DEBARK UNIVERSITY

COLLEGE OF BUSINESS AND ECONOMICS

DEPARTMENT OF ECONOMICS

DETERMINANTS OF FACTORS AFFECTING OFF-FARM PARTICIPATION DECISIONS

OF FARM HOUSEHOLDS (A CASE STUDY OF DEBARK WOREDA)

A RESEARCH THESIS SUBMITTED TO THE DEPARTMENT OF ECONOMICS OF

DEBARK UNIVERSITY IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR
THE BACHELOR OF ARTS (BA) DEGREE IN ECONOMICS

PREPARED BY;
1 TEREFE BERGENE 1301459
2 ATKILT BELACHEW 1301012
3 BINIANM ASSEFA 1302269

ADVISOR: Mr. SHEGAW G. (MSc)

March, 2025
DEBARK, ETHIOPIA
 DECLARATION
We have carried out research work on determinant of factors that affect off farm participation
decision of farm households (in the case of Debark woreda) for partial fulfillment of the
requirement of the Bachelor of Art ―(BA)‖ degree in Economics with the guidance and
supervision of the research advisor, Mr. SHEGAW G. (MSc)

1. Name Of Students:

1 Terefe Bergene: Date ____________________ Signature____________________

2 Atkilt Belachew: Date ____________________ Signature____________________

3 Biniyanm Assefa: Date____________________ Signature____________________

2. ADVISOR: Mr. SHEGAW G. (MSc)

Date______________________Signature____________________

3. EXAMINER:

Name of Examiner___________________

Date____________Signature_____________

I
 ACKNOWLEDGMENT
First and foremost, we give all credit and praise to our Lord Jesus Christ, whose grace, strength,
and guidance have been the backbone of this research. We are profoundly thankful for His
presence in our lives. We also extend our heartfelt appreciation to our advisor, Mr. Shegaw G. ,
for his essential guidance, feedback, and unwavering support. His expertise and willingness to
share knowledge have been crucial. Our sincere gratitude goes out to our families and friends for
their ongoing encouragement and emotional support. Finally, we recognize all who contributed
to this research, whether through assistance, resources, or motivation. Your support has been
invaluable, and we sincerely appreciate each of you.

II
 ACRONYM

GDP Gross Domestic Product

MOARD Ministry of Agriculture and Rural Development
IFAD International Fund for Agricultural Development
FAO Food and Agriculture Organization
WFP World Food Programme
NGO Non-Governmental Organization
GTP Growth and Transformation Plan
WHS World Health Survey
SSA Sub-Saharan Africa
USA United States of America
UNESCO United Nations Educational, Scientific and Cultural Organization
VIF Variance Inflation Factor
LR Likelihood Ratio
LPM Linear Probability Model
EC Ethiopian Calendar
NLI Non-Labor Income
OFP Off-Farm Participation

III
 Table of content
DECLARATION ................................................................................................................................................ I
ACKNOWLEDGMENT ..................................................................................................................................... II
ACRONYM ................................................................................................................................................. III
List of Tables ............................................................................................................................................... VI
List of Figures .............................................................................................................................................. VII
Abstract ...................................................................................................................................................... VIII
CHAPTER ONE ............................................................................................................................................... 1
INTRODUCTION ............................................................................................................................................. 1
1. 1 Background of the study .................................................................................................................... 1
1.4 Research question ............................................................................................................................... 4
1.8 Organization of the study .................................................................................................................... 5
Chapter Two .................................................................................................................................................. 6
Review Literature .......................................................................................................................................... 6
2.1 Theoretical Literature Review ............................................................................................................. 6
2.1.1 Introduction to Theory of Off-Farm ............................................................................................. 6
Determinants of Off-Farm Participation ............................................................................................... 8
2.3 Conceptual Framework ..................................................................................................................... 10
Chapter Three ............................................................................................................................................. 12
Research Methodology ............................................................................................................................... 12
3.1 Description of Study Area ................................................................................................................. 12
3.1.1 Geography and Climate ............................................................................................................. 12
3.1.2 Demographic Characteristics .................................................................................................... 12
3.1.3 Agricultural Practices and Economy .......................................................................................... 13
3.1.4 Off-Farm Activities and Economic Diversification...................................................................... 13
3.2 Research Design ................................................................................................................................ 14
3.3 Sources of Data ................................................................................................................................. 14
3.4 Sampling Technique and Sample Size ............................................................................................... 14
3.5 Methods of Data Collection .............................................................................................................. 15
3.5.1 Primary Data .............................................................................................................................. 15

IV
 3.5.2 Secondary Sources of Data ........................................................................................................ 15
3.6 Method of Data Analysis ................................................................................................................... 15
3.7 Model Specification .......................................................................................................................... 16
3.7.1 Specification of the Logit Model ................................................................................................ 17
CHAPTER FOUR ........................................................................................................................................... 20
DATA ANALYSIS AND PRESENTATION ......................................................................................................... 20
4.1. Descriptive analysis .......................................................................................................................... 20
4.2 Econometric analysis............................................................................................................................. 25
4.2.1 Logit regression analysis of off-farm participation of respondents ........................................... 25
4.2.2 Odds ratio result and its interpretation ....................................................................................... 26
4.2.3 Marginal effect and its interpretation ....................................................................................... 27
CHAPTER FIVE CONCLUSION AND RECOMMENDATIONS ........................................................................... 31
5.1 CONCLUSION ..................................................................................................................................... 31
REFERENCE.............................................................................................................................................. 32
APPENDEX1.............................................................................................................................................. 34
APPENDIX2 ............................................................................................................................................... 38
APPENDEX1.............................................................................................................................................. 40

V
List of Tables Page
Table 4.1 Off farm participation……………………………………………………………20

Table 4.2 Gender description ………………………………………………..…………….21

Table 4.3 Age structure and off farm participation……………………………..…………..21

Table 4.4 Description of farm land holding structure……………………………..…..……22

Table 4.5 Description of the education level………………………………………………..22

Table 4.6 Non labor income and participation………………………………….…...………23

Table 4.7 Reason to participate………………………………………………….…..….…..24

Table 4.8 reasons for to be non- participant………………………………………..……….24

VI
List of Figures page

Figure 2.1 Conceptual frame work that shows dependent and

independent variables……………………………………………….11

Figure 3.1 Location map of the study area………………………………...……13

VII
 Abstract
The research examined the effect of various factors influencing household involvement in off-
farm activities among rural families in the Debark Woreda of the Amhara Region. The primary
information was gathered using a structured questionnaire and interviews with key informants.
Both descriptive statistics and an econometric model (Logit model) were employed to assess this
data at the household level. This model addresses selection bias arising from unobserved factors
that could potentially influence a household‘s involvement. Among the male participants, 126
out of 154 are engaged in off-farm activity, representing 62. 38% of the entire sample and 81.
82% of total participants.
Additionally, the findings from the model indicated that various key factors, such as the area of
cultivated land, the age of the head of the household, the educational attainment of the household
head, and the number of dependents, significantly affected off-farm participation. Specifically,
families with larger tracts of land and higher educational levels among heads of households were
more inclined to take part in off-farm activities. In contrast, families with a greater number of
dependents were less inclined to engage in off-farm work.
Ultimately, the research recommended that initiatives aimed at encouraging off-farm
involvement could be vital in enhancing the livelihood conditions of families in the study region.
By improving educational access, backing agricultural progress, and boosting opportunities for
off-farm income generation, these initiatives could bolster the rural economy and alleviate
poverty. The study underscored that fostering off-farm activities could result in increased
productivity, enhanced economic resilience, and stronger structural transformation within the
local economy, ultimately benefiting families in the Debark Woreda.

VIII
 CHAPTER ONE

INTRODUCTION
1. 1 Background of the study
Agriculture represents the primary economic sector in Ethiopia. Over 85% of the workforce is
involved in agriculture, primarily in crop production, animal rearing, and mixed farming
(Ministry of Finance, 2012). This agricultural sector is the main contributor to the economy,
providing about 41% of the GDP and accounting for 80% of exports and 85% of the employment
in the nation (moard, 2019). However, most of the rural population relies on marginal on-farm
economic activities. The agricultural sector is marked by shrinking farm sizes as the population
increases (ifad, 2011), low output per farm, and a significant prevalence of subsistence farming
(moard, 2010).
In Ethiopia, farming is characterized by traditional methods, primarily depending on animal
traction and rain-fed agriculture. The nation experiences fluctuations in agricultural production
and productivity due to weather or human-induced shocks, resulting in insufficient income for
rural households (beyene, 2008). Additionally, the reduction in agricultural output caused by
drought, inconsistent rainfall, inadequate production technology, and soil erosion significantly
contributes to unemployment, underemployment, poverty, and food insecurity (World Ahmad,
2010; Asnake, 2010; World Bank, 2008; Fentie and Sandra, 2016).
Consequently, farm households engaging in rural off-farm activities play a crucial role in
creating employment, generating income, and enhancing agricultural income activities. These
off-farm activities engaged in both self-employment and wage employment on milling, weaving,
handicraft, trade in grain and livestock, general trade, income from share cropped out land,
collecting and selling firewood, charcoal, bakery, salt trade and pottery, selling local food and
drinks (fentie & sundara, 2016; tewele, 2012; yizengaw, 2014). In Debark Woreda, common
rural off-farm activities include: milling, Weaving, handicrafts, Trade in grains and livestock,
general trade, earnings from sharecropped land, collection and sale of firewood, charcoal
production and sales, bakery products, salt trade, pottery, Selling local food and beverages (e.g.,
tella, tej, and other traditional drinks), guiding tourists, renting out pack animals (mules/horses),
selling souvenirs, running small lodges, and offering transportation services and selling local
food and beverages
Despite the significant potential for off-farm activities to impact the food security status of farm
households and alleviate poverty, there is a lack of clear development policies that recognize and
integrate rural off-farm activities as a vital part of the rural economy and a source of employment
in Ethiopia. This is because policymakers in our country primarily focus their attention on
agricultural sector strategies as the main avenue for reducing rural poverty and ensuring food
security in the country. However, the country is experiencing fluctuations in agricultural income

Page 1
due to population growth from limited arable land and climate variability, as well as human-
induced shocks, resulting in a continuous rise in rural poverty and food insecurity issues (ifad,
2011; kassie et al. , 2017).
Thus, off-farm activities can serve as a coping mechanism to help manage unexpected income
losses for survival and enhance food security (zeraiand gebere egziabher, 2011). Numerous
studies examine the economic effects of rural household involvement in off-farm activities on
farm household food security through household expenditure (amurtiya, 2015; endiris et al.
cogent food and agriculture 2021).
1. 2 Statements of the problem
In rural Ethiopia, agriculture primarily serves as the main livelihood approach for the majority of
households. Nevertheless, throughout the years, the shortcomings of agriculture in delivering a
sustainable income, especially when faced with obstacles such as climate variability, low
productivity, and lack of market access, have compelled numerous rural households to pursue
alternative income sources. Off-farm involvement, which encompasses activities like wage labor,
small enterprises, and non-agricultural services, has emerged as a vital coping mechanism for
farming households (hussien et al. , 2020). Despite its increasing relevance, the degree to which
economic elements, including household income, land size, access to credit, and market
accessibility, affect the choice to participate in off-farm activities is still inadequately
investigated, particularly in areas like Debark woreda in the Amhara region of Ethiopia.

