Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, CLIMATE SCIENCE (climatescience.oxfordre.com). (c) Oxford University Press USA, 2016. All Rights Reserved. Personal use only; commercial use is strictly prohibited. Please see applicable Privacy Policy and Legal Notice (for details see Privacy Policy).

date: 22 November 2017

Audience Segmentation and Climate Change Communication

Summary and Keywords

Scientists and policy makers face significant challenges when attempting to engage the public about climate change. An important first step is to understand the number and nature of the audiences one plans to target—a process known as audience segmentation. Segmentation involves identifying, within an audience or target population, homogenous subgroups that share similar demographic and/or psychographic profiles. After segmenting an audience, climate change communicators can target their messages based on the unique characteristics of each subgroup. For example, to stimulate engagement and behavior change, messages aimed at audiences that are skeptical about climate change may require different content and framing than messages aimed at audiences already deeply concerned about climate change.

The notion of matching message content to audience characteristics has a long history, dating back to the Ancient Greeks. More recently, audience segmentation has played a central role in targeted advertising and also social marketing, which uses marketing principles to help “sell” ideas and behaviors that benefit society. Applications to climate change communication are becoming more common, with major segmentation and communication initiatives being implemented across the globe.

Messages crafted to meet the needs of specific audience segments are more likely to be read, understood, and recalled than generic ones, and are also more likely to change behavior. However, despite these successes, the approach has not been uniformly embraced. Controversies have emerged related to the cost effectiveness of segmentation strategies, choice of segmentation variables, potential effects related to social stigmatization, whether segmentation encourages shallow (as opposed to deep) change, the extent to which segments are “found” as opposed to socially constructed by researchers, and whether interindividual differences are best conceptualized in terms of categories or dimensions.

Keywords: audience segmentation, social marketing, climate change, profiling analysis, message tailoring and targeting, public engagement, behavior change, communication

What Is Audience Segmentation?

Audience segmentation involves dividing a target population into relatively homogeneous, mutually exclusive subgroupings that share common profiles based on demographics, values, beliefs, behaviors, and/or position in the consumer decision-making process. Communicators then design messages and select communications channels that best match the characteristics of specific segments, a process known as message targeting. In some instances, communicators will craft messages for specific individuals, as opposed to larger segments. This is referred to as message tailoring, and is becoming increasing common with advances in Internet marketing. The primary goal of tailoring and targeting is to increase the persuasive and behavioral impact of marketing campaigns by matching message content to audience needs.

Segmentation can assist climate change communicators and program planners to make four main strategic decisions.

  1. 1. Who should be targeted? Segmentation research provides information about the number of distinct audiences present in a population, as well as the characteristics and relative size of each. This information can be used to guide resource allocation decisions. For example, an organization may decide it can maximize on-the-ground impact by targeting a large disengaged but receptive segment as opposed to focusing on a smaller group of entrenched skeptics.

  2. 2. How to optimize messages and intervention programs for each audience selected for targeting? Each audience segment is characterized by a unique combination of demographic, psychological, and behavioral attributes. These attributes can be helpful in guiding the development of engagement strategies specifically designed to address the knowledge needs and potential behavioral barriers facing each segment. Some segments may be unaware that climate change is occurring, whereas others may be motivated to act, but lack specific knowledge about what to do.

  3. 3. How to ensure the messages and interventions reach selected audiences? Different audience segments may have their own unique preferences for where they obtain information about climate change. Some rely heavily on Facebook, others watch cable news, and others prefer reading traditional print media. Using the wrong communication channel may prevent climate change messages from reaching their intended recipients, and may create unintended consequences if, for example, a climate change message framed for an alarmed audience is presented to a highly skeptical or dismissive one.

  4. 4. How to select messengers for each audience segment? Not all audiences will perceive certain climate communicators as credible and trustworthy. Al Gore is revered by some groups, and reviled by others. Audience segmentation helps climate change communicators select messengers with the relevant expertise, values, and personal experiences needed to build and maintain trust with their audiences.

A Brief History of Audience Segmentation

Persuasion and Marketing

The basic principles of audience segmentation date back to the 4th century B.C., when Aristotle, in his Art of Rhetoric, urged communicators to create messages to match the political orientations of their audiences (Aristotle, 2014/350 bce). He stressed that each form of government, whether democracy, oligarchy, or tyranny, had specific customs, institutions, and interests, all of which should be considered when attempting to persuade.

Smith (1956) introduced market segmentation as a marketing concept in the mid-1950s. During this period, most segmentations were based on demographics such as age, gender, education level, and socioeconomic status. But, in an influential paper, Yankelovich (1964) argued that this approach was too narrow and that consumers’ values, tastes, and preferences also should be taken into account. Ten years later, Demby (1974) popularized the term “psychographics” to describe the defining of segments by combining demographics with elements of psychology such as values, attitudes, beliefs, and behaviors. The terms “baby boomers,” “generations X and Y,” and “millennials” are all psychographic concepts in that they combine elements of birth age with specific values, attitudes, and behaviors that characterize each group.

One of the most well-known applications of psychographic methodology is the Values, Attitude, and Lifestyles (VALS) framework, developed by Mitchell (1983) at the Stanford Research Institute. VALS distinguishes between nine American lifestyles based on a range of values, attitudes, behaviors, and demographics. The Advertising Age selected VALS as one of the 10 marketing breakthroughs of the 1980s, and the methodology continues to be marketed to businesses and organizations as a key tool for guiding strategic planning, product positioning, and communications.

In the digital age, segmentation continues to play a central role in advertising strategies; marketers work closely with companies like Google and Facebook to tailor messages based on individuals’ online profiles and activities. Widespread commercial use suggests that this is a cost-effective and successful approach for increasing sales revenue and profit. However, solid empirical evidence relating to effectiveness of tailored and targeted advertising, at least in the public domain, was surprisingly sparse.

Nowhere has segmentation in the digital environment been used in a more sophisticated manner than in political campaigns. Campaign organizations and political data management firms regularly buy and trade personal information about prospective voters, and use this information to identify and target audience segments (Kreiss & Jasinski, 2016). As digital technologies continue to encroach into modern life, campaigns continue to collect unprecedented amounts of personal information about potential voters and their likely support for each candidate, and to pinpoint messages and advertisements using SMS and ads tied to social media and search engine behavior. The sector has undergone explosive growth: In the 12 years from 2004, Democrats in the United States founded 67 firms and organizations dedicated to digital campaigning and data analytics. In the 2008 presidential campaign, Obama’s campaign team hired 131 data staffers: a number that had grown to 342 staffers by the time of the 2012 presidential race (Kreiss & Jasinski, 2016).

Interestingly, and relevant to climate change communication campaigns—where views often diverge along political lines—a recent survey indicated that 86% of Americans were opposed to having political advertisements tailored to their interests. This suggests “a deep discomfort over behavioral targeting and tailored advertising when it comes to politics” (Turow, Delli Carpini, Draper, & Howard-Williams, 2012, p. 4).

Applications in Physical and Social Sciences

Segmentation strategies also have played a prominent role in the physical and social sciences for purposes unrelated to persuasion and marketing. A few notable examples follow. Aristotle introduced a basic taxonomy for classifying living organisms in his History of Animals, which provided the foundation for Linnaeus’s hierarchical system of kingdom, class, order, genus, and species (Pellegrin & Preus, 1986). Within the social sciences, Dewey introduced the concept of “multiple publics”—groups that share similar values and interests with respect to a given social or political issue (Dewey & Rogers, 2012). And psychologists have a long history of interest in personality typologies. Carl Jung developed a psychological theory that proposed eight personality types based on combinations of (1) introversion versus extraversion, and (2) four psychological functions: sensation, intuition, thinking, and feeling (Jung, 2017). Jung’s model in turn provided the conceptual foundation for the Myers-Briggs Type Indicator, which continues to be used extensively in workplace settings, despite concerns about its reliability and validity (Pettinger, 2005).

Social Marketing

In the early 1970s, interest grew in applying traditional marketing principles to issues of public interest. Kotler and Zaltman (1971) coined the term social marketing, promoting it as a marketing-oriented framework for developing and implementing strategies for progressive social change. The difference between social marketing and traditional marketing primarily comes down to objective. Whereas traditional marketing aims to increase the awareness, consideration, and ultimately sales of products and services, and therefore increase profit, social marketing aims to benefit society and individuals in market segments by influencing behaviors.

The most widespread users of social marketing have been health communicators targeting behaviors such as smoking, alcohol abuse, the transmission of sexually transmitted diseases, and physical activity (Lefebvre & Flora, 1988; Mathijssen, Janssen, van Bon-Martens, & van de Goor, 2012; Rimal et al., 2009). Within the health domain, psychological models often serve as the basis for determining the segments that are to receive targeted communications. For example, the Transtheoretical Model (Prochaska, DiClemente, & Norcross, 1992) divides the health behavior change process into five distinct stages: Pre-contemplation (not planning to change), Contemplation (intending to change within the next six months), Preparation (intending to change soon and preparing for change), Action (currently going through the change process), and Maintenance (sustaining change for six months or more). Social marketers segmenting on this basis often attempt to match communications to each stage of change, intending to move individuals along or through the various stages until action and maintenance are ultimately achieved. Meta-analytic research indicates that targeted health messages achieve significantly greater behavior change than generalized, non-targeted messages, although the effects are modest in magnitude, at least for print (Noar, Benac, & Harris, 2007) and computer interventions (Krebs, Prochaska, & Rossi, 2010).

Applications to Climate Change

Climate change communication researchers and practitioners have become increasingly interested in applying social marketing techniques, including audience segmentation, in an effort to increase public engagement. Table 1 contains an updated summary of this research, building upon an earlier review by Hine et al. (2014).

Most climate change segmentation research has identified groups that share similar beliefs, attitudes, and behaviors related to climate change within large national samples. The Yale Project on Climate Change Communication (2009), conducted in collaboration with George Mason University, is the most influential and well-known of such programs. Data for its initial segmentation study were collected from a large nationally representative sample of U.S. residents in 2008. Segmentation based on 36 variables assessing climate change beliefs, issue involvement, policy preference, and behaviors revealed six distinctive segments: Alarmed, Concerned, Cautious, Disengaged, Doubtful, and Dismissive. Collectively labeled the Six Americas, the segments reflect quantitative shifts from generally high to generally low levels of concern, issue engagement, and degree of certainty that global warming is occurring.

