceptical. The difference in effects generated from Indonesian and non-Indonesian participants could also be the reason for the insignificant result generated from the full dataset. However, due to the low statistical power possessed by non-Indonesian participants, this paper cannot draw any conclusion because insufficient sample size may lead to an inaccurate representation of the population. However, if it's true, then the different result might shed new light on a potential variable that could influence the model. Therefore, taking into account nationality or cognitive style as a moderator to be included in the model (changing the model into a moderated mediation model) could produce more accurate results.
Conclusion
As global warming has become more and more severe, the importance of individual effort on combating global warming, specifically individuals’ likelihood to switch to non-fossil renewable energy, is increasing. However, people might avoid doing any action because of their beliefs that their individual effort will not make a difference. Looking at existing past studies that have been mentioned before, this research predicts that dynamic norms could influence people's likelihood to switch to non-fossil renewable energy through increasing people’s response efficacy as the mediator. However, this paper only managed to find evidence for the relationship of response efficacy and people’s likelihood to switch to non-fossil renewable energy. It means that the higher one's beliefs that their effort can influence global warming, the higher the likelihood to switch to non-fossil renewable energy.
Even though this study failed to find evidence of the effect of dynamic norms on response efficacy and people's likelihood to switch to non-fossil renewable energy, its shows different effect of dynamic norm on people’s likelihood to switch to non-fossil renewable energy (while ignoring the mediator) when the dataset was split into separated group of Indonesian and non-Indonesian participants, dataset containing Indonesian participants indicate a positive effect, while dataset containing non-Indonesian participants indicate a negative effect. However, taking into account the low statistical power possessed by non-Indonesian participants, this paper cannot draw any conclusion from the dataset containing the non-Indonesian participants. Nevertheless, the different result might shed new light on a potential variable (nationality or cognitive style) that could moderate the model.
Practical Implications
existing research that also supports the relationship of response efficacy and people's behavior change. Moreover, this paper also might find a potential variable (nationality or cognitive style) that could moderate the influence of dynamic norms on people's behavior.
Furthermore, this paper examined a context related to pro-environmental behavior, where, until now, there was not yet examined (people's likelihood to switch to non-fossil renewable energy). This would help researchers to more effectively encourage effort on addressing global warming, even climate change, which is becoming more and more severe nowadays. This would also benefit both the government and private parties who have the same goal of increasing the production, utilization, and adoption of non-fossil renewable energy while reducing the fossil-fueled energy consumption.
Limitations and Future researches
cognitive style (Indonesians and non-Indonesian, which mostly consist of Europeans). Since this paper did not include any variable (nationality or cognitive style) that might act as a moderator in the initial conceptual model, this could become a possible explanation for the insignificant outcome between the independent and the dependent variable tested from the full dataset. And also considering the small sample size of the non-Indonesian participants, therefore, including nationality or cognitive style as a mediator with a more appropriate sample size for the non-Indonesian participants is encouraged.
to non-fossil renewable energy or not, stating that there will be no additional cost considering that price could strongly influence people's purchase decisions (Safitri, 2018).
References
Abramson, L., Seligman, M., & Teasdale, J. (1978). Learned helplessness in humans: Critique and reformulation. Journal of Abnormal Psychology, 87(1), pp.49-74.
Brekke, K., Kipperberg, G. & Nyborg, K. (2010). Social Interaction in Responsibility Ascription: The Case of Household Recycling. Land Economics, 86(4), pp.766-784.
Chambliss, C., & Murray, E. (1979). Efficacy attribution, locus of control, and weight loss. Cognitive
Therapy and Research, 3(4), pp.349-353.
Chaudhuri, A. (2011). Exp. Econ. 14, 47.
Choi, I., Koo, M., & Jong An Choi. (2007). Individual Differences in Analytic Versus Holistic Thinking. Personality And Social Psychology Bulletin, 33(5), 691-705. doi: 10.1177/0146167206298568
Cialdini, R., Reno, R., & Kallgren, C. (1990). A focus theory of normative conduct: Recycling the concept of norms to reduce littering in public places. Journal of Personality And Social
Psychology, 58(6), 1015-1026. DOI: 10.1037/0022-3514.58.6.1015
Cohen, P., West, S. G., & Aiken, L. S. (2014). Applied multiple regression/correlation analysis for the behavioral science. Psychology Press
Eia.gov. (2020). EIA - Annual Energy Outlook 2019. [online]
Faul, F., Erdfelder, E., Buchner, A., & Lang, A. (2009). Statistical power analyses using G*Power 3.1: Tests for correlation and regression analyses. Behavior Research Methods, 41(4), pp.1149-1160.
