Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
British Journal of Education, Learning and Development Psychology
ISSN: 2682-6704
Volume 7, Issue 3, 2024 (pp. 40-57)
40 Article DOI: 10.52589/BJELDP-CXGDO8CI
DOI URL: https://doi.org/10.52589/BJELDP-CXGDO8CI
www.abjournals.org
ABSTRACT: Blended Learning is an undisputedly useful and
effective pedagogical approach for the 21st-century classroom.
However, its adoption in many state universities in Kenya is
worryingly low. This study aimed to develop a pedagogical model
that would accelerate the adoption of blended learning in public
universities in Kenya. Bandura’s Social Learning Theory was
used to understand students’ perception, self-efficacy, and
previous experience variables in a blended learning environment.
The methodology used was exploratory sequential mixed research
design. Third-year bachelor of education students (N=7385) in
public universities in Kenya formed the population for the study.
The researcher used multiple-stage sampling and the Nassiuma
formula was used to select 3rd-year education students (n=218).
Data was collected using semi-structured questionnaires. Data
was analyzed by Structural Equation Modelling (SEM) to design
an appropriate pedagogical model out on institutional
characteristics. The study revealed three significant paths: 1)
University preparedness and students’ perception (regression
estimate = .399; P<.05; 2) university preparedness and students’
self-efficacy (regression estimates = .389; P<.05); and 3) blended
learning adoption and students’ perception (regression estimates
= .55; P<.05). Students and lecturers responded that “Poor
internet connection,” and “Lack appropriate infrastructure and
equipment,” as the main barriers. In conclusion, the
implementation of blended learning highly depends on the
interaction of students’ perceptions and universities’
preparedness. The study suggested that universities should focus
on promoting an environment that focuses on university
preparedness and perception/attitudes. Further studies should be
done on appropriate BL models for TVET and secondary schools
in Kenya.
KEYWORDS:
Modelling, blended learning, characteristics,
implementation, public universities.
MODELLING INSTITUTIONAL CHARACTERISTICS THAT ACCELERATES
IMPLEMENTATION OF BLENDED LEARNING IN PUBLIC UNIVERSITIES IN
KENYA
Ndwiga Murithi Moses*, Ogeti Khaemba, and Syomwene Anne
*Department of Curriculum, Instruction and Educational Media, Moi University, P.O. Box
3900 - 30100, Eldoret, Kenya.
Cite this article:
Ndwiga M. M., Ogeti K.,
Syomwene A. (2024),
Modelling Institutional
Characteristics that Influenced
Implementation of Blended
Learning in Public
Universities in Kenya. British
Journal of Education,
Learning and Development
Psychology 7(3), 40-57. DOI:
10.52589/BJELDP-
CXGDO8CI
Manuscript History
Received: 16 May 2024
Accepted: 14 Jul 2024
Published: 26 Jul 2024
Copyright © 2024 The Author(s).
This is an Open Access article
distributed under the terms of
Creative Commons Attribution-
NonCommercial-NoDerivatives
4.0 International (CC BY-NC-ND
4.0), which permits anyone to
share, use, reproduce and
redistribute in any medium,
provided the original author and
source are credited.
British Journal of Education, Learning and Development Psychology
ISSN: 2682-6704
Volume 7, Issue 3, 2024 (pp. 40-57)
41 Article DOI: 10.52589/BJELDP-CXGDO8CI
DOI URL: https://doi.org/10.52589/BJELDP-CXGDO8CI
www.abjournals.org
INTRODUCTION
Background information
A model is a desired working system under specified conditions. Structural Equation Modelling
(SEM) is the testing of a multivariate structure with causal connections between variables that
are both latent and observed. In this context, the researcher sought to come up with the best
student characteristics variables as contemplated in Bandura’s Social Learning theory and their
causal connection to the blended learning approach for curriculum delivery under conditions
of public universities in Kenya. Structural equation modelling entailed factorial analysis
(covariance analysis) and causal relationship analysis using partial least square methods in a
structural mode (Hair et al., 2021).
