Article

Heterogeneities in Farmers’ Preference for Advisory Services: A

Choice Experiment of Vegetable Growers in North-Western Ethiopia
Ermias Tesfaye Teferi 1,2, * , Tigist Damtew Worku 3 , Solomon Bizuayehu Wassie 2 , Bernd Muller 4 ,
Abdul-Rahim Abdulai 5 and Céline Termote 6

1 International Center for Tropical Agriculture (CIAT), Addis Ababa P.O. Box 5689, Ethiopia
2 Department of Agricultural Economics, College of Agriculture and Environmental Sciences, Blue Nile Water
Institute, Bahir Dar University, Bahir Dar P.O. Box 79, Ethiopia; sbizuayehu@gmail.com
3 Department of Agricultural Economics, College of Agriculture, Food and Climate Science, Injibara University,
Injibara P.O. Box 40, Ethiopia; tigistdamtew16@gmail.com
4 Faculty of Agriculture, Food, and Nutrition, Weihenstephan-Triesdorf University of Applied Sciences,
85354 Freising, Germany; bernd.mueller@hswt.de
5 International Center for Tropical Agriculture (CIAT), Cali 6713, Colombia; a.abdulai@cgiar.org
6 The Alliance of Bioversity International and the International Center for Tropical Agriculture,
Nairobi P.O. Box 823-00621, Kenya; c.termote@cgiar.org
* Correspondence: e.tesfaye@cgiar.org

Abstract: This study investigated vegetable farmers’ willingness-to-pay (WTP) for private agricultural
advisory services in northwestern Ethiopia. Understanding farmers’ preferences is a crucial step for
modernizing agricultural advisory services and transforming smallholder agri-food systems into a
sustainable path. Discrete choice experiment data from 393 farm households were analyzed using a
random parameter logit model (RPL). The results revealed that vegetable farmers are willing to pay
for practice-oriented private advisory services. The result also showed the existence of heterogeneity
in farmers’ preference for features of vegetable advisory services. Household heads’ educational
status and age influenced preferences for advisory service features. The result is substantiated by
the fact that merely 25.5% of the sample households acquired formal education. Farmers in general
Citation: Teferi, E.T.; Worku, T.D.; preferred extension services with frequent expert visits and practical engagement on-farm as opposed
Wassie, S.B.; Muller, B.; Abdulai, A.-R.; to digitized options that rely on short message service (SMS) and voice-based guidance. Additionally,
Termote, C. Heterogeneities in farmers are willing to pay an ETB 120.89 and ETB 203.94 monthly fee for an extension service that
Farmers’ Preference for Advisory emphasizes fruity and root and tuber vegetables, respectively, as opposed to leafy vegetables. The
Services: A Choice Experiment of findings imply initiatives that push for commercializing agricultural advisory services should strive to
Vegetable Growers in North-Western achieve a balance between the practical application and digitization of extension services accounting
Ethiopia. Sustainability 2024, 16, 10821. for the heterogeneous preferences of smallholder farmers.
https://doi.org/10.3390/
su162410821
Keywords: mobile-based extension service; practice-oriented extension service; random parameter
Academic Editor: Piotr Prus logit; willingness-to-pay; private agricultural advisory; preference heterogeneity

Received: 25 October 2024

Revised: 30 November 2024
Accepted: 8 December 2024
1. Introduction
Published: 10 December 2024
Efficient agricultural advisory services (AAS) play a vital role in food system transfor-
mation through facilitating the adoption of sustainable technologies that enhance resilience
and productivity, thereby using improving rural livelihoods and sustainable agriculture
Copyright: © 2024 by the authors. as a catalyst for pro-poor economic growth [1–5]. Improved AAS is especially critical in
Licensee MDPI, Basel, Switzerland.
lower-and middle-income countries (LMICs) where agriculture is an engine of growth [6].
This article is an open access article
Ref. [7] defines agricultural extension as “systems that should facilitate the access of farmers,
distributed under the terms and
their organizations, and other market actors to knowledge, information, and technologies;
conditions of the Creative Commons
facilitate their interaction with partners in research, education, agribusiness, and other
Attribution (CC BY) license (https://
relevant institutions; and assist them in developing their own technical, organizational,
creativecommons.org/licenses/by/
4.0/).
and management skills and practices”.

Sustainability 2024, 16, 10821. https://doi.org/10.3390/su162410821 https://www.mdpi.com/journal/sustainability

Sustainability 2024, 16, 10821 2 of 15

Poor public agricultural extension service in African nations poses a significant obstacle
to achieving productivity gains. Recent research showed that the introduction of private
extension services in legume-based and cereal farming systems in countries like Ghana,
Mali, Tanzania, and Niger resulted in an increase in the adoption of enhanced agricultural
inputs, such as pesticides [8–12]. Similarly, a study conducted in Burkina Faso found that the
agricultural extension service offered by private groups was superior to the service provided
by the government [13]. Studies in Kenya and Cape Verde demonstrated the effectiveness of
private AAS, when combined with public extension systems, in improving efficiency [14,15].
In Ethiopia, the establishment of AAS dates back to the imperial regime in 1953 [16].
Over the years, the publicly operated system has undergone significant changes, partic-
ularly in terms of geographic coverage. Despite its long history and increasing access in
several parts of the country, research has shown that the system has struggled to keep up
with evolving needs, primarily due to factors such as overworked extension agents, limited
collaboration with research, lack of institutional innovation, and inadequate facilities at
farmer training centers [16–18].
The horticultural crops sector suffers most from inefficient AAS [2,19–21]. Despite mas-
sive diet shifts in LMICs, experts encourage fruit and vegetable crops for their health and
economic benefits [22–24]. Many sub-Saharan African countries, including Ethiopia [25]
and others with inefficient AAS, struggle to produce these essential foods. The low re-
sponsiveness of the public AAS to the diverse and dynamic needs of smallholders sustains
the previously ineffective management of diseases and pest outbreaks, outdated inputs
and agronomic information, lack of improved storage and transportation systems, and
dysfunctional markets.
Therefore, some have called for building private, responsive, and demand-driven
services to address the inefficiencies created by the current public AAS [17,26]. While calls
for private extension gain momentum in Ethiopia, their feasibility remains questionable,
especially among farmers used to free public extension. Specifically, are farmers willing to
pay for private extension services to receive tailored and efficient information? How might
farmers respond to such services? Previous studies in Ethiopia have alluded to farmers’
willingness-to-pay (WTP) for AAS. For example, ref. [27] found that only 10% of farmers
are willing to pay for current AAS in eastern Ethiopia. Most respondents to their survey
(64.2%) were favorably disposed to pay if the service was satisfying them and increased
their profits (see also [28–30]).
Similarly, several studies conducted across various Sub-Saharan African (SSA) coun-
tries generally highlight a favorable disposition among farmers to invest in enhanced
commercial agricultural advisory services despite the considerable discrepancies in the
amount, demanded features, and coverage of the services. For example, studies in Liberia,
Tanzania, Nigeria, Rwanda, and Zambia revealed that about 78.7%, 92%, 95.1%, 46% and
64% of farmers, respectively, were observed to be willing to pay for enhanced advisory
services [31–34]. Conversely, in certain countries, including Ghana and certain parts of
Tanzania, studies have identified discrepancies between the expectations and actual farm-
ers’ interests in paying for advisory services [5,35]. Consequentially, the existing body of
literature underscores the pressing necessity for investigating the demanded features of
enhanced advisory services and the corresponding willingness to pay for the services that
address the needs of smallholder farmers.
The findings of previous similar studies generally imply a low willingness for less
effective public advisory services and a potential high willingness for enhanced commercial
advisory services, but most studies fail to address which features of AAS are demanded
by farmers, especially those engaged in vegetable production. This study responds to
this need by elucidating the requisite attributes of advisory services in Ethiopia focusing
on vegetable cultivation, which requires specialized skills, knowledge, and care due to
the crops’ susceptibility to both natural and market-related risks, as they are primarily
produced for commercial purposes. The research employed a discrete choice experiment
approach to examine this critical niche associated with high-end vegetable produce. The
Sustainability 2024, 16, 10821 3 of 15

study investigated whether vegetable farmers in northwest Ethiopia are prepared to pay
for a paid extension service that provides timely, adequate, and appropriate information.

