Living alongside predators can inflict substantial costs on local communities through loss of livelihoods and risk of life (Thirgood et al., Reference Thirgood, Woodroffe, Rabinowitz, Woodroffe, Thirgood and Rabinowitz2005; Inskip & Zimmermann, Reference Inskip and Zimmermann2009; Acha et al., Reference Acha, Temesgen and Bauer2018). These costs can often lead to negative attitudes (Kansky & Knight, Reference Kansky and Knight2014) or negative behaviours towards predators (Inskip & Zimmermann, Reference Inskip and Zimmermann2009; Marchini & Macdonald, Reference Marchini and Macdonald2012). Negative behaviours can include the retaliatory or pre-emptive killing of predators (e.g. Loveridge et al., Reference Loveridge, Valeix, Elliot and Macdonald2017) and it is believed that this has resulted in the decline or local extirpation of many carnivore populations (Treves & Karanth, Reference Treves and Karanth2003). Although attitudes can be a useful predictor of behavioural intentions (Kraus, Reference Kraus1995) this is not necessarily the case as the Theory of Planned Behaviour predicts that behavioural intentions can also be influenced by subjective norms and perceived behavioural control (Ajzen, Reference Ajzen1991; Fishbein & Ajzen, Reference Fishbein and Ajzen2011). Hence it is possible that the underlying determinants of attitudes and behavioural intentions differ (Liu et al., Reference Liu, McShea, Garshelis, Zhu, Wang and Shao2011; Harvey et al., Reference Harvey, Briggs-Gonzalez and Mazzotti2017). Therefore, understanding the factors that influence both attitudes and behavioural intentions towards wildlife, independent of each other, is essential in informing conservation interventions (Manfredo, Reference Manfredo2009; Harvey et al., Reference Harvey, Briggs-Gonzalez and Mazzotti2017).
As negative interactions with predators are often considered to be the root cause of negative attitudes and behaviours towards predators (Madden & McQuinn, Reference Madden and McQuinn2014), minimizing these negative interactions has become central to conservation efforts (Eklund et al., Reference Eklund, López-Bao, Tourani, Chapron and Frank2017). Minimizing these negative interactions can broadly be achieved through exclusion or coexistence strategies. Exclusion strategies rely on humans and wildlife being separated through, for example, the erection of physical barriers such as fences (e.g. Massey et al., Reference Massey, King and Foufopoulos2014), the removal of wildlife from anthropogenic landscapes (e.g. Treves & Karanth, Reference Treves and Karanth2003), the removal of people from wildlife areas (e.g. Karanth, Reference Karanth2007) or a combination of these. However, exclusion policies can lead to negative attitudes towards conservation (e.g. Ite, Reference Ite1996) and could have significant ecological consequences. Fencing, for example, is a particularly contentious issue (Creel et al., Reference Creel, Becker, Durant, M'Soka, Matandiko and Dickman2013; Packer et al., Reference Packer, Loveridge, Canney, Caro, Garnett and Pfeifer2013a, Reference Packer, Swanson, Canney, Loveridge, Garnett and Pfeiferb) as it leads to habitat fragmentation and loss of connectivity between protected areas (Creel et al., Reference Creel, Becker, Durant, M'Soka, Matandiko and Dickman2013). This is particularly important in dryland and migratory ecosystems (Durant et al., Reference Durant, Becker, Creel, Bashir, Dickman and Beudels Jamar2015) such as the Maasai Mara in Kenya. Coexistence, on the other hand, is a ‘dynamic but sustainable state in which humans and large carnivores co-adapt to living in shared landscapes where human interactions with carnivores are governed by effective institutions that ensure long-term carnivore population persistence, social legitimacy, and tolerable levels of risk’ (Carter & Linnell, Reference Carter and Linnell2016). Although it is difficult to influence the adaptation of carnivores, human adaptation can be promoted by minimizing the negative effects of predators and fostering positive attitudes through various, usually financial, incentives, a strategy that has proliferated in the Maasai Mara (Homewood et al., Reference Homewood, Trench and Brockington2012; Courtney, Reference Courtney2015).
