The Devil Made Me Do It 

Influence of Values on Interpretation and Behaviors for Tasmanian Devils 


Jeffrey C. Skibins 
1403 Belk Building
Mail Stop 540
East Carolina University
Greenville, NC 27858-4353
252-737-1374 office
skibinsj18@ecu.edu 
ORC ID: 0000-0002-8870-3741

Betty Weiler
Southern Cross University

Kevin Markwell
Southern Cross University

Lian Wilson
Zoos Victoria

Abstract
Zoos are becoming more intentional about embedding messaging in their interpretation to promote pro-conservation behaviors, essentially acting as agents of social change. Values theory suggests that, be effective, interpretation including these pro-conservation messages needs to broadly align with visitors’ values. Using the Schwartz value system, this study modeled the relationships between visitors’ values, perceptions of interpretation, emotional connectivity to Tasmanian Devils, and behavioral intent, including a comparison of different types of on-site and post-visit pro-conservation behaviors. Most visitors held moderate to strong conservation values, which were predictive of positive perceptions of interpretation, emotional connectivity, and pro-conservation behaviors. However, the results suggest that while visitors’ values align with their perceptions of interpretation, they are only weak predictors of behavioral intent. Visitors’ perceptions of interpretation and behavioral intent aligned more strongly with emotional connectivity than with their values. Overall, behavioral intentions were low for all pro-conservation behaviors. Liking a Facebook post about the conservation of Tasmanian Devils was the behavior most likely to be performed.

Keywords
behaviors, emotional connection, interpretation, Tasmanian Devils, values, zoo-based conservation, Zoos Victoria

Introduction
Zoos, given their mandate of keeping and displaying wildlife, offer an ideal portal to reach broad cross-sections of the global public with conservation-oriented interpretation that harnesses the zoo visit to gain support for and action on conservation issues. As zoos continue to evolve into centers of conservation, they are well positioned to act as agents of social change (Fraser & Wharton, 2007; Rabb & Saunders, 2005). This is further enhanced by the staggering visitor numbers now reported by wildlife attractions. As per the World Association of Zoos and Aquariums and the Association of Zoos and Aquariums, annual worldwide visitation rates to accredited institutions exceeds 700 million visitors, with an average visit lasting four hours (AZA, 2019; WAZA, 2019).

Many zoos are actively looking to be more impactful in their use of interpretation, and this includes a better understanding of their publics’ values and how these values relate to behavioral intentions and behaviors. The problem, or gap, that this study seeks to fill is that of understanding the relationship between visitors’ pre-existing values, which are relatively stable, and their perceptions, emotional connections, and pro-conservation behaviors, which are what zoo interpretation typically seeks to influence. Such understanding can then be channeled into developing interpretation that better aligns with the values held by intended audiences.

Interpretation and Behavior Change
A fundamental tool zoos have in their arsenal for garnering conservation support and action is on-site interpretation. Beck and Cable (2011) define interpretation as an educational activity that reveals meanings about natural resources through various media. Another leading definition, from the National Association for Interpretation, defines interpretation as “a mission-based communication process that forges emotional and intellectual connections between the interests of the audience and the meanings inherent in the resource” (NAI, 2019). In both cases, zoo-based interpretation is centered on connecting visitors (emotionally and cognitively) to animals. Ultimately, many zoos seek to use the connection visitors form with wildlife to drive conservation. This model is summarized in the quote apocryphally attributed to Freeman Tilden (1957) “through interpretation, understanding; through understanding, appreciation; through appreciation, protection” (p. 38).

The premise that interpretation can influence an individual’s connection to a resource (e.g. wildlife) and thus stimulate protective action is a core tenet of conservation psychology (Ham, 2009; Saunders, 2003), and the foundation of many behavior models. Several behavior models have been used to explain and predict a wide range of consumer, social, and environmental behaviors, and their efficacy assessed from a conservation psychology perspective (Kaiser, Hübner, & Bogner, 2005). The theory of planned behavior (Ajzen, 1991), value belief norm theory (Stern, Dietz, Abel, Guagnano, & Kalof, 1999), and the elaboration likelihood model (Petty & Cacioppo, 1986) are commonly used theoretical frameworks to align interpretation to desired outcomes (e.g. behavior change), for example, see MacDonald, Milfont, & Gavin (2016), Powell & Ham (2008), and Walker & Moscardo (2014). Such behavior models provide a useful framework to study the applied nature of the zoo visit and in particular the influence of interpretation on visitors’ conservation attitudes and behaviors, including such behaviors as protecting old growth trees in order to preserve nesting sites, choosing products made using sustainable forestry practices, and donating money to a conservation organization.

Most of these frameworks posit that values are antecedent predictors of behavior change (e.g., cessation or adoption of an action). However, most scholars suggest that values are relatively stable and not subject to significant change or influence, particularly as a result of one-off exposure to interpretation (Ajzen, 1991; Manfredo et al., 2017; Stern et al., 1999). Rather than attempting to change values, Manfredo et al. (2016) suggest that conservation would be better served by aligning messages to visitor values. Aligning interpretation to visitors’ underlying values has been shown to increase receptivity to messages and positively influence behavior change (Ham & Weiler, 2002; Orams, 1997).