Ferguson, M., & Branscombe, N. (2010). Collective guilt mediates the effect of beliefs about global warming on willingness to engage in mitigation behavior. Journal of Environmental
Psychology, 30(2), 135-142. DOI: 10.1016/j.jenvp.2009.11.010
Forbes.com. (2019). Climate Emergency Declarations: How Cities Are Leading The Charge. [online]
Geiger, N., Swim, J., & Fraser, J. (2017). Creating a climate for change: Interventions, efficacy and public discussion about climate change. Journal Of Environmental Psychology, 51, 104-116. DOI: 10.1016/j.jenvp.2017.03.010
Goodwin, D.L., Insull, W., Russell, M., & Probstfield, J. (1990). Predicting Adherence to Prescribed Regimens Using the Health Perceptions Questionnaire (HPQ). In: Shumaker SA (ed.) The Handbook of Health Behavior Change. New York, NY: Springer Publishing Company, 64-83.
Hayes, A. F. (2013). An introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York, NY: Guilford Press.
about climate change. Journal Of Environmental Psychology, 42, 57-65. DOI: 10.1016/j.jenvp.2015.02.003
Ipcc.ch. (2018). Global Warming of 1.5°C: An IPCC Special Report — IPCC site. [online]
Ipcc.ch. (2019). Special Report on Climate Change and Land — IPCC site. [online]
Maddux, J., & Rogers, R. (1983). Protection motivation and self-efficacy: A revised theory of fear appeals and attitude change. Journal of Experimental Social Psychology, 19(5), 469-479. DOI: 10.1016/0022-1031(83)90023-9
Maier, S., & Seligman, M. (1976). Learned helplessness: Theory and evidence. Journal Of
Experimental Psychology: General, 105(1), 3-46. DOI: 10.1037/0096-3445.105.1.3
Mortensen, C., Neel, R., Cialdini, R., Jaeger, C., Jacobson, R., & Ringel, M. (2017). Trending Norms: A Lever for Encouraging Behaviors Performed by the Minority. Social Psychological And
Personality Science, 10(2), 201-210. DOI: 10.1177/1948550617734615
Nyborg, K., Anderies, J., Dannenberg, A., Lindahl, T., Schill, C., & Schluter, M. et al. (2016). Social norms as solutions: Policies may influence large-scale behavioral tipping.
Ostrom, E. (2000). Collective Action and the Evolution of Social Norms. Journal Of Economic
Perspectives, 14(3), 137-158. DOI: 10.1257/jep.14.3.137
Paluck, E. (2009). Reducing intergroup prejudice and conflict using the media: A field experiment in Rwanda. Journal of Personality and Social Psychology, 96(3), pp.574-587.
RENEWABLES 2019 GLOBAL STATUS REPORT. (2019). Retrieved from
Rotter, J. (1966). Generalized expectancies for internal versus external control of reinforcement. Psychological Monographs: General and Applied, 80(1), pp.1-28.
Safitri, I. (2018). The Influence of Product Price on Consumers’ Purchasing Decision. Review of
Integrative Business and Economics Research, 7, 328-337.
Scheffer, M. (2009). Critical Transitions in Nature and Society. Princeton Univ. Press, Princeton, NJ.
Shrum, L. J. (2002). Media consumption and perceptions of social reality: Effects and underlying processes. In J. Bryant & D. Zillmann (Eds.), Media effects: Advances in theory and research (2nd ed., pp. 69–95). Mahwah, NJ: Erlbaum.
Sparkman, G., & Walton, G. (2017). Dynamic Norms Promote Sustainable Behavior, Even if It Is
Counternormative. Psychological Science, 28(11), 1663-1674. DOI:
10.1177/0956797617719950
Sparkman, G., & Walton, G. (2019). Witnessing change: Dynamic norms help resolve diverse barriers to personal change. Journal Of Experimental Social Psychology, 82, 238-252. DOI: 10.1016/j.jesp.2019.01.007
Strecher, V., McEvoy DeVellis, B., Becker, M., & Rosenstock, I. (1986). The Role of Self-Efficacy in Achieving Health Behavior Change. Health Education Quarterly, 13(1), pp.73-92.