In Debark woreda, a region primarily defined by agriculture, farm households encounter

difficulties stemming from poor agricultural performance, restricted access to financial
resources, and underdeveloped market infrastructure. While some households actively engage in
off-farm income-generating endeavors, others continue to depend solely on agriculture for their
income. This variation in off-farm participation indicates that economic factors might be
significantly influencing these choices, yet the nature and degree of these factors remain unclear.
The current literature on rural Ethiopia frequently highlights variables such as landholding size,
access to credit, and income levels as possible determinants of off-farm participation; however,
the interaction of these factors within the specific setting of Debark woreda has yet to be
examined (bekele et al. , 2018; davis et al. , 2010).

Furthermore, although research has recognized off-farm participation as a crucial coping strategy
for farm households (gebremedhin et al. , 2013), there is a scarcity of studies on how the
economic conditions particular to regions like Debark woreda affect these choices. The available
research often lacks regional detail and does not take into account local economic variables that
may be distinctive to Debark woreda, such as the degree of infrastructure development, regional
policies, or access to informal credit systems.

The research gap exists in the absence of empirical data and thorough analysis regarding the

Page 2
economic factors that influence the decision for off-farm participation in Debark woreda, a
region characterized by unique agricultural and socio-economic features. Grasping these factors
is essential for formulating policies that can effectively aid in the economic diversification of
rural households and mitigate their susceptibility to agricultural risks. The main drive for
executing this research originates from the acknowledgment of the increasing significance of off-
farm activities in enhancing the livelihoods of rural households in Ethiopia. As rural regions
confront growing challenges in agriculture—such as climate change, land degradation, and
market variations—off-farm participation may provide a means for diversifying household
income and boosting resilience (alemu et al. , 2021). However, comprehending the economic
determinants that sway such decisions is vital for crafting targeted interventions that can
encourage sustainable off-farm activities. Debark woreda, with its agricultural foundation yet a
complex socio-economic landscape, acts as a fitting case study to investigate these dynamics.

The research aims to provide insights into how economic factors, such as income, land size, and
access to financial services, affect the likelihood of farm households in debark woreda engaging
in off-farm activities. The findings will contribute to the development of region-specific policies
and strategies aimed at enhancing off-farm participation, which is vital for promoting economic
diversification and improving the resilience of rural households in Ethiopia. By focusing on
debark woreda, the study also aims to fill a significant gap in the literature regarding the role of
regional-specific factors in influencing off-farm participation decisions. While existing studies
have examined off-farm participation in Ethiopia and other parts of Sub-Saharan Africa, the
research on the economic determinants of off-farm participation, particularly in rural Ethiopia,
remains limited. Previous studies, such as those by hussien et al. (2020) and bekele et al. (2018),
highlight general factors like landholding size, income diversification, and access to credit, but
they do not provide an in-depth analysis of how these factors interact within specific local
contexts. Furthermore, studies often overlook important variables such as market access,
transportation infrastructure, and regional development policies, all of which could significantly
influence off-farm participation in a specific area like Debark Woreda.

There is a clear gap in understanding the economic dynamics of off-farm participation in rural
Debark Woreda. While off-farm activities are seen as a coping mechanism for risk
diversification and income improvement, the economic factors driving this decision have not
been sufficiently explored, particularly in the context of this specific rural area, which has its
own unique agricultural and socio-economic conditions. The research addressed this gap by
conducting a region-specific study that will offer both theoretical insights into the economic
determinants of off-farm participation and practical recommendations for policymakers working
in Debark Woreda.

1.3 Objectives of the Study

Page 3
1.3.1 General Objective
The general objective of this research is to examine the determinants of off-farm participation
decisions among farm households in Debark Woreda.

1.3.2 Specific Objectives

To achieve this, the study aims to address the following specific objectives:
• To identify the factors that affect the off-farm participation decisions of farm households in
Debark Woreda, and

• To assess the involvement of smallholder farm households in non-farm activities within the
study area.

1.4 Research question

The review will try to answer the following questions:

• What are the major factors that affect off-farm participation decision in farm households?

• What are non-farm activities of small holder farm found in the study area?

1.5 Significance of the Study

This study provides valuable insights into the participation of rural households in off-farm
activities in Debark Woreda. By analyzing the factors influencing off-farm employment, the
research enhances understanding of the determinants shaping rural labor decisions. The findings
contribute to the existing literature on off-farm participation in Ethiopia by offering region-
specific evidence. Moreover, they serve as a practical resource for microfinance institutions,
donors, and policymakers by highlighting the effects of non-labor income on off-farm
engagement. This information is crucial for designing effective financial support mechanisms
and development programs.

For rural households, the study aids in informed decision-making, helping them evaluate whether
to engage in off-farm activities based on economic conditions and potential benefits.
Additionally, researchers and scholars can use this study as a reference for further investigations
into rural labor markets and income diversification strategies. The findings also provide policy-
relevant recommendations for government bodies and institutions, supporting the development
of strategies aimed at improving rural livelihoods and employment opportunities in Debark
Woreda.

1.6 Scope of the study

Page 4
The study focused solely on the economic factors influencing the off-farm participation decisions
of farm households. The research was conducted in the North Gonder Zone, specifically within
Debark Woreda, located in the Amhara Region of Ethiopia. The scope of the study was
geographically restricted to rural households in this area to ensure that the data collected was
representative of the rural agricultural communities in the region.

The rationale for choosing the North Gonder Zone, Debark Woreda included factors such as the
area's agricultural dominance and the need for off-farm diversification, vulnerability to climate
variability, limited previous research in the area, policy relevance, and homogeneous
socioeconomic characteristics. By limiting the study to the North Gonder Zone, Debark Woreda,
the research provided a detailed, context-specific understanding of the factors influencing off-
farm participation among rural households, contributing to more tailored policy
recommendations.

1.8 Organization of the study

The study is organized into five chapters, each addressing different aspects of the research
process. The first chapter represents the introduction that consists of background of the study,
statement problem, objectives of the study, and significance of the study and the scope of the
study. The second chapter presents the literature review that deals with theoretical and empirical
review of off-farm employment participation decision of farm households. The third chapter
presents methodology and description area of the study. As well as, the fourth chapter presents
and analyzes the research findings. It includes the presentation of statistical analysis, discussion
of the economic factors affecting off-farm participation, and their implications for farm
households in Debark Woreda. It compares the findings with existing literature and interprets the
results in the context of the local economic environment. Finally, the last chapter summarizes the
key findings of the study, provides conclusions drawn from the research, and offers
recommendations for policy, practice, and future research. It also discusses the limitations of the
study and suggests areas for further investigation.

Page 5
 Chapter Two

Review Literature

2.1 Theoretical Literature Review

2.1.1 Introduction to Theory of Off-Farm

The neoclassical farm household model anticipated that a farm household decides to work off the
farm or on the farm depending on the marginal return from farming and labor. For an individual,
the decision to engage in off-farm work is determined by comparing the market wage rate with
the reservation wage. The reservation wage is the marginal value of time when no individual is
engaged in off-farm employment. A person will take part in off-farm work when the reservation
wage is less than the market wage (Abebe, 2002). In Africa, off-farm economic activities, as a
method of income diversification, are extremely vital for enhancing the livelihoods of the rural
poor (Asenso-Okyere and Samson, 2012; Diao and Ninpratt, 2007).

Moreover, it can act as a source of inputs for agricultural production and offer employment
opportunities for those who lack arable land and do not wish to depend solely on agriculture.
Despite its importance, participation in off-farm activities in Africa is low; and Haggblade et al.
(2007) indicated that 37% of rural household income is derived from off-farm activities, while
surprisingly, less than 20% of the labor force is involved. Similarly, in Nigeria, research by
Adewunmi et al. (2011) discovered that involvement in off-farm activities, particularly in wage
employment for both skilled and unskilled workers, has led to a reduction in rural poverty by
11.02% and 10.68%, respectively; those who participate in off-farm activities reduce poverty
more effectively than those who do not (Alaba and Kayode, 2011).

In Ethiopia, which is categorized as a developing region, particularly in Sub-Saharan Africa,

agriculture is the fundamental backbone supporting 85% of the Ethiopian population, allowing
them to provide sustenance for the remaining 15% of the urban population. In this context, the
Ministry of Agriculture and Rural Development (2010) recognized smallholder farmers as
responsible for producing the majority of the country‘s total output. Additionally, the World
Bank (2010) concluded that 90% of foreign earnings and 70% of the raw materials used in
industry come from agriculture. Thus, 85% of Ethiopians consider agriculture to be the solution
for achieving food security and even prosperity in the long run. Reflecting this deep-seated
belief, in recent times, the size of arable land and agricultural production has been declining,
forcing society to either shift to off-farm activities or seek food aid to satisfy the food needs of
households.

