Advertisement

Environmental Sustainability

, Volume 2, Issue 4, pp 355–368 | Cite as

Determinants of household wetland resources use and management behavior in the Central Rift Valley of Ethiopia

  • Fitsum DechasaEmail author
  • Feyera Senbeta
  • Dawit Diriba
Original Article
  • 524 Downloads

Abstract

Sustainability of wetland resources requires pro-environmental behavior of use and management. This study examins determinants of household wetland resources use and management behavior in the Central Rift Valley of Ethiopia. The study used data generated from 405 randomly selected sample households complemented with that collected through Participatory Rural Appraisals. Multiple linear regression and the Sobel Mediator Test analyses were employed to investigate factors that determine household behavior. The study showed that majorities (94.1%) of the households believe that the wetlands are already degraded; however, only 54.1% of the households have high level of pro-environmental behavior. The result of multiple regression analysis revealed that household wetland use and management behavior is significantly and positively influenced by age, family size, gross annual income, deriving benefit from wetlands, number of livestock owned, farmland size, knowledge about wetlands and their ecosystem services, attitude and participation intention to wetland resources management. Off-farm job participation and distance to wetlands negatively influence wetland use and management behavior. The study result suggests the need to devise strategies to reinforce locals’ pro-environmental behavior of use and management of wetland resources of the study area. Accordingly, measures that enhance and promote knowledge, attitude and participation intention should be targeted to fortify the pro-environmental behavior of locals. Providing incentives in the form of tax holiday, payment for environmental service and creating off-farm livelihood options in their locality could be the most proactive measures to promote locals’ pro-environmental behavior while safeguarding livelihood and easing pressures on wetlands.

Keywords

Wetlands Pro-environmental behavior Determinants Households Ethiopia 

Introduction

Wetland ecosystems are highly valued for their local and global importance having immense potential for ecological and socio-economic sustainability (Nhamo et al. 2017). Worldwide wetlands contribute in diverse ways to the lives of millions of people. In rural communities of developing countries the poor heavily depend on wetlands and floodplain resources for their livelihood and daily sustenance (Tekalegn 2009; Khan 2011). In most contexts, however, unwise use and management of wetlands has often threaten their existence and negatively impact on their role in providing sustainable ecological function, supporting biodiversity and livelihoods of local people (Marti 2011; Nhamo et al. 2017; Ogato 2013). Hence, wise use1 and management of wetlands has become a prime importance since their degradation and loss have profound social and economic impacts (Turner et al. 2004).

Ethiopia is among the most populous countries in Africa, where majority of the population lives in rural areas being dependent on natural resources such as wetlands. To the rural local communities of the country wetlands provide various socio-economic values and services; among others are food, fodder, water (for domestic use, animals and farming), fuel wood, medicine and materials for construction (Amsalu and Addisu 2014). Notwithstanding the varied values and services, the wetlands of the country have been continually degraded and lost (Amsalu and Addisu 2014; Tekalegn 2009; Wondafrash 2003). According to Tekalegn (2009), the main impediment for sustainability of Ethiopian wetlands is the high priority given to the short-term economic benefits than the sustainability issues. The Central Rift Valley (CRV) of Ethiopia is one of the areas with an extensive biodiversity-rich lakes and lacustrine wetlands having multitude of ecosystem services. However, large part of the CRV lakes, associated wetland ecosystems and their biodiversity are highly threatened and degraded, thereby jeopardizing sustainability of their services (Hengsdijk and Jansen 2006; Jansen et al. 2007; Gebretsadik and Mereke 2017; Pascual-Ferrer et al. 2014). Because the CRV is a closed basin, any improper use of resources of the area has far-reaching impacts on ecosystem goods and services of wetlands that potentially undermine sustainability of these resources (Jansen et al. 2007).

The unwise use of wetland resources has brought about worldwide calls for sustainable use and management efforts (Mulugeta 2004; Turner et al. 2004). According to the report of IWMI (2014), however, there is no universal response for the challenges of sustainability in all wetlands. Sustainable management of wetlands rather needs local level responses which call for local communities’ involvement. The ecosystem approaches for wise use and management of wetlands promoted through the Ramsar Convention underscore, among others, the need for effective involvement of local communities in all stages of wetland management (Ramsar Convention Secretariat 2010). Because local communities are the immediate beneficiaries of wetlands, the uses and managements of these resources are usually and ultimately determined by the local communities (Chandool 2007; Zhu et al. 2016).

According to Dixon (2005), in many parts of the developing countries including Ethiopia local communities have used and managed wetlands for generations. However, what often remains ambiguous is why some wetland management systems remain sustainable, whilst others are severely degraded and lost. It is presumed that socio-economic and site-specific environmental factors may give rise to the differences in wetlands use and management behavior of people who use the wetlands. In view of these, and given the existing degradation in the CRV Lakes and associated wetlands, ensuring long-term productivity of wetland ecosystems presupposes study on wetlands use and management behavior of local people and the factors that determine it. This is an important research gap that has been seldom addressed in previous studies. Empirical research on the local peoples’ behavior for the use and management of wetland resources and the driving factors of this behavior give crucial evidence for policy makers to formulate and implement sound and effective policies and strategies for reversing resource degradation (Adem 2017; Zhu et al. 2016). The objective of the present study is therefore to examine wetlands use and management behavior of households (HHs) and the factors that determine them in the CRV of Ethiopia.

Conceptual framework

According to Eilam and Trop (2012) any active responsiveness or participation to preservation or conservation of environmental resources is considered as a pro-environmental behavior (PEB); and so people with PEB show responsible actions that do not disturb the environment. The present study focuses on wetland resources which are increasingly being threatened and their sustainability necessitates PEB of people who are using the resources. It is, however, a complex undertaking to try to fully understand the environmental behavior of people since it can be influenced by several factors (Bantider et al. 2013). Scholars have long sought to understand environmental behavior by developing different frameworks and models. The behavioral model which has support of several empirical works in a range of environmental issues is the Theory of Planned Behavior (TPB), proposed by Icek Ajzen. In the present study conceptualization of factors that are presumed to determine (pro-) environmental behavior is underpinned by TPB. However, the study did not purely confine within this framework. Gifford and Nilsson (2014) noted that it is difficult to fully account, in this model, for variation in PEB; and so the model could be extended to include other personal and social factors. Ajzen (2005) also affirmed that for better understanding of determinants of TPB can be complemented by identifying relevant background factors in the behavior domain of interest.

In this study PEB to use and management of wetland resources is hypothesized to be a function of knowledge about wetlands and their ecosystem services, environmental attitude and participation intention to wetland management- which Blankenberg and Alhusen (2018) refer as ‘psychological factors’. Environmental knowledge, which is an understanding about the environment, is a necessary precursor to PEB. In this regard Adem (2017) and Lawson (2014) asserted that with a better knowledge and understanding about the environment or resources, there will be a better appreciation of their values, thereby developing a more environmentally responsible behavior. The TPB also assumed that “human beings usually behave in a sensible manner that they take account of available information” (Ajzen 2005: 117). Apart from environmental knowledge people’s PEB is also presumed to be shaped by the impact of attitude, which is one’s positive or negative view/evaluation/of performing a particular behavior of interest. According to Ajzen (2005), a behavior is strongest when it is evaluated positively. The TPB, however, recognizes that the influence of knowledge and attitude on specific behavior is mediated (attenuated) by the presence of other more immediate factor: intention towards performing that behavior. Participation intension which refers to the willingness to perform an activity is assumed to capture the motivational factors that influence a behavior. Hence, the TPB postulates that intention to perform a behavior is the most important immediate determinant of that action; the stronger the intention to engage in a behavior, the more likely is the performance (Ajzen 1991, 2005). This posit of the TPB about the mediating effect of intention seems to confound with empirical works (Bagherian et al. 2011; Suwarto 2013) that evidenced strong direct influences of knowledge and attitude on behavior. Given these seemingly conflicting premises the present study tried to examine the direct effect of knowledge and attitude on behavior as well as the mediating effect attributed to intention.

