Stop the Denial!
The Effect of Assertive Language in Pro-Environmental Messages on
Reactance, Attitude, and Behavior Intention
Master’s Thesis
University of Amsterdam - Graduate School of Communication
Master Track Persuasive Communication
Antonina Siermann
Student ID: 12265039
Supervisor: Ivana Bušljeta Banks
Abstract
Encouraging individuals to act environmentally friendly is challenging. Past research reveals that one of the reasons public messages aimed at changing behavior fail to achieve the desired effect is the use of assertive language. Since this controlling and forceful phrasing of
messages tends to elicit reactance, an aversive arousal in response to threats to freedom and autonomy, receivers are less likely to comply with the advocated stance. This study aims to examine this relationship of assertive language on reactance, attitude, and behavior intention in the context of promoting pro-environmental behavior. Additionally, a frequently used moderator in communication research is included which might mitigate the arousal of reactance towards assertive language: source credibility. Since highly credible sources are considered to be knowledgeable and competent, the assertive language used may be seen as justified, and not as threatening or impolite. The conceptual model used in this study is analyzed in the context of plastic pollution and the use of reusable water bottles as an advised solution. In line with prior research to assertive language and reactance theory, results
indicate that reactance mediates the effect of assertive language on attitude towards reusable bottles and behavior intention to use a reusable bottle in the future: Participants exposed to a highly assertive message experienced significantly more reactance than participants in the low-assertive condition, which negatively affected their attitudes and behavior intentions. However, source credibility showed no significant effect in this relationship so that a moderated mediation effect was excluded.
Keywords: assertive language, reactance, source credibility, attitude, behavior intention
Introduction
‘Drop Dirty Palm Oil Now’, ‘Take Action!’, and ‘Stop Bankrupting Our Oceans’ are common slogans used by policy-makers and environmental stakeholders in an attempt to change individuals’ behavior (Greenpeace International, 2019; WWF – World Wildlife Fund, 2019). However, prior research in the field of health-communication, psycholinguistics, and advertising has shown that such highly assertive phrases, including commands and orders, result in low receiver compliance (Kronrod, Grinstein, & Wathieu, 2012; Dillard & Shen, 2005; Quick & Considine, 2008; Zemack-Rugar, Moore, & Fitzsimons, 2017). The
psychological mechanism behind this effect is called reactance, an aversive arousal that elicits when individuals feel that their autonomy or self-determination is being threatened (Brehm, 1966; Miller et al., 2007). The Theory of Psychological Reactance further states that
individuals experiencing reactance are motivated to reestablish threatened freedoms through resistance techniques, such as message rejection, source derogation, or behaving exactly opposite of how they are asked to (Brehm & Brehm, 1981; Ungar, Sieverding, Schweizer, Stadnitski, 2015). However, research of psychological reactance in the field of social issues are limited, although assertive language is extensively used in pro-environmental messages.
Furthermore, studies indicate that assertive language does not always lead to the experience of reactance: Messages that contain explicit statements and direct commands can also be perceived as clear and important, which positively affects receivers’ assessment of the source (Miller et al., 2007). Similarly, assertive expressions used by highly credible sources can be viewed as justifiable, and therefore may not elicit reactance in receivers. In
comparison, assertive language used by low-credibility sources can be perceived as
unjustified, leading to defensive reactions (Jenkins & Dragojevic, 2013). However, according to Sternthal, Phillips, and Dholokia (1978), source credibility only matters when the claims made in the message are highly discrepant from receivers’ initial opinion. If the arguments
made are in line with receivers’ opinion, both high and low-credibility sources are likely to reach the same persuasion effects.
Taking the prior findings collectively, the purpose of this research is to extend current reactance literature by investigating the relationship of assertive language on receivers’ reactance, attitude, and behavior intention in a pro-environmental context. The study further examines if this relationship is dependent on receivers’ assessment of the source as source credibility has proven to have a considerable impact on the effects of assertive language. Since the conceptual model is analyzed in the context of plastic pollution and the use of reusable water bottles as an advised solution, the study focuses on receivers’ attitude towards reusable bottles (ARB) and their behavior intention to use a reusable bottle in the near future
(BI). The research question of this study is as follows:
To what extent do pro-environmental messages written in high-assertive language have an influence on receiver’s reactance, attitude, and behavior intention compared to pro-environmental messages written in low-assertive language? What role does source credibility play in this relationship?
Investigating this relationship addresses several gaps in literature. First, the majority of academic research on reactance theory is largely limited to the field of health communication (Rains & Turner, 2007; Dillard & Shen, 2005; Grandpre et al., 2003) and commercial
advertising (Fitzsimons & Lehmann, 2004; Meirick & Nisbett, 2011). Studies that examine psychological reactance with regard to social issues such as pro-environmental campaigns have been minimal.
Second, research on the effect of assertive language on receivers’ reactions yields inconsistent findings (Kim, Baek, Yoon, Oh, & Choi, 2017; Katz, Kronrod, Grinstein & Nisan, 2018). Extending knowledge on assertive phrasing in pro-environmental messages and its impact on public opinion can, therefore, further illuminate the linguistic research. From a
practical point of view, the findings of the study reveal how and when assertive language can increase compliance, which is especially important when persuading individuals to act in an environmentally friendly manner. Thus, the study provides a direction for policy-makers and organizations to effectively design environmental messages in the future and reach maximum persuasion effects.
Third, while source credibility is extensively investigated in the field of
communication, a review of the literature provides no findings of this concept being studied in a pro-environmental context combined with assertive language.
The thesis starts with a theoretical grounding of the central concepts used in the study, while the proposed hypotheses are consecutively presented. The theoretical framework is followed by the description of the chosen research design and the procedure thereof. In the results section, analyses of the collected data and the findings of the study are presented. The thesis concludes with a discussion of the findings, including theoretical and managerial implications, the limitations of the study, and suggestions for future research.
Theoretical Background Pro-Environmental Messages
Since human behavior is the main factor causing environmental problems such as climate change or plastic pollution, there is an increasing need for governments,
organizations, and green political parties to effectively craft messages that promote behavior change towards a more sustainable and environmentally friendly lifestyle (Vlek & Steg, 2007; Bhatnagar & McKay-Nesbitt, 2016). However, persuading individuals to act in an
environmentally responsible manner proves a challenging task since the beneficiary thereof is not always directly the individual him/herself but society or the environment at large (Lucas, Brooks, Darnton, & Jones, 2008; Kronrod, Grinstein, & Wathieu, 2012). Additionally, a large range of studies reveals that public messages are often written in assertive language, including
commands, explicit statements, and orders, which leads to reluctant reactions of receivers’ (Wilson & Kunkel, 2000; Zhang & Sapp, 2013; Grandpre et al. 2003). Since these assertive expressions force receivers to comply with the advocated stance, the message can be
perceived as threatening and impolite, which can in turn increase negative outcomes, such as message rejection or source derogation (Dillard & Shen, 2005; Jenkins & Dragojevic, 2013). Other findings on pro-environmental campaigns reveal more promising strategies to effectively craft persuasive messages. For example, the results of a study by Gifford and Comeau (2011) show that message framing has a considerable impact on perceived climate change competence, engagement, and behavioral intentions. The authors concluded that motivational-oriented frames (i.e. messages that involve solutions, values, and portray
positive changes in lifestyle) tend to be more effective in encouraging climate-change-related behaviors than sacrifice-oriented frames (i.e. messages that focus on sacrifices by citizens). Moreover, Cialdini (2003) recommends public service communicators to use both descriptive (i.e. portray what people actually do) and injunctive (i.e. characterize typically approved behaviors, what people are supposed to do) social norms in messages to mobilize action against a problem. More specifically, his studies showed that behavior intention (e.g., to recycle, or to stop littering) was higher when practitioners made clear that the green behavior is a) prevalent and b) approved by others. Compared to assertive language as a persuasion strategy, these techniques do not necessarily threaten receivers’ feelings of autonomy and self-determination, and therefore may prevent the aversive arousal of reactance (Brehm, 1966).