Follow-up studies by the Yale/George Mason group have monitored how the proportion of respondents in each segment has changed across time (Leiserowitz et al., 2014; Leiserowitz, Maibach, & Roser-Renouf, 2010; Leiserowitz, Maibach, Roser-Renouf, Feinberg, & Howe, 2013a, 2013b; Leiserowitz, Maibach, Roser-Renouf, Feinberg, & Rosenthal, 2015; Leiserowitz, Maibach, Roser-Renouf, & Hmielowski, 2012; Leiserowitz, Maibach, Roser-Renouf, & Smith, 2011; Roser-Renouf et al., 2014; Roser-Renouf, Maibach, Leiserowitz, & Rosenthal, 2016). Segment proportions fluctuated slightly over the 8-year period from 2008 to 2016, but no clear trajectory toward increasing denial or acceptance of climate change was evident. In 2016, membership of the Alarmed segment had increased after a slump and returned to a level that was comparable to the high levels observed in 2008.

Studies Segmenting on Climate Change-Related Beliefs, Behaviors, and Policy Preferences

Large national samples from Australia (Morrison, Duncan, & Parton, 2013, 2015; Sherley, Morrison, Duncan, & Parton, 2014), India (Leiserowitz, Thaker, Feinberg, & Cooper, 2013), and Germany (Metag, Füchslin, & Schäfer, 2015) have been segmented using variations of the Yale/George Mason group’s measures and methodology. In a comparative study, Morrison and colleagues (2013) found that the climate change attitudes and behaviors of Australians were less polarized than U.S. residents, with fewer respondents classified as Alarmed or Concerned and more classified in the centrist Cautious and Disengaged groups. Segmentation of a sample of the Indian population, using a subset of the Yale/George Mason group’s measures, identified six segments: Informed, Experienced, Undecided, Unconcerned, Indifferent, and Disengaged. Notably, the proportion of Disengaged respondents in India was considerably larger in rural settings (19%) than in urban settings (10%) (Leiserowitz et al., 2013). Employing a representative sample of German residents, Metag and colleagues (2015) factor-analyzed responses to a set of questions that largely corresponded with the Yale/George Mason Group’s measures. Cluster analysis of the resulting seven factor scores identified five segments that were collectively labeled the Five Germanys: Alarmed, Concerned Activists, Cautious, Disengaged, and Doubtful. This finding revealed a relatively high degree of concern about climate change in the German public; the highly dismissive segment that emerged in the United States, Australian, and Indian publics was absent in Germany.

Samples from other countries have been segmented using a broad range of segmentation variables and methodologies. The diversity of resulting solutions highlights the important influence of variable choice and profiling strategy on segmentation outcomes. In one of the most ambitious projects, the British Broadcasting Corporation’s Climate Asia initiative conducted interviews with over 33,000 urban and rural residents from China, Indonesia, Nepal, Bangladesh, Vietnam, Pakistan, and India (Colom & Pradhan, 2013; Copsey, Dalimunthe, Hoijtink, & Stoll, 2013; Copsey, Hoijtink, Shi, & Whitehead, 2013; Gambhir & Kumar, 2013; Manum, Stoll, & Whitehead, 2013; Zaheer & Colom, 2013). Cluster analyses were conducted on responses to perceived impact of climate change, barriers related to information and resources, community cooperation, likelihood to take action in response to future changes, and current behavioral responses. Five segments were identified in all seven countries: Surviving, Struggling, Adapting, Willing, and Unaffected.

In New Zealand, residents’ responses to two survey questions assessing beliefs about the reality and anthropogenic causes of climate change formed the basis of four distinct segments: Climate Believers, Undecided/Neutral, Climate Skeptics, and Anthropogenic Climate Skeptics (Sibley & Kurz, 2013). In a large combined U.S. and Swiss sample, latent class analysis of responses to climate change-related measures of altruism, moral values, and social approval produced four segments that were associated with low to high willingness to pay for carbon offsets (Blasch & Ohndorf, 2015). This use of cognitive profiling variables contrasts with a study in Finland that grouped young adults into segments based on their engagement in 14 climate change mitigation behaviors (Korkala, Hugg, & Jaakkola, 2014). Similarly, latent class analysis of variables assessing engagement in household pro-environmental behaviors revealed four segments of the Hungarian public (Tabi, 2013). Interestingly, residential CO2 emissions did not differ across these segments. This underscores the importance of incorporating measures of ecological impact into research projects. Changing behavior does not always directly translate into expected environmental changes. It is important to understand when and why this does not occur.

Studies Incorporating Non-Climate-Change Variables

The research reviewed in the previous subsection segmented respondents according to their climate change-related beliefs, behaviors, and policy preferences. Other studies have taken a different tack and employed a wider range of psychological variables, some of which are not explicitly related to climate change. In Australia, Hine and colleagues (Hine et al., 2016; Hine, Phillips, et al., 2013; Hine, Reser, et al., 2013) used sets of profiling variables that included environmental values, trust, emotional responses, and spatial and temporal discounting, as well as the more standard climate change belief variables. Five segments (Dismissive, Doubtful, Uncertain, Concerned, and Alarmed) emerged in one sample (Hine, Reser, et al., 2013) and three segments (Dismissive, Uncommitted, and Alarmed) were found in a second sample using a different sampling strategy and a modified set of profiling variables (Hine et al., 2016). Significant differences emerged across segments on a range of validation dimensions, including climate change mitigation behaviors and energy policy preferences.

In the United Kingdom, the Department for Environment, Food and Rural Affairs (DEFRA) took a similarly broad approach, by segmenting 3,600 English residents according to their environmental attitudes, beliefs, and behaviors; including (but not restricted to) climate change (DEFRA, 2008). The seven emergent segments (Positive Greens, Waste Watchers, Concerned Consumers, Sideline Supporters, Cautious Participants, Stalled Starters, and Honestly Disengaged) differed in motivations, barriers, and the degree to which they engaged in climate change mitigation behaviors (Barr, Gilg, & Shaw, 2011). Poortinga and Darnton (2016) segmented a large representative sample of the Welsh population according to various human values about sustainability and community, along with perceptions of climate change and energy security. They found six segments (Enthusiasts, Pragmatists, Aspirers, Community Focused, Commentators, and Self-Reliant) that exhibited widely varying sociodemographic characteristics and patterns of environmental behavior.

Additionally, researchers have used resource-related characteristics to segment the public. For example, researchers have identified six segments of Swiss energy consumers that varied in the importance they placed on various energy-saving efforts, values, and convenience (Sütterlin, Brunner, & Siegrist, 2011); three groups of potential green electricity adopters in Hungary that reported different preferences for renewable energy (Tabi, Hille, & Wüstenhagen, 2014); and five segments of Australian residents that exhibited different levels of support for water-saving policy initiatives (Dean, Lindsay, Fielding, & Smith, 2016). Other studies have identified key characteristics of audience segments that may be persuaded to adopt pro-environmental practices, such as technology (Axsen, Tyreehageman, & Lentz, 2012) or plug-in electric cars (Axsen, Bailey, & Castro, 2015).

Theory-Driven Segmentation Studies

Theory-driven climate change segmentation studies are much less common. Using the Risk Perception Attitude framework (Rimal & Real, 2003), Mead and colleagues (2012) classified 523 American parent and adolescent pairs into four groups according to their risk perceptions and personal efficacy beliefs about managing climate change: Indifferent, Proactive, Avoidant, and Responsive. Adolescents in the responsive and avoidant groups sought significantly more information about climate change than the indifferent group. Kahan and colleagues (2011) applied the cultural cognition model to classify 1,500 U.S. respondents into groups based on their preferences for social order. Consistent with previous studies (Kahan, Braman, Gastil, Slovic, & Mertz, 2007; Kahan, Wittlin, et al., 2011; Leiserowitz, 2006), individuals with an Egalitarian-Communitarian worldview expressed significantly higher levels of belief in climate change than individuals classified as Hierarchical-Individualists. Finally, EcoAmerica (2008, 2011) used the Values Attitude Lifestyle System (VALS) (Strategic Business Insights, 1978) to classify two large samples of U.S. residents into eight segments: Innovator, Thinker, Believer, Achiever, Striver, Experiencer, Maker, and Survivor. Innovators tended to report the highest levels of concern about and belief in climate change, and achievers tended to report the least concern and belief.

Communicating With Audience Segments

Interest is growing in how specific audience segments respond to different types of climate change communications and interventions. The Yale/George Mason group’s initial report (2009) concluded that overall levels of media use were similar across the Six Americas segments, but they differed in the attention they placed on specific forms of news, information seeking, trust in information sources, and frequency of attending to particular genres. For example, members of the Alarmed segment actively sought out and attended to political, environmental, and scientific news, and tended to trust scientists and environmental organizations as sources of information. On the other hand, individuals who were dismissive about climate change tended to restrict themselves to media sources reflecting their own point of view and trusted their own friends and families as sources of information. A longitudinal analysis of the Six Americas’ 2008 and 2011 data found that the two most engaged segments interpreted their personal experiences of climate change in a way that reaffirmed their existing climate change beliefs, whereas personal experiences influenced the beliefs of the four least engaged segments (Myers, Maibach, Roser-Renouf, Akerlof, & Leiserowitz, 2013). Given that approximately 75% of the U.S. population belongs to the four least engaged segments, this finding suggests that conveying information about local climate change effects may be an effective mass communication strategy for the American national audience.

Metag et al. (2015) found that all Five Germanys segments used television as their main source of climate change information. However, similar to the Yale/George Mason group’s (2009) findings, individuals who were alarmed about climate change sought information in the mass media most frequently and doubtful respondents searched the least. Based on their findings, Metag and colleagues suggested that the three least engaged segments might be motivated by different communication methods. Because individuals who are doubtful about climate change do not intentionally seek information about the issue, they might be reached by concern-inducing television messages presented incidentally during their everyday media use. In contrast, disengaged individuals might respond most favorably to entertainment (e.g., The Day after Tomorrow) and to campaigns that present basic information about climate change and recommend inexpensive methods for changing behavior. Finally, communication campaigns for the cautious should focus on providing information about environmental problems and everyday behaviors to mitigate climate change.