In Maasailand, efforts to reduce the costs of livestock predation by predators include compensation and insurance schemes (e.g. Maclennan et al., Reference Maclennan, Groom, Macdonald and Frank2009; Bauer et al., Reference Bauer, Müller, Van Der Goes and Sillero-Zubiri2017), fortification of livestock enclosures (e.g. Lichtenfeld et al., Reference Lichtenfeld, Trout and Kisimir2015) and the use of deterrents (e.g. Ogada et al., Reference Ogada, Woodroffe, Oguge and Frank2003). Benefits are largely associated with wildlife-based activities such as tourism. Wildlife, and especially big cats, attract tourists (Mossaz et al., Reference Mossaz, Buckley and Castley2015). High predator densities in the Maasai Mara (Broekhuis & Gopalaswamy, Reference Broekhuis and Gopalaswamy2016; Elliot & Gopalaswamy, Reference Elliot and Gopalaswamy2017) together with the spectacle of the annual wildebeest migration have resulted in the Maasai Mara National Reserve being the highest-earning wildlife area in Kenya (Narok County Council & Trans-Mara County Council, 2012). However, some question whether tourism-related benefits really translate into coexistence with wildlife (Homewood et al., Reference Homewood, Trench and Brockington2012).
Other than socio-economic costs and benefits, factors such as conservation politics could play a role in people's attitudes and behavioural intentions towards predators. Goldman et al. (Reference Goldman, de Pinho and Perry2013), who describe conservation politics as the ‘uneven power dynamics related to land-use and conservation decision making’, report cases where Maasai killed lions Panthera leo as a political statement as they felt increasingly alienated by management and let down by government. There are many cases in Maasailand where the Maasai are resentful towards conservation efforts, especially when they are prohibited from cultural practices or excluded from key resources, such as grazing (Homewood et al., Reference Homewood, Trench and Brockington2012; Goldman et al., Reference Goldman, de Pinho and Perry2013; Hazzah et al., Reference Hazzah, Dolrenry, Kaplan and Frank2013; Bedelian & Ogutu, Reference Bedelian and Ogutu2017). It has been shown that providing access to resources (Measham & Lumbasi, Reference Measham and Lumbasi2013), including the local community in decision-making (Treves et al., Reference Treves, Wallace and White2009) and providing a sense of predator ownership (Dolrenry, Reference Dolrenry2013), can influence positive attitudes and behavioural intentions towards predators. As such, it is increasingly being recognized that a collaborative, rather than a top-down, approach whereby local communities are actively involved in conservation-related decisions, is required for conservation initiatives to be successful (Redpath et al., Reference Redpath, Linnell, Festa-Bianchet, Boitani, Bunnefeld and Dickman2017). Taking a collaborative approach can also overcome human–human conflict, the conflict between humans over the value of predators, which can hinder conservation efforts (Redpath et al., Reference Redpath, Bhatia and Young2015).
In the Maasai Mara the relationship between coexistence strategies and people's attitudes and behaviours towards predators is unclear. Here we assess whether people's attitudes and behaviours towards predators are influenced by the costs, benefits or factors relating to conservation politics. We hypothesized that costs, as measured by livestock lost to predators, would have a negative effect on people's attitudes and behaviour towards predators, whereas the perceived presence of benefits would have a positive effect. As management regimes of the various wildlife areas differ in their level of inclusion of the community we hypothesized that this, and a person's sense of ownership of wildlife, would influence people's attitudes and behavioural intentions towards predators.
The Maasai Mara landscape in south-west Kenya is home to the Maasai people who, historically, were a semi-nomadic pastoralist community but are now largely sedentary (Seno & Shaw, Reference Seno and Shaw2002). Traditionally, Maasai predominantly kept cattle but more recently there has been a shift to small stock, particularly sheep Ovis aries and goats Capra hircus, as these are more drought resistant (Bedelian & Ogutu, Reference Bedelian and Ogutu2017; Broekhuis et al., Reference Broekhuis, Cushman and Elliot2017).