Structural equation modelling is becoming a popular and plausible educational research tool in
building pragmatic multidimensional models (Panchenko, 2023). Once data is valid and
representative, the researcher was able to do factorial analysis, that is; correlation, variance,
covariance, and regression using SEM. According to Panchenko (2023), SEM helped
determine effective new methods in teaching and learning. A knowledge synthesis of 132
journal articles found SEM an appropriate methodology for determining effective innovative
factors in teaching and teachers’ education (Yin & Huang, 2021). SEM was also used to a fit
model of Turkish state university students enrolled for online learning in two faculties of
education (Yilmaz, 2021). In Vietnam, the SEM model yielded a first-order model of blended
learning for Hanoi University of Science and Technology (Long & Hanh, 2020).
Problem statement
Mainstreaming blended learning in low-income universities for effective learning and teaching
is still a challenge (Oduor, Ayiro & Boit, 2018). The lack of role models and models in the
blended learning approach have been reported as being responsible for the low uptake.
University with recorded exemplary implementation of blended learning for benchmarking in
sub-Saharan Africa in the technological landscape is yet to be identified (Ayere, 2020).
Knowledge was still scanty on the interaction of lecturers-related, students-related, and
institutions-related factors towards blended learning adoption (Oduor, Ayiro, & Boit, 2018).
Further, evidence emphasized that the lack of a context-based interactive approach toward
blended learning was a pedagogical gap (Namyssova et al., 2019). Successful implementation
of blended learning in a public university context requires the identification of strong predictors
of learners’ behaviours and the designing of an ecologic model for faculty to use for delivering
courses by both face-to-face and virtual methods (Saleem & Masadeh, 2021). This study,
therefore sought to model an appropriate mix of student and institution-related factors that
would bring the best out of using both face-to-face and online teaching methods in public
universities.
Main Objective
The purpose of this paper is to come up with a pedagogical model that explains institutional
characteristics that influenced the implementation of blended learning in public universities.
British Journal of Education, Learning and Development Psychology
ISSN: 2682-6704
Volume 7, Issue 3, 2024 (pp. 40-57)
42 Article DOI: 10.52589/BJELDP-CXGDO8CI
DOI URL: https://doi.org/10.52589/BJELDP-CXGDO8CI
www.abjournals.org
Specific objective
1. To determine the relationship between university preparedness and students’ perception
of blended learning in public universities in Kenya.
2. To determine the relationship between university preparedness and students’ self-
efficacy towards blended learning in public universities in Kenya.
3. To determine the relationship between students’ perception of blended learning and the
adoption of blended learning in public universities in Kenya.
Research Hypothesis
H01: There is no significant relationship between university preparedness and students’
perception of blended learning in public universities in Kenya.
H02: There is no significant relationship between university preparedness and students’ self-
efficacy towards blended learning in public universities in Kenya.
H03: There is no significant relationship between students’ perception of blended learning and
the adoption of blended learning in public universities in Kenya.
Proposed model
According to the conceptualization of the institutional framework, a basic model was created
within the framework of Bandura’s Social learning theory as shown in Figure 1.
Figure 1: Proposed model for institutional characteristics and adoption of blended
learning in public universities
The figure 1 demonstrates that the adoption of blended learning happens in public universities
when students have the right perception, self-efficacy and vicarious/previous experience.
Besides the three determinants, university preparedness was an important prerequisite to the
adoption of Blended learning. The first set of arrows showed that students’ perceptions
(attitudes) either directly or indirectly through university preparedness determined the adoption
of blended learning. second pathway is self-efficacy leading to the adoption of blended
learning, either directly or through university preparedness. Self-efficacy was used as students’
self-belief to organize and execute blended learning activities to produce desirable learning
British Journal of Education, Learning and Development Psychology
ISSN: 2682-6704
Volume 7, Issue 3, 2024 (pp. 40-57)
43 Article DOI: 10.52589/BJELDP-CXGDO8CI
DOI URL: https://doi.org/10.52589/BJELDP-CXGDO8CI
www.abjournals.org
outcomes. The third pathway was students’ previous experience influencing either directly or
indirectly the uptake of blended learning in public universities in Kenya. This proposed model
will be tested through structural equation modelling using sampled data collected from BED
students in Kenyan public universities.
