









































Pa
ge

 
1



Pa
ge

 
15

 American Journal of  
Food Science and Technology (AJFST)

From Indigenous to Improved Breeds: Adoption Dynamics and Intensity of  Artificial 
Insemination Among Farmers in Western Kenya

G. O. Awuor1, E. Kipkemei1, A. Serem1, A. K. Kipkoech1* 

Volume 4 Issue 2, Year 2025
ISSN: 2834-0086 (Online)

DOI: https://doi.org/10.54536/ajfst.v4i2.5280
https://journals.e-palli.com/home/index.php/ajfst

Article Information ABSTRACT

Received: May 09, 2025

Accepted: June 14, 2025

Published: September 01, 2025

Adoption of  Artificial Insemination (AI) has the potential to upgrade local dairy breeds 
for improved milk production in western Kenya. This study employed the double hurdle 
probit model to analyze factors influencing the adoption and intensity of  AI technology. 
A multistage random sampling technique was employed to identify sample units among 
adopters and non-adopters of  AI technology. Data was obtained from cross-sectional survey 
of  378 farmer households. Results of  the probit model showed that age, education level, 
experience, milk sales, AI cost, worker’s skill on heat detection, semen type, AI reliability, and 
availability of  the inseminator positively and significantly influenced AI technology adoption. 
Only training on livestock production negatively and significantly influenced AI technology 
adoption. Results of  the truncated regression showed that age, education level, experience, 
and training on livestock production positively and significantly influenced the intensity of  
AI technology use. Group membership and the availability of  the inseminator negatively 
and significantly influenced the intensity of  AI technology adoption. It is concluded 
information is the most critical factor influencing adoption of  AI. Building more trust and 
confidence about AI technologies will lead to increased adoption. The study recommends 
the improvement of  farmer education through introduction of  effective farmer training 
and information sessions. There is need to conduct training needs assessments before the 
trainings are carried out so as to capture the farmers’ interest together with their socio-
economic environments. 

Keywords

Adoption Intensity, AI Technology, 
Probit, Western Kenya

1 Department of  Agricultural Economics and Resource Management, Moi University, P.O Box 3900, Eldoret, 30100, Kenya
* Corresponding author’s e-mail: akkipkoech@gmail.com

INTRODUCTION 
The primary objective of  improving the livestock sector 
is meeting the increasing demand for livestock products 
for achieving food security and the sustainability of  the 
economy. In Kenya, the livestock sector contributes over 
30% of  the farm gate value of  agricultural commodities, 
about 10% of  the national Gross Domestic Product 
(GDP), at least 50% of  the agricultural GDP, and employ 
50% of  the labor force in agriculture (KALRO, 2024). 
The dairy sub-sector are critical in the economy given the 
ever-increasing human demand of  milk and the potential 
to generate frequent and sustainable household incomes. 
Milk contributes substantially to the economy, accounting 
for 27% of  the value added by livestock and 10% of  the 
total agricultural value added (FAO, 2016).
In western Kenya, dairy sub-sector faces significant 
challenges despite its potential to contribute to food 
security and economic growth. While cows are the 
most prevalent dairy animal, farmers typically keep just 
two or three cows, meaning that farmers experience a 
higher average cost of  production compared to farmers 
in developed countries where average dairy herd size is 
around 90 cows in the UK and 300 in the USA (FAO, 
2016). Herd quality is characterized by low productivity, 
with smallholder farms predominantly keeping exotic-
zebu (indigenous breed) crosses (41.7%), Friesian (34.3%), 
and Ayrshire (22.4). The average milk production per cow 
in is estimated at 6.47 liters per day, with a lactation yield 

of  2,400 liters (Wanjala & Njehia, 2014). These figures are 
significantly lower than the global average of  40 liters per 
day per cow and up to 14,000 liters per lactation. 
This disparity is attributed to genetic limitations of  the 
available animal breeds Low adoption of  high-yield exotic 
breeds due to cultural preferences of  the indigenous Zebu 
breed and inadequate artificial insemination services 
(Kiplagat et al., 2025), inefficient breeding services, 
inadequate management practices leading to poor-quality 
fodder production and seasonal scarcity (Odero-Waitituh, 
2017) and insufficient extension and advisory services. 
Studies have shown that the herd quality influence the 
ability of  farmers to improve their milk production 
and herd size through influencing the rate of  spread of  
reproductive diseases, conception rates, calf  mortality 
rates, growth and maturity rates. 
The use of  new technologies to improve herd quality 
has been found to increase productivity, lower the 
risk of  livestock diseases, and ensures environmental 
sustainability in productive areas (Kimunya, 2014). There 
are a number of  reproductive technologies available 
to transfer desirable genetic materials such as artificial 
insemination, embryo transplant, and vitro fertilization, 
of  which only artificial insemination (AI) is the most 
commonly used technique in developing countries 
including Kenya. Particularly used as a reproduction 
method in dairy farming, AI provides significant economic 
contributions to milk production and to farmers by 



Pa
ge

 
16

https://journals.e-palli.com/home/index.php/ajfst

Am. J. Food. Sci. Technol. 4(2) 15-25, 2025

genetically improving animals (Howley et al., 2012). The 
advancement and application of  artificial insemination 
have significantly transformed cattle production and 
genetic enhancement, especially within the dairy industry 
in developed nations (Henning et al., 2010).

Potential of  Artificial Insemination (AI) To Improve 
Herd Quality And Productivity 
Artificial Insemination is the process by which semen is 
artificially introduced into the female reproductive tract 
for the purpose of  conception (Shehu et al., 2010). The 
process has become a popular avenue in the endeavour to 
the improve herd quality and milk production in western 
Kenyan. The technology allows for future change of  
the herd breed by introducing semen from genetically 
superior bulls in to local cows. The semen is carefully 
selected to ensure they have the desired traits such as 
higher milk yield, disease resistance, and better adaptation 
to local environments. 
In western Kenya, the government promote AI services 
on dairy cattle, highlighting the focus on boosting the 
productivity of  the dairy sub-sector. The use of  AI has 
helped to address challenges such spread of  reproductive 
diseases, poor conception rates, social issues from 
sharing bulls, lack of  control over progeny quality, limited 
financial resources, inadequate veterinary support, and 
poor performance of  dairy breeds (Gahakwa et al., 
2014). The use of  AI services enables farmers to access 
superior genetics that would otherwise be unavailable or 
unaffordable through natural mating. For instance, the 
cost of  purchasing and maintaining a bull are inhibitably 
high. The adoption of  AI service among farmers has 
resulted to emergence of  crossbreeds heifer cows 
with significantly improved milk production potential 
compared to indigenous Zebu breed. There has been a 
significant reduction in inbreeding, which was responsible 
for the undesirable traits inherent in the local cow breeds. 
For the purpose of  this research the study sought to 
understand the frequency of  use of  artificial insemination 
by the AI expert in consultation with the farmer. There has 
been increase in promotional activities for the AI service 
by both government and private sector players through 
training of  professionals in AI, provision of  high-quality 
semen and training for farmers on various aspects of  AI. 
There has also been a policy shift in to allowing for the 
privatization of  AI services that improved accessibility. 
Heat synchronization of  heifers and cows was tried in 
Siaya County among other counties, with the Government 
of  Kenya collaborating with ILRI to upgrade the local 
zebus using fixed time artificial insemination (FTAI). 
This initiative saw the increase in the number of  artificial 
inseminations in western Kenya (in particular at Alego-
Usonga sub-County) with a peak being witnessed in 
2016 as presented in Table 1. Despite the potential of  
AI technology to upgrade local zebus through selective 
breeding and meet growing milk demand, farmers have 
not fully embraced the technology, with sizeable number 
of  farmers still opting for use of  traditional bulls, or even 

preferring to maintain their traditional herd. 
Several studies have explored the drivers of  AI technology 
in different contexts, highlighting the complex interplay 
of  socio-economic, institutional, and technical factors 
(Gebre et al., 2022; Ingabire et al., 2018; Bayan, 2018; 
Mwanga et al., 2019). These factors can include farmer 
characteristics (age, education, and experience), access 
to credit and extension services, cost of  AI services, 
availability of  liquid nitrogen, infrastructure, and cultural 
beliefs (Ayantunde et al., 2008). Understanding these 
factors is crucial for designing and implementing effective 
interventions to promote AI technology adoption and 
maximize its impact on livestock productivity. 
In Alego-Usonga Sub-County, while livestock keeping 
is a common practice, there is a need for more in-depth 
research to understand the specific factors that influence 
adoption of  AI technology. There is need to capture the 
local and unique context of  Alego-Usonga, which might 
have distinct socio-economic, environmental, and cultural 
characteristics. A comprehensive analysis of  the factors 
influencing AI adoption and intensity of  use within this 
specific sub-county is essential. This study sought to 
address this gap by investigating the socio-economic, 
technical, and institutional factors that influence farmers’ 
decisions to adopt and consistently utilize AI technology 
in Alego-Usonga sub-County to improve their dairy herd 
and ultimately milk yield.

MATERIALS AND METHODS
Study Area
The study area was Alego-Usonga sub-county that covers 
an area of  599 square kilometers within Siaya county in 
western Kenya. The areas has a population of  224,363 
people according to the 2029 Kenya National Bureau of  
Statistics (KNBS) census of  2019 (KNBS, 2019). The 
population is made of  smallholder farmers practicing 
mixed farming where they keep livestock and farm crops. 
There are 3,512 households in the study area producing 
26% of  the milk consumed in the area from the 81% of  
marketed of  milk produce. The farmers are known to hire 
labour for dairy management activities. The smallholder 
dairy farmers keep 1-5 cows on an average of  0.5 to 3 
acres. The improved cows produce 3.4 liters per cow per 
day compared to the indigenous zebu cows that produce 
an average of  2.4 liters/cow/day (ASDSP, 2014).
The area has a bimodal rainfall pattern with two rainy 
seasons in each year typically separated by drier periods, 
leading to alternating wet and dry phases throughout the 
year. The rains occur between March and May and again 
between October and December and is well distributed 
within the Northern parts of  Ugunja, Gem and Ugenya 
receiving the highest amount between 1600-2000mm and 
rainfall gradually reducing to Southern parts of  Bondo 
and Rarieda which receive the lowest amounts between 
800-1200mm annually. The temperature ranges from 
15 to 32 degree Celsius with an annual average of  28 
degrees Celsius (Directorate of  Livestock Production 
and Veterinary, 2019). The weather characterized by 



Pa
ge

 
17

https://journals.e-palli.com/home/index.php/ajfst

Am. J. Food. Sci. Technol. 4(2) 15-25, 2025

warm temperatures, consistent rainfall and high humidity 
provide reliable moisture for pasture growth, ensuring 
that grasses and forage crops can thrive throughout the 
rainy seasons.

Data Collection, Sampling Technique and Size 
This study utilized a cross-sectional study design. 
Data was collected using a structured questionnaire 
administered to smallholder farmers. The questionnaire 
gathered information on socio-economic, technical, and 
institutional characteristics of  smallholder farmers. The 
study employed a multi-stage sampling technique to 
identify adoptors and non-adopters of  AI technology. 
A sample size of  378 was randomly selected. This study 
utilized the formula by Kothari (2004) to estimate the 
sample size as follows: 

Where,
N = population Size; n = the desired sample size; z – 
Standard normal deviate at 0.05 significance level; p – the 
proportion in the target population estimate to have a 
characteristic being measured (an assumption of  p =0.5 
was made); q = 1 - p; e = level of  statistical significance 
desired; z - statistic at 95% confidence level is 1.96.
The level of  significance being 0.05 therefore; the sample 
size, n was estimated as: 

Data Processing and Analysis
After data collection, cleaning of  the data was done using 
Microsoft Excel software. STATA software version 15 
was utilized to analyze quantitative data collected from 
the household survey. Both descriptive and inferential 
statistics were applied to the data to identify factors 
influencing the adoption of  AI technology and intensity 
of  use in Alego-Usonga. 
The Cragg (1971) Double Hurdle model was employed in 
this study to analyze factors influencing the adoption and 
the intensity of  use of  AI technology in Alego Usonga 
Sub County. According to Cragg (1971), adoption is a 
process involving two stages/tiers; the first is decision 
on whether or not to adopt the technology, and second 
is to what extent to adopt. The model assumes that the 
decision not to adopt is a deliberate choice, thus the zeros 
from non-participants are considered as corner solution 
in the utility maximizing model. The model curbs bias 
in the continuous second tier dependent variable by 
linking a value to the piled up data, thus maintaining all 
the data within the sample. The Cragg model is flexible, 
assuming that there are no restrictions regarding the 
components of  independent variables in each estimation 
stage. The model requires a joint application of  the 
probit and truncated regression models, sequentially or 
simultaneously. The probit model equation that was used 
in the study is given as: 

Where subscript i is the ith household, is the latent discrete 
adoption choice variable, Yi

* is the observed adoption 
variable which takes a value of  1 if  the farmer adopted AI 
technology and 0 otherwise, Xi is Kx1 vector of  factors 
that influence AI technology adoption, βi is a Kx1 vector 
of  parameters to be estimated, ε is the error term.
These coefficients βi represent changes in the latent 
variable (unobserved z-score) underlying the binary 
outcome and thus, the effect of  each predictor on the 
latent propensity (z-score) to adopt AI. The change in 
probability that a farmer will adopt AI or not arising from 
the change in the variables Xi depends on their effect on 
the cumulative normal distribution function (Φ) evaluated 
at the mean predictor values.The marginal effect of  an 
independent variable Xk on the probability is:

Where, Φ is the standard normal density
The PDF of  the standard normal at βX is:

Marginal effect of  change in Xx= Φ(βX) x βk 
In the second hurdle, the truncated regression model was 
employed to determine factors that influence the intensity 
of  AI technology adoption among farmers who adopted 
the technology. The truncated regression is given by:

Where, Ii is the intensity of  AI technology adoption, and 
depends on Ii

* is the latent variable being greater than 
zero on the condition that a decision is made to adopt AI 
technology.Xi is a vector of  parameters to be estimated 
for the intensity of  adoption, is a vector of  factors that 
influence the intensity of  use of  AI technology, μ is the 
error term which has a normal distribution.
The double-hurdle regression equation is specified as 
follows:

....(1)

....(4)

....(5)

....(6)

....(7)

....(3)

....(2)

Where, Yi is the AI adoption which takes a value of  1 
for adopters and 0 otherwise, Ii is intensity of  use of  
AI measured by the number of  times a farmer used the 
technology, βi is a vector of  parameters to be estimated,
Xi is vector of  the explanatory variables, and εi is the error 
term.

RESULTS AND DISCUSSIONS
Results in Table 1 presents the descriptive statistics for 
the variables used in the analysis. The results indicate 
that, on average, the majority of  small-scale dairy farmers 



Pa
ge

 
18

https://journals.e-palli.com/home/index.php/ajfst

Am. J. Food. Sci. Technol. 4(2) 15-25, 2025

were aged 40-49 years , with 46% being male. Most of  
the farmers had attained primary-level education and 
had more than five years of  experience in dairy farming. 
The age and education level of  the farmers suggests that 
the typical dairy farmer in this sample is a mature adult 
with basic to moderate educational attainment, which 
can influence their ability to adopt new technologies and 
management practices. The gender distribution (mean 
of  0.46) indicates a relatively balanced representation 
of  men and women, though slightly more women may 
be involved. On average, 39.5% of  male small-scale 
dairy farmers were members of  various farmer groups 
that is relatively low, which could affect access to shared 
resources, training, and collective bargaining for better 
prices or services. The average herd size is 3.7, with a high 
standard deviation (2.08), indicating significant variation 
in the scale of  dairy operations among households. Milk 
sales average at Kshs. 455.5, but with a large standard 
deviation (267.8), showing that while some households 
are quite productive, others are less so, possibly reflecting 
differences in whether a farmer has improved or 
indigenous breed, differences in herd size, management 

skills, or market access.
The costs associated with breeding among the dairy 
farming are notable, with artificial insemination (AI) 
services of  about Kshs. 2000 and the cost of  a bull was 
Kshs. 854.5, which a high standard deviation of  Kshs. 
376.3 for the cost of  the bull. The high standard deviation 
pints to lack of  standard prices for breeding that expose 
farmers to high charges from scrupulous breeders and the 
high financial investment required for herd improvement 
and breeding. 
The average worker’s skill in heat detection is 1.6 (in a 
scale of  1-4, 1 =poor, 2=fair, 3=good, 4=excellent) with 
considerable variability (std. error = 0.7), suggesting that 
most farmers are considered to have some knowledge 
on heat detection but majority need additional training. 
This skill is crucial for effective breeding and maximizing 
milk production, underlining the importance of  capacity 
building in dairy farm management. This particular 
variable for the success of  AI technology given that a 
farmers ought to have an excellent heat detection skill to 
ensure the inseminator is called in on the right time for 
successful breeding.

Table 1: Descriptive Statistics and Measurement of  Variables Influencing Adoption and Intensity of  AI
Variable (Characteristics of  
household head)

Measurement Sample Mean Std. Dev

Age 1 = 20 – 29 years, 2 = 30 – 39 years, 3 = 40 
– 49 years, 4 = above 50 years

3.3 0.76

Gender 1 = Male, 0 = Female 0.46 0.5
Education level 1 =non-formal, 2=primary, 3=secondary

4 = college/university
2.3 1.0

Years of  experience in dairy farming 1 = less than 1 year, 2= 1 – 5 years, 3= 
more than 5 years

2.7 0.5

Group membership 1 = yes, 0 = no 0.4 0.49
Herd size In numbers 3.7 2.08
Milk Sales In Ksh 455.5 267.8
Cost of  AI service 1 = Ksh. 1000 – Ksh. 2000, 2= above Ksh. 

2000
1.8 1.3

Cost of  bull Cost in Ksh 854.5 376.3
Worker’s skill on heat detection 1 = poor, 2 = good, 3 = very good

4 = excellent
1.6 0.7

Semen type 1 = poor, 2 = good, 3 = very good 1.8 0.4
Perception about AI reliability If  it is reliable

1 = yes, 0 = no
0.45 0.5

Supplements use If  they give supplement feeds
1 = yes, 0 = no

0.9 0.3

Farmer access to extension Access to extension service
1 = yes, 0 = no

0.9 0.3

Farmer has attended atlease one 
farmer training 

Attended any training 
1 = yes, 0=no

0.72 0.45

Farmers access other support services 
(apart from extension and training)

If  support services are available
1 = yes, 0 = no

0.07 0.25

Information on availability of  
Inseminator 

If  the inseminator is available
1 = yes, 0 = no

0.9 0.3

Source: Author’s Computation (2025)



Pa
ge

 
19

https://journals.e-palli.com/home/index.php/ajfst

Am. J. Food. Sci. Technol. 4(2) 15-25, 2025

The results also shows high access to extension services 
by about 90% of  the farmers. But this is surprising 
given that the although there is a high extension access, 
the perception about the reliability of  AI remained low 
(45%). This points out at the need to re-look at the mode 
of  extension service, the packaging of  the extension 
information and assess the factors that influence the 
effectiveness of  the service. Results showed that 72% 
of  the farmers had attended at least one training by the 
extension service. Only 7% of  farmers access other 
support services beyond extension and training. There is 
a high preference of  non-sexed semen of  1.8 (in a scale 
of  1=sexed, 2=non sexed) pointing out to the limiting 
possibility of  high cost of  sexed semen. Results showed 
that there was a high awareness among farmers (90%) on 
the availability of  AI service within reach of  the farmers 
meaning that the adoption and non-adoption of  AI was 
deliberate decisions by farmers. 
 
Factors Influencing Adoption of  AI Technology
Results of  the probit regression from using Equation 
3 used to estimate the factors influencing the adoption 
of  AI technology in the study area is provided in Table 
2 below. The results revealed that ten out of  seventeen 
variables considered and analyzed were statistically 
significant and influenced adoption of  AI technology in 
the study area. Age, education level and experience of  the 
household head, milk sales, cost of  AI service, cost of  
use of  bull service, worker’s skill on heat detection, semen 
type, perception about AI reliability and information 
on availability of  inseminator positively influenced AI 
technology adoption. Further, only training negatively 

influenced AI technology adoption in the study area that 
calls for a further analysis of  the effectiveness of  training 
in achieving the intended outcomes of  promoting genetic 
improvements of  dairy herd in the study area. Training 
may negatively influence artificial insemination (AI) 
adoption if  it is poorly designed, irrelevant, or not tailored 
to farmers’ needs and socio-economic circumstances. 
Poorly designed training my lower farmers’ confidence in 
AI technology and may cause resistance, confusion and 
mistrust of  the technology or the AI service and service 
providers. 
Age was established to have a positive effect and 
statistically significant at 5% level of  significance in 
explaining the adoption decision of  AI technology. 
This showed that a shift to the higher next age category 
increases the probability of  adopting AI technology 
by 25.3%. The findings suggests that older farmers are 
more likely to adopt AI technology. The argument is 
that older farmers have more experience, have a better 
access to resources, or a greater willingness to invest in AI 
technology due to accumulated resources and knowledge. 
The current finding is in line with those on factors for 
adoption of  artificial insemination technology among 
pig farmers by Sharma et al. (2020) who found out that 
age of  the household head positively and significantly 
determined the adoption of  AI technology in small scale 
pig production systems in India. The current finding 
is also consistent with those on factors influencing 
adoption of  AI by smallholder livestock farmers in 
dryland production systems of  Kenya by Abot (2020) 
who reported a positive influence of  age on the adoption 
of  AI technology.

Table 2: Probit Regression Results on Factors Influencing AI technology Adoption
Factors influencing 
adoption of  AI (Xi)

Coef. (βi) St. Err. t-value p-value [95% Conf Interval] Marginal 
effects 

Age 0.253** 0.102 2.49 0.013 0.054 0.452 0.105
Gender 0.095 0.186 0.51 0.609 -0.269 0.459 0.039
Education level 0.201*** 0.03 6.80 0.000 0.143 0.258 0.083
Years of  experience in dairy 
farming 

0.121** 0.059 2.05 0.041 0.005 0.237 0.050

Group membership 0.154 0.212 0.73 0.468 -0.262 0.569 0.064
Herd size -0.018 0.045 -0.39 0.696 -0.106 0.071 -0.007
Milk sales 0.001*** 0.000 2.94 0.003 0.000 0.002 0.000
Cost of  AI service 0.542*** 0.205 2.64 0.008 0.14 0.943 0.225
Cost of  bull 0.001* 0.000 1.67 0.094 0.000 0.001 0.000
Worker’s skill 0.198** 0.099 1.99 0.047 0.003 0.393 0.082
Semen type 0.345*** 0.086 4.03 0.000 0.178 0.513 0.143
Perception about AI 
reliability

1.862*** 0.289 6.45 0.000 1.296 2.428 0.772

Supplements 0.283 0.271 1.04 0.296 -0.248 0.815 0.117
Farmers access to extension 
access

-0.332 0.29 -1.15 0.252 -0.900 0.236 -0.138

Farmers attend at least one 
farmer training 

-0.496** 0.225 -2.20 0.028 -0.937 -0.054 -0.206



Pa
ge

 
20

https://journals.e-palli.com/home/index.php/ajfst

Am. J. Food. Sci. Technol. 4(2) 15-25, 2025

Information about 
inseminator availability

0.85*** 0.241 3.53 0.000 0.378 1.323 0.353

Farmers access other support 0.204 0.379 0.54 0.59 -0.538 0.946 0.085
Constant -4.239*** 1.095 -3.87 0.000 -6.385 -2.093
Mean dependent var 0.696 SD dependent var 0.461
Pseudo r-squared 0.357 Number of  obs 332
Chi-square 145.649 Prob > chi2 0.000
Akaike crit. (AIC) 300.309 Bayesian crit. (BIC) 372.606

βX = 1.33 ϕ(X) = 0.414
*** p<.01, ** p<.05, * p<.1 denotes significance at 1%, 5% and 10%, respectively.

Education level of  the household head was found to 
have a positive effect and statistically significant at 1% 
level of  significance in explaining the adoption decision 
of  AI technology. This showed that an increase in 
education level is likely to increase the probability of  AI 
technology adoption by 3.9%. Studies have shown that 
education increases the awareness and promotes attitude 
change which creates a favourable environment for 
technology adoption (Mwanga et al., 2019; Okello et al., 
2021). Farmers with high levels of  education are likely to 
have more knowledge and skills and hence their higher 
probability of  adopting AI technology compared to those 
with low levels of  education. The current finding are 
consistent with those on the determinants of  utilization 
of  agricultural technologies among smallholder dairy 
farmers in Kenya by Okello et al. (2021) who established 
that education level positively influenced the utilization of  
AI. Similarly, a study on adoption of  artificial insemination 
and the intensity of  use in Ethiopia by Gebre et al. (2022) 
reported a positive relationship between the households’ 
level of  literacy and adoption and intensity of  use of  AI.
Years of  experience in dairy farming was established to 
have positive influence on AI technology adoption at 5% 
level of  significance. This means that for every additional 
years in dairying experience, the likelihood of  adopting 
AI technology increases by 5.0%. Experienced farmers 
may have observed or learned about the benefits of  
AI over time which enhance their chances of  adopting 
the technology. Additionally, longer involvement in 
dairy farming may provide farmers with better financial 
resources to afford AI services. The current finding is 
consistent with the findings on factors for adoption of  
artificial insemination technology among pig farmers 
by Sharma et al. (2020) who found out that farming 
experience positively and significantly determined the 
adoption of  AI technology in small scale pig production 
systems in India. The current result is also similar to those 
on factors affecting the use of  artificial insemination of  
farmers in dairy farming in Turkey by Özsayın (2020) 
who reported that dairy farming experience had a positive 
effect on the use of  artificial insemination. However, other 
studies have reported negative influence of  experience 
on adoption of  technologies. Kaaya et al. (2005) reported 
that experience negatively influences the utilization of  AI 
in Uganda. Similarly, a study on adoption and intensity of  

improved fish feeds use in Western Kenya by Wafula et al. 
(2021) established that experience negatively influenced 
the intensity of  using improved fish feeds in Kenya. It is 
imperative that as farmers accumulate years of  experience 
in dairy farming, they experiment several strategies in 
their quest to improve earnings, including improving 
dairy herd using AI. Farmers adopt and permanently 
use new agricultural technologies when they test them 
and find them working and perceive clear economic and 
practical benefits (Castellini et al. 2025). 
Milk sales was found to have a positive effect and 
statistically significant at 1% level of  significance in 
explaining the AI technology adoption decision. This 
implies that a 1 Ksh increase in milk sales is likely to 
influence the probability of  adopting AI technology by 
a very low percentage (<0.1%). Income from milk is an 
incentive and will determine the probability that a farmer 
will use AI technology or not. Farmers who receive low 
incomes are not motivated to use this technology since 
they do not realize much from dairy farming as compared 
to those who earn more. The current finding is consistent 
with the findings by Tefera (2013) who reported that 
income from milk sales positively influenced the farmer 
decision to adopt AI in Ethiopia. A study on multi-
country investigation of  factors influencing breeding 
decisions by smallholder dairy farmers in sub-Saharan 
Africa by Mwanga et al. (2019) reported that income from 
selling dairy products positively influenced the use of  AI 
in Ethiopia and negatively influenced the choice of  AI as 
a breeding option in Tanzania, Uganda, and Kenya.
The cost of  AI was found to be statistically significant 
at 1% level of  significance and positively influenced AI 
technology adoption by 22.5%. The cost of  any dairy 
technology will always determine its uptake and hence if  
the technology is affordable it will be embraced by the 
farmer, and if  it is expensive only few farmers will adopt 
it. The current finding is divergent to the findings on the 
adoption of  improved technologies and profitability of  the 
catfish processors in Ondo State, Nigeria by Olutumise et 
al. (2020) who reported that cost of  equipment negatively 
influenced the decision to adopt and the rate of  adopting 
improved fish processing technologies in Nigeria. The 
current findings are also inconsistent with those on multi-
country investigation of  factors influencing breeding 
decisions by smallholder dairy farmers in sub-Saharan 



Pa
ge

 
21

https://journals.e-palli.com/home/index.php/ajfst

Am. J. Food. Sci. Technol. 4(2) 15-25, 2025

Africa by Mwanga et al. (2019) who found out that cost 
of  AI service negatively influenced the choice of  AI as 
a breeding option. The positive relationship between 
cost of  AI and the probability of  its adoption show that 
adoption decisions are strongly influenced by the expected 
profitability and cost-benefit considerations. While high 
costs is expected to be a barrier, farmers are more likely 
to evaluate benefits and costs of  a technology and adopt 
technologies with higher costs if  these technologies 
promise greater returns or efficiency gains that justify the 
investment (Pope & Sonka 2020).
Worker’s skill on heat detection was statistically significant 
at 1% level of  significance and positively influenced 
the probability of  adopting AI technology by 8.2%. 
Successful conception is determined by proper heat 
timing and timely insemination. Accurate heat detection 
leads to better reproductive management, shorter calving 
intervals, and increased milk production. A worker with 
good skill on heat detection will enhance the chances of  a 
successful conception as compared to the one with poor 
skills. Additionally, the worker spends most of  his/her 
time with the cows and monitors the behaviour of  the 
animal and hence is able to detect the possibilities that it 
is on heat or not. The success of  AI technology depends 
majorly on the ability of  the farmer to detect heat and 
invite an inseminator on time. Without the skills, the 
success rate of  AI can be highly compromised leading to 
economic losses and mistrust of  the technology. 
Semen type was statistically significant at 1% level of  
significance and positively influenced the adoption of  AI 
technology by 14.3%. There are various types of  semen 
available to the farmers for the select for insemination. 
It is important to note that farmers made deliberate 
decisions to adopt Ai service based on expected benefits. 
The sexed semen was adopted by farmers who had 
perceived clear pathway to improving their dairy herd 
and milk production and are likely to continuously use AI 
service. Moreover, farmers who used sexed semen were 
likely to be knowledgeable also preferring semen from 
genetically superior bulls that offer traits such as higher 
milk production, disease resistance, or faster growth 
thereby obtaining additional benefits from AI and help 
further promote adoption of  AI.
Perception about AI reliability was statistically significant 
at 1% level of  significance and positively influenced the 
adoption of  AI technology by 77.2%. When farmers 
believe and trust the capability of  a technology to improve 
the performance of  their enterprises they are likely to 
adopt and continue using it. This could be attributed to 
the fact that farmers like trying out technologies that are 
believed to benefit them in the long-run. Additionally, 
farmers are likely to adopt a given technology if  it is 
being promoted by the government and trusted agencies. 
The perception about reliability is linked to access to 
extension service, farmer training and the availability of  
information about AI service and the associated benefits. 
Farmers are more likely to adopt agricultural technologies 
if  they receive correct information through the right 

sources (Livondo et al. 2015).
Training on livestock production was established to be 
statistically significant at 5% level of  significance and 
negatively influenced AI technology adoption by 20.6%. 
Training impacts knowledge to farmers about existing 
technologies and hence enabling them to use them. 
The current finding is divergent to those on factors 
affecting small dairy farmers’ adoption and intensity 
of  artificial insemination technology in Ethiopia by 
Herana and Kumari (2017) who reported that training 
positively influenced adoption of  AI. The current 
finding is also in line with the finding by Sharma et al. 
(2020) who reported that participation in training and 
demonstration programmes positively influenced the 
adoption of  AI technology in small scale pig production 
systems in India. Moreover, the findings on adoption of  
artificial insemination service for cattle crossbreeding by 
smallholder farmers in Ethiopia by Abraha et al. (2020) 
who established that formal training positively influenced 
AI technology adoption contradict the current result.
Inseminator availability was statistically significant at 1% 
and positively influenced the probability of  adopting AI 
technology by 35.3%. This finding could be attributed 
to the fact that most farmers stated that the inseminator 
was available when called. Conception of  a cow once 
heat signs are detected was time bound and tend to 
put farmers in a panic model. The information about 
availability determines chances for the farmer inviting 
the inseminator. Supporting this findings are those on 
factors affecting adoption of  artificial insemination 
technology by dairy farmers in Tanzania by Temba (2011) 
who revealed that proximity to AI service providers and 
access to information were significant factors affecting 
AI adoption by dairy farmers.
Factors influencing the Intensity of  AI Technology Use
Table 3 presents the truncated regression results on 
the factors influencing the intensity of  AI technology 
adoption in the study area. The table of  results shows that 
nine out of  seventeen variables considered and analyzed 
were statistically significant and influenced the intensity 
of  AI technology adoption on the level in the study 
area. Among the variables that positively influenced the 
intensity of  AI technology include; age, education level, 
experience, milk income, and training. Further, group 
membership, AI cost, worker’s skill on heat detection, and 
availability of  the inseminator negatively influenced the 
intensity of  AI technology use.
Age positively influenced the intensity of  using AI 
technology at 1% level of  significance. This shows 
that a year increase in age increases the intensity of  AI 
technology use by 5%. The argument is that as farmers 
increase in age their experience also adds up. Farmers 
with a good history with AI use are likely to intensify 
as opposed to those with bad experiences. The current 
finding conform those by Chen et al. (2020) who reported 
a positive effect of  age on the intensity of  tea consumption 
among men and women in China. The current finding 
however is divergent to finding by Mahoussi et al. (2021) 



Pa
ge

 
22

https://journals.e-palli.com/home/index.php/ajfst

Am. J. Food. Sci. Technol. 4(2) 15-25, 2025

who found out that age square had a negative quadratic 
relationship with the intensity of  use of  improved maize 
seeds in Benin.
Education level was established to have a positive 
influence on the intensity of  AI technology use at a 1% 
significance level. This implies that for each additional 
year in education, the intensity of  using AI technology 
increased by 4.2%. This could be attributed to the fact 
that education improves the knowledge and skills of  the 
farmer. The current finding is consistent with the findings 
on adoption of  artificial insemination technology and 
its intensity of  use in Ethiopia conducted by Gebre et 
al. (2022) who reported that literacy level increases the 
intensity of  AI technology adoption. The current findings 
also conform to the finding by Mahama et al. (2020) who 
noted that education level positively influenced the extent 
of  AI adoption. However, the current finding contradicts 
the findings on factor influencing adoption of  AI by 
smallholder farmers in dryland production systems of  
Kenya by Abot (2020) who found that education level 
negatively influenced the extent of  AI adoption.
Experience was also found to positively influence 
the intensity of  AI technology use at 1% level of  
significance. This means that a year increase in dairy 
farming experience is likely to increase the intensity of  
AI technology use by 5.8%. This could be attributed to 
the fact that more experienced farmers tend to continue 
employing technologies that they perceive beneficial with 
time. The current finding are consistent with the findings 
by Olutumise et al. (2020) who reported a positive 
influence of  experience on the intensity of  adoption of  
improved technologies and profitability of  the catfish 
processors in Nigeria. The current finding is divergent 
from the findings by Tefera (2013) who reported that 
experience of  keeping crossbred cattle in the past years 
had negative effect on the extent of  adoption of  AI in 
Ethiopia. The current findings also contradict those on 
factors influencing adoption decision of  AI technology 
and the extent of  adoption by Bayan (2018) who found 
out that with one additional year older from start of  a 
dairy farm, the probability of  adoption and intensity of  
adoption goes down.
Group membership negatively influenced the intensity of  
AI technology use by 3.8% at a 5% level of  significance. 

Cooperatives offer a range of  benefits to its members 
including marketing of  output, inputs, education and 
new technologies. The decision to adopt and intensify 
dairy technologies may be independent of  the influence 
of  cooperatives. The current findings is consistent 
with the finding on the determinants of  utilization 
of  agricultural technologies among smallholder dairy 
farmers in Kenya by Okello et al. (2021) who reported a 
negative influence of  group membership on utilization 
of  dairy technologies in Kenya. The current finding is 
also convergent to those on adoption and intensity of  
improved fish feeds use in Western Kenya by Wafula et al. 
(2021) who reported that group membership negatively 
influenced the adoption and intensity of  using improved 
fish feeds. The current finding are inconsistent with the 
finding on the determinants of  artificial insemination use 
by smallholder dairy farmers in Ethiopia by Tefera (2013) 
who reported that being a member of  dairy cooperative 
positively influenced the extent of  AI use. The current 
findings are also divergent to the findings on factors 
affecting adoption of  artificial insemination technology 
by dairy farmers in Tanzania by Temba (2011) who 
reported that group membership positively influenced 
adoption and intensity of  AI.
Training was established to have a positive influence 
on the intensity of  AI technology use at a 10% level 
of  significance. This means that attending trainings 
increases the intensity of  AI technology use by 5.6%. 
The argument is that farmers who have been trained have 
better access to information and agricultural knowledge 
about dairy farming as opposed to their non-trained 
counterparts. The current finding is consistent with the 
findings on factors affecting small dairy farmers’ adoption 
and intensity of  artificial insemination technology in 
Ethiopia by Herana and Kumari (2017) who reported 
that access to AI training positively influenced intensity 
of  AI in Ethiopia. The current finding also conforms 
to the findings by Gebre et al. (2022) who reported 
that access to training positively influences the intensity 
of  AI technology adoption in Ethiopia. However, a 
study by Dumara and Zenbaba (2020) established that 
attendance in training had a positive and significant effect 
on the adoption and intensity of  adoption of  malt barley 
technology in Ethiopia.

Table 3: Truncated Regression Results on Factors Influencing Intensity of  AI Technology Adoption
AI Intensity factors Xi Coef. βi St.Err. t-value p-value [95% Conf Interval]
Age 0.05*** 0.013 3.80 0.000 0.024 0.076
Gender -0.023 0.016 -1.46 0.145 -0.054 0.008
Education level 0.042*** 0.01 4.44 0.000 0.024 0.061
Experience 0.058*** 0.017 3.40 0.001 0.025 0.091
Group membership -0.038** 0.017 -2.17 0.03 -0.072 -0.004
Herd size -0.003 0.004 -0.76 0.448 -0.011 0.005
Milk sales 0.000** 0.000 2.07 0.039 0.00 0.00
AI cost -0.029* 0.016 -1.80 0.071 -0.06 0.002
Cost of  bull 0.000 0.000 0.75 0.456 0.00 0.00



Pa
ge

 
23

https://journals.e-palli.com/home/index.php/ajfst

Am. J. Food. Sci. Technol. 4(2) 15-25, 2025

Inseminator availability was found to have a negative 
influence on the intensity of  AI technology use by 4.8% 
at a 5% level of  significance. This could be attributed to 
some cases of  dishonesty by the inseminators. Farmers 
who have experienced any form of  cheating from the 
service providers are less likely to intensify the adoption 
of  AI technology (Kaaya et al., 2005). The findings by 
Abot (2020) on factor influencing adoption of  AI by 
smallholder farmers in dryland production systems of  
Kenya, who reported that dishonesty from the service 
providers negatively influenced the intensity of  AI 
adoption supports our findings.