Subdivision and privatization of communally owned land surrounding the Maasai Mara National Reserve occurred in 1999 and gave rise to concerns that livestock production would decrease (Seno & Shaw, Reference Seno and Shaw2002) and cultivation and fencing would increase, leading to a decline in wildlife (Lamprey & Reid, Reference Lamprey and Reid2004). As a result, tourism companies partnered with local landowners to form wildlife conservancies where landowners receive a fixed, monthly payment for leasing their land for wildlife-based activities on the condition that the landowner moves off that land and does not cultivate or develop it (Thompson et al., Reference Thompson, Serneels, Kaelo, Trench, Homewood, Kristjanson and Trench2009; Osano et al., Reference Osano, Said, de Leeuw, Ndiwa, Kaelo and Schomers2013). At the time this study was conducted, there were nine conservancies of varying size and with their own private management (Fig. 1). Each conservancy comprises individually-owned parcels of land (46–850 landowners). Different lease agreements by the conservancies have resulted in varied management policies. Conservancy-specific management policies are not publicly available for all the conservancies (but see Osano et al., Reference Osano, Said, de Leeuw, Ndiwa, Kaelo and Schomers2013; Bedelian, Reference Bedelian2014; Bedelian & Ogutu, Reference Bedelian and Ogutu2017), but some of the differences are as follows: lease fees vary per conservancy (USD 30–50 per ha per year); some conservancies utilize a percentage of the tourism revenue to run various community development programmes (Bedelian, Reference Bedelian2014; Courtney, Reference Courtney2015); some conservancies have a board that comprises both tourism partners and landowners, where landowners are included in the decision-making process (Bedelian, Reference Bedelian2014); some conservancies allow controlled grazing of cattle or enforce a livestock depredation insurance policy, whereas others do not. As with the Maasai Mara National Reserve, illegal grazing does occur within the conservancies, resulting in fines that can cause animosity, especially during dry periods when resources are scarce (Bedelian & Ogutu, Reference Bedelian and Ogutu2017). In the Maasai Mara National Reserve and the conservancies, the tourist lodges offer employment, frequently to local people, and pay additional lease fees to landowners (Thompson & Homewood, Reference Thompson and Homewood2002; Narok County Council & Trans-Mara County Council, 2012; Bedelian, Reference Bedelian2014).
There are no physical barriers between the wildlife areas and the surrounding community, and people, livestock and wildlife come into contact, which can result in negative interactions. In the Maasai Mara, Kolowski & Holekamp (Reference Kolowski and Holekamp2006) found that the main predator species that caused conflict were spotted hyaenas Crocuta crocuta (53%), leopards Panthera pardus (32%) and lions (15%; Kolowski & Holekamp, Reference Kolowski and Holekamp2006) but cheetahs Acinonyx jubatus also occasionally predate on small stock (Broekhuis et al., Reference Broekhuis, Thuo and Hayward2018a). Broekhuis et al. (Reference Broekhuis, Cushman and Elliot2017) found that during a 3-month period most households lost an average of 3.5% of their livestock to depredation. This loss can have significant financial implications, especially as livestock is the main source of income (Homewood et al., Reference Homewood, Trench and Brockington2012). Furthermore, Maasai people invest a cultural value in livestock that exceeds their economic worth, making losses difficult to bear (Galaty, Reference Galaty1982).
Data were collected using semi-structured interviews (Supplementary Material 1) conducted in June and July 2015. Ten Maasai men from the community, who had previous experience in conducting questionnaire-based interviews, were employed to conduct the survey. Before the survey the interviewers attended a full-day workshop in which they were tested on their interviewing and note-taking skills and their ability to use a GPS. The questionnaire was written in English but the interviews were conducted in Maasai. During the workshop each question was discussed and translated into Maasai to ensure consistency in translations between interviewers.