LITERATURE REVIEW
Empirical Review
Students’ Perception and Blended Learning
A critical analysis explored UK university students’ perceptions of blended learning. A
pragmatic worldview and mixed methods were used to carry out the study. Convenient
sampling helped to identify 1917 respondents to the study. Questionnaires and FGDs were used
to collect data. The findings revealed that the students were positive about blended learning
because they did not see it to be intrinsically detrimental. Again they approved BL because it
was flexible and inclusive (Syska & Pritchard, 2023). The European study setting may not
apply to Africa, despite insightful findings. Therefore, there is a need for another study focusing
on blended learning for curriculum delivery in public universities in Africa.
From the lenses of students, Lu (2021) sought to establish students’ perceptions of the social,
pedagogical, and technical design of blended learning and its impact on critical thinking. Using
a mixed method design the study collected data via a Web-Based Learning Environment
Instrument from 90 first-year non-English major students at Normal University in China. The
findings showed students’ positive impressions of the designs and expressed that the BL
environment fostered critical thinking (Lu, 2021). According to Lu (2021), students received
and perceived technical support to be satisfactory, online material available, and conveniently
enjoyed learning ‘anywhere’ and ‘anytime.’
Bhagat (2020) surveyed 7 faculty members and 31 MBA students enrolled in BL courses in
2019 at Uganda Management Institute on the learners’ attitude towards blended learning
courses. The results showed that students’ general experience was positive; the reason being
the flexibility to learn anywhere and anytime. In addition, most students found courses
delivered via BL to be relevant (71.7%). BL made the students attentive (54.8%), confident
(58.06%), and connected with others (87.09%). Generally, the learners were satisfied (Bhagat,
2020). Like the previous studies, the study also suffered from self-grading and inadequate
sampling which limits its results to be generalized on the population of lectures and students.
To address the deficits a broader probability sampled mixed study needed to be carried out.
Among 19 universities that offered bachelor of nursing in Kenya, experimental research was
done in two public and two private universities on how they utilized blended learning on
undergraduate nurses for post-intervention outcomes. The respondents were 486 nursing
students in their fourth year and enrolled in the NRSG 400 course that was concerned with
education concepts and instruction styles. The study revealed that most nurse students
n=302(62.1%) were motivated to embrace blended learning. However, 75.1% of them
experienced challenges while using the blended learning mode of delivery (Kaniaru, Karani,
Mirie, & Nyangina, 2019).
British Journal of Education, Learning and Development Psychology
ISSN: 2682-6704
Volume 7, Issue 3, 2024 (pp. 40-57)
44 Article DOI: 10.52589/BJELDP-CXGDO8CI
DOI URL: https://doi.org/10.52589/BJELDP-CXGDO8CI
www.abjournals.org
Students’ Self-Efficacy and Blended Learning
Self-efficacy is a key construct in Bandura’s social cognitive theory. Self-efficacy is a self-
belief to organize and execute the ‘courses of action needed to produce given accomplishments
and having exclusive power to predict one’s behavior’ (Bandura, 1977, P3). The belief is made
up of four constructs: enactive mastery experience (performance accomplishments, vicarious
experience, verbal persuasions, and physiological and affective state (Bandura, 1977).
A review of antecedent literature on self-efficacy has different findings. For example, Katsarou
(2021) sought to establish the influence of self-efficacy and computer anxiety on Greek L2
students’ self-perceived satisfaction and digital competence in higher education through a
cross-sectional study. The survey involved 331 undergraduates from the faculty of agricultural
and forestry sciences at Democritus University of Thrace. The findings revealed that self-
efficacy positively influenced IT attitude and usage (Katsarou, 2021). This study has good
insights into Bandura’s social cognitive theory and self-efficacy among public university
undergraduates. However, the study assumes that attitude and use of IT are equivalent to
attitude and use of blended learning.