Assertive Language
In linguistic research, ‘assertiveness’ is defined as ‘confident and forceful behavior’ (Stevenson, 2010), whereby ‘assertive’ is equated with being aggressive, self-assured, or pushy (Merriam-Webster’s, 2019). Derived from these definitions, assertive language is
described as direct, demanding and forceful, and is apparent in messages that contain the imperative verb-form, such as ‘do’ and ‘must’ (Kronrod, Grinstein, & Wathieu, 2012; Dillard & Shen, 2005). In the scope of pro-environmental campaigns, non-government organizations like Greenpeace and WWF make use of this linguistic tool, which is apparent in their
campaign slogans ‘Take Action’, or ‘Stop Bankrupting Our Oceans’ (Greenpeace International, 2019; WWF – World Wildlife Fund, 2019).
Green advertisements and campaigns using assertive language aim to guide individuals towards performing a certain behavior by giving clear directions without providing an option to refuse (Katz, Kronrod, Grinstein, & Nisan, 2018). Commands and orders, but also statements that demean individuals who hold contradicting views, are components of assertive language. On the contrary, low-assertive messages emphasize receiver’s freedom of choice, and are written in a more polite and gentle manner, including words like ‘might’, ‘please’ or ‘possibly’ (Jenkins & Dragojevic, 2013; Dillard & Shen, 2005).
Some studies, however, reveal that assertive language can have positive effects. Miller et al. (2007) found that compared to implicit, low-controlling messages, explicit messages that contain clear requests, direct commands, and convey one single meaning, receive more attention, are perceived as more important, and generate more positive assessments of the source. Furthermore, Kim et al. (2017) found that reactance to assertive messages is
dependent on cultural psychology, in that people in western cultures, who see themselves as unique and independent, feel a higher threat to freedom triggered by assertive language than people in eastern cultures, who focus on group harmony and interpersonal relations and thus find threats to individual freedom less disturbing (Markus & Kitayama, 1991). Persuasiveness of assertive language also depends on the receiver’s perceived importance of the presented topic (Kronrod, Grinstein, & Wathieu, 2012). Pushy requests do not seem to elicit reactance
in receivers who view the issue as important, while receivers who perceive the issue of lower importance respond better to low-assertive language.
However, the fact that policy-makers and environmental organizations frequently use assertive language is rather remarkable since the majority of empirical research has shown that the use of gentle, less assertive language in messages increases compliance and achieves greater persuasiveness than highly assertive phrasing (Wilson & Kunkel, 2000; Zhang & Sapp, 2013; Grandpre et al. 2003). The unfavorable outcome of persuasive attempts that backfire due to assertive language is extensively documented in marketing literature (Fitzsimons & Lehmann, 2004) and communication science (Dillard & Shen, 2005), with specific focus on health-related behaviors (Quick & Considine, 2008).
Several authors have utilized the Politeness Theory to describe their findings why, and when receivers tend to demonstrate resistance to persuasive attempts. The theory posits that individuals possess both a fundamental need for autonomy (i.e., negative face) and a desire to be approved, liked, and respected (i.e. positive face). It is argued that assertive language can trigger an overall threat to face, which is consequently seen as impolite. This in turn can lead to a sequence of negative outcomes, which includes the derogation of the source, derogation of the message, and, finally, to less favorable attitudes toward the advocated stance (Brown & Levinson, 1987; Jenkins & Dragojevic, 2013).
Other authors claim that the Theory of Psychological Reactance is the reason why assertive language fails to persuade receivers. Since assertive language creates pressure to comply with the request or position advocated, individuals feel that the freedom to make their own choices or behave in a certain way is threatened, which in turn leads to reactance (Brehm & Brehm, 1981; Dillard & Shen, 2005; Zemack-Rugar, Moore, & Fitzsimons, 2017).
Reactance describes a motivational state to restore a threatened freedom by engaging in resistance techniques, such as derogating the source of threat, ignoring the persuasive attempt,
or engaging in more of the undesired behavior (Brehm & Brehm, 1981; Burgoon, Alvaro, Grandpre, Voulodakis, 2002).
To conclude, past research has repeatedly shown that assertive language leads to undesirable responses in receivers, which is mainly justified by the Theory of Psychological Reactance and Politeness Theory. In line with existing literature, it is hypothesized that:
H1: Exposure to a pro-environmental message that promotes the use of reusable bottles written in high-assertive (vs. low-assertive) language will lead to a more negative ARB and
lower BI.
Psychological Reactance
Based on the theory by Brehm (1966), psychological reactance is defined as an aversive arousal that occurs in response when behavioral freedom is threatened or eliminated by others. In other words, reactance describes the reaction individuals have when they experience a threat to their autonomy and self-determination due to feeling pushed towards doing something (Miller, 2015).
The Theory of Psychological Reactance consists of four essential stages: perceived freedom, threat to freedom, reactance, and restoration of freedom (Brehm, 1966). Freedom, in this context, is considered a concrete action, emotion, or attitude, which individuals possess only if they are aware of it and perceive the self-efficacy to enact that specific freedom (Brehm & Brehm, 1981; Dillard & Shen, 2005). Any force that impede performing a
perceived freedom constitutes a threat. A threat to freedom, in turn, motivates individuals to restore that freedom either directly through adapting the opposed behavior, i.e. the ‘forbidden act’, or indirectly by e.g., derogating the source of threat, contesting the existence of threat, or exerting a different freedom to regain a feeling of self-determination and choice (Dillard & Shen, 2005; Wicklund, 1974; Quick & Stephenson, 2007). The magnitude of reactance arousal is dependent on: a) the perceived importance of the threatened behavior, b) the
proportion of free behaviors that are threatened, and, c) the receiver’s competence to form meaningful judgements on the attitudinal issue (Smith, 1977; Wicklund & Brehm, 1968; Miron & Brehm, 2006; Burgoon, Alvaro, Grandpre, & Voulodakis, 2002).
In the context of persuasion, reactance mediates the effects of threat to freedom on outcomes such as attitudes, beliefs, or behaviors (Dillard & Shen, 2005). The investigation of this relationship has been increasingly popular in the field of advertising (Fitzsimons & Lehmann, 2004; Meirick & Nisbett, 2011), psychology (Sensenig & Brehm, 1968; Steindl et al., 2015; Kayser, Graupmann, Fryer, & Frey, 2016), and health communication (Rains & Turner, 2007) with focus on anti-smoking campaigns targeting adolescents (Grandpre et al., 2003), campaigns promoting dental flossing, or responsible drinking (Dillard & Shen, 2005). In line with existing literature, it is expected that psychological reactance mediates the effect of assertive language on receivers’ ARB and BI.
H2: The effect of assertive language in a pro-environmental message promoting the use of reusable bottles on ARB and BI is mediated by reactance, in that a) high-assertive language
will arouse more reactance than low-assertive language, which in turn b) leads to a more negative ARB and lower BI.
Source Credibility
Source credibility describes the communicator’s degree of favorable characteristics that influences the extent to which a receiver accepts or derogates the message (Ohanian, 1990; Wood & Burkhalter, 2014). Early research concludes that there are three factors leading to perceived credibility of the communicator: expertise (i.e. the degree of which the source is perceived as making valid claims), trustworthiness (i.e. the level of confidence the speaker conveys when making these assertions), and attractiveness (i.e. the physical appearance of the source) (Ohanian, 1990; Hovland, Janis, & Kelley, 1953).
