Using Image Elicitation and Metaphor Analysis to Reveal Cross-cultural Differences in Perceptions of Climate Change: Implication for Environmental Communication

George Anghelcev (Penn State University, USA) & Sela Sar (University of Illinois, USA)

Summary of the Project


First and foremost, the research makes a methodological argument. It illustrates that the ZMET protocol (Zaltman, 1997; Zaltman and Coulter, 1995) has an exceptional ability to complement survey research – the default option in examining public perceptions of environmental issues – by overcoming and compensating the inherent methodological limitations of the latter. We argue that ZMET-based protocols should be seen a complementary effort in the developmental stages of pro-environmental campaign planning. Indeed, the method helps guide the cultural tayloring of campaign messages. Messages which are designed to resonate with the cultural beliefs of the targeted group have been shown to be more credible, likeable and persuasive.

In addition, the findings revealed and array of themes that can be used by practitioners to develop social marketing messages congruent with the climate change views shared by each of the two cultures.

Finally, above and beyond its specific focus on cultural differences in perceptions of climate change, the study is illustrates the benefits of employing this image-based elictiation technique when exploring cross-cultural differences in perceptions of other socially relevant issues on the public policy agenda.

Research Question(s)

The goal was to uncover commonalities and differences in perceptions of climate change endorsed by youth in the US and South Korea.

Theoretical Framework

Research on climate change attitudes indicates the possibility that mental representations of climate change are partially anchored in psychological universals which transcend culture and cultural values (Norenzayan and Heine, 2005) as they are shaped by a global media discourse (Stamm et al., 2000). Yet, it appears that cultural factors should also significantly influence perceptions of climate change (e.g., Anghelcev, Chung, Sar and Duff, 2015). Consequently, communication strategies aimed at promoting support for climate change policies should rely on research aimed at understanding how culturally-connoted meanings might influence public response to, and understanding of, climate change mitigation campaigns (Orr, 2003). Such insights are most likely attained if a range of research methods are employed. As we mentioned elsewhere (Anghelcev et al, 2015), research on public perceptions of climate change has consisted primarily of quantitative surveys (Henry, 2000). Over-reliance on quantitative approaches is problematic in strategic communication because checklist surveys: […] do not adequately measure [the public’s understanding of the topic] but rather reflect, at best, respondents’ recognitions of abstract terms and, at worst, their ability to guess and select […] responses provided on the survey questionnaire (Whitmarsh, 2009, p. 415; see also Peattie et al., 2009). In addition, Whitmarsh (2009) observed that close-ended surveys are inherently reactive, in that respondents can only express their opinions in response to statements generated by experts. The discovery of cultural biases in most research on public perceptions of climate change is thus severely limited to observing support or lack of support to the viewpoints of experts, which are much less likely to demonstrate cultural variability than laypeople’s perceptions. Several other previously noted limitations of quantitative survey research point to the need for a more exploratory, qualitative approach. They include the inability of survey to tap into experiential, emotional and interpretive knowledge, as well as the restrictions imposed by using language as a reflection of mental perceptions (for a more detailed discussion, see Anghelcev et al., 2015).


The paper makes a methodological argument for the use of a six-step image elicitation protocol, in combination with metaphor analysis, to identify cultural differences and commonalities in perceptions of climate change among youth in one Western and one East Asian country, based on the success we have had in the past with this method in discovering the strong cultural connotations which imbue public perceptions of climate change in South Korea (see Anghelcev et al, 2015). The six steps of the ZALTMAN technique we employed are outlined below (based on Zaltman, 1997; Zaltman and Coulter, 1995):

Step 1 – Storytelling: informants explained how each image reflected their understanding of climate change

Step 2 – Missing image: if participants wanted to include images they could not find, a sketch of the desired image was drawn and included for the rest of the interview

Step 3 – Construct elicitation: comparisons among images were used to elicit further meaning

Step 4 – Most important image: participants identified the three most important images and explained why they were considered the most important

Step 5 – Sensory images: sensory feelings associated with climate change were elicited

Step 6 – The collage (summary image): participants arranged images in a collage and explained how images were related to one another


The study builds on previous research conducted in South Korea only, which revealed “a multifaceted mental model of climate change, whereby factual, interpretive and emotional knowledge is organized around themes of loss, human greed, affective distress, and iconic representations of tragic endings. The causal dynamics of climate change are construed along a continuum of psychological distance, with antecedents placed in proximity and effects assigned to distant temporal, geographical and psychological spaces” (Anghelcev et al, 2015). These perceptions are reflective of the Korean Sample; US respondents’ mindset was only partially represented by these themes, suggesting a strong presence of culturally-shaped differences. A detailed comparison of the two mindsets will be presented at the conference.