A correlational study was done in Turkey. The aim was to assess the influence of reflective
thinking, problem-solving, metacognitive awareness, and community of inquiry on learners’
academic self-efficacy in blended learning. The study involved 217 undergraduates in the
faculty of education enrolled in Turkish language and math for primary schools and were doing
introductory computer courses. The sampling was purposive. According to the study findings,
a community of inquiry, metacognitive awareness, problem-solving skills, and reflective
thinking strongly and positively correlated with self-efficacy among undergrads (Gizem,
Yilmaz, Ustun, & Yimaz, 2023). In this study, self-efficacy is a dependent variable instead of
an independent variable. Secondly, it used purposive sampling subjecting it to serious bias.
Thirdly it used correlational design which only establishes relationships and not cause effect.
These weaknesses point to the need for another study that is robust and makes self-efficacy the
subject and independent variable.
A similar study using pretest and posttest design was done in Boston, USA. The study aimed
to examine changes in self-efficacy for service learners involved in various community
services. The researchers interviewed 228 students from one state university and 4 community
colleges across 19 courses. The study revealed that the motivating potential of courses
moderated self-efficacy (Cronstaves, Metchik, Lynch, Bedezos, & Richards, 2023). The study
focuses on the role of motivation potential of courses on self-efficacy and service learning.
Again, the study was self-reporting research in the northeastern United States whose results are
susceptible to self-bias and may not credibly apply in Africa.
Phan (2023) did a comparative study on self-efficacy among Taiwan and Vietnam engineering
students. The study used mixed methods and an 11-point Likert scale questionnaire to collect
information from 222 engineering students. T-test and regression analysis was used and
demonstrated that the number of prior MOOCs, English proficiency levels, self-regulation, and
age predicted self-efficacy (Phan, 2023). Like precedent studies, self-efficacy is a dependent
variable. Therefore, it does not tell how it influences the use of blended learning among public
universities.
British Journal of Education, Learning and Development Psychology
ISSN: 2682-6704
Volume 7, Issue 3, 2024 (pp. 40-57)
45 Article DOI: 10.52589/BJELDP-CXGDO8CI
DOI URL: https://doi.org/10.52589/BJELDP-CXGDO8CI
www.abjournals.org
Previous Experience of Students and Blended Learning
In Social Learning Theory (SLT), previous experience is the vicarious experience; the
influence of students towards hating and liking blended learning depended on other previously
completed tasks. Past experiences included their successful encounters with digital devices to
interconnect with comrades and lecturers on a social platform. The previous experiences and
performances with technical device tools not only give the students requisite skills for blended
learning but also cause social persuasion or power of others (peers and mentors) on students’
ultimate behaviour (Koutroubas & Galanakis, 2022).
How did learners’ previous experience with BL influence their use of blended learning? This
question was answered by an exploratory case study in Australia. The study involved 20
students enrolled in the Bachelor of Law program’s introductory unit. The case study used
focused group discussions and questionnaires and found that most of the students were direct
high school leavers who had not had prior BL encounters. However, their previous experience
did not influence their use of BL. Instead, students were quick to learn BL’s benefits and used
BL tools such as videos and quizzes to catch up (Pechenkina, Scardamaglia & Gregory, 2018).
This study was done in an Australian setting which was different from Africa. Secondly, a
sample of 20 students is too small to infer for all public university students in Kenya.
Shedrout (2021) also used an exploratory case study, to examine experiences of elementary
teacher candidates on technology tools. Twenty-seven teacher candidates enrolled in a teacher
education program at the Catholic Liberal Arts College in the Midwest participated in the study.
Previous experience of the teacher candidates influenced their use of blended learning. The
previous experience made them familiar with digital tools and usage (Shedrout, 2021). The
case study was largely qualitative, excluding the strengths of quantitative methods. It also used
a very small sample(n=27) during COVID-19. The results may not be valid in normal post-
COVID-19 times and a large population of public universities in Kenya.
In Jordan, a descriptive survey study was done to investigate the online component challenges.