We were interested in the effect of livestock depredation on attitudes and behavioural intentions, so 820 households that kept livestock were randomly selected for the survey (see Broekhuis et al., Reference Broekhuis, Cushman and Elliot2017, for details). The most senior male of each household was interviewed, as they own the livestock and land, with the interviewer returning at a later date if he was not present. Interviews where data were missing were excluded from the analysis (N = 17). To minimize incentives for exaggerating answers, prior to each interview the interviewee was informed that the survey was independent of the government or any management companies and that no compensation would be provided for livestock losses. Before the interview the purpose was fully explained by the interviewer and the respondents were therefore not misled in any way. Respondents were also informed that they could withdraw from the interview at any time without penalty and that they did not have to answer any questions if they did not want to. The interview commenced only once informed consent was given verbally. All data were treated with strict anonymity, by removing any personal identifiers, and confidentiality. All respondents agreed to be interviewed.
Assessing attitudes and behavioural intentions
As predators are not restricted to the wildlife areas and edge effects can have significant negative impacts on protected predator populations (Loveridge et al., Reference Loveridge, Hemson, Davidson, Macdonald, Macdonalds and Loveridge2010), our questionnaire was designed to examine the attitudes and behavioural intentions of people living within or on the periphery of wildlife areas (Fig. 1). Predators are present throughout this area (Madsen & Broekhuis, Reference Madsen and Broekhuisin press) and we therefore decided to measure attitudes towards predators by asking the question ‘should people and livestock live alongside predators?’. We defined a ‘no’ or ‘yes’ answer as being a negative or positive attitude respectively. Individuals that were ‘unsure’ (N = 33) were excluded from further analyses. To determine behavioural intentions, respondents were asked ‘What do you do when your livestock has been killed by a predator?’ to which respondents could select one or more of the following answers: nothing, chase predator away, call authorities, call predator projects, kill predator, or other. For the analysis, the answers were condensed into those that answered they would kill a predator and those that did not select this answer. Killing predators is illegal in Kenya and doing so can incur a fine of KES 20 million (USD 222,000; exchange rate USD 1 = KES 90 in 2015), life imprisonment or both (Kenyan Wildlife and Conservation Act, 2013 (No. 47 of 2013)). Therefore, this question was asked and interpreted as a hypothetical behavioural intention rather than a self-reported behaviour.
People's behaviour towards predators can be influenced by several factors, including the costs and benefits associated with predators, inclusion in conservation decisions and having a sense of ownership of predators. To quantify the costs of living with predators, respondents were asked how many livestock they had lost to predators in the 3 months prior to the interview. Respondents were also asked whether they thought there were any benefits associated with predators and, if so, what these benefits were. This was an open question and responses were grouped based on their similarities. When asked about their primary source of income respondents could choose from the following categories: tourism sector, agriculture, pastoralism, business or other. The respondents were then asked whether they leased land to a conservancy and, if so, which conservancy. Respondents who were members of Lemek, Ol Chorro Oiroua, Olarro, Oloisukut and Enonkishu conservancies (N = 23) were excluded from analysis because of the small sample size, leaving 747 non-conservancy members and members of Naboisho, Ol Kinyei, Olare-Motorogi and Mara North. Lastly, respondents were asked whether they thought that the predators belonged to the community or the authorities.
Attitudes and behavioural intentions towards predators were analysed separately using a Generalized Linear Model with a binomial error structure and logit link function. For attitudes the response was either yes (people should live alongside predators = 1) or no (people should not live alongside predators = 0). Similarly, for behavioural intentions, the response was either yes (would kill a predator = 1) or no (would not kill a predator = 0) in response to livestock depredation. The predictor variables used to determine attitudes and behavioural intentions were self-reported livestock loss, primary source of income, conservancy membership, and perceptions of whether there were benefits associated with predators and whether the predators belonged to the community. Additionally, we included respondents’ attitudes in the behavioural intentions model. The candidate models were ranked using the Akaike Information Criterion (AIC) and when one model was superior this was used, otherwise parameter estimates were averaged across models with AIC differences < 2 correcting for model weights (Burnham & Anderson, Reference Burnham and Anderson2002). Parameter estimates (β) are presented with their 95% confidence intervals (CI) and are considered to be statistically significant if the 95% CI does not overlap zero. All statistical analyses were performed in R 3.3.0 (R Development Core Team, 2016).