A large number of studies reveals that individuals produce more favorable attitudes and are more persuaded by a message that is presented by a highly trustworthy and/or expert source than by a source that is perceived as low in credibility (Hovland & Weiss, 1951; Kelman & Hovland, 1953; Choo, 1964). One explanation for this effect is that individuals associate trust and confidence with highly credible sources, which inhibits activation of questioning or scrutinizing the assertions made and, therefore, facilitates acceptance of the message (Harmon & Coney, 1982). Contrarily, sources low in credibility are considered to lack knowledge and competence, which activate receivers to doubt or derogate the claims made in the message (Clark & Evans, 2014). However, other studies indicate that highly credible sources have a greater effect on persuasion only when the claims made in the
message are strongly discrepant from the recipient’s initial opinion. When the message claims are less discrepant, meaning that both the assertions in the message and the recipient’s attitude are more or less coherent, both high and low-credibility sources are likely to induce the same amount of persuasion (McGinnies, 1973; Sternthal, Phillips, Dholakia, 1978).
Regarding the language used in messages, Hewgill and Miller (1965) found that recipients were more likely to accept the advocated position when high-credibility sources presented a strongly threatening message that stressed the consequence of noncompliance. When the message was perceived as only mildly threatening, attitudes of participants changed significantly less towards the advocated stance. Similar results were observed in a study by Powell and Miller (1967), in which threatening messages presented by highly credible sources enhanced persuasion. However, these findings are inconsistent considering the Theory of Psychological Reactance, which states that threatening messages often result in unfavorable outcomes and a decrease in the evaluation of the source of restriction (Wicklund, 1974; Brehm & Brehm, 1981; Fitzsimmons & Lehmann, 2004).
In the context of public information campaigns, some studies indicate that non-profit organization sources are perceived as more credible than for-profit organization sources since
the genuine interest of social welfare is more attributed to non-profit than for-profit
organizations (Haley, 1996; Hammond, 1987; Reid, Soley, & Vander Bergh, 1981). These findings reveal that receivers use their so called ‘persuasion knowledge’ to identify the tactics and motives behind persuasion attempts (Friestad & Wright, 1994). Similarly, a study by Metzger and Flanagin (2013) revealed that due to limited time and cognitive capacity, people evaluate credibility by activating different heuristics that refer to reputation, consistency, or persuasive intent of the source. It is generally argued that when a clear persuasion intent is detected, which means that the source is perceived to gain more through successful persuasion than consumers or society, trustworthiness of the source decreases (Haley, 1996).
Following the first and second hypothesis and in line with previous research, it is expected that high-assertive language will lead to high reactance, but that the strength of reactance will attenuate when receivers perceive the source as trustworthy and having expertise (see Figure 1).
H3: Exposure to a pro-environmental message promoting the use of reusable bottles written in high-assertive (vs. low-assertive) language will lead to higher reactance, but the effect on reactance will be mitigated when receivers perceive the source as highly credible (vs. low in credibility).
Figure 1. Moderated mediation model of the study Assertive Language in pro-environmental messages Reactance Attitude & Behavior Intention Source Credibility H1 H2a H2b H3
Method
Before the study was conducted, a non-representative study found that plastic
pollution is one of the biggest environmental concerns at the moment. This is probably due to the fact that plastic pollution recently received worldwide attention, sparking discussion in mainstream media (Knapton, 2019; Laville, 2019; Wright, 2019). In contrast to other environmental problems, such as climate change or deforestation, plastic pollution is classified as a problem with clearly identifiable solutions, such as reducing the amount of plastic shopping bags (Kamaruddin, Yusuf, 2010; Ritch, Brennan, MacLeod, 2009), changing people’s recycling behaviors (Ofstad, Tobolova, Nayum, Klöckner, 2017), litter prohibition (Löhr et al., 2017; Gunggut, Hing, Saufi, 2013), or encouraging the installation of bottle refill stations (Uehara, Ynacay-Nye, 2018).
Given the relevance and actuality of this environmental issue, the pro-environmental message in this study focuses on plastic pollution and the use of reusable instead of
disposable water bottles.
Research Design & Participants
To test the hypotheses, an online experiment with a 2 (assertive language: high vs. low) x 2 (source credibility: high vs. low) between-subjects factorial design was created by using the online survey tool Qualtrics (see Table 1).
Table 1.
2 x 2 between-subject factorial design of the study
Source Credibility Assertive Language high low
high 1 2
Through a snowball sampling technique on social media (including Facebook and Instagram) a convenience sample was recruited. A moderated mediation approach was used to empirically test how source credibility moderates the mediation effect of assertive language on ARB and BI through reactance (see Figure 1).
Experimental Stimuli
Assertive Language. The level of assertiveness was manipulated with two short pro-environmental messages about the use of reusable water bottles, which were either written in a highly assertive, demanding language or in a less-assertive, suggestive manner. The
message was self-created, whereby some facts about the amount of plastic waste were taken from recent research reports (Jambeck et al., 2015; Laville & Taylor, 2017). In line with Miller et al. (2007), the high-assertive language condition contained imperatives and controlling phrases, including ‘should’, ‘have to’, and ‘must’. On the contrary, the low-assertive message was written more politely and consisted of recommendations and suggestions, using terms such as ‘might’, ‘worth’, and ‘please consider’ (Kim et al., 2017; Baek, Yoon, & Kim, 2015; Dillard & Shen, 2005). Except for these manipulated textual elements, other aspects of the message such as background, format, and text length remained constant across conditions.
Source Credibility. Given the topic of plastic pollution, source credibility was manipulated by using the non-profit organization ‘World Wildlife Fund’ (WWF) as a highly credible, reputable source and ‘Evian’, a for-profit brand of mineral water as a low-credibility source. Since WWF is known for its mission to preserve the environment and wildlife, it is expected that the source is perceived as more credible in this context than Evian, a brand known for selling mineral water in disposable bottles. The promoted behavior of using reusable water bottles that can be refilled with tap-water is, thus, contradicting to Evian’s mission of persuading consumers to purchase its bottled water. The layout of WWF’s and
Evian’s current website was used to simulate an actual message from the respective source. The experimental stimuli of the study can be found in Appendix A.
Pre-Test
Prior to the experiment, a pre-test study was conducted to examine if the manipulations of assertive language and source credibility were perceived as intended. Twenty participants, collected via a snowball method, were randomly assigned to one of the four conditions, followed by the respective manipulation check items. These participants were excluded from the main study.
Assertive Language. Adapted from a study by Dillard and Shen (2005), four items that measure ‘threat to freedom’ were used as a manipulation check to assess the level of assertiveness in the pro-environmental message, e.g.: ‘The message tried to make a decision for me’. Participants were asked to indicate their level of agreement on a 7-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree). A summary of all items can be found in Appendix B.
Source Credibility. The items used to assess perceived source credibility are derived from a study by Clark and Evans (2014). Depending on the condition, participants were asked to indicate to what extent they believe WWF or Evian to be a) ‘credible’, b) ‘trustworthy’, and c) an ‘expert on the issue’ on a 7-point Likert scale ranging from 1 (not at all) to 7 (very much).
Results Pre-Test. An exploratory factor analysis with Direct Oblimin rotation indicated that the scale that measured assertiveness was unidimensional, explaining 81.29% of the variance (Eigenvalue = 3.25). All four items were loading on the same factor which implies a strong association. Furthermore, the 7-point scale proved to be reliable as indicated by a Cronbach’s Alpha of .92. The manipulation turned out to be successful as an
= 5.53, SD = 1.40) as significantly more assertive than the low-assertive message (M = 3.35, SD = 1.73), t(18.00) = -3.10, p = .006.
