Heliyon 8 (2022) e08924

Contents lists available at ScienceDirect

Heliyon
journal homepage: www.cell.com/heliyon

Research article

Community perceptions towards the impacts of ecotourism development in

the central highlands of Ethiopia: the case of Lake Wanchi and its
adjacent landscapes
Abebe Tufa Angessa a, *, Brook Lemma b, Kumelachew Yeshitela c, Mahammed Endrias a
a
Department of Natural Resource Management, Ambo University, Ambo, Ethiopia
b
College of Natural and Computational Sciences, Addis Ababa University, P. O. Box 1913, 1110, Addis Ababa, Ethiopia
c
Ethiopian Institute of Architecture, Building Construction and City Development (EiABC), Addis Ababa University, P. O. Box 518, Addis Ababa, Ethiopia

A R T I C L E I N F O A B S T R A C T

Keywords: Scientific studies harmonizing biodiversity conservation, ecotourism development, and local livelihoods in areas
Biodiversity conservation of natural endowments have significant importance in the welfare of society. The objective of this study was to
Community perception investigate the local community's perception of the impacts of the Wanchi Ecotourism Association (WETA) and
Ecotourism development
test the relationship between these perceptions and some socio-demographic predictor variables. Data was
Lake Wanchi
Multinomial regression
collected using household surveys, focus group discussions, key informant interviews, field observations, and
Local livelihoods document analysis. Simple statistical analysis, such as descriptive statistics, cross-tabulations, multiple response
sets, and chi-square tests were used to analyze the socio-demographic characteristics and opinions on ecotourism
development. Multinomial logistic regression was used to analyze the local community's perceptions towards the
impacts of ecotourism using socio-demographic variables as predictors of the community responsiveness. The
non-quantifiable information was analyzed using qualitative descriptions. There was enough evidence for the high
resident's support for ecotourism development and perceptions of its positive impacts, while there was limited
community participation, less economic benefit, and inequitable sharing of the revenues generated from WETA.
The examination in the distribution of observed and expected perception responses on the impacts of ecotourism
showed statistically significant differences (x2 ¼ 110.833, df ¼ 3, P ¼ 0.000). The multinomial logistic regression
revealed that the variables of educational level, duration of stay in the study landscape, ecotourism benefit, and
place of residence from the ecotourism attraction center have a significant association with respondents'
perception toward impacts of ecotourism. The study supported the social exchange theory, in that those who
benefited from ecotourism viewed it as a development preference, and so are more likely to have positive atti-
tudes concerning ecotourism.

1. Introduction biodiversity and ecosystem services (Christ et al., 2003; ECA, 2011;
Admasu, 2020). Since ecotourism-based livelihoods are given opportu-
The term ecotourism developed in the late 1980s as a direct conse- nities to benefit directly from nature, they presumably have an incentive
quence of the world's recognition of sustainable development and global to stop external threats to biodiversity (Torquebiau and Taylor, 2009;
environmental practices (Diamantis, 1999). Since its inception, Lapeyre, 2010; Mbaiwa and Stronza, 2010).
ecotourism has gained adequate attention in academic circles and Empirical evidence shows that economic incentives from ecotourism
policy-makers as it balances between conservation and development by have the potential to stimulate income diversification and risk manage-
creating synergistic relationships among natural landscapes, local resi- ment among households (Christ et al., 2003; Kiss, 2004; Lapeyre, 2010).
dents, and the tourism industry (Zacarias and Loyola, 2017). It offers an Economic diversification, income-generation, and job security, in turn,
opportunity to develop products that can contribute to socio-economic assist the local communities to reduce the over-exploitation of nature's
development, local livelihood improvement, and visitor experience resources. The premise is that natural areas are the reason why tourists
while safeguarding local culture and maintaining the natural state of come to visit and that if the communities receive benefits from protecting

* Corresponding author.
E-mail address: [email protected] (A.T. Angessa).

https://doi.org/10.1016/j.heliyon.2022.e08924
Received 11 June 2021; Received in revised form 3 December 2021; Accepted 5 February 2022
2405-8440/© 2022 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).
A.T. Angessa et al. Heliyon 8 (2022) e08924

those natural areas; their livelihood strategy will no longer depend on Lehager” in 2020 by the government of Ethiopia, which aims to develop
their exploitation of the local natural resource (Goodwin, 2002; Rechlin three major natural landscapes (including Gorgora and Koisha sites) as
et al., 2008). It was stated that, even though, understandings of the key part of sustainable landscape management, biodiversity conservation and
environmental concepts are limited, most local communities have posi- livelihood improvement approach. Besides, the United Nations World
tive attitudes towards ecotourism development (including interaction Tourism Organization (UNWTO, 2021) initiative also designated Lake
with eco-tourists) and biodiversity conservation (Makindi, 2016). Wanchi and its adjacent landscapes as one of the best tourism villages of
Ecotourism is not, therefore, simply another niche market within the 2021 on December 2, 2021; by considering its role in safeguarding rural
tourism industry; rather a way of life, a set of practices and principles villages, along with their landscapes, natural and cultural diversity, and
that, if properly planned and effectively implemented will harmonize the their local values and activities (https://www.unwto.orginewsi
triple objectives of tourism industry, namely, local livelihoods, biodi- unwto-announces-list).
versity conservation and the travel industry (Christ et al., 2003; Honey, A few pieces of research related to ecotourism development were
2008; Makindi, 2016). conducted at Lake Wanchi and its surrounding landscapes (Ogato et al.,
Despite its potential for providing and enhancing livelihood oppor- 2014; Ketema, 2015a, 2015b). These studies indicated that contrary to
tunities, ecotourism is not always the solution to conservation challenges Lake Wanchi's and the surrounding landscapes' greatest potential for
and local development, especially in developing countries, like Ethiopia community-based ecotourism development, there were hindrances that
(ECA, 2011; SCBD, 2015; Admasu, 2020). It has been widely documented impede the sustainability of WETA. Hence, detailed assessment and
that tourism development does not only bring positive impacts but also evaluation on the implementation of WETA to identify its benefits and
has potentially negative impact on the environment and resident com- costs to the environment and local residents may help to minimize the
munities (Lankford and Howard, 1994; Ko and Stewart, 2002). Some negative environmental and socio-cultural footprint of WETA and
scholars even label tourism as a “double-edged sword” that involves both enhance its contribution to ecological conservation, while realizing sus-
positive and negative aspects for the host communities (Wang and Pfis- tainable local development strategies.
ter, 2008). Similarly, the ecotourism industry may create downturns to Furthermore, understanding the knowledge and perceptions of the
the sustainability of the local environment and the living conditions of residents of ecotourism destinations towards ecotourism is crucial for the
residents, unless it is properly planned, effectively implemented, and success of ecotourism development and environmental management of
community-centered in terms of management, decision-making, and the destination areas (Vodouh^e et al., 2010; Holladay and Ormsby,
benefit-sharing mechanisms (ECA, 2011). There are many reported in- 2011). It also helps policy and decision-makers to develop and imple-
cidents where forms of ecotourism, which are not sufficiently ment a plan that would incorporate the concerns of the residents and
community-focused, are harming the environment, and where local/- realize sustainable biodiversity conservation and environmentally
indigenous communities are not receiving sufficient benefit from the friendly livelihood strategies (Harun et al., 2018). Closely associated
projects (WWF, 2001). The implementation of most ecotourism projects with community attitude is the social exchange theory that has been
has received much criticism, as they failed to provide benefits for local widely accepted as an appropriate hypothetical framework by re-
communities (Monteros, 2002; Krüger, 2005), and a significant lack of searchers to explain community perception toward the impact of tourism
support for nature conservation (Mühlh€ausler and Peace, 2001). There development (Wang et al., 2006). From tourism perspectives, social ex-
are also concerns regarding the direct benefits to local communities and change theory suggests that residents' attitude toward tourism and their
potential negative impacts associated with eco/tourism including crea- subsequent level of support for its development will be influenced by
tion of a dependency syndrome on handouts among the local people, and their expectations in terms of the benefits or costs incurred in return for
erosion of indigenous cultural values through misconduct and bad in- the services they offer (Ap, 1992; Andereck et al., 2005). In other words,
fluence (Makindi, 2016). In community-focused ecotourism, the com- tourism effects were viewed positively when the exchange of resources is
munity is taking care of its natural resources to gain income through high for the host community in either the adjusted or disturbed exchange
operating an ecotourism enterprise and using that income to improve its relation, whereas tourism effects were viewed negatively if the exchange
livelihoods. The approach is to make livelihoods drive conservation of resources is low (Ap, 1992). Most perception studies focus on inves-
rather than simply being compatible with it. tigating the differences in communities' attitudes according to their
However, most ecotourism in developing countries, including socio-economic and demographic attributes (Wang et al., 2006). There-
Ethiopia, is characterized by haphazard planning, lack of environmental fore, the objectives of this study were to: (1) assess the local community's
standards and monitoring, stark seasonality, and domination of tourism understanding and attitude on ecotourism development and its impacts
in the overall economy. Besides, although ecotourism-related research (both positive and negative) in the Lake Wanchi and adjacent landscape;
has been a research focus for the past few decades in Ethiopia, scientific (2) evaluate the involvement of the communities and their benefits from
studies with the aim of generating knowledge and evidence in WETA, and (3) analyze the major socio-demographic characteristics that
ecotourism destinations are still largely lacking (ECA, 2011; Admasu, could have a significant effect on community perceptions towards the
2020). This has hindered the planning of the ecotourism sector for impacts of ecotourism development.
biodiversity conservation, environmental management, and
socio-economic development of the local communities and the country at 2. Study area and methods
large (ECA, 2011). One of such data deficit ecotourism destinations in
Ethiopia is Lake Wanchi and its adjacent landscapes, located at the 2.1. Study area
central highlands of the country (Ketema, 2015a; Angessa, 2020).
The unique natural and cultural landscapes, historical features, and The study was conducted in Wanchi district, Oromia, Ethiopia with
rich flora and fauna make Lake Wanchi and its surrounding landscapes special reference to Lake Wanchi and its adjacent landscapes. Lake
among the most popular ecotourism destinations in Ethiopia. As a result, Wanchi is located in the central highlands of the country at about 155 km
Wanchi Ecotourism Association (WETA), which is a local community- to the southwest of Addis Ababa (8o 45’ – 8o 490 N and 37o 50’ - 37o 55’
centered initiative, was established in 2002 with the objective of E). It is found in the Hero Wanchi rural Kebele (the lowest administrative
fostering the conservation of natural and cultural resources as well as unit) and is bordered by 12 other adjacent Kebeles (Figure 1). Lake
supporting the livelihoods of local communities. According to the Wanchi is a picturesque crater lake formed as a result of volcanic erup-
WETA's policy, 8% of the service providers' revenue and half of the tions (Abebe et al., 1998). According to Degefu et al. (2014), the lake has
entrance fee go directly to the community in a transparent way for the a surface area of about 560 ha and mean and maximum depths of 28 m
improvement of its services. Lake Wanchi and its adjacent landscape is and 107 m, respectively. The landscape is a highland area characterized
one of the three nationally launched ecotourism projects “Gebeta by mountain hills, steep slopes, gorges, and valleys. The altitude ranges