CONCLUSION
The study highlights key factors influencing AI adoption 
among small-scale dairy farmers, emphasizing the role 
of  age, education, farming experience, and economic 
considerations. Older and more experienced farmers 
were more likely to adopt AI technology, while education 
and experience played a crucial role in both adoption 
and intensity of  AI technology use. Despite having 
relatively small herds, farmers exhibited significant 
variations in milk productivity and income, with high 
AI costs presenting a financial barrier. Challenges such 
as poor heat detection skills, limited awareness of  semen 
types, and concerns over AI reliability further impacted 
adoption rates. Access to extension services and training 
was generally high, yet government support remained 
minimal. Interestingly, while inseminator availability 
was not a major issue, group membership and livestock 
production training negatively influenced AI intensity, 
suggesting that certain collective or traditional farming 
practices may deter AI usage. The findings underscore 
the need for targeted interventions, including improved 
farmer education, financial support, and enhanced AI 
service reliability, to increase AI adoption and maximize 
its benefits in small-scale dairy farming.
The single important factor influencing adoption of  
AI services was the perception about the reliability of  
AI services where a unit change in this variable would 

increase probability of  farmers adopting AI by 77.2%. 
Other factors with high impact in influencing adoption 
are information about inseminator availability (35.3%) 
and access to extension service (20.6%). The benefit will 
be increased awareness and positive impact on adoption 
of  artificial insemination technology in the study area. 
Farmers should be encouraged to form more farmer 
groups by promoting the benefits that are likely to be 
accrued when people form groups, such as markets for 
their outputs, education and training, and dissemination 
of  dairy technologies such as AI. Farmers should 
enhance the skills of  their workers by allowing them 
also to attend trainings on livestock production and AI 
technology. Enhance training effectiveness by conduct 
training needs assessments before the trainings are 
carried out so as to capture the farmers’ interest together 
with the environment. The government should support 
AI activities by subsidizing the cost of  AI and feeds to 
individual farmers and groups, funding trainings and 
workshops, and provide capacity building for the trainers. 

REFERENCES
Abot, D. M. (2020). Factors Influencing Adoption of  Artificial 

insemination by Smallholder Livestock Farmers in Dryland 
Production Systems of  Kenya. Unpublished MSc. Thesis; 
University of  Nairobi.

Abraha, B., Muluken, G. and Jemal, Y. (2020). Adoption 
of  Artificial Insemination Service for Cattle 
Crossbreeding by Smallholder Farmers in Laelay-
Maichew District. Tigray, Ethiopia. Journal of  
Development and Agricultural Economics, 12(2), 104 – 112. 
https://doi.org/10.5897/JDAE2020.1183 

Agricultural Sector Development Support Programme 
(ASDSP). (2014). Report on milk production rates in Kenya. 
Ministry of  Agriculture, Livestock and Fisheries, Kenya. 

Ayantunde, A. A., Fernandez-Rivera, S., Hiernaux, P. H., 
& Tabo, R. (2008). Implications of  restricted access 
to grazing by cattle in the wet season in the Sahel. 
Journal of  Arid Environments, 72(5), 523–533. https://
doi.org/10.1016/j.jaridenv.2007.06.006 

Worker’s skill -0.016* 0.009 -1.83 0.067 -0.034 0.001
Semen type -0.003 0.007 -0.38 0.702 -0.016 0.011
AI reliability -0.019 0.02 -0.95 0.343 -0.057 0.02
Supplements use 0.03 0.023 1.27 0.204 -0.016 0.075
Extension access -0.021 0.023 -0.93 0.353 -0.067 0.024
Training 0.056*** 0.021 2.61 0.009 0.014 0.098
Inseminator availability -0.048** 0.021 -2.26 0.024 -0.09 -0.006
Government support 0.02 0.031 0.66 0.506 -0.04 0.08
Constant 1.096*** 0.09 12.18 0.000 0.92 1.272
Sigma 0.137*** 0.005 25.77 0.000 0.127 0.147
Mean dependent var 1.021 SD dependent var 0.144
Number of  obs 332 Chi-square 33.543
Prob > chi2 0.014 Akaike crit. (AIC) -338.117

*** p<.01, ** p<.05, * p<.1 denotes significance at 1%, 5% and 10%, respectively.



Pa
ge

 
24

https://journals.e-palli.com/home/index.php/ajfst

Am. J. Food. Sci. Technol. 4(2) 15-25, 2025

Bayan, B. (2018). Factors Influencing Extent of  Adoption 
of  Artificial Insemination (AI)  Technology among 
cattle farmers in Assam. Indian Journal of  Economics 
and Development, 14(3), 528 – 534. https://doi.
org/10.5958/2322-0430.2018.00166.X 

Castellini, G., Raffaelli, S., Mancinelli, V., Boncinelli, F., 
Corsi, C., & Gallerani, G. (2025). Determinants of  
consumer and farmer acceptance of  new production 
technologies: A systematic review. Frontiers in 
Sustainable Food Systems, 5, 1557974. https://doi.
org/10.3389/fsufs.2025.1557974 

Chen, L., Guan, X., Zhuo, J., Han, H., Gasper, M., 
Doan, B., Yang, J., & Ko, T.-H. (2020). Application 
of  double hurdle model on effects of  demographics 
for tea consumption in China. Journal of  Food Quality, 
2020, 1–6. https://doi.org/10.1155/2020/9862390 

Cragg, J. G. (1971). Some statistical models for limited 
dependent variables with application to the demand 
for durable goods. Econometrica, 39(5), 829–844. 
https://doi.org/10.2307/1909582

Directorate of  Livestock Production and Veterinary, 
Ministry of  Agriculture, Kenya. (2019). Climate report 
for Western Kenya livestock production. Government 
Printer. 

Dumara, A., & Zenbaba, O. S. (2020). Factors affecting 
adoption and its intensity of  malt barley technology 
package in Malga Woreda, Southern Ethiopia. Journal 
of  Agricultural Extension and Rural Development, 6(1), 
1–2. https://doi.org/10.11648/j.ijae.20160103.15 

Food and Agriculture Organization of  the United 
Nations. (2016). The global dairy sector: Facts. FAO.

Gahakwa, D., Asiimwe, T., Nabahungu, N. L., Mutimura, 
M., Isibo, T., Mutaganda, A., & Ngaboyisonga, C. 
(2014). A decade of  agricultural research in Rwanda: 
Achievements and the way forward. In B. Vanlauwe 
et al. (Eds.), Challenges and opportunities for agricultural 
intensification of  the humid highland systems of  Sub-
Saharan Africa (pp. 69–80). Springer. https://doi.
org/10.1007/978-3-319-07662-1_6 

Gebre, Y. H., Gebru, G. W., & Gebre, K. T. (2022). 
Adoption of  artificial insemination technology and its 
intensity of  use in Eastern Tigray National Regional 
State of  Ethiopia. Agriculture & Food Security, 11(44). 
https://doi.org/10.1186/s40066-022-00384-3 

Henning, J. I. F., Mare, F. A., & Willemse, B. J. (2010). 
Profitability analysis of  different reproduction methods 
with Dohne Merinos. Paper presented at the AEASA 
Conference 2010, Cape Town, South Africa.

Herana, T., & Kumari, S. (2017). Factors affecting small 
dairy farmers’ adoption and intensity of  adoption of  
artificial insemination technology: A case study of  
Southern Ethiopia. International Journal of  Agricultural 
Science and Research, 7(6), 335-346. https://journals.
indexcopernicus.com/api/file/viewByFileId/191667 

Howley, P., O’Donoghue, C., & Heanue, K. (2012). 
Factors affecting farmers’ adoption of  agricultural 
innovations: A panel data analysis of  the use of  
artificial insemination among dairy farmers in Ireland. 

Journal of  Agricultural Science, 4(171). https://doi.
org/10.5539/jas.v4n6p171 

Ingabire MC, Liu Y, Pesha JC, Hardi A (2018). Factors 
affecting adoption of  artificial insemination 
technology by small dairy farmers in Rwanda. A 
case of  Rwamagana district. Journal of  Economics 
and Sustainable Development 9(12):46-53. https://doi.
org/10.7176/JESD/9-12-06 

Kaaya, H., Bashasha, B., & Mutetikka, D. (2005). 
Determinants of  utilization of  AI services among Ugandan 
dairy farmers. Department of  Veterinary Services and Animal 
Industry. Department of  Agricultural Economics 
and Agribusiness. Faculty of  Agriculture, Makerere 
University, Kampala, Uganda. pp. 34 - 43.

KALRO. (2024). Livestock. Retried from https://www.
kalro.org/divisions/livestock/ 

Kimunya, E.G. (2014). Effects of  Patterns of  Adoption of  
Dairy Farming Technologies among Small-Scale Farmers. 
Nairobi, Kenya: University of  Nairobi.

Kiplagat, M. D., Cheruiyot, R., Matura, F. K., Kimani, F., & 
Gitau, M. G. (2025). The status of  dairy development 
in the highly dairy and potential dairy counties in 
Kenya. Journal of  Agriculture Science & Technology, 24(1), 
122–138. https://doi.org/10.4314/jagst.v24i1.7 

KNBS. (2019). 2019 Kenya Population and Housing Census, 
Volume II: Distribution of  Population by Administrative 
Units. KNBS, Nairobi.

Livondo, J. L., Kipkoech, A., & Macharia, E. W. (2015). 
Factors affecting communication channels preference 
by farmers in adoption of  agricultural technology for 
Striga control: A case of  Bungoma County, Kenya. 
Journal of  Current Research in Agricultural Science, 11(2), 
45-58. https://doi.org/10.5897/JAERD2015.0697 

Mahama, A., Awuni, J. A., Mabe, F. N., & Azumah, S. 
B. (2020). Modelling adoption intensity of  improved 
soybean production technologies in Ghana: A 
generalized Poisson approach. Heliyon, 6(7), e03543. 
https://doi.org/10.1016/j.heliyon.2020.e03543  

Mahoussi, F. E., Adegbola, P. Y., Aoudji, A. K. 
N., Kouton-Bognon, B., & Biaou, G. (2021).  
Modeling the adoption and use intensity of  improved 
maize seeds in Benin, West Africa: A double-hurdle 
approach. African Journal of  Food, Agriculture, Nutrition 
and Development, 21(4). https://doi.org/10.18697/
ajfand.99.20520 

Mwanga, G., Mujibi, F. D. N., Yonah, Z. O., & Chagunda, 
M. G. G. (2019). Multi-country investigation of  
factors influencing breeding decisions by smallholder 
dairy farmers in sub-Saharan Africa. Tropical animal 
health and production, 51(2), 395–409. https://doi.
org/10.1007/s11250-018-1703-7 

Odero-Waitituh, J. A. (2017). Dairy farming in Kenya: 
An analysis of  resource allocation, productivity and 
profitability. Journal of  Agricultural Extension and Rural 
Development, 9(7), 151–161. https://doi.org/10.5897/
JAERD2017.0869 

Okello, D., Owuor, G., Larochelle, C., Gathungu, E., 
& Mshenga, P. (2021). Determinants of  utilization 



Pa
ge

 
25

https://journals.e-palli.com/home/index.php/ajfst

Am. J. Food. Sci. Technol. 4(2) 15-25, 2025

of  agricultural technologies among smallholder 
dairy farmers in Kenya. Journal of  Agriculture and 
Food Research, 6, 100213. https://doi.org/10.1016/j.
jafr.2021.100213 

Olutumise, A. I., Adene, I. C., Ajibefun, A. I., & Amos, 
T. T. (2020). Adoption of  improved technologies 
and profitability of  catfish processors in Ondo 
State, Nigeria: A Cragg’s double-hurdle model 
approach. Scientific African, 10, e00576. https://doi.
org/10.1016/j.sciaf.2020.e00576 

Özsayın, D. (2020). Factors affecting the use of  artificial 
insemination of  farmers in dairy farming.International 
Journal of  Agriculture, Environment and Food Sciences, 4(3), 
340–347. https://doi.org/10.31015/jaefs.2020.3.13 

Pope, M., & Sonka, S. (2020). Quantifying the economic 
benefits of  on-farm digital technologies. Farmdoc 
daily 10, 40. Department of  Agricultural and 
Consumer Economics, University of  Illinois at 
Urbana-Champaign. https://farmdocdaily.illinois.
edu/2020/03/quantifying-the-economic-benefits-of-
on-farm-digital-technologies.html 

Sharma, P. R., Singh, M., Sinha, P. K., Mollier, R. T., 
& Rajkhowa, D. J. (2020). Factors for adoption of  
artificial insemination technology in pig: Evidence 

from small-scale pig production system. Tropical 
Animal Health and Production, 52, 2755–2765. https://
doi.org/10.1007/s11250-020-02391-7 

Shehu, M. B., Kezi, M. D., & Bidoli, T. D. (2010). 
Challenges to farmers’ participation in artificial 
insemination (AI) biotechnology in Nigeria: An 
overview. Journal of  Agricultural Extension, 14(2). 
https://doi.org/10.4314/jae.v14i2.64128 

Tefera, S. S. (2013). Determinants of  artificial insemination use by 
smallholder dairy farmers in Lemu-bilbilo District, Ethiopia. 
Unpublished MSc Thesis. Egerton University.

Temba, A. E. M. (2011). Factors Affecting Adoption of  
Artificial Insemination Technology by Dairy Farmers in 
Kinondoni District. MA Thesis; Sokoine University of  
Agriculture, Morogoro, Tanzania. 

Wafula, B. N., Ngeno, V., Serem, A., & Kipkorir, P. (2021). 
Adoption and intensity of  improved  fish feeds 
use in Western Kenya. East African Agricultural and 
Forestry Journal, 85(1–4), 187–198. https://doi.org/1
0.1080/00128325.2021.1894567 

Wanjala, S. P. O., & Njehia, K. B. (2014). Herd 
characteristics on smallholder dairy farms in Western 
Kenya. Journal of  Animal Science Advances, 4(8), 996–
1003. https://doi.org/10.5455/jasa.20140827111904