Anghelcev, G., Chung, M. Y., Sar, S., & Duff, B. R. (2015). A ZMET-based analysis of perceptions of climate change among young South Koreans: implications for social marketing communication. Journal of Social Marketing,5(1).

Christensen, G.L. and Olson, J.C. (2002), “Mapping consumers’ mental models with ZMET”, Psychology and Marketing, Vol. 19 No. 6, pp. 477-501.

Henry, A.D. (2000), “Public perceptions of global warming”, Human Ecology Review, Vol. 7 No. 1, pp. 25-30.

Norenzayan, A., & Heine, S. J. (2005). Psychological universals: What are they and how can we know?. Psychological bulletin131(5), 763.

Orr, M. (2003), “Environmental decline and the rise of religion”, Zygon, Vol. 38 No. 4, pp. 895-910.

Peattie, K., Peattie, S. and Ponting, C. (2009), “Climate change: a social and commercial marketing communications challenge”, EuroMed Journal of Business, Vol. 4 No. 3, pp. 270-286.

Zaltman, G. (1997), “Rethinking market research: putting people back in”, Journal of Marketing Research, Vol. 34 No. 4, pp. 424-437.

Zaltman, G. and Coulter, R.H. (1995), “Seeing the voice of the customer: metaphor-based advertising research”, Journal of Advertising Research, Vol. 35 No. 4, pp. 35-51.

Methodological Reflections

Methodological Potentials

To the best of our knowledge, this is the first study employing the ZMET technique in the cross-cultural communication literature. The method appears to have been used only once (see Anghelcev et. al, 2015) to investigate pro-environmental topics, but that study was confined to only one cultural space.

From a conceptual standpoint, this method has several major advantages over verbocentric approaches (for a detailed discussion, see Anghelcev et al, 2015). First, drawing from a long tradition of image-based thought elicitation research in psychotherapy, anthropology and cultural sociology, it recognizes the superiority of images over language to encompass and project complex, emotionally-laden, culturally-connoted structures of mental knowledge. Images, rather than words are used to elicit the foundational elements that define one’s understanding of the topic. We were thus able to use images to identify both commonalities and differences in perceptions that have a direct association to culture. The method also allows for the discovery of non-factual knowledge, unlike most verbocentric research. Non-factual knowledge such as emotions, goals, feelings, behavioral scripts, symbols, imagery and sensory experiences are likely important determinants of climate change attitudes and behaviors (Christensen and Olson, 2002; see Anghelcev et al, 2015). The method also allows researchers to tap into layers of knowledge beyond the realm of the subjects‘ immediate awareness. We were able to reveal emotional thoughts and interpretive meanings subjects admittedly were unaware they endorsed prior to engaging in the introspective thought process triggered by the image-guided discussions.

Consequently, we found that the hybrid ZMET protocol is able to capture and reflect the complexity of consumers’ culturally-laden mental representations of climate change better than exclusively verbocentric approaches (surveys, focus groups, interviews).

Methodological Challenges

Translation and back-translation of interview scripts were one of the main challenges. Second, the method requires trained researchers with experience in conducting ZMET interviews, so much practice is needed if graduate students are employed. Truly bicultural interviewers should conduct the research – they must have internalized both cultures and speak both languages fluently. Multiple researchers should assist in the identification of themes and underlying core metaphors to limit subjective biases of certain individual researchers. Overall, financially cheap but timely and demanding of highly qualified personnel.

Methodological Limitations

ZMET is an exploratory technique, based on convenience samples and relying on subjective interpretation of findings. Thus, no claim regarding the generalizability of the findings can be made unless further steps are taken to establish the degree to which the themes uncovered via ZMET are endorsed by sociologically representative samples of respondents from the two cultures.

Kommentare sind geschlossen