The same analyses were used for the manipulation check of source credibility. An exploratory factor analysis with Direct Oblimin rotation showed that the scale was
unidimensional, explaining 97.45% of the variance (Eigenvalue = 2.92). Additionally, the 7-point scale proved to be reliable as indicated by a Cronbach’s Alpha of .99. An independent-samples t-test reported a successful manipulation of source credibility since participants perceived the high-credibility source (WWF) (M = 6.30, SD = .56) as significantly more credible than the low-credibility source (Evian) (M = 2.56, SD = 1.50), t(10.22) = -7.00, p < .001. Accordingly, the experimental stimuli were qualified to use in the main study.
Procedure
In the final study, participants were informed about the general idea of the study, the duration of the questionnaire, their anonymity, as well as the requirements of being at least 18 years old and fluent in English in order to complete the questionnaire. If participants provided informed consent, several demographic questions such as age and gender were asked. After that, participants were instructed that the following questions would refer to a
pro-environmental message they were about to see on the next page. Accordingly, participants were randomly assigned to one of the four conditions. Right after the exposure to the stimuli, participant’s reactance, ARB, and BI were measured. The manipulation check items of
assertiveness and source credibility were asked afterwards to avoid bias of the dependent variables. The study ended with a clear debriefing, in which participants were informed that the pro-environmental message and the allocation of the sources WWF and Evian were completely fictitious and only created for the purpose of the study. The experiment took about eight minutes to complete. The entire questionnaire can be found in Appendix D.
Measurements
Psychological Reactance. A shortened version of the scale by Hong and Faedda (1996) was used to measure psychological reactance. Although the scale was conceptualized to measure reactance as a personality trait, the eight items used were reformulated to assess reactance as a situational state. Trait reactance describes individuals’ proneness to reactance and is therefore seen as an antecedent variable whereas state reactance refers to individuals’ aversive arousal in a specific situation, and therefore functions as an outcome variable (Hong & Faedda, 1996; Dillard & Shen, 2005; Miron & Brehm, 2006). Example items are ‘The pro-environmental message I just read triggers a sense of resistance in me’ or ‘I consider the advice in the pro-environmental message to be an intrusion’. The items were assessed on a 7-point Likert scale ranging from 1 (= strongly disagree) to 7 (= strongly agree) (see Appendix B for all items).
Attitude towards Reusable Bottles. To assess participants’ ARB, they were asked
how bad/good, foolish/wise, negative/positive, unfavorable/favorable, unnecessary/necessary, harmful/beneficial, and undesirable/desirable they found the advocated behavior. The items were measured on a 7-point semantic differential items scale. The scale was adapted from a study by Baek, Yoon, and Kim (2015), which examined participants’ attitude towards recycling.
Behavior Intention. If participants indicated they purchase plastic water bottles either ‘daily’, ‘4-6 times a week’, ‘2-3 times a week’, ‘once a week’, or ‘occasionally’, their
intention to use a reusable bottle in the near future was measured as the mean of three
statements on a 7-point Likert scale (1 = very unlikely, 7 = very likely). The items were based on a study by Baek, Yoon, and Kim (2015), which focused on recycling behavior, and
reformulated so that the participants’ intention to use a reusable bottle could be measured (see Appendix B for all items).
Control variables. Keeping in line with the findings of Kronrod, Grinstein, and Wathieu (2012), issue importance was included as a control variable. Participants were asked to indicate the extent to which they perceived the reduction of plastic pollution as an
important topic on a 7-point Likert scale ranging from 7 (= extremely important) to 1 (= extremely unimportant).
Results Sample
The sample consisted of 225 participants. From the total number of participants, six did not provide consent, 16 did not finish the questionnaire or had missing data on important variables, and 22 were not familiar with the portrayed source either in the low or
high-credibility condition. These 44 participants were excluded from the data analysis, leaving 181 participants in the final sample, of which 69.6% were female (Nf = 126) (see Table 3 for the
number of participants per condition). Participants’ age ranged from 20 to 78 years, with an average age of 27.4 (SD = 8.89). Most of the participants completed a Bachelor’s Degree (55.2%), followed by a Master’s Degree (26.5%). The majority of the participants identified themselves as white (80.7%), while the next largest group of participants identified
themselves as Asian (8.3%). The remainder was mixed or preferred to not answer the question.
Table 3
Number of participants per condition
High Credibility (HC) Low Credibility (LC) Total
High Assertiveness (HA) 46 43 89
Low Assertiveness (LA) 48 44 92
Construction of Variables
Reactance. An exploratory factor analysis with Direct Oblimin rotation indicated that the scale that measured reactance was unidimensional, explaining 58.77% of the variance (Eigenvalue = 4.70). All items were loading on the same factor, which implies a strong association. Furthermore, the scale proved to be reliable as indicated by a Cronbach’s Alpha of .89. The means of all eight items were computed to create a new variable for reactance (M = 2.81, SD = 1.26). Higher scores on this variable correspond to a higher experience of reactance, and lower scores indicate a lower experience of reactance.
Attitude towards Reusable Bottles. Participants’ ARB was assessed by seven items
on a 7-point semantic differential scale (Baek, Yoon, & Kim, 2015). An exploratory factor analysis with Direct Oblimin rotation showed that the scale was unidimensional, explaining 72.29% of the variance (Eigenvalue = 5.06). Furthermore, as indicated by a Cronbach’s Alpha of .93, the scale proved to be highly reliable. The new variable for ARB (M = 6.58, SD = 0.71)
was created by computing the means of the items. Higher scores on the variable indicated a positive ARB and lower scores a negative ARB.
Behavior Intention. Before analyses with the variable of BI were conducted, the one negatively framed item was reverse coded, so that a higher score on the item referred to a higher BI. An exploratory factor analysis with Direct Oblimin rotation indicated that the scale was unidimensional, explaining 69.57% of the variance (Eigenvalue = 2.09). A strong
association of the three items can be expected since the three items are loading on one factor. The scale proved to be reliable as indicated by a Cronbach’s Alpha of .76. The means of the three items were computed to create the new variable for BI (M = 5.61, SD = 1.34). Higher scores on the variable mean a higher BI, while lower scores correspond to a lower BI.
Participants’ current behavior of using a reusable bottle was assessed by asking how often they generally buy disposable water bottles. Thirty participants indicated to ‘never’
purchase any plastic bottles and were therefore excluded from all analyses with BI as an outcome variable.
Randomization Check
To determine whether the randomization of participants across conditions in terms of age was successful, a one-way Analysis of Variance (ANOVA) with ‘conditions’ as
independent variable, and ‘age’ as dependent variable was conducted. The ANOVA showed that participants’ mean age did not significantly differ per condition, F (3.18) = 2.26, p = .084.
Additionally, a Chi-square test was conducted with ‘conditions’ as independent variable and ‘gender’ as outcome variable. The test revealed that males and females were equally distributed among the four conditions, χ 2 (3, N = 181) = 4.48, p = .214.
To conclude, the randomization was successful.
Manipulation Check
Four items were included to check if participants perceived the conditions with different levels of assertive language as intended (Dillard & Shen, 2005). An exploratory factor analysis with Direct Oblimin rotation showed that the four items form a
one-dimensional scale (Eigenvalue = 3.10), explaining 77.44% of the variance. The 7-point scale proved to be reliable as indicated by a Cronbach’s Alpha of .90. Higher scores on the scale correspond to higher perceived assertive language and vice versa. Results of an independent-samples t-test revealed that the manipulation was successful: Participants perceived the high-assertive message (M = 3.64, SD = 1.77) as significantly more high-assertive than the low-high-assertive message (M = 2.57, SD = 1.35), t(164.52) = -4.53, p < .001.