2. Methods
2.1. Study Area, Sampling, and Data
For this study, we collected data from two districts in the Amhara region of northwestern
Ethiopia, Fogera and North Mecha (Figure 1). These districts, near the Lake Tana, Rib,
and Koga irrigation schemes provide access to irrigation water facilities, allowing for the
cultivation of a variety of vegetables, including onions, tomatoes, cabbage, and potatoes [36].
The districts’ specialization in the cultivation of vegetables in the Amhara region, coupled with
the growing demand for efficient advisory service from producers, made them a pertinent
site for this study. We used a choice experiment approach to gather data from smallholder
vegetable producers. Using a Modified Fedorov algorithm, the Ngene v.1.4 (Ngene is a
software for generating experimental designs that are used in stated choice experiment for the
purpose of estimating choice models, particularly of the logit type [37]) software generated
18-choice tasks with an efficient d-error-minimizing design [38].

Figure 1. Location of the study area. The left panel shows the study districts of Mecha and Fogera
within the Amhara region in northwestern Ethiopia. The right panel shows the two districts, the
main roads and their location near Lake Tana.

Investigators interviewed the sampled household heads for the in-person survey.
Two-stage random sampling was utilized to choose sample households. Initially, two
vegetable-growing districts were targeted. In the second stage, our team randomly picked
three kebeles from each district. The study selected 393 households from selected kebeles
using systematic random sampling, ensuring sample households were proportional to
kebele populations. Households from North Mecha made up 46% of the sample, while
those from Fogera accounted for 54%.

2.2. Discrete Choice Experiment Design

The choice experiments assume that a service or policy links to attributes and attribute
levels [39]. Investigators asked sample respondents to choose among various competing
alternatives, such as different vegetable advisory services. Identifying relevant attributes
and levels is vital for designing a stated preference study. Our approach determined the
critical features of vegetable advisory services based on preference elicitation research
on extension services and on advice from horticulture specialists [40–43]. Accordingly,
the main attributes of the vegetable advisory service were identified as the frequency of
advisory service, the advisory knowledge/information exchange approach, the types of
vegetables covered, and the service fee in Ethiopian currency (ETB; Birr) expressed in
purchasing power parity for non-OECD countries (PPP (Ethiopia ET: Purchasing Power
Parity for non-OECD countries. A total of ETB 14.17 equals USD 1) of ETB 14.17). We
developed three alternatives for the choice experiment: two vegetable extension services,
and the option to refuse service. Respondents were asked to indicate their preferred
alternatives in the choice experiment.
Of the five vegetable advisory attributes included in the experimental design, three had
three levels, one had four, and the remaining had two. We designed the choice sets using
Ngene software, resulting in 18 choice sets. The sets were grouped into three blocks, each
with six choice sets, which were randomly provided to respondents. The number of choice
sets was limited to six to avoid response fatigue. A single respondent was presented with
Sustainability 2024, 16, 10821 4 of 15

six choices out of the 18 in the survey. Hence, 393 respondents generated 7074 observations
(see Tables 1 and 2).

Table 1. The DCE attributes and their levels/categories.

Attributes Values Used to Represent Each Level Description of Each Level

Focus of vegetable advisory service
1, 2 Input-focused and management-focused
(MVAS)
Frequency of advisory service (FAS) −1, 0, 1 Per week, per two weeks, per month
Approach of knowledge exchange (AKE) 0, 1, 2 On shops, on-field, by mobile
Types of Vegetables (TVAS) 0, 1, 2 On FV, on LV, and on R and T vegetables
40% less, 20% less, proposed amount,
Service fee for advisory in ETB (P) 80, 100, 120, 140
14.5% more

Table 2. Example of single choice set.

Attributes Advisory Service Alternative 1 Advisory Service Alternative 2 Alternative 3: Opt-out

Focus on vegetable Inputs Focus on vegetable management.

Main focus of vegetable advisory

service

Advice service per two weeks Advice service per one-month duration
Frequency of advisory service
Once Every 2 weeks Once Every month

Farmers go to one-stop shops Advice delivered on-field

Approach of knowledge exchange

Main focus on roots and tubers

Main focus on fruity vegetables

Types of vegetables in the advisory

service

ETB 100 per month ETB 140 per month

Service fee for advisory per month

Choice question:

The discrete choice experiment (DCE) attributes and their levels include fruity vegeta-
bles (FV), leafy vegetables (LV), and roots and tubers (R and T); ETB stands for Ethiopian
birr, and P is the price.

2.3. Empirical Strategy

Random utility theory [44] and Lancaster’s theory of characteristics [45] are the theo-
retical foundations for DCE. Our approach assumes the farmer will choose the vegetable
Sustainability 2024, 16, 10821 5 of 15

advisory service alternative that provides the highest level of utility. This choice depends on
the properties or attributes of the options provided and the characteristics of the individual
making the choices. Decision makers (farmers in this case) demonstrate their preferences for
specific attributes and their willingness to accept trade-offs for these attributes embedded in the
alternatives by choosing one alternative over another [46]. Following previous research [47,48],
we define a latent variable Vnjs* as an indirect utility function associated with selecting alter-
native j by farmer n from a given set of choices. Consequently, farmer n is provided with K
possible alternatives within a choice set s. Producers are presumed to optimize their utility in
every conceivable scenario, commonly referred to as a choice set. Thus:
(
* = max V * * *


1 i f Vnjs n1s , Vn2s , . . . , VnKs
Vnjs = (1)
0 otherwise

* > V * ∀ K ̸ = j. In Equation (1), actual utility

farmer n will choose alternative j if Vnjs nKs
is observed through the utility maximizing choice Vnjs , but indirect utility Vnjs * is not

directly observable. Indirect utility is assumed to be linear, whereas marginal utility is

monotonic (i.e., not increasing nor decreasing), yielding corner solutions in which one
choice is selected [49,50]. Based on this assumption, the utility function of farmer n can be
written as
* ′
Vnjs = Xnjs β + ε njs (2)
′ represents the vector of attributes of each choice for the jth alternative observed
where Xnjs
by farmer n, β is the vector of parameters that explain the influence each attribute has on
the choice (e.g., frequency of advisory service), and ε njs is the unobserved (stochastic) com-
ponent of utility, independent of the observed components (i.e., X) and equally distributed
across individuals and choices. The unobserved component reflects the hypothesis that
unobserved variations and errors occur in a farmer’s preference for an alternative in any
given choice set.
The random parameter logit model allows us to examine farmers’ preferences and will-
ingness to improve vegetable extension services by varying randomly assigned attributes.
A paper by [51] used this model to evaluate the heterogeneity of choices among individuals.
The model shows which attribute preferences are associated with specific types of farmers
(e.g., male and female). The probability that farmer n chooses alternative j in choice set s is
assumed to be [52]:

prob Vnjs = 1 X ′ i1s , X ′ i2s , . . . , X ′ iks , ∧


′
Z exp( X ijs β (3)
=  f (β|∧)dβ)
k ′
∑k=1 exp( X iks β

′ β is the marginal utility parameter at various attribute levels. Λ refers to the

where Xiks
parameters that characterize the distribution of random parameters, such as the mean and
covariance of β [53]. In this study, we specify the parameters (i.e., the attributes of vegetable
advisory services) and their respective attribute levels. In Equation (3), the probability
was numerically approximated using a maximum-likelihood simulation. The coefficients
corresponding to each attribute were assumed to follow a normal distribution. By doing so,
their sign can be either positive or negative, indicating preferences for each attribute. The
random parameter logit model was estimated using Stata 15.
Because of the non-cardinal nature of utility, the coefficients generated by a random
parameter logit (RPL) have a limited economic interpretation [50]. To gain insights into
the behavior of sampled individuals, economic trade-offs are calculated by dividing non-
Sustainability 2024, 16, 10821 6 of 15

monetary attributes (e.g., types of vegetables in the vegetable advisory service) by monetary
ones, such as the service fee [54].

βi ( atributes)
MWTP attributes = (4)
β6( price)

The values indicate a positive, negative, or null WTP ratio. Instead of giving everyone
an identical value for all their traits, the RPLs show whether the coefficients for each
person follow a testable distribution [52]. In this study, it is defined as the additional
amount of money a farmer is willing to pay for a marginal change in the vegetable advisory
service [55]. When an attribute parameter has a positive ratio, it means that people are
ready to pay more for an attribute that increases their utility—more payment for practical
farm advice, for example [56]. The monetary cost that people are willing to bear to accept a
modification to a feature that reduces its value may be represented as a negative ratio [57].

3. Results
3.1. Characteristics of Sample Households
Table 3 provides an overview of the demographic and socioeconomic characteris-
tics of the sample households. Almost all household heads (97%) were male, while fe-
males accounted for only 3%. The mean family size was 7.4 members, while on average,
3.8 members of the household were in the working age group of above 17 years old. The
observed value was comparatively higher than the mean size of Ethiopian families (4.6)
at the national level [58]. The average age of the sample household heads was 44 years,
while on average household heads had 23 years of agricultural experience. The age of the
household heads ranged from 19 to 81 years.

Table 3. Demographic and socioeconomic characteristics of sample households. N = 393.

Variable Mean Std. Dev. Min Max

Age of HH 43.7 12 19 81
Farm experience 23 12.1 1 58
Family size 7.4 2.1 1 11
Working family members 3.8 1.6 1 10
Tropical livestock units (TLU) 5.2 2.3 0 16.7
Total cultivated land 1.2 0.6 0.08 5.8
Land for vegetables 0.39 0.02 0.01 2

Livestock ownership ranged from households with none to those with over 16 tropical
livestock units (TLU). The average livestock holding was 5.2 based on the total number
of households. Along the same lines, the average size of cultivated landholdings held
by households in the sample is 1.2 ha, over twice as much as the national landholding,
which is 0.55 ha [59]. Sample households allotted an average of 0.4 ha of land for irrigated
farming out of the total cultivated land, with a minimum of 0.01 ha and a maximum of 2 ha
allocated to the same purpose.
Half of all heads of household sampled were unable to read and write. Of the half of
household heads who could read and write, 25% had completed formal education (Table 4).

Table 4. Education status of sampled household heads.

Education Status HH Frequency Percent Cumulative

Cannot read and write 198 50.4 50.4
Can read and write, but with
95 24.2 74.6
no formal education
Have formal education 100 25.5 100.0
Sustainability 2024, 16, 10821 7 of 15

3.2. Households Access to Basic Services

The survey assessed farmers’ access to general and vegetable-specific advisory services,
access to credit, membership in at least one local association (such as village savings,
religious or non-religious associations, etc.), and participation in at least one activity that
generates income outside farming. Forty-two percent of the households included in the
sample did not make use of any of the extension services offered in their area, while
the remaining 58% of the households took advantage of at least one extension service.
Similarly, 34% of the households included in the sample did not receive any extension
services specifically related to the production and marketing of vegetables. In comparison,
66% of the households received these services. The survey showed that 84.5% of households
had access to credit, while 58.3% of those households showed an interest in obtaining credit
for vegetable production. However, less than 5% of households received credit for the
agricultural production of vegetables (Table 5).

Table 5. Institutional characteristics of HH (N = 393).

Yes No
Variables
Frequency Percent Frequency Percent
Access to extension/training 229 58.27 164 41.73
Access to credit 332 84.48 61 15.52
Membership in local association 356 90.59 37 9.41
Advisory service related to vegetable
258 65.65 135 34.35
production and marketing
Credit for vegetable farming 18 4.58 375 95.42
Off-farm participation 78 19.85 315 80.15

3.3. Vegetable Advisory Service Choices and Preferences

The findings from the DCE data are presented in Table 6. A positive coefficient
suggests a positive impact of an attribute increase on the selection of a vegetable advice
service, whereas a negative coefficient indicates a negative influence. On occasions where
the attribute has three discrete levels—such as the approach used for knowledge exchange
(AKE) and the vegetable focus in the advisory service (TVAS)—one level is used as a
comparison reference to the other levels.

Table 6. RPL model results with attributes only.

Attributes Only Model

Mean of Fixed Parameter
Coefficient Standard Error Z P > (Z) WTP
Advisory Fee −0.0080 0.009 −2.04 0.041
Mean of random parameter
Input delivery focused 0.2024 0.3368 0.60 0.548 25.47
Frequency of advisory service 0.0150 0.0044 3.41 0.001 1.89
Advice by mobile phone −0.9488 0.1714 −5.54 0.000 −119.39
One-stop shops −0.3984 0.2009 −1.98 0.047 −50.14
Fruity vegetables 0.9607 0.0948 10.14 0.000 120.89
Roots and tuber vegetables 1.6208 0.1276 12.71 0.000 203.94
Standard Deviation random
Coefficient Standard Error Z P > (Z)
parameter
Input delivery focused 0.85835 0.28376 3.02 0.002 ***
Frequency of advisory service 0.00319 0.00037 8.71 0.000 ***
Advice by mobile phone 0.99874 0.15475 6.45 0.000 ***
One-stop shops 0.03019 0.23527 0.13 0.898
Fruity vegetables 0.76422 0.14829 5.15 0.000 ***
Roots and tuber vegetables 1.27165 0.14543 8.74 0.000 ***
Log likelihood = −1546.0302
LRchi2(6) = 255.40
Prob > chi2 = 0.0000
Notes: Dependent variable is choice; Standard errors in parentheses *** p < 0.01; or are statistically significant at
1%, 5%, and 10% level of significance, respectively.
Sustainability 2024, 16, 10821 8 of 15

For estimation purposes, the continuous variables price fee for advisory services (P)
and frequency of the advisory services (FAS) were coded by their levels. However, the
levels of the categorical attributes were coded as a dummy variable, where one was given
for their presence and zero otherwise. Management-oriented services delivered in the field
and leafy vegetable services were used as base levels for the focus of vegetable advisory
services, the approach of knowledge exchange, and types of vegetables in the advisory
service attributes. The price attribute is fixed in the estimation. All other variables have a
normal distribution. RPL was specified as a fixed or random coefficient for the attributes of
vegetable extension services.
Before estimating the parameters, the choice data were checked. The inspection results
showed that the opt-out alternative was chosen in approximately 0.83% of the cases, or
59 observations. The other attributes—including the focus of vegetable advisory service
(MVAS), frequency of advisory service (FAS), approach of knowledge exchange (AKE), and
types of vegetables in the advisory service (TVAS)—are set to random.
The results showed that the frequency of advisory service, face-to-face practice-
oriented mode (on-field) of advisory, leafy vegetables-focused services (the base level), and
monthly advisory fee significantly influenced the type of service preferred by vegetable-
producing farmers (Table 6). Advisory service features that were input-delivery focused
did not influence preferences. The coefficient of the fixed variable of price attribute is
negative and significant at the 5% level (−0.008), which indicates that the respondents’
preferences decrease as the price increases. This result is consistent with the theory of
demand; demand increases as price decreases and vice versa.
In the study area, vegetable producers preferred frequent visits from advisory providers
at the 1% significance level. As hypothesized, vegetable producers responded to increases
in advisory fees by shifting to self-management and traditional options. The negative
coefficients of the advice delivered through mobile calls and SMS and the advice offered
to farmers going to one-stop shops imply that practical advisory types, accompanied by
expert field-level visits, were preferred more by vegetable producers than mobile-assisted
and one-stop shop services, with 1% and 5% levels of significance, respectively. However,
this result contradicts the current efforts in Ethiopia to replace field-level advisory visits
with digitally assisted services.
Research conducted in northern Ethiopia has shown that farmers are willing to pay for
mobile-based advisory services [30]. However, compared to face-to-face advisory services,
the research results indicate that farmers prefer practice-oriented face-to-face advice.
Moreover, the significant standard deviations associated with random parameters
indicate unobserved sources of preference heterogeneity in the sample. The RPL model
results also indicated that each variable exhibited a preference for heterogeneity depending
on whether the variable had statistically significant standard deviations. Here, advice
delivered by farmers going to a one-stop shop is not statistically significant. Other attributes
are significant at the 1% level, such as advice delivered through mobile calls, SMS frequency
of the advisory services, the focus of the advisory (which is input delivered focus), the
type of vegetable (root type), and fruit type vegetable attributes. This result indicates
that the attributes exhibit preference heterogeneity. However, the insignificant standard
deviation for the variable on farmers going to the one-stop shops implies that all farm
households, regardless of their socioeconomic background, agree that the one-stop shop
extension service has no more important vegetable extension service attributes than the
other attributes.

3.4. Heterogenous Effects on Choices of Vegetable Advisory Services: Interaction of Mixed Logit Results
The RPL model was estimated using the included interaction terms. Table 7 shows the
heterogeneous effects on advisory service choice with respect to socioeconomic status based
on the mixed logit/random parameter logit model. These interaction terms allowed us to
consider that people with different demographic, socioeconomic, and attitudinal variables
may have different utilities concerning vegetable-producing advisory services. The interaction
Sustainability 2024, 16, 10821 9 of 15

between the frequency of the advisory service attribute and gender is positive and statistically
significant at the 5% level (Table 7). This finding indicates that male-headed household farmers
have a higher preference for frequent vegetable extension advisories.

Table 7. RPL attribute vs. sociodemographic interaction result.

Variables Variable Definition Mean SD

FeeA Monthly fee for vegetable advisory service −0.0139 * (0.0025)
SeXFrAS Sex interacted with frequency of adv. service 0.0029 ** (0.001)
AgeVI Age interacted with input focused adv. service 0.0207 (0.0149)
Age of household head interacted with frequency of adv.
AgeFrAS −5.77 × 10−5 * (0.0003)
service
EducOnM Education interacted with advice by mobile −0.373 *** (0.125)
EducFrAS Education interacted with frequency of adv. service 0.0016 ** (0.0007)
ExtTrainingXFrAS Extension training interacted with frequency of adv. service 0.001 (0.0007)
Extension training interacted with root tuber
ExtTrainingXRTV −0.599 *** (0.219)
vegetable-focused advisory
TLUXVI Livestock size interacted with frequency of adv. service 0.0938 (0.0738)
Extension on vegetables interacted with frequency of adv.
ExtencontactforvegeXFrAS −0.00045 (0.0004)
service
Land allocated for vegetables interacted with root tuber
totallandforvegetablesXRTV −0.121 (0.375)
vegetable-focused advisory
Input delivery focused −1.763 ** (0.718) 1.154 *** (0.304)
−0.007 ***
Frequency of adv. Service 0.0219 *** (0.0035)
(0.0011)
Advice through by mobile −0.0677 (0.239) 1.036 *** (0.157)
One-stop shops 1.025 *** (0.103) 0.911 *** (0.142)
Fruity vegs 2.061 *** (0.242) 1.412 *** (0.156)
Log likelihood = −1544.6939
LR chi2(5) = 225.15
Observations = 7074
Note: Standard errors in parentheses *** p < 0.01, ** p < 0.05, * p < 0.1. Table 6 shows heterogeneous effects on
advisory service choice with respect to socioeconomic status based on the mixed logit/random parameter logit.

The interaction between the frequency of advisory services and age of the household
head was negative and statistically significant at the 10% level of significance. As farmers
grow older, their preferences for more frequent vegetable advisory services become lower
than those of the youngest. However, the interaction between education status and frequency
of advisory services is positive and statistically significant at the 5% level. This result shows
that farmers who have received education prefer frequent vegetable advisory services.
The interaction between education and advisory services delivered through mobile
devices is negative and statistically significant at the 1% level, implying that farmers who
have received education are less concerned about advisory services delivered through
mobile devices. This result indicates that farmers with more education prefer the field-visit
mode of deliveries by field experts over receiving information over mobile phones.
The interaction between extension training and root-type vegetable advisory attributes
had a negative significance at 1%, indicating that farmers who received extension training
had less preference for root and tuber-type vegetable extension services than those who
did not receive extension training. However, the statistically significant standard deviation
estimates of the interaction between those attributes and variables suggest that the level of
heterogeneity of the attribute varies beyond what can be explained by the variables.

3.5. Estimation of Marginal Willingness-to-Pay for Advisory Services

Table 8 displays the marginal willingness to pay (MWTP) values expressed in Ethiopian
birr, which indicate the intensity of respondents’ preferences for each attribute. Respon-
dents anticipated that they would obtain less value from an advisory visit for vegetable
expansion with a lower MWTP. In contrast, a higher MWTP for the same characteristic
suggests that respondents anticipate greater utility. One indicator of limited WTP was the
frequency of the advisory service attribute (1.89 visits). The results showed that farmers are
not willing to pay higher margins for frequent advisory visits after they started obtaining
the required vegetable advisory services. Compared to the leafy type of vegetable, farmers
would rather pay a higher margin for acquiring the advisory service for fruity vegetables,
which is ETB 120.89, and for root-type vegetables, which is ETB 203.94.
Sustainability 2024, 16, 10821 10 of 15

Table 8. Estimation of marginal willingness to pay.