In total, data from 747 interviews were used to determine people's attitudes and behavioural intentions towards predators in the Maasai Mara. Of these, 587 respondents (75.78%) had lost at least one head of livestock to predators in the 3 months prior to the interview and 443 respondents (59.30%) thought there were benefits associated with having predators. The benefits that the respondents named were largely related to tourism, employment, income and development, but some respondents also mentioned ecological benefits (i.e. predators kill herbivores) and aesthetics (Table 1). The majority of respondents derived their primary source of income from pastoralism (n = 553, 74%) followed by tourism (n = 112, 15%), business (n = 46, 6%) and agriculture (n = 22, 3%). Of the 747 respondents, 392 (52%) did not lease land to a conservancy and over half of the respondents (n = 429, 57.43%) felt that the predators in the area belonged to them rather than to the government or the conservancies.
More than half of the respondents (n = 432, 57.52%) thought that people and livestock should live alongside predators. Their response to this question was not associated with the number of livestock that were lost (β = 0.008, CI = −0.004–0.020) but the belief that predators were associated with benefits did have a positive influence (β = 1.709, CI = 1.294–2.124). In other words, the majority of respondents who said that predators bring benefits also thought that people and their livestock should coexist with them (Fig. 2). Similarly, the majority of the respondents who said that predators belonged to the community thought that people, livestock and predators should coexist (β = 1.286, CI = 0.888–1.683; Fig. 2). In addition, a higher proportion of respondents who were members of a conservancy responded ‘yes’ to the question about whether people and livestock should coexist with predators than did respondents who were not members of a conservancy, but this effect differed by conservancy (χ2 = 135.08, df = 4, P < 0.001; Fig. 2). Of the four conservancies, the majority of the respondents who were members of Olare-Motorogi, Naboisho and Mara North thought that people and livestock should live alongside predators but this was not the case for members of Ol Kinyei. Respondents who were members of Ol Kinyei differed significantly from the three other conservancies (P < 0.001) in that they thought that people and livestock should not live alongside predators, which was similar to respondents who were not members of a conservancy (P = 1.000). Occupation also had an effect (χ2 = 30.59, df = 4, P < 0.001; Fig. 2) and a higher proportion of respondents working in the tourism industry thought that people and livestock should live alongside predators compared to agriculturists (P = 0.001), pastoralists (P = 0.001), business people (P = 0.001) and other groups (P = 0.034).
Only a small number (n = 79, 10.30%) said they would kill a predator if it killed their livestock. The majority of respondents (72.52%) said that they would call the authorities (Table 2). The behavioural intention of killing a predator if it killed a respondent's livestock was not influenced by the number of livestock that were killed (β = 0.008, CI = −0.006–0.021), the benefits associated with predators (β = −0.309, CI = −0.914–0.196), whether the respondent thought that predators belonged to the community (β = 0.143, CI = −0.390–0.677) or the respondent's occupation (χ2 = 5.712, df = 4, P < 0.223). However, respondents’ behavioural intentions were influenced by their attitude towards predators (W = −0.571, CI = −1.048–−0.101) as the majority of respondents that did not think that people and livestock should coexist with predators said that they would kill a predator if it killed their livestock.
* This includes the Mara Cheetah Project, the Mara Lion Project and the Mara Hyena Project.
The majority of respondents agreed that people and livestock should live alongside predators and this attitude influenced a respondent's behavioural intention towards predators if it killed their livestock. It is possible that, because it is illegal to kill predators, some respondents did not respond truthfully to the hypothetical question about whether they would kill a predator if it killed their livestock. Despite this, the results were significant and are likely to be conservative. In addition, the results are corroborated by other studies, especially those conducted in Maasailand (e.g. Romañach et al., Reference Romañach, Lindsey and Woodroffe2007; Marchini & Macdonald, Reference Marchini and Macdonald2012; Hazzah et al., Reference Hazzah, Dolrenry, Naughton, Edwards, Mwebi and Kearney2014; Hazzah et al., Reference Hazzah, Bath, Dolrenry, Dickman and Frank2017).