Swim, J., & Geiger, N. (2017). From Alarmed to Dismissive of Climate Change: A Single Item Assessment of Individual Differences in Concern and Issue Involvement. Environmental
Communication, 11(4), 568-586. DOI: 10.1080/17524032.2017.1308409
Appendix A
Table 1. Descriptive Variables and Correlation Table of The Full Dataset
Number of Observations, Means. Standard Deviations, and Bivariate Correlations for the Research Variables
Variables N Mean SD 1 2 3 4 5 1. Gender 118 1.62 .488 1 2. Age 118 25.41 7.356 -.121 1 3. Dynamic norms 118 .42 .496 .002 -.078 1 4. Response Efficacy 118 5.130 1.123 -.112 .049 .033 1 5. Likelihood to switch 118 4.489 1.157 -.041 .019 .048 .184* 1
Table 2. Robustness Table of The Full Dataset
Robustness Table
Variables Response Efficacy
(M)
Likelihood to Switch (DV)
Independent Variable
Dynamic Norms (IV) .075 .098
(.210) (.214) Response Efficacy (M) .188* (.094) Constant 5.098*** 3.483*** (.137) (.501) Observations 118 118 F-statistic .129 2.121 R2 .001 .036
Table 3. Mediation Result Table of the Full Dataset
Mediation Result
Variables B SE t p
Dynamic Norms (DV) to Response Efficacy (M) .075 .210 .359 .721
Constant 3.483 .501 6.954 .000
Dynamic Norms .098 .214 .459 .647
Response Efficacy .188 .094 1.992 .049
Bootstrap
Effect SE LL 95% CI UL 95% CI
Bootstrap results for indirect effect
Likelihood to switch .014 .045 -.090 .100
Dynamic Norms and People’s Likelihood to
Switch to Non-Fossil Renewable Energy:
The Mediating Effect of Response Efficacy
Master Thesis Defense, 3
rdFebruary 2020
MSc Marketing Management
Introduction
The acceleration of climate crisis which is more severe than
anticipated
2 Global warming as
collective problem
The CO2 level in our atmosphere in 2018 is the highest it has been in at
least 3 million years (NOAA climate.gov, 2019)
Electricity is one of the notable
contributors to global warming (EIA, 2019).
focus to the context of people's likelihood to switch to non-fossil
renewable energy
the majority is still NOT contributing and it makes people
“
Are people more likely to act more pro-environmental
if they feel their actions are not in vain?
And how to make people believe that their actions
make a difference?
A high response efficacy leads to people taking actions
4
Theoretical Background
Hypothesis 1:
People with high response efficacy are more likely to switch to non-fossil renewable energy
Low response efficacy 1. How to make people to act more pro-environmental?
Reluctant to Act If the act will not produce the desired outcome,
even when they are actually want to and know that a particular act is advantageous.
Response efficacy “The degree of one's beliefs
that specific goals or progress will be achieved or
a specific threat will be avoided by doing the recommended action”
5
Theoretical Background
Hypothesis 2:
People who are put under dynamic norms condition are more likely to have higher response efficacy
2. How to influence people’s response efficacy?
Dynamic normscan provide insight into the increasing number of people who use non-fossil renewable energy
Dynamic norms “the information of a growing minority doing a collective change in their behavior”
(Sparkman & Walton, 2019)
Pre-conformity where people conform to the arising norm before it becomes a norm as if the new behavior is
the current norm.
Feeling of helplessness
caused by the uncertainty of not knowing whether others also doing the
Conceptual Model & Hypotheses
H1: People with high response efficacy are more likely to switch to non-fossil renewable energy.
H2: People who are put under dynamic norms condition are more likely to have higher response efficacy.
H3: Dynamic norm has an indirect positive effect on people’s likely to switch to non-fossil renewable energy through response efficacy
118 Participants
7
Methodology & Analysis
M=25.4
years old 45 Males and 73 Females 21 non-Indonesian97 Indonesian
Online Survey using Qualtrics
◎
Mediation analysis◎
Random assignment◎
Limited to those who indicated concern about climate change (Swim and Geiger, 2017)◎
A power analysis using G*Power with a medium effect size (f2=.15, a=.05, power level of .8) -> a minimum of 68 participants was needed◎
Independent sample t-testSurvey Flow
Link
8
Manipulation(Mortensen et al., 2017)
Dynamic norm: “Interestingly, the Renewables Global Status Report (GSR) from REN21 (2019) has found that 26%
of people globally have switched to renewable energy
sources. This number has increased from 22% in 2013. Showing an 18.2% growth rate in the course of five years.”