In Australia, Morrison and colleagues (2013) examined how to engage household segments in climate change policy. They found the perceived trustworthiness of celebrities, scientists, and left-wing politicians decreased steadily across the segments from Alarmed to Dismissive. They argued that the Cautious segment was the most politically salient voter segment because its members were relatively open to changing their climate change views and were generally supportive of both government and opposition policies. Consequently, Morrison’s research group (Sherley et al., 2014) further investigated cautious respondents by identifying cautious subsegments (Stay-at-home Parents, Professionals, and Retirees). The group also determined archetypal prototypes and responsiveness to marketing stimuli for each subsegment. Overall, the word “farmer” and images of farmers evoked positive emotions and a desire to act, and only one call to action—“Deteriorating atmosphere is a major issue in the world today”—was received positively by all cautious respondents.

In an another Australian study focusing on how message content influences climate change adaptation intentions, Hine and colleagues (2016) found messages with strong negative emotive content or provided specific adaptation advice increased adaptation intentions in all three of their viewer segments (dismissive, uncertain, and alarmed). The study also found that including information about local impacts and not mentioning climate change was effective in increasing engagement in dismissive audiences. Similarly, Bain and colleagues (2012) demonstrated that pro-environmental messages framed in terms of social welfare and economic development, as opposed to avoiding risks associated with climate change, were more likely to be accepted by Climate Change Deniers.

Using the DEFRA segmentation framework, Horton and Doran (2011) found that shifts between segments could be facilitated by targeting individuals’ beliefs about fairness related to sustainable consumption and climate change. Following focus group sessions, numbers of individuals classified as Positive Greens and Waste Watchers increased and respondents classified as Stalled Starters and Disengaged decreased. Flora and colleagues (2014) found that exposing high school students to an engaging 50-minute entertainment-education presentation on climate science increased their knowledge of climate science, positive engagement with climate change, and almost all assessed conservation behaviors. Maibach and colleagues’ (2011) segmentation tool was used to group students into the Six Americas audience segments approximately 2.5 days before and after the edutainment. Thirty-eight percent of students moved into a more engaged segment after the edutainment, whereas only 13% moved to a less engaged segment. The largest shifts into more engaged segments were from the initially Disengaged and Doubtful groups.

Despite these encouraging results, not all communication interventions have produced desired attitudinal shifts. Lorenzoni and Hulme (2009) used a prescreening survey and factor analysis to create a climate change typology of respondents from the United Kingdom and Italy that consisted of four groups: Deniers, Doubters, Uninterested, and Engaged. They then presented scenarios about future socioeconomic and climate change impacts to all respondents. Although the scenarios elicited considerable deliberation in all groups, they did not generally produce changes in group members’ original views. Forums addressing climate change scenarios were also conducted by Hobson and Niemeyer (2012) in an effort to stimulate “discourse migration” between the five climate change skeptical discourse segments. Although migrations occurred, they were rarely sustained and skeptical positions remained. Overall, more of this type of work is required to determine which types of messages and interventions work most effectively with which audiences, and to what extent message effects can be sustained over time.

In summary, interest in how audience segmentation principles can be applied to climate change has increased considerably during the past decade. A broad range of segmentation variables and solutions have been proposed, some focusing primarily on climate change beliefs, behaviors, and policy preferences, and others using more general value-orientation and attitudinal variables to predict behaviors relevant to climate change. To date, most research has focused on identifying audience segments and making recommendations about how to best engage with each segment. Theoretically driven segmentation and empirical evaluations of the impact of different message frames and channels on segments have been far less common, and represent important areas for future research.

Controversies and Unresolved Issues

Despite its intuitive appeal and growing evidence supporting its effectiveness, audience segmentation has not been universally embraced as a research or engagement tool. In this section several controversies and unresolved issues associated with the approach are highlighted.

Is Segmentation Worth the Extra Time and Effort?

Designing, implementing, and evaluating a social marketing program involving audience segmentation, and tailored or targeted messaging can be costly. The time, money, and human resources required for such a strategy may exceed the budgets of many organizations. To justify the additional resources required to identify segments and match messages to them, the increase in impact should outweigh the added cost.

To date there has been little research directly assessing the cost effectiveness of segmentation and message tailoring/targeting within the realm of climate change communication. However, findings from communications studies from other areas, such as health, are instructive. Ishikawa and colleagues (2012) conducted a randomized control trial comparing the cost effectiveness of a tailored reminder for breast cancer screening with a non-tailored reminder in a non-adherent population. They found that the tailored reminder, relative to the non-tailored one, reduced the cost of recruiting each new mammography client from $52 to $30. However, a similar intervention, by the same research team, aimed at increasing screening for colorectal screening was no more effective than an unmatched control message (Hirai et al., 2016).

Results from a meta-analysis evaluating the effectiveness of segmentation/tailoring in the health domain found that tailored print communications only resulted in better outcomes than non-tailored communications by an average effect size (r) of .07 (Noar et al., 2007)—a statistically reliable, but surprisingly small improvement. If this small effect size is indicative, this suggests that segmentation, tailoring, and targeting may be best suited to large-scale communication efforts. For example, Hine and colleagues (2014) noted that interventions with small effect sizes can still have large practical impacts if applied on a large enough scale. They noted that in a population of 2 million people, r = .07 translates to 70,000 more people engaging in the target behavior compared to a non-tailored control. Given that many of the costs associated with matching messages to audience segments are present irrespective of the size of the target populations, the approach is likely to be less cost effective for smaller audiences. But prior to dismissing the approach as inappropriate for small populations, it is worth noting that (1) effect sizes associated with tailoring and targeting will likely increase as communicators continue to refine their methodologies, making such work more cost effective; and (2) creative strategies have been suggested for implementing segmentation strategies on “shoestring” budgets (Slater, 1996).

How Should Segmentation Variables Be Selected?

Many segmentation studies fail to articulate a clear conceptual or theoretical basis for selecting the variables used to segment their samples. Much of the literature has an ad hoc exploratory feel. Although exploratory segmentation can be appropriate in some circumstances, it is important to appreciate that climate change communication programs are usually designed with specific goals in mind, and it is important to select segmentation variables that are aligned with these goals. As Hine and colleagues (2014) noted:

If the goal is to educate, climate change knowledge should be included in one’s set of profiling variables to identify knowledge gaps, and the specific audience segments in which they occur. If the goal is to facilitate behaviour change, segmentation should be focused on key drivers and barriers associated with desired behaviours; recognizing, of course, that lack of knowledge may constitute a key barrier to action. For other campaigns, such as Common Cause (Crompton, 2010), which have broader aims related to changing cultural values and worldviews, it would be essential to know the current distribution of these values and worldviews within the target audience.

(p. 450)

Psychological and behavioral theories represent a second useful way to help guide the choice of segmentation variables. Theory-driven segmentation is also underutilized climate change communication. This is somewhat surprising given the extensive collection of theories relevant to understanding individuals’ perceptions and responses to climate change. For example, in a recent review, Michie, West, Campbell, Brown, and Gainforth (2014) identified over 80 psychological theories and models relevant to understanding behavior and behavior change. An earlier review by Darnton (2008) identified over 60. This rich theoretical base could be extremely useful in shaping future climate change segmentation work.

Does Segmentation Increase Stigmatization and Social Distance?

Classifying individuals into distinct segments or types has significant implications for how they are viewed and treated by others. The American Psychological Association has linguistic guidelines against the use of labels that essentialize categorical membership at the expense of people’s individuality and personhood. Maass, Suitner, and Merkel (2014) argued that noun labels are particularly problematic in this respect.

Nouns are the grammatical form used to indicate objects and hence their application to human beings may promote an object-like perception. For instance, referring to someone as “the depressive” may prevent listeners from noting other characteristics of the person .… As we do for objects, when we label a person with a noun, we depersonalise him/her, thus equating the person to his/her conditions or group membership.

(p. 336)

Howell, Ulan, and Powell (2014) found that individuals who used noun labels held more stigmatizing attitudes and exhibited less empathy toward those experiencing mental illness. Noun labels, compared to adjective and verb labels, have also been found to encourage stereotyping, and the perception of personal qualities that are more fundamental and permanent (Carnaghi et al., 2008). The same principles apply to climate change segmentations in which individuals are sometimes labeled as “skeptics” or “alarmists.” By using such labels, segmentation practitioners may inadvertently reinforce divisive stereotypes, increase social distance, and undermine the ability of community members to work together to create collectively beneficial solutions.

Corner and Randall (2011) made a similar point in their critique of social marketing applications to climate change. They argued that segmentation might be fundamentally discriminatory, because it treats individuals differently based on selected personal characteristics: an approach that can marginalize groups identified as difficult to access or influence. Furthermore, segmenting audiences, and targeting messages to specific segments, may accentuate differences between members of these segments and diminish a sense of shared collective responsibility within communities. This accentuation can ultimately undercut the empathy and social capital needed for pro-environmental change.

Although we generally agree with the view that segmentation has the potential to widen the divides within a community, we do not believe that such outcomes are inevitable. Rather than devising strategies that accentuate differences, communicators could promote a more unified collective mindset about climate change, such as that advocated by the Common Cause program (Ling, Franklin, Lindsteadt, & Gearon, 1992). Under this approach, messages tailored and targeted to segments would aim to bring groups closer together, not divide them further. Hine and colleagues’ (2016) preaching to different choirs study illustrates this point. Although they found that several message types resonated more strongly with skeptical audiences than alarmed and uncommitted ones, they also identified types that increased climate change adaptation intentions across all segments. In short, segmentation need not be used exclusively to tailor messages to distinct audiences. Understanding segment similarities and differences can also be used to identify messages that work for everyone.

Does Segmentation Encourage “Shallow” as Opposed to “Deep” Change?