2.1.2 Definition of Off-Farm Participation and Income

Off-farm participation refers to individuals engaging in activities outside of their own

agricultural work environment. It encompasses economic activities conducted by employment
with others or establishing one's own business activities not situated on personal farmland. Off-
Page 6
farm employment is categorized into two types: self-employment and wage employment. Off-
farm income is generated when a farmer, their spouse, or another family member engages in
work away from the farm, thus providing additional financial resources for the household. This
income is also known as plural activity. It represents the segment of a farm household's income
acquired off the farm, which includes non-farm wages and salaries, pensions, and interest
earnings generated by farming families. Pensions are not included in the computation of off-farm
income (Amsalu et al., 2013).

Self-Employment

Self-employment is the process of earning a livelihood by doing something independently. In the

context of business, self-employment involves generating income by utilizing one's own capital
or borrowed funds, employing personal knowledge, skills, efficiency, and accepting risks
(Norsida, 2000).

Different types of self-employment

In the United States, there are numerous ways one can achieve self-employment. It represents a
specific kind of labor market engagement that has its own tax classification, covering a wide
range of occupations and industries. Therefore, individuals such as artists, musicians,
accountants, doctors, mechanics, real estate agents, consultants, lawyers, and others can fall
under the self-employed category. Hence, one can be self-financing (Fekadu, 2016/17).
Employment may also occur in sectors such as manufacturing, retail trade, professional services,
and personal services.

Other forms of self-employment include small-scale trading, selling fruits, mining stones, selling
wood, and charcoal among others. Small-scale trading is a prevalent socioeconomic activity that
supports a large number of low-income individuals in rapidly urbanizing developing nations (D.
Mbisso, 2011). Wood fuel comprises resources like firewood, charcoal, chips, pellets, and
sawdust. Currently, the burning of wood accounts for the largest consumption of energy sourced
from solid fuel biomass (Fekadu, 2016/17).

Self-Employment among Immigrant and Ethnic Minorities

Self-employment is notably common among recent immigrants and ethnic minorities within the
United States. It is observed that self-employment is more prevalent among immigrants
compared to their second-generation children who are born in the United States. Nonetheless, the
second-generation offspring of Asian immigrants continue to pursue self-employment across
various sectors and professions (Abebe, 2016/17).

Taxation

In the United States, the self-employment tax is usually established at 15.3%, which is
approximately equivalent to the combined contributions made by both the employee and the
employer under tax regulations. Generally, only 92.35% of the self-employment income is
Page 7
subject to these tax rates. Moreover, half of the self-employment tax, specifically the portion
equivalent to the employer's share, can be deducted against income (Wikipedia,
2016/17/Temporary Payroll Tax Cut Continuation Act of 2011).

Determinants of Off-Farm Participation

Number of emigrants: the likelihood of securing an urban job is directly linked to urban
unemployment. Thus, employers who relocate to urban areas are more likely to invest in off-
farm activities. (Todaro, 2009, Economic Development)

Education of the household head: when the household head is educated, participation in off-farm
activities by the household will increase. This is because educated men and women possess
advanced production skills across various fields such as computers, software, biotechnology, and
financial development. (Todaro, 2009, p. 654, Economic Development).

Sex of the household head: households led by females generally exhibit lower participation in
off-farm activities. This is due to their lower labor productivity on farms and restricted access to
capital, resulting in higher investment returns. (Todaro, 2009, pp. 341).

Age of the household head: as the age of the household head increases, the likelihood of
participation decreases. Younger heads of household tend to be more associated with off-farm
income (Nahume, 2016).

Number of dependents in the household: a greater number of dependents in the household results
in the need for more income, which diminishes the marginal value of leisure, thereby
necessitating an increase in off-farm labor supply. (Beyene, 2008).

Size of cultivated land: this refers to the extent of land cultivated by the household. A farmer
with a larger farm participates less frequently in off-farm activities because they must dedicate
more time to farm activities (Abebe, 2002).

Non-labor income: this refers to remittances from outside the family or from household members
who emigrate to foreign countries. When emigrant members of the household are abroad, they
are more likely to find job opportunities (Todaro, M., 2009). Consequently, these emigrants send
money back to their families, which encourages participation in off-farm activities.

2. 2 Empirical Literature Review

Literature that focuses on off-farm employment has concentrated on the factors influencing
participation in off-farm work and the supply of off-farm labor. Numerous studies have
examined the decision to participate in off-farm activities among farm households in Ethiopia,
Page 8
including those by Tesfaye W. (2016), Berihun (2016), Mohammedawol (2015), Babagunde et
al. (2010), and Nahume (2016), among others.

As the engine of economic development and poverty alleviation, in developing nations,

agriculture ought to be connected with sectors that possess direct or indirect associations. Based
on Babatunde et al. (2010), financial resources seem to be the primary constraint on farming;
thus, income generated from off-farm activities contributes to enhanced farm production,
elevated household income, and diminished risk of crop failure. Therefore, off-farm activities
represent one of the areas where agriculture is considered to be integrated. The World Bank
(2008) has indicated that, in many developing nations, the significance of off-farm activities is
rising and is estimated to make up 30 to 50% of rural earnings.

Tesfaye (2016) examined the impact of off-farm income on the commercialization of

smallholder farmers in rural Ethiopia. He conducted his analysis through a double-hurdle model
using data from an Ethiopian rural household survey. The findings of this study revealed that
additional off-farm income adversely impacts market surplus. This suggests that farmers utilize
off-farm income primarily for consumption rather than for investment purposes.

Babatunde R. O. et al. (2010) researched the factors influencing participation in off-farm

employment among smallholder farming households in Kwarastate, Nigeria. Their investigation
employed both descriptive and multivariate logit models. According to the descriptive analysis,
65% of households engage in off-farm employment; the econometric analysis indicates that
household size, production asset quantity, gender of the household head, education level of the
household head, access to electricity and piped water, and proximity to the nearest urban market
significantly influence off-farm participation.

Beyene (2008) assessed the factors determining the decision of farm households to participate in
farm work by applying a logit model to analyze the data. He concluded that the household's
human capital and education do not affect the off-farm work participation decision. The health
status of the family is, undeniably, a crucial element in off-farm participation.

Berhun et al. (2016) studied the decision to participate in off-farm activities and its effect on crop
yield in Northern Ethiopia. They utilized descriptive statistics, the logit model, and the ordinary
least squares method to evaluate the data. This research found that off-farm participation was
positively influenced by gender, education, working individuals, the number of pack animals,
and access to credit. Conversely, age and land size negatively impacted off-farm participation.
Additionally, off-farm participation also had a negative effect on crop production.

Amsalu et al. (2013) investigated the factors that influence the likelihood of household
participation in off-farm work using data collected from rural Ethiopia, applying descriptive
Page 9
analysis and the logit model. The study's results indicated that individual characteristics,
household composition, credit availability, off-farm equipment value, and location factors
significantly affected participation.

Delil (2001) explored the determinants of the likelihood of off-farm employment in Oromia. He
found that variables such as family size, religion (Orthodox), and access to credit positively
influenced the probability of off-farm employment. However, variables like the size of cultivated
land, a married household, a head of household working on their account, coffee and chat
producer households, fertilizer-using households, and the number of cattle owned by households
negatively and significantly influenced the likelihood of farm households participating in off-
farm work in the study area.

Abebe (2002) examined off-farm participation using Hickman's two-step regression analysis. He
demonstrated that the education level of household members is a primary factor influencing the
off-farm wage rate for both male and female family members. However, the education level of
the household head does not impact participation in off-farm activities. The presence of children
receiving training in handcraft skills positively influences off-farm participation. Factors such as
land size, age, and the number of dependents negatively affect the decision to engage in off-farm
participation.

Several researchers have explored off-farm activities in Ethiopia and their determinants. The
researchers who investigated off-farm activities participation did not analyze the impact of the
number of immigrants in a household. Beyene (2008) outlined in his study the reasons that
compel individuals to participate in off-farm activities. Nonetheless, he did not explain the
reasons for the inability to engage in off-farm activities. This paper aims to address this aspect
additionally, and we will also examine the theoretical effect of non-farm activity on farm
production.

2.3 Conceptual Framework

Dependent and Independent Variable
Off-farm participation serves as the dependent variable, influenced by several independent
variables that determine household engagement in off-farm activities. These explanatory
variables may include household characteristics, and economic factors.

Figure 2.1: Conceptual Framework

Page 10
 Age of
household
head

Number of Sex of
dependents in Off Farm household
the household participation head

Non-labor
income

Figure 2.1 conceptual frame work that shows dependent and independent variables.

dependent variable (off-farm participation) is typically measured as a binary (yes/no) or

continuous variable (extent of participation, income share from off-farm activities).
Independent variables may include:
Household factors: (Age of Household, Sex of Household Head, Education Level, Number of
Dependents in the Household, Number of Emigrants from household.

Economic factors: (Size of Cultivated Land, Number of Animals in Household, Non-labor

Income.

Page 11
 Chapter Three

Research Methodology

3.1 Description of Study Area

Debark woreda is situated in the northern region of Ethiopia, specifically within the North
Gondar Zone of the Amhara Region. It lies at the foot of the Simien Mountains, a UNESCO
world heritage site, which contributes to its geographical and cultural importance. The Woreda
features a varied agro-ecological setting, which is crucial in shaping the livelihoods of its
residents. The study area is primarily rural, with agriculture being the main economic activity.
However, diversification of the economy, especially through activities outside of farming, has
become an increasingly vital income source for numerous households.

3.1.1 Geography and Climate

Debark Woreda is found at different elevations that range from approximately 2,000 to over
4,000 meters above sea level, affecting the climate and agricultural trends in the area. The
climate changes from temperate to highland (cool), with two primary seasons: a wet season
(from June to September) and a dry season (from October to May). These climatic factors are
essential for establishing the agricultural productivity of the region, which mainly relies on rain-
fed farming. The precise geographic position of Debark woreda can be indicated by its latitude
and longitude. Debark woreda is located at approximately: latitude: 13. 0833° N and longitude:
37. 9167° E.