The same analyses were used for the manipulation check of source credibility. An exploratory factor analysis with Direct Oblimin rotation indicated that the scale of three items measuring source credibility was one-dimensional (Eigenvalue = 2.53), explaining 84.20% of
the variance. The items were loading on the same factor which indicated a strong association. Furthermore, the 7-point scale proved to be highly reliable (Cronbach’s Alpha of .91). An independent-samples t-test revealed that the manipulation of source credibility was also successful: Participants perceived the high-credibility source (WWF) (M = 5.85, SD = 1.01) as significantly more credible than the low-credibility source (Evian) (M = 4.32, SD = 1.46), t(151.53) = -8.13, p < .001.
Hypotheses Testing
Hypothesis 1. To test whether there was a direct effect of assertive language (high vs. low) on participants’ ARB and BI, an independent-samples t-test was conducted. The results
revealed that after reading the pro-environmental message, the ARB of the 92 participants in
the low-assertive language condition (M = 6.62, SD = 0.65) did not significantly differ from the ARB of the 89 participants in the high-assertive language condition (M = 6.54, SD = 0.77),
t(179) = .77, p = .440.
Similarly, participants’ BI did not significantly differ after seeing a pro-environmental message in either high or low-assertive language, t(149) = 1.82, p = .071. This means that the participants in the low-assertiveness condition (M = 5.81, SD = 1.36) were equally motivated to use a reusable water bottle in the near future as the participants in the high-assertiveness condition (M = 5.41, SD = 1.30). The first hypothesis of the study is, therefore, rejected.
Although the causal step approach by Baron and Kenny (1986) states that a significant direct effect of the independent on the dependent variable is preconditioned in order to infer a mediation effect, Hayes (2013) introduces more rigorous approaches that legitimate a
conclusion of an indirect mediation effect, even if there is an insignificant total effect (c path). One of those approaches is called bootstrapping, which was applied for the second and third hypothesis testing. The number of bootstrap samples remained at 5000, while the model generated bias corrected in 95% bootstrap confidence intervals.
Hypothesis 2. To investigate a possible indirect effect of assertive language on ARB
and BI mediated by reactance, a mediation test was performed using the 4th model of the PROCESS macro in SPSS (Hayes, 2013). Results of the multiple regression analysis indicated that assertive language was a significant predictor of reactance, b = .66, SE = .18, t(179) = 3.65, p < .001 (for N = 181), b = .55, SE = .20, t(149) = 2.78, p = .006 (for N = 151), in that participants in the high-assertive language condition experienced significantly more reactance than participants in the low-assertive language condition. Additionally, reactance significantly predicted ARB, b = -.19, SE = .04, t(178) = -4.51, p < .001, and BI, b = -.53, SE =
.08, t(148) = - 6.78, p < .001, since participants who experienced higher reactance had a less favorable ARB and lower BI than those experiencing low reactance. The confidence intervals
surrounding the indirect effect did not include zero, which means that a significant mediation effect of reactance could be reported for ARB, indirect = -.12, SE = .05, 95% CI [-.22, -.05],
and BI, indirect = -.29, SE = .12, 95% CI [-.56, -.08].
To conclude, participants assigned to the high-assertive language condition experienced more reactance and consequently scored lower in their ARB and BI than
participants in the low-assertive language condition. Therefore, hypotheses 2a and 2b were confirmed.
Hypothesis 3. The third hypothesis stated that a pro-environmental message written in high-assertive language (vs. low-assertive language) will lead to higher reactance, but the effect on reactance will decrease when receivers perceive the source as high in credibility (vs. low in credibility). To test this moderated mediation effect, the 7th model of the PROCESS macro for SPSS was run (Hayes, 2013), in which ‘source credibility’ was included as a moderator.
The multiple regression analysis revealed no moderation effect: Neither a significant effect of source credibility on reactance, b = -.11, SE = .26, t(177) = -.43, p = .668 (N = 181),
b = -.00, SE = .28, t(147) = -.00, p = .997 (N = 151), nor a significant interaction effect of assertive language and source credibility on reactance was found, b = .27, SE = .37, t(177) = .73, p = .467 (for N = 181), b = .11, SE = .40, t(147) = .28, p = .777 (for N =151). Since the confidence intervals surrounding the conditional indirect effects include zero, no significant moderated mediation effect was observed. More specifically, there was neither a significant effect of assertive language on ARB, indirect = -.05, SE = .07, 95% CI [-.19, .09] nor on BI
indirect = -.06, SE = .22, 95% CI [-.49, .36], mediated by reactance and moderated by source credibility. Therefore, the third hypothesis was rejected.
Figure 2 shows the coefficients of the moderated mediation analysis in the conceptual model, whereby the blue numbers represent the analysis for ARB and the orange numbers the
analysis for BI. An overview of the hypotheses and results can be found in Appendix C.
Notes: *p < .01, **p < .001
Figure 2. Results of Moderated Mediation Analysis
Additional Findings
To control for the effect of issue importance in the moderated mediation analysis, the variable was included as a covariate in the 7th model of the PROCESS macro for SPSS (Hayes, 2013). The multiple regression analysis reported that issue importance was a
significant predictor of reactance, b = -.15, SE = 0.07, t(176) = -2.00, p = .047 (N = 181), ARB,
.55* Assertive Language in pro-environmental messages Reactance .04 -.19** .66** -.53** -.10 ARB Analysis for (N = 181) Analysis for BI (N = 151) .11 .27 Attitude & Behavior Intention Source Credibility
b = .16, SE = .04, t(177) = 4.11, p < .001 (N = 181), and BI, b = .21, SE = .07, t(147) = 2.83, p = .005 (N = 151).
However, a lower sample size yielded no significant effect of issue importance on reactance, b = -.14, SE = .08, t(146) = -1.71, p = .089 (N = 151). For both sample sizes, no moderated mediation effect was found, indirect = -.05, SE = .06, CI [-.17, .07] (N = 181), indirect = -.08, SE = .20, CI [-.47, .31] (N = 151).
Conclusion & Discussion
Previous research has repeatedly shown that assertive language leads to unfavorable outcomes since commands and forceful expressions arouse reactance in receivers. The aim of this study was to contribute to this developing research field by investigating the use of this linguistic tool in pro-environmental messages, whereby the moderating role of source credibility was included. The research question was: ‘To what extent do pro-environmental messages written in high-assertive language have an influence on receiver’s reactance, attitude, and behavior intention compared to pro-environmental messages written in low-assertive language? What role does source credibility play in this relationship?’
Theoretical implications
Results indicated that although no direct effect of assertive language on ARB and BI
occurred, an indirect effect with reactance as a mediator was observed. The results mirror previous findings of assertive language and reactance theory in that people exposed to assertive language felt that their freedom was being threatened, which in turn elicited the unpleasant arousal of reactance. Consequently, the state of high reactance led to more
aversion towards the advocated stance since ARB and BI decreased. In contrast, participants in
the low assertive language condition showed lower reactance. As expected, the suggestive message constituted whether a threat nor a pressure to comply to the promoted position,
which in turn positively affected their ARB and BI (Brehm & Brehm, 1981; Dillard & Shen,
2005; Zemack-Rugar, Moore, & Fitzsimons, 2017).