Social marketing in the environmental domain is sometimes criticized for achieving superficial or shallow change by focusing too narrowly on changing specific problematic behaviors, as opposed to underlying worldviews and values that may have a more substantial and enduring effect on behavior change. Thogerson and Crompton (2009) posited that communications tailored to segments might achieve short-term success but do so at the possible cost of negative longer-term effects: While short-term campaigns might win the day they can unintentionally reinforce worldviews that are incompatible with sustainable lifestyles. For example, the DEFRA segmentation identified a group of individuals labeled Waste Watchers, who engaged in pro-environmental behaviors primarily for financial gain (Department for Environment Food and Rural Affairs, 2008). If communicators were to repeatedly target this group with emphasis on financial benefits, they would risk inadvertently reinforcing a highly individualistic “what’s in it for me” mindset. If, at some point in the future, incentives were no longer present, this group’s pro-environmental behavior might be undermined. Crompton (2010) noted that such inadvertent reinforcement is particularly problematic for bigger-than-self problems such as climate change, which will require a shift away from rational self-interest to coordinated collective responses.

However, it is important to appreciate that audience segmentation need not be used to reinforce existing worldviews that are incompatible with sustainability and/or long-term climate change action. Crompton (2010) acknowledged that audience segmentation could prove to be a valuable approach for facilitating transformational changes within communities. By understanding how members of each segment view and understand the world, climate change communicators are in a strong position to identify language and metaphors that prime broad, community-minded, values that help segmented members conceptualize and respond to issues like climate change in more constructive ways. That is, segmentation can be used as a starting point for conversations leading to deeper shifts in worldviews and values that are more compatible with sustainability and a stable climate.

Are Segments “Found” or “Constructed”?

Debate exists over whether segments are “found” in audiences, via statistical techniques such as cluster analysis, or are “constructions,” based on the theories and methods of researchers. Social marketing assumes that distinct groups exist in populations and that these groups can be “discovered” using appropriate empirical methods. A contrary perspective, from the critical social sciences (Gandy, 2001), is that segmentation analyses depend on the variables included and excluded, and that decisions about what to include and exclude are often influenced by theory and practical constraints. Furthermore, the number and nature of segments may change in response to targeted communications, with new segments emerging, individuals moving between segments, and/or once prominent segments becoming extinct as public understanding evolves. Thus, identified segments are largely the product of theoretical and methodological choices made during the research process.

We believe that segmentation, like all other areas of science, combines discovery and construction. Genuine differences do exist between individuals in terms of their values, beliefs, preferences, and behaviors, and these differences can be measured. But we also accept that audience segmentation is not value neutral: Our choice of theories and methods can have profound impacts on segmentation solutions and therefore on the communities we study (Thogerson & Crompton, 2009). As a result, climate change communicators and policy makers must recognize that no single segmentation solution is objectively correct; many valid solutions may exist.

Segments versus Dimensions?

Within psychology, there has been a long-standing debate about the nature of personality and clinical disorders. Some have argued that many clinical diagnostic categories and personality types are more appropriately conceptualized as dimensional latent variables (Haslam, Holland, & Kuppens, 2012). That is, differences between individuals are better conceived as quantitative differences along one or more continua as opposed to qualitative differences reflecting distinct category membership. It is worth noting that climate change segmentation works implicitly assume the existence of multiple homogeneous audiences that differ qualitatively from each other. This assumption seems perfectly reasonable for some segmentation solutions, such as Morgan and colleagues’ (2015) distinction between four major groups of Australian farmers: Non-Green Dismissive, Uncommitted, Green Adopters, and Profit-Driven Adopters. However, for other segmentation solutions—for example, the Yale/George Mason group’s 6 Americas and Hine et al.’s (2014) 5 Australias—which include segments that range from Dismissive to Alarmed, the case for qualitatively distinct dimension, as opposed to single dimension of climate change concern, becomes more difficult to accept as truly categorical. A range of techniques have been developed by Meehl and his colleagues to assess the relative fit of categorical and dimensional models of latent variables (Meehl, 1992, 1995). To date, these techniques have not been applied to climate change segmentation solutions.

General Conclusions

Interest is growing in applying audience segmentation to improve climate change communication. Individuals vary considerably in their interest and motivations related to climate change, and advocates of segmentation believe that it is important to identify the number and nature of public audiences prior to developing climate change engagement strategies. Applications from marketing, politics, and health communication indicate that tailoring/targeting messages to homogeneous subtypes is effective, although the magnitude of the effects are perhaps more modest than one might expect.

Application of segmentation methodology to climate change communication is still in its infancy. An impressive and rapidly growing literature has revealed that distinct audience segments are present in a broad array of countries around the world. However, too few studies have explicitly measured the immediate impact of tailored messages (compared to non-tailored) on different segments, let alone the longer-term impacts on values, beliefs, behavior, or segment membership. Also, relatively few studies have explicitly used theory to guide the choice of segmentation and validation variables, which partly explains the variability in segments identified in the literature. As climate change segmentation research becomes more mature, we expect to see much more work addressing these limitations.

Concerns have been raised about segmentation’s cost effectiveness, its potential to increase stigmatization and social distance in communities, its ability to deliver deep and lasting change, and the extent to which segmentation solutions are dependent on researchers’ choice of measures and methodologies. Although these concerns are valid, we do not believe they are substantial enough to disqualify segmentation as a valuable technique for enhancing climate change communication. Methods for segmentation, tailoring, and targeting are constantly being refined to improve cost efficiency and increase impact. Segment labels can be carefully chosen to minimize stigmatizing effects, and messages can be specifically designed to promote community cohesion, and also to prime commonly shared values that can help drive collective action toward positive social and environmental outcomes.

Finally, it is also true that segmentation solutions are heavily dependent on the methods employed by climate change researchers and communicators. But this is true of all areas of science. To the extent that “objective reality” exists, scientists always have an obstructed view. Nevertheless, audience segmentation solutions need not be 100% objectively valid to be functionally useful. Segmentation helps us understand how groups differ in terms of their values, beliefs, and behaviors related to climate change, and provides a basis for engaging with these groups more effectively. In some instances, we will be led astray. But over time, we will learn from our mistakes, refine our methods and measures, and get better and better at the task of understanding the number and nature of audiences in the populations we study, and how to best engage them.

Table 1: Summary Climate Change Segmentation Studies

Study

N

Population

Sampling

Profiling Variables

Analysis

Segments

Objective(s)

Arbuckle et al. (2014)

4,778

U.S. Farmers

NP Mail survey

34 items derived from Maibach et al. (2011) assessing environmental hazards; climate change beliefs; and concern about risk to agriculture, response efficacy, and support for adaptive/mitigative action.

Latent class analysis

6 segments: Concerned, Uneasy, Uncertain, Unconcerned, Confident, and Detached

To identify groups of farmers with similar responses to climate change. To compare climate change responses across segments.

Ashworth, Jeanneret, Gardner, and Shaw (2011)

1,602

Australian residents

NP Online survey

Questions assessing knowledge and concern about climate change.

Two-stage cluster analysisf:

  1. 1. Unspecified hierarchical

  2. 2. k-means

4 dimensional segments ranging from Engaged to Doubtful

To identify groups with similar responses to climate change.

Axsen et al. (2015)

634 (sub-sample)

Canadian potential plug-in electric vehicles (PEV) buyers

  • NP Canadian Plug-in Electric Vehicle Study in 2013.

  • Online surveys with mailed material.

  • Initial design exercise and segmentation identified potential PEV buyers.

  • Clustering then conducted using Axsen et al.’s (2012)

  • 4 environment and technology variables.

k-means cluster analysis

6 clusters: Pro-environmental PEV (Strong, Tech-enviro, Concerned) and Non-environmental PEV (Techie, Open, Unengaged)

To identify groups of potential PEV buyers with similar lifestyle and motivational characteristics.

Axsen et al. (2012)

711

House-holds, San Diego, USA

P Online survey

4 variables assessing engagement in environment- and technology-oriented lifestyle, liminality, and environmental concern.

k-means cluster analysis

5 clusters: Pro-environmental (Engaged, Aspiring, Low-tech) and Non-environmental (Traditional, Techies)

  • To identify groups of consumers with similar attitudes toward pro-environmental technologies (PETs).

  • To identify household and PET-related characteristics of each group.

Bain et al. (2012)

347

Australian residents

NP Online survey

Single question assessing belief in climate change.

Grouped according to response to screening question

2 segments: Climate Change Believers and Deniers

To examine pro-environmental action intentions in response to message frames.

Barnes and Toma (2012)

540

Scottish dairy farmers

P Phone survey

11 factor scores based on climate change perceptions, attitudes, and values.

k-means cluster analysis

6 segments, e.g., Regulation Skeptic, Commercial Ecologist, Innovator, Disengaged, Negativist, Positivist

To identify groups with similar responses to climate change.

  • BBC Climate Asia Project (2013)

  • (Colom & Pradhan, 2013; Copsey, Dalimunthe, et al., 2013; Copsey, Hoijtink, et al., 2013; Gambhir & Kumar, 2013; Manum et al., 2013; Zaheer & Colom, 2013)

33,513

Residents from:

  • Bangla-desh

  • China

  • India

  • Indonesia

  • Nepal

  • Pakistan

  • Vietnam

NP/P Interviews

5 variables assessing perceived impact of climate change, barriers, community cooperation, likelihood to take action in response to future changes, and current behavioral responses.

k-means cluster analysis

5 segments: Surviving, Struggling, Adapting, Willing, Unaffected

To identify groups with similar responses to climate change. To compare urban and rural respondents. To recommend communication strategies. To monitor segment changes over time.

Blasch and Ohndorf (2015)

1,894

U.S. and Swiss residents

NP Online survey

Climate change awareness and concern, ascribed responsibility, perceived social norm for offsetting, and percentage of consumers who offset.

Latent class analysis

4 dimensional segments ranging from Low Offset to High Offset

To identify groups of individuals with similar responses to climate change and carbon offsetting social norms. To compare carbon offset behaviors across segments.

Coles, Zschiegner, and Dinan (2014)

416

Tourism businesses, Southwest England

NP Interview and question-naire

Questions assessed the uptake of innovations that could reduce environmental load.

Two-stage cluster analysis:

  1. 1. Ward’s

  2. 2. k-Means

Stage 1 found three clusters of mitigation systems and innovations.

  • To identify groups of businesses with similar climate change mitigation behaviors.