Respondents’ attitudes towards predators were influenced by multiple variables but not by the cost of losing livestock. This is similar to findings in other areas (Zimmermann et al., Reference Zimmermann, Walpole and Leader-Williams2005) but it is possible that attitudes could be influenced by loss relative to overall wealth rather than by absolute losses (Romañach et al., Reference Romañach, Lindsey and Woodroffe2007; Inskip & Zimmermann, Reference Inskip and Zimmermann2009). In addition, costs were measured over a 3-month time-frame and it is possible that attitudes are shaped by historic or prolonged costs (Kansky & Knight, Reference Kansky and Knight2014). Alternatively, experiencing damage might not always be a driver of attitudes (Kansky et al., Reference Kansky, Kidd and Knight2014) as intangible costs, such as fear or alienation, rather than tangible costs, such as direct monetary losses, might be a better predictor of attitudes (Kansky & Knight, Reference Kansky and Knight2014).
Our results showed that benefits and employment in the tourism industry had a positive influence on respondents’ attitudes towards predators. However, in terms of employment, only a small percentage of the people interviewed (15%) obtained their main income from being employed in the tourism industry. This is supported by previous findings in the Maasai Mara that income from tourism-based activities only contribute to a small proportion of households (Homewood et al., Reference Homewood, Trench and Brockington2012) and favour a small elite (Lamprey & Reid, Reference Lamprey and Reid2004). Therefore, employment in the tourism industry is likely to influence only the attitudes of a small proportion of the population. However, the majority of the benefits mentioned by the respondents were related to tourism either directly through development, sale of craftwork or visits to cultural bomas, or indirectly through leasing land to conservancies for wildlife-based activities. Because of these socio-economic benefits, tourism has increased rapidly in the Maasai Mara (Lamprey & Reid, Reference Lamprey and Reid2004). Although using tourism to link conservation and better livelihoods is a popular tool, it is not always successful (Kiss, Reference Kiss2004). Therefore, care needs to be taken in putting too much emphasis on tourism, as high volumes of tourism can have negative ecological impacts (Broekhuis, Reference Broekhuis2018b; Buckley et al., Reference Buckley, Morrison and Castley2016) that can consequently destroy the tourism product and can be negatively affected by both national and international events (Homewood et al., Reference Homewood, Trench and Brockington2012). As the tangible benefits (i.e. employment) were mentioned more often than the intangible benefits (i.e. aesthetic value) we suggest that alternative, wildlife-based incomes, such as conservation payments, should be considered. For example, in Belize, people who are part of a jaguar Panthera onca camera-trap programme receive payments for captures of jaguars and other key wildlife (Harvey et al., Reference Harvey, Briggs-Gonzalez and Mazzotti2017).
A higher proportion of respondents who were members of a conservancy believed that people and livestock should live alongside predators compared to those who did not lease land to a conservancy. This is probably because of the multitude of benefits offered to conservancy members (Bedelian, Reference Bedelian2014; Courtney, Reference Courtney2015). However, this was not consistent across the conservancies examined. Conservancies can provide important resources for livestock in terms of grass, water and salt licks, especially during the dry season, but prohibiting access to these resources can create animosity (Bedelian & Ogutu, Reference Bedelian and Ogutu2017). For example, Hazzah et al. (Reference Hazzah, Dolrenry, Kaplan and Frank2013) found that restricted access to resources during the dry season was linked to both negative attitudes and behavioural intentions towards lions. As access to resources differ per conservancy, this could reflect the differences in attitudes towards predators between the different conservancy members. In addition, animosity could be increased when herders are fined for grazing in areas or during times when access is prohibited (Bedelian & Ogutu, Reference Bedelian and Ogutu2017). Similarly, the presence of compensation (through an insurance scheme that members pay into) and the inclusion of landowners in the decision-making process varied across the wildlife areas. This inclusion could be an important factor to consider, especially as a higher number of respondents who felt that the predators belonged to the community, rather than the authorities, thought that people, livestock and predators should coexist. Inclusion, through a sense of ownership and community engagement, is an important part of effective conservation strategies (Madden & McQuinn, Reference Madden and McQuinn2014) and our results suggest that conservation politics are likely to influence people's attitudes. We propose that once management plans are made public, the factors influencing these differences are explored, as this would aid future management and coexistence strategies.