3.1.2 Demographic Characteristics

Based on the 2007 Population and Housing Census of Ethiopia, Debark Woreda had a total
population of 159,193, which includes 80,274 men and 78,919 women. The rural population of
Debark woreda heavily relies on subsistence agriculture. According to the most recent census,
the populace of Debark woreda is predominantly ethnic Amhara, with a few additional ethnic
groups present. The population density is relatively low, attributed to the difficult terrain, but
there are settled areas surrounding the woreda's administrative center and close to transportation
routes. The average household size in Debark woreda is fairly large, suggesting that a significant
labor force is accessible for both farming and off-farm tasks. Most of the population engages in
agriculture, but younger members of households are increasingly pursuing off-farm job
opportunities in nearby urban centers, such as Gondar or other areas in Ethiopia.

Page 12
3.1.3 Agricultural Practices and Economy

Agriculture serves as the cornerstone of the economy in Debark woreda. Households generally
participate in both crop cultivation and livestock farming. Principal crops cultivated in the
woreda consist of maize, barley, wheat, teff, and peas. The livestock sector holds significance as
well, with cattle, sheep, goats, and donkeys being the primary livestock raised. Nonetheless,
agricultural productivity is frequently limited by factors like land fragmentation, tenure issues,
and fluctuations in climate. Central Statistical Agency of Ethiopia (CSA). (2021).

3.1.4 Off-Farm Activities and Economic Diversification

Participation in off-farm activities in Debark woreda has been increasing in recent years,
motivated by the necessity to supplement farm earnings. Off-farm endeavors encompass
construction jobs, transportation services, small-scale trading, local artisan enterprises, and wage
labor in the nearby towns of Gondar and other urban locations. Many households depend on the
income generated from these ventures to bridge the gaps in agricultural output or to enhance
their overall household welfare.

Figure 3. 1 Location Map of The Study Area.

Page 13
3.2 Research Design

The research utilized a quantitative research design. More specifically, it was a cross-sectional
survey study that gathered data from a sample of farm households at a particular moment in time
to examine the connections between economic factors and off-farm participation.

3.3 Sources of Data

Data was obtained from both primary and secondary sources. The primary data sources included
interviews and questionnaires. Secondary data sources encompassed prior research, official
statistics, books, and other pertinent materials collected for the research.

3.4 Sampling Technique and Sample Size

The research applied a two-stage sampling technique to choose sample respondents. In the first
phase, simple random sampling was used to select three kebele administrations (specifically
zebena, debir, and gomya) from a total of 22 kebeles in the area under study. During the second
phase, a probability proportional to size sampling method was applied to ascertain the number of
households within the selected kebeles, followed by the selection of all respondent households
using a lottery method in those kebeles. This indicates that 202 respondent households were
randomly chosen from the total number of households in the three selected kebele
administrations. In this research, the sample size was calculated using the Yamane formula
(Yemane, 1967). The formula is expressed as:

n = N/1+N (e)²

= 18115/1+18115(0.07)²

= 201.8081

= 202

Where: n = represents the statistically acceptable sample size.

N = signifies the total number of target households (18,115) in the study area.

e = indicates the level of precision/margin of error (7%).

Sample size by Kebele;

NO Kebele Household Proportion Sample proportion

Page 14
 1 zebena 563 26% 52
2 debir 990 44% 89
3 gomya 670 30% 61
Total 2223 100% 202

3.5 Methods of Data Collection

Data were collected for this study with the aim of grasping the engagement in off-farm activities
and the factors that affect it. Both primary and secondary data sources were utilized.

3.5.1 Primary Data

Primary data were obtained via structured questionnaires distributed to chosen households in the
research area, as well as through interviews with key informants. Moreover, focus group
discussions were conducted with 8–12 participants in each of the three selected kebeles. The
structured questionnaires gathered cross-sectional information on socioeconomic, demographic,
and institutional factors that impacted off-farm participation. These questionnaires were given to
representative respondents.

3.5.2 Secondary Sources of Data

Secondary data were collected from a variety of published and unpublished documents,
including reports from the district human resource and administration office.

3.6 Method of Data Analysis

The raw quantitative and qualitative data gathered from the surveyed households were edited,
coded, entered, cleaned, and analyzed with Stata 15 software. An econometric model was
employed for the analysis. Both descriptive and inferential statistics were used to scrutinize the
survey data. Descriptive statistics, like frequency and percentage, were applied for dummy
variables qualitatively, while minimum, maximum, mean, and standard deviation were utilized
for continuous variables to encapsulate and display the quantitative data. Inferential statistics,
including the t-test for continuous variables and the chi-square test for independent discrete
variables, were used to examine the significance of associations with the dependent variable.

Page 15
3.7 Model Specification

The research method applied in this study addressed the factors influencing off-farm
participation for farming households (whether or not they engaged in off-farm activities). A
binary choice model, specifically the logit model, was utilized to estimate the model according to
Gujarati D. N. (2004). Although numerous binary regression models exist in econometrics,
where the dependent variable is categorical and takes a value of either 0 or 1, the logit model was
favored over the linear probability model (LPM) when the response could signify one of two
outcomes representing success or failure. The LPM was considered less suitable due to the
following concerns:

• Non-normality of the disturbances

• Constant marginal effects • Probabilities exceeding one

• A questionable R² value as a measure of goodness of fit

Hence, the logit model estimates the probability of engaging in off-farm activities as a function
of household, economic, and social factors. The model transforms the dependent variable (off-
farm participation) into log-odds form, ensuring that the predicted values remain between 0 and
1.

Since the probability (pi ) of off-farm participation lies between 0 and 1, the odds ratio (Ofp) is
modeled as:

Ln (pi/1−pi) =α+β1Edu+β2Sex+β3Age+β4Dep+β5Mip+β6Farm+β7Ani+β8Nli+Ui

Interpretation

The estimated model can be delineated as follows, where l signifies the natural logarithm of the
ratio of the probability of engaging in off-farm activities to the probability of not engaging. The
error term is represented as Ui. The variables consist of:

• Ofp = off-farm participation

• Edu = education level of households

• Age = age of household head

• Farm = land size of household

• Sex = sex of household head

Page 16
• Dep = number of dependents in the household

• Mip = count of emigrants from the household

•Ani= Number of Animals in Household

•Nli= Non-labor Income

• α = reflects the odds ratio supporting participation in off-farm activities

• β₀, β₁,. . . , β8 = the variation in the log-odds ratio favoring success for a unit alteration in the
related independent variables

3.7.1 Specification of the Logit Model

The goal of the logit model was to guarantee that the likelihood of an event happening, given the
values of explanatory variables, remained constrained between 0 and 1. This was accomplished
through the cumulative logistic distribution function, which upheld the condition that 0 ≤ pi ≤ 1,
even as zi extended from negative infinity to positive infinity. The odds ratio was computed as
the quotient of the probability of an event occurring (pi) to the probability of it not occurring (1 -
pi), and taking the natural log of this ratio produced the logit. The logit was linearly associated
with the explanatory variables (xi), with Ui denoting the residual (error term).

3.8 Description and Expected Sign of Variables Used in the Analysis

As discussed in Section 1.3, the main objective of the study was to identify the determinants of
off-farm participation among farm households. The variables used in the analysis, along with
their expected signs on participation and the hours of work in off-farm activities, are as follows:

 Off-farm Participation: Variables influencing off-farm participation included age, age

squared, sex of household head, education level, location, health status, handicraft
training, presence of children, number of draft animals, distance to the nearest market,
number of dependents, non-labor income, amount of credit, and size of cultivated land
(Abebe, 2002; Tesfaye, 2009).
 Age of Household Head: Age represented general experience, which increased the
marginal value of each activity. At younger ages, the probability of participating in off-
farm work increased (Abebe, 2002). Therefore, it was hypothesized that age of the
household head had a negative impact on participation in off-farm activities.
 Size of Cultivated Land: Increasing land size was expected to decrease the probability
of participating in off-farm activities, as larger landholdings required more time for farm

Page 17
 work (Beyene, 2008). Hence, it was hypothesized that the size of cultivated land
negatively affected off-farm participation.
 Sex of Household Head: As a dummy variable, male was represented by 1 and female
by 0. Males were hypothesized to have a higher probability of participating in off-farm
activities, given they had more opportunities to search for non-farm employment (Ellis,
2000; World Bank, 2008).
 Number of Animals in Household: The possession of draft animals and other livestock
was expected to negatively affect off-farm work participation, as livestock required more
time for agricultural activities (Beyene, 2008). Hence, it was hypothesized that the
number of animals had a negative impact on off-farm participation.
 Number of Emigrants from Households: The number of emigrants was expected to
positively influence off-farm participation. Emigrants often move to areas with better
employment opportunities, which could increase the household's access to non-farm
income and encourage others to follow suit (Taylor & Martin, 2001; Ellis, 2000).
 Number of Dependents in the Household: A higher number of dependents, particularly
those under 15 or over 60, was expected to negatively affect off-farm participation, as
households with more dependents faced greater economic pressure, limiting their ability
to allocate labor to off-farm activities (Huffman, 1980; Ellis, 2000).
 Education Level: Education was expected to have a positive relationship with off-farm
work participation, as educated household heads were more likely to diversify income
sources and make rational decisions to maximize income (Schultz, 1961; Ellis, 2000).
 Non-labor Income: Households with substantial non-labor income (e.g., government
allowances or donations) were hypothesized to be less likely to engage in off-farm
activities, as their economic needs were partially met without relying on variable labor
income (Becker, 1965; Benjamin, 1992).

3.9 Diagnosis Test

3.9.1 The Likelihood Ratio Test

To test the correctness of the regression results, the likelihood ratio (LR) test was used, which
follows a chi-square distribution with the degree of freedom equal to the number of explanatory
variables. This test was performed by estimating two models and comparing the fit of one model
to the fit of another. It compared the log-likelihoods of the two models and observed the
significance. The formula for the LR test was:
LR = -2ln(l(m1)/l(m2)) = 2(ll(m2) - ll(m1)), where l(m*) represents the likelihood of the
respective model.