  • To compare behaviors and business attributes across segments.

Dean et al. (2016)

5,194

Australian residents

P Online survey

Questions assessed cognitive, emotional, and behavioral engagement with water use and conservation.

Hierarchical cluster analysis (Ward’s)

5 clusters: Disengaged, Aware but Inactive, Active but not Engaged, Engaged but Cautious, and Highly Engaged.

To identify groups of Australians with similar responses to water use. To compare individual characteristics and water policy support across groups.

DEFRA (2008)

3600

UK residents

P Interviews

44 items assessing attitudes, beliefs, and behaviors about the environment (including climate change).

Unspecified cluster analysis

7 dimensional segments ranging from Positive Greens to Honestly Disengaged

To identify groups with similar responses to climate change and the environment.

  • 1707

  • 1221

  • U.S. residents

  • U.S. residents

P Online survey

Used VALS: 35 items assessing values and lifestyles and 4 demographic questions.

Unspecified cluster analysis

8 segments: Thinker, Innovator, Believer, Achiever, Striver, Experiencer, Maker, Survivor

To compare climate change responses across segments.

Flora et al. (2014)

1,241

U.S. high school students

NP Class survey

Maibach et al.’s (2011) 36 items assessing global warming & energy beliefs, engagement, behaviors, and societal responses

Latent class analysis and discriminant function algorithm based on Maibach et al. (2011), assessed before and after exposure to climate science edutainment.

6 dimensional segments ranging from Dismissive to Alarmed

To determine whether exposure to climate science edutainment positively influences youths’ climate change knowledge, beliefs, involvement, and behavior.

Fritz and Koch (2014)

38

Developed countries, national data

Various data sources

Questions assessed indicators or prosperity: ecological sustainability, social inclusion, quality of life, and economic development.

Hierarchical cluster analysis (Ward’s)

5 clusters: Latin America and Turkey, Rich western and northern countries, other rich countries, Russia and China, and eastern and southern European countries.

  • To group countries into “prosperity regimes.”

  • To compare indicators across groups, and assess relationships between indicators.

Haney (2015)

99

Firms that answered the Carbon Disclosure Project survey in 2003, 2006 and 2009

NP Survey, format not reported

7 variables: Interpretation of climate change as Opportunity or Threat and difference between them, tentative and certain language, attention to climate change risks, and total CO2 emissions.

k-means

  • 2003—2 clusters: Threat and Opportunity.

  • 2006—3 clusters: Balance, Opportunity and High Opportunity.

  • 2009—3 clusters: Opportunity Low Attention, Opportunity Medium Attention and Opportunity High Attention.

To identify groups of firms with similar perceptions of climate change. To compare innovative sustainable products and services across segments.

Hine et al. (2013)

1031

Australian residents

NP Online survey

14 variables assessing a range of cognitive and affective responses to climate change and the environment

Latent profile analysis

3 dimensional segments: Dismissive, Uncommitted, and Alarmed

To identify groups with similar responses to climate change and the environment. To identify effective climate change message characteristics for each segment.

Hine et al. (2013)

3096

Australian residents

NP Online survey

12 variables assessing a range of cognitive and affective responses to climate change and the environment.

Latent profile analysis

5 dimensional segments ranging from Dismissive to Alarmed

To identify groups with similar responses to climate change and the environment.

Hine et al. (2016)

1031

Australian residents

NP Online survey

15 variables assessing a range of cognitive and affective responses to climate change and the environment.

Latent profile analysis

3 dimensional segments: Dismissive, Uncommitted, and Alarmed

To identify groups with similar responses to climate change and the environment. To identify effective climate change message characteristics for each segment.

Hobson and Niemeyer (2012)

44

Australian residents

NP Interviews

33 statements relevant to climate change.

Q methodology (Q sort followed by Q factor analysis)

5 overlapping skeptical discourses: Emphatic Negation, Unperturbed Pragmatism, Earnest Acclimatization, Noncommittal Consent

To identify the main skeptical discourses used by Australians. To compare responses to future scenarios (2050 to 2100) depicting various climatic variables.

Horton and Doran (2011)

64

English residents; excluded climate change skeptics and green activists

NP Focus groups

Items assessed attitudes and beliefs regarding the environment, environmental issues, and behaviors. DEFRA block method.

Unspecified

7 dimensional segments ranging from Positive Greens to Honestly Disengaged

To compare climate change responses in relation to fairness across segments. To explore attitude change in response to focus group sessions.

Hyland, Jones, Parkhill, Barnes, and Williams (2016)

286

Welsh livestock farmers

NP Survey, format not reported

4 variables derived from PCA of 29 items on climate change awareness, environmental responsibility, productivism, and perceived risk

Two-stage cluster analysis

  1. 1. Ward’s

  2. 2. k-means

4 clusters: The Environmentalist, The Dejected, The Countryside Steward, and The Productivist

To identify groups of farmers with similar responses to climate change. To compare responses to climate change across groups.

Kahan, Jenkins-Smith, et al. (2011)

1500

U.S. residents

P Online survey

2 composite variables assessing Hierarchy-Egalitarianism and Individualism-Communitarianism cultural worldviews.

Median splits

4 segments: Hierarchical- Individualist, Hierarchical- Communitarian, Egalitarian- Individualist, and Egalitarian- Communitarian

To compare perceptions of scientific consensus about climate change risk across segments.

Kelly et al. (2014)

3,594

U.S. zoo and aquarium visitors

NP Pencil/paper surveys

Maibach et al.’s (2011) 15 item global warming screening tool assessing attitudes toward global warming

Latent class analysis and discriminant function algorithm based on Maibach et al. (2011)

6 segments: Concerned, Uneasy, Uncertain, Unconcerned, Confident, and Detached

To classify zoo and aquarium visitors into “Global Warming’s Six Americas” segments. To compare responses to nature, animals, and global warming across segments.

Korkala et al. (2014)

1,623

Young adults drawn from Espoo cohort, Helsinki, Finland, 1991

NP Mail survey

14 items assessing engagement in each of 14 climate change mitigation actions

Latent class analysis

  • Males—3 classes: Inactive, Semi-active and Active

  • Females—2 classes: Semi-active and Active.

To identify groups of people (by gender) who perform similar climate change mitigation actions. To identify sociodemographic characteristics that predict group membership.

Leiserowitz, Thaker, et al. (2013)d

3138

Indian residents

P Interviews

21 single items and 2 composite variables.

Cluster analysis, plus one additional segment (Disengaged) based on non-responses and don’t know responses

6 dimensional segments: Disengaged, Indifferent, Unconcerned, Undecided, Experienced, and Informed.

To identify groups with similar responses to climate change. To compare urban and rural respondents.

Lorenzoni and Hulme (2009)

341

U.K. and Italian adults and high school students

NP Interviews and discussion groups

9 items relating to the importance of climate change, human influences on the climate, and personal and global effects of climate

Factor analysis, followed by selection of respondents who scored high and low on each factor

4 segments: Deniers, Doubters, Uninterested, and Engaged

To determine how members of each group responded to a set of future scenarios related to socioeconomic and climate change outcomes.

  • Maibach et al., (2011)

  • Segmentation tool also used by Leiserowitz, Maibach, et al. in:

  • September 2008

  • January 2010

  • June 2010

  • May 2011

  • March 2012/Nov 2011

  • September 2012

  • April 2013

  • November 2013

  • October 2014

  • March 2015

  • March 2016

  • 2129

  • 2129

  • 1001

  • 1024

  • 1000

  • 1008

  • 1058

  • 1045

  • 830

  • 1275

  • 1263

  • 1203

US residents

P Online survey

36 variables assessing global warming and energy-related beliefs, engagement, behaviors, and societal responses

Latent profile analysis

6 dimensional segments ranging from Dismissive to Alarmed

To identify groups with similar responses to climate change. To develop a tool to identify these segments in independent samples.

Mead et al. (2012)

523

U.S. parent-adolescent pairs

P Online survey

Risk Perception Attitude Framework: 29 items assessing perceptions of risk and beliefs about personal efficacy.

Median splits of both the risk perception and efficacy belief scores

4 segments: Indifferent, Proactive, Avoidant, Responsive

To compare climate change information-seeking across segments. To examine similarity between parent and adolescent climate change responses.

Metag et al. (2015)

3,000

German residents

P Telephone survey

7 variables derived from PAF of questions largely corresponding with Maibach et al. (2011): Climate change concern, political activism on energy, everyday car use, ecological conservatism, environmental concern, abstention from long car/plane journeys, and use of eco-power.

Hierarchical cluster analysis (Ward’s)

5 segments: Alarmed, Concerned Activists, Cautious, Disengaged, and Doubtful

To identify groups with similar responses to climate change and the environment. To compare perceptions of climate change media and interpersonal communication channels across groups.

Milfont, Milojev, Greaves, and Sibley (2015)

6,489

New Zealand residents

P Data from New Zealand Attitudes and Values Study, 2009.

Sibley & Kurz’s (2013) 2 items assessing beliefs about the reality of climate change and its anthropogenic cause.

Latent class analysis

  • 4 segments: Climate Believers, Undecided/Neutral, Climate Skeptics, and Anthropogenic Climate Skeptics

To compare distinct demographic, personal values, and personality traits across groups.

Morgan, Hine, Bhullar, and Loi (2015)

551

Australian farmers

NP Online survey or mail survey

11 variables assessing a range of psychological and behavioral measures related to low emission agricultural practices (LEAP).

Latent profile analysis

4 segments: Non-Green Dismissive, Uncommitted, Green Adopters, and Profit-Driven Adopters

To identify groups of farmers who share similar values, beliefs, and personality orientations relevant to LEAP adoption.

Morrison, Duncan, Sherley, and Parton (2013)

1927

Australian residents

NP Online survey

Maibach et al.’s (2011) 36 items assessing global warming and energy beliefs, engagement, behaviors, and societal responses.

Latent class analysis and discriminant function algorithm based on Maibach et al. (2011)

6 dimensional segments ranging from Dismissive to Alarmed

To identify groups with similar responses to climate change. To compare Australian and U.S. segments.