We focused our survey on men as Maasai men own the land and livestock and are often associated with the killing of predators, especially lions (e.g. Goldman et al., Reference Goldman, de Pinho and Perry2013). However, recent research has shown that women value benefits, such as conservancy membership, differently to men (Keane et al., Reference Keane, Gurd, Kaelo, Said, de Leeuw and Rowcliffe2016) and that women may be less tolerant towards predators than men (e.g. Carter & Allendorf, Reference Carter and Allendorf2016; Harvey et al., Reference Harvey, Briggs-Gonzalez and Mazzotti2017). We would therefore advise that women are included in future surveys relating to attitudes and behavioural intentions towards predators. In addition, we measured attitude and behavioural intentions as a dichotomous response but it is likely that both attitudes and behavioural intentions lie on a continuous scale (Treves, Reference Treves2012). By assessing attitudes and behavioural response on a continuous scale, the strength of these could be measured (Harvey et al., Reference Harvey, Briggs-Gonzalez and Mazzotti2017; Hazzah et al., Reference Hazzah, Bath, Dolrenry, Dickman and Frank2017).
It is encouraging that more than half of the respondents thought that people, livestock and predators should coexist, especially as exclusion strategies, such as fencing of Africa's wildlife areas (Packer et al., Reference Packer, Loveridge, Canney, Caro, Garnett and Pfeifer2013a), are increasingly advocated. However, fencing of an ecosystem such as the Maasai Mara, where free movement is critical for the access of transient resources would be detrimental to both wildlife and people (Notenbaert et al., Reference Notenbaert, Davies, De Leeuw, Said, Herrero and Manzano2012). As predator populations are negatively impacted by edge effects (Loveridge et al., Reference Loveridge, Hemson, Davidson, Macdonald, Macdonalds and Loveridge2010), the overall willingness of respondents to coexist with predators bodes well for their conservation. However, a relatively large proportion of respondents (42.48%) felt that people, livestock and predators should not coexist. In the long term, this could have an impact on predator populations as people may wish to exclude predators from community land that has not been set aside for wildlife-based activities. Our results suggest that focusing on the benefits, rather than the costs, associated with predators, could be a more effective strategy to foster attitudes favouring coexistence. In addition, promoting a sense of ownership of predators may enhance a willingness to coexist. We acknowledge that more work is needed to understand better the complexities of coexistence within this landscape. We propose that future research is conducted within a Theory of Planned Behaviour framework by including factors such as subjective norms and perceived behavioural control (Ajzen, Reference Ajzen1991) and we hope that our findings will provide a starting point.
We thank the National Council for Science and Technology, Narok County Government, the Maasai Mara Wildlife Conservancies Association and the community for permission to carry out this study, Basecamp Foundation Kenya for providing funding to allow us to conduct the interviews, and the African Wildlife Foundation for funding the barazas (local meetings) in which the results were presented and discussed with the community. Additional funding was provided by WWF and through donations to the Kenya Wildlife Trust. We thank Francis Sopia for feedback on the questionnaire and for fruitful discussions, Clevers Ntokoiwuan, Njapit Pariken, Miton Taki, Pesi Saning'o, Daniel Momposhi, Charles Nkoitoi, Noosaron Simoren, Wilson Rotiken, Danson Kaelo and David Naurori for conducting the interviews, the respondents for volunteering their time, Fiona Tande for assisting with data entry, and the reviewers for their valuable critiques.
Conceived and designed the study: FB, MK, DKS and NBE. Trained the interviewers: MK and DKS. Supervised data collection: FB, MK and DKS. Informal analysis: MK and DKS. Formal analysis: FB and NBE. Data validation: FB and MK. Wrote the paper: FB and NBE. Revised and edited the paper: MK and DKS.
Conflicts of interest
Permissions to conduct the questionnaire survey were granted by the National Council for Science and Technology (Permit No.: NACOSTI/P/15/9772/5795), Narok County Government, the Maasai Mara Wildlife Conservancies Association and the community.