Khóa luận tốt nghiệp: Impact of human capital on farmer’s decision towards sustainable farming: a meta-analysis

Presently, numerous meta-analysis studies explore the impact of human capital on sustainable agricultural decisions 7⁄2: Anh Nguyen et al., 2022; Lich Hoang-Khac et al., 2021; Hongyun Ha

TOPIC: IMPACT OF HUMAN CAPITAL ON FARMER’S DECISION

TOWARDS SUSTAINABLE FARMING: A META-ANALYSIS

Lecturer : PhD Nguyen Thi Anh Tuyet

Student : Nguyen The Phong

Class : QH2020E KT CLC 3

Training System : CLC

Hanoi, October 31, 2023

TOPIC: IMPACT OF HUMAN CAPITAL ON FARMER’S DECISION

TOWARDS SUSTAINABLE FARMING: A META-ANALYSIS

Lecturer : PhD Nguyen Thi Anh Tuyet

Student : Nguyen The Phong

Class : QH2020E KT CLC 3

Training System : CLC

Hanoi, October 31, 2023

by the Faculty of Political Economics throughout the entire process of researching andwriting my thesis on “Impact of Human capital on farmer decision towards sustainable

farming: a Meta-Analysis” The school and the faculty have created optimal conditions,demonstrating deep care for my academic journey

Foremost, I extend my heartfelt thanks to my mentor, Dr Nguyen Thi Anh Tuyet,

who guided and supported me during the completion of my graduation thesis Herunwavering dedication significantly contributed to the quality of my work I am also

grateful to other teachers who imparted essential knowledge, laying the foundation notonly for my thesis but also for my future endeavors

I extend my appreciation to the board of directors at the University of Economics

- Vietnam National University, Hanoi, and the various departments for providing thenecessary resources and facilities These conducive conditions offered me the

opportunity and environment to study and apply theoretical knowledge in practice

Despite my efforts, the limitations of my knowledge and theoretical abilities are

reflected in certain shortcomings within the thesis I eagerly anticipate constructivecontributions from my teachers to enhance the completeness of my graduation thesis

In conclusion, I would like to express my best wishes for the good health and continuedsuccess of the teachers on the Board of Directors and functional departments at the

University of Economics - Vietnam National University, Hanoi

Once again, I sincerely thank you for your invaluable support!"

Hanoi, October 14 ,2023

StudentPHONG NGUYEN THE

LIST OF ABBREVIATIONS

INTRODUCTION

1.Background of the study

2.Research gap

3.Objective of the study

4.Objective and scope of the study

5.Structure of the topic

3.4 Estimation results of the weighted fractional regression with interaction terms

between three traits of human capitals

CHAPTER 5: POLICY IMPLICATION

CONCLUSION

REFERENCES

LIST OF CHARTS AND TABLES

Table 2.1 Keyword and online search strategy

Table 2.2 Type of sustainable agriculture techniques/practices

Table 3.1 Definition and descriptive statistics

Table 3.2 Estimation results of the weighted fractional regressions with

standardized,unstandardized variables and their marginal effects

Table 3.3 Interaction of educ_exp, exten_exp, educ_ext

LIST OF IMAGES

Figure 2.1 PRISMA flow diagram of data collection

Figure 2.2 Symmetrical funnel plot of adoption rate of sustainable agriculture

1.Background of the study

Agriculture stands as a pivotal and indispensable industry for human life, wielding

a decisive role in societal development and survival In the current global context, peopleworldwide grapple with extreme weather conditions, crop failures due to climate change,and pollution of water and air sources, leading to an escalating threat of food insecurity(Masset, 2010), These challenges imperceptibly exert pressure on the agricultural sector,giving rise to numerous complex issues The demand for food is surging due topopulation growth and increased individual consumption (Benjamin et al., 2015) Thisdemand is often met by expanding agricultural production through practices such as land

expansion, routine tillage, and heightened use of inputs like pesticides and chemicalfertilizers to expedite the harvest process (Vaclav Voltr et al., 2021) However, such

practices are unsustainable and inflict substantial damage on the environment Crop,livestock, and aquaculture systems further contribute to water pollution by releasing

chemicals, organic matter, drug residues, sediments, and saline drainage (Naveen Kumar

et al., 2021) With regional and local variations among countries, agricultural productionhas become ensnared in a vicious cycle Environmental degradation resulting from theexternal impacts of traditional agriculture necessitates increasingly unsustainableactivities to maintain production levels and support people's livelihoods Considering

factors like global population growth, economic expansion, and rapid climate change,

the demand for food, energy, and water resources is projected to surge by 40%, 57%,and 70%, respectively, within the next 20 years (Alberto Boretfi & Lorenzo Rosa, 2019)

Consequently, minimizing and reversing the adverse environmental impacts ofagricultural activities on land and water resources emerges as one of the most urgentactions to shield the Earth from degradation

Traditional agricultural methods have been widely employed due to their

accessibility and low costs However, despite this convenience, these methods presentsignificant challenges to the environment, human health, and natural resources (Carla

Costa et al., 2014) Notably, they contribute to issues such as climate change, saltwaterintrusion, land degradation, and water pollution According to a study by the

Environmental Protection Agency (4, 2027), the agricultural sector is responsible for

a quarter of greenhouse gas emissions, predominantly stemming from traditional

practices and activities like rice production, cattle raising, and agricultural landmanagement Despite some modifications in traditional agriculture, these challengespersist (Athanasios Theocharopoulos et al., 2012) Therefore, to address these issues and

strike a balance between current and future needs, prioritizing sustainable agriculture isimperative

According to the US Farm Bill, sustainable agriculture is defined as an "integratedsystem of plant and animal production practices with site-specific applications that, over

the long term, will: meet the food and fiber needs of humans; improve environmentalquality; ensure efficient use of non-renewable and on-farm resources, while integrating

appropriate natural biological controls and cycles; maintain the economic viability offarm operations; and improve the quality of life for farmers and society as a whole."

Similar to the US farm bill, research by (Habib M Alshuwaikhat and Ismaila Abubakar,2008) also indicates that sustainable agriculture involves "management processes that

coordinate with natural processes to conserve all resources, minimize waste andenvironmental impact, prevent problems, and promote agricultural ecosystem resilience,self-regulation, evolution, and sustainable production to feed, nourish, and meet the

needs of all people" (Erik Lichtenberg, 2014) Numerous studies, such as those by (7

Fiisun Tatlidil, Ismet Boz, and Hasan Tatlidil, 2009), as well as (R G Adeola and S I.

Adetunbi, 2015), highlight the significant role of sustainable agriculture in the economy

It serves as a crucial industry that provides raw materials, primarily for the processing

industry Agricultural products become essential input sources for the processingindustry, thereby enhancing the value of agricultural products, especially for export

(Miet Maertens, Bart Minten, Johan Swinnen, 2012) Concerning society, thedevelopment of sustainable agriculture is a specific contribution of farmers to societalprogress (Ezatollah Karami & Marzieh Keshavarz, 2009) It ensures fairness in

development, increases income for farmers, improves the quality of life, addresseshunger, and reduces poverty while narrowing the gap between rich and poor and among

natural resources Research by (Nawab Khan et al, 2021) emphasizes that these

sustainable agriculture types represent a crucial progressive direction for the agriculturalindustry, aiming to establish a system of environmentally friendly food production thatensures food safety and sustainability These methods don't just focus on optimizing crop

productivity and animal husbandry (Ana Trigo et al., 2021) but also place specialemphasis on the impact of agricultural activities on the environment and human health

(Martin Purvis and Richard Smith, 200) These farming methods have gainedpopularity within communities By employing modern technologies and smart

management practices, farmers can achieve stable yields while simultaneouslysafeguarding the environment Studies in the field of sustainable agriculture conducted

by (Carla Azeda et al, 2021) have affirmed that these practices not only enhance crop

productivity but also contribute to the overall performance of the agriculture industry.Most importantly, they aid in environmental protection and ensure that agricultural

products are safe and beneficial for human health Sustainable agriculture is more thanjust a trend; it is a commitment to the future of food systems and the natural environment

In contemporary times, many countries worldwide have recognized the importance

of adopting sustainable agricultural practices, aiming to enhance productivity whilesafeguarding water resources, the environment, and land (J Preity ef al., 2003)

Nevertheless, substantial variations in productivity exist between European and Asiancountries (Hasan A Faruq and Peter J Telaroli, 2011; Lajos Barath and Imre Ferto,

2016)

European Union (EU) countries have a clear focus on agriculture, particularlyemphasizing sustainable practices (Eileen Kennedy et al., 2020) In countries like theUnited States, there is a notable encouragement for the adoption of high technology in

the agricultural sector (K Knickel et al., 2075) The integration of high technology intoagricultural production has significantly supported the industry by reducing working

hours and enhancing the professionalism of farmers (Dmytro Serebrennikoy et al., 2020).One notable technological advancement successfully applied in agriculture is the use of

phone applications for livestock management (Nadia Adnan and her colleagues, 2019).This innovation is transforming the landscape of farming from the practices of the 90s,enabling increased productivity and quality of agricultural products with less manual

labor (Benoit A Aubert et al., 2012) Moreover, it is the government's commitment tosupporting agricultural development that has contributed to the success of the European

agricultural industry Governments in these countries have created favorable conditions

for farmers to thrive by proposing land conservation and maintenance policies and

prioritizing strategic projects to serve the agricultural sector effectively Thesesupportive policies, combined with governmental attention, have propelled the robustdevelopment of European agriculture (Jacqui Dibden et al., 2009) Furthermore, farmers

in European regions exhibit self-reliance, hard work, and a commitment to innovation inagricultural applications (Paolo Gaibazzi, 20/4) They demonstrate proficiency in using

machinery (Julie Ingram, 2008) and possess a strong understanding of both agriculturalpractices and economics (Gunna Breustedt and Thomas Glauben, 2007) This

knowledge is applied to production processes, resulting in reduced labor while increasingoverall productivity Through the diligence and technological acumen of farmers,coupled with the government's sustained focus on agricultural production and the

application of scientific and technical advancements, European agriculture hasestablished itself as a leader in high-tech farming

In contrast to the robust development observed in European countries, sustainableagricultural productivity in regions such as Asia and Africa lags significantly, facing a

host of challenges and difficulties According to (Harsimran Singh et al, 2072), it hasbeen highlighted that effective resource management remains a challenge in countries

like India, leading to land and water depletion, environmental damage, and decliningproductivity (41s K Biswas, 2077) Many farming households in these regions still

adhere to traditional farming methods, contributing to unsustainable exploitation andresulting in resource degradation (Simachew Bantigegn Wassie, 2020) Moreover,

limited access to new technologies compounds the issue Research by (Yakubu B Issakaetal., 2016) indicates that the low adoption of organic agricultural production technology

in Ghana is a major factor contributing to low productivity Farmers' low education

levels are identified as the primary barrier to accessing technology (Evans Sackey Teye

& Philip Tetteh Quarshie, 2021) The educational attainment of farmers in these regions

remains modest, constraining the adoption and practice of high technology (Ben White,2012) Additionally, policies aimed at facilitating people's access to new, high-quality

technologies in these countries are scant Importantly, these policies prove ineffectivewhen high technologies are continually imported without concurrent improvements in

farmers' education levels (Raphael Lencucha et al., 2020), leading to a misallocation ofstate budget resources without resolving the core issues (Guy M Robinson, 2018).Anoteworthy concern is that many countries in these regions are grappling with severe

climate change, yet government policies remain ineffective in addressing this pressing

problem

Hence, there is a pressing need for a systematic study to synthesize and present thedifferences between factors, specifically assessing which factors act as prerequisites forvariations in productivity between regions Presently, numerous meta-analysis studies

explore the impact of human capital on sustainable agricultural decisions (7⁄2: Anh

Nguyen et al., 2022; Lich Hoang-Khac et al., 2021; Hongyun Han and associates, 2023;Cynthia M Creze & William R Horwath, 2021) However, there has been nocomprehensive research on human capital affecting farmers’ decisions, encompassing

their behavior The existing research does not specifically concentrate on thecharacteristics of human capital, resulting in an incomplete overview and a limited

theoretical understanding of this crucial aspect (L.S Prokopy, 2008) Consequently,research on the characteristics of human capital contributing to sustainable farming

decisions among farmers in developing and emerging countries is rare With numerousfactors influencing farmers’ decisions on sustainable agriculture and their productivity

outcomes, farmers often struggle to discern which factors are pivotal, leading toinefficiencies in crop productivity and the ineffectiveness of implementing sustainable

farming practices compared to other countries Hence, my research: “Impact of humancapital on farmer’s decision towards sustainable farming: A meta-analysis” is crucial

in studying key behavioral factors in human capital to draw conclusions that elucidate

the reasons for productivity differences among farmers in regions such as Europe andAsia

Meta-analysis and systematic review methodologies are increasingly prevalent in

various fields The central concept involves statistically combining results from two ormore individual research sources, facilitating evidence-based practices, and resolving

uncertainties in research outcomes This approach relies on empirical insights drawnfrom existing scientific sources—in my case, investigating the specific behavioral

factors of farmers and seeking to explain how this behavior influences decisions related

to sustainable farming, employing regression models to analyze fundamental differences

between studies Therefore, my contribution lies in providing an empirical meta-analysisthat delves into the impact of farmers' personal preferences on the decision to adoptsustainable agricultural practices

2.Research gap

Over the past decades, there have been numerous efforts aimed at enhancing

research on factors influencing the adoption of sustainable farming practices by farmers,particularly focusing on the human capital aspect Academic studies to date share acommon goal of assessing the factors affecting farmers’ adoption of sustainable farming

practices (Elisa M et al., 2021; Melanie et al., 2021; Haijun, B et al., 2018) However,these studies have not specifically concentrated on the characteristics of human capital,

and they have not synthesized which factors are pivotal in enhancing or hinderingfarmers’ sustainable agricultural practices, leading to an incomplete and theoretically

limited understanding of the mechanisms and reasons behind these decisions (Daniel, C

et ai, 2022) These assessments are fragmented across disciplines (Hasmin, T &

Yusriadi, Y., 2022), except for the research by (Adam B., Linda, S P., & Flores, K.,2012), which dates back over a decade

Moreover, previous research on sustainable agriculture has predominantly relied

on foreign literature (Dessart, 2079), focusing on countries with significant experienceand advancements in agricultural revolution Previous studies have primarily addressed

influencing factors without delving into specific characteristics of human capitalaffecting the decision to adopt sustainable farming practices Therefore, research on thebehavioral factors within human capital influencing decisions to practice sustainable

agriculture in developing and emerging countries is scarce The variations in landownership, climate, culture, and farming practices necessitate adjustments to research

factors while exploring new elements to align with the study population Overall, pastresearch has often concentrated on specific methods such as organic farming, precision

agriculture techniques, and green control technologies However, these are just parts ofthe larger picture of farmers' decisions The proposed study seeks to expand the scope

by examining a more diverse set of factors and developing a comprehensive concept

The research will not only explore the impact of human capital but will alsoinvestigate the characteristics of these human capital factors on farmers’ decisions,

aiming to gain a deeper understanding of their interactions Distinct factors such as riskpreferences, time preferences, and loss aversion will provide a more holistic view of

individual decisions made by farmers Importantly, the study will not focus solely onindividual aspects but will also attempt to connect these findings into an overarching

framework This approach allows us to build a stronger evidential base, clarifying notonly the critical factors for sustainable agriculture adoption but also decoding thecomplex interconnections between them My research not only contributes to advancing

knowledge but also serves as a vital resource for policies and support strategies In the

face of increasingly complex challenges in sustainable agriculture, this study iscommitted to providing detailed insights into the significant factors influencing thecomprehensive and sustainable development of the agricultural sector

3.Objective of the study

This research aims to identify human capital factors and examine the behavioral

characteristics within human capital that influence decisions to adopt sustainable

agricultural practices The objective is to propose appropriate policy implications to

increase the adoption rate of sustainable agricultural practices among farmers This, inturn, contributes to addressing global food challenges, protecting environmental

resources, mitigating climate change, and producing safe and healthy agriculturalproducts for consumers

To achieve the aforementioned goals, the study needs to accomplish the followingspecific objectives:

e Firstly, systematize the existing knowledge base on factors influencing decisions

to practice sustainable agriculture Develop a new theoretical framework onhuman capital factors and behavioral elements within human capital

e Secondly, examine publication bias in research to ensure the representativeness

and integrity of published data, avoiding the formation of biased understandings

of results

e Thirdly, Identify the most significant factors contributing to and hinderingindividuals in adopting sustainable agricultural practices through regression

analysis in a Meta-analysis model

e Lastly, based on the determined results, propose potential policies for relevant

stakeholders such as government entities, non-governmental organizations, and

policy planners to encourage farmers to increase the adoption rate of sustainable

agricultural practices

4.Objective and scope of the study

About the research subject: The study focuses on analyzing human capital factors andbehavioral characteristics within human capital that influence decisions to adopt

sustainable agricultural practices in agricultural countries, primarily concentrating on theMonsoon, Savannah, Rainforest, and Temperate climate zones

For research scope:

- About sapce: The primary data are sourced from published articles available on

research platforms such as Web of Science and Google Scholar

- About time: The study is conducted within the timeframe from July 2023 to October

2023

5.Structure of the topic

This research includes 03 main parts: Introduction, Content and Conclusion

In the content, the topic specifically includes 4 chapters as follows:

- Chapter 1: Theoretical basis

- Chapter 2: Data and methods

- Chapter 3: Results and Discussion

- Chapter 4: Policy implication

In a recent study, researchers synthesized experience and knowledge, applying

sustainable agriculture principles in rural communities from a human capital perspective.They explored the impact of enhancing efficiency and effectiveness in sustainable

agriculture through education and training (Raidimi E N & Kabiti HH M., 2019) Otherstudies have revealed a correlation between years of education and production costs in

agriculture (Kirtti Ranjan Paltasingh & Phanindra Goyari 2018) analyzed the inverserelationship between total production costs and farmers' years of education, suggesting

that as the number of schooling years increases, the real cost of agricultural productionproportionally decreases Educational variables, often measured by years of schooling

or level of training, are commonly used to assess farmers' educational attainment in

practice (Paula T Ross, Tamera Hart-Johnson, Sally A Santen & Nikki L Bibler Zaidi,

2020; Idris Géksu, Naif Ergiin, Zafer Ozkan, Halis Sakiz, 2027) Additionally, studies

have delved into the impact of human capital, such as specialized knowledge and skills,

on efficiency and productivity in adopting sustainable agriculture (Barbara Wieliczko

ORCID and Zbigniew Florianczyk, 2022) The results consistently indicate that humancapital positively influences sustainable agriculture performance Investing in education

can thus facilitate the success and development of sustainable agriculture, increaseproductivity, and yield sustainable benefits for rural communities Therefore, thehypothesis I propose is:

Hypothesis 1:(+) Farmers' educational level has a positive impact on people's

intention to do sustainable agriculture

Experience gained from years of working in traditional agriculture can act as adeterrent when transitioning to sustainable agriculture Farmers accustomed to

traditional methods may be hesitant to embrace change, given their comfort withpractices they have employed for decades (O Nywmba et al., 2078) The reluctance oftenstems from a fear of the risks associated with testing new methods (O Nywmba et al.,

2018) However, experience can also be a valuable resource in the move towards

sustainable agriculture Farmers with extensive experience can leverage their knowledge

and skills from traditional agriculture to adopt more systematic and sustainable farming

methods (Renata Anibaldi et al., 2021) Moreover, their familiarity with localcooperatives, understanding of local conditions, and established relationships within thecommunity can contribute to a more effective adaptation to sustainable agriculture,

thereby enhancing its success

Research by (Liese Coulter et al., 2079) also indicates that individuals learn frompast experiences and adapt to climate risks (Anja Riihlemann & Joanne C Jordan, 2019).This suggests that households with more experience in climate change are less likely to

be adversely affected by droughts and natural disasters, particularly by less severedroughts (Anja Riihlemann & Joanne C Jordan, 2019) Furthermore, research by (LieseCoulter et al 2019) demonstrates that, over time, individuals accumulate extensiveexperience in preventing and addressing problems and risks in agriculture moreeffectively Experienced farmers become adept at dealing with diseases and pests, and

research by (C Svensson et al., 2018) indicates that they can optimize production bychoosing suitable crop varieties, using fertilizers and pesticides effectively, and

managing water and land resources (1o Li eí al., 2020) Therefore, experiential variableswithin human capital play a crucial role in shaping individuals' decisions to embracesustainable agriculture by enabling them to learn and adapt to changing environmental

conditions in the field Therefore, the hypothesis I make is:

Hypothesis 2:(+) The experience of farming households has a positive impact onpeople's intention to do sustainable agriculture

Professional training programs in the field of sustainable agriculture are commonly

designed and implemented by both governmental and non-governmental organizations.Technical consulting courses and programs for farmers aim to develop human resourcesand enhance understanding of sustainable agricultural methods and techniques (Burton

E Swanson, 2008) Numerous studies have focused on estimating the impact of

agricultural extension programs on the adoption of sustainable agriculture For instance,

in the study by (/ Fusun Tatlidil et al., 2009), the authors examined provinces in Turkeyand assessed the impact of extension courses and promotional programs on agricultural

development Their findings revealed that farmers who received training throughextension programs positively influenced productivity in all regions studied

Additionally, (Lori Beaman et ai, 2027) highlights that farmer households regularlyparticipating in agricultural extension programs often broaden their connections within

agricultural networks This expanded network facilitates the sharing of experiences andmutual support in agricultural projects (Gloria Otieno et di, 2021) Moreover,agricultural extension programs emphasize sustainable agricultural methods and

environmental protection (R.H Khwidzhili & S Worth, 2079), encouraging participants

to adopt practices that minimize pollution, protect water resources, and preserve land

(R.H Khwidzhili & S Worth, 2019) Therefore, the hypothesis that I make after analysis:

Hypothesis 3: (+) Extensive training sessions have a significant and positive impact

on farmers' intention to do sustainable agriculture

Moreover, research by (Renate Strobl ,2002) suggests that specific personalcharacteristics, such as risk avoidance behavior or loss aversion, guide individuals inmaking safer investment decisions However, (Jim Yong Kim, 20/8) has indicated that

individuals with higher risk levels tend to invest less in human capital This is becauserisk-averse individuals are less inclined to accept the uncertainties associated with

investing in education and skills development, doubting that participating in agriculturalextension programs offered by local or other organizations would enhance their

productivity (Murari Suvedi et al., 2017) In contrast, individuals with lower levels ofrisk avoidance are more likely to invest in human capital Research by (Kira A Sullivan-Wiley & Anne G Short Gianotti, 2018) has revealed that the productivity of farming

households with lower levels of risk avoidance is generally higher than that ofindividuals with a high level of risk avoidance

Previous meta-analysis studies have not thoroughly explored and appear to haveoverlooked the interrelationship between farmers' personality traits and human capital

Given the global threat to food security, especially in developing and developedcountries due to increasing food demand, understanding personality traits in human

capital, particularly risk aversion and loss aversion, can enhance agriculturalproductivity This insight is crucial for policymakers and governments to formulatesolutions that address the challenges in agriculture and promote a better understanding

of personality traits among the populace Therefore, the hypothesis we make is:

Hypothesis 4: (+) Behavioral factors in human capital have a positive impact on

farmers’ decisions to do sustainable agriculture

CHAPTER 2: RESEARCH METHOD

2.1 Data collection

Table 2.1 Keywords and online search strategy

Not Numbers of potentially relevant studie:

1 “tisk prefernce” OR “risk aversion” OR “risk loving” OR “risk neutral”

2 “loss aversion”

3 “time preference” OR “time pref” OR “future bias” OR “present bias”

A “organic farming” OR “sustainable famring “conservation farming” OR “conservation agriculture” OR

“eco-friendly farming” OR “permaculture farming” OR “agroecology farming” OR “environmental farming” OR “crop diversification”

B “clean manufacturing technologies” OR “green chemistry” OR “water conservation technologies” OR

“waste management technologies” OR “soi management technologies” OR “improved variety technology” OR “water-saving irrigation technology” OR “soil testing and formulated fertilization

technology” OR “rainwater harvesting and supplementary irrigation technology”

Cc “green control techniques” OR “ eco-friendly soil-management practices” OR “diversification

practices” OR “higher-yielding varieties seeds”

Data ‘Search straegy Number | Numbers of | Extend search with Web | Scarches updated (Performed source of potentially of science and other on 25th October 2023)

studies | relevant studies sources (duplicated

removed) Numbers of | Number of

non-article and working papers

Total after removing articles that do 129 14

not have enough information to

caculate effect size

total numbers of studies after removing duplicates and not related to sustainable farming at all.

Firstly, I researched relevant databases to investigate the relationship between

individual characteristics and sustainable agriculture decisions I started searching forsome basic keywords like: “risk aversion”, “loss aversion”, “time preference” and

“sustainable farming” in Google Scholar and Web of Science The selection of articlessuitable for the research direction is carried out according to the PRISMA step process

to conduct meta-analysis to minimize bias in selected studies throughout or within the

scope of the study ( David Moher et al., 2009) PRISMA is a set of 27 checklist itemsintended to guide authors in reporting systematic reviews or meta-analyses One of theadvantages of PRISMA compared to previous meta-analysis studies is that it raisesauthors' awareness of the risk of bias in the studies they use in their articles

After searching for related research articles, I discovered 1008 related articles onkeywords and titles, mainly illustrated in (Figure 2.1) The steps to implement PRISMA

I performed are as follows: The first exclusion step is called screening the articles found,searching through some of the following criteria: written in English, published articles,

research on personal interests, priorities time horizons, farmers' loss aversion, andsustainable agricultural practices The second exclusion step, called eligibility, was

performed for three reasons, namely removing duplicate articles (22 studies) and missingvariables risk aversion and adoption rate (789 studies) In the process of searching for

keywords related to time preference because there are too few research articles withenough of this variable, so my research focuses mainly on risk aversion and loss

aversion Finally, I came to the conclusion that I drew the 143 most relevant studies,including 314 observations in my research, fully including the following criteria: written

in English, research on the application of scientific methods sustainable farming

methods, published articles, full data for reports and experimental research in

agricultural countries

Figure 2.1 PRISMA flow diagram of data collection.

Title/AbstractKeywords approach:

Total: (n=1008)

Sustainable Farming.

“Article title at Duplication, “Articles removed after ‘Contains Individual preference: Risk

: — — Nga aversion/Loss aversion/Time preferenc

EỊ (n=986) Empirical research and case location in

5

agriculture country.

&

‘The research reports the sample size (n)

and regression coefficient (b) and mean of

+ Aatcle not satisying eligibility CC đi preference, adoption (SEs

Abstract/Keywords approach criteria (Missing risk aversion là nhan dan

Total: (n=964) 7) and adoption rate)

¬_:-Exclusion criteria:

(a=789) #Nen-empirical studies and case are

2 ‘uncertain

EỊ + ‘vague variables in the original study

Full articles satisfied cligibility Article excluded after reading a

criteria after reading main body >) body (n =32) h h itoat) main body (n =3) models (ie, not using probit,

to observe bias in a meta-analysis According to the study of (Christopher J Weir et al.,

2018), I use median values, median ranges, and sample sizes

In the agricultural sector, the combination of education, experience and openness

to sustainable decision-making plays a key role in shaping a more environmentallyfriendly agricultural sector (Enrica Chiappero- Martinetti & Anna Sabadash, 2014)

Research articles on human capital will often have individual reports on variables such

as education, experience and expansion, or a combination of two or three of these

variables to estimate possible uses in my meta-analysis Many reports and studies usehuman capital as a deciding factor in adopting sustainable farming methods, but the

question is that in a model analyzing the determinants of efficiency, whether their studyuse at least one of these three factors? Therefore, to capture the main effects described

in hypotheses 1, 2 and 3, I used level of education, experience’s farmer, and extension

to see whether one of the three variables used in aggregated studies, there is a significantand positive impact of human capital factors on people's intention to do sustainable

agriculture

In contrast to the above three hypotheses, I build interactions between personality

traits of human capital, according to (Francois J Dessart et al., 2019), arguing that traitssuch as psychological anxiety farmers' risk aversion or loss aversion is expected to make

them more inclined to adopt more systematic and environmentally friendly farmingmethods Because they believe that using advanced farming methods will avoid more

risks and losses than conventional farming methods In analyzing the relationshipbetween personality traits and AD, linear regression analysis is considered appropriate

in analyzing the empirical model (Douglas C Montgomery et al., 2021) According to(A K Wagner PharmD et al., 2002) segmented regression models are also used to studythe effects of determinants

Besides the factors of human capital and its characteristics, i also try to determinethe impact of other important factors affecting farmers’ AD, such as arable land area,

land area, etc farmer's land, research location, methods and techniques used, technical

specifications and number of observations (1 R Raupach et al., 2005) According to

(Xiao Lyu et al., 2021), forms of sustainable agriculture are systematically classified,they are consistent with the criteria of comparing AD between methods or types offarming with together From there, it also helps us see the forms of sustainable farming

that are currently favored by farmer households According to my research, afterinvestigating from 143 articles including 314 observations, forms of sustainable

technology application account for the most (46.85%), ranked second is sustainable cropmanagement methods (23.77 %), third is crop diversification (14.68%), followed by

other forms of sustainable agriculture such as Organic farming (8.4%) andEnvironmental Practices (6.3%), specific measures are presented in (Table 2.2) (U/laKou Griffiths, Rosa Legood, Catherine Pitt, 2016) has provided criteria to distinguish

the income level of each country, showing the wealth of economic conditions of theresearch location in my analysis It helps us know whether geographical variation has an

experimental effect on ADs In empirical analyzes of the implementation of generalbehavior change measures, models such as logit are widely used (/redric J Janzen &

Hal S Stern, 1998) The logit model uses the logistic function to model relationshipsbetween a dependent variable and one or more independent variables (Carian Mood,

2009), especially it is very suitable for binary data types

Table 2.2 Type of sustainable agriculture techniques/practices

Type Sub-type of sustainable agriculture Count Percentage (/total

practices (papers) founded paper)

1 Organic farming - Organic drystock farming 12 8.4%

- Green control techniques

- Soil management practices

- Sustainable land management

- Best management practices

- Improve fertilizer use

- Sustainable intensification

- Water conservation measures

- Reduce fertilizer input

- Climate-smart conservation

agriculture

- Crop insurance programmes

- Agriculture green development

3 Crop diversification - Higher-yielding varieties seeds 21 14.68%

- Plan new crop varieties

- Mixed cropping

- Improved sunflower varieties

- Drought tolerant maize varieties

- Hybrid maize seed

- Improved maize seed

- Improved maize varieties

- Modern rice varieties

- Improved rice varieties

- Stress-tolerant rice varieties

- Intercropping

- Straw return

4 Environmental - Water saving 9 6.3%

Practices - Cleaner production

- Pro-Environmental agriculture

- Eco-friendly agriculture

production

- Eco-friendly agriculturalpractices

- Agriculture green production

- Energy-efficient

- Technical efficiency

- Controlled-release fertilizer

5 Sustainable - Improved seeds technology 67 46.85%

Technology - Fertilizer technology

- Natural resource managementtechnology

- Green agriculture technology

- Risk management Strategies

technology

- Production system

- Breed production

- Fertilizer technology

- Modern rice technology

- Mixed cropping technology

- Cautionary savings technology

- Straw return technology

- Climate smart technology

- Water-saving irrigationtechnology

- Hybrid rice technology

- Rainwater harvesting andsupplementary irrigation

- Land conservation technology

- Soil testing technology

2.2 Publication bias

Related issues of meta-analysis shortcomings in empirical data collection,

publication bias must be considered to ensure a robust meta-analysis Reluctant to relyfor data on published articles, a meta-analysis must include a publication bias test for

robustness This type of bias stems from the logic behind the funnel chart: if large (i.e.,high precision) studies are more likely to be published, then the number of studies withpositive results will be greater small number of studies or with negative results in

journals And if this happens, the funnel chart will show asymmetry In other words, an

unbalanced funnel chart is a sign of a publication problem In this research article, I applyfunnel charts to detect deviations (Figure 2.2) plots the number of observations andstudy AD estimates collected in this meta-analysis The symmetric inverted funnel

model implies that AD bias decreases with increasing estimation accuracy (Jonathan J.Deeks et al., 2005) Following (Jonathan A.C Sterne & Matthias Egger, 2001), 1 include

the standard error (and variance) of AD as the vertical axis and the logarithm of the oddsratio as the horizontal axis The asymmetry of the funnel plots suggests the presence of

publication bias as the precision of AD estimates is higher as study size increases I alsoapplied Egger's test, suggested by (James E Pustejovsky & Melissa A Rodgers, 2018)

to the asymmetric test funnel plot The results show that a small study effect exists in ourdata with p-value < 0.05

According to research by (Agriculture Production Region) also found that

agricultural production in different locations will have potential differences due tocharacteristics of culture, people, land, climate, etc Therefore, we attribute small study

effects to differences in study sites To address this issue, I include the variable continent

to control for the heterogeneity of effect sizes across small studies With the Egger test,

I observed homogeneity in different subsets with p-value = 0.237 > 0.05, this result

implies that the possibility of publication bias is low

In addition to publication bias trends, multicollinearity implies that one or more

variables in an empirical regression model are implicitly predicted from anotherpredictor to a relatively high prior precision The limitation of multicollinearity is

pointed out in the magnitude of the estimated coefficients, which can changesignificantly with only a small adjustment in the data used Therefore, the estimated

results are biased with R-squared solutions or fitting unreliable results Multicollinearityrarely occurs in meta-analysis because the predictors in the regression are collected from

independent studies (Christpher Glen Thompson et al., 2017) My research uses threevariables which are education level, experience, and extension from all available

reputable articles Therefore, examining the correlation of these variables is necessary in

2.3 Meta-regression model

Meta-analytic studies have also shown the link between human capital and their

productivity performance (Forouzan Safari Ojgaz & Hassan Rangriz, 2020) which hasestimated the effectiveness of the study to be estimated by estimated odds ratio in meta-

analysis Similarly in my analysis, I estimate the effect size of the study using theexplanatory variables called education, experience, and extension to observe their

relationship with the dependent variable, our dependent variable is the “adoption rate”

of farmers practicing sustainable agriculture My dependent variable has a mean

Adoptionratel of 0.473 and a 95% confidence interval [0.05; 0.99]

Meta-regression models were used to detect the importance and signs of human capital

on the Adoptionratel of people engaged in sustainable agriculture Therefore, we have a

logistic regression model that characterizes the adoption of organic agriculture byfarmers, specified by the following formula:

ADij= B, + ,Educsigi + B,Expsigy + B,Extsigiy + B,Agelj + ;Age12¡ +

B Hhsizely + P„Farmsize1j + B ,Riskaversion; + „Lossauej+ B, Dif fyeari + B,, Continent + Dh=1 AnFnij + dị + ej (1)

Where: ADij is the average AD score (or mean) of the observation I reported in the

sustainable agricultural research study j, the intercept BO measures the average scale ofeducational level and the interaction their value would be one if the study took into

account the impact of variables such as agricultural extension, farmer experience andeducation level Human capital includes the statistical significance of the experience of

farmers Expsigij, the statistical significance of agricultural extension programs thatfarming households participate in Extsigij, the statistical significance of education level

of farming households Educsigi, age of household heads in the Age lij studies, twice the

age of household heads in the Agel2ij studies and household size in the Hhsizelij

studies The only variable in physical capital is the size of agricultural land of theresearched farming household Farmsize1ij Individual characteristics in human capital

include risk aversion among farmers Riskaversionij and loss aversion among farmersLossaveij In addition, control variables include the number of years difference betweenthe year the author completed writing the article and the year the data were collected

Diffyearij and the types of climate zones in the Continentij studies Fnij denotes n controlvariables and eij is the meta-regression error term Descriptive statistics and detailed

definitions of these variables are presented in (Table 3.1)

Tobit models are often used to estimate the influence of determinants on AD because

the efficiency score is tightly bounded to a range of 0 to 1 (Boris E Bravo - Ureta et al.events, 2007) However, in this study, we apply a segmented regression model to handlethe dependent variable - mean AD, determined on the closed interval ADi [0,1] (Leslie

E Papke & Jeffrey M Wooldridge , 1996) A descriptive introduction to the segmentedregression model is often described in the analysis of (Jeffrey M Wooldridge, 2009) The

fractional model with the dependent variable xi being a fraction bounded between 0 and

1, it means ADi [0,1], has the following structure:

E(AD/Zi) = H(Zif) (2)

In which, Zi represents a set of regression variables including explanatoryvariables (Expsigii , Extsigi and Educsigi ), interaction terms between theseexplanatory variables and the control variables (Fn) For making the logistic link H(.) issatisfied 0, H(.) = exp(.)/1 + exp(.) < 1 (VJeffrey M Wooldridge, 2009), The fractionallogistic model can be written as follows:

E(AD/Z) = —— B (3)

1+ eZj BpRecommended estimator for B is Quasi Maximum Likelihood Estimator (QMLE),

which maximizes the following Bernoulli log-likelihood function (Penny McCullagh &William S Little, 2009):

1) = ADj log[H(Z’i Ø)] + (1 - ADj )log[1 - H(Z”:/)] (4)