Surprisingly, no moderating effect of source credibility was found. Although
participants perceived WWF as significantly more trustworthy than Evian, source credibility did not change the effect of assertive language on reactance. There are several implications that could explain this finding. First, when viewing the pro-environmental post, participants may put all their attention to the message itself, whereby the source was left unnoticed. Therefore, only the used language, and not the source, lead to reactance and a consequent decrease in ARB and BI. Second, the surprisingly high ARB (M = 6.6) indicate that participants
possibly already felt positively towards reusable bottles prior to the study. Following this assumption, receivers gave less care to who the message was presented by as long as the arguments made were coherent with their own beliefs. This explanation is in line with previous findings of McGinnies (1973), and Sternthal, Phillips, and Dholakia (1978), which state that source credibility has no systematic effect if the degree of discrepancy between message claims and one’s initial opinion is low. Third, although the manipulation check was significant, participants also perceived the ‘non-credible’ source Evian as relatively
trustworthy (M = 4.32). This finding is contradicting to results of early research, which state that for-profit organizations are often perceived as untrustworthy by default (Haley, 1996; Hammond, 1987; Reid, Soley, & Vander Bergh, 1981). One may conclude that consumers appreciate Evian’s pro-environmental effort of encouraging consumers to reuse their bottles. In line with the findings of Jenkins and Dragojevic (2013), the assertive phrasing used in the pro-environmental message by Evian was, thus, seen as equally justified as the assertive language used by WWF. Finally, as Metzger and Flanagin (2013) point out, information about the source is necessary in order to make legitimate judgements about the sender’s credibility. Since this information was unavailable in the study, participants who only knew Evian or
WWF by name were unable to form an opinion about the credibility of the company, which may explain the insignificant results.
Issue importance as a control variable yielded similar effects as found in the study by Kronrod, Grinstein, and Wathieu (2012): People who perceived plastic pollution as an important topic, experienced less reactance and a greater shift in ARB, as well as BI, after
reading a highly assertive message than people who perceived the environmental issue as less important. In line with other findings, the success of persuasion strategies varies depending on the attributed importance of the specific issue at stake (Lecheler, de Vreese, & Slothuus, 2009; Zuwerink & Devine, 1996). Therefore, the moderating and/or mediating role of issue importance should preferably be included in future studies aimed at determining persuasion effects on public opinions.
Managerial Implications
The findings of this study have several implications for policy-makers designing pro-environmental messages. The significant indirect effect of assertive language on ARB and BI
via reactance implies that the phrasing of messages matters when it comes to persuading people to act environmentally friendly. From this study, one can deduce that
pro-environmental messages achieve higher persuasive effects when the message is formulated in a suggestive manner that emphasizes self-determination and choice. Highly assertive
language including commands and imperatives tend to alienate receivers due to the aversive arousal of reactance, which in turn negatively affects attitudes and behavior intentions.
Since source credibility had an insignificant effect, it is derived that receivers care more about the message itself than the sender who produced it. However, this finding should be treated with caution, as this insignificant effect may be due to the fact that the arguments made in the message were coherent with receivers’ initial beliefs. If the message claims are highly discrepant from receivers’ opinions, it is suggested that high-credibility sources are
more persuasive than low-credibility sources (McGinnies 1973, Sternthal, Phillips, &
Dholakia, 1978). According to this finding, policy-makers are advised to think carefully about what arguments are presented with what source. If it is assumed that the audience shares the same opinion as the sender of the message, for-profit organizations are likely to reach as many positive effects posting a pro-environmental message as non-profit organizations, regardless of whether the source is seen as high or low in credibility. In other words, individuals are prone to agree with messages that reflect their own opinions. When similar opinions are not assumed, a high-credibility source will probably lead to higher persuasion effects than a low-credibility source (Haley, 1996; Hammond, 1987).
Finally, since assertive language did not seem to negatively affect people that viewed the topic of plastic pollution as important, practitioners are advised to first create enough awareness about the environmental issue and stress the urgency to do something against the problem before using commands and imperatives in a campaign message. If people are frequently confronted with the environmental issue, personal importance on the topic is likely to increase as well, which reduces reactance and enhances the willingness to comply
(Kronrod, Grinstein, & Wathieu, 2012).
Limitation and Future Research
There are some limitations to the study that potentially impacted the results and could be improved in future research. First is the relatively small sample size. Although the four conditions consisted of enough participants to draw reliable conclusions, a larger sample size would not only increase the validity but also the reliability of the results (Cohen, 1988).
Second, the surprisingly high scores of ARB (M = 6.6) and BI (M = 5.6) imply that
participants were already greatly aware of plastic pollution as an environmental issue and the addressed solution of using a reusable bottle before the study. Therefore, it might be
familiar environmental issue, e.g., deforestation or the impact of palm oil plantations, would yield similar findings.
A third limitation is that, due to limited time and budget, the study focused only on behavior intention, instead of assessing actual behavior. Ajzen, Brown, and Carvajal (2004) point out that the expressed willingness to perform a specific behavior does not automatically lead to its actual performance. Therefore, future research should place more focus on the evaluation of real behavior instead of self-reported questionnaire responses about the intention to carry out a specific behavior.
Finally, the significant effect of assertiveness on reactance in this study may be due to the fact that the sample consisted of 81% white people. In line with Kim et al. (2017), people of western, individualistic cultures tend to feel a higher threat to freedom triggered by
assertive language than people from eastern, collectivistic cultures. Therefore, it would be worth expanding the study to include individuals of different cultural backgrounds.
References
Ajzen, I., Brown, T., Carvajal, F. (2004). Explaining the discrepancy between intentions and actions: The case of hypothetical bias in contingent valuation. Personality and Social Psychology Bulletin, 30(9), 1108–1121. doi:10.1177/0146167204264079
Baek, T. H., Yoon, S., & Kim, S. (2015). When environmental messages should be assertive: Examining the moderating role of effort investment. International Journal of
Advertising, 34(1), 135–157. doi.org/10.1080/02650487.2014.993513
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51(6), 1173–1182.
Bhatnagar, N., McKay-Nesbitt, J. (2016). Pro-environment advertising messages: The role of regulatory focus. International Journal of Advertising, 35(1), 4–22.
doi:10.1080/02650487.2015.1101225
Brehm, J. W. (1966). A theory of psychological reactance. Oxford, England: Academic Press. Brehm, J. W. & Brehm, S. S. (1981). Psychological reactance – a theory of freedom and
control. New York, Academic Press.
Brown, P., & Levinson, S. C. (1987). Politeness: Some universals in language usage, Vol. 4, Cambridge University Press. doi:10.1525/ae.1988.15.4.02a00420
Burgoon, M., Alvaro, E., Grandpre, J., & Voulodakis, M. (2002). Revisiting the theory of psychological reactance: Communicating threats to attitudinal freedom. In Dillard, J.P, & Pfau, M. (Eds.), The Persuasion Handbook: Developments in Theory and Practice. California, Sage Publications. doi:10.4135/9781412976046.n12
Choo, TH. (1964). Communicator credibility and communication discrepancy as determinants of opinion change. The Journal of Social Psychology, 64(1), 65–76.
doi:10.1080/00224545.1964.9919544
Cialdini, R. B. (2003). Crafting normative messages to protect the environment. Current Directions in Psychological Science, 12(4), 105–109. doi:10.1111/1467-8721.01242 Clark, J. K., & Evans, A. T. (2014). Source credibility and persuasion: The role of message
position in self-validation. Personality and Social Psychology Bulletin, 40(8), 1024– 1036. doi:10.1177/0146167214534733
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum.
Dillard, J. P. & Shen, L. (2005). On the nature of reactance and its role in persuasive health communication. Communication Monographs, 72(2), 144–168.