Morrison, Duncan, and Parton (2013)

1,927

Australian respondents

P Online survey

Maibach et al.’s (2011) 36 items assessing global warming and energy beliefs, engagement, behaviors, and societal responses

Latent class analysis and discriminant function algorithm based on Maibach et al. (2011)

6 dimensional segments ranging from Dismissive to Alarmed

To compare preferences for political party and climate change policies, and trusted information sources, across segments.

Morrison et al. (2015)

1,927

Australian respondents

P Online survey

Maibach et al.’s (2011) 36 items assessing global warming and energy beliefs, engagement, behaviors, and societal responses.

Latent class analysis and discriminant function algorithm based on Maibach et al. (2011)

6 dimensional segments ranging from Dismissive to Alarmed

To examine how religious affiliations differed across segments.

Muriuki, Dowd, and Ashworth (2016)

993

Residents of Greater Brisbane, Australia

P Online survey

11 variables derived from PCA of items economic, environmental and social sustainability, social identities, and lifestyle choices.

Latent class analysis

4 segments: Mainstream Pragmatics, Sustainability Advocates, The Social Sustainability Frugals, and The Indifferent

To identify groups of Brisbane residents with similar psychographic, social identities and lifestyle characteristics.

Myers, Nisbet, Maibach, and Leiserowitz (2012)

1127

U.S. residents

P Online survey

Maibach et al.’s (2011) 36 items assessing global warming and energy-related beliefs, issue involvement, efficacy and conservation behaviors, and preferred societal responses.

Latent class analysis

6 dimensional segments ranging from Dismissive to Alarmed

To compare the effectiveness of climate change message framings across groups.

Poortinga and Darnton (2016)

1,538

Welsh residents

P Face-to-face interviews

13 indices of personal values, views on sustainability and sustainable living, attitudes to climate change and energy security, and attitudes to community and place.

Two-stage cluster analysis:

  1. 1. Ward’s

  2. 2. k-means

6 segments: Enthusiasts, Pragmatists, Aspirers, Community Focused, Commentators, and Self-Reliant

To identify groups with similar views on sustainability. To identify each group’s socio-demographics and environmental behaviors. To develop a brief screening tool to identify the segments.

Sherley et al. (2014)

502 (sub-sample)

Australian residents

P Online survey and interviews

  • Maibach et al.’s (2011) segmentation procedure identified Cautious participants.

  • 7 sociodemographic variables then used to identify cautious subsegments.

Latent class analysis

3 subsegments: Stay-at-home Parents, Professionals, and Retirees

To identify subsegments of Cautious respondents. To identify archetypal prototypes and responsiveness to various marketing stimuli for each subsegment.

Sibley and Kurz (2013)

6,072

New Zealand residents

P Data from New Zealand Attitudes and Values Study, 2009.

2 items assessing beliefs about the reality of climate change and its anthropogenic cause.

Latent class analysis

4 segments: Climate Believers, Undecided/Neutral, Climate Skeptics, and Anthropogenic Climate Skeptics

To identify groups with similar beliefs about the reality and anthropogenic cause of climate change.

Sütterlin et al. (2011)

1,292

Swiss households

P Survey

17 variables assessed various energy-saving actions, motives, values, norms, efficacy beliefs, convictions, and comfort.

Hierarchical cluster analysis (Ward’s)

6 segments: idealistic, selfless inconsequent, thrifty, materialistic, convenience-oriented indifferent, and problem-aware well-being-oriented

To identify groups of energy consumers with similar energy-saving characteristics.

Tabi et al. (2014)

414

German electricity consumers

P Face-to-face interviews

A range of sociodemographic, Psychographic, and behavioral variables.

Latent class analysis followed by hierarchical Bayes estimation

5 segments: Adopters, Potential Adopters (Truly Greens, Price Sensitive Greens, Local Patriots), and Likely Non-Adopters

  • To identify groups of electricity consumers with similar characteristics.

  • To identify differences between green electricity adopters and potential adopters.

Tabi (2013)

1,012

Hungary residents

P Interviews

Questions assessed engagement in several household pro-environmental behaviors (e.g., recycling, energy use, transport).

Latent class analysis

4 segments: Browns, Beginners, Energy Savers, and Supergreens

To identify groups with similar pro-environmental behaviors. To compare CO2 emissions across groups.

Thornton et al. (2011)

3,923

English residents

P Interviews

Composite variables derived from PCA of items on environmental attitudes and behaviors, and transport behavior.

Unspecified cluster analysis

9 segments: Older, less mobile car owners, Less affluent urban young families, Less affluent older skeptics, Affluent empty nesters, Educated suburban families, Town and rural heavy car use, Elderly without cars, Young urbanites without cars, Urban low income without cars,

  • To compare climate change responses across segments.

  • To compare attitudes to transport choices across groups.

Waitt et al. (2012)

878

Households, Wollon-gong, Australia

P Mail survey

Items assessing socioeconomic characteristics, pro-environmental household consumption practices, judgments of climate change and place-based attachments.

Two-stage cluster analysis:

  1. 1. Ward’s

  2. 2. k-Means

  • Stage 1 identified four groups of consumption practices.

  • Stage 2 clustered households into Strong Engagement, Modest Engagement, and Limited Engagement.

To compare climate change responses across segments.

Wilson, Howard, and Burnett (2014)

303

Corn and soybean farmers, Ohio, USA

P Mail survey

Questions assessing demographics, attitudes toward taking additional action, and the goals of profit and environmental stewardship.

Latent class analysis

2 segments: Majority and Minority

To determine characteristics and adaptive actions of homogenous groups of farmers.

Yue et al. (2016)

2,380

Residents of Minnesota and Texas, USA; and Ontario, CA.

NP Experimental auction and paper survey

Data from an experimental auction of a set of plants with various production methods, product origin, and plant containers.

Latent class analysis

3 classes: Import-Liking Consumers, Mainstream Consumers and Eco- Local Consumers

To identify groups of consumers with similar willingness to pay for plants with sustainable attributes.

Notes:

(*) For sampling, NP = nonprobability sample and P = probability sample. All samples comprised adults aged 18 years and over except for Flora et al. (2014), Mead et al. (2012), and Lorenzoni and Hulme (2009).

Acknowledgments

The authors thank Edward Maibach for his insightful feedback on an earlier draft of this article.

Corner, A., & Randall, A. (2011). Selling climate change? The limitations of social marketing as a strategy for climate change public engagement. Global Environmental Change, 21(3), 1005–1014.Find this resource:

Hine, D. W., Reser, J. P., Morrison, M., Phillips, W. J., Nunn, P., & Cooksey, R. (2014). Audience segmentation and climate change communication: Conceptual and methodological considerations. Wiley Interdisciplinary Reviews-Climate Change, 5(4), 441–459.Find this resource:

Maibach, E. W., Leiserowitz, A., Roser-Renouf, C., & Mertz, C. K. (2011). Identifying like-minded audiences for global warming public engagement campaigns: An audience segmentation analysis and tool development. PLoS ONE, 6(3), e17571.Find this resource:

Slater, M. D. (1996). Theory and method in health audience segmentation. Journal of Health Communication: International Perspectives, 1(3), 267–284.Find this resource:

References

Arbuckle, J. G., Hobbs, J., Loy, A., Morton, L. W., Prokopy, L. S., & Tyndall, J. (2014). Understanding corn belt farmer perspectives on climate change to inform engagement strategies for adaptation and mitigation. Journal of Soil and Water Conservation, 69(6), 505–516.Find this resource:

Aristotle. (2014/350 B.C.E.). The art of rhetoric. Toronto: HarperCollins Canada.Find this resource:

Ashworth, P., Jeanneret, T., Gardner, J., & Shaw, H. (2011). Communication and climate change: What the Australian public thinks. CSIRO Report EP112769. National Research Flagships: Energy Transformed.Find this resource:

Axsen, J., Bailey, J., & Castro, M. A. (2015). Preference and lifestyle heterogeneity among potential plug-in electric vehicle buyers. Energy Economics, 50, 190–201.Find this resource:

Axsen, J., Tyreehageman, J., & Lentz, A. (2012). Lifestyle practices and pro-environmental technology. Ecological Economics, 82, 64–74.Find this resource:

Bain, P. G., Hornsey, M. J., Bongiorno, R., & Jeffries, C. (2012). Promoting pro-environmental action in climate change deniers. Nature Climate Change, 2(8), 600–603.Find this resource:

Barnes, A. P., & Toma, L. (2012). A typology of dairy farmer perceptions towards climate change. Climatic Change, 112(2), 507–522.Find this resource:

Barr, S., Gilg, A., & Shaw, G. (2011). Helping people make better choices: Exploring the behaviour change agenda for environmental sustainability. Applied Geography, 31(2), 712–720.Find this resource:

Blasch, J., & Ohndorf, M. (2015). Altruism, moral norms and social approval: Joint determinants of individual offset behavior. Ecological Economics, 116, 251–260.Find this resource:

Carnaghi, A., Maass, A., Gresta, S., Bianchi, M., Cadinu, M., & Arcuri, L. (2008). Nomina sunt omina: On the inductive potential of nouns and adjectives in person perception. Journal of Personality and Social Psychology, 94(5), 839.Find this resource:

Coles, T., Zschiegner, A.-K., & Dinan, C. (2014). A cluster analysis of climate change mitigation behaviours among SMTEs. Tourism Geographies, 16(3), 382–399.Find this resource:

Colom, A., & Pradhan, S. (2013). How the people of Nepal live with climate change and what communication can do. London: BBC Media Action Lab.Find this resource:

Copsey, T., Dalimunthe, S., Hoijtink, L., & Stoll, N. (2013). How the people of Indonesia live with climate change and what communication can do. London: British Broadcasting Corporation.Find this resource:

Copsey, T., Hoijtink, L., Shi, X., & Whitehead, S. (2013). How the people of China live with climate change and what communication can do. London.Find this resource:

Corner, A., & Randall, A. (2011). Selling climate change? The limitations of social marketing as a strategy for climate change public engagement. Global environmental change, 21(3), 1005–1014.Find this resource:

Crompton, T. (2010). Common cause: The case for working with our cultural values. Surrey, U.K.: WWF.Find this resource:

Darnton, A. (2008). GSR behaviour change knowledge review. Reference report. An overview of behaviour change models and their uses. London: HMT Publishing Unit, U.K. Government Social Research Unit.Find this resource:

Dean, A. J., Lindsay, J., Fielding, K. S., & Smith, L. D. G. (2016). Fostering water sensitive citizenship—community profiles of engagement in water-related issues. Environmental Science & Policy, 55(1), 238–247.Find this resource:

Demby, E. (1974). Psychographics and from whence it came. In W. D. Wells (Ed.), Life style and psychographics (pp. 9–30). Chicago: American Marketing Association.Find this resource:

Department for Environment Food and Rural Affairs. (2008). A framework for pro-environmental behaviours. London. Retrieved from www.defra.gov.uk.Find this resource:

Dewey, J., & Rogers, M. L. (2012). The public and its problems: An essay in political inquiry. University Park: Pennsylvania State University Press.Find this resource:

EcoAmerica. (2008). The American climate & environmental values survey. Retrieved from http://ecoamerica.org/wp-content/uploads/2013/02/ACVS_Summary.pdf.

