Loading...

Farmers’ Perceptions and Adaptations to Climate Change through Conservation Agriculture

The Case of Guto Gida and Sasiga Districts, Western Ethiopia

Master's Thesis 2013 37 Pages

Agrarian Studies

Excerpt

Table of Contents

i. Acknowledgement

i. List of Acronyms

ii. Table of Contents

iii. Abstract

1. Introduction

2. Materials and Methods

3. Result and Discussion
3.1. Conservation Agriculture as Adaptation Strategy to Climate Change
3.2. Results of the Econometric Model
3.2.1. Farmers’ Perception and Adaptation to Climate Change
3.2.2. Farmers’ Participation in Conservation Agriculture
3.3. Climate Change Adaptation Measures and Causes of Non Adaptation

4. Conclusion

5. References

6. Appendix

iv. List of Tables

1. Summary statistics of continuous variables and their mean difference test used in selection equation for the Heckman two stage selection model -

2. Summary statistics of dummy and categorical variables used in selection equation for the Heckman two stage selection model -

3. Summary statistics of continuous variables and their mean difference test used in outcome equation for the Heckman two stage selection model -

4. Summary statistics of dummy and categorical variables used in outcome equation for the Heckman two stage selection model -

5. Summary statistics of continuous variables and their mean difference test used binary logit model -

6. Summary statistics of dummy and categorical variables used binary logit model

7. Result of Heckman two stage sample selection model

8. Binary logistic regression for conservation agriculture

i. Acknowledgement

Many thanks and appreciation goes to the following institutions and individuals whom without their help and support, the successful completion of my study would not have been possible. I am very grateful to my advisors for their guidance and encouragement to accomplish this work. I am also highly indebted to Ato Abera Gemechu, Socio-economic Researcher at Debre zeit Agricultural Research Center for his support in supplying me with necessary related journals and articles.

ii. List of Acronyms

illustration not visible in this excerpt

v. Abstract

Ethiopia, one of the developing countries, is facing serious natural resource degradation problems. The main objective of this study was to examine the farmer’s perceptions and adaptation to climate change through conservation agriculture. The data used for the study were collected from 142 farm households heads drawn from five kebeles. Primary data and secondary data were used. In addition to descriptive statistics, Heckman two stage sample selection model was employed to examine farmer’s perceptions and adaptations of climate change. Farmers level of education, household nonfarm income, livestock ownership, extension on crop and livestock, households’ credit accessibility, perception of increase in temperature and perception of decrease in precipitation significantly affect the adaptation to climate change. Similarly, farmers’ perception of climate change was affected significantly by information on climate, farmer to farmer extension, local agro -ecology, number of relatives in development group and perception of change in duration of season. A binary logit model was employed for farmers’ participation in conservation agriculture shows education level, number of active family labour and main employment of farmers were significant variables in determining participation in conservation agriculture

Key words: Climate Change, Conservation Agriculture, Heckman and Binary Logit, Western Ethiopia

1. Introduction

Human beings of current world are faced by the depletion of natural resource (Abera, 2003). Agriculture is among the factors affecting the environment in satisfying human needs, while “climate is the primary determinant of agricultural productivity” (Apata et al., 2009).

Ethiopia, one of the developing countries, is “facing serious natural resource degradation problems” (Anemut, 2006). The diversity in altitude accompanied with climatic and ecological variations which affect production is among the features of the country (Shibru & Kifle, 1998). One of Ethiopia's principal natural resources is its rich endowment of agricultural land. Agriculture is the backbone of the Ethiopian economy and is given special attention by the government to spearhead the economic transformation of the country. However, land degradation, especially soil erosion, soil nutrient depletion and soil moisture stress, is a major problem confronting Ethiopia. The proximate causes of land degradation include cultivation of steep slopes and erodible soils, low vegetation cover of the soil, burning of dung and crop residues, declining fallow periods, and limited application of organic or inorganic fertilizers.

Climate is a primary determinant of agricultural productivity. The rate and magnitude of change in climate characteristics determines agronomic and economic impacts from climate change (Bruce et al., 2001). Though climate change is a threat to agriculture and non-agricultural socio-economic development, “agricultural production activities are generally more vulnerable to climate change than other sectors” (Ayanwuyi et al., 2010).

Literature on farmers' perceptions about climate change and participation on conservation agriculture in Ethiopia in general and in the Oromia Region in particular are very few. There are no empirical studies conducted on farmers' perceptions of climate change and their adoption decision on agricultural conservation strategies in Guto Gida and Sasiga districts.

The purpose of this study is therefore, to examine the farmers’ perceptions and adaptation to climate change through conservation agriculture in which the following specific objectives, examine farmers’ perceptions and adaptations to climate change, investigate farmers’ perception towards conservation agriculture as adaptation strategy to climate change and analyze the determinants of farmers’ participation in conservation agriculture, were studied.