Fitzsimons, G. J., & Lehmann, D. R. (2004). Reactance to recommendations: When unsolicited advice yields contrary responses. Marketing Science, 23(1), 82–94. doi:10.1287/mksc.1030.0033
Friestad, M., Wright, P. (1994). The Persuasion Knowledge Model: How people cope with persuasion attempts. Journal of Consumer Research, 21(1), 1–31. doi:10.1086/209380 Gifford, R., & Comedau, L. A. (2011). Message framing influences perceived climate change
competence, engagement, and behavioral intentions. Global Environmental Change, 21(4), 1301–1307. doi:10.1016/j.gloenvcha.2011.06.004
Grandpre, J., Alvaro, E. M., Burgoon, M., Miller, C. H., & Hall, J. R. (2003). Adolescent reactance and anti-smoking campaigns: A theoretical approach. Health
communication, 15(3), 349–366. doi:10.1207/S15327027HC1503_6 Greenpeace International. (2019). Retrieved March 15, from
https://www.greenpeace.org/international/act/
Gunggut, H., Hing, C. K., Saufi, D. S. N. S. A. M. (2013). Internalization and anti littering campaign implementation. Procedia – Social and Behavioral Sciences, 85, 544–553. doi:10.1016/j.sbspro.2013.08.383
Haley, E. (1996). Exploring the construct of organization as source: Consumer's
understandings of organizational sponsorship of advocacy advertising. Journal of Advertising, 25(2), 19–35. doi:10.1080/00913367.1996.10673497
Hammond, S. L. (1987). Health advertising: The credibility of organizational sources. In McLaughlin, M. L. (Ed.), Communication Yearbook, 10, pp. 613–628. Beverly Hills, CA: Sage.
Harmon, R. R., & Coney, K. A. (1982). The persuasive effects of source credibility in buy and lease situations. Journal of Marketing Research, 19(2), 255–260.
Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis: A regression‐based approach. New York, NY: The Guilford Press.
Hewgill, M. A., & Miller, G. R. (1965). Source credibility and response of fear-arousing communication. Speech Monographs, 32(2), 95–101.
doi:10.1080/03637756509375436
Hong, S.-M., & Faedda, S. (1996). Refinement of the Hong Psychological Reactance Scale. Educational and Psychological Measurement, 56(1), 173–182.
doi:10.1177/0013164496056001014
Hovland, C. I., & Weiss W. (1951). The influence of source credibility on communication effectiveness. Public Opinion Quarterly, 15(4), 635–650. doi:10.1086/266350 Hovland, C. I., Janis, I. L., & Kelley, H. H. (1953). Communication and persuasion;
psychological studies of opinion change. New Haven, CT, US: Yale University Press. Jambeck, J. R., Geyer, R., Wilcox, C., Siegler, T. R., Perryman, M., Andrady, A., Narayan,
R., & Law, K. L. (2015). Plastic waste inputs from land into the ocean. Science, 347 (6223), 768–771. doi:10.1126/science.1260352.
Jenkins, M., & Dragojevic, M. (2013). Explaining the process of resistance to persuasion: A politeness theory-based approach. Communication Research, 40(4), 559–590. doi:10.1177/0093650211420136
Kamaruddin, R., & Yusuf, M. M. (2010). Selangor Government’s “No plastic bag day” campaign: motivation and acceptance level. Procedia – Social and Behavioral Sciences, 42, 205–211. doi:10.1016/j.sbspro.2012.04.183
Katz, D., Kronrod, A., Grinstein, A., & Nisan, U. (2018). Still waters run deep: Comparing assertive and suggestive language in water conservation campaigns. Water, 10(3), 1– 10 (275). doi:10.3390/w10030275
Kayser, D. N., Graupmann, V., Fryer, J. W., & Frey, D. (2016). Threat to freedom and the detrimental effect of avoidance goal frames: reactance as a mediating variable. Frontiers in Psychology, 7, 632. doi:10.3389/fpsyg.2016.00632
Kelman, H. C., & Hovland, C. I. (1953). “Reinstatement” of the communicator in delayed measurement of opinion change. The Journal of Abnormal and Social Psychology, 48(3), 327–335. doi:10.1037/h0061861
Kim, Y., Baek, T. H., Yoon, S., Oh, S., & Choi, Y. K. (2017). Assertive environmental advertising and reactance: Differences between South Koreans and Americans. Journal of Advertising, 46(4), 550–564. doi:10.1080/00913367.2017.1361878 Knapton, S. (2019). Plastic pollution has got 10 times worse in seas around Britain since
2000. The Telegraph. Retrieved from https://www.telegraph.co.uk.
Kronrod, A., Grinstein, A., & Wathieu, L. (2012). Go Green! Should environmental messages be so assertive? Journal of Marketing, 76(1), 95–102. doi:10.1509/jm.10.0416
Laville, S., & Taylor, M. (2017). A million bottles a minute: world’s plastic binge ‘as dangerous as climate change’. The Guardian International Edition. Retrieved from https://www.theguardian.com/international.
Laville, S. (2019). ‘Biodegradable’ plastic bags survive three years in soil and sea. The Guardian. Retrieved from https://www.theguardian.com/international.
Lecheler, S., de Vreese, C., & Slothuus, R. (2009). Issue Importance as a Moderator of Framing Effects. Communication Research, 36(3), 400–425.
doi:10.1177/0093650209333028
Löhr, A., Savelli, H., Beunen, R., Kalz, M., Ragas, A., & Van Belleghem, F. (2017). Solutions for global marine litter pollution. Current Opinion in Environmental Sustainability, 28, 90–99. doi:10.1016/j.cosust.2017.08.009
Lucas, K., Brooks, M., Darnton, A., Jones, J. E. (2008). Promoting pro-environmental
behavior: existing evidence and policy implications. Environmental Science & Policy, 11, 456–466. doi:10.1016/j.envsci.2008.03.001
Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications for cognition, emotion, and motivation. Psychological Review, 98(2), 224–253. doi:10.1037/0033-295X.98.2.224
McGinnies, E. (1973). Initial attitude, source credibility and involvement as factors in persuasion. Journal of Experimental Social Psychology, 9(4), 285–296. doi:10.1016/0022-1031(73)90066-8
Meirick, P. C., & Nisbett, G. S. (2011). I approve this message: Effects of sponsorship, ad tone, and reactance in 2008 presidential advertising. Mass Communication and Society, 14(5), 666–689. doi:10.1080/15205436.2010.530381
Webster’s collegiate dictionary (10th ed.). (1993). Springfield, MA: Merriam-Webster.
Metzger, M. J., Flanagin, A. J. (2013). Credibility and trust of information in online
environments: The use of cognitive heuristics. Journal of Pragmatics, 59, 210–220. doi:10.1016/j.pragma.2013.07.012
Miller, C. H., Lane, L. T., Deatrick, L. M., Young, A. M., & Potts, K. A. (2007).