2
A.T. Angessa et al. Heliyon 8 (2022) e08924

Figure 1. Map of the study area.

from 2810 – 3385 m above sea level. The average annual temperature is Ethiopian highlands Wattled ibis (Bostrychia carunculata) are some of the
17.8  C with a maximum of 23.4  C and a minimum of 12.2  C, whereas bird species that are common to the area (Ketema, 2015b).
the mean annual rainfall is 1402 mm with peck rainfall in July and
August (Angessa, 2020). Unimodal type of rainfall with longer rainy 2.2. Methods
periods extending from June to September is the characteristic of the
landscape. According to the international slope classifications standards 2.2.1. Data collection and sampling techniques
of WOCAT (2007), the slope of the area ranges from flat (0–3%) at the This study employed information from socio-economic data collected
lake water surface to very steeply sloped terrain (>60%) near the ridges. through household surveys (HHS) supplemented with focus group dis-
Lake Wanchi and its adjacent landscapes are among the most exciting cussions (FGDs); key informant interviews (KIIs), direct field observa-
ecotourism destinations in Ethiopia, featuring spectacular natural land- tions, and document analysis. Prior to the actual data collection, a
scapes (mountains, valleys, waterfalls and Lake Wanchi itself with its detailed reconnaissance survey was conducted to obtain baseline infor-
islands and peninsulas). The natural vegetation; mineral waters and hot mation on population, faunal and vegetation distribution, topography,
springs; ancient churches and monasteries; and the eye-catching huts accessibility, climate, and infrastructures about the study landscape. In
with Enset (Ensete ventricosum) dominated home gardens make Lake addition, identification of subjects and possible sampling sites for the
Wanchi and its adjacent landscapes one of the most ecotourism potential questionnaire administration; determination of sample size for the HHS;
destinations. The lake provides an attractive landscape scenery, nature a decision of the required number of FGDs and KIIs were made during the
and culture with its ability of ecosystem in accepting managed level of reconnaissance survey. A pilot survey was also conducted by randomly
visitors, and cultural landscapes being shaped. Owing to the topographic selecting 12 respondents who were not included in the main sample
attractiveness of the landscape, most tourists call Lake Wanchi and its group to assess the validity of the research instruments (Hamito, 2001).
adjacent landscape as “The Switzerland of Africa”, while others call it as This enabled us to evaluate and modify those questions, in which the
“The hidden Garden of Eden” (Angessa, 2020). respondents encountered difficulty to respond. The HHS enumerators
Sub-afro-alpine vegetation dominated by Erica species is the charac- were recruited and trained on the objective of the research, on how to fill
teristic vegetation type of the study landscape. However, there are also the questionnaire, how to approach sensitive questions, such as illegal
indigenous plant species such as Hagenia abyssinica, Hypericum revolutum, resource utilization, and on those questions considered as taboos by the
Protea gaguedi, Rosa abyssinica, Lobelia giberroa, Solanecio gigas, Rumex local community. The questionnaire was interpreted into the local lan-
nervosus and Rubus species (Angessa et al., 2020) that have tourism guage (Afan Oromo), and the enumerators were monitored and evalu-
importance, especially during their flowering seasons. Colobus guereza, ated by the researchers during the data collection process. Data collection
Tragelaphus sylvaticus, Sylvicapra grimmia, Anubis baboon, Common jackal, instruments were evaluated for ethical considerations and approved by
and Common hyena are the commonly observed wild mammals. Gyps the research ethical review board of the college of agriculture and vet-
africanus, Milvus aegyptius, Tauraco leucotis, and the endemic bird to the erinary sciences of Ambo University.

3
A.T. Angessa et al. Heliyon 8 (2022) e08924

Data for socio-economic characteristics, ecotourism knowledge, and employed for the selection of key informants to include knowledgeable
community perceptions towards the impacts of WETA on the environ- individuals from different social groups including elders, community
ment, economy, and socio-cultural aspects of the local community was leaders, village chiefs, WETA managers and experts, tour guides, and
collected using HHS. The sample size was determined using the standard experts from the district culture and tourism office. The main focus points
formula by (Yamane, 1967) stated in Eq. (1) below. Sample households of the interviews included the demographic and socio-economic condi-
were selected randomly from a list of registered total resident households tions of the people, ecotourism resources and activities; the perception of
obtained from the Hero Wanchi Kebele Office. Similarly, sample re- the local communities on the ecotourism development; the positive and
spondents were selected randomly from those selected sample house- negative impacts of WETA with respect to the environment, local econ-
holds, so as to give each person or family in the households an equal omy, and socio-cultural aspects of the local community; the major chal-
chance of being selected for the sample survey. The fieldwork was carried lenges and opportunities in the development of ecotourism; and
out between October and December 2019, based on a simple random strategies on the future sustainable management of Lake Wanchi and its
sampling of 240 households. The purpose of the survey was clarified for adjacent landscapes.
respondents before raising any questions of interest to gain his/her In addition, against the social exchange theory background, described
confidence and interest. A semi-structured questionnaire containing both earlier, the major socio-demographic characteristics that were believed
open-ended and fixed response questions was used to collect data to affect the respondent's understanding of the impacts of ecotourism
through on-site surveys by questioning the respondents in face-to-face development were considered for multinomial logistic regression anal-
style interviews. The questionnaire was administered randomly to the ysis (Table 1).
survey respondents. Finally, the household's survey result was
cross-checked with the information obtained from the FGDs and KIIs, 2.2.2. Data analysis
field observations, and other pertinent secondary sources to capture the All the data collected were summarized and carefully coded to
consistency in the responses. perform the appropriate statistical computations, after which it was
subjected to a two-stage analysis. In the first stage, statements dealing
N with respondent's socio-demographic characteristics and the knowledge
n¼ (1)
1 þ NðeÞ2 and awareness related to ecotourism were analyzed using simple statis-
tical analysis, such as descriptive statistics, cross-tabulations, multiple
where, n is the sample for the households, N is total number of house- response sets, and the non-parametric chi-square tests. Descriptive sta-
holds of Hero Wanchi Kebele and e is the level of precision. The level of tistics and cross-tabulations were carried out to produce frequency
precision is a guarantee for the representativeness of a sample from the counts and percentages based on tables and graphical presentations
selected population. The accepted level of precision for representative (Field, 2000). Chi-square test was used to determine the degree of as-
samples usually ranges from 1 to 10% (Limaei et al., 2014). The sociations between the dependent and independent variables, and to
commonly used 5% level of precision was used for this research. estimate how closely an observed distribution matches an expected dis-
Purposive sampling was used for selecting focus group discussants. tribution. In the second stage, multinomial logistic regression was used to
Participants were selected based on their home distance from the center analyze the local community's perceptions towards the impacts of
of the ecotourism site, their influence in the society, their knowledge of ecotourism development using socio-demographic variables affecting the
ecotourism and ecotourism-related activities, their age, and duration of respondent's understandings. The data collected through FGD, KII, and
stay in the study landscape. Ten pre-designed open-ended questions were field observation (non-quantifiable information) was analyzed and
used for FGDs. Four FGDs consisting of 6, 7, 7, and 8 individuals, interpreted using narrations, context analysis, and qualitative
respectively, were conducted. Participants were invited to discuss issues descriptions.
according to their convenience. Most often, community leaders were
approached in advance, and development agents of the respective 2.2.2.1. Multinomial logistic regression analysis. Prior to the multinomial
Kebeles were communicated to organize and create conducive environ- logistic regression analysis, the distributions in the observed and ex-
ment for the discussion. Similarly, a purposive sampling technique was pected responses among the respondents perception on the impacts of

Table 1. Socio-demographic factors affecting respondent's perception and their theoretical assumptions.

Variables Scale Levels/categories Research hypothesis/Theoretical assumptions

Gender Nominal 1 ¼ Female Female respondents will be less likely to report the positive impacts of ecotourism development as
2 ¼ male compared to their male counterparts.

Educational status Ordinal 1 ¼ Primary level Respondents with a higher education level are more likely to report the positive impacts of ecotourism
2 ¼ Secondary level than those with a lower education level/with no formal education.

3 ¼ Tertiary level
4¼ No formal education
Occupation Nominal 1 ¼ Tour-guide There might be perception differences in reporting the impacts of ecotourism based on the respondent's
2 ¼ Merchant occupation.

3 ¼ Employer
4 ¼ Student
5 ¼ Farmer
Duration of stay Ordinal 1 ¼ < 10 years Respondents with longer duration of stay in the study area may have more supportive attitude towards
2 ¼ 10–20 years ecotourism development than those with a shorter duration of stay.

3 ¼ 20–30 years
4 ¼ Since birth
Ecotourism benefit Nominal 1 ¼ Yes Respondents receiving economic benefits from ecotourism are more likely to report the positive impacts
2 ¼ No of ecotourism than those not receiving economic benefits.

Ecotourism distance (km) Numeric Ranging from Home distance from the ecotourism attraction center may influence respondents' perception towards the
0.5km to 9km impacts of ecotourism development.

4
A.T. Angessa et al. Heliyon 8 (2022) e08924

ecotourism were checked using the nonparametric chi-square test, and whereas, very few of the respondents (1.7%) were older than 65 years.
showed statistically significant differences. Multinomial logistic regres- Gender-wise comparison of the age category indicated that relatively
sion was performed as a follow-up to the chi-square test to analyze the more proportions of males were represented in the 31–45 age category,
local community's perception towards the impacts of ecotourism using whereas females are more represented in the 46–65 age category
socio-demographic factors as predictors of the community responsive- (Figure 2a).
ness. The traditional 0.05 criterion of statistical significance was Contrary to the rural Ethiopian situations, where people with non-
employed for all tests. The entire model assumptions were checked, and formal education are greater in proportion, in the Lake Wanchi and its
found that no any of the assumptions have been violated. Then, the adjacent landscape a greater percentage (61.3%) of the respondents
correlations among the model predictors (independent variables) were attained primary education (grade 1–8 level); whereas, only 23% of the
checked for multicollinearity using Spearman's rank correlation co- respondents reported no formal education. The educational achieve-
efficients, and those predictors with a correlation coefficient above the ments of 14.6% of respondents ranged from grade 9 to grade 12 (sec-
conventionally accepted value of 0.4 cut-off level (Schober et al., 2018) ondary education). A very small proportion of the household heads
were removed (for instance age and membership status). The (0.8%) had attended tertiary education, which is above grade 12,
goodness-of-fit and the model-fitting information statistics were including some college and university level of formal education. As is
assessed, in order to check whether the model reasonably approximates common in most parts of Ethiopia, compared to the female respondent's
the behavior of the data. The Goodness of fit was explored and in no case males constituted the larger proportions in both the primary and sec-
this test found significant, implying that the model fits the data well. The ondary levels of education, while females constituted greater proportion
model fitting information table also confirmed that the addition of the of the respondents who have no formal education (Figure 2b). Unex-
predictors to a model that contained only the intercept significantly pectedly, the proportion of female respondents who attained a tertiary
improved the fit between the model and the data, that is, the final model level of education was slightly greater than that of male respondents
is outperforming the null model. In a final step, the entire independent (Figure 2b).
variables (without interactions) were fitted to the model using the cus- Given the focus of the survey was on household heads, the majority of
tom/stepwise–main effects with backward elimination, thus filtering out respondents (84.6%) were married households, which is consistent with
the statistically not significant independent variables (at P < 0.05). All the survey expectations. Low proportions of the respondents reported
the statistical analysis was performed using IBM SPSS statistics version that they were either single (12%), widowed (2.9%), and divorced
23. (0.4%). A large percentage (85%) of the respondents were farmers by
occupation, followed by students, which constituted 6.3% of the sample
3. Results households, while traders, employers, and tour guides together consti-
tuted 8.7% of the respondents (Table 2).
3.1. Socio-demographic profile of the sample households Information on the respondent's duration of stay in the study area
might indicate the level of knowledge and understandings about the
Male households constituted the largest proportion (84.2%), while socio-economic undertakings in the study landscape. In line with this, the
females constituted 15.8% of the survey respondents. The gender bias majority of the household heads (60.4%) have lived in the study land-
was the expected result in the Ethiopian household heads condition, scape since birth. A further 8.8% and 10.8% of the survey respondents
where the socio-cultural practices take for granted male heads of reported that they have lived in the study area for the last 20–30 and
households as responsible actors than female counterparts. In terms of 10–20 years, respectively. Only 20% of the survey respondents reported
the age profile of the survey households, relatively larger proportions as they have lived less than 10 years in the study landscape. Those re-
(45.4%) were included between the age categories of 31–45 years. spondents who were not lived in the study area since birth were asked to
Nearly proportional numbers of respondents, 23%, and 30%, were justify the reasons why they came to settle in the current study landscape.
included between the age categories of 15–30 and 46–65, respectively, Of these respondents, a greater proportion (72.6%) came to the study

Figure 2. Respondents (a) Gender by age category, (b) Educational status by gender. Note: Educational status: Non-formal ¼ no formal education, Primary ¼ 1–8
grade level, Secondary ¼ 9–12 grade level, Tertiary ¼ above grade 12.

5
A.T. Angessa et al. Heliyon 8 (2022) e08924

3.2. Local community's knowledge and perceptions on ecotourism

Table 2. Demographic and socio-economic characteristics of the sample house- development
holds (N ¼ 240).
Variables Responses N percent The local community's knowledge on eco/tourism and essential re-
Marital status Single 29 12.1% sources and their perception towards the impacts (positive or negative) of
Married 203 84.6% WETA were analyzed using the opinion of sample households. Results
Divorced 1 0.4% from the 240 HHS analysis indicated that the local community in Lake
Widowed 7 2.9% Wanchi and its adjacent landscapes were well aware of tourism and its
Occupation Farmer 204 85.0% impacts on landscape and local livelihoods. The majority of the re-
Trader 13 5.4% spondents (80.4%) know what tourism means. Tourism understood and
Employer 7 2.9% described as visiting and enjoying attractive and beautiful landscapes
Student 15 6.3% nearly by one-sixths of the respondents (14%); both natural and man-
Tour-guide 1 0.4%
made places that attract people by one-fourth of the respondents (25%);
Major Livelihoodsa Crop production 218 63.4%
natural and cultural resources that serve as a source of income by 43.3%,
and any recreational activity enjoyed by the people roughly by one-tenth
Livestock rearing 85 24.7%
(11.9%) of the respondents. Almost all of the respondents (97.1%)
Beekeeping 6 1.7%
remarked on the appropriateness of the establishment of ecotourism that
Monthly salary 5 1.5%
encompasses sustainability at Lake Wanchi and its adjacent landscapes.
Handcraft 3 0.9%
However, relatively a smaller proportion of respondents (32.1%) re-
Ecotourism 16 4.7%
ported that they are benefiting from the ongoing ecotourism and related
Petty trade 11 3.2%
activities. Of the respondents who reported they are benefiting, 29% of
Duration of stay in the study area Since birth 145 60.4%
them stated - collective benefits such as infrastructure and facilities that
20–30 years 21 8.8% put into consideration the conservation of natural habitats, and 24% of
10–20 years 26 10.8% them reported – exemption from land tenure taxes, which they wish to
<10 years 48 20.0% compensate society by conserving natural resources. The rest reported
If not since birth reason for coming Marriage case 10 10.5% employment opportunities, percent share from the cash revenue WETA
Get access to land 69 72.6% generates, biodiversity conservation, and income generation from the
Attracted by WETA and Hero 16 16.8% different ecotourism services (Table 3). The majority of respondents
Wanchi town (73.3%) reported that they were not a member of WETA, whereas, 90%
Do you have farmland Yes 168 70% of them expressed their willingness to join the WETA as members.
No 72 30% When asked about the presence of conflicts of interest over the use of
Farmland condition in the past 10 Increased 2 1.2% natural resources in Lake Wanchi and its adjacent landscapes, a slightly
years Decreased 37 22.0% higher proportion of the respondents (37.1%) reported the presence of
No change 129 76.8%
Farmland enough to support your Yes 0 0.0%
households? No 168 100.0% Table 3. Respondents knowledge and perception of WETA's development (N ¼
240 unless otherwise stated).
a
Dichotomy group tabulated at value 1.
Inquiry Replies N percent
Do you know what tourism means? Yes 193 80.4%
No 47 19.6%
Does ecotourism appropriate for Lake Yes 233 97.1%
landscape to get access to land, while 16.8% of them were reported as Wanchi? No 7 2.9%
they were attracted by the establishment of WETA and Hero Wanchi rural
Do you get benefit from ecotourism Yes 77 32.1%
town in the past two decades. Relatively a lower proportion of the activities? No 163 67.9%
respondent (10.8%) claimed that they came to the study landscape due to
Are you a member of WETA? Yes 64 26.7%
marriage cases.
No 176 73.3%
Most of the surveyed households (88.1%) were engaged in agricul-
Do you need to be a member of WETA? (n Yes 158 90.0%
tural activities including crop production and livestock rearing as their
¼ 176) No 18 10.0%
main source of livelihoods followed by ecotourism that constituted 4.7%
Ecotourism benefitsa (n ¼ 163) Employment 9 3.6%
of the livelihoods of the sample households. The remaining sample
opportunities
households depended on beekeeping (1.7%), employment (1.5%),
Collective benefits 73 28.9%
handcraft (0.9%), and petty trading (3.2%) for their survival. The ma-
Revenue sharing 5 2.0%
jority of the survey households (70.0%) reported that they have their
Biodiversity 47 18.6%
own farmland, where, the landholding size varied from 0.5 ha to 3.5 ha
conservation
with an average of 1.4 ha per household. On the other hand, the
land tenure tax 61 24.1%
household survey result indicated that there is a larger family size that exemption
ranged from 1 - 17 individuals with a mean of seven persons per Service-income 58 22.9%
household. Hence, all of the respondents who own farmland argued that
Conflict over the use of resources Yes 89 37.1%
the landholding size is not sufficient enough to sustain their family
No 80 33.3%
members. In terms of the conditions of landholding size, approximately
I have no idea 71 29.6%
three fourth (76.8%) of those respondents who have their own farmland
Conflict resolution mechanismsa local traditional system 102 65.0%
reported that there was no change in the landholding size in the past 10
legal or court system 42 26.8%
years, while nearly one-third (22%), and small (1.2%) reported a
ecotourism code of 13 8.3%
decrease and an increase in the landholding size, respectively. Table 2
conduct
shows the key demographic and socio-economic characteristics of the
a
surveyed households. Dichotomy group tabulated at value 1.

6
A.T. Angessa et al. Heliyon 8 (2022) e08924

conflicts of interests, whereas, somewhat proportional percentage of the

respondents, i.e., 33.3% and 29.6% reported the absence of any sort of Table 4. Case processing summary for the multinomial logistic regression.
conflicts, and as if they have no idea over such matters, respectively. The Variables N Marginal
majority of respondents (65.0%) stated the local traditional system or Percentage
“jaarssummaa” as a remedy to disputes in the society about sharing Respondents perception on the Positive impact 40 16.7%
natural resources of the watershed, while nearly one-fourth (26.8%) impacts of ecotourism Negative impact 15 6.3%
stated that they prefer to use the legal system of conflict resolution, and Both positive and 125 52.1%
smaller proportions (8.3%) stated WETA's tourism code of conduct can negative impact
serve as the conflict management system in the study landscape. I have no idea 60 25.0%
Regarding the impacts of WETA on the local environment, economy, Gender Female 38 15.8%
and socio-cultural aspects of the local community, over half of the re- Male 202 84.2%
spondents reported that WETA has both positive and negative impacts, Educational status Primary education 143 59.6%
while one-fourths of them reported as if they have no idea about its Secondary education 39 16.3%
impacts. Only small proportions of the respondents (6.25%) reported the Tertiary education 2 0.8%
negative impacts of WETA on the study landscape and its resident com- No formal education 56 23.3%
munity (Figure 3). Occupation Tour-guide 1 0.4%
Trader 13 5.4%
3.3. Perception of local communities on the development of WETA and its Employer 7 2.9%
impacts Student 15 6.3%
Farmer 204 85.0%
Prior to multinomial logistic regression analysis, the examination Duration of stay Less than 10 years 48 20.0%
in the distribution of observed and expected perception responses of For the last 10–20 26 10.8%
the respondents on the impacts of WETA was estimated to check years
whether they have significant differences or not and showed a sta- For the last 20–30 21 8.8%
tistically significant difference (x2 ¼ 110.833, df ¼ 3, P ¼ 0.000). This years
chi-square test result leads to the further exploration of whether re- Since birth 145 60.4%
spondent's perception significantly differed based on the socio- Benefit from ecotourism Yes 77 32.1%
demographic characteristics of the respondents (such as gender, edu- No 163 67.9%
cation level, occupation, duration of stay, receiving a benefit or not Valid 240 100.0%
from ecotourism, and place of residence). Hence, a multinomial lo- Missing 0
gistic regression was performed to model the relationship between the Total 240
respondent's socio-demographic characteristics and perception to- Subpopulation 123a
wards the impacts of WETA's development (Table 4). The Goodness-
a
of-fit table showed the consistency of the data with the model as- The dependent variable has only one value observed in 93 (75.6%)
subpopulations.
sumptions (Pearson x2 ¼ 310.932, df ¼ 342, P ¼ 0.885). The model

Figure 3. Respondents perception on the impacts of ecotourism at Lake Wanchi and its surrounding landscapes.

7
A.T. Angessa et al. Heliyon 8 (2022) e08924

development by classifying 87.2% correctly. The correct predictions

Table 5. Multinomial logistic regression Likelihood Ratio Tests of statistically were less frequent (15.0%) for those respondents who reported the
significant independent variables.
“positive impacts” of ecotourism development. However, the model
Effect Model Fitting Criteria Likelihood Ratio Tests showed poor performance in classifying the respondents who reported
AIC of BIC of -2 Log Chi- df Sig. the “negative impacts” of ecotourism (6.7%). Taking into account only
Reduced Reduced Likelihood of Square the null model, one would correctly predict 25.0% of the cases, while
Model Model Reduced using the logistic model to make such predictions resulted in 58.8%
Model
correct prediction, i.e., the model gets 33.8% more, as compared to the
Intercept 385.298 479.276 331.298a .000 0 .
modal category.
Educational 418.171 480.823 382.171 50.873 9 .000
status
3.3.2. The parameter estimates
Duration of 384.025 446.676 348.025 16.726 9 .053
The parameter estimates table was computed to present the estimated
stay
multinomial logistic regression coefficients of the model. As there were
Ecotourism 395.184 478.719 347.184 15.886 3 .001
benefit four levels of the dependent variable for this study there are three sets of
Ecotourism 388.618 472.153 340.618 9.319 3 .025 logistic regression model coefficients (three logits) for each independent
distance variable in the parameter estimates table, one for each comparison to the
reference category (Table 7). The model treated the responses to the
The chi-square statistic is the difference in -2 log-likelihoods between the final
model and a reduced model. The reduced model is formed by omitting an effect
impacts of WETA, “I have no idea” as the reference category. The first set
from the final model. The null hypothesis is that all parameters of that effect are 0. of coefficients are found in the "positive impacts" row (representing the
a
This reduced model is equivalent to the final model because omitting the comparison that WETA has positive impacts category to the reference
effect does not increase the degrees of freedom. category, I have no idea). The second set of coefficients is found in the
"negative impacts” row (representing the comparison that WETA has
negative impacts category to the reference category). The third set of
coefficients is found in the “both negative and positive impacts” row (this
fitting information table also confirmed that the full model statistically
time representing the comparison that WETA has both positive and
significantly predicted the dependent variables better than the null
negative impacts category to the reference category).
model (with no variables added), Pearson x2 (24, N ¼ 240) ¼ 93.638, Because the parameter estimates are relative to the reference cate-
Nagelkerke R2 ¼ 0.358, p < 0 .001. gory, the standard interpretation of the multinomial logit is that for a unit
Moreover, from the Likelihood Ratio Tests (which shows the overall change in the independent variable, the logit of outcome relative to the
statistically significant independent variables), statistically significant reference category is expected to change by its respective parameter es-
unique contributions were made by educational status, ecotourism timate (which is in log-odds units), given the other variables in the model
benefit, distance to the center of ecotourism, and partially by the dura- are held constant. Parameters with significant negative coefficients
tion of stay in the study area (Table 5). This implies that the removal of decrease the likelihood of that response category with respect to the
these four independent variables from the model would significantly reference category, and parameters with positive coefficients increase the
reduce the fit of the model to the data. However, gender and occupation likelihood of that response category with respect to the reference cate-
were the variables that do not significantly predict respondents’ gory. Since the coefficient does not have a simple interpretation, the
perception of the impacts of WETA. exponentiated values of the coefficients or the odds ratio (Table 7) are
normally considered for interpreting multinomial logistic regression
3.3.1. The classification table for the multinomial logistic regression results.
The classification table shows the practical results of using the The odds ratio of a coefficient indicates how the risk of the outcome
multinomial logistic regression model (Table 6). For each case, the pre- falling in the comparison group compared to the risk of the outcome
dicted response category is chosen by selecting the category with the falling in the reference group changes with the variable in question. An
highest model-predicted probability. Of the cases used to create the odds ratio >1 indicates that the risk of the outcome falling in the com-
model 25 of the 60 respondents (41.7%) who “have no idea” about the parison group relative to the risk of the outcome falling in the referent
impacts of ecotourism development were classified correctly. In partic- group increases as the variable increases (i.e., the comparison outcome is
ular, the model did extremely well at identifying those respondents who more likely). An odds ratio <1 indicates that the risk of the outcome
reported “both the positive and negative” impacts of ecotourism falling in the comparison group relative to the risk of the outcome falling
in the referent group decreases as the variable increases (i.e., the refer-
ence category outcome is more likely).
Table 6. Classification table of the perception of local communities on the im-
pacts of WETA. 3.3.2.1. Positive impacts relative to the reference category (I have no
Observed Predicted idea). Statistically significant logistic regression coefficients were
Positive Negative Both negative I have Percent observed for the respondent's educational status, duration of stay in the
impact impact and Positive no idea Correct study area, and whether they get a benefit from WETA or not (Table 7).
impact The educational status specifically represents the primary level of ed-
Positive impact 6 0 34 0 15.0% ucation (p < 0.001), and secondary level of education (p < 0.001). The
Negative 0 1 9 5 6.7% multinomial logit for primary education was 4.073 and that of sec-
impact ondary education was 4.267 units higher for those respondents who
Both positive 5 0 109 11 87.2% reported the “positive impacts” of ecotourism relative to those who
and negative
“have no idea” about the impacts of ecotourism. Given the other vari-
impact
ables in the model are held constant, a one-level increase in the
I have no idea 0 0 35 25 41.7%
educational status of the respondents, the odds ratio (relative risk) of
Overall 4.6% 0.4% 77.9% 17.1% 58.8%
Percentage
being reporting the “positive impacts” of WETA would be 58.7 and 71.3
times more likely, for respondents with primary education, and sec-
Note: Cells in the diagonal and bold are correct predictions of the model, while ondary education, respectively, as compared to those who “have no
cells off the diagonal (not bold) are in correct predictions.
idea” about the impacts of WETA. Therefore, at the 0.05 alpha levels,

8
A.T. Angessa et al. Heliyon 8 (2022) e08924

Table 7. Parameter estimates of the multinomial logistic regression model.

Community Perception toward the impacts of ecotourisma B Std. Error Wald df Sig. Exp(B) 95% CI for Exp(B)

Lower Bound Upper Bound

Positive impacts
Intercept -3.959 1.378 8.250 1 .004
[Education ¼ primary] 4.073 1.102 13.653 1 .000 58.704 6.769 509.122
[Education ¼ secondary] 4.267 1.211 12.418 1 .000 71.339 6.646 765.812
[Education ¼ tertiary] 3.931 .000 . 1 . 50.975 50.975 50.975
[Educational ¼ no formal] 0b . . 0 . . . .
[Duration of stay ¼ < 10 years] .169 .666 .064 1 .800 1.184 .321 4.363
[Duration of stay ¼ 10–20 years] .284 .837 .115 1 .735 1.328 .257 6.853
[Duration of stay ¼ 20–30 years] 2.041 .848 5.791 1 .016 7.696 1.460 40.555
[Duration of stay ¼ since birth] 0b . . 0 . . . .
[Tourism benefit ¼ yes] 2.094 .593 12.451 1 .000 8.116 2.537 25.968
[Tourism benefit ¼ no] 0b . . 0 . . . .
Distance from ecotourism -.124 .134 .849 1 .357 .883 .679 1.150
Negative impact
Intercept .025 1.067 .001 1 .981
[Education ¼ primary] .537 .652 .677 1 .410 1.711 .476 6.142
[Education ¼ secondary] -16.75 3015.87 .000 1 .996 5.298E-8 .000 .c
[Education ¼ tertiary] .012 .000 . 1 . 1.012 1.012 1.012
[Educational ¼ no formal] 0b . . 0 . . . .
[Duration of stay ¼ < 10 years] 1.126 .789 2.033 1 .154 3.082 .656 14.483
[Duration of stay ¼ 10–20 years] -16.73 3935.48 .000 1 .997 5.444E-8 .000 .c
[Duration of stay ¼ 20–30 years] 1.440 .896 2.586 1 .108 4.222 .730 24.424
[Duration of stay ¼ since birth] 0b . . 0 . . . .
[Tourism benefit ¼ yes] .073 .791 .009 1 .926 1.076 .228 5.065
[Tourism benefit ¼ no] 0b . . 0 . . . .
Distance from ecotourism -.292 .156 3.513 1 .061 .747 .551 1.013
Both positive and negative impacts
Intercept .580 .711 .666 1 .415
[Education ¼ primary] 1.775 .421 17.823 1 .000 5.902 2.588 13.456
[Education ¼ secondary] 2.038 .601 11.494 1 .001 7.674 2.363 24.925
[Education ¼ tertiary] 19.071 6105.77 .000 1 .998 191610792.3 .000 .c
[Educational ¼ no formal] 0b . . 0 . . . .
[Duration of stay ¼ < 10 years] .383 .487 .618 1 .432 1.467 .565 3.811
[Duration of stay ¼ 10–20 years] .586 .571 1.051 1 .305 1.796 .586 5.505
[Duration of stay ¼ 20–30 years] .362 .734 .243 1 .622 1.436 .341 6.052
[Duration of stay ¼ since birth] 0b . . 0 . . . .
[Tourism benefit ¼ yes] 1.204 .476 6.402 1 .011 3.333 1.312 8.467
[Tourism benefit ¼ no] 0b . . 0 . . . .
Distance from ecotourism -.265 .102 6.799 1 .009 .767 .628 .936
a
The reference category is: I have no idea.
b
This parameter is set to zero because it is redundant.
c
Floating point overflow occurred while computing this statistic. Its value is therefore set to system missing.

the null hypothesis was rejected and concluded that at least one of the factor of 7.70 for those respondents reported the “positive impacts” of
regression coefficient has been found to be statistically different from WETA relative to those who reported “I have no idea” about the impacts
zero for the educational status of the respondents with respect to the of WETA. Given the 0.05 alpha level, the null hypothesis was rejected and
“positive impacts” of the ecotourism category relative to the reference concluded that at least one of the regression coefficient has been found to
category, when the other variables in the model are held constant. be statistically different from zero for the respondent's duration of stay in
Although the addition of the duration of stay in the study area did the study area with respect to the “positive impacts” of WETA category,
only partially significantly improved the model compared to the null relative to the “I have no idea” category, given that the other variables in
model, the logistic regression coefficient for the duration of stay in the the model are held constant.
study area was statistically significant (p < 0.05) representing those re- The estimated logistic regression coefficients of ecotourism benefit
spondents who have to reside for 20–30 years being quite knowledgeable are also highly significant (p < 0.001) representing those respondents
(Table 7). The multinomial logit for the 20–30 years duration of stay in who are benefiting from WETA. The multinomial logit for the re-
the study area is 2.04 units higher for those respondents reported the spondents who reported the benefit from WETA was nearly doubled (2.1
“positive impacts” of WETA relative to those who “have no idea” about units higher) to report the “positive impacts” of WETA relative to those
the impacts of WETA. When the other variables in the model are held who “have no idea” about the impacts of WETA, given the other variables
constant, the odds ratio for a one year increase for the 20–30 years in the model are held constant. A one-unit increase in a benefit for those
duration of stay in the study area would be expected to increase by a respondents who reported the benefit from the ecotourism development,

9
A.T. Angessa et al. Heliyon 8 (2022) e08924

the relative risk of being reporting the “positive impacts” of WETA would negative” impacts of ecotourism category, relative to “I have no idea
be 8.12 times more likely as compared to those who “have no idea” about category”, given that the other variables in the model are held constant.
the impacts of WETA. In general, given the 0.05 alpha level and the other However, no statistically significant logistic regression coefficients were
variables in the model are held constant, the null hypothesis was rejected reported toward the “negative impacts” of WETA with respect to the “I
and concluded that at least one of the regression coefficient has been have no idea” about the impact of WETA.
found to be statistically different from zero for ecotourism benefit with
respect to the “positive impacts” of ecotourism category relative to the “I 4. Discussion
have no idea” category.
Ecotourism schemes planned and implemented according to the
3.3.2.2. Both positive and negative impacts relative to the reference cat- principles of ecological sustainability, community participation, eco-
egory. Statistically significant logistic regression coefficients were nomic and socio-cultural security of the host community will have a net
detected for the respondent's educational status, ecotourism benefit, and positive contribution to the well-being of the society while satisfying the
home distance from the center of WETA. The educational status precisely needs and aspirations of the eco-tourists (Christ et al., 2003). The pro-
represents the primary level of education (p < 0.001), and secondary motion of such a socio-culturally and environmentally sound ecotourism,
level of education (p < 0.01). The multinomial logit for primary educa- especially in areas of significant natural endowments like that of Lake
tion and secondary education was 1.78 and 2.04 units higher, respec- Wanchi and its adjacent landscapes, offers many opportunities in biodi-
tively, for those respondents who reported “both the positive and versity conservation while realizing sustainable local development stra-
negative” impacts of WETA relative to those who “have no idea” about tegies. However, many obstacles prevent the local communities and
the impacts of WETA. Given the other variables in the model are held other relevant ecotourism stakeholders from actually attaining such
constant, the odds ratio (relative risk) for a one-level increase for primary practical ecotourism development schemes. We investigated the local
education and secondary education would be expected to increase by a community's perception toward the impacts of the Wanchi ecotourism
factor of 5.9 and 7.67, respectively, for respondents to report “both the Association (WETA) at the biologically diverse, ecologically fragile, and
positive and negative impacts” of WETA relative to report “I have no socio-politically marginalized landscape in the central highlands of
idea” about the impacts of WETA. Given the 0.05 alpha level, the null Ethiopia. Generally, there were evidences for the high resident's support
hypothesis was rejected and concluded that at least one of the regression for ecotourism, while there was less community participation and benefit
coefficient has been found to be statistically different from zero for the from the undergoing ecotourism development. These findings and their
educational status of the respondents with respect to “both the positive implications for landscape management, biodiversity conservation, and
and negative” impacts of the WETA category relative to the “I have no local livelihood improvements were discussed in relation to other studies
idea” category, given that the other variables in the model are held elsewhere.
constant. The local communities in Lake Wanchi and its surrounding land-
When the other variables in the model are held constant, the esti- scapes were well aware of the “pros and cons” of ecotourism development
mated logistic regression coefficients of ecotourism benefit representing and were highly supportive of the establishment of WETA, where, they
those respondents who are getting benefit from WETA was statistically express their positive attitude to support the undergoing ecotourism ac-
significant (p < 0.05). The multinomial logit for the respondents who tivities at the study landscape (see Table 3). Nevertheless, the study
reported the benefit from WETA was 1.2 units higher for reporting “both indicated that a smaller proportion of the residents were involved and
the positive and negative” impacts of WETA relative to those respondents economically benefited from the revenues generated and services pro-
who “have no idea” about the impacts of ecotourism. Given the other vided by WETA. The focus group discussants and key informants
variables in the model are held constant, the odds ratio for one unit of remarked that these differences in ecotourism benefit formed two
increase in benefit for those respondents who reported they are getting different community groups, that is, the economically, socially, and
benefit from WETA would be expected to increase by a factor of 3.33 to politically marginalized (those who were alienated from the ecotourism
report “both the positive and negative” impacts of WETA relative to those development) and the empowered (those who benefited from the
who “have no idea” about the impacts of WETA. Generally, at the 0.05 ecotourism developments). Similarly, a study by Goodwin (2002) in
alpha level, the null hypothesis was rejected and concluded that at least different National Parks also specified that even though ecotourism
one of the regression coefficient has been found to be statistically presents additional income and employment opportunities, rural pop-
different from zero for ecotourism benefit with respect to “both the ulations remain largely marginalized from ecotourism-related de-
positive and negative” impacts of the WETA category, relative to the “I velopments. For instance, 99% of the revenues generated from
have no idea” category, given that the other variables in the model are ecotourism in Komodo national park (Indonesia) goes to the nearby town
held constant. residents, and not to those residents living within the park who were the
Similarly, the estimated logistic regression coefficients of ecotourism most underprivileged by restrictions of resource use within the park and
distance were statistically significant (p < 0.01) representing re- who lack the resources and opportunity to develop ecotourism facilities
spondents' home distance from the center of WETA attraction. The because of park restrictions (Goodwin, 2002). Furthermore, Kiss (2004)
multinomial logit for the respondent's home distance from the center of specified that usually ecotourism projects generate modest cash benefits,
WETA was 0.265 units lower for those respondents who reported “both and these are frequently grabbed by a relatively small proportion of the
the positive and negative” impacts of WETA relative to those who “have community.
no idea” about the impacts of WETA. Stated differently, respondents who This study also showed that the majority of the residents of Lake
are residing near to WETA attraction center are more likely to report Wanchi and its adjacent landscapes were not members of WETA, yet, a
“both the positive and negative impacts” of ecotourism as compared to larger proportion of the community showed their willingness to join
those who “have no idea” about the impacts of WETA. When the other WETA. This could be justified by the fact that the landholding size in the
variables in the model are held constant, a 0.5 km increase in home study landscape is not sufficient to support the livelihood of the house-
distance of the respondents, the relative risk of being reporting “both the holds (Table 2). Besides, due to the steep slope and highly fragile nature
positive and negative” impacts of WETA would be 0.77 times less likely of the landscape, some rural livelihood activities such as agriculture are
compared to those respondents who “have no idea” about the impacts of not suitable for the landscape. Thus, the desire to become a member of
WETA. Given the 0.05 alpha level, the null hypothesis was rejected and WETA could be what is expected from such rural communities with
concluded that at least one of the regression coefficient has been found to relatively disadvantaged backgrounds to secure their family livelihoods.
be statistically different from zero for the respondents home distance Likewise, Goodwin (2002) stated that tourists are affluent customers
from the center of WETA attraction with respect to the “both positive and with cash to expend; it would be amazing if large numbers of people in

10
A.T. Angessa et al. Heliyon 8 (2022) e08924

marginalized rural communities were not interested in finding a means of demographic characteristics. The hypothesis that residents with a
securing some of their disposable income. higher educational level may be more supportive of ecotourism devel-
Biodiversity conservation, employment opportunities, income gen- opment and are more likely to report the positive impacts of ecotourism
eration from ecotourism activities, exemption of land tenure taxes, and than those with a lower education level was highly confirmed. The
collective benefits such as infrastructure improvements and public fa- findings indicated that respondents with a higher educational level were
cilities were some of the limited benefits provided by WETA to the local more motivated and supportive of WETA. This could be attributed to the
residents. Studies also indicated that ecotourism offers many opportu- fact that individuals with higher educational levels have more opportu-
nities such as employment, job-creation, income-generation, and alter- nities to learn and understand the potential advantages and disadvan-
native means of livelihoods to those that may result in the destruction or tages of ecotourism that can accrue to a community, than those
overuse of natural resources (IUCN, 2012; Eshetu, 2014). Therefore, respondents with lower/no formal education. This positive correlation of
helping communities earn income from ecotourism projects provides high educational level with the support for WETA was consistent with
both an incentive for conservation and an economic alternative to studies elsewhere (Teye et al., 2002; Andriotis and Vaughan, 2003; Bi-xia
destructive activities (Kiss, 2004). Regarding the collective benefits, and Zhen-mian, 2017). Generally, it was confirmed that as the educa-
some scholars who studied residents perception toward tourism impacts tional level of the respondents increases, it is expected that respondents
also reported the resident's supportive attitude toward the contribution are more likely to develop a positive attitude towards ecotourism
of tourism for improvements in a set of community infrastructure and development.
public facilities (Andereck and Vogt, 2000; Andereck et al., 2005; Wang The hypothesis that respondent's home distance from the center of
and Pfister, 2008). The findings of this study also indicated that there was ecotourism attraction may influence residents' perception towards the
a conflict of interests over the use of natural resources among the impacts of ecotourism development was partially supported by this study
different stakeholders in Lake Wanchi and its adjacent landscapes. The result. Respondents who live relatively at a closer distance to WETA's
focus group discussants stated that the conflicts are mainly of between attraction center were more conscious about the impacts of ecotourism in
WETA and non-WETA members, between WETA and investors, between that they perceive “both the positive and negative” impacts of WETA than
the Orthodox Church and WETA, and between Orthodox Church and those who live relatively at a further distance from the attraction center.
investors”. The key informants also added that there are also conflicts of Their closeness to the center of WETA usually predisposes them to
interest among the WETA members over the arrangements of the economically dependent on ecotourism and, hence, they develop more
ecotourism service delivery system. This result was consistent with the positive attitudes towards WETA. At the same time, they are also more
knowledge that tourism induces conflict in the host community (Jafari, exposed to the negative influences of WETA that drive them to perceive
2001). Previous studies in the study landscape also reported the presence ecotourism negatively. More generally, as the resident's home distance
of conflicts of interest among the different stakeholders over the use of from the center of WETA's attraction increases, it is expected that resi-
natural resources (Ogato et al., 2014; Ketema, 2015a; Angessa et al., dents have less or no idea about the impacts of WETA. This result is
2019). Conflicts of interest over the use of natural resource in the local consistent with an earlier study, which stated residents residing farther
context refers any sort of disputes over the use of local natural resources, from the center of the tourism zone, were less concerned about the im-
particularly of the use of communal properties such as land (for grazing pacts of tourism development (Haley et al., 2005).
and cultivation), forests (for timber and fuel wood collection), and spring Partial confirmation was obtained regarding the hypothesis of the
water (for drinking purpose). resident duration of stay in the study landscape. Statistically significant
With regard to ecotourism impacts, the findings indicated that the logistic regression coefficients for the duration of stay in the study area
local residents were conscious of ecotourism development and recognized were observed for the “positive impacts” of the WETA category relative
both the positive and negative impacts of WETA on the local environment, to the “reference category” for those respondents who resided for 20–30
economy, and socio-cultural aspects of the local community (see Figure 3). years. More generally, as the respondent's duration of stay in the study
Amazingly, a similar result was reported by Andereck et al. (2005) in his landscape increases, it is an indication that respondents are more likely to
study on residents' perceptions of community tourism impacts in Arizona, support the existence of WETA. Likewise, a study by Davis et al. (1988)
USA. The economic and environmental benefits gained from WETA could indicated that long-term residents showed a more supportive attitude
be the reasons for the local community to develop a positive perception toward tourism development than those newly settled in the community.
towards WETA's impacts (see also Almeida et al., 2015). In addition to the The hypothesis that residents receiving economic benefits from
historical and cultural features and natural landscape experiences, there ecotourism development will have a more supportive attitude toward
are additional products that can be sold by the local people supporting ecotourism development and are more likely to report the positive im-
them to get cash income for family subsistence and diversify their liveli- pacts of ecotourism development than those receiving no economic
hoods (Goodwin, 2002). The slight negative impacts perceived by the benefits from the ecotourism venture was largely confirmed. The findings
residents may be associated with the conflicts of interest among the specified that those respondents who benefited from WETA were more
communities and socio-cultural aspects that are affecting the traditions inclined to support WETA than those who were not benefited. Similarly, a
and social associations of the resident communities. Likewise, a study by study by Wang et al. (2006) found that personal benefit was found to be
(Doǧan, 1989) identified that tourism development has negative impacts closely and positively related to residents' attitudes toward tourism
on the socio-cultural aspects of the local community such as habits, daily development and tourism impacts. In general, it can be remarked that in
routines, social lives, community relationships, beliefs, and values. The the study area as the respondent's economic benefits increase from the
key informants also added that there are considerable changes in land ecotourism development undertakings, it is expected that they are more
ownership taking place in Lake Wanchi landscape, with non-local land likely to develop a positive attitude towards WETA and become less
grabbers buying lakeshore wetlands for tourism lodges that raise the price concerned with the negative impacts of WETA.
of land and other consumable goods in the locality (see also Saveriades,
2000). However, the residents of Lake Wanchi and its surrounding land- 5. Conclusion and recommendations
scapes prefer to support ecotourism, putting its benefits ahead of its costs
(see also Almeida et al., 2015). Here, it was stated that the local resident's Ecotourism development that is not properly planned, effectively
attitude towards tourism development may not be determined only by implemented, and community-centered in terms of management,
supposed benefits and costs but also influenced by several moderating decision-making, and benefit-sharing mechanisms, may create downturns
factors (Lankford, 1994). to the sustainability of the local environment and to the living conditions
The study findings indicated that residents’ perceptions towards the of local residents. Studies of local resident's perception of the impacts of
impacts of WETA were significantly differed based on the socio- ecotourism are useful in setting up strategies that seek to minimize the

11
A.T. Angessa et al. Heliyon 8 (2022) e08924

negative environmental and socio-cultural footprints of ecotourism and Brook Lemma and Kumelachew Yeshitela: Conceived and designed
enhance its positive contribution. It offers also many opportunities in the experiments.
ecological conservation, and landscape maintenance while realizing sus- Mahammed Endrias: Conceived and designed the experiments; Per-
tainable local development strategies. Indeed, this study investigated the formed the experiments; Contributed reagents, materials, analysis tools
local community's perception toward ecotourism impacts at Lake Wanchi or data.
and its adjacent landscapes, central highlands of Ethiopia. In the study
area, there were enough evidence to conclude there was a positive rela- Funding statement
tionship between supports for ecotourism development and perceptions of
its positive impacts by the local community. Nonetheless, it was detected This research did not receive any specific grant from funding agencies
that there was limited community participation, less economic benefit, in the public, commercial, or not-for-profit sectors.
and inequitable sharing of the revenues generated from the ecotourism
development in the study area. However, in order for ecotourism to be Data availability statement
sustainable and contribute to maintaining the landscape and improving
local livelihoods more of the economic benefits of ecotourism should be Data will be made available on request.
allocated to the local people by empowering them to benefit from the
income generated and economic diversification.
Declaration of interests statement
A systematic analysis of the perception of local residents on
ecotourism impacts using socio-demographic factors as predictors of
The authors declare no conflict of interest.
their perception may help planners, policy, and local decision-makers to
identify real concerns and requirements of individuals in order for
Additional information
appropriate policies and decisions to take place. In this study, multino-
mial logistic regression was used to assess whether community percep-
No additional information is available for this paper.
tions and attitudes towards ecotourism development impacts
significantly differed based on the socio-demographic characteristics of
Acknowledgements
the respondents. Here, the variables of educational level, duration of stay
in the study landscape, ecotourism benefit (whether receiving a benefit
We acknowledge financial and logistic support provided by Ambo and
or not from WETA), and place of residence from WETA attraction center
Addis Ababa Universities, the German Academic Exchange Service
have a significant association with respondents’ perception toward the
(DAAD) and IDEA WILD. We greatly appreciate Leuphana University for
impacts of WETA development. Moreover, the study was consistent with
its logistics support for the first author during his short-term research stay
the social exchange theory in that those who received benefits from
in Lueneburg, Germany. We sincerely thank all the local communities,
WETA viewed ecotourism as a development preference, and so are more
and the Kebele and Wereda authorities for their support and permission
likely to have positive attitudes concerning ecotourism.
for our research.
To ensure the success of WETA and maintain the Lake Wanchi and its
adjacent landscapes to play an important role in keeping the conditions of
References
the natural and cultural landscape, conservation of biodiversity, and
improving of local livelihoods, while fulfilling the requirements of Abebe, T.W., Mazzarini, F., Piero, M., 1998. The Yerer-Tullu Wellel volcanotectonic
ecotourism, the study suggest the following considerations: (1) improving lineament: a transtensional structure in central Ethiopia and the associated magmatic
the participation of the local communities and allow better access to activity. J. Afr. Earth Sci. 26 (1), 135–150.
Admasu, S., 2020. Ecotourism as a tool to substantially funding conservation endeavors in
ecotourism resources. (2) Resolving the conflicts of interest of the different
Ethiopia. J. Tourism Hospit. 9 (3), 432.
stakeholders, and establishing strong and cooperative partnerships among Almeida, F., Balbuena, A., Cortes, R., 2015. Resident’s attitudes towards the impacts of
the different stakeholders and decision-makers involved in the ecotourism tourism. Tourism Manag. Perspect. 13, 33–40.
Andereck, K.L., Valentine, K.M., Knopf, R.C., Vogt, C.A., 2005. Residents’ perceptions of
development. (3) Scaling up of WETA to a more sustainable form of
community tourism impacts. Ann. Tourism Res. 32 (4), 1056–1076.
ecotourism so as to enhance community livelihoods and ecological resil- Andereck, K., Vogt, C., 2000. The relationship between residents’ attitudes toward
ience. (4) Regional or local decision-makers must use evidence-based tourism and tourism development options. J. Trav. Res. 39, 27–36.
decisions and devise appropriate policies, strategies, and investments Andriotis, K., Vaughan, R.D., 2003. Urban residents’ attitudes toward tourism
development: the case of Crete. J. Trav. Res. 42 (2), 172–185.
that could balance the land-use trade-offs between environmental con- Angessa, A.T., 2020. Use of Land-Use Land-Cover Mapping and Biodiversity Analysis as
servation and satisfying immediate human needs. Tools for Community-Based Ecotourism Planning of Lake Wanchi Watershed , Central
Highlands of Ethiopia. PhD Dissertation. Addis Ababa University.
Angessa, A.T., Lemma, B., Yeshitela, K., 2019. Land-use and land-cover dynamics and
5.1. Limitations of this research their drivers in the central highlands of Ethiopia with special reference to the Lake
Wanchi watershed. Geojournal 1225–1243.
Angessa, A.T., Lemma, B., Yeshitela, K., Fischer, J., May, F., Shumi, G., 2020. Woody plant
In selecting sample households our target population was not the total
diversity, composition and structure in relation to environmental variables and land-cover
population, rather the total households in the study area. Because of the types in Lake Wanchi watershed, central highlands of Ethiopia. Afr. J. Ecol. 1–12, 00.
absence of complete information on the total population and resource Ap, J., 1992. Residents’ perceptions on tourism impacts. Ann. Tourism Res. 19 (4),
665–690.
constraints, we decided our target population to be the total households
Bi-xia, C., Zhen-mian, Q., 2017. Community attitudes toward ecotourism development
of the study area. This has implications on the validity, to be specific and environmental conservation in nature reserve: a case of Fujian Wuyishan
external validity or generalizability of our results, i.e., the findings from National Nature Reserve, China. J. Mt. Sci. 14 (7), 1405–1418.
the survey may not be generalized with confidence to the population of Christ, C., Hillel, O., Matus, S., Sweeting, J., 2003. Tourism and Biodiversity: Mapping
Tourism’s Global Footprint. Washington, DC.
interest. Davis, D., Allen, J., Cosenza, R.M., 1988. Segmenting local residents by their attitudes,
interests, and opinions toward tourism. J. Trav. Res. 27 (2), 2–8.
Declarations Degefu, F., Herzig, A., Jirsa, F., Schgerl, M., 2014. First limnological records of highly
threatened tropical high-mountain crater lakes in Ethiopia. Trop. Conserv. Sci. 7 (3),
365–381.
Author contribution statement Diamantis, D., 1999. The concept of ecotourism: evolution and trends. Curr. Issues
Tourism 2 (2–3), 93–122.
Abebe Tufa Angessa: Conceived and designed the experiments; Per- Doǧan, H.Z., 1989. Forms of adjustment: sociocultural impacts of tourism. Ann. Tourism
Res. 16 (2), 216–236.
formed the experiments; Analyzed and interpreted the data; Contributed ECA, 2011. Ecotourism in the Intergovernmental Authority on Development (IGAD) Region:
reagents, materials, analysis tools or data; Wrote the paper. an untapped potential with considerable socio-economic opportunities. Addis Ababa.

12
A.T. Angessa et al. Heliyon 8 (2022) e08924

Eshetu, A.A., 2014. Ecotourism as a viable strategy for livelihood diversification and Makindi, S.M., 2016. Local communities, biodiversity conservation and ecotourism: a
sustainable natural resource management in Ethiopia. J. Environ. Sci. Water Res. 3 case study of the Kimana Community Wildlife Sanctuary, Kenya. Afr. J. Hospital.
(2), 40–52. Tour. Leisure 5 (3) (special edition) http//: www.ajhtl.com.
Field P, A, 2000. Discovering Statistics using SPSS for Wiondows: Advanced Techniques Mbaiwa, J., Stronza, A., 2010. The effects of tourism development on rural livelihoods in
for the Biginner. Sage Publications, London. the Okavango Delta, Botswana. J. Sustain. Tourism 18 (5), 635–656.
Goodwin, H., 2002. Local community involvement in tourism around National Parks: Monteros, R.L.E., 2002. Evaluating ecotourism in natural protected areas of La Paz Bay,
opportunities and constraints. Curr. Issues Tourism 5 (3–4), 338–360. Baja California Sur, Mexico: ecotourism or nature-based tourism? Biodivers. Conserv.
Haley, A.J., Snaith, T., Miller, G., 2005. The social impacts of tourism: a case study of 11 (9), 1539–1550.
Bath, UK. Ann. Tourism Res. 32 (3), 647–668. Mühlh€ausler, P., Peace, A., 2001. Discourses of ecotourism: the case of fraser island,
Hamito, Desta, 2001. Research Methods in Forestry: Principles and Practices with Queensland. Lang. Commun. 21 (4), 359–380.
Particular Reference to Ethiopia. Larenstein University, Deventer. Ogato, G.S., Fikadu, A., Tesfaye, G., Abebe, W., 2014. Promoting rural local development:
Harun, R., Chiciudean, G.O., Sirwan, K., Arion, F.H., Muresan, I.C., 2018. Attitudes and the case of Wonchi Ecotourism Society. Prime J. Soc. Sci. 3 (4), 662–673.
perceptions of the local community towards sustainable tourism development in Rechlin, M.A., Tayler, D., Lichatowich, J., Hoon, P., Pandey, Shiva, S., de Leon, B.,
Kurdistan regional government , Iraq. Sustainability 10 (2991), 1–13. Taylor, J., 2008. Community-based Conservation: Is it More Effective, Efficient and
Holladay, P.J., Ormsby, A.A., 2011. A comparative study of local perceptions of Sustainable?.
ecotourism and conservation at Five Blues Lake National Park, Belize. J. Ecotourism Saveriades, A., 2000. Establishing the social tourism carrying capacity for the tourist
10 (2), 118–134. resorts of the east coast of the Republic of Cyprus. Tourism Manag. 21 (2), 147–156.
Honey, M., 2008. Ecotourism and Sustainable Development: Who Owns Paradise?, second SCBD, 2015. Tourism Supporting Biodiversity - A Manual on Applying the CBD
ed. Island Press, Washington. Guidelines on Biodiversity and Tourism Development. Secretariat of the Convention
IUCN, 2012. Guidelines for Application of IUCN Red List Criteria at Regional and National on Biological Diversity, Montreal.
Levels: Version 4.0. Schober, P., Boer, C., Schwarte, L.A., 2018. Correlation coefficients: appropriate use and
Jafari, J., 2001. The scientification of tourism. In: Smith, V.L., Brent, M. (Eds.), Hosts and interpretation. Anesth. Analg. 126 (5), 1763–1768.
Guests Revisted: Tourism Issues of the 21st Century. Cognizant Communication Teye, V., S€
onmez, S.F., Sirakaya, E., 2002. Residents’ attitudes toward tourism
Corporation, New York, pp. 27–41. development. Ann. Tourism Res. 29 (3), 668–688.
Ketema, T.D., 2015a. Development of community based ecotourism in Wenchi Torquebiau, E., Taylor, R.D., 2009. Natural resource management by rural citizens in
crater lake , Ethiopia : challenges and prospects. J. Hospit. Manag. Tourism 6 (4), developing countries: innovations still required. Biodivers. Conserv. 18 (10),
39–46. 2537–2550.
Ketema, T.D., 2015b. Inside the commons of ecotourism development in Ethiopia: UNWTO, 2021. UNWTO Announces List of “Best Tourism Villages” 2021. Available at:
strategic communal empowerment or marginalization? Evidence from Wenchi https://www.unwto.orginewsiunwto-announces-list(Assessed on 02-12, 2021).
community based ecotourism. Int. J. Biodivers. Conserv. 7 (11), 429–434. Vodouh^e, F.G., Coulibaly, O., Adegbidi, A., Sinsin, B., 2010. Community perception of
Kiss, A., 2004. Is community-based ecotourism a good use of biodiversity conservation biodiversity conservation within protected areas in Benin. For. Pol. Econ. 12 (7),
funds? Trends Ecol. Evol. 19 (5), 232–237. 505–512.
Ko, D.W., Stewart, W.P., 2002. A structural equation model of residents’ attitudes for Wang, Y., Pfister, R.E., 2008. Residents’ attitudes toward tourism and perceived personal
tourism development. Tourism Manag. 23 (5), 521–530. benefits in a rural community. J. Trav. Res. 47 (1), 84–93.
Krüger, O., 2005. The role of ecotourism in conservation: panacea or Pandora’s box? Wang, Y., Pfister, R.E., Morais, D.B., 2006. Residents’ attitudes toward tourism
Biodivers. Conserv. 14 (3), 579–600. development: a case study of Wanshington, NC. In: Proceedings of the 2006 Northern
Lankford, S.V., 1994. Attitudes and perceptions toward tourism and rural regional Recreation Research Symposium, pp. 411–418.
development. J. Trav. Res. 32 (3), 35–43. WOCAT (World Overview of Conservation Approaches and Technologies), 2007. In:
Lankford, S.V., Howard, D.R., 1994. Developing a tourism impact attitude scale. Ann. Linger, H., Critchley, W. (Eds.), Where the Land Is Greener: Case Studies and Analysis
Tourism Res. 21 (1), 121–139. of Soil and Water Conservation Initiatives Worldwide. WOCAT, Berne.
Lapeyre, R., 2010. Community-based tourism sustainable solution to maximise WWF, 2001. Guidelines for Community-Based Ecotourism Development. WWF
impacts locally? The Tsiseb Conservancy case , Namibia. Dev. South Afr. 27 (5), International, Gland.
757–822. Yamane, T., 1967. Statistics: An Introductory Analysis. Harper and Row, New York.
Limaei, S.M., Ghesmati, H., Rashidi, R., Yamini, N., 2014. Economic evaluation of natural Zacarias, D.A., Loyola, R., 2017. How ecotourism affects human communities. In:
forest park using the travel cost method (case study; Masouleh forest park, north of Ecotourism’s Promise and Peril. Springer International Publishing, Rio de Janeiro,
Iran). J. For. Sci. 60 (6), 254–261. pp. 133–151.

13