Other tests, such as the Lagrange Multiplier, were also used to test a single model. Additionally,
to test the goodness of fit, the pseudo R² was used instead of the conventional R², as the logit
model primarily focuses on inference. For hypothesis testing, standard z-statistics were used,
Page 18
which followed a normal distribution (Maddala, 1997). The analysis was conducted using Stata
version 15, Microsoft Excel 2010, and Microsoft Word 2010.

3.9.2 Heteroscedasticity
Heteroscedasticity was tested using the Breusch-Pagan (hettest) test, which performed a chi-
square test. If the test result indicated a p-value greater than 5%, it was concluded that there was
no heteroscedasticity problem. If heteroscedasticity was detected, it was addressed by using
robust standard errors during the regression analysis.

3.9.3 Multicollinearity
Multicollinearity was tested using the variance inflation factor (VIF). If the VIF was greater than
10, it indicated the presence of multicollinearity. In such cases, collinear variables were dropped
to resolve the issue. The study examined whether multicollinearity existed and addressed it as
necessary.

3.9.4 Normality Test

Normality was tested using the kernel density test. If non-normality was detected, the researcher
transformed the variables to resolve the issue.

3.9.5 Goodness of Fit of the Model

The goodness of fit of the model was checked by computing the coefficient of determination
(R²). In line with recommendations from most scientific studies, a value greater than 70% was
considered desirable. If the coefficient of determination was found to be less than this threshold,
the sample size was increased to improve the model's fit.

Page 19
 CHAPTER FOUR

DATA ANALYSIS AND PRESENTATION

4.1. Descriptive analysis

The purpose of descriptive analysis is to elaborate and understand socioeconomic condition of
household characteristics (participation on off farm activity, age, sex, education, cultivated land
size, non-labor income, dependents in the household and other variables in the study)
theoretically in the form of table and graph if possible.

Off farm participation: As shown in the table below the respondents who participate in off
farm activity in Debark woreda, constitute 62.87%of the total sample, while the remaining
37.13% of respondents do not participate in off farm activity. Respondents set different reasons
for their decision of being in or out of off farm activity.

Table 4.1 Off farm participation

Off farm participation frequency Percentage

Participant 127 62.87%
Non participant 75 37.13%
Total 202 100%
Source; own survey 2017 E.C

Classification of population by sex: When we classify the total population by sex on the
table below, all most 57.92% of our respondent are male headed only 42.08% are female headed
because of death of husband and divorce and again we had not got female headed respondent
when a husband alive. From female headed respondents 53 female participates in off farm
activity which constitutes 26.24% of participant from total sample. This shows us gender plays a
major role in participation decision although those respondents set other reasons also as
complementary not to participate in off farm activity. Among male respondents,74 out of 117,
are participant in off farm activity constituting 58.27% from total sample and36.63% of total
participants. the table below summarizes the above fact.

Page 20
Table 4.2 Gender description

Gender frequency Percentage Participant % of participant from % of participant

total participant
from total sample

Male 117 57.92% 74 58.27% 36.63%

Female 85 42.08% 53 41.73% 26.24%
Total 202 100% 127 100% 62.87%
Source; own survey 2017 E.C (significant at 10%,5% and 1%) and You can show stata result
below (Appendix 1)

Effects of age on off-farm activity participation: According to our study age plays a vital role in
households‘ participation decision of the study area of a society. When the age of the household
leader goes, probability of participating in off farm activity lowers continuously and family
members are forced to do and substitute fathers‘ role in family by being out of off farm activity.
The data collected shows that all most 63.37% of off farm participants are aged (15-64) years
old.

Table 4.3 Age structure and off farm participation

Age freque percentage participant % of participant % of participant

ncy from total sample from totalparticipant
Working age population (15-64 128 63.37% 109 53.96% 85.83%
age) participants
Young age population (Below 74 36.63% 18 8.91% 14.17%
15 ) and Elderly population(
Above 64) age participants
Total 202 100 127 62.87% 100%
As you look on table above from 74 (old and young age) respondents 18 participates are
participate in off-farm activity which is around14.17% from total participant‘s and 8.91% of total
sample size. Working age group (15-64 years old) constitute 63.37% of the total population and
85.83% of respondents participate in off farm activity. This shows effect of the age under young

Page 21
and elder age population has adversely affect the participation decision of a family. The mean
age of our respondent group is 36.80 years old as shown below.

Effects of farm land size on off-farm activity participation:

From table 4.4 as settled in the table below, more than 51.98% of respondents have land less than
2 hectares. From 107 respondents participate in off farm activity constituting52.97% from the
total participant and 48.02% of respondents have arable land of more than 2 hectares from 20
respondents only participate in off farm activity constituting 9.9% from total sample and 15.45%
from total participant.

Table 4.3 Description of farm land holding structure

Size in hectare Frequency percentage Number of % of participant % of participant

from total sample from total
participants
participant
0-2 105 51.98% 107 52.97% 84.55%
Above 2 97 48.02% 20 9.9% 15.45%
Total 202 100% 127 62.87% 100%
Source; own survey 2017 E.C (significant at 10%,5%and 1%)

This shows respondents having less arable land are participating in off farm activity actively as
alternative or complementary job to their life. Mean of the target group land holding is 3.53
hectare.

Effects of education level on off-farm activity participation:

Table 4.5 Description of the education level

Education Freque % cover from No of % of participants from % of participants

total sample participants total participant from total sample
ncy
Degree and above 110 54.46% 94 73.64% 46.53%
Diploma 84 41.58% 30 23.64% 14.85%

Page 22
 Lower secondary and 8 7.27% 3 2.73% 1.49%
preparatory education
Total 202 100% 127 100% 62.87%

Source; own survey 2017 E.C(significant at 10%,5% and 1%) and You can see stata result on
(Appendix1)below.

From the above data we see that degree and above educated participants constitute 54.46% of the
total population and 73.64% of participants. And diploma participant constitute 41.58% of total
sample and 23.64% of participants. As education level increase the participation number of
house hold increase. This shows us education positively plays a major role on participating in off
farm activity.

Effects of Non labor income on off-farm activity participation:

Table 4.6 Non labor income and participation

NLI frequency percentage participant %of participant % of participant

from total sample from participants
Lessthan10000 83 41.09% 69 41.09% 54.5%
10000-20000 34 16.83% 40 24.26% 31.5%
Above 20000 85 42.08% 18 10.89% 14%
Total 202 100% 127 76.24% 100%
Source; own survey 2017 E.C (significant at 10%,5% and 1%)

The table above says that 41.09% of off farm participant receive less than 10000 birr non labor
income. Respondents with non-labor income above 20000 constitute only42.08% of off farm
participants, while middle non labor income receivers constitute 16.83% 0f total participants.
This shows us as non-labor income increase the probability of being participant in off farm
activity goes to decrease.

When we look immigrated number in a family all most half of the respondent have immigrant to
other place especially to town. Also half of respondents have at least one dependent to their
family. Our criterion for dependency is the age and disability. i.e a member with less than 18
Page 23
years or above 70(even if he/she is a family leader) or a disabled one is categorized under
dependent. Lastly, almost all respondents have at least 1 or 2 animal of any kind. more or less
90% of respondents with above 15 house animals do not participate in off farm activity. Instead
they devote the remaining time out of farming by looking after and collecting food for their
animals.

Reason to participation

4.7 Reason to participate

Reason to participate frequency Percentage

To help family 28 22.08%
Favorable conditions 48 37.66%
Limited farm income 45 35.71%
Other reason 6 4.55%
Total 127 100%
Source; own survey 2017 E.C

According to our survey, favorable conditions account for 37.66% and limited farm income
contributes to 35.71% of participants engaging in off-farm activities, while those who actively
participate in off-farm activities with the aim of helping their family represent 22.08%. The
remaining 4.55% of participants provide various reasons such as personal benefit, using the
activity as leisure, and other individual motivations. Overall, when we examine their motivations
for being active participants in off-farm activities, favorable conditions for participation
comprise the largest portion, indicating that they engage in off-farm activities when they have
favorable circumstances, such as proximity to off-farm activity areas and additional work
opportunities or job creation, particularly for seasonal participants, while many people's
motivation is also limited farm income, indicating that they are utilizing off-farm activities as a
supplementary source of income or, in essence, to diversify their income.

Table 4.8 reasons for to be non-participant

Reason frequency Percentage

Time inadequacy 13 18.48%
Unfavorable conditions 8 10.87%

Page 24
 Enough income 8 10.87%
Age 24 32.61%
Sex 20 27.17%
Total 73 100%
Source; Own survey 2017 EC

Based on the data, the leading reason for non-participation in Debark woreda is age. Individuals
who are unable to engage in off-farm activities due to age make up 32.61% of the total non-
participants and about 27.27% of all respondents. The second most significant reason for non-
participation is sex. According to our study, respondents who are unable to participate in off-
farm activities due to their gender account for 27.17% of non-participants and 22.73% of the
total respondents. The underlying reason is societal cultural views on women's employment
outside the home. Time inadequacy and unfavorable conditions also contribute to 10.87% of
non-participants.

4.2 Econometric analysis

This section aims to analyze factors that influence farmers' decisions regarding off-farm
participation. Logit regression analysis is used for estimation purposes.

4.2.1 Logit regression analysis of off-farm participation of respondents

This section presents findings from econometric results concerning the determinants of farmers'
off-farm participation decisions. The section includes the logit model utilized in the study, and
the results of the logit regression analysis of the specified estimated model are presented below.
The logit results below display coefficients, standard errors, pseudo r2, the standard error of the
logit regression, and the number of observations in the analysis. The regression results of the
study demonstrate that the model is statistically significant at the 5% level of significance or
within a 0.05 margin of error because the model's probability is less than 5% (0.0000chi2 value
of 0.0000 indicates strong statistical significance, enhancing the reliability and validity of the
model).
Even though the model has overall statistical significance, not every variable is statistically
significant on its own. Based on the results, age, sex, non-labor income, arable land, and
education are identified as significant factors influencing farmers' off-farm participation
decisions at a 5% significance level, as their respective p-values are below 5%. Non-labor
income, age, and the extent of arable land, along with being male and educated, positively
influence decision-making, while the number of immigrants, the number of animals, and the
number of dependents are not significant at the 5% level because their respective p-values (p. |z|)
exceed 5%.

Page 25
. logit ofp sex Age edu dep mip Farm Ani Nli

Iteration 0: log likelihood = -133.24671

Iteration 1: log likelihood = -68.967882
Iteration 2: log likelihood = -67.405952
Iteration 3: log likelihood = -67.364724
Iteration 4: log likelihood = -67.364669
Iteration 5: log likelihood = -67.364669

Logistic regression Number of obs = 202

LR chi2(8) = 131.76
Prob > chi2 = 0.0000
Log likelihood = -67.364669 Pseudo R2 = 0.4944

ofp Coef. Std. Err. z P>|z| [95% Conf. Interval]

sex -.5071946 .6099328 -0.83 0.406 -1.702641 .6882518

Age -.1142955 .0218586 -5.23 0.000 -.1571375 -.0714535
edu .9809372 .4817539 2.04 0.042 .036717 1.925157
dep -.4912017 .2327248 -2.11 0.035 -.947334 -.0350695
mip -.2308746 .180283 -1.28 0.200 -.5842228 .1224736
Farm -.207241 .1046499 -1.98 0.048 -.412351 -.002131
Ani -.008529 .0349808 -0.24 0.807 -.0770901 .0600321
Nli -.0000867 .0000199 -4.35 0.000 -.0001257 -.0000476
_cons 7.710862 1.19723 6.44 0.000 5.364333 10.05739

4.2.2 Odds ratio result and its interpretation

Based on our data stata64 gives the odds ratio logistic regression result below;

. logistic ofp sex Age edu dep mip Farm Ani Nli

Logistic regression Number of obs = 202

LR chi2(8) = 131.76
Prob > chi2 = 0.0000
Log likelihood = -67.364669 Pseudo R2 = 0.4944

ofp Odds Ratio Std. Err. z P>|z| [95% Conf. Interval]

sex .6021826 .3672909 -0.83 0.406 .1822017 1.990233

Age .8919943 .0194977 -5.23 0.000 .8545866 .9310396
edu 2.666955 1.284816 2.04 0.042 1.037399 6.856228
dep .6118906 .1424021 -2.11 0.035 .3877735 .9655383
mip .793839 .1431157 -1.28 0.200 .557539 1.130289
Farm .8128238 .0850619 -1.98 0.048 .6620919 .9978713
Ani .9915073 .0346837 -0.24 0.807 .9258064 1.061871
Nli .9999133 .0000199 -4.35 0.000 .9998743 .9999524
_cons 2232.465 2672.774 6.44 0.000 213.6488 23327.53

Note: _cons estimates baseline odds.

Education level; effect of education (giving 1 for educated) is positive on off farm participation
decision as we look from the regression result; and the odds ratio result is saying us ; being
educated increases odds ratio in favor of being participant by 2.6669ceteris paribus. The
implication is that family headed by educated one has higher probability to be participant. This is
because educated heads have more chance and ability to participate as well to advice the family
to participate in off farm activity.

Page 26
Age; as we seen from the coefficient regression result age has a negative effect and from odds
ratio; ceteris paribus an increase in age by one year reduces odds ratio in favor of participating
by 0.891994 keeping other things constant. This is because when age goes peoples loss effort
and time for off farm participation.

Cultivated land size; land has a negative effect on being participant. From the odds ratio result;
when land holding of a family increases by one hectare the odds ratio in favor of off farm
activity reduced by 0.812823ceteris paribus. This shows us peoples having more land spend
more of their time on farming related works rather than off farm in Debark woreda.

Family dependence size: dependence size mean the family members out of work age stage. (i.e
the age level under 18 years old and above 64 years old.) has a negative effect on being
participant on off-farm activity. From the odds ratio result; when dependent family member
increases by one the odds ratio in favor of off farm activity reduced by 0.6118906ceteris
paribus. This shows us increase in family dependence ratio has inversely related with
participation on off-farm activity.

Non labor income; has negative impact on off farm participation. Odds ratio result shows that
an increase in non-labor income by one birr results in a decrease in odds ratio in favor of
participating by 0.9999027 other things being constant.

4.2.3 Marginal effect and its interpretation

Though we discuss the logit result of regression, the important one for recommendation in logit
model is marginal effect. This is because it is the marginal effect, neither the logit nor odds ratio
which shows the probability. In terms of marginal effect the model is;

OFP = 0.21EDU -0.107SEX -0.0245AGE-0.1056DEP-0.04969MIP-0.0445FARM-0.0018ANI-

0.186(10)-4Nli

Education is an important factor that affects participation decision of a farmer. Being educated
has a positive effect, then the probability of being participant increase by 21.09% ceteris paribus.
Age is another important variable in the model with negative effect, and an increase in age of a
person by one year reduces the probability of being participant by 2.45%, keeping other things
constant.

Another important variable is farm land size, having a negative effect on being participant,
keeping other things constant an increase in arable land holding of a person by one hectare
reduces the probability of being participant by 4.45%.Lastly, non-labor income has a negative
effect on being participant, an increase in non-labor income by l birr, reduces the probability of
being participant by 0.0018 % ceteris paribus. The result is found below.

Page 27
. mfx

Marginal effects after logistic

y = Pr(ofp) (predict)
= .6869777

variable dy/dx Std. Err. z P>|z| [ 95% C.I. ] X

sex* -.107092 .12704 -0.84 0.399 -.356078 .141894 .579208

Age -.024578 .00516 -4.76 0.000 -.034688 -.014468 36.802
edu .2109401 .10461 2.02 0.044 .005901 .415979 .623762
dep -.1056277 .05107 -2.07 0.039 -.205722 -.005534 1.12376
mip -.0496471 .03843 -1.29 0.196 -.124973 .025679 1.15842
Farm -.044565 .02235 -1.99 0.046 -.088369 -.00076 3.53465
Ani -.0018341 .00752 -0.24 0.807 -.016567 .012899 9.82673
Nli -.0000186 .00000 -4.29 0.000 -.000027 -.00001 16165.7

(*) dy/dx is for discrete change of dummy variable from 0 to 1

.
Hetroscedasticity

Heteroscedasticity issues arise when there is varying variance within the error term. This means that the
variance of the error terms changes as the explanatory variable changes. As indicated by our results
below, Chi2=0.96 and prob>chi2=0.3270. When the hettest value exceeds 0.05, there is no issue with
heteroscedasticity.
. hettest

Breusch-Pagan / Cook-Weisberg test for heteroskedasticity

Ho: Constant variance
Variables: fitted values of ofp

chi2(1) = 0.96
Prob > chi2 = 0.3270

Multicollinearity

Multicollinearity is the relation of explanatory variables with each other or it is the problem
thatarises due to the presence of perfect linear relationship among explanatory variables. This
testshows the testing of interdependence of explanatory variables. The interdependence
ofexplanatory variables examined by variance Inflating factor (VIF).VIF should be< 10. If
VIF=1, perfectly no multicollinearity.This means explanatory variables are independent. If
VIF>10there is Perfect multicollinearity, that means impossible to determine the individual
impact of explanatory variables on dependent variables. When VIF is greater than or equal to ten
(VIF ) or tolerate (1/ VIF) is less than or equal to 0.10, accept alternative that
ismulticolinearity.When VIF or tolerance (1/VIF) greater than 0.10, accept the null hypothesis.
Page 28
That is nomulticollinearity. In this study variance inflating factor (VIF) was employed to
testmulticollinearity of independent variables. It shows that there is no problem of
multicollinearity,i.e. VIF of each variables less than 10 or (1/VIF) >0.10(see the result below).
. corr
(obs=202)

sex Age edu dep mip Farm Ani ofp Nli

sex 1.0000
Age -0.3956 1.0000
edu 0.5194 -0.3115 1.0000
dep -0.0044 0.1324 -0.0768 1.0000
mip 0.2265 -0.2331 0.1405 0.1710 1.0000
Farm -0.4796 0.5638 -0.3018 0.1757 -0.2866 1.0000
Ani -0.0938 0.3768 -0.1894 0.0777 -0.2378 0.4472 1.0000
ofp 0.2790 -0.6469 0.2886 -0.2388 0.0803 -0.4682 -0.2689 1.0000
Nli -0.0112 0.2024 0.0350 -0.0246 -0.0355 0.0845 -0.0749 -0.3582 1.0000

.
. vif

Variable VIF 1/VIF

Farm 1.98 0.505725

sex 1.78 0.561659
Age 1.69 0.590889
edu 1.43 0.698888
Ani 1.43 0.701405
mip 1.19 0.838771
dep 1.11 0.898499
Nli 1.09 0.916182

Mean VIF 1.46

Normality Test

Higher values of the t-score indicate that a large difference exists between the two sample sets.
The smaller the t-value, the more similarity exists between the two sample sets.

Page 29
. ttest sex=.5792079

One-sample t test

Variable Obs Mean Std. Err. Std. Dev. [95% Conf. Interval]

sex 202 .5792079 .0348219 .4949128 .5105447 .6478711

mean = mean(sex) t = 0.0000

Ho: mean = .5792079 degrees of freedom = 201

Ha: mean < .5792079 Ha: mean != .5792079 Ha: mean > .5792079
Pr(T < t) = 0.5000 Pr(|T| > |t|) = 1.0000 Pr(T > t) = 0.5000

.
Goodness of Fit of The Model

Goodness-of-fit tests are commonly used to test for the normality of residuals or to determine
whether two samples are gathered from identical distributions.

. estat gof, outsample

Logistic model for ofp, goodness-of-fit test

number of observations = 202

number of covariate patterns = 182
Pearson chi2(182) = 151.64
Prob > chi2 = 0.9509

Page 30
 CHAPTER FIVE
CONCLUSION AND RECOMMENDATIONS

5.1 CONCLUSION
The primary objective of this study is to analyze the key factors influencing rural household
involvement in off-farm employment. The exploration of off-farm activities contributes
significantly to household income. This will motivate households to engage in off-farm
activities, resulting in beneficial indirect effects. Particularly when rural financial markets are
flawed, income from off-farm sources can be partially reinvested in agriculture, thereby
increasing farm production and income as well. A related policy suggestion for this and
comparable situations is that there remains considerable potential for income enhancement
through the direct encouragement of crop and livestock initiatives, which are currently the
primary income sources for the impoverished. In addition to farming, household members
similarly engage in off-farm activities to enhance their livelihoods.
The research demonstrates that the majority of households participating in off-farm activities
derived their average annual household income from crop and livestock farming in the study
region. Off-farm activities contributed to self-employment and the provision of labor to
supplement household income. Some challenges hindering farming and off-farm activities
include: insufficient credit facilities, inadequate knowledge of improved technology, limited
market access, high raw material costs, discrimination, and low social status within the
community, among others. The findings of the study further indicated that households engaged
in off-farm activities experience improved conditions and better living standards due to the
additional income generated from these activities.

5.2 RECOMMENDATIONS
Drawing upon the study's findings, the following recommendations are proposed:
We suggest that rural development policies aimed at alleviating poverty should concentrate on
both on-farm and off-farm sectors, as off-farm activities have been elevating household income
and reinvested into agricultural production. Therefore, we advocate for a policy that employs an
off-farm participation strategy to ensure the sustainability of farmers' food security through
educational services, recognition and support of special skills, and the provision of credit
facilities as well as infrastructure like roads and electricity that improve alternative livelihood
opportunities for farming households involved in off-farm activities while alleviating their
challenges, thus enhancing their food security status. Household members should be motivated to
engage in off-farm activities to boost their income and improve their living standards. The
establishment of accessible credit schemes can aid in creating off-farm enterprises. Off-farm
activities should be diversified, and rural households adequately informed about their benefits to
livelihoods. The constraints identified by households should be addressed by all stakeholders to
enhance the living conditions of the households.
Page 31
 REFERENCE
Adem, M. (2018).Determinants of Income Diversification and Its Effect on Food Security of
Small Holder Farmers: The Case of AsayitaWoreda, Afar Regional State. MSc. thesis in
Agricultural Economics.

Asnake, W. (2010). Participation into off-farm activities in rural Ethiopia: who earns more? ma
thesis. Erasmus University: Hague.

Berhane, Z. (2010). Rural non-farm diversification in Ethiopia: what determines participation

and returns? Centre for African and oriental studies. Addis Ababa University.

Beyene (2008), Determinants of off-farm participation decision of farm household in Ethiopia

College of Business and Economics, Mekelle University, P.O.Box 450, Mekelle, Ethiopia

CSA (2007), Angacha District data office about agricultural product

Innocent Matshe, Trevor Yong (2002), Off-farm labor allocation decision in small scale rural
household in Zimbabwe

Mbisso, (2011), Petty trading in market places: space denegation, use and management at
TemelseSereo Market place in Dare Salam, Tanzania, and Department of Architecture Chalmers
University of Technology

MoARD.(2010).Ministry of agriculture and rural development of Ethiopia 2010.Annual Progress

report.

NorsidaMan (200), Factors that affecting the decision making in on off-farm employment among
Paddy farmers in KemasinSemerek in Malaysia

TesfayeWoldeyohannes (2009), Effect off-farm income on smallholder commercialization:

Panel evidence from rural households in Ethiopia

Zerai, B., &Gebereegziabher, Z. (2011). Effects of Non-Farm Income on Households Food

Security in Eastern Tigray, Ethiopia: An Entitlement Approach: Food Science and Quality
Management, 2224–6088 (Paper) 2225–0557 (Online) Vol.1

Zewdie, E., &Sivakumar, S. (2017). Determinants of off farm participation of rural farm
households in shebedino district of sidama zone, southern Ethiopia. International Journal of
Development Research, 07(9), 15157–15165. ISSN: 2230-9926

Tewodros, T., &Fikadu, T. (2014). Determinants of households food security and coping
strategies for food shortfall in mareko district. Guraghe Zone

Page 32
Southern Ethiopia. Journal of Food Security, 2(3), 92– 99.https://doi.org/10.12691/jfs-2-3-4
2014

Tithy, D., Naznin, S., & Md., E. H. (2016). Analysis of the impact of income diversification
strategies on food security status of rural households in bangladesh: a case study of Rajshahi
district.

American Journal of Theoretical and Applied Business, 2(4), 46–56. https://doi.org/10.11648/j.

ajtab.20160204.13

World Bank. (2008). ‗World development report 2008: agriculture for development‘, the
International Bank for Reconstruction and Development.

Berhane, Z. (2010). Rural non-farm diversification in ethiopia: what determines participation and
returns? centre for african and oriental studies. Addis Ababa University.

Yemane, T. (1967).Statistics: An Introductory Analysis (2nded.). Harper and Row Inc.

Yizengaw, B. (2014). Determinants of household income diversification and its effect on food
security status in rural ethiopia: evidence from ethiopia rural household survey. In MSc in
economic policy analysis (pp. 15–30). Addis Ababa, Ethiopia: Addis Ababa

University.

Zeleke, T., Jemal, Y., & Lemma, Z. (2017). Impact of livelihood diversification on rural
households‘ food security in fedisweroda, oromiya regional state, ethiopia. journal of Poverty.
Investment and Development, 32,36-37. ISSN 2422-846X

Zerai, B., &Gebereegziabher, Z. (2011).Effects of Non Farm Income on Households Food

Security in Eastern Tigrai, Ethiopia: An Entitlement Approach: Food

Science and Quality Management, 2224–6088 (Paper) 2225–0557 (Online) Vol.1

Zewdie, E., &Sivakumar, S. (2017). Determinants of off farm participation of rural farm
households in shebedino district of sidama zone, southern Ethiopia.

International Journal of Development Research, 07(9), 15157–15165. ISSN: 2230-9926

2007 Population and Housing Census - Amhara Region Statistical Report.

Central Statistical Agency of Ethiopia (CSA). (2021). Agricultural Sample Survey 2020/2021:
Report on Area and Production of Major Crops (Private Peasant Holdings, Meher Season).
CSA.

Page 33
APPENDEX1
. logit ofp sex Age edu dep mip Farm Ani Nli

Iteration 0: log likelihood = -133.24671

Iteration 1: log likelihood = -68.967882
Iteration 2: log likelihood = -67.405952
Iteration 3: log likelihood = -67.364724
Iteration 4: log likelihood = -67.364669
Iteration 5: log likelihood = -67.364669

Logistic regression Number of obs = 202

LR chi2(8) = 131.76
Prob > chi2 = 0.0000
Log likelihood = -67.364669 Pseudo R2 = 0.4944

ofp Coef. Std. Err. z P>|z| [95% Conf. Interval]

sex -.5071946 .6099328 -0.83 0.406 -1.702641 .6882518

Age -.1142955 .0218586 -5.23 0.000 -.1571375 -.0714535
edu .9809372 .4817539 2.04 0.042 .036717 1.925157
dep -.4912017 .2327248 -2.11 0.035 -.947334 -.0350695
mip -.2308746 .180283 -1.28 0.200 -.5842228 .1224736
Farm -.207241 .1046499 -1.98 0.048 -.412351 -.002131
Ani -.008529 .0349808 -0.24 0.807 -.0770901 .0600321
Nli -.0000867 .0000199 -4.35 0.000 -.0001257 -.0000476
_cons 7.710862 1.19723 6.44 0.000 5.364333 10.05739

. tab ofp

ofp Freq. Percent Cum.

0 75 37.13 37.13
1 127 62.87 100.00

Total 202 100.00

. tab sex

sex Freq. Percent Cum.

0 85 42.08 42.08
1 117 57.92 100.00

Total 202 100.00

Page 34
. tab Age

Age Freq. Percent Cum.

21 8 3.96 3.96
22 2 0.99 4.95
23 17 8.42 13.37
24 22 10.89 24.26
25 6 2.97 27.23
27 18 8.91 36.14
28 17 8.42 44.55
29 6 2.97 47.52
30 6 2.97 50.50
31 12 5.94 56.44
32 2 0.99 57.43
34 2 0.99 58.42
35 2 0.99 59.41
37 4 1.98 61.39
39 4 1.98 63.37
41 2 0.99 64.36
42 2 0.99 65.35
43 4 1.98 67.33
44 4 1.98 69.31
45 2 0.99 70.30
47 6 2.97 73.27
48 7 3.47 76.73
49 6 2.97 79.70
50 2 0.99 80.69
51 10 4.95 85.64
55 3 1.49 87.13
56 4 1.98 89.11
58 4 1.98 91.09
61 4 1.98 93.07
63 2 0.99 94.06
64 2 0.99 95.05
67 2 0.99 96.04
71 2 0.99 97.03
73 6 2.97 100.00

Total 202 100.00

. tab dep

dep Freq. Percent Cum.

0 78 38.61 38.61
1 47 23.27 61.88
2 57 28.22 90.10
3 16 7.92 98.02
4 2 0.99 99.01
5 2 0.99 100.00

Total 202 100.00

Page 35
. tab mip

mip Freq. Percent Cum.

0 72 35.64 35.64
1 70 34.65 70.30
2 31 15.35 85.64
3 20 9.90 95.54
4 7 3.47 99.01
5 1 0.50 99.50
9 1 0.50 100.00

Total 202 100.00

. tab Farm

Farm Freq. Percent Cum.

0 16 7.92 7.92
1 41 20.30 28.22
2 48 23.76 51.98
3 12 5.94 57.92
4 6 2.97 60.89
5 19 9.41 70.30
6 24 11.88 82.18
7 14 6.93 89.11
8 18 8.91 98.02
9 4 1.98 100.00

Total 202 100.00

. tab Ani

Ani Freq. Percent Cum.