For example, Clayton, Fraser, and Burgess (2011) found zoo visitors’ environmental identity and connection to an animal were positively correlated with an interest in conservation. Other studies have documented increases in visitor awareness, empathy, and knowledge following zoo-based interpretation (Fraser, Gruber, & Condon, 2007; Gwynne, 2007; Woods, 2002). Other research has shown mixed results as to the efficacy of interpretive experiences influencing behaviors, and the overall likelihood of visitors adopting pro-conservation behaviors (Smith, Broad, & Weiler, 2008; Smith, Curtis, & van Dijk, 2010; Smith, Weiler, & Ham, 2011). Conservation caring, a measure of visitors’ emotional connection to wildlife, has been found to be a stronger predictor of behaviors compared to interpretation. However, interpretation has a significant influence on conservation caring (Skibins, Powell, & Hallo, 2013; Skibins, Dunstan, & Pahlow, 2017).

Another challenge that has not been taken up in published literature is how to address conservation issues that require human intervention, but do not have human behaviors at the root cause of the threat. Tasmanian devil conservation is one such example. The Tasmanian devil is the world’s largest living carnivorous marsupial, is endemic to the state of Tasmania, Australia, and is officially listed as endangered. However, the greatest threat to Tasmanian devil survival, Devil Facial Tumor Disease (DFTD), is not directly attributable to human action; rather, it is a contagious disease transmitted by contact with individual animals already infected with the disease. DFTD has nearly destroyed the entire wild population.

What makes this situation rare, from a conservation standpoint, is that, although DFTD is not rooted in anthropogenic causes, the survival of the species still requires broad public support and action. In particular, financial support from private sources are needed to supplement government funding for breeding disease-free populations and re-introducing them to the wild, and for researching and monitoring their health and survival as a species. Public support, including financial donations, is key, and zoo interpretation can play an important role in fostering pro-conservation attitudes and behaviors toward the species. The influence of visitors’ values on perceptions of interpretive messaging, connection with wildlife, and preferences for pro-conservation behaviors is generally poorly understood, and how values relate to interpretation of a species threatened by non-anthropogenic causes is unknown. Given the increasing evidence that wildlife conservation and threats to biodiversity more generally are largely dependent on human responses to these issues, it is imperative to continue to build understanding of the antecedents to pro-conservation behavior, including the role that both values and interpretation can play in enhancing this (St. John, Edwards-Jones, & Jones 2011).

Values
Given that zoos are significant visitor attractions with broad appeal across nationalities and ethnic groups, this study made use of Schwartz’s (2006, 2012) values instrument, which has been shown to be robust across cultures. Schwartz identified 10 personal value orientations (self-direction, stimulation, hedonism, achievement, power, security, conformity, tradition, benevolence, and universalism), with six features being common to all 10 value orientations: 1) values are beliefs, 2) values are desirable goals, 3) values transcend specific actions and situations, 4) values are standards or criteria, 5) values are ordered by importance, and 6) values are relatively important in guiding actions.

Schwartz’s value orientations provide a way of understanding how people see themselves in relation to the environment, and may be useful in determining how an organization like a public zoo can use interpretation to align with and perhaps reinforce values, which can then in turn help foster a more sustainable world (Pahlow & Dunstan, 2015). Values are measured with multiple items, with a full profiling of an individual’s values requiring at least 56 items. In order to minimize respondent burden, many studies focus on a subset of relevant values. Ballantyne, Hughes, Lee, Packer, and Sneddon (2018) used a subset of Schwartz’s values to evaluate zoo visitors’ behaviors. Manfredo, Teel, and Dietsch (2016) suggest aligning messaging to specific value orientations to improve broad conservation efforts.

This study focused on two of Schwartz’s value orientations, benevolence and universalism, as these have been shown to be of particular relevance to visitor-based conservation actions (Clayton, Litchfield, & Geller, 2013; Walker & Moscardo, 2014; Yocco, Bruskotter, Wilson, & Heimlich, 2015). The benevolence value orientation focuses on the welfare of those with whom one is in frequent contact. It emphasizes voluntary concern for others’ welfare and promotes cooperative and supportive behaviors. The concern expressed for others can be extended to include wildlife (Manfredo, Teel, & Dietsch, 2016). The universalism value orientation has its defining goal as understanding, appreciation, tolerance, and protection for the welfare of all people and for nature. Universalism values are triggered by the recognition of the scarcity of natural resources. Universalism’s generic focus is contrasted with the personalized focus of benevolence.

As noted in Schwartz’s value features five and six, individual importance and impact of values is relative. In order to identify the strength of value orientations, Schwartz created the Schwartz Value Survey (SVS). The SVS uses nine-point scale items that are compressed at the bottom end and allow for expression of opposition to the value. The SVS is designed to capture how people explicitly report their own values. It has been shown to be valid across diverse samples. However, as feature six indicates, value importance is relative and the subsequent differential effect on visitor outcomes is unknown.

Study Objectives
This study had two main objectives: 1) quantify the influence of zoo visitors’ values on perceptions of interpretive messaging, connection to wildlife (conservation caring), and pro-conservation behavioral intentions, and 2) evaluate visitors’ preferences for pro-conservation behaviors for the Tasmanian Devil. Positive perceptions of interpretive messaging, conservation caring, and pro-conservation behavioral intentions are common visitor-based outcomes that zoos aim to achieve through interpretation and are expected to contribute to wildlife conservation (Skibins & Powell, 2013; Smith et al., 2008; Weiler & Smith, 2009). Statistically significant relationships between a visitor’s values and these outcome variables would have implications for message development and delivery.