In addition to the measures of environmental knowledge, attitude and intention, other background factors are expected to influence wetlands use and management behavior. The potential importance of such factors is recognized by Ajzen (2005) asserting  that different background factors influence the way people perform a behavior. However, owing to the presence of vast number of potentially relevant factors, it is difficult to know which should be considered. Hence, selection of relevant factors in a study of behavior of interest can be guided by empirical findings. Getacher and Tafere (2013) in their study tried to categorize background factors into three: demographic, biophysical and economic. In the present study the background factors that are assumed to potentially influence behavior can be seen as characteristics (demographic and socio-economic) of individuals or HHs and biophysical/environmental factors (distance from the lake and associated wetlands, location) (Fig. 1).
Fig. 1

Schematic presentation of determinants of wetland use and management behavior;

Source: Elaborated based on Ajzen (2005) and Getacher and Tafere (2013)

Materials and methods

Study area

The CRV is a sub-basin of the Ethiopian Rift Valley that is part of the Great East African Rift Valley System (Jansen et al. 2007). The sub-basin is located between 7°10′ to 8°30′N and 38°15′ to 39°25′E. It covers an area of approximately 10,000 km2 with altitude ranging from 1500 m.a.s.l. in the lowest part to about 4000 m.a.s.l. (Mount kaka) in the eastern side of the valley. Arid and semi-arid climate characterize the sub-basin with mean annual rainfall of 900 mm. The temperature of the area ranges from 11.9 to 29.5 °C in the lower part to 4.3–26 °C in areas above 2500 m (Pascual-Ferrer et al. 2014). The dominant vegetation is Acacia woodlands and savannas. Farming, livestock rearing, fisheries and forestry constitute the main economic activities in the area. The CRV consists of a chain of lakes and rivers which serve as habitat for wide varieties of wildlife and provide water for different economic activities, irrigation being the main water consumer (Jansen et al. 2007; Pascual-Ferrer et al. 2014).

The study was conducted specifically around lakes Ziway and Abijata where rapid decline of the lakes and associated wetlands threaten the persistence of many of these ecosystems with consequent impacts on the riverine and wet grassland habitats (close to the lakes) (Menegesha et al. 2014). Ziway is an open lake which is fed by Ketar and Meki rivers, whereas Lake Abijata receives a major water inflow (on average two-third) from lake Ziway via river Bulbula (Hengsdijk and Jansen 2006; Legesse et al. 2004). The study covers three woredas (districts)—Arsi Negelle, Ziway Dugda and Adami Tulu Jido Kombolcha (ATJK) that border the two lakes (Fig. 2).
Fig. 2

Map of the study area

Sampling procedure

The study population includes the local communities living in the three districts. Because the unit of analysis is HH, we considered HHs in the study districts as the survey population and followed Kothari (2004) to estimate the sample size from the study population. Including 10% as a contingency to compensate for non-response, sample size of 422 was obtained through the calculation. To select the sample HHs for the survey, first a total of 30 kebeles (the smallest administrative units in Ethiopia) that border lakes Ziway and Abijata were identified from the three study districts. Second, six kebeles, three from lake Ziway and three from lake Abijata bordering kebeles were randomly selected as sample study kebeles. Finally, using a probability proportional to size sampling method, 422 HHs were randomly sampled from the sample kebeles (Table 1). Participants of Participatory Rural Appraisals (PRAs) were selected from the sample Kebeles in consultation with development agents (DAs) of the study areas. A total of nine PRAs, each comprising six to ten persons were conducted with different groups of the study community (elderly, youth, women, and fishermen).
Table 1

Distribution of sample HHs by study districts and kebeles

Districts

Lake

Kebeles

Total HHs

Sample HHs

Arsi Negelle

Abijata

Daka Dellu Harangema

585

55

Galle fi Qello

564

53

ATJK

Abijata

Desta Abijata

914

86

Ziway

Abine Germama

904

85

Ziway Dugda

Ziway

Senbero

882

83

Burqa Lemefo

638

60

Total

 

4487

422

Data collection and reliability of the survey

The data for this study was collected using cross-sectional HH survey and PRAs conducted from December 2016 to February 2017. The survey questionnaire included open and close-ended question items that elicit different information including individual and HH characteristics (demographic and socio-economic); knowledge about wetlands and their ecosystem services; attitude, intention and behavior in the use and management of wetlands. Because knowledge, attitude, intention and behavior are latent constructs that hardly have direct and objective measurement, each construct was assessed by multiple question items, each of which measures some more defined part of the construct. The knowledge test constituted question items with response options of “yes”, “no” or “don’t know”; whereas Likert scale question items with five point (1–5) response categories were used to assess attitude and intention, and Likert scale question items with four point (1–4) response categories to assess behavior. Here in, Likert scale question items were used for attitude, intention and behavior, because in each of the items the constructs are reflected/represented/ as points along a continuum whereby individual evaluate a particular issue in a continuum level, in the Likert scale, that runs from extremely pro-issue/favored/ to extremely anti-issue/disfavored/ (Javaras, 2004). In the Likert scale question items of behavior the middle value, ‘neutral’, was excluded since the question items ask something that the respondents have to report on whether or not they have actually practiced/participated in/ the activities mentioned in the items. To make items in the scale of behavior assessment as practical as possible detailed discussions were held with local people and DAs, during reconnaissance survey, regarding indicators of HH’s participation in management of natural resources particularly the lakes and associated wetlands. Accordingly, contribution (money or labor) to resource conservation; adherence to laws, rules, etc., on resource use; refrain from resource damaging activities are among the mentioned indicators on which the question items were based on in the present study.

While the survey predominantly generated the relevant information for this research purpose, some data might not be fully captured. Hence, to fully understand the communities’ behavior in the use and management of wetland resources and the reason(s) behind that behavior, thereby complement and substantiate the survey data qualitative data was collected using different PRA techniques. Accordingly, trend analysis was used to get idea about resource use and management dynamics. Impact diagram, on the other hand, was used to collect information about local communities’ view on wetland resources management scenarios and their possible impact.

Prior to the main data collection the questionnaire was piloted using 30 individuals to check the internal consistency of the items. Based on the analysis of the pilot data we found good internal consistency with respective Cronbach’s alpha coefficients of 0.78, 0.84, 0.78 and 0.74 for the items in the knowledge, attitude, intention and behavior scales. Hence, with minor revision the questionnaire was administered for the main data collection. After the data collection inter-item correlation analysis was made for multi-collinearity check of the items in all the four scales; and all items revealed desirable coefficients less than 0.8. The internal consistency of the individual item to the respective construct in each scale was also assessed through an item-total correlation analysis, using 0.20 as the lowest desirable value (Everitt 2002; Field 2005). All 19 items in the attitude, 18 in the intention and 12 in the behavior scales revealed acceptable item-total correlation values more than 0.20 with respective Cronbach’s alpha of 0.93, 0.92 and 0.90 for the scales. Conversely, two items (K13 and K14) from 19 knowledge test items revealed item-total correlation values of 0.01 and 0.12, which fell below 0.20. Because this shows the inconsistency of the two items with the scale overall, these items were dropped and only 17 were retained for further analyses with the scale’s alpha value of 0.91.

Empirical modeling of pro-environmental behavior (PEB)

Quantitative data analysis was done using descriptive and inferential statistics after the data was edited, cleaned, coded and entered to a computer using SPSS version 24. Although the survey data was collected from 422 HHs, only 405 were used for analysis after excluding 17 questionnaires with missing and inconsistent information. Descriptive statistics: frequency, percentage, total scores (TS), mean score (MS) and standard deviation (SD) were used to describe HH behavior to the use and management of wetlands. The TS for every respondent was calculated by adding score for responses to the 12 questions. Accordingly, the minimum and maximum possible TS are 12 (1 × 12) and 48 (4 × 12) respectively; a higher score indicated a better PEB to the use and management of wetland resources. By adapting measures developed in previous related studies (Adem 2017; Birhanu 2014; Evangelista et al. 2016) using TS we assign low (12 ≤ TS ≤ 24) and high (25 ≤ TS ≤ 48) for HHs’ level of behavior. Multiple linear regression (MLR) analysis was used to examine factors that determine HH behavior in the use and management of wetlands. Because MLR requires the dependent variable to be at an interval or ratio scale (continuous variable), we calculated TS of the Liker scale items in the behavior constructs. Hence, in our MLR model the response variable—HH wetland resources use and management behavior is a continuous variable measured by the TS of Likert scale items in the behavior construct. Explanatory variables which were hypothesized to determine behavior are presented in Table 2. For the variable knowledge, attitude and intention, we also used TS that were calculated for each construct by adding the scores of all items in the respective construct. TS were used because the items under each constructs are many in number hence item by item analysis is impractical. The following model specification was assumed for the HH PEB:
Table 2

Description of variables, measurements and hypothesized relationship

Dependent variable

Description and measurement

Behavior

HH wetland resources use and management behavior (a continuous variable measured by TS of Likert scale items of the behavior construct)

Independent variables

Description and measurement

Type

Hypothesized relationship

Identified relationship in studies

Positive

Negative

AGE

Age of HH head (in years)

Continuous

(+)

Atmis et al. (2007); Badal et al. (2006);Okuma and Muchapondwa (2017); Udo (2014)

Samuel (2017)

SEX

Sex of HH head (male = 1, otherwise = 0)

Dummy

(Indeterminate)

Damena (2012); Getacher and Tafere (2013); Okuthe et al. (2013)

Udo (2014)

EDUC

Education level of HH head (in years spent in school)

Continuous

(+)

La peyre et al. (2011); Okuma and Muchapondwa (2017); Samuel (2017) Van Liere and Dunlap (1980); Zhu et al. (2016)

 

FAMSIZ

Family size (in number)

Continuous

(+)

Getacher and Tafere (2013); Okuma and Muchapondwa (2017); Okuthe et al. (2013); Ranjit (2014); Zidanaet al. (2007)

 

GHI

Estimated gross annual income of HH (in ETBa)

Continuous

(+)

Bagherianet al. (2009); Ranjit (2014)

Zhu et al. (2016)

BENEF

Whether HH derive any benefit from the lake/wetlands(HH derive benefit = 1, otherwise = 0)

Dummy

(+)

Agrawal and Chatre (2006); Getacher and Tafere (2013)

 

OFFJOB

Whether HH engaged in off-farm job (HH has off-farm job = 1, otherwise = 0)

Dummy

(-)

 

Badal et al. (2006); Samuel (2017)

TLUb

Total Livestock owned by HH (in TLU)

Continuous

(+)

Damena (2012); Samuel (2017); Ranjit (2014)

 

FARSIZ

Size farmland owned by HH (in hectare)

Continuous

(+)

Okuthe et al. (2013); Ranjit (2014)

Udo (2014)

TRAIN

Attend training on wetland (attend = 1, otherwise = 0)

Dummy

((+)

Badal et al. (2006); Damena (2012)

 

EXTCON

Contact with extension worker (Have contact = 1, otherwise = 0)

Dummy

(+)

Badal et al., (2006); Damena (2012); Faham et al. (2008); Samuel (2017)

 

DIST

Distance from home to the nearest edge of Lake or associated wetlands (in minutes to walk)

Continuous

(−)

Getacher and Tafere (2013); Ranjit (2014)

 

LOCAT

Bordering Lake (Lake Ziway = 1 Lake Abijata = 0)

Dummy

(Indeterminate)

  

KNOW

Knowledge about wetlands (in TS of knowledge test items)

Continuous

(+)

Ajzen (2005); Bagherian et al. (2009); Suwarto (2013)

 

ATTIT

Attitude to wetland management (in TS of attitude items)

Continuous

(+)

Bagherian et al. (2009); Bagherian et al. (2011); Suwarto (2013)

 

INTEN

Intention to wetland management (in TS of intention items)

Continuous

(+)

Ajzen (1991, 2005)

 

aETB is Ethiopian Birr that is the unit of currency; b Conversion Factor is calculated based on Jahnke (1982), hence Calf = 0.20, Heifer = 0.50, Ox = 0.70, Cow = 0.70, Sheep = 0.10, Goat = 0.10, Camel = 1.00, Horse 0.80, Mule = 0.70, Donkey = 0.50, Chicken = 0.01

$$Y_{i} = \beta_{0} + \beta_{1 } X_{1} + \beta_{2} X_{2} + \cdots + \beta_{k} X_{k} + \varepsilon_{i }$$
(1)
where Yi = the dependent variable (HH PEB) measured by TS in the behavior Likert scale for HH i; β0= the intercept of the model, X1…Xk= explanatory variables included in the model, β1βk= coefficients of the explanatory variables and ɛi = residual term.

On the other hand, to examine indirect effect we used Sobel Mediator Test analysis. The indirect effect represents the relationship between an independent and a dependent variable that is mediated by a mediator variable. So, in the indirect effect, the independent variable (herein refers to knowledge or attitude) is hypothesized to have indirect effect on the dependent variable (behavior) due to the influence of the mediator (participation intention). The Sobel Mediator test provides a method to examine whether the reduction in the effect of the independent variable on the dependent variable, after including the mediator in the model, is a significant reduction; and so whether the mediation effect is statistically significant. Hence, the Sobel test estimation was undertaken to test whether or not participation intention is significant mediator in explaining the relationship of knowledge or attitude with behavior. The Sobel test uses unstandardized coefficients and standard errors obtained from two regression models (Sobel 1982), specified as follow:

Model 1: Simple linear regression with independent variable predicting the mediator:
$$M_{i } = \beta_{0} + \alpha X_{i} + \varepsilon_{i}$$
(2)
where Mi = the mediator variable (intention), measured by TS in the Likert scale of intention, for HH i; β0= the intercept of the model; Xi = the independent variable (either knowledge or attitude), measured by TS for HH i; α = vector of coefficients of the independent variables and ɛi = residual term.
Model 2: MLR with independent variable and mediator predicting the dependent variable:
$$Y_{i} = \beta_{0} + \alpha X_{i} + bM_{i} + \varepsilon_{i}$$
(3)
where Yi is the dependent variable (behavior) measured by TS in the behavior Likert scale for HH i; β0 is the intercept of the model; b is coefficient of the mediator (Mi).
In the Sobel Mediator Test analysis unstandardized coefficient (a) and standard error (Sa) values of the independent variable (knowledge or attitude) from model 1 and the mediator’s unstandardized coefficient (b) and standard error (Sb) values from model 2 (Fig. 3) were used. The statistic was done using SPSS and an online program. The coefficients for the indirect effects represent the change in the TS of participatory behavior for every unit change in the TS of knowledge or attitude that is mediated by intention.
Fig. 3