0 5 2.48 2.48
1 16 7.92 10.40
2 13 6.44 16.83
3 6 2.97 19.80
4 10 4.95 24.75
5 10 4.95 29.70
6 22 10.89 40.59
7 14 6.93 47.52
8 24 11.88 59.41
9 5 2.48 61.88
10 4 1.98 63.86
11 7 3.47 67.33
12 2 0.99 68.32
13 3 1.49 69.80
14 5 2.48 72.28
15 5 2.48 74.75
16 3 1.49 76.24
17 11 5.45 81.68
18 5 2.48 84.16
19 7 3.47 87.62
20 2 0.99 88.61
21 7 3.47 92.08
22 1 0.50 92.57
23 8 3.96 96.53
24 1 0.50 97.03
25 1 0.50 97.52
27 2 0.99 98.51
29 1 0.50 99.01
30 2 0.99 100.00

Total 202 100.00

Page 36
. tab Nli

Nli Freq. Percent Cum.

0 11 5.45 5.45
1000 11 5.45 10.89
1400 1 0.50 11.39
1900 1 0.50 11.88
2000 2 0.99 12.87
2100 4 1.98 14.85
2400 2 0.99 15.84
2500 6 2.97 18.81
2800 2 0.99 19.80
2900 2 0.99 20.79
3000 3 1.49 22.28
3400 1 0.50 22.77
3410 1 0.50 23.27
3500 4 1.98 25.25
3800 2 0.99 26.24
4000 1 0.50 26.73
4100 2 0.99 27.72
5000 10 4.95 32.67
7000 2 0.99 33.66
8000 1 0.50 34.16
8500 2 0.99 35.15
9000 1 0.50 35.64
10000 11 5.45 41.09
11000 8 3.96 45.05
12000 4 1.98 47.03
12500 3 1.49 48.51
13500 2 0.99 49.50
14000 2 0.99 50.50
15000 3 1.49 51.98
17000 9 4.46 56.44
19000 3 1.49 57.92
21000 4 1.98 59.90
22500 3 1.49 61.39
23000 5 2.48 63.86
24000 6 2.97 66.83
25000 17 8.42 75.25
27688 2 0.99 76.24
28000 3 1.49 77.72
29000 10 4.95 82.67
30000 1 0.50 83.17
31500 3 1.49 84.65
32000 7 3.47 88.12
33000 5 2.48 90.59
34100 1 0.50 91.09
35000 4 1.98 93.07
35122 4 1.98 95.05
35500 3 1.49 96.53
37000 5 2.48 99.01
41000 2 0.99 100.00

Total 202 100.00

Page 37
 APPENDIX2
. logistic ofp sex Age edu dep mip Farm Ani Nli

Logistic regression Number of obs = 202

LR chi2(8) = 131.76
Prob > chi2 = 0.0000
Log likelihood = -67.364669 Pseudo R2 = 0.4944

ofp Odds Ratio Std. Err. z P>|z| [95% Conf. Interval]

sex .6021826 .3672909 -0.83 0.406 .1822017 1.990233

Age .8919943 .0194977 -5.23 0.000 .8545866 .9310396
edu 2.666955 1.284816 2.04 0.042 1.037399 6.856228
dep .6118906 .1424021 -2.11 0.035 .3877735 .9655383
mip .793839 .1431157 -1.28 0.200 .557539 1.130289
Farm .8128238 .0850619 -1.98 0.048 .6620919 .9978713
Ani .9915073 .0346837 -0.24 0.807 .9258064 1.061871
Nli .9999133 .0000199 -4.35 0.000 .9998743 .9999524
_cons 2232.465 2672.774 6.44 0.000 213.6488 23327.53

Note: _cons estimates baseline odds.

. reg ofp sex Age edu dep mip Farm Ani Nli

Source SS df MS Number of obs = 202

F(8, 193) = 27.04
Model 24.9193133 8 3.11491416 Prob > F = 0.0000
Residual 22.2341521 193 .11520286 R-squared = 0.5285
Adj R-squared = 0.5089
Total 47.1534653 201 .234594355 Root MSE = .33942

ofp Coef. Std. Err. t P>|t| [95% Conf. Interval]

sex -.0236469 .0645459 -0.37 0.714 -.1509528 .1036589

Age -.0161618 .0021774 -7.42 0.000 -.0204563 -.0118673
edu .0984257 .0509874 1.93 0.055 -.0021383 .1989898
dep -.0586174 .0228343 -2.57 0.011 -.1036543 -.0135806
mip -.0289307 .0206873 -1.40 0.164 -.069733 .0118715
Farm -.0248264 .0126914 -1.96 0.052 -.0498581 .0002053
Ani -.0024114 .0039384 -0.61 0.541 -.0101793 .0053564
Nli -.0000104 2.04e-06 -5.09 0.000 -.0000144 -6.37e-06
_cons 1.554792 .0994348 15.64 0.000 1.358673 1.75091

. hettest

Breusch-Pagan / Cook-Weisberg test for heteroskedasticity

Ho: Constant variance
Variables: fitted values of ofp

chi2(1) = 0.96
Prob > chi2 = 0.3270

Page 38
. corr
(obs=202)

sex Age edu dep mip Farm Ani ofp Nli

sex 1.0000
Age -0.3956 1.0000
edu 0.5194 -0.3115 1.0000
dep -0.0044 0.1324 -0.0768 1.0000
mip 0.2265 -0.2331 0.1405 0.1710 1.0000
Farm -0.4796 0.5638 -0.3018 0.1757 -0.2866 1.0000
Ani -0.0938 0.3768 -0.1894 0.0777 -0.2378 0.4472 1.0000
ofp 0.2790 -0.6469 0.2886 -0.2388 0.0803 -0.4682 -0.2689 1.0000
Nli -0.0112 0.2024 0.0350 -0.0246 -0.0355 0.0845 -0.0749 -0.3582 1.0000

. mfx

Marginal effects after regress

y = Fitted values (predict)
= .62871287

variable dy/dx Std. Err. z P>|z| [ 95% C.I. ] X

sex* -.0236469 .06455 -0.37 0.714 -.150155 .102861 .579208

Age -.0161618 .00218 -7.42 0.000 -.020429 -.011894 36.802
edu .0984257 .05099 1.93 0.054 -.001508 .198359 .623762
dep -.0586174 .02283 -2.57 0.010 -.103372 -.013863 1.12376
mip -.0289307 .02069 -1.40 0.162 -.069477 .011616 1.15842
Farm -.0248264 .01269 -1.96 0.050 -.049701 .000048 3.53465
Ani -.0024114 .00394 -0.61 0.540 -.010131 .005308 9.82673
Nlisex=.5792079
. ttest -.0000104 .00000 -5.09 0.000 -.000014 -6.4e-06 16165.7

(*) dy/dx is for discrete change of dummy variable from 0 to 1

One-sample t test

Variable Obs Mean Std. Err. Std. Dev. [95% Conf. Interval]

sex 202 .5792079 .0348219 .4949128 .5105447 .6478711

mean = mean(sex) t = 0.0000

Ho: mean = .5792079 degrees of freedom = 201

Ha: mean < .5792079 Ha: mean != .5792079 Ha: mean > .5792079
Pr(T < t) = 0.5000 Pr(|T| > |t|) = 1.0000 Pr(T > t) = 0.5000

. estat gof, outsample

Logistic model for ofp, goodness-of-fit test

number of observations = 202

number of covariate patterns = 182
Pearson chi2(182) = 151.64
Prob > chi2 = 0.9509

Page 39
 APPENDEX1

DEBARK UNIVERSITY
COLLEGE OF BUSINESS AND ECONOMICS
DEPARTMENT OF ECONOMICS
On the research title determinant of factors that affect off farm participation
decision of farm household

Dear, respondents this is questionnaire prepared for you on the research title (determinant of
factors that affect off farm participation decision of farm household) from debark university,
business and economics college, department of economics undergraduate students Terefe
Bergene, Atkilt Belachew And Biniam Assefa

Questionnaires

1. Off-farm Participation

Do you participate in any off-farm activities?

Yes

No

If no, what are the main reasons for not participating in off-farm activities?
Please specify:

2. Gender of Household Head

What is the gender of the household head?

Male

Female

Do you think the gender of the household head influences participation in off-farm activities?
Please explain:

Page 40
3. Age of Household Head

What is the age of the household head…………..?

How does the age of the household head affect their decision to participate in off-farm activities?
Please explain:

4. Size of Cultivated Land

How many hectares of land does your household cultivate……………….?

How does the size of your cultivated land affect your participation in off-farm activities?
Please explain:

5. Education Level of Household Head

What is the highest level of education achieved by the household head?

Lower secondary and preparatory education

Diploma

Degree and above

How does education level influence participation in off-farm activities?

Please elaborate:

6. Non-labor Income

Does your household receive any non-labor income (e.g., pensions, government transfers,
remittances)? how much with in year…………………?

how does this income affect your decision to participate in off-farm activities?
Please explain:

7. Dependents in the Household

How many dependents (children, elderly, etc.) does your household have…………………..?

Page 41
How does the number of dependents in your household affect participation in off-farm activities?
Please explain:

8. Reason for Participation in Off-farm Activities

What are the main reasons you participate in off-farm activities? (Select all that apply)

To help family

Favorable conditions (e.g., nearby opportunities, good infrastructure)

Limited farm income

Other (Please specify)

If you selected "Other," please explain:

9. Reason for Not Participating in Off-farm Activities

If you do not participate in off-farm activities, what are the main reasons for not participating?
(Select all that apply)

Time inadequacy

Unfavorable conditions

Enough income from farming

Age

Sex

Other (Please specify)

If you selected "Other," please explain:

10. Number of Animals in Household

Page 42
How many animals (livestock) does your household own?-----------

Does owning livestock affect your participation in off-farm activities?

Please explain:

11. Migration from Household

Has anyone from your household migrated to another area or country?

write numbers of migrants…………………………….?

If yes, how much numbers of migrant-----------------,and how does migration influence your household’s
decision to participate in off-farm activities?
Please explain:

Page 43