Study Site
Zoos Victoria (ZV) is a zoo-based conservation organization with three properties—Melbourne Zoo, Healesville Sanctuary, and Werribee Open Range Zoo—all located in Victoria, Australia. This study was conducted at Healesville Sanctuary (average annual visitation 500,000), which is designed to provide a uniquely Australian experience. Healesville Sanctuary departs from the traditional zoo practice in that its collection is not dominated by exotic charismatic megafauna, large-bodied vertebrates, and less threatened species (Colleony, Clayton, Couvet, Saint Jalme, & Prevot, 2017). Rather, the zoo exhibits mainly native Australian wildlife and its interpretation is strongly focused on local endangered species and recovery programs, one of which is supporting the recovery of Tasmanian Devils.



Healesville Sanctuary holds the second largest captive group of Tasmanian Devils in Australia and has the highest breeding success of any of the 31 captive breeding programs (Zoos Victoria, 2014). The Tasmanian Devil exhibit consists of several enclosures along a circular path with one entry/exit point. Keeper Talks (i.e., first-person interpretation) are conducted at the first enclosure on the path. There are 10 full-color interpretive panels throughout the exhibit that detail Tasmanian devil natural history, conservation issues, and DFTD. Signage is almost entirely static (i.e., does not include video or sound) with multi-sensory interpretation largely absent except during the Keeper Talks, although several of the signs have interactive flaps and actions. Collectively, the messaging is informative and compelling, and strongly focused on conveying the primary threat (DFTD), building emotional connections with the devil, and providing visitors with actions they can undertake to help address the threat and thereby help “save” the devil. For example, “DFTD is a contagious fatal cancer that has wiped out most of Tasmania’s wild devils and may leave them extinct in the wild”; “Breeding and research is helping us survive”; “One day we’ll be back in the wild”; and “You can help fight extinction by making a donation.” Messaging applies interpretive principles and uses interpretive techniques that are known to be effective in attracting attention and impacting visitors understanding, attitudes, and behavioral intentions (Skibins et al., 2012). This includes content that is relevant to the audience, cognitive and affective messaging, providing examples of actions visitors can take, and providing an opportunity to take action on-site. The latter consists of a donation box near the main viewing platform, in which inserting a coin activates a recording of Tasmanian devil sounds. The donation box panel includes the zoo’s “Fighting Extinction” logo, text stating Zoos Victoria is a not-for-profit organization, and that donations support the zoo’s threatened species conservation programs (see Figure 1).

Methods
A self-completed questionnaire was administered to visitors aged 18 and over at Healesville Sanctuary over a two-month period as they exited the Tasmanian devil exhibit. A census sampling protocol was used. Census sampling is a useful approach when all visitors during the sampling period can be intercepted at a central location (Salkind, 2016).

Measurements for visitors’ values were adapted from the Schwartz Value Survey (SVS). SVS items were rated on a nine-point Likert-type scale: –1 (opposed to my values) to 7 (of extreme importance) and follow Schwartz’s schema for measuring values (Schwartz, 1992, 2006, 2012). The three dependent variables included perceptions of interpretive messaging, conservation caring (emotional connection to Tasmanian Devils), and pro-conservation behavioral intentions. All items were rated on a nine-point Likert-type scale: 1 (strongly disagree/extremely unlikely) to 9 (strongly agree/extremely likely).

Variables

Visitor Values
We measured the SVS value orientations of universalism and benevolence using a total of eight items (four items for each orientation).

Visitor Perceptions of Interpretive Messaging
This composite variable, hereafter perceptions of interpretive messaging (PIM), included items addressing visitors’ self-perceptions of knowledge gained, personal responsibility, and Healesville Sanctuary’s responsibilities for Tasmanian devil survival. Visitors rated each of the 12 items on a nine-point Likert-type (agree-disagree) scale.

Emotional Connectivity
Emotional connectivity to Tasmanian devils was measured using conservation caring. Conservation caring is a scale that measures visitors’ emotional connection to a species. The scale, consisting of six items, has been validated in in situ and ex situ settings for a wide variety of species, and has been shown to be a strong predictor of behaviors (Skibins & Powell, 2013; Skibins et al., 2013; Skibins et al., 2017).

Pro-Conservation Behavioral Intentions (PCBI)
We measured visitors’ self-reported intentions of performing eight different pro-conservation behaviors. These eight behaviors were used to create the composite variable, PCBI. Behavioral intentions can be an acceptable proxy for actual behaviors and can overcome logistical challenges in collecting data at the time of behavior execution (Ballantyne, Packer, & Falk, 2011; Smith & Sutton, 2008). Behaviors were based upon actions that zoo visitors could actually perform at Healesville Sanctuary or at home. Behaviors included five traditional options such as donating various amounts of money and adopting (i.e., sponsoring for an annual fee) an animal (also a form of donation), and three social media-related options that had a direct connection to conservation, such as “liking” a Healesville Sanctuary/Tasmanian devil conservation post on Facebook.