Fruit Root
Input-Based Frequency Advisory Through Advisory Through
Vegetable-Focused Vegetable-Focused
Advisory Services Advisory Visits Mobile One-Stop Shops
Advisory Service Advisory Service
WTP ETB 25.47 ETB 1.89 ETB −119.39 ETB −50.14 ETB 120.89 ETB 203.94
Lower limit ETB −78.28 ETB 1.13 ETB −264.53 ETB −142.42 ETB 4.88 ETB 3.08
Upper
ETB 129.22 ETB 2.65 ETB 25.75 ETB 42.15 ETB 236.90 ETB 404.80
Limit

This result suggests that the inclusion of preferred features of advisory services can
improve the acceptance and potential of future vegetable productivity. Compared to
input-focused advising services, vegetable producers in the Mecha and Fogera regions are
prepared to spend five and eight times as much for fruit-based and root-based services,
respectively, as for management-focused ones. For mobile advisory services, the average
WTP was ETB −119.39, whereas, for one-stop shop services, it was ETB −50. Based on
the field expert visit results, farmers did not want certain traits as much as they did at
the basic level. In addition, the findings showed that farmers were not very interested
in the frequency of the practice, as shown by their relatively low WTP, even after they
began obtaining the required advising information and practices. Advising services that
concentrate on the root and fruit types of vegetables and those that are practice-based tend
to have a greater WTP from vegetable growers.

4. Discussion
Agriculture remains the backbone of many sub-Saharan African countries, as it con-
tributes to maintaining food security, generating employment opportunities, producing
foreign exchange, and providing raw materials for industries [60–62]. Therefore, food
systems are transformative for most countries in the region. However, the accessibility
of agricultural information to smallholder farmers, who form the sector’s core, remains a
critical barrier [63,64] to ensuring equity and inclusivity in food systems transformation.
Many regional governments prioritize identifying the optimal conditions to ensure that
the right information reaches farmers at the right time to aid production decisions. The
Ethiopian government has struggled to facilitate access to information through extension,
prompting calls for private extension to expand its reach to as many farmers as possible.
Therefore, this study examined the conditions under which smallholders in Ethiopia are
willing to pay for privately provided extension services.
The choice experiment showed the heterogeneity in farmers’ WTP for extension. As
anticipated through prior literature on Ethiopian farmers’ WTP for extension services [28–30],
vegetable farmers showed some WTP for private extensions delivered through various media.
However, a notable finding of the study was the conditionality associated with willingness.
Specifically, the results indicate that farmers’ willingness is moderated by the elements of the
service, such as the mode of delivery, cost of the service, and type of information provided.
There was no blanket acceptance or WTP for extension services. The findings of this study
align with previous research [17], which emphasizes the need to enhance agricultural extension
services in Ethiopia by offering specialized training tailored to the unique situation. Hence,
efforts to establish private extensions will likely fail without paying attention to the differences
in farmers’ preferences [65]. This argument is further supported by the propensity of farmers
to opt out of services when they are not designed to meet the right conditions.
Furthering the understanding of the specificities of farmers’ WTP is the effect of
socioeconomic (age and education) interaction with extension service attributes. Specifically,
there was a significant association between advisory service frequency and age. As farmers
grow older, their preferences for more frequent vegetable advisory services become lower
than those of younger farmers. The age element, consistent with prior results [28,29,31],
could be attributed to experiences gained from farming as people aged. Farmers gain more
experience with age, reducing their need for external knowledge and information. However,
as climate change and many other shocks continue to emerge, these relationships can be
Sustainability 2024, 16, 10821 11 of 15

tested over time and must be considered an important topic for research. Novel research
must be conducted to understand how emergent shocks, with the risk of undermining
the existing knowledge of older farmers, can influence responses to external knowledge.
The results show a significant association between educational status and the frequency
of advisory services that farmers require and are willing to pay. Specifically, educated
farmers prefer vegetable advisory services more frequently. Educated farmers may know
more about the vulnerability of vegetables to different diseases during production and
the importance of extension services for reducing this vulnerability [59]. However, the
relationship between education, frequency, and WTP was not generic for all types of
extension services. The negative result of the interaction between education and advisory
services delivered through mobile phones indicated that farmers with more education
prefer the field visit mode of delivery. This critical finding, when applied to a general
sample, is worth discussing further.
Farmers were generally more willing to pay for advisory services delivered by expert
field-level visits than for mobile-assisted and one-stop shop services, with 1% and 5% levels
of significance, respectively. The potential source of this might be inadequate digital acces-
sibility resulting from the substandard digital infrastructure experienced by smallholder
farmers in Ethiopia [66]. This result is essential for the broader efforts to enhance access to
advisory services in Africa and Ethiopia. Governments and international development enti-
ties in the last decade have favored and promoted widespread digitalization of agriculture,
with extension and advisories at the center of such programs [67–69] Digital extensions and
advisories aim to reach hard-to-reach farmers left out of the overstretched public extension
schemes. Ironically, some have questioned whether such digital efforts are truly built in
the interests of farmers. For example, ref. [70] found that farmers tend to prefer traditional
face-to-face meetings. There are several reasons for this preference. Farmers will not
benefit from the delivery of digital extensions in languages (mostly English) inaccessible
to smallholders or when that information is poorly delivered (one-way communication).
Digital extension services suffer when farmers lack the knowledge to take advantage of
them, among many other shortcomings of digital extension [69–71]. While this finding
casts doubt on the feasibility and acceptability of digital extension for vegetable farmers in
Ethiopia, it is essential to note that it does not entirely discount the value that such services
may offer farmers. Similarly, lack of digital literacy was found to be the major barrier for
slow adoption of digital extension services in Ghana, Kenya, and Uganda [71,72].
Finally, farmers’ WTP for private extension depends on product/crop lines. Specif-
ically, vegetable producers in the Mecha and Fogera areas were prepared to spend five
and eight times more, respectively, on extension advice on leafy vegetables (such as lettuce
and cabbage) than on root and fruit vegetables (such as potatoes). The result implies that
improved vegetable extension services with a focus on leafy vegetables (i.e., the comparison
base) are considered desirable advisory services. Although the reason for these preferences
was not apparent from the experiment, it could be based on the variations in the suscepti-
bility of these different crops to disease risk and the agronomic care required. According to
one case study [73], leafy vegetables are increasingly exposed to new pathogens, such as
Alternaria leaf spot and Fusarium wilt that attack basil and lettuce vegetables, and require
constant adaptation and new knowledge [73–75]. Whatever the reason for this finding,
extension services should be designed for the specific needs of farmers rather than blanket
services that do not consider differences in preferences.

5. Conclusions
This study found heterogeneity in farmers’ WTP for private, improved extension
services, with the type of crop information (root vs. leafy vegetables) and mode of exten-
sion delivery (face-to-face vs. digital) among some critical differences that impact farmers’
preferences. Farmers emphasized the importance of in-person extension service deliv-
ery. Younger and educated farmers showed more WTP for frequent private extensions,
underscoring socioeconomic variation, including age and level of education.
Sustainability 2024, 16, 10821 12 of 15

The results provide critical insights into the future potential design and implementa-
tion of private extension services in Ethiopia and in several sub-Saharan African countries.
Farmers’ WTP for private extension must not be viewed blindly, regardless of hetero-
geneities within the smallholder sector and population. Instead, any attempt to improve
such a service must be discussed in the context of questions such as what type of extension
farmers demand. How can an extension service be delivered? What type of products will
such a service be based on? Specialists must design extension services for the specific needs
of farmers, rather than blanket programs that ignore differences in preferences.
The study nevertheless exhibits certain limitations, including the cross-sectional nature
of the survey data and the challenge that some survey participants encounter difficulties
in comprehensively conceptualizing the proposed advisory services in the choice situa-
tions. This entails the need to augment the willingness to pay for advisory services from
choice experiments with actual experimental data that pilots different attributes and their
acceptance by users in future studies.

Author Contributions: Conceptualization, E.T.T. and S.B.W.; methodology, E.T.T. and S.B.W.; soft-
ware, E.T.T. and T.D.W.; validation, B.M., E.T.T. and S.B.W.; formal analysis, E.T.T. and T.D.W.;
investigation, E.T.T. and T.D.W.; resources, B.M.; data curation, E.T.T. and T.D.W.; writing—original
draft preparation, T.D.W. and E.T.T.; writing—review and editing, E.T.T. and A.-R.A.; visualization,
E.T.T.; supervision, C.T., E.T.T. and S.B.W.; project administration, B.M.; funding acquisition, B.M.,
E.T.T. and S.B.W. All authors have read and agreed to the published version of the manuscript.
Funding: This research was funded by DAAD grant number 57571345.
Institutional Review Board Statement: The study was conducted in accordance with the Declaration of
Ethiopian Ministry of Education, and approved by the Research Ethical Review committee of Agricultural
Economics Department of Bahir Dar University (BDU/AGECO/Et202219, 15 September 2022).
Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.
Data Availability Statement: The original contributions presented in the study are included in the
article, further inquiries can be directed to the corresponding author.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Yitayew, A.; Abdulai, A.; Yigezu, Y.A. The Effects of Advisory Services and Technology Channeling on Farm Yields and Technical
Efficiency of Wheat Farmers in Ethiopia. Food Policy 2023, 116, 102436. [CrossRef]
2. Buehren, N.; Goldstein, M.; Molina, E.; Vaillant, J. The Impact of Strengthening Agricultural Extension Services on Women
Farmers: Evidence from Ethiopia. Agric. Econ. 2019, 50, 407–419. [CrossRef]
3. Venkatesan, P.; Sivaramane, N.; Sontakki, B.S.; Rao, C.S.; Chahal, V.P.; Singh, A.K.; Sivakumar, P.S.; Seetharaman, P.; Kalyani, B.
Aligning Agricultural Research and Extension for Sustainable Development Goals in India: A Case of Farmer FIRST Programme.
Sustainability 2023, 15, 2463. [CrossRef]
4. Wang, Z.; Wang, J.; Zhang, G.; Wang, Z. Evaluation of Agricultural Extension Service for Sustainable Agricultural Development
Using a Hybrid Entropy and TOPSIS Method. Sustainability 2021, 13, 347. [CrossRef]
5. Mapiye, O.; Makombe, G.; Molotsi, A.; Dzama, K.; Mapiye, C. Towards a Revolutionized Agricultural Extension System for the
Sustainability of Smallholder Livestock Production in Developing Countries: The Potential Role of Icts. Sustainability 2021, 13, 5868.
[CrossRef]
6. Danso-Abbeam, G.; Ehiakpor, D.S.; Aidoo, R. Agricultural Extension and Its Effects on Farm Productivity and Income: Insight
from Northern Ghana. Agric. Food Secur. 2018, 7, 74. [CrossRef]
7. FAO (Food and Agriculture Organization). Ethiopia Country Brief ; FAO: Addis Ababa, Ethiopia, 2010.
8. Tabe-Ojong, M.P.; Nyam, Y.S.; Lokossou, J.C.; Gebrekidan, B.H. Farmer Advisory Systems and Pesticide Use in Legume-Based
Systems in West Africa. Sci. Total Environ. 2023, 867, 161282. [CrossRef]
9. Abed, R.; Sseguya, H.; Flock, J.; Mruma, S.; Mwango, H. An Evolving Agricultural Extension Model for Lasting Impact: How
Willing Are Tanzanian Farmers to Pay for Extension Services? Sustainability 2020, 12, 8473. [CrossRef]
10. Jensen, P.F.; Prowse, M.; Larsen, M.N. Smallholders’ Demand for and Access to Private-Sector Extension Services: A Case Study
of Contracted Cotton Producers in Northern Tanzania. J. Agrar. Change 2019, 19, 122–134. [CrossRef]
11. Yegbemey, R.N.; Yabi, J.A.; Heubach, K.; Bauer, S.; Nuppenau, E.A. Willingness to Be Informed and to Pay for Agricultural
Extension Services in Times of Climate Change: The Case of Maize Farming in Northern Benin, West Africa. Clim. Dev. 2014, 6,
132–143. [CrossRef]
Sustainability 2024, 16, 10821 13 of 15