2. Materials and Methods

This paper used both primary and secondary data. Primary data was collected by structured questionnaire. Detailed information on household and farm characteristics, household socio-economic and demographic characteristics, location characteristics and farm management practices and other related information were collected through interview of sample household heads.

The study was conducted in Guto Gida and Sasiga districts, East Wollega Zone of Oromia Regional State. These districts were purposefully selected due to the fact that in these areas the environment has been degraded largely and the occurrence of climate change that affect agricultural production during the year 2010 and 2011 in three kebeles of Guto Gida district. Systematic random sampling technique was employed to draw sample of household heads. From a total of 50 peasant associations in these districts nine peasant associations were selected randomly. From these sampled peasant associations based on formula by Kothari (2004) 142 households were selected proportionally.

Two types of econometric models were used for this study. The first model, Heckman Two Stage Selection Model, analyzes farmers’ perception and adaptation to climate change, whereas the second model, Binary Logistic Regression Model, examines the farmers’ participation in conservation agriculture in Guto Gida and Sasiga districts of Oromiya Regional State.

Statistical Package for Social Science (SPSS) version 16.0 and stata version 10.0 were employed for the analysis of this study. Along with the econometric models, descriptive statistics tools were employed to have clear picture of household demographic characteristics, socio-economic and farm characteristics, perception and adaptation of climate change and participation in conservation agriculture. Mean, standard deviation, percentage, t-test, χ2 test, Wald test, correlation matrix and charts were employed to analyze data.

Adaptation to climate change involves a two-stage process: first, perceiving change and, second, deciding whether or not to adapt by taking a particular measure. This leads to a sample selectivity problem, since only those who perceive climate change will adapt, whereas we need to make an inference about adaptation by the agricultural population in general, which implies the use of Heckman’s sample selectivity probit model (Maddison, 2006). The probit model for sample selection assumes that an underlying relationship exists, the latent equation given by

Abbildung in dieser Leseprobe nicht enthalten - (1)

such that we observe only the binary outcome given by the probit model as

Abbildung in dieser Leseprobe nicht enthalten

The dependent variable is observed only if j is observed in the selection equation

Abbildung in dieser Leseprobe nicht enthalten- (3)

Where x is a k -vector of regressors or independent variables that is affect farmers perception and adaptation to climate change, z is an m vector of regressors, u1 and u2 are error terms. When ρ≠0, the standard probit techniques applied to equation (1) yield biased results (Deressa et al., 2008). Thus, the Heckman probit provides consistent, asymptotically efficient estimates for all parameters in such models. Thus, the Heckman two stage selection model was employed to analyze the perception and adaptation to climate change in the Guto Gida and Sasiga districts.

For this study, the first stage of the Heckman probit model considers whether the farmer perceived a climate change; this is the selection model. The second-stage model looks at whether the farmer tried to adapt to climate change, and it is conditional on the first stage, that is, a perceived change in climate. This second stage is the outcome model (Deressa et al., 2008).

There are two dependent variables; farmers’ perception of climate change and farmers’ adaptation to climate change. Farmers’ perception of climate Change (climate_perception) is selection equation and dichotomous in nature and represented in the model 1 for perceived farmer, otherwise 0. Farmers’ adaptation to climate change (climate_adaptation) is outcome equation and dichotomous in nature and explains whether farmers adapted climate change or not. It is valued 1 in the model if farmer adapted climate change, 0 otherwise. The explanatory variables for the selection equation include different socio-demographic and environmental factors based on the literature on factors affecting the awareness of farmers to climate change or their risk perceptions. The explanatory variables of the outcome equation are chosen based on the climate change adaptation literature and data availability. These variables include: education of the head of the household, household size, gender of the head of the household, non-farm income, livestock ownership, extension on crop and livestock production, access to credit, farm size, distance to input and output markets, temperature and precipitation.

A logistic regression analysis was employed to identify the factors that influence farmer’s participation in conservation agriculture as an adaptation to climate change. The farmers’ participation in conservation agriculture is dependent variable which takes a value of 1 if the farmer was participated and 0 if farmer did not participated. The basic model of the logit estimation (Gujarati, 2004) is as follows:

Abbildung in dieser Leseprobe nicht enthalten

Similarly,

Abbildung in dieser Leseprobe nicht enthalten

By dividing (4) by (5) we get

Abbildung in dieser Leseprobe nicht enthalten. (6)

Where Pi is the probability that household participate in conservation agriculture and then (1-Pi) is the probability that household is non participant in conservation agriculture and e is the exponential constant.