Analyses
Data were screened for missingness and for univariate, and multivariate outliers (Tabachnick & Fidell, 2007; Vaske, 2008). Sixty-two cases were removed, using listwise deletion, for exceeding ± 3 S. D. and the Mahalanobis Distance value (χ 2 (43) = 77.42, p < .001). The final sample size was 419. Composite variables were assessed using scale reliability analysis in SPSS v22. Reliability iterations were stopped when there was no meaningful improvement to Cronbach alpha scores from item removal. Final Cronbach alpha scores for all composites (i.e., visitor values, PIM, conservation caring, and PCBI) exceeded 0.8. Linear regressions were performed to uncover relationships between variables. ANOVAs with Bonferroni post-hoc comparison and pairwise t-tests were performed to assess differences in means.

A K-means cluster analysis was performed using SVS value orientations following Wu (2012). This technique has been shown to be useful for assessing psycho-social constructs in tourism experiences (Brida, Osti, & Barquet, 2010; Kibicho, 2006). K-means cluster analyses allow the researcher to specify the number of groups, and subsequently allows for cases with similar scores to be placed into relatively homogeneous groups or clusters. Assignment to a group is based on the shared similarities for that cluster, as well as the degree of difference from other clusters. The analysis seeks to maximize intra-group similarity and inter-group differences. Fredline (2012) advocates for the use of cluster analysis in tourism settings as a meaningful and effective approach to assess nuanced differences within psycho-social variables, such as values, rather than relying on geographic or demographic segmentation. By creating groups that are similar, solutions can be more targeted and specialized.

As per Wu (2012), the number of clusters is based, in part, on the sample size of potential clusters. It is preferred to have clusters with 20 or more individuals and of the same approximate sample size. However, neither condition is mandatory (Fredline, 2012; Wu, 2012). For this study, we created three clusters, high (n=221), medium (n=199), and low (n=57) values scores, which were subsequently evaluated for differences in responses across the dependent variables. 

Results

Survey Sample and Composite Variable Descriptives
A total of 481 zoo visitors (43% response rate) completed the survey. The sample was 52% female, 48% male; mean age was 40.5 years; 76% reported completing undergraduate or post-graduate level education; 86% resided in Australia; and 54% report an annual household income greater than $75,000 AUD.



Sample sizes, means, standard deviations, and Cronbach alpha values are reported for the composite variables respectively (Table 1): benevolence (410, 5.61 ± 1.32, .94), universalism (403, 5.25 ± 1.40, .93), PIM (365, 7.94 ± .85, .88), conservation caring (401, 5.33 ± 1.70, .92), and PCBI (399, 3.97 ± 1.77, .87). As shown in Table 1, respondents generally held positive benevolence and universalism values and are above average in conservation caring (emotional connection with the devil), but had low intentions to undertake pro-conservation actions.



Influence of Values on Dependent Variables
Linear regressions were performed to uncover relationships between value orientations and the dependent variables. For individual parameter estimates, see Table 2. SVS values were shown to significantly predict PIM (F(2, 344) = 24.44, p < .001, R2 = .12), conservation caring (F(2, 377) = 63.67, p < .001, R2 = .25), and PCBI (F(2, 377) = 16.53, p < .001, R2 = .076). When both SVS value orientations, PIM, and conservation caring were regressed on PCBI, the model was significant (F(4, 322) = 52.32, p < .001, R2 = .39); universalism (β = -.16 , p < .001) and conservation caring (β = .62 , p < .001) were the only significant predictors.







Cluster Analyses
High, medium, and low clusters were generated for benevolence and universalism values (Table 3). Socio-demographic characteristics were consistent across clusters, with the exception of gender in cluster 3 (low values). Dependent variables were assessed across clusters. The overall pattern was for cluster 3 (low value scores) to report less positive PIM levels, conservation caring and PCBI (p < .05) than clusters 1 (high) and 2 (medium). Cluster 1 had the highest scores for PIM and conservation caring. Cluster 3 had the lowest scores for PCBI, but there was no difference between Clusters 1 and 2. For a complete list of results see Table 4. ANOVAs with Bonferroni post-hoc comparisons were performed for each behavior across clusters (see Table 5). Respondents in cluster 3 (low) had significantly lower scores for all behavioral intentions compared to clusters 1 (high) and 2 (medium).



Differences Between Individual PCBI
Pairwise comparisons of individual pro-conservation behavioral intentions were performed at the total sample level. The purpose of this analysis was to uncover potential differences in visitors’ likelihood of performing pro-conservation behaviors for Tasmanian devils. For a complete list of individual comparisons see Table 6. The following patterns were observed. The least likely behavior to be performed was donating $250 to help fund research (3.16 ± 1.93, p < .05 in comparison to all other PCBI). The behavior that had the highest reported level of intention was “liking” a post on Facebook (5.58 ± 2.86, p < .05 in comparison to all other PCBI). Within financial donation behaviors, visitors were most likely to make a one-time donation of $50 to $60. Use of Twitter was the least likely of the social media behaviors (3.60 ± 2.78, comparisons within social media category all p < .05).

Discussion
This study had two main objectives. The first was to determine the relationship between zoo visitors’ values and perceptions of interpretive messaging, connection to wildlife, and behavioral intentions. The second was to evaluate visitors’ preferences for conservation behaviors they could perform to aid the Tasmanian devil.

Influence of Values on Conservation Outcomes
The SVS value orientations demonstrated high levels of statistical rigor and were successful in measuring latent values of zoo visitors. This was expected as these instruments have been validated in numerous contexts (Schwartz, 2006, 2012; Schwartz et al., 2012). Visitors’ mean scores for SVS values (both benevolence and universalism) were just over five, indicating a moderate to high level of importance. Additionally, the majority of respondents (85%) were assigned to the high and medium value clusters (Table 3), which may suggest a trend in zoo visitors of holding moderate to strong benevolence and universalism values. However, further investigations are needed to better define this pattern.

Results from the linear regressions support the assertion that visitors’ values have a positive and direct relationship with PIM, conservation caring, and PCBI. However, universalism was only a weak predictor of the dependent variables, and benevolence was not found to be a significant predictor of any of the dependent variables. Although a negative coefficient for universalism appears in the model with all the IVs, the value is too small to be considered meaningful. Future interpretive messaging could consider focusing more on universalism value statements. A focus on one value set may also help interpretive messaging better adhere to being thematic, organized, relevant, and enjoyable as advocated by Ham (2013).

In general, all the R2 values are low and indicative of only “weak” relationships (i.e., R2 < .40) (Vaske, 2008); no R2 value exceeded .40 (Table 2). In fact, the model with values as the only IV and PCBI as the DV performed the poorest (R2 = .076), suggesting values are not strongly related to behavioral intentions. Values’ effect on PIM was also very weak (R2 = .12). Overall, these findings support previous studies that show values and experiential variables (such as interpretation and observed species) are more strongly related to conservation caring, and only weakly associated with PCBI (Skibins et al., 2016; Skibins et al., 2013).

In general, the findings of this study are consistent with how values are modeled in communication and behavior models (Petty & Cacioppo, 1986; Stern et al., 1999; Stern, Dietz, & Guagnano, 1998) and previous values studies (Sponarski, Vaske, Bath & Musiani, 2014; Walker & Moscardo, 2014). As noted earlier, values are relatively stable over time and function as anchor points in most models. Thus, while values do exert an influence on communication efficacy and behavior change, they are several steps removed. Data from this study suggest interpretive messaging that is generally framed by pro-conservation values, but which targets constructs such as emotional connectivity may be more effective in generating desired visitor outcomes compared to messaging that focuses solely on values, as seen in Jacobs and Harms (2014) and Pfatteheicher, Sassenrath, and Schindler, 2015.

Visitor Outcomes Based on Values Clusters
Clusters were generated using a K-Means cluster analyses, based on visitors’ value scores. The purpose of these analyses was to provide practitioners an additional perspective in considering ways of engaging audiences. With the exception of gender in Cluster 3, values were uniformly distributed in visitors to Healesville Sanctuary and high/medium/low scores were not restricted to a specific subset of visitors. In other words, values were relatively evenly distributed regardless of age, education, and income levels. This would indicate a reduced need for a values-based messaging campaigns targeted to individual demographics. This is consistent with Doran (2009) who found SVS values were not linked to specific demographics.

Once clusters were established, the mean scores for PIM, conservation caring, and PCBI were compared across clusters. All scores for PIM were very high regardless of value cluster, which suggests an overall high-quality interpretive experience at Healesville Sanctuary. Visitors who reported high value scores (Cluster 1) reported the highest PIM. Low (Cluster 3) and medium (Cluster 2) clusters did not differ, suggesting that the values threshold for perceptions of interpretation is relatively high. That is to say, visitors who reported the strongest value orientation scores rated interpretation higher than visitors with medium or low value scores. This may be an example of “preaching to the choir,” in that visitors with high benevolence and universalism scores may be predisposed to the zoo’s mission, and thus favorably regard interpretive messaging (Ballantyne, Packer, & Falk, 2011; Skibins et al., 2013).

For conservation caring, all three clusters (high/medium/low) were significantly different from each other, and all hovered around the neutral point. Visitors who reported high value scores reported the highest conservation caring scores, followed by medium then low clusters. This pattern is expected and supports how values underlie emotional responses, particularly for wildlife species such as the Tasmanian devil that are not necessarily charismatic (Burns, 2006; Schultz, 2001).

No differences were revealed between high and medium clusters for PCBI responses, effectively creating only two groups (low and medium/high). This suggests the values threshold for PCBI is low, whereas for PIM it was high. As medium and high clusters had the greatest number of individuals, this would indicate the majority of zoo visitors are equally likely to engage in pro-conservation behaviors. The minority segment of visitors with low value scores appear to be very unlikely to participate in behaviors. However, it is important to reiterate that values were only weak predictors of behaviors, as were all other variables in this study.

It should also be noted that the mean scores for all PCBIs across value clusters were “unlikely” to “neutral” except for “liking” a Facebook post (see Tables 4 and 5). The mean score for the PCBI composite was also below the neutral point, whereas conservation caring was just above the neutral point of 5. In comparison, PIM was strongly positive. Thus, even though PIM is generally high, and conservation caring is neutral, visitors do not appear to be likely to perform any of the behaviors tested in this study. This may be due to perceptions of the behaviors in this particular study, and/or perceptions of efficacy of conservation behaviors in general (cf. Smith, et al., 2012; Weiler & Smith, 2009; Smith et al., 2010; Smith et al., 2010).

Visitor Preferences for Pro-Conservation Behaviors
This study assessed visitors’ preferences for a set of proposed behaviors intended to contribute to Tasmanian devil conservation. Behaviors included “new” behaviors (social media action) and more traditional behaviors (financial contributions to specific field conservation measures). Overall, visitors’ intention to perform any behavior was low and no single behavior emerged as being highly likely to be adopted. Even simple behaviors, such as sharing information or photos about devils, were found to be unlikely. The behavior that had the highest likelihood of being performed was “liking” a Facebook post. This could be explained by the ease of execution and low level of commitment. It may also speak to the disparity between social media platforms, as behaviors related to Twitter and email were significantly lower.

Although the interpretation is strongly focused on conveying DFTD as the primary threat to Tasmanian devils, the lack of an anthropogenic cause of DFTD may have resulted in visitors expressing low levels of PCBI. It may have been difficult for visitors to relate to the behaviors posed because none precisely targets the cause of the problem or precisely how their action will remove or reduce the threat of DFTD. This situation is different to nearly every other endangered species in that, although human intervention is indeed needed to save the devil, human behaviors are not the underlying cause of the devils’ demise. Thus, visitors may not feel the same level of culpability, or perhaps guilt, as they do in relation to orangutans, mountain gorillas, or Leadbeater’s possums, all for whom Zoos Victoria have successfully leveraged interpretation for conservation outcomes (Banks & Dunstan, 2014; Mellish, Sanders, Litchfield, & Pearson, 2017; Pearson, Lowry, Dorrian, & Litchfield, 2014).

Conclusion
Zoos must continue to provide wildlife viewing experiences that are enjoyable for visitors. However, they are increasingly motivated and even compelled to harness those experiences to impact visitors (Gray, 2017). Moreover, pro-conservation messages are expected to deliver not only better understanding, but greater empathy towards wildlife and a propensity to take action. Visitors in this study report being very favorable regarding their interpretation experience at the zoo, but on average visitors’ levels of emotional connection to the Tasmanian devil is only moderate, and the likelihood of undertaking pro-conservation behaviors is low.

Generally, the findings of this study suggest a low return on investment for interpretive planners who use values to drive message development and delivery. However, replication is needed to ascertain whether this is true for visitor cohorts whose value orientations are less homogeneous, e.g., visitors with differing cultural backgrounds. In addition, findings may be different for species with different threats to their survival and thus options for conservation behaviors that have greater resonance with visitors. Nonetheless, the findings of the present study suggest that there is likely to be greater efficacy in focusing more directly on caring about the species, that is, fostering empathy, rather than attempting to align to values. One such outlet for caring/empathy messaging is social media. Social media is emerging as a new vehicle for zoos to facilitate a deeper and more meaningful connection between visitors, collections, and conservation, as well as providing an avenue for visitor action (Green, Crawford, Williamson, & DeWan, 2019).

Future studies could explore how exposure to different interpretive elements, messages, and media influence visitor responses in general, and behavioral intentions specifically. The present study did not isolate the effects of specific interpretive variables on the dependent variables. While other studies have investigated these relationships (Skibins, et al., 2012), results can be variable and context-specific, suggesting the need for more research generally but particularly for non-anthropogenic driven issues such as the one investigated in the present study.
 
References
Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), 179–211. 

Association of Zoos and Aquariums (AZA). (2019.). Retrieved June 19, 2019 from https://www.aza.org/partnerships-visitor-demographics

Ballantyne, R., Hughes, K., Lee, J., Packer, J., & Sneddon, J. (2018). Visitors’ values and environmental learning outcomes at wildlife attractions: Implications for interpretive practices. Tourism Management, 64, 190–201.

Ballantyne, R., Packer, J., & Falk, J. H. (2011). Visitors’ learning for environmental sustainability: Testing short- and long-term impacts of wildlife tourism experiences using structural equation modelling. Tourism Management, 32(6), 1243–1252.

Banks, C., & Dunstan, E. (2014). Zoos Victoria influencing consumer palm oil purchasing behaviour. Oryx, 48(02), 171–171. 

Beck, L., & Cable, T. T. (2011). The gifts of interpretation: Fifteen guiding principles for interpreting nature and culture. Urbana: Sagamore.

Brida, J. G., Osti, L., & Barquet, A. (2010). Segmenting resident perceptions towards tourism—a cluster analysis with a multinomial logic model of a mountain community. International Journal of Tourism Research, 12(5), 591–602.

Burns, G. L. (2006). The fascination of fur and feathers: managing human-animal interactions in wildlife tourism settings. Australian Zoologist, 33(4), 446–457.

Clayton, S., Fraser, J., & Burgess, C. (2011). The Role of Zoos in Fostering Environmental Identity. Ecopsychology, 3(2), 87–96. 

Clayton, S., Litchfield, C., & Geller, E. S. (2013). Psychological science, conservation, and environmental sustainability. Frontiers in Ecology and the Environment, 11(7), 377–382. 

Colléony, A., Clayton, S., Couvet, D., Saint Jalme, M., & Prévot, A.-C. (2017). Human preferences for species conservation: Animal charisma trumps endangered status. Biological Conservation, 206, 263–269. 

Doran, C. J. (2009). The role of personal values in fair trade consumption. Journal of Business Ethics, 84(4), 549–563.

Fa, J. E., Gusset, M., Flesness, N., & Conde, D. A. (2014). Zoos have yet to unveil their full conservation potential. Animal Conservation, 17(2), 97–100. 

Fraser, J., Gruber, S., & Condon, K. (2007). Exposing the tourist value proposition of zoos and aquaria. Tourism Review International, 11(3), 279–293. 

Fraser, J., & Wharton, D. (2007). The Future of Zoos: A New Model for Cultural Institutions. Curator: The Museum Journal, 50(1), 41–54. 

Fredline, L. (2012). Cluster analysis. In L. Dwyer, A. Gill, & N. Seetaram (Eds.), Handbook of research methods in tourism (pp. 212–226). Cehltenham: Edward Elgar.

Gray, J. (2017). Zoo Ethics: The Challenges of Compassionate Conservation. Ithaca: Comstock Publishing Associates.

Green, K. M., Crawford, B. A., Williamson, K. A., & DeWan, A. A. (2019). A meta-analysis of social marketing campaigns to improve global conservation outcomes. Social Marketing Quarterly, 25(1), 69–87.

Gwynne, J. A. (2007). Inspiration for conservation: Moving audiences to care. In A. Zimmermann, M. Hatchwell, L. Dickie, & C. West (Eds.), Zoos in the 21st century: Catalysts for conservation? (Vol. 15). Cambridge: Cambridge University Press.

Ham, S. H. (2009). From interpretation to protection: Is there a theoretical basis? Journal of Interpretation Research, 14(2). 

Ham, S. H. (2013). Interpretation: Making a difference on purpose. Fulcrum: Golden. 

Ham, S. H., & Weiler, B. (2002). Interpretation as the centerpiece of sustainable wildlife tourism. In R. Harris, T. Griffin, & P. Williams (Eds.), Sustainable tourism: A global perspective (pp. 35–44). Oxford: Elsevier Science Limited.

Jacobs, M. H., & Harms, M. (2014). Influence of interpretation on conservation intentions of whale tourists. Tourism Management, 42, 123–131.

Kaiser, F. G., Hübner, G., & Bogner, F. X. (2005). Contrasting the theory of planned behavior with the value-belief-norm model in explaining conservation behavior. Journal of Applied Social Psychology, 35, 2150 – 2170.

Kibicho, W. (2006). Tourists to Amboseli National Park: A factor-cluster segmentation analysis. Journal of Vacation Marketing, 12(3), 218–231.

MacDonald, E., Milfont, T., & Gavin, M. (2016). Applying the Elaboration Likelihood Model to increase recall of conservation messages and elaboration by zoo visitors. Journal of Sustainable Tourism, 24(6), 866–881.

Manfredo, M. J., Bruskotter, J. T., Teel, T. L., Fulton, D., Schwartz, S. H., Arlinghaus, R., . . . Kitayama, S. (2017). Why social values cannot be changed for the sake of conservation. Conservation Biology, 31.4 (2017): 772–780.

Manfredo, M. J., Teel, T. L., & Dietsch, A. M. (2016). Implications of human value shift and persistence for biodiversity conservation. Conservation Biology, 30(2), 287–296. 

Mellish, S., Sanders, B., Litchfield, C. A., & Pearson, E. L. (2017). An investigation of the impact of Melbourne Zoo’s “Seal-the-Loop” donate call-to-action on visitor satisfaction and behavior. Zoo Biology, 36(3), 237–242. 

National Association for Interpretation (NAI) (2019.). Retrieved June 25, 2019 from https://www.interpnet.com/NAI/interp/About/About_Interpretation/nai/_About/what_is_interp

Orams, M. B. (1997). The effectiveness of environmental education: Can we turn tourists into ‘Greenies’? Progress in Tourism and Hospitality, 3, 295–306. 

Pahlow, K., & Dunstan, E. (2015). Zoos Victoria Community Conservation Master Plan. 

Pearson, E. L., Lowry, R., Dorrian, J., & Litchfield, C. A. (2014). Evaluating the conservation impact of an innovative zoo-based educational campaign: “Don’t Palm Us Off” for orangutan conservation. Zoo Biology, 33(3), 184–196. 

Petty, R. E., & Cacioppo, J. T. (1986). The elaboration likelihood model of persuasion. Advances in Experimental Social Psychology, 19, 123–205. 

Pfattheicher, S., Sassenrath, C., & Schindler, S. (2016). Feelings for the suffering of others and the environment: Compassion fosters proenvironmental tendencies. Environment and Behavior, 48(7), 929–945.

Powell, R. B., & Ham, S. H. (2008). Can ecotourism interpretation really lead to pro-conservation knowledge, attitudes and behaviour? Evidence from the Galapagos Islands. Journal of Sustainable Tourism, 16, 467–489.

Rabb, G. B., & Saunders, C. D. (2005). The future of zoos and aquariums: conservation and caring. International Zoo Yearbook, 39(1), 1–26. 

Salkind, N. (2016). Exploring research (9th ed). Boston: Pearson.

Saunders, C. D. (2003). The emerging field of conservation psychology. Human Ecology Review, 10, 137–149. 

Schultz, P. W. (2001). The structure of environmental concern: Concern for self, other people, and the biosphere. Journal of Environmental Psychology, 21(4), 327–339. 

Schwartz, S. H. (1992). Universals in the content and structure of values: Theoretical advances and empirical tests in 20 countries. Advances in experimental social psychology, 25, 1–65. 

Schwartz, S. H. (2006). A theory of cultural value orientations: Explication and applications. Comparative sociology, 5(2), 137–182. 

Schwartz, S. H. (2012). An overview of the Schwartz theory of basic values. Online Readings in Psychology and Culture, 2(1), 11. 

Schwartz, S. H., Cieciuch, J., Vecchione, M., Davidov, E., Fischer, R., Beierlein, C., . . . Demirutku, K. (2012). Refining the theory of basic individual values. Journal of Personality and Social Psychology, 103(4), 663. 

Skibins, J. C., Dunstan, E., & Pahlow, K. (2017). Exploring the Influence of Charismatic Characteristics on Flagship Outcomes in Zoo Visitors. Human Dimensions of Wildlife, 1–15. 

Skibins, J. C., Powell, R., & Hallo, J. (2016). Lucky 13: conservation implications of broadening “Big 5” flagship species recognition in East Africa. Journal of Sustainable Tourism, 24(7), 1024–1040. 

Skibins, J. C., & Powell, R. B. (2013). Conservation caring: Measuring the influence of zoo visitors’ connection to wildlife on proconservation behaviors. Zoo Biology, 32(5), 528–540. 

Skibins, J. C., Powell, R. B., & Hallo, J. C. (2013). Charisma and conservation: charismatic megafauna’s influence on safari and zoo tourists’ pro-conservation behaviors. Biodiversity and Conservation, 22(4), 959–982. 

Skibins, J. C., Powell, R. B., & Stern, M. J. (2012). Exploring empirical support for interpretation’s best practices. Journal of Interpretation Research, 17(27), 25–44. 

Smith, A., & Sutton, S. (2008). The role of flagship species in the formation of conservation intentions. Human Dimensions of Wildlife, 13, 127–140. 

Smith, L., Broad, S., & Weiler, B. (2008). A Closer Examination of the Impact of Zoo Visits on Visitor Behaviour. Journal of Sustainable Tourism, 16(5), 544–562. 

Smith, L., Curtis, J., & van Dijk, P. (2010). What the Zoo Should Ask: The Visitor Perspective on Pro-wildlife Behavior Attributes. Curator: The Museum Journal, 53(3), 339–357. 

Smith, L., Van Dijk, P., & Curtis, J. (2010). When does the zoo start to nag? Testing the limits for pro-wildlife behaviour requests. Journal of the International Zoo Educators’ Association, 46, 47–50.

Smith, L., Weiler, B., & Ham, S. (2011). The rhetoric versus the reality: A critical examination of the zoo proposition. In W. Frost (Ed.), Zoos and tourism: Conservation, education, entertainment? (pp. 59–68). Bristol: Channel View Publications.

Smith, L., Weiler, B., Smith, A., & van Dijk, P. (2012). Applying Visitor Preference Criteria to Choose Pro-wildlife Behaviors to Ask of Zoo Visitors. Curator: The Museum Journal, 55(4), 453–466.

Sponarski, C. C., Vaske, J. J., Bath, A. J., & Musiani, M. M. (2014). Salient values, social trust, and attitudes toward wolf management in south-western Alberta, Canada. Environmental Conservation, 41(4), 303–310.

St. John, Freya & Edwards-Jones, Gareth & Jones, Julia. (2010). Conservation and human behaviour: Lessons from social psychology. Wildlife Research, 37: 658–667.

Stern, P. C., Dietz, T., Abel, T., Guagnano, G. A., & Kalof, L. (1999). A Value-Belief-Norm Theory of Support for Social Movements: The Case of Environmentalism. Human Ecology Review, 6(2), 81–98. 

Stern, P. C., Dietz, T., & Guagnano, G. A. (1998). A brief inventory of values. Educational and Psychological Measurement, 58(6), 984(1). 

Tabachnick, B. G., & Fidell, L. S. (2007). Using multivariate statistics (5th ed.). Boston: Pearson Education, Inc.

Tilden, F. (1957). Interpreting our heritage (3rd ed.). Chapel Hill: The University of North Carolina Press.

Vaske, J. J. (2008). Survey research and analysis: Applications in parks, recreation and human dimensions. State College: Venture Publishing, Inc.

Walker, K., & Moscardo, G. (2014). Encouraging sustainability beyond the tourist experience: ecotourism, interpretation and values. Journal of Sustainable Tourism, 22(8), 1175–1196.

Weiler, B., & Smith, L. (2009). Does more interpretation lead to greater outcomes? An assessment of the impacts of multiple layers of interpretation in a zoo context. Journal of Sustainable Tourism, 17(1), 91–105. 

Woods, B. (2002). Good zoo/bad zoo: Visitor experiences in captive settings. Anthrozoos, 15(4), 343–360. 

World Association of Zoos and Aquariums (WAZA). (2019.). Retrieved June 19, 2019 from http://www.waza.org/en/site/about-waza

Wu, J. (2012). Advances in K-means clustering: a data mining thinking. Springer Science & Business Media.

Yocco, V. S., Bruskotter, J., Wilson, R., & Heimlich, J. E. (2015). Why Should I Care? Exploring the Use of Environmental Concern as a Frame of Communication in Zoos. The Journal of Environmental Education, 46(1), 56–71.

Zoos Victoria. (2014). 2014–2019 conservation master plan. Victoria, Australia: State of Victoria.