12. Bakang, J.-E.A.; Wongnaa, C.A.; Tham-Agyekum, E.K.; Fatimatu, S.; Obeng, J.A.; Nsafoah, E.B.; Antwi, M.A.; Bakang, J.-E.A.;
Wongnaa, C.A.; Tham-Agyekum, E.K.; et al. Determinants of Maize Farmers Willingness to Pay for Private Extension Services in
Ejisu Municipality, Ghana. Heliy 2024, 10, e37464. [CrossRef] [PubMed]
13. Sylla, A.Y.; Mahama Al-Hassan, R.; Egyir, I.S.; Anim-Somuah, H. Perceptions about Quality of Public and Private Agricultural
Extension in Africa: Evidence from Farmers in Burkina Faso. Cogent Food Agric. 2019, 5, 1685861. [CrossRef]
14. Muyanga, M.; Jayne, T.S. Private Agricultural Extension System in Kenya: Practice and Policy Lessons. J. Agric. Educ. Ext. 2008,
14, 111–124. [CrossRef]
15. Conradie, B.; Galloway, C.; Renner, A. Private Extension Delivers Productivity Growth in Pasture-Based Dairy Farming in the
Eastern Cape, 2012–2018. Agrekon 2022, 61, 2063143. [CrossRef]
16. Berhane, G.; Ragasa, C.; Abate, G.T.; Assefa, T.W. The State of Agricultural Extension Services in Ethiopia and Their Contribution to
Agricultural Productivity; ESSP Working Paper 118; International Food Policy Research Institute (IFPRI): Washington, DC, USA, 2018.
17. Davis, K.E.; Babu, S.C.; Ragasa, C. Agricultural Extension: Global Status and Performance in Selected Countries; International Food
Policy Research Institute: Washington, DC, USA, 2020. [CrossRef]
18. Leta, G.; Kelboro, G.; Stellmacher, T.; Hornidge, A.-K. The Agricultural Extension System in Ethiopia: Operational Setup,
Challenges and Opportunities. In ZEF Development Studies; Center for Development Research (ZEF), University of Bonn: Bonn,
Germany, 2017.
19. Aliyi, I.; Faris, A.; Ayele, A.; Oljirra, A.; Bayessa, M. Profitability and Market Performance of Smallholder Vegetable Production:
Evidence from Ethiopia. Heliyon 2021, 7, e08008. [CrossRef]
20. Gebru, K.M.; Leung, M.; Rammelt, C.; Zoomers, A.; van Westen, G. Vegetable Business and Smallholders’ Food Security:
Empirical Findings from Northern Ethiopia. Sustainability 2019, 11, 743. [CrossRef]
21. Gebremariam, Y.A.; Dessein, J.; Wondimagegnhu, B.A.; Breusers, M.; Lenaerts, L.; Adgo, E.; Van Passel, S.; Minale, A.S.; Nyssen, J.
Undoing the Development Army: A Paradigm Shift from Transfer of Technology to Agricultural Innovation System in Ethiopian
Extension. Environ. Dev. Sustain. 2023, 26, 1–27. [CrossRef]
22. Keller, K.L.; Lehmann, D.R. Brands and Branding: Research Findings and Future Priorities. Mark. Sci. 2006, 25, 740–759.
[CrossRef]
23. Deribe, H.; Mintesnot, A. Review on Contribution of Fruits and Vegetables on Food Security in Ethiopia. J. Biol. Agric. Healthc.
2016, 6, 49–58.
24. Wolfenden, L.; Barnes, C.; Lane, C.; McCrabb, S.; Brown, H.M.; Gerritsen, S.; Barquera, S.; Véjar, L.S.; Munguía, A.; Yoong, S.L.
Consolidating Evidence on the Effectiveness of Interventions Promoting Fruit and Vegetable Consumption: An Umbrella Review.
Int. J. Behav. Nutr. Phys. Act. 2021, 18, 11. [CrossRef]
25. Bekele, M. Challenges and Opportunities of Marketing Fruit and Vegetables at Logia, Northeastern Ethiopia. The Case of Onion,
Tomato and Banana. J. Mark. Consum. Res. 2016, 22, 51–58.
26. Uddin, E.; Gao, Q.; Mamun-Ur-Rashid, M.D. Crop Farmers’ Willingness to Pay for Agricultural Extension Services in Bangladesh:
Cases of Selected Villages in Two Important Agro-Ecological Zones. J. Agric. Educ. Ext. 2016, 22, 43–60. [CrossRef]
27. Tolera, T.; Temesgen, D.; Rajan, D.S. Factors Affecting Farmers’ Willingness to Pay for Agricultural Extension Services: The Case
of Haramaya District, Ethiopia. Agric. Sci. Res. 2014, 3, 268–277.
28. Temesgen, D.; Tola, T. Determinates of Small Holder Farmers Willingness to Pay for Agricultural Extension Services: A Case
Study from Eastern Ethiopia. Afr. J. Agric. Res. 2015, 10, 2152–2158. [CrossRef]
29. Gebreegziabher, K.T.; Mezgebo, G.K. Smallholder Farmers Willingness to Pay for Privatized Agricultural Extension Services in
Tigray National Regional State, Ethiopia. J. Agric. Ext. 2020, 24, 29–38. [CrossRef]
30. Abebe, A. Farmers’ Willingness to Pay for Mobile Phone-Based Agricultural Extension Service in Northern Ethiopia. Cogent Food
Agric. 2023, 9, 2260605. [CrossRef]
31. Gosbert, L.S.; Athman, K.A.; Jumanne, M.A. Factors Determining Crop Farmers Willingness to Pay for Agricultural Extension
Services in Tanzania: A Case of Mpwapwa and Mvomero Districts. J. Agric. Ext. Rural Dev. 2019, 11, 239–247. [CrossRef]
32. Ogunmodede, A.M.; Tambo, J.A.; Adeleke, A.T.; Gulak, D.M.; Ogunsanwo, M.O. Farmers’ Willingness to Pay towards the
Sustainability of Plant Clinics: Evidence from Bangladesh, Rwanda and Zambia. Int. J. Agric. Sustain. 2022, 20, 1360–1372.
[CrossRef]
33. Ozor, N.; Garforth, C.J.; Madukwe, M.C. Farmers’ Willingness to Pay for Agricultural Extension Service: Evidence from Nigeria.
J. Int. Dev. 2013, 25, 382–392. [CrossRef]
34. Sumo, T.V.; Ritho, C.; Irungu, P. Determinants of Smallholder Rice Farmers’ Willingness-to-Pay for Private Extension Services in
Liberia: The Case of Gibi District. Sustainability 2023, 15, 14300. [CrossRef]
35. Alhassan, N.J.; Samuel, A.D.; Franklin, N.M.; Isaac, G.K.A. Commercializing Innovations from Agricultural Research in Northern
Ghana and Farmers’ Willingness to Pay. Afr. J. Bus. Manag. 2016, 10, 140–150. [CrossRef]
36. Kassahun, B.; Tegegne, G.-E. Socio-Economic Base-Line Survey of Rural and Urban Households in Tana Sub-Basin, Amhara National
Regional State; Forum for Social Studies (FSS): Addis Ababa, Ethiopia, 2014.
37. ChoiceMetrics. The Cutting Edge in Experimental Design. In Ngene 1.4 User Manual & Reference Guide; Choice Metrics Pty. Ltd:
Sidney, Australia, 2024.
38. ChoiceMetrics. Ngene 1.2 User Manual & Reference Guide; Choice Metrics Pty. Ltd: Sidney, Australia, 2018.
Sustainability 2024, 16, 10821 14 of 15

39. Louviere, J.J.; Pihlens, D.; Carson, R. Design of Discrete Choice Experiments: A Discussion of Issues That Matter in Future
Applied Research. J. Choice Model. 2009, 4, 1–8. [CrossRef]
40. Mugunieri, G.L.; Omiti, J.M. Decentralization, Community-Based Organizations and Access to Agricultural Extension Services in
Eastern Kenya. In Decentralization and the Social Economics of Development: Lessons from Kenya; Cornell University: Ithaca, NY, USA,
2007; pp. 64–83. [CrossRef]
41. Tesfaye, A.; Hansen, J.; Kagabo, D.; Birachi, E.; Radeny, M. Rwanda Climate Services for Agriculture: Farmers Willingness to Pay for
Improved Climate Services; CCAFS Working Paper No. 314; CGIAR Research Program on Climate Change, Agriculture and Food
Security: Wageningen, The Netherlands, 2020.
42. James, P.A.; Smart, J.C.; Smith, J.; Bulling, M.; Beed, F.D.; Luwandagga, D. The Effect of Participation in the Ugandan National
Agricultural Advisory Services on Willingness to Pay for Extension Services. Afr. J. Agric. Resour. Econ. 2011, 6, 1–19.
43. Altobelli, F.; Marta, A.D.; Heinen, M.; Jacobs, C.; Giampietri, E.; Mancini, M.; Cimino, O.; Trestini, S.; Kranendonk, R.; Chanzy, A.;
et al. Irrigation Advisory Services: Farmers Preferences and Willingness to Pay for Innovation. Outlook Agric. 2021, 50, 277–285.
[CrossRef]
44. List, J.A.; Sinha, P.; Taylor, M.H. Using Choice Experiments to Value Non-Market Goods and Services: Evidence from Field
Experiments. B.E. J. Econ. Anal. Policy 2006, 6, 0000102202153806371132. [CrossRef]
45. Lancaster, K.J. A New Approach to Consumer Theory. J. Political Econ. 1966, 74, 132–157. [CrossRef]
46. Altobelli, F.; Lall, U.; Dalla Marta, A.; Caracciolo, F.; Cicia, G.; D’urso, G.; Giudice, T. Del Willingness of Farmers to Pay for
Satellite-Based Irrigation Advisory Services: A Southern Italy Experience. J. Agric. Sci. 2018, 156, 723–730. [CrossRef]
47. Schaafsma, M.; Ferrini, S.; Turner, R.K. Assessing Smallholder Preferences for Incentivised Climate-Smart Agriculture Using a
Discrete Choice Experiment. Land Use Policy 2019, 88, 104153. [CrossRef]
48. Teferi, E.T.; Kassie, G.T.; Pe, M.E.; Fadda, C. Are Farmers Willing to Pay for Climate Related Traits of Wheat? Evidence from Rural
Parts of Ethiopia. Agric. Syst. 2020, 185, 102947. [CrossRef]
49. Useche, P.; Barham, B.L.; Foltz, J.D. Trait-Based Adoption Models Using Ex-Ante and Ex-Post Approaches. Am. J. Agric. Econ.
2013, 95, 332–338. [CrossRef]
50. Silberg, T.R.; Richardson, R.B.; Lopez, M.C. Maize Farmer Preferences for Intercropping Systems to Reduce Striga in Malawi. Food
Secur. 2020, 12, 269–283. [CrossRef]
51. Hensher, D.A.; Greene, W.H. The Mixed Logit Model: The State of Practice. Transportation 2003, 30, 133–176. [CrossRef]
52. Train, K. Discrete Choice Methods with Simulation, Second Edition; Cambridge University Press: Cambridge, UK, 2009; Volume
9780521766, ISBN 9780511805271.
53. Waldman, K.B.; Ortega, D.L.; Richardson, R.B.; Snapp, S.S. Estimating Demand for Perennial Pigeon Pea in Malawi Using Choice
Experiments. Ecol. Econ. 2017, 131, 6. [CrossRef] [PubMed]
54. Knowler, D.; Philcox, N.; Nathan, S.; Delamare, W.; Haider, W.; Gupta, K. Assessing Prospects for Shrimp Culture in the Indian
Sundarbans: A Combined Simulation Modelling and Choice Experiment Approach. Mar. Policy 2009, 33, 9. [CrossRef]
55. Gibson, J.M.; Rigby, D.; Polya, D.A.; Russell, N. Discrete Choice Experiments in Developing Countries: Willingness to Pay Versus
Willingness to Work. Envrion. Resour. Econ. 2016, 65, 697–721. [CrossRef]
56. Nahuelhual, L.; Loureiro, M.L.; Loomis, J. Using Random Parameters to Account for Heterogeneous Preferences in Contingent
Valuation of Public Open Space. J. Agric. Resour. Econ. 2004, 29, 537–552.
57. Rigby, D.; Burton, M. Preference Heterogeneity and GM Food in the UK. Eur. Rev. Agric. Econ. 2005, 32, 269–288. [CrossRef]
58. CSA. Population Projection of Ethiopia for All Regions at Woreda Level from 2014–2017; CSA: Addis Ababa, Ethiopia, 2017.
59. CSA. Agricultural Sample Survey; Volume I, Report on Area and Production for Major Crops (Private Peasant Holdings, Main
Season); CSA: Addis Ababa, Ethiopia, 2017.
60. Modi, R. The Role of Agriculture for Food Security and Poverty Reduction in Sub-Saharan Africa. In The Palgrave Handbook of
Contemporary International Political Economy; Palgrave Macmillan: London, UK, 2019.
61. Diao, X.; Hazell, P.; Resnick, D.; Thurlow, J. The Role of Agriculture in Development: Implications for Sub-Saharan Africa; Research
Report; International Food Policy Research Institute: Washington, DC, USA, 2007. [CrossRef]
62. Christiaensen, L.; Demery, L.; Kuhl, J. The (Evolving) Role of Agriculture in Poverty Reduction-An Empirical Perspective. J. Dev.
Econ. 2011, 96, 6. [CrossRef]
63. Ndimbwa, T.; Mwantimwa, K.; Ndumbaro, F. Channels Used to Deliver Agricultural Information and Knowledge to Smallholder
Farmers. IFLA J. 2021, 47, 153–167. [CrossRef]
64. Megerssa, G.R.; Gemede, B.G.; Jarsa, A.W. Challenges of Smallholder Farmers in Accessing Agricultural Information in Southern
Ethiopia. Agric. Sci. Technol. 2020, 12, 87–95. [CrossRef]
65. Steinke, J.; Achieng, J.O.; Hammond, J.; Kebede, S.S.; Mengistu, D.K.; Mgimiloko, M.G.; Mohammed, J.N.; Musyoka, J.; Sieber, S.;
van de Gevel, J.; et al. Household-Specific Targeting of Agricultural Advice via Mobile Phones: Feasibility of a Minimum Data
Approach for Smallholder Context. Comput. Electron. Agric. 2019, 162, 991–1000. [CrossRef]
66. Tamene, L.D.; Ashenafi, A. Digital Agriculture Profile: Ethiopia; Alliance of Bioversity International and CIAT: Addis Ababa,
Ethiopia, 2022.
67. Abdulai, A.R.; KC, K.B.; Fraser, E. What Factors Influence the Likelihood of Rural Farmer Participation in Digital Agricultural
Services? Experience from Smallholder Digitalization in Northern Ghana. Outlook Agric. 2023, 52, 57–66. [CrossRef]
Sustainability 2024, 16, 10821 15 of 15

68. Abdulai, A.R. Toward Digitalization Futures in Smallholder Farming Systems in Sub-Sahara Africa: A Social Practice Proposal.
Front. Sustain. Food Syst. 2022, 6, 866331. [CrossRef]
69. Tsan, M.; Benjamin, K.; Totapally, S.; Hailu, M.; Addom, B.K. The Digitalisation of African Agriculture Report 2018–2019;
CTA/Dalberg Advisers: Wageningen, The Netherlands, 2019; Volume 53.
70. McCampbell, M.; Schumann, C.; Klerkx, L. Good Intentions in Complex Realities: Challenges for Designing Responsibly in
Digital Agriculture in Low-Income Countries. Sociol. Rural. 2022, 62, 279–304. [CrossRef]
71. Kansiime, M.K.; Mugambi, I.; Rware, H.; Alokit, C.; Aliamo, C.; Zhang, F.; Latzko, J.; Yang, P.; Karanja, D.; Romney, D. Challenges
and Capacity Gaps in Smallholder Access to Digital Extension and Advisory Services in Kenya And Uganda. Front. Agric. Sci.
Eng. 2022, 9, 642–654. [CrossRef]
72. Abdulai, A.-R. Are Rural Smallholders Ready for Agricultural Digitalization? Farmer (In)Competencies and the Political Economy
of Access in Digital Agricultural Extension and Advisories in Northern Ghana. In Digital Communication for Agricultural and Rural
Development; Chowdhury, A., Gow, G.A., Eds.; Routledge: London, UK, 2024; pp. 175–192.
73. Gilardi, G.; Garibaldi, A.; Gullino, M.L. Emerging Pathogens as a Consequence of Globalization and Climate Change: Leafy
Vegetables as a Case Study. Phytopathol. Mediterr. 2018, 57, 146–152.
74. Ebert, A.W. Vegetable Production, Diseases, and Climate Change. Front. Econ. Glob. 2017, 17, 103–124. [CrossRef]
75. Golding, J.; Tesoriero, L.; Daniel, R. Leafy Vegetables. In Pathology of Fresh Horticultural Produce; CRC Press: Boca Raton, FL, USA,
2019; pp. 339–372.

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.