The two computing models commonly used in the adoption studies are the probit and logit models. But the results obtained from the two models are very similar since the normal and logistic distributions from which the models are derived are very similar (Gujarati, 2004). As a result, only the logit model will be reported in the paper even if both models will be estimated for the purpose of comparison.

In this analysis before estimating the model, it was necessary to check the existence of multicollinearity among the hypothesized explanatory variables. Multicollinearity problem arises when at least one of the independent variables is a linear combination of the others; with the rest that we have too few independent normal equations and, hence, cannot derive estimators for our entire coefficient. VIF shows how the variance of an estimator is inflated by the presence of multicollinearity (Gujarati, 2004). The speed with which variances and covariances increase can be seen with the variance-inflating factor (VIF)Abbildung in dieser Leseprobe nicht enthalten, which is defined as Abbildung in dieser Leseprobe nicht enthalten where Abbildung in dieser Leseprobe nicht enthaltenAbbildung in dieser Leseprobe nicht enthalten is the coefficient of determination in the regression. The larger the value of VIFj, the more troublesome or collinear the explanatory variables is (Gujarati, 2004).

Farmers’ participation in conservation agriculture (Participation_CA), for logit analysis has a dichotomous nature measuring the willingness of a farmer to participate in conservation agriculture as a measure of adaptation of climate change. The probability of participation in conservation agriculture practices dependent on several household, farm and location characteristics. The independent variables included in this model were age, sex, marital, total family size, level of education, topography of arable land, farming experience, farm size in hectares, extension services and technology promoters, membership in farmer organization, main employment, and active family labor.

3. Result and Discussion

From all sampled respondents 69 were taken from Guto Gida and the left 73 were sampled from Sasiga district. Out of all these 109 respondents perceived the change in climate while the remaining 33 did not perceived the change in climate (Appendix a). Farmers who perceived change in climate have around 8 mean numbers of relatives of household head in development group while it was around 6 for those who did not perceive the change in climate. The maximum number of relatives of respondent household heads who did not perceive climate change was 24 while it was 23 for those who perceived the change (Appendix a). The t-test values indicated that the difference in number of relatives of households in development group between those who did not perceive the change in climate and those who perceived the climate change was significant at 1 percent probability level (Table 1). The average farm income during last production period (2012/13) for the household those who did not perceived the change in climate was 4,126.57 and the mean of farm income of those who perceived the change in climate was 8,909.20. The t-test values indicated that the difference in farm income between those who did not perceive the change in climate and those who perceived the climate change was significant at 1 percent probability level (Table 1).

The mean of nonfarm income during last production period (2012/13) for farmers who did not perceive and who perceived change in climate was 2,930.30 and 4,380.96 respectively. The t-test values indicated that the difference in nonfarm income of households between those who did not perceive the change in climate and those who perceived the climate change was significant at less than10 percent probability level (Table1).

Table 1. Summary statistics of continuous variables and their mean difference test used in selection equation for the Heckman two stage selection model (n=142)

illustration not visible in this excerpt

Source: Own Survey, 2013

The maximum level of education of the respondent household who did not perceive change in climate was those attained grade 9-10 while the maximum level of education for household head who perceived the occurrence of climate change was those with certificate. Out of all households who perceive the occurrence of the climate change 39.45 percent were those who attended grade 1-8 (Appendix a). The χ2 test shows significant difference between households who perceived the climate change to those who did not perceive the change (Table 2).

Change in duration of season was perceived differently among the respondent households in the study area. The χ2 statistic (11.636) and its small significance level (p< .001) indicate that it is very unlikely that these variables are independent of each other. This shows the existence of relationship between a household’s perception of climate change and their perception in change in duration of season (Table 2).

Having information on climate change is one way through which farmers perceive the change in climate. Variability in accessibility of information on climate change between those who did not perceived the change in climate and those who did was the same The χ2 statistic (56.119) and its small significance level (p< .001) indicates existence of relationship between a household’s perception of climate change and their availability of information on climate change (Table 2).

Farmer to farmer extension helps the farmers to share experience and information between them in perceiving environmental problems occurring in their area. The χ2 test shows significant difference between households who received farmer to farmer extension to those who did not take the extension (Table 2).

Table 2 Summary statistics of dummy and categorical variables used in selection equation for the Heckman two stage selection model (n=142)

illustration not visible in this excerpt

Source: Own Survey, 2013

Perceiving climate change is prerequisite for adaptation of climate change. Out of total of 109 respondents who perceived the change in climate was 75 respondents adapted the change through taking adaptation measures while 34 of from 109 respondents did not adapt the change (Appendix a).