EcoAmerica. (2011). The American climate & environmental values survey. Retrieved from http://ecoamerica.org/wp-content/uploads/2013/02/ACEVS_Report.pdf.

Flora, J. A., Saphir, M., Lappé, M., Roser-Renouf, C., Maibach, E. W., & Leiserowitz, A. A. (2014). Evaluation of a national high school entertainment education program: The alliance for climate education. Climatic Change, 127(3), 419–434.Find this resource:

Fritz, M., & Koch, M. (2014). Potentials for prosperity without growth: Ecological sustainability, social inclusion and the quality of life in 38 countries. Ecological Economics, 108, 191–199.Find this resource:

Gandy, O. H. (2001). Dividing practices: Segmentation and targeting in the emerging public sphere. In L. Bennett & R. Entmann (Eds.), Mediated politics: Communications in the future of democracy (pp. 141–159). Cambridge, U.K.: Cambridge University Press.Find this resource:

Gambhir, V., & Kumar, P. (2013). How the people of India live with climate change and what communication can do. London: BBC Media Action Lab.Find this resource:

Haney, A. B. (2015). Threat interpretation and innovation in the context of climate change: An ethical perspective. Journal of Business Ethics, 1–16.Find this resource:

Haslam, N., Holland, E., & Kuppens, P. (2012). Categories vs dimensions in personality and psychopathology: A quantitative review of taxometric research. Psychological Medicine, 42, 903–920.Find this resource:

Hine, D. W., Phillips, W. J., Cooksey, R., Reser, J. P., Nunn, P., Marks, A. D. G., et al. (2016). Preaching to different choirs: How to motivate dismissive, uncommitted, and alarmed audiences to adapt to climate change? Global Environmental Change, 36, 1–11.Find this resource:

Hine, D. W., Phillips, W. J., Reser, J. P., Cooksey, R., Marks, A. D. G., Nunn, P., et al. (2013). Enhancing climate change communication: Strategies for profiling and targeting Australian interpretive communities. Gold Coast, Australia: National Climate Change Adaptation Research Facility.Find this resource:

Hine, D. W., Reser, J. P., Morrison, M., Phillips, W. J., Nunn, P., & Cooksey, R. (2014). Audience segmentation and climate change communication: conceptual and methodological considerations. Wiley Interdisciplinary Reviews-Climate Change, 5(4), 441–459.Find this resource:

Hine, D. W., Reser, J. P., Phillips, W. J., Cooksey, R., Marks, A. D. G., Nunn, P., et al. (2013). Identifying climate change interpretive communities in a large Australian sample. Journal of Environmental Psychology, 36, 229–239.Find this resource:

Hirai, K., Ishikawa, Y., Fukuyoshi, J., Yonekura, A., Harada, K., Shibuya, D., et al. (2016). Tailored message interventions versus typical messages for increasing participation in colorectal cancer screening among a non-adherent population: A randomized controlled trial. BMC Public Health, 16(1), 1.Find this resource:

Hobson, K., & Niemeyer, S. (2012). What skeptics believe: The effect of information and deliberation on climate change skepticism. Public Understanding of Science, 22, 396–412.Find this resource:

Horton, K., & Doran, M. (2011). Climate change and sustainable consumption: What do the public think is fair? Retrieved from http://www.jrf.org.uk/.

Howell, A. J., Ulan, J. A., & Powell, R. A. (2014). Essentialist beliefs, stigmatizing attitudes, and low empathy predict greater endorsement of noun labels applied to people with mental disorders. Personality and Individual Differences, 66, 33–38.Find this resource:

Hyland, J. J., Jones, D. L., Parkhill, K. A., Barnes, A. P., & Williams, A. P. (2016). Farmers’ perceptions of climate change: Identifying types. Agriculture and Human Values, 33(2), 323–339.Find this resource:

Ishikawa, Y., Hirai, K., Saito, H., Fukuyoshi, J., Yonekura, A., Harada, K., et al. (2012). Cost-effectiveness of a tailored intervention designed to increase breast cancer screening among a non-adherent population: A randomized controlled trial. BMC Public Health, 12(1), 1.Find this resource:

Jung, C. (2017). Psychological types. London: Routledge.Find this resource:

Kahan, D. M., Braman, D., Gastil, J., Slovic, P., & Mertz, C. K. (2007). Culture and identity-protective cognition: Explaining the white-male effect in risk perception. Journal of Empirical Legal Studies, 4(3), 465–505.Find this resource:

Kahan, D. M., Jenkins-Smith, H., & Braman, D. (2011). Cultural cognition of scientific consensus. Journal of Risk Research, 14, 147–174.Find this resource:

Kahan, D. M., Wittlin, M., Peters, E., Slovic, P., Ouellette, L. L., Braman, D., & Mandel, G. N. (2011). The tragedy of the risk-perception commons: Culture conflict, rationality conflict, and climate change. Temple University Legal Studies Research Paper No. 2011–26; Cultural Cognition Project Working Paper No. 89; Yale Law & Economics Research Paper No. 435; Yale Law School, Public Law Working Paper No. 230. Available at SSRN: http://ssrn.com/abstract=1871503 or http://dx.doi.org/10.2139/ssrn.1871503.Find this resource:

Kelly, L. D., Luebke, J. F., Clayton, S., Saunders, C. D., Matiasek, J., & Grajal, A. (2014). Climate change attitudes of zoo and aquarium visitors: Implications for climate literacy education. Journal of Geoscience Education, 62(3), 502–510.Find this resource:

Korkala, E. A. E., Hugg, T. T., & Jaakkola, J. J. K. (2014). Voluntary climate change mitigation actions of young adults: A classification of mitigators through latent class analysis. PLoS ONE, 9(7).Find this resource:

Kotler, P., & Zaltman, G. (1971). Social marketing: An approach to planned social change. Journal of Marketing, 35(3), 3–12.Find this resource:

Krebs, P., Prochaska, J. O., & Rossi, J. S. (2010). A meta-analysis of computer-tailored interventions for health behavior change. Preventive Medicine, 51(3), 214–221.Find this resource:

Kreiss, D., & Jasinski, C. (2016). The tech industry meets presidential politics: Explaining the Democratic party’s technological advantage in electoral campaigning, 2004–2012. Political Communication, Online, 1–19.Find this resource:

Lefebvre, C. R., & Flora, J. A. (1988). Social marketing and public health intervention. Health Education & Behavior, 15(3), 299–315.Find this resource:

Leiserowitz, A. (2006). Climate change risk perception and policy preferences: The role of affect, imagery, and values. Climatic Change, 77(1–2), 45–72.Find this resource:

Leiserowitz, A., Maibach, E., & Roser-Renouf, C. (2009). Global warming’s six Americas 2009: An audience segmentation analysis. Yale University and George Mason University. Retrieved from http://climatecommunication.yale.edu/wp-content/uploads/2016/02/2009_05_Global-Warmings-Six-Americas.pdf.

Leiserowitz, A., Maibach, E., & Roser-Renouf, C. (2010, January 10). Global warming’s six Americas. Yale University and George Mason University. New Haven, CT: Yale Project on Climate Change. Retrieved from http://environment.yale.edu/climate/files/SixAmericasJan2010.pdf.Find this resource:

Leiserowitz, A., Maibach, E., Roser-Renouf, C., Feinberg, G., & Howe, P. (2013a). Climate change in the American mind: Americans’ global warming beliefs and attitudes in April 2013. Yale University and George Mason University. New Haven, CT: Yale Project on Climate Change. Retrieved from http://climatecommunication.yale.edu/wp-content/uploads/2016/02/2013_05_Americans%E2%80%99-Global-Warming-Beliefs-and-Attitudes-in-April-2013.pdf.Find this resource:

Leiserowitz, A., Maibach, E., Roser-Renouf, C., Feinberg, G., & Howe, P. (2013b). Global warming’s Six Americas in September 2012. Yale University and George Mason University. New Haven, CT: Yale Project on Climate Change.Find this resource:

Leiserowitz, A., Maibach, E., Roser-Renouf, C., Feinberg, G., & Rosenthal, S. (2015). Climate change in the American Christian mind: March, 2015. Yale University and George Mason University. New Haven, CT: Yale Project on Climate Change. Retrieved from http://climatecommunication.yale.edu/wp-content/uploads/2015/04/Global-Warming-Religion-March-2015.pdf.Find this resource:

Leiserowitz, A., Maibach, E., Roser-Renouf, C., Feinberg, G., Rosenthal, S., & Marlon, J. (2014). Climate change in the American mind: Americans’ global warming beliefs and attitudes in November, 2013. Yale University and George Mason University. New Haven, CT: Yale Project on Climate Change. Retrieved from http://environment.yale.edu/climate-communication-OFF/files/Climate-Beliefs-November-2013.pdf.Find this resource:

Leiserowitz, A., Maibach, E., Roser-Renouf, C., & Hmielowski, J. (2012). Global warming’s six Americas, March 2012 and November 2011. Yale University and George Mason University. New Haven, CT: Yale Project on Climate Change. Retrieved from http://environment.yale.edu/climate/files/Six-Americas-March-2012.pdf.Find this resource:

Leiserowitz, A., Maibach, E., Roser-Renouf, C., & Smith, N. (2011). Global warming’s six Americas, May 2011. Yale University and George Mason University. New Haven, CT: Yale Project on Climate Change. Retrieved from http://environment.yale.edu/climate/files/SixAmericasMay2011.pdf.Find this resource:

Leiserowitz, A., Thaker, J., Feinberg, G., & Cooper, D. (2013). Global warming’s six Indias. New Haven, CT: Yale University Press.Find this resource:

Ling, J. C., Franklin, B. A. K., Lindsteadt, J. F., & Gearon, S. A. N. (1992). Social marketing: Its place in public health. Annual Review of Public Health, 13, 239–255.Find this resource:

Lorenzoni, I., & Hulme, M. (2009). Believing is seeing: Laypeople’s views of future socio-economic and climate change in England and in Italy. Public Understanding of Science, 18, 383–400.Find this resource:

Maass, A., Suitner, C., & Merkel, E. (2014). Does political correctness make (social) sense? In J. P. Forgas, O. Vincze, & J. Laszlo (Eds.), Social cognition and communication (pp. 331–346). New York: Psychology Press.Find this resource:

Maibach, E. W., Leiserowitz, A., Roser-Renouf, C., & Mertz, C. K. (2011). Identifying like-minded audiences for global warming public engagement campaigns: An audience segmentation analysis and tool development. PLos ONE, 6(3), e17571.Find this resource:

Manum, M. A. A., Stoll, N., & Whitehead, S. (2013). How the people of Bangladesk live with climate change and what communication can do. London: BBC Media Action Lab.Find this resource:

Mathijssen, J. J. P., Janssen, M. M., van Bon-Martens, M. J. H., & van de Goor, I. A. M. (2012). Adolescents and alcohol: An explorative audience segmentation analysis. BMC Public Health, 12(742), Provisional pdf.Find this resource:

Mead, E., Roser-Renouf, C., Rimal, R. N., Flora, J. A., Maibach, E. W., & Leiserowitz, A. (2012). Information seeking about global climate change among adolescents: The role of risk perceptions, efficacy beliefs, and parental influences. Atlantic Journal of Communication, 20(1), 31–52.Find this resource:

Meehl, P. E. (1992). Factors and taxa, traits and types, differences of degree and differences in kind. Journal of Personality, 60(1), 117–174.Find this resource:

Meehl, P. E. (1995). Bootstraps taxometrics: Solving the classification problem in psychopathology. American Psychologist, 50(4), 266.Find this resource:

Metag, J., Füchslin, T., & Schäfer, M. S. (2015). Global warming’s five Germanys: A typology of Germans’ views on climate change and patterns of media use and information. Public Understanding of Science.Find this resource:

Michie, S., West, R., Campbell, R., Brown, J., & Gainforth, H. (2014). ABC of behaviour change theories. London: Silverback Publishing.Find this resource:

Milfont, T. L., Milojev, P., Greaves, L. M., & Sibley, C. G. (2015). Socio-structural and psychological foundations of climate change beliefs. New Zealand Journal of Psychology, 44(1), 17–30.Find this resource:

Mitchell, A. (1983). The nine American lifestyles: Who we are and where we are going. New York: Warner.Find this resource:

Morgan, M. I., Hine, D. W., Bhullar, N., & Loi, N. M. (2015). Landholder adoption of low emission agricultural practices: A profiling approach. Journal of Environmental Psychology, 41, 35–44.Find this resource:

Morrison, M., Duncan, R., & Parton, K. (2013). Targeting segments in the Australian community to increase support for climate change policy. Australasian Marketing Journal (AMJ), 21(4), 212–217.Find this resource:

Morrison, M., Duncan, R., & Parton, K. (2015). Religion does matter for climate change attitudes and behavior. PLoS ONE, 10(8), e0134868.Find this resource:

Morrison, M., Duncan, R., Sherley, C., & Parton, K. (2013). A comparison of the attitudes toward climate change in Australian and the United States. Australasian Journal of Environmental Management, 20, 87–100.Find this resource:

Muriuki, G., Dowd, A.-M., & Ashworth, P. (2016). Urban sustainability—a segmentation study of Greater Brisbane, Australia. Journal of Environmental Planning and Management, 59(3), 414–435.Find this resource:

Myers, T. A., Maibach, E. W., Roser-Renouf, C., Akerlof, K., & Leiserowitz, A. A. (2013). The relationship between personal experience and belief in the reality of global warming. Nature Climate Change, 3(4), 343–347.Find this resource:

Myers, T. A., Nisbet, M. C., Maibach, E. W., & Leiserowitz, A. A. (2012). A public health frame arouses hopeful emotions about climate change: A letter. Climatic Change, 113(3–4), 1105–1112.Find this resource:

Noar, S. M., Benac, C. N., & Harris, M. S. (2007). Does tailoring matter: Meta-analytic review of print health behavior change interventions. Psychological Bulletin, 133(4), 673–693.Find this resource:

Pellegrin, P., & Preus, A. (1986). Aristotle’s classification of animals: Biology and the conceptual unity of the Aristotelian corpus. Berkeley, CA: University of California Press.Find this resource:

Pettinger, D. (2005). Cautionary comments regarding the Myers-Briggs Type Indicator. Consulting Psychology Journal, 5(3), 210–221.Find this resource:

Poortinga, W., & Darnton, A. (2016). Segmenting for sustainability: The development of a sustainability segmentation model from a Welsh sample. Journal of Environmental Psychology, 45, 221–232.Find this resource:

Prochaska, J. O., DiClemente, C. C., & Norcross, J. C. (1992). In search of how people change: Applications to addictive behaviors. American Psychologist, 47(9), 1102–1114.Find this resource:

Rimal, R. N., Brown, J., Mkandawire, G., Folda, L., Böse, K., & Creel, A. H. (2009). Audience segmentation as a social-marketing tool in health promotion: Use of the risk perception attitude framework in HIV prevention in Malawi. American Journal of Public Health, 99(12), 2224–2229.Find this resource:

Rimal, R. N., & Real, K. (2003). Perceived risk and efficacy beliefs as motivators of change: Use of the risk perception attitude (RPA) framework to understand health behaviors. Human Communication Research, 29, 370–399.Find this resource:

Roser-Renouf, C., Maibach, E., Leiserowitz, A., Feinberg, G., Rosenthal, S., & Kreslake, J. (2014, October). Global warming’s six Americas: Perception of the health consequences of global warming and update on key beliefs. Yale University and George Mason University. New Haven, CT: Yale Project on Climate Change. Retrieved from http://environment.yale.edu/climate-communication-OFF/files/Six-Americas-October-2014.pdf.Find this resource:

Roser-Renouf, C., Maibach, E., Leiserowitz, A., & Rosenthal, S. (2016). Global warming’s six Americas and the election. Yale University and George Mason University. New Haven, CT: Yale Project on Climate Change. Retrieved from http://climatecommunication.yale.edu/publications/six-americas-2016-election/.Find this resource:

Sherley, C., Morrison, M., Duncan, R., & Parton, K. (2014). Using segmentation and prototyping in engaging politically-salient climate-change household segments. Journal of Nonprofit & Public Sector Marketing, 26(3), 258–280.Find this resource:

Sibley, C. G., & Kurz, T. (2013). A model of climate belief profiles: How much does it matter if people question human causation? Analyses of Social Issues and Public Policy, 13(1), 245–261.Find this resource:

Slater, M. D. (1996). Theory and method in health audience segmentation. Journal of Health Communication: International Perspectives, 1(3), 267–284.Find this resource:

Smith, W. R. (1956). Product differention and market segmentation as alternative marketing strategies. Journal of Marketing, 21(1), 3–8.Find this resource:

Strategic Business Insights. (1978). VALS. Retrieved from http://www.strategicbusinessinsights.com/vals/about.shtml.

Sütterlin, B., Brunner, T. A., & Siegrist, M. (2011). Who puts the most energy into energy conservation? A segmentation of energy consumers based on energy-related behavioral characteristics. Energy Policy, 39(12), 8137–8152.Find this resource:

Tabi, A. (2013). Does pro-environmental behaviour affect carbon emissions? Energy Policy, 63, 972–981.Find this resource:

Tabi, A., Hille, S. L., & Wüstenhagen, R. (2014). What makes people seal the green power deal?—Customer segmentation based on choice experiment in Germany. Ecological Economics, 107, 206–215.Find this resource:

Thogerson, J., & Crompton, T. (2009). Simple and painless? The limitations of spillover in environmental campaigning. Journal of Consumer Policy, 32, 141–146.Find this resource:

Thornton, A., Evans, L., Bunt, K., Simon, A., King, S., & Webster, T. (2011). Climate change and transport choices: Segmentation model—a framework for reducing CO2 emissions from personal travel (Report No. PPRO 04/06/21). Retrieved from https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/49971/climate-change-transport-choices-full.pdf.

Turow, J., Delli Carpini, M. X., Draper, N., & Howard-Williams, R. (2012). Americans roundly reject tailored political advertising at a time when political campaigns are embracing it. Philadelphia: University of Pennsylvania, Annenberg School for Communication.Find this resource:

Waitt, G., Caputi, P., Gibson, C., Farbotko, C., Head, L., Gill, N., & Stanes, E. (2012). Sustainable household capability: Which households are doing the work of environmental sustainability? Australian Geographer, 43(1), 51–74.Find this resource:

Wilson, R. S., Howard, G., & Burnett, E. A. (2014). Improving nutrient management practices in agriculture: The role of risk-based beliefs in understanding farmers’ attitudes toward taking additional action. Water Resources Research, 50(8), 6735–6746.Find this resource:

Yankelovich, D. (1964). New criteria for market segmentation. Harvard Business Review, 42, 83–90.Find this resource:

Yue, C., Campbell, B., Hall, C., Behe, B., Dennis, J., & Khachatryan, H. (2016). Consumer preference for sustainable attributes in plants: Evidence from experimental auctions. Agribusiness, 32(2), 222–235.Find this resource:

Zaheer, K., & Colom, A. (2013). How the people of Pakistan live with climate change and what communication can do. London: BBC Media Action Lab.Find this resource: