Introduction The over population of wild and feral animals is increasing as an environmental problem in many parts of the world due the pressure on native flora and ecosystems. (BBC, 2013; Hall, 2015; Kaji, Saitoh, Uno, Matsuda, & Yamamura, 2010). Examples include deer in Northern Japan and Northeast USA, the urban fox in England, possums in New Zealand and the crown of thorns starfish in the Great Barrier Reef. This phenomenon is also happening in Australia. Recent news reports of huge kangaroo populations devastating grazing land in western Queensland (Arthur, 2015) and a spike in Koalas eating away their habitat in the Cape Otway area of Victoria (Paul, 2015) have highlighted this problem. While the overpopulation of koalas is causing environmental damage to natural gum trees, to the point that they will not regenerate, it is difficult to enforce population control because these animals hold such as positive place as an Australian symbolic animal. Hence, there is some controversy whether they should be culled by environment advocates as part of an ongoing population/environmental management program. To help facilitate appropriate wildlife management in light of the controversial environmental problems, Non-Government Organisations (NGOs) play an essential role as a conduit between government and local citizens. For example, Australian Koala Foundation contributes to the conservation and management of the wild Koala through conducting strategic research for Koala management, conservation and community education in Australia (Australian Koala Foundation, 2015). However, NGOs are currently faced with challenges, such as acquisition of funds to support such work and gathering supporters for volunteer activities. This is because the competitiveness in the not-for-profit-sector has intensified due to an increase in the number of NGOs and oligopoly of donation share by high-profile NGOs (Foster & Fine, 2007; Sunder, 2015). To deal with these challenges NGO’s are resorting to more commercial types of marketing communications such as the use of celebrity endorsement or using celebrities as spokespeople. Although using celebrities as spokespeople for the NGO sector has become a common advertising strategy (De los Salmones, Dominguez, & Herrero, 2013; Wheeler, 2009), research into what kind of characteristics of spokesperson would lead effectively to change customer's attitude and behavioural intentions is limited. This celebrity/cause match is especially important for many environmental NGOs who have to deal with controversial environmental problems (e.g. wildlife management for overpopulated animals). This research examines the differences between the relevant expertise and perceived attractiveness of the celebrity spokesperson and its effect on the public’s perception of trustworthiness of the NGO. As the role of the celebrity spokesperson to encourage the public’s intentions to donate increases another issue arises: can the same strategy be used to solicit the donation of time (by volunteers)? This latter dilemma is something that is rarely experienced in the for-profit or commercial sector. This study presents a conceptual model that may help to identify answers to these questions and will extend the current research on celebrity endorsement. It should also bring out new academic insights about the process of building source credibility and a detailed evaluation of the spokesperson’s role in creating a two dimensional approach to behavioural intentions. Literature review Celebrity endorsement is a common advertising technique used by many organisations to build an association between a well-known and well-liked personality and the company’s brand in order to increase consumer’s awareness and liking for the brand (Spry, Pappu, & Bettina Cornwell, 2011). By utilizing the endorsement of a celebrity spokesperson, the product/service, band and/or company is able to leverage the positive attributes and characteristics of the spokesperson to the advantage of that product, band and/or company image (Erdogan, 1999; Ohanian, 1990). Recently, this strategy of utilizing celebrities as credible spokespeople has been adopted by many socially purposed organisations and NGOs (De los Salmones, et al., 2013; Samman, Auliffe, & MacLachlan, 2009; Wheeler, 2009). In order to evaluate the effectiveness of the spokesperson, source credibility is used as a key measure. The source credibility is defined as ‘a communicator's positive characteristics that affect the receiver's acceptance of a message’ (Ohanian, 1990, p. 41). The concept of source credibility has been established through the development of two important models: source credibility model and source attractiveness model. The source credibility model, proposed by Hovland, Janis, and Kelley (1953), suggests that the effectiveness of a message depends on perceived level of expertise and trustworthiness in an endorser (Erdogan, 1999). Hovland, et al. (1953) analyze the factors which lead to the perceived credibility of the spokesperson and defined that two essential items of source credibility namely, expertise and trustworthiness. Moreover, the source attractiveness model, takes a social psychological perspective (McGuire, 1985) and is defined as another important factor that is likely to affect customers’ perception of the brand. The source attractiveness model explains that the effectiveness of a message depends on source's 'familiarity', 'likability', 'similarity', and 'attractiveness' to the respondent (Erdogan, 1999). This model brings about the idea that attractiveness is also a factor determining source credibility. Ohanian (1990) combines these two models and defines the construct of endorser source credibility as consisting of three sub-dimension items (i.e. expertise, trustworthiness, attractiveness). By adopting the source credibility model, researchers have studied the effects of spokesperson on customer's attitudes and behavioural intentions in various research settings and conditions (Amos, Holmes, & Strutton, 2008; Erdogan, 1999; Lafferty & Goldsmith, 1999; Ohanian, 1991). However, a comprehensive review of the literature reveals that some gaps still exist in the work that has been undertaken in this field. Firstly, rather than source credibility consisting of three independent variables (expertise, trustworthiness, attractiveness) Busch and Wilson (1976) and Johnson and Grayson (2005) suggest that for trust building within service marketing and sales research, the constructs of expertise and attractiveness are regarded as antecedent factors of trustworthiness. This approach has not been used in research into celebrity endorsement. Secondly, although most studies have dealt with concept of behavioural intentions as a unidimensional concept, this unidimensional way could make the actual effects of spokesperson unclear because different behavioural intentions may hold different meanings, for different potential sponsors resulting in different forms of donation. In service quality research Zeithaml, Berry, and Parasuraman (1996) confirm that different types of service quality may influence differently any one of five different behavioural intentions (i.e. loyalty, switch, pay more, external response, and internal response). Thus, different types of celebrity endorser (expert or attractive) could also influence different types of behavioural intentions. For example the attributes of celebrity endorser may have an influence on whether someone would be more inclined to donate money, to donate time (volunteer for the organisation), and/or spread positive word of mouth. Conceptual framework & hypotheses development Although this study investigates the effects of the various combinations of endorser's characteristics on customer’s attitudes and behavioural intentions, research shows that different endorser’s characteristics do affect different customer attitudes and behaviour (Eisend & Langner, 2010; Lord & Putrevu, 2009). Lord and Putrevu (2009) find that attractiveness drives customer's behavioural intentions when customer's motivation is transformational (i.e. affective), whereas expertise and trustworthiness are influential when their motivation is informational (i.e. cognitive). Eisend and Langner (2010) reveal that attractiveness is a determinant of positive customer attitudes in the immediate condition (i.e. ad effects after just 60 seconds), whereas expertise is effective in the delayed condition (i.e. ad effects after one or three days). They also found that a high-expertise and high-attractiveness endorser is particularly effective towards customer attitudes in both conditions. Thus, considering the different effects on customer’s attitudes by different endorser’s characteristics our research objective is to examine which type of spokesperson is more effective in influencing a customer's attitudes and behavioural intentions for NGOs dealing with controversial environmental problems. In addition to solving the research objective, the research will also fill the research gaps indicated through the literature review. Our model (See Figure 1) shows that both the expertise and attractiveness of a celebrity spokesperson will lead to consumers’ perception of the trustworthiness of that spokesperson. Considering the insights from the perspective of trust building processes in the services marketing and sales literature, the celebrity’s expertise and/or attractiveness as antecedent factors of trustworthiness should also have an effect on organisation credibility through trustworthiness as a mediator. Moreover, we propose that the effects of a spokesperson’s expertise and/or attractiveness will influence differently the organisational credibility depending on the level of issue controversy they deal with. For example, Wheeler (2009) found that a celebrity endorser that showed a logical connection with the organisation increased organisational credibility and behavioural intentions. Therefore it is expected that for an environmental NGO that has to deal with controversial issues (such as reducing numbers of animals in habitats under stress from overgrazing) an expert spokesperson may well be a better fit and gain more trust and therefore more credibility for the organisation than an attractive one. However, where the issue the spokesperson is dealing with is not controversial this pattern may well reverse with an attractive celebrity being the more effective spokesperson. From this point of view, the level of issue controversy that the NGO deals with is hypothesized as a moderator between both expertise and attractiveness and trustworthiness. The customer’s perception of both the trustworthiness/credibility of the celebrity chosen as spokesperson and the credibility of the organisation will influence their (the customer’s) attitudes toward the organisation. In the model we follow Amos, et al. (2008) and Erdogan (1999) in asserting that positive organisational credibility will positively influence attitudes toward the firm. From the view that a unidimensional behavioural intention could make actual outcomes unclear, three behavioural intentions (i.e. donation of time, donation of money, word of mouth) options are proposed. This distinction is especially important when considering that many NGOs unlike for-profit organisations have a need to both increase financial resources and human resources. The NGO/celebrity endorser conceptual model is presented in figure 1. Conclusion & future research direction The conceptual model developed in this study research has implications for both academics and managers. Firstly, as an academic contribution, the idea that expertise and attractiveness contributes to trustworthiness, which is reported in trust building in service marketing, should raise new considerations about source credibility building process in celebrity endorsement research. Moreover, the moderating influence of the degree of controversy will play an important part in that relationship, especially for NGOs having to deal with controversial issues such as culling wildlife to protect the environment. The approach that multiple types of behavioural intentions are important outcomes may also enable a more detailed evaluation of the effects of celebrity spokespeople. The interactions between spokesperson characteristics (e.g. expertise, attractiveness) and the three behavioural intentions should be further investigated within the celebrity endorser field of research. Secondly, since the literature which evaluates the effect of celebrity endorsement for NGOs in the context of controversial social issues has been still under-researched, the outcomes of this research will be valuable for most NGO managers struggling with same issues. Moreover, by applying multiple behavioural intentions, more detailed insights how to increase each behavioural intention (i.e. donation of money, donation of time, positive word of mouth) through assigning different spokesperson with different characteristics (i.e. attractiveness and expertise) may enable NGOs to more flexible in assigning a suitable spokesperson. The characteristics of the spokesperson may need to change depending on the resource acquisition requirements the NGO has at that time. Hence, this research will provide meaningful insights from the both academic and managerial perspectives. As a future research direction, in order to validate this conceptual model it is suggested that a between subjects experiment be conducted. The experiment could consist of a scenario whereby a spokesperson is trying to gain public support for the activities of an environmental NGO. Here one could manipulate the expertise (high vs low) and the attractiveness (high vs low) of the spokesperson and the message (controversial vs non-controversial) to establish if differences in the spokesperson characteristics would have an effect on the different behavioural intentions of the respondents.

• Toyohiko Sugimoto(University of Queensland, Australia)
• Alastair Tombs(University of Queensland, Australia)