Static norm:“Interestingly, the Renewables Global Status Report (GSR) from REN21 (2019) has found that 26% of
people globally are currently using renewable energy
sources.” Short text Eval & Summary RE Additional questions DV Alertness
Response Efficacy(Hornsey et al., 2015)
(1) I believe my actions have an influence on climate change."
(2) It is hard to imagine that my individual action can make a difference with respect to addressing climate change"
(reverse coded).
(3) There is little point in me taking action against climate change because so many others will not”(reverse coded)
(Ferguson and Branscombe, 2010) Likelihood to Switch
(1) “I intend to switch to a utility that only uses renewable energy sources for energy production.”
(2) “I intend to install solar/wind power at home.”
9
Result & Discussion
Dynamic Norm Response Efficacy Likelihood to switch to non-fossil renewable energy Dynamic Norm Likelihood to switch to non-fossil renewable energy
(p= .605)
H3: (indirect B=.014, ns) (direct B=.098, ns) H3: not supportedH2: not supported H1: Supported
10 Dynamic Norm Response Efficacy Likelihood to switch to non-fossil renewable energy Dynamic Norm Likelihood to switch to non-fossil renewable energy Dynamic Norm Response Efficacy Likelihood to switch to non-fossil renewable energy Dynamic Norm Likelihood to switch to non-fossil renewable energy (p<.10) H3: (indirect B=.014, ns) (direct B=.366, ns) (p<.01) H3: (indirect B=.014, ns) (direct B=-1.460, p<.05)
“It is widely accepted that Asians hold a holistic way of thinking while Westerners have the tendency to have an analytic way of thinking.”
(Munro, 1985; Nakamura, 1964; 1985; Needham, 1962; as cited in Choi, Koo, An Choi, 2007).
“
Limitation
11
Future Research
Inaccurate response efficacy measurement Not salient manipulation
Use summary as manipulation check The absence of control condition
Consist of participants who might have very different cognitive style
Small sample size for non-Indonesian participants
Using online survey
More in-context response efficacy measurement The use of different colors and tools to visualize
manipulation better
The use of specific question as manipulation check Including control condition
Including cognitive style/nationality as potential moderator
More appropriate sample size for each participants group
12
References
Eia.gov. (2020). EIA - Annual Energy Outlook 2019. [online]
Ipcc.ch. (2019). Special Report on Climate Change and Land — IPCC site. [online]
Sparkman, G., & Walton, G. (2019). Witnessing change: Dynamic norms help resolve diverse barriers to personal change. Journal Of Experimental Social Psychology, 82, 238-252. DOI: 10.1016/j.jesp.2019.01.007
Swim, J., & Geiger, N. (2017). From Alarmed to Dismissive of Climate Change: A Single Item Assessment of Individual Differences in Concern and Issue Involvement. Environmental Communication, 11(4), 568-586. DOI: 10.1080/17524032.2017.1308409
Hayes, A. F. (2013). An introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York, NY: Guilford Press.
Hornsey, M., Fielding, K., McStay, R., Reser, J., Bradley, G., & Greenaway, K. (2015). Evidence for motivated control: Understanding the paradoxical link between threat and efficacy beliefs about climate change. Journal Of Environmental Psychology, 42, 57-65. DOI: 10.1016/j.jenvp.2015.02.003 Mortensen, C., Neel, R., Cialdini, R., Jaeger, C., Jacobson, R., & Ringel, M. (2017). Trending Norms: A Lever for Encouraging Behaviors Performed by the Minority. Social Psychological And Personality Science, 10(2), 201-210. DOI: 10.1177/1948550617734615
Ferguson, M., & Branscombe, N. (2010). Collective guilt mediates the effect of beliefs about global warming on willingness to engage in mitigation behavior. Journal of Environmental Psychology, 30(2), 135-142. DOI: 10.1016/j.jenvp.2009.11.010
Choi, I., Koo, M., & Jong An Choi. (2007). Individual Differences in Analytic Versus Holistic Thinking. Personality And Social Psychology Bulletin, 33(5), 691-705. doi: 10.1177/0146167206298568