An illustration of mediation model

Prior to running MLR model the data was checked for basic assumptions. Shapiro–Wilk and Kolmogorov–Smirnov tests were used for the assumption of the normality of the distribution of the scores of the dependent variable (participation behavior). The values of the two tests were not statistically significant (P > 0.05) so the variable participation behavior was assumed as normally distributed. The assumption for Linearity was checked with Normal Probability Plot of the Regression Standardized Residual. No major deviation from normality was found, suggesting the linearity of the relationship between the independent variables and the level of participation behavior. The assumption for the absence of outliers was checked using scatterplot of the Standardized residuals. A very few outliers were detected which fall outside ± 3 SD which did not necessitate further action. The absence of collinearity or multicollinearity among the independent variables was checked using variance inflated factor (VIF). The VIF values for all the variables were far less than 10 (the highest VIF value is for variable age that is 2.126). Hence, multicollinearity was not suspected to be a problem in the regression analysis, thereby all the independent variables were retained in the model for the final analysis.

On the other hand, qualitative data collected through PRAs was recorded into note books. Hence, the notes were first translated from Amharic to English text. To develop general understanding of the data, whole text was reviewed through repeated close reading; and short memos were prepared which best help in forming broader categories of information. All data pertaining to particular theme or category were assembled together to capture similarities or differences in groups’ responses within a given category and summarize information pertaining to each category. Then, through narrative description the entire qualitative data was analyzed, synthesized and meaningful inferences were drawn that were integrated with quantitative data to handle the research problem.

Results and discussion

Descriptive results

Household wetland resources use and management behavior

Local communities’ PEB in the use and management of resources is highly acknowledged since local communities are the foremost beneficiaries (Chandool 2007). In the present study the descriptive analysis revealed that the maximum and minimum TS of HHs in the behavior scale were 46 and 12, respectively, with respective highest and lowest possible TS of 48 and 12. The overall MS of behavior among the survey HHs was 29 with a SD of 5.58. A further examination of HHs’ level of PEB to the use and management of wetland resources revealed that out of the total (405) survey HHs considered in the study, 54.1% have high level of PEB (TS ≥ 25), yet a fairly large proportion (45.9%) of the HHs had low level of PEB (TS ≤ 24). In view of the recognition of the local communities’ great role for sustainability of wetland resources, the proportion of HHs with low PEB suggests the need for measures to encourage and enhance locals’ PEB.

Unlike casual visitors, because the local people live close to wetland ecosystems for longer time they can easily notice changes in the status of wetlands; and so may develop better PEB (Mulugeta 2004). In view of this, one important facet of inquiry of the present study is to assess HHs’ awareness of any change in the status of wetland ecosystems in the study area and to investigate whether their awareness of the changes corresponds to better PEB. Accordingly, respondents were asked if they have noticed any change over the past 10 years. The majority (94.1%) of the HHs considered the deteriorating status of the ecosystems while only 2.7% and 3.2% felt improvement and no change, respectively. Despite the fact that a very high proportion (94.1%) of the sample HHs reported the degradation of the wetland ecosystems, only about 54.1% the HHS have high level of PEB. The study results therefore revealed discrepancy in HHs’ awareness of the deteriorating status of the ecosystems and their level of PEB to the use and management of wetlands. These findings conform to results of Evangelista et al. (2016) that knowledge is not always translated to actual practice/behavior.

Likewise PRA discussants affirmed that despite their claim for degradation of wetland resources, they lack incentive to be prudent in the use and management of the resources. Most participants mentioned their frustration due to the ever-increasing and unlimited access to and use of wetland resources, especially extraction of water from lake Abijata for Soda Ash Plant and from lake Ziway for small- and large-scale irrigation farming. There were also participants who relate the degradation of wetlands to climate change and so they end up doing nothing otherwise solicit divine intercession and wait for better conditions. On the other hand, some participants claimed that it is the responsibility of the government to take measures and design strategies for proper use and management of wetland resources. Similarly, Lawson (2014) indicated that majority of study participants believed that the management of coastal natural resources is the sole responsibility of the government. This is because they believed that the money they paid in the form of tax should be used for resource management and conservation programs. In general, it is uncontested that there is a gap between locals’ realization of the degradation of the wetland ecosystems and their behavioral response to wisely use and manage resources.

Empirical model results

Determinants of household PEB to wetland resources use and management

In studying the PEB of HHs in the use and management of wetland resources, we looked at the determinants of this behavior. Table 3 presents summary of descriptive statistics of the explanatory variables used in the MLR model.
Table 3

Descriptive summary of explanatory variables used in the MLR model

Variables

Mean

SD

Dependent variable

 HH wetland resources use and management behavior (TS in behavior scale)

29.00

9.584

Independent variables

  

 HH head age (years)

39.59

10.59

 HH head sex (male = 1)

0.67

0.469

 HH head education (years in school)

5.70

2.71

 Family size (number)

6.69

2.39

 Gross annual income of HH (ETB)

22,386.8

15,065.58

 Benefit (HH derive benefit = 1)

0.80

0.404

 Off-farm job (HH has off-farm job = 1)

0.53

0.500

 Livestock owned (TLU)

3.68

3.24

 Farmland size (Hectare)

1.78

1.41

 Training on wetland (Attend = 1)

0.26

0.440

 Extension contact (Have contact = 1)

0.28

0.448

 Distance to wetland (minutes to walk)

46.82

38.43

 Location (Lake Ziway area = 1)

0.54

0.499

 Knowledge about wetlands (TS in knowledge test)

10.92

5.21

 Attitude to wetland management (TS in attitude scale)

66.08

14.69

 Intention to wetland management (TS in intention scale)

57.27

14.93

From Table 3 it can be observed that 39.59 is the mean age of the HH heads among the sample population. Out of 405 HHs considered in this study 67.4% are male-headed HHs. The finding for educational status revealed that the highest level of education among the sample population is secondary education and average highest grade attended is 5.7. The mean family size is 6.69, which is above the national average of 4.7 persons per HH (CSA 2008). Based on estimated gross annual HH income (GHI) (generated from farm, livestock and off-farm) the average GHI among the sample population is 22,386.8 ETB, yet the SD suggest the wide disparity in income. Benefit as a dummy variable, herein refers to whether HH derive any benefit from the lake or associated wetlands. In view of that, HH that report deriving benefit accounted for 80% of the total sample. Slightly more than half (53%) of the HHs have off-farm job by which they eke out their living. Livestock is an important component of the livelihood system in the study area. Majority (95.3%) of the study households own livestock of different kind. The average livestock holding among the survey HHs was 3.68 TLU. Of the total HHs who own livestock (386) more than half (51.6%) reported that they usually feed (their livestock) in the lakeside or wetland areas. The average farmland size among 403 HHs is 1.78 hectares which is slightly higher than the national average of 1.17. We also observed  that HHs that attended trainings related to wetlands and have contact with extension worker constitute only about 26% and 28%, respectively. The sample HHs travel on an average 46.82 min from their home to the nearest of the lake or wetland. HHs that accounted for 54% are residents of lake Ziway bordering kebeles, while 46% of the HHs live in lake Abijata bordering kebeles. The average TS of HHs on the knowledge test, attitude and intention Likert scales are 10.92, 66.08 and 57.25 points, respectively.

The result of the MLR model is presented in Table 4. Among the 16 variables controlled in the model, eleven variables (AGE, FAMSIZ, GHI, BENEF, OFFJOB, TLU, FARSIZ, DIST, KNOW, ATTIT and INTEN) were found to be statistically significant predictors of HHs’ behavior in the use and management of wetlands. Looking at the unstandardized Beta (β) coefficients (second column in Table 4), it can be seen that all the significant predictors have the expected effect on the level of participation behavior. On the other hand, the standardized Beta (B) values (third column) show the unique contribution of the independent variables in explaining the dependent variable-HH behavior. Looking at these values, age is the variable with the highest significant unique contribution (B = 0.236), followed by intention (B = 0.150), whereas gross annual HH income (GHI) made the lowest significant unique contribution (B = 0.086). Here under are the findings and discussion on significant variables.
Table 4

Results of MLR analysis for PEB model

Explanatory variables

Β

B

S.E.

T

P-values (two-tailed)

AGE

0.214***

0.236

0.047

4.530

0.000

SEX

− 1.174

− 0.057

0.760

− 1.545

0.124

EDUC

0.220

0.062

0.135

1.633

0.104

FAMSIZ

0.426**

0.107

0.192

2.222

0.027

GHI

0.010**

0.086

0.001

2.089

0.038

BENEF

2.256**

0.095

0.021

2.210

0.028

OFFJOB

− 2.195**

− 0.114

0.044

− 2.602

0.010

TLU

0.288**

0.097

0.138

2.077

0.039

FARSIZ

0.565**

0.089

0.271

2.086

0.038

TRAIN

0.124

0.006

0.892

0.139

0.889

EXTCON

0.392

0.018

0.840

0.467

0.641

DIST

− 0.030***

− 0.122

0.010

− 2.951

0.003

LOCAT

0.041

0.002

0.846

0.049

0.961

KNOW

0.261***

0.142

0.085

3.091

0.002

ATTIT

0.090***

0.138

0.027

3.290

0.001

INTEN

0.102***

0.150

0.026

3.891

0.000

B =standardized Beta, β =Unstandardized Beta; *** significant at 1%; ** significant at 5% level

R = 0.834, R2 = 0.696, Adjusted R2 = 0.675 , F = 34.010, Sig = 0.000

Age of household head

The age hypothesis, in this study, posits that HH’s level of PEB to wetland resources increases with increase in HH head’s age based on the premise that concern about resources increases with age. Moreover, older people have longer life (and farming) experiences that could help them to adopt new techniques of resource management and conservation. As expected, our result shows a significant (P < 0.001) positive effect of age on PEB. This result conforms to the findings of number of studies (Atmis et al. 2007; Badal et al. 2006; Okuma and Muchapondwa 2017; Udo 2014). Conversely, Samuel (2017) found a significant negative effect of age and stated that older farmers are less likely to engage in or adopt sustainable resource management practices that could be due to their short planning horizon as contrast to the younger ones. Our finding suggests the need for efforts that target on younger people of the community to behave pro-environmentally in the use and management of the lakes and associated wetland resources.

Family size

Although it is not always the case, large family size is an indicative of labor availability in a family that could have important role in resource conservation and management. HHs with large family size might not face labor constraint to engage in resource conservation and management activities. The study result revealed a significant (P < 0.05) and positive effect of family size on HH’s PEB. Given that the labor for almost all farm and off-farm activities in the study area is generally provided by the family rather than hired labor, lack of adequate family labor could constraint HH’s PEB to wetland resources. In line with this Okuthe et al. (2013) noted that family labor assumes great importance in resource management particularly in a condition where low income constraints financial liquidity for hiring wage laborers. The present study result concur with the findings of Getacher and Tafere (2013); Okuma and Muchapondwa (2017); Okuthe et al. (2013); Ranjit (2014) and Zidana et al. (2007) that found a significant positive influence of family size on the level of participation in resource management.

Gross annual income of HH (GHI)

Gross income is one of the key variables predicting HH PEB to wetland resources. In agreement with the a priori hypothesis GHI has a significant (P < 0.05) and positive effect on HH behavior. A plausible reason for this finding could be that economically poor people usually have high level of dependence on natural resources that could trigger higher exploitation opposed to prudent use of resources. The other possible reason is that, low income for a HH could constrain the ability of the HH to contribute money for resource conservation and management activities; hence an increase in HH’s income increases participation in resource conservation and management. The present study finding is in tandem with Bagherian, et al. (2009) and Ranjit (2014) who found that income has a significant and positive impact on the level of participation in resource management.

Benefit

It is indicative that people who derive benefit from a given resource are more likely to exhibit high PEB in the use and management of the resource. Herein, Bagherian et al. (2009) noted that people are not expected to reveal PEB toward resource management if they are not benefited out of their participations. In our finding benefit is a significant (P < 0.05) positive predictor of HH PEB. The result revealed that ceteris paribus HH that derive any benefit from wetlands are predicted to score 2.256 more points for the TS of participation behavior than HH that do not derive any benefit. This result lends support to the findings of related studies (Agrawal and Chhatre 2006; Getacher and Tafere 2013). From the findings of PRAs, however, we learned that locals’ effort in the wise use and management of wetland resources depend not only on the mere generation of benefit but also the extent to which the livelihood of the HH depend on these resources. In line with this Agrawal and Chhatre (2006) stated that, subsistence rather than general benefits from resources prompt greater efforts to protect and manage the resources, hence people make greater efforts when they assess the resource to be more useful for subsistence and livelihood.

Participation in off-farm job

PEB of HH is also found to be significantly (P < 0.05) influenced by availability of off-farm job. The negative coefficient implies that, ceteris paribus the TS for the level of participation behavior decreases by 2.195 points for HH that has off-farm job. Hence, HHs’ level of PEB to wetland resources decreases with availability of off-farm income source. The study result conforms to the findings of Badal et al. (2006) and Samuel (2017). The negative influence could possibly be due to the fact that off-farm jobs/activities/ usually demand people to move out from village; and so HHs who are involved in off-farm jobs may encounter time and labor constraints to participate in resource conservation and management activities. In view of this, promoting and creating off-farm livelihood options in their locality is recommended. This will not only enhance /or ensure/ participation, but also safeguard and improve the livelihood of people thereby easing the pressure on wetlands as many HHs could shift their livelihood to off-farm activities that would otherwise (totally) depend on wetlands. On the other hand, HHs’ level of PEB could also be reduced if opportunity costs of the HHs increase due to availability of off-farm activities. Moreover, HHs that have off-farm jobs may feel secure by the income they generate from these sources, thereby less likely to behave pro-environmentally. Herein, Okuma and Muchapondwa (2017) noted that HHs employed in off-farm jobs are less likely to be active in resource management activities due to possibility of exit options from farm work.

Livestock owned in TLU

Livestock owned (measured in TLU) is another variable that was hypothesized to influence the level of PEB to wetland resources. Consistent with the a priori hypothesis TLU has a significant (P < 0.05) positive influence on HH level of PEB; implying that HHs’ level of PEB to wetland resource increases with increase in the number of livestock owned by the HH. This is possibly because with higher number of livestock, the probability of dependence on these resources for livestock grazing and watering is high and so a higher PEB is a requisite for greater and sustainable availability of water and fodder. The finding mirrors the results of the works by Damena (2012); Ranjit (2014) and Samuel (2017) that the number of livestock owned as a significant positive determinant of the level of participation behavior in wise use of resources.

Farm landholding size

The size of total farmland owned by HH is the other significant (P < 0.05) positive predictor of HH PEB to wetland resources. The possible reasons are farmers with large size of farmland holding could expect more benefit from wise use and management of the lakes and associated wetlands because they tend to use water for irrigation and they are more likely to have farmland around the wetland areas of the lakefront. Previous studies revealed mixed results. The findings of Okuthe et al. (2013) and Ranjit (2014) conform to the present study finding while Udo (2014) found a significant negative effect of increase in farmland size with the participation in resource management.

Distance of HH from the lake or wetlands

Using minutes to walk as proxy for distance from home to the nearest edge of the lake or associated wetlands, a negative influence of distance on HH level of PEB to wetland resources was hypothesized. The study result revealed a significant negative coefficient (P < 0.001) implying that the farther the HH’s home is from the nearest edge of the lake or associated wetlands, the lower the level of PEB. This is most likely because a distant HH might cost more in terms of time and energy to travel and participate in resource conservation and management activities. The other plausible reason could be that it is less likely for a distant HH to generate more benefit from these resources and so they could have less incentive for PEB. The study result concurs with earlier findings (Getacher and Tafere 2013; Okuma and Muchapondwa 2017; Ranjit 2014).

Knowledge of wetlands and their ecosystem services

On a premise that knowledge of an issue is an important reason to act in a particular way; we hypothesized that HH PEB to wetlands is influenced by the level of knowledge of wetlands and their ecosystem services. The study finding revealed a significant (P < 0.001) positive influence suggesting that better knowledge of wetlands and their ecosystem services enhances PEB to wetlands. The finding is consistent with theories and empirical works. It corroborates with TPB which posit: human being usually behave by considering available knowledge and information (Ajzen 2005). In the linear models of PEB environmental knowledge is thought to lead to positive environmental behavior (Lawson 2014). The finding is also in tandem with works of Bagherian et al. (2009) and Suwarto (2013) that found individuals with greater knowledge of environmental issues are more likely to engage in responsible environmental behaviors.

Attitude to wetlands and their management

Attitude to wetlands and their management is the other variable that significantly (P < 0.001) and positively influence HH level of PEB to wetland resources. This result is consistent with the premise of TPB that with a more positive evaluation the strongest will be the behavior (Ajzen 2005). The finding is also in tandem with the works of Bagherian et al. (2009; 2011) and Suwarto (2013). Our study findings from PRAs however, revealed a conflicting result that the knowledge and attitude is not in unison with PEB. Herein, Kollmuss and Agyeman (2002) noted that people’s behavior is sometimes in conflict with their professed knowledge and attitudes. In the present study PRA discussants of fisher men groups, for example, stated that despite their knowhow about the detrimental effect of using monofilament fishnet (narrow mesh size) and despite its use being penal, people still use it. Some participants, particularly residents of lake Ziway area, have also stated that although they perceive the negative impact of large scale irrigation farming on the lake and associated wetlands, they rent out their farmland found near the lakeshore area for large scale irrigation farming to generate income for their living. There were also participants who claim farming in the lake retreat areas because these are more productive and need less effort to irrigate. Moreover, most PRA discussants affirmed that rather than using the ‘cut and carry’ method of feeding their animals, they usually prefer to freely graze in the wetland areas. Although the locals know and believe that such practices are threatening for sustainability of wetlands, they still perform these practices. According to their argument this is because of community’s urge to meet livelihood. Hence, this could be an important area of policy interventions by which alternative livelihood options need be designed.

Intention to participate in wetland resources management
Participation intention in management of wetland resources was hypothesized to positively influence HH level of PEB to wetland resources because with higher level of willingness and motivation to perform an activity the more likely the chances of its success. Verifying the a priori hypothesis, the study results revealed a significant (P < 0.001) positive influence. This finding corroborates the assertion of the TPB that stronger the intention to engage in a behavior, the higher is the performance because people usually behave in accordance with their intentions (Ajzen 1991, 2005). TPB also theorized that intention is the immediate antecedent of an overt behavior, hence the influence of both knowledge and attitude on a given behavior is mediated by intention, albeit empirical evidences on the predictive ability, in a linear relationship, of knowledge and attitude on PEB. The mediation effects of intention in the relationship between knowledge and PEB as well as between attitude and PEB were examined by the Sobel Mediator Test analyses (Tables 5, 6).
Table 5

Mediation effect of intention in the relationship between knowledge and behavior

 

Test statistics

Std. error

p value

Sobel test

4.68916563

0.03533464

0.00000274

Aroian test

4.66551965

0.03551373

0.00000308

Goodman test

4.71317482

0.03515465

0.00000244

Table 6

Mediation effect of intention in the relationship between attitude and behavior

 

Test statistics

Sed. error

p value

Sobel test

4.46682902

0.01393177

0.00000495

Aroian test

4.54484356

0.01399916

0.0000055

Goodman test

4.58913666

0.01386405

0.00000445

The result of the Sobel Mediator Test revealed statistically significant results for both knowledge (P < 0.001) and attitude (P < 0.001) suggesting the presence of mediation effect. These imply that, HH that have high level of knowledge about wetlands and their ecosystem services or a more favorable attitude to management of wetlands also possess high level of PEB provided that they have high participation intention. This result validates the mediating effects of intention that is posited by the TPB (Ajzen 2005). Hence, it is highly recommended to design measures that enhance and motivate the study community to participate in wise use and management of wetlands.

Conclusions and recommendations

Wetlands, in different parts of the world including Ethiopia, are under continued degradation owing to high demand and unsustainable exploitation of these resources. Long-term productivity and sustainability of wetlands therefore require wise use and management of the ecosystems. In particular, local communities are the foremost users of wetland resources; and hence they should be at the forefront in sustainability and management efforts of wetland ecosystems. This study examined HHs wetland resources use and management behavior and the factors that determine this behavior in the CRV of Ethiopia. The study provides empirical evidences that can help resource planners and managers to design sound policy and strategies that can address the existing challenges of wetlands’ sustainability, thereby contribute to sustainable development outcomes.

It is apparent that majorities of the study HHs felt the deteriorating status of the ecosystem of the lakes and associated wetlands, yet a fairly large proportion of the HHs revealed low level of PEB to wetland resources implying a lag in PEB. A mere awareness and understanding of the degradation of the wetland ecosystems is futile in the absence of a PEB; and hence a more pro-environmental action is required. The results of the study have further showed that knowledge, attitude and participation intention have significant positive influence on PEB. However, it has also been evidenced that HH participation intention is a significant mediator in explaining the relationship of both knowledge and attitude with PEB. This finding implies that the influence of knowledge or attitude on PEB of a HH is influenced by the level of participation intention to wise use and management of wetland resources. Accordingly, the most proactive and effective strategies need to target on awareness creation and environmental education programs that promote locals’ knowledge, attitude and intention thereby strengthen their PEB in the use and management of wetlands.