The maximum family size for household head those who did not adapt the change in climate was 9 and the minimum family size was 2 (Appendix b). The average family size of those who did not adapt to climate change was around 5 and the family sizes of the household head those who did not adapt the change deviates from its mean by 1.805. However, the maximum family sizes of respondent household those who perceive the change in climate was 16 while the minimum was 3. The standard deviation of family size of those farmers who adapt to climate change was 2.147. This shows that the family size of respondents who did adapt the change in climate deviates larger from its mean than those who did not adapted the change in climate. The t-test values indicated that the difference in family size of households between those who did not adapt the change in climate and those who adapted the climate change was significant at 1 percent probability level (Table 3). The maximum farm size for those farmers who did not adapt the change in climate was 3 hectare while it was 7.25 hectare for those who adapted the change in climate (Appendix b). As the result of the survey shows the mean farm size of respondents who adapted the change in climate was 1.537 hectare which is greater than mean farm size of respondents who did not adapt the change in climate.

The mean of nonfarm income for farmers who did not adapt and who adapted the change in climate was 2,132.740 and 5,400 respectively. The standard deviation of the household nonfarm income for farmers who did not adapt the change in climate was 1,871 and 4,131 for farmers who adapted the change in climate. The t-test values indicated that the difference in nonfarm income of households between those who did not adapt the change in climate and those who adapted the climate change was significant at 1 percent probability level (Table 3). The mean distance from input market for those who did not adapt the change in climate was 12.519 km while it was 15.195 km for those households who adapted the change in climate. The standard deviation of the respondent households distance from input market was 10.36 for those who did not adapt the change in climate and 11.88 for those who adapt the change in climate change. The mean distance from output market for those who did not adapt the change in climate was 10.61 km while it was 15.12 km for those households who adapted the change in climate. The standard deviation of the respondent households distance from output market was 10.16 for those who did not adapt the change in climate and 14.04 for those who adapt the change in climate.

Table 3 Summary statistics of continuous variables and their mean difference test used in outcome equation for the Heckman two stage selection model (n=142)

illustration not visible in this excerpt

Source: Own Survey, 2013

Farmers those who adapted the change in climate were 75 out of which 6.6 percent were female households while those who did not adapt were 34 out of which 17.65 percent were female headed households. The maximum level of education of the respondent household who did not adapt climate change was those households who have certificate while the maximum level of education for household head who adapted the occurrence of climate change was those households who attained grade 11-12. The standard deviation of education level of household adapted change in climate was 0.97 while it was 1.163 for those farmers who did not adapt change in climate. This shows that variability of level of education of households was larger for those who did not adapt the change in climate than those who adapted the change. Out of all households who adapted the occurrence of the climate change 53.33 percent were those who attended grade 1-8 while 58.82 percent (20 out of 34) of all who did not adapt climate change were those who were illiterate (Appendix a). This implies that illiterate households have more probability not to adapt climate change than those with higher level of education.

Extension on crop and livestock is one way through which households exchange information to each other. Out of 109 household heads those perceived climate change 74 of them were those who get extension on crop and livestock. From the total of 109 farmers who did not adapt and adapted climate change 74 were those who received extension on crop and livestock and 35 of them were those who did not receive the extension.

The availability of credit may facilitate the favorable condition to adapt climate. As per the result of household survey reflected 57.04 percent of the total households were those with no availability of credit. The χ2 statistic (11.855) and its small significance level (p< .001) indicate existence of relationship between a household’s who with access to credit and those without the access (Table 4). From total of 69 respondents from Guto Gida 55 households perceived climate change. Out of these who perceive the change in climate 51 of them perceive increase in temperature and three of them respond as there is no change in temperature. From total 54 farmers in Sasiga district who perceive change in climate 42 of them respond as temperature is increasing, 6 of them perceived decrease in temperature and the left 6 farmers responded as there was no change in temperature (Appendix c).

From all the respondents on perception of change in precipitation 41 respondents from Guto Gida and 49 respondents from Sasiga districts were those who perceive decrease in precipitation. 9 respondent from Guto Gida and 2 from Sasiga district were perceived increase in precipitation while the rest 5 from Guto Gida and 3 from Sasiga district were those who did not observe change in precipitation (Appendix c).

[...]


[1] Farmers who did not perceive climate change

[2] Farmers who perceived the change in climate

[3] Farmers who did not perceive climate change

[4] Farmers who perceived the change in climate

[5] Households who did not adapt climate change

[6] Households who adapted climate change

Details

Pages
37
Year
2013
ISBN (eBook)
9783656705314
ISBN (Book)
9783656711605
File size
651 KB
Language
English
Catalog Number
v277938
Grade
Tags
farmers’ perceptions adaptations climate change conservation agriculture case guto gida sasiga districts western ethiopia

Author

Share

Previous

Title: Farmers’ Perceptions and Adaptations to Climate Change through Conservation Agriculture