Psychological reactance and promotional health messages: The effects of controlling language, lexical concreteness, and the restoration of freedom. Human
Communication Research, 33(2), 219–240. doi:10.1111/j.1468-2958.2007.00297.x Miller, C. H. (2015). Persuasion and psychological reactance. The effects of explicit, high-controlling language. In Schulze R., & Pishwa H. (Eds.) The exercise of power in communication. Palgrave Macmillan, London. (pp. 465–501)
doi:10.1057/9781137478382_11
Miron, A. M., & Brehm, J. W. (2006). Reactance Theory – 40 Years Later. Zeitschrift für Sozialpsychologie, 37(1), 9–18. doi:10.1024/0044-3514.37.1.9
Ofstad, S. P., Tobolova, M., Nayum, A., & Klöckner, C. A. (2017). Understanding the mechanisms behind changing people's recycling behavior at work by applying a
comprehensive action determination model. Sustainability, 9(2), (204), 1–17. doi:10.3390/su9020204
Ohanian, R. (1990). Construction and validation of a scale to measure celebrity endorsers’ perceived expertise, trustworthiness, and attractiveness. Journal of Advertising, 19(3), 39–52. doi:10.1080/00913367.1990.10673191
Powell, F. A., & Miller, G. R. (1967). Social approval and disapproval cues in anxiety-arousing communications. Speech Monographs, 34(2), 152–159.
doi:10.1080/03637756709375535
Quick, B. L., & Stephenson, M. T. (2007). Further evidence that psychological reactance can be modeled as a combination of anger and negative cognitions. Communication Research, 34(3), 255–276. doi:10.1177/0093650207300427
Quick, B. L., & Considine, J. R. (2008). Examining the use of forceful language when
designing exercise persuasive messages for adults: A test of conceptualizing reactance arousal as a two-step process. Health Communication, 23(5), 483–491.
doi:10.1080/10410230802342150.
Rains, S. A., & Turner, M. M. (2007). Psychological reactance and persuasive health communication: A test and extension of the intertwined model. Human
Communication Research, 33(2), 241–269. doi:10.1111/j.1468-2958.2007.00298.x Reid, L. N., Soley, L. C., & Vander Bergh, B. G. (1981). Does source affect response to direct
advocacy print advertisements? Journal of Business Research, 9(3), 309–319. doi:10.1016/0148-2963(81)90024-2
Ritch, E., Brennan, C., & MacLeod, C. (2009). Plastic bag politics: modifying consumer behavior for sustainable development. International Journal of Consumer Studies, 33(2), 168–174. doi:10.1111/j.1470-6431.2009.00749.x
Sensenig, J., & Brehm, J. W. (1968). Attitude change from an implied threat to attitudinal freedom. Journal of Personality and Social Psychology, 8(4), 324–330.
doi:10.1037/h0021241
Smith, M. J. (1977). The effects of threats to attitudinal freedom as a function of message quality and initial receiver attitude. Communication Monographs, 44(3), 196–206. doi:10.1080/03637757709390131
Steindl, C., Jonas, E., Sittenthaler, S., Traut-Mattausch, E., & Greenberg, J. (2015).
Understanding psychological reactance. New developments and findings. Zeitschrift für Psychologie, 223(4), 205–214. doi:10.1027/2151-2604/a000222
Sternthal, B., Phillips, L. W., & Dholakia, R. (1978). The persuasive effect of source credibility: a situational analysis. The Public Opinion Quarterly, 42(3), 285–314. doi:10.1086/268454
Stevenson, A. (2010). assertiveness. Oxford Dictionary of English, (3 ed.). Oxford University Press.
Uehara, T., Ynacay-Nye, A. (2018). How water bottle refill stations contribute to campus sustainability: A case study in Japan. Sustainability, 10(9), 1–19.
doi:10.3390/su10093074
Ungar, N., Sieverding, M., Schweizer, F., & Stadnitski, T. (2015). Intervention-elicited reactance and its implications: Let me eat what I want. Zeitschrift für Psychologie, 223(4), 247–256. doi:10.1027/2151-2604/a000226
Vlek, C., & Steg, L. (2007). Human behavior and environmental sustainability: problems, driving forces, and research topics. Journal of Social Issues, 63(1), 1–19.
doi:10.1111/j.1540-4560.2007.00493.x
Wicklund, R. A., & Brehm, J. W. (1968). Attitude change as a function of felt competence and threat to attitudinal freedom. Journal of Experimental Social Psychology, 4(1), 64–75. doi:10.1016/0022-1031(68)90050-4
Wicklund, R. A. (1974). Freedom and reactance. Oxford, England: Lawrence Erlbaum. Wilson, S. R., & Kunkel, A. W. (2000). Identity implications of influence goals: Similarities
in perceived face threats and facework across sex and close relationships. Journal of Language and Social Psychology, 19(2), 195–221.
doi:10.1177/0261927X00019002002
Wood, N.T., Burkhalter, J. N. (2014). Tweet this, not that: A comparison between brand promotions in microblogging environments using celebrity and company-generated tweets. Journal of Marketing Communications, 20 (1-2), 129–146. doi:
10.1080/13527266.2013.797784
Wright, G. (2019). Etihad Airways: Is plastic-free flying the future? BBC News. Retrieved from https://www.bbc.com.
WWF – World Wildlife Fund. (2019). Retrieved March 15, from http://www.wwf.eu/campaigns/
Zemack-Rugar, Y., Moore, S. G., & Fitzsimons, G. J. (2017). Just do it! Why committed consumers react negatively to assertive ads. Journal of Consumer Psychology, 27(3), 287–301. doi:10.1016/j.jcps.2017.01.002
Zhang, Q., & Sapp, D. A. (2013). Psychological reactance and resistance intention in the classroom: Effects of perceived request politeness and legitimacy, relationship distance, and teacher credibility. Communication Education, 62(1), 1–25. doi:10.1080/03634523.2012.727008
Zuwerink, J. R., & Devine, P. G. (1996). Attitude importance and resistance to persuasion: It's not just the thought that counts. Journal of Personality and Social Psychology, 70(5), 931–944. doi: 10.1037/0022-3514.70.5.931
Appendix A
Stimulus material, condition one to four
Appendix B
Summary of items per scale
Variable Scale Items Cronbach's
Alpha Assertive
Language
Dillard & Shen (2005) The message threatened my freedom to choose .90
The message tried to make a decision for me
The message tried to manipulate me
The message tried to pressure me
Source Credibility
Clark & Evans (2014) I perceive Evian/WWF to be a credible source of information .91 I perceive Evian/WWF to be a trustworthy source of information
I perceive Evian/WWF to be an expert on the issue
Psychological Reactance
Hong & Faedda (1996) The pro-environmental message I just read triggers a sense of resistance in me .89 I find contradicting the pro-environmental message stimulating
I consider the advice in the pro-environmental message to be an intrusion
Reading the pro-environmental message frustrates me because it makes me feel that I cannot make a free and independent decision
It irritates me that the pro-environmental message points out things which are obvious to me
do just the opposite
I resist the attempts of the pro-environmental message to influence me
It makes me angry that the pro-environmental message is held up as a model for me to follow
Attitude Baek, Yoon, & Kim (2015)
I think using reusable instead of disposable water bottles to combat plastic pollution is bad vs. good, foolish vs. wise, negative vs. positive, unfavorable vs. favorable,
unnecessary vs. necessary, harmful vs. beneficial, undesirable vs. desirable
.93
Behavior Intention
Baek, Yoon, & Kim (2015)
I intend to reduce my plastic-bottle consumption and instead use a reusable bottle in the next few weeks
.76
I consider to use a reusable bottle more often instead of purchasing plastic water bottles in the next few weeks
I am likely to keep on purchasing plastic water bottles and not use a reusable bottle in the next few weeks
Appendix C
Summary of Hypotheses and Results
Hypothesis Summary Confirmed / Rejected
H1
Independent-samples t-test
no significant effect of assertive language on ARB and BI
x
H2a
Mediation Analysis with PROCESS, model 4 Assertive language significant predictor of reactance (p < .001 for N = 181, p < 0.01 for N = 151).
✓
H2b
Mediation Analysis with PROCESS, model 4 Reactance significant predictor of
ARB and BI (p < .001). Significant indirect effect
of assertive language on ARB and BI mediated by
reactance.
✓
H3
Moderated Mediation Analysis with PROCESS, model 7
No significant moderated mediation effect