A more PEB in the use and management of wetlands is evidenced with an increase in the age of HH head. Hence, importantly, policy measures for sustainability of wetland resources need to target younger group of the community to develop responsibility thereby act pro-environmentally in the use and management of the resources. This may be through the use of different incentive mechanisms that promote PEB of the youth, such as tax subsidy or tax holiday and payment for environmental services for an individual and collective pro-environmental action. PEB of HH is also induced by benefit gain from wetlands. However, sustainable provision of benefits from wetlands is possible only if wetland resources are wisely used and managed. Hence, local peoples’ pro-environmental action in the use and management of wetlands is paramount to enhance benefits. In view of these, integrated efforts are required from governmental and non-governmental organizations, as well as academic and private institutions to foster locals’ PEB thereby increase benefit from wetlands. These may be through technical support or devising programs that sensitize local community for a more pro-environmental action. In general, planners and policy makers should formulate robust institutional arrangements for better PEB of the locals in the use and management of wetland resources. This is because, the existing lag in wetland policy in Ethiopia; absence of responsible institution and clearly defined tenure right over wetlands may partly deter the local community to invest on conservation and management. In view of this it is exigent to devise a wetland policy and legal frameworks that gear towards creating management system ensuring wise and lawful use of resources and people are incentivized for wetland conservation, whereas penalized for any practice that threatens the sustainability of these very important ecosystems.

Footnotes

  1. 1.

    Wise use and management of wetlands is their sustainable use for the benefit of human being in a way compatible with the maintenance of their natural properties (Ramsar Convention Secretariat 2010).

Notes

Acknowledgements

The authors are very grateful for the study community; the sample respondents; agricultural development agents and local administrators of the study area for their willingness and assistances during the field work and data collection. The authors also wish to warmly thank anonymous reviewers whose constructive comments immensely contributed to improvement of this article; remaining errors are ours. We would also like to thank Dr. Solomon Mekonnen, from Wolayita Sodo University, for kindly editing the language of this article.

Funding

Bahir Dar University [2015/16] and the thematic research project of Addis Ababa University [2016–2018] are greatly acknowledged for financial support of this research.

Compliance with ethical standards

Conflict of interest

On behalf of all authors, the corresponding author states that there is no conflict of interest.

References

  1. Adem MS (2017) Environmental Knowledge, Attitude and Awareness of Farmers in Chencha Woreda, GamoGofa Zone, South Ethiopia. Int J Sci Res Publ 7(1):69–76Google Scholar
  2. Agrawal A, Chhatre A (2006) Explaining success on the commons: community forest governance in the Indian Himalaya. World Dev 34(1):149–166CrossRefGoogle Scholar
  3. Ajzen I (1991) The theory of planned behavior. Organ Behav Hum Decis Process 50(2):179–211CrossRefGoogle Scholar
  4. Ajzen I (2005) Attitudes, personality and behavior, 2ndedn. Open University Press, LondonGoogle Scholar
  5. Amsalu T, Addisu S (2014) A Review of wetland conservation and Management Policy in Ethiopia. Int J Sci Res Publ 4(9):1–6Google Scholar
  6. Atmis E, Daademir I, Lise W, Yldran O (2007) Factors affecting women’s participation in forestry in Turkey. Ecol Econ 60(4):787–796CrossRefGoogle Scholar
  7. Badal PS, Kumar P, Bisaria G (2006) Dimensions and determinants of peoples’ participation in watershed development programmes in Rajasthan. Agric Econ Res Rev 19:57–69Google Scholar
  8. Bagherian R, Bahaman AS, Asnarulkhadi AS, Shamsuddin A (2009) Factors influencing local people’s participation in watershed management programs in Iran, American-Eurasian. J Agric Environ Sci 6(5):532–538Google Scholar
  9. Bagherian R, Goodarzi M, Shadfar S (2011) Relationship between attitude toward watershed management programs and level of participation. Middle-East J Sci Res 9(3):324–329Google Scholar
  10. Bantider A, Mohammed Y, Baudouin A (2013) Environmental knowledge and behavior among rural youths in Southern Ethiopia: Cases from Gedeo Zone. Norwegian University of Science and Technology. ActaGeographica-Trondheim, Theses and reports, Series A, No. 27Google Scholar
  11. Birhanu A (2014) Environmental knowledge, attitude and participatory behavior towards land degradation in Injibara Secondary and Preparatory School. Northwestern Ethiopia. J Environ Earth Sci 4(17):89–95 (ISSN 2224-3216 (Paper) ISSN 2225-0948) Google Scholar
  12. Blankenberg A, Alhusen H (2018) On the determinants of pro-environmental behavior—a guide for further investigations. University of Goettingen, Germany, Discussion paper 350. Center for European, Governance and Economic Development Research. ISSN: 1439-2305Google Scholar
  13. Chandool C (2007) Household’s resource use, conservation attitude and perceiver socio-economic impact of the protected areas policy in the Nariva swamp, Trinidad and Tobago. Dissertation, University of FloridaGoogle Scholar
  14. CSA (Central Statistical Agency) (2008) Summary and statistical report of the 2007 population and housing census. Federal Democratic Republic of Ethiopia Population Census Commission, Addis AbabaGoogle Scholar
  15. Damena D (2012) Determinants of farmers’ land management practices: the case of Tole District, South West Sgewa, Oromia Natonal Regional State, Ethiopia. J Sustain Dev Africa 14(1):76–96Google Scholar
  16. Dixon AB (2005) Wetland sustainability and the evolution of indigenous knowledge in Ethiopia. Geogr J 171(4):306–323CrossRefGoogle Scholar
  17. Eilam E, Trop T (2012) Environmental attitudes and environmental behavior—which is the horse and which is the cart? Sustainability 4:2210–2246.  https://doi.org/10.3390/su4092210(ISSN 2071-1050) CrossRefGoogle Scholar
  18. Evangelista RJP, Cruz AC, Lasco RD (2016) Knowledge, attitude and practice of smallholder farmers regarding climate change and agroforestry: a case study of farmers in Peñablanca, Cagayan. J Environ Sci Manag (Special Issue 1): 42–55 (ISSN 0119-1144)  Google Scholar
  19. Everitt BS (2002) The Cambridge dictionary of statistics. 2nd edn. Cambridge University Press, CambridgeGoogle Scholar
  20. Faham E, Hosseini SM, Darvish AK (2008) Analysis of factors influencing rural people’s participation in national action plan for sustainable management of land and water resources in Hable-Rud Basin, Iran. Am J Agric Biol Sci 3(2):457–461 (ISSN 1557-4989) CrossRefGoogle Scholar
  21. Field A (2005) Discovering statistics using SPSS, 2nd edn. Sage, LondonGoogle Scholar
  22. Gebretsadik T, Mereke K (2017) Threats and opportunities to major Rift Valley Lakes wetlands of Ethiopia. Agric Res Tech Open Access J 9(1):001–006.  https://doi.org/10.19080/ARTOAJ.2017.09.555751 CrossRefGoogle Scholar
  23. Getacher T, Tafere A (2013) Explaining the determinants of community based forest management: evidence from Alamata, Ethiopia. Int J Commun Dev 1(2):63–70.  https://doi.org/10.11634/233028791301431 CrossRefGoogle Scholar
  24. Gifford R, Nilsson A (2014) Personal and social factors that influence pro-environmental concern and behavior: a review. Int J Psychol.  https://doi.org/10.1002/ijop.12034 CrossRefGoogle Scholar
  25. Hengsdijk H, Jansen H (2006) Agricultural development in the Central Ethiopian Rift Valley: a desk-study on water-related issues and knowledge to support a policy dialogue. Plant Research International B.V, Wageningen URGoogle Scholar
  26. IWMI (International Water Management Institute) (2014) Wetlands and people. Colombo, Sri Lanka.  https://doi.org/10.5337/2014.202
  27. Jahnke HE (1982) Livestock production systems and livestock development in tropical Africa (vol. 35), Kieler Wissenschaftsverlag Vauk,  KielGoogle Scholar
  28. Jansen HC, Hengsdijk H, Legesse D, Ayenew T, Hellegers P, Spliethoff PC (2007) Land and water resources assessment in the Ethiopian Central Rift Valley: Project: Ecosystems for water, food and economic development in the Ethiopian Central Rift Valley (No. 1587), Alterra (ISSN 1566-7197)Google Scholar
  29. Javaras KN (2004) Statistical analysis of Likert data on attitudes. Dissertation, University of OxfordGoogle Scholar
  30. Khan MH (2011) Participatory wetland resource governance in Bangladesh: an analysis of community-based experiments in Hakaluki Haor. Dissertation, Faculty of Environment, Earth and Resources; Natural Resources Institute, University of Manitoba, CanadaGoogle Scholar
  31. Kollmuss A, Agyeman J (2002) Mind the gap: why do people act environmentally and what are the barriers to pro-environmental behavior? Environ Educ Res 8(3):239–260CrossRefGoogle Scholar
  32. Kothari CR (2004) Research methodology: methods and techniques. New Age International Publisher, New DelhiGoogle Scholar
  33. La Peyre MK, Reams MA, Grace JB (2011) Identifying determinants of nations’ wetland management programs using structural equation modeling: an exploratory analysis. Environ Manag 27:859–868CrossRefGoogle Scholar
  34. Lawson ET (2014) When rhetoric meets reality: attitude change and coastal zone management in Ghana. Environ Nat Resour Res 4(4):37–50.  https://doi.org/10.5539/enrr.v4n4p37(ISSN 1927-0488 E-ISSN 1927-0496) CrossRefGoogle Scholar
  35. Legesse D, Vallet-Coulomb C, Gasse F (2004) Analysis of the hydrological response of a tropical terminal lake, Lake Abijata (Main Ethiopian Rift Valley) to changes in climate and human activities. J Hydrol Process 18:487–504CrossRefGoogle Scholar
  36. Marti AM (2011) Wetlands: a review. With three case studies: The People’s Republic of China, The United States of America, and Ethiopia, International Resource ManagementGoogle Scholar
  37. Mengesha G, Bekele A, Fraser G, Mamo Y (2014) Land use, land cover and climate change impacts on the bird community in and around Lake Zeway, Ethiopia. Int J Biodiv Conserv 6(3):256–270CrossRefGoogle Scholar
  38. Mulugeta S (2004) Socio-economic determinants of wetland cultivation in Kemise, Illubabor Zone, South Western Ethiopia. Eastern Africa Soc Sci Res Rev 20(1):93–114CrossRefGoogle Scholar
  39. Nhamo L, Magidi J, Dickens C (2017) Determining wetland spatial extent and seasonal variations of the inundated area using multispectral remote sensing. Water SA 43(4):1–10CrossRefGoogle Scholar
  40. Ogato GS (2013) The human ecology of wetlands in Least Developed Countries in time of climate change: policy and strategy implications for wise use and conservation of wetlands. Am J Hum Ecol 2(4):127–138.  https://doi.org/10.11634/216796221302476 CrossRefGoogle Scholar
  41. Okuma B, Muchapondwa E (2017) Determinants of successful collective management of forest resources: Evidence from Kenyan community forest associations. Working Paper 698.Ersa Economic Research Southern AfricaGoogle Scholar
  42. Okuthe IK, Kioli F, Abuom P (2013) Socio cultural determinants of the adoption of integrated natural resource management technologies by small scale farmers in Ndhiwa Division, Kenya. Curr Res J Soc Sci 5(6):203–218 (ISSN: 2041-3238) Google Scholar
  43. Pascual-Ferrer J, Pérez-Foguet A, Codony J, Raventós E, Candela L (2014) Assessment of water resources management in the Ethiopian Central Rift Valley: environmental conservation and poverty reduction. Int J Water Resour Dev 30(3):572–587.  https://doi.org/10.1080/07900627.2013.843410 CrossRefGoogle Scholar
  44. Ramsar Convention Secretariat (2010) National wetland policies: Developing and implementing National wetland policies. Ramsar handbooks for the wise use of wetlands, 4th edn., vol. 2. Ramsar Convention Secretariat, GlandGoogle Scholar
  45. Ranjit Y (2014) Determinants of people’s participation in forest protection and management: a study in Kaski, Nepal. Econ J Dev Issues 17 & 18:175–186Google Scholar
  46. Samuel T (2017) Determinants of adoption of sustainable land management (SLM) practices among smallholder farmers’ in Jeldu District, West Shewa Zone, Oromia Region, Ethiopia. Glob J Sci Front Res 7:60–76 (ISSN:2249-4626) Google Scholar
  47. Sobel ME (1982) Asymptotic confidence intervals for indirect effects in structural equation models. Sociol Methodol 13:290–312CrossRefGoogle Scholar
  48. Suwarto WA (2013) The Effect of environmental knowledge and attitude towards pro-environmental behavior with social economic status as moderation in peasant community in Banjarsari Regency, Surakarta. J Educ Pract 17:179–189Google Scholar
  49. Tekalegn B (2009) Wetland values, functions and challenges for sustainable use. In: Proceeding of national advocacy workshop on wetlands and climate change, compiled by, G/Meskel L and Geda B Dec 2–3, 2009Google Scholar
  50. Turner RK, Bateman IJ, Georgiou S, Jones A, Langford IH, Matias NG, Subramanian L (2004) An ecological economics approach to the management of a multi-purpose coastal wetland. Reg Environ Change 4:86–99.  https://doi.org/10.1007/s10113-004-0075-x CrossRefGoogle Scholar
  51. Udo NC (2014) Socio-Economic factors influencing farmers’ participation in community-based programme in Abia and Cross River States of Nigeria. J Agric Ext 18(1):48–61 (ISSN 1119-944X) Google Scholar
  52. Van Liere KD, Dunlap RE (1980) The social bases of environmental concern: a review of hypotheses, explanations, and empirical evidence. Public Opin Quart 44(2):43–59CrossRefGoogle Scholar
  53. Wondafrash M (2003) Wetlands, birds and important bird areas in Ethiopia. In: Abebe YD, Geheb K (eds) Proceedings of a seminar on the resources and status of Ethiopia’s Wetlands, IUCN, pp 1–11Google Scholar
  54. Zhu H, Guan Wei X (2016) Factors influencing farmers’ willingness to participate in wetland restoration: evidence from China. Sustainability.  https://doi.org/10.3390/su8121325 CrossRefGoogle Scholar
  55. Zidan A, Kaunda E, Phiri A, Khalil-Edriss A, Matiya G, Jamu D (2007) Factors influencing cultivation of the Lilongwe and Linthipe river banks in Malawi: case study of Salima district. J Appl Sci 7(21):3334–3337.  https://doi.org/10.3923/jas.2007.3334.3337 CrossRefGoogle Scholar

Copyright information

© Society for Environmental Sustainability 2019

Authors and Affiliations

  1. 1.Center for Environment and Development StudiesAddis Ababa UniversityAddis AbabaEthiopia

Personalised recommendations