The study had 263 participants who were students enrolled in sports science BL classes at the
University of Jordan. Information was gathered with the help of questionnaires and analyzed
with the aid of SPSS and AMOS software. Students who had no previous experience in BL
encountered significant challenges in the use of BL for learning in sport science studies at the
University of Jordan (Bayyat, Muaili & Aldabbas, 2021); meaning that previous experience
significantly and positively influenced students’ use of blended learning in Jordan. The
limitation of this study is that it is exclusively qualitative and applicable to Jordan settings.
There is a need for a mixed-method study that applies to the implementation of a blended
learning approach in public universities in Kenya.
Among students of Sultan Qaboos University - Oman, a study was done to discover variables
that affected the adoption of BL in higher education institutions. The research was animated
by the Theory of Planned Behaviour (TPB). Data was collected on demographics, attitudes,
subjective norms, beliefs, perceived behavioural control, behavioural intention, self-efficacy,
and actual usage from 362 social science students. The data was analyzed by Pearson
correlation and multiple regression. The analysis revealed that previous experience positively
influenced social science students at Sultan Qaboos University to use blended learning (Hamad,
Shehata, & Hosni, 2024). The exclusive quantitative approach and Oman contextualization
make the results of the study not applicable to Kenyan public universities with utmost validity.
British Journal of Education, Learning and Development Psychology
ISSN: 2682-6704
Volume 7, Issue 3, 2024 (pp. 40-57)
46 Article DOI: 10.52589/BJELDP-CXGDO8CI
DOI URL: https://doi.org/10.52589/BJELDP-CXGDO8CI
www.abjournals.org
While investigating the reasons for liking or disliking a learning environment in a local
university, Chaw and Tang (2023) found out that previous experience and particularly prior
use of web applications influenced students’ digital readiness. The study used an exploratory
sequential mixed methods research design where data was collected from 117 diploma,
bachelor's, and master's students using focus group discussions and online questionnaires
(Chaw & Tang, 2023). Likes and previous experiences of students in Singapore may not apply
in Kenya due to geographical and developmental pedestal differences. The study also assumes
that previous experience in web applications is the same as previous experience in blended
learning. Therefore, there is a need for another mixed-method study focussing on blended
learning in public universities in Kenya.
University Preparedness and Blended Learning
University preparedness meant institutional readiness which entailed vision, policies,
structures, infrastructure, partnerships, and technical support systems that favour or frustrate
the acceptance and implementation of blended learning at public universities. Perris and Mohee
(2021) guide that quality in higher education embracing blended learning can only be assured
when BL is anchored on university vision; policies and structures; infrastructures, partnership;
research and innovation; program relevance and curriculum; learning support; and professional
development (Mohee & Perris, 2021).
Across the European Higher Education Area (EHEA), institutions heavily invested in
equipment, infrastructure, and professional development. However, the approach suffered from
strained funding, inability to design a concerted institutional approach, and inadequate staff
(Gaebel, Zhang, Stoeber, & Morrisroe, 2021). The level of blended learning in Europe is at an
advanced level; to the extent of developing a customized model. A 3-round Delphi study
carried out between December 2018 and July 2019 on 28 European experts revealed that
Europe had developed a European Maturity Model (EMM). EMM defined how blended
education was designed and implemented in institutions of higher learning. The model
systematically mapped blended learning activities, conditions, strategies, and policies.
Maturity was the degree of formality and optimization of evidence-based decision-making
design, recording, and CQI. The model helped to guide instructors to align course objectives,
learning activities, and assessments with target student groups. The model had 21 subdivisions
that were grouped under course, program, and institutional levels (Dijkstra & Goeman, 2020).
Lecturers were actors at the course level. Coordinators, deans, and departmental heads were
actors at the program level.
In a cross-institutional study among engineering students at Purdue, Trine, and McGill
universities preparedness dimensions were observed as critical success factors in BL
implementation in higher education institutions. Blended learning was positively approved as
a “freeform environment” for teaching and learning. However, university preparedness as
extracurricular pressures and responsibilities, time constraints, and technical support affected
the application of blended learning. The investigator also discovered a lack of structures to
realign online and face-to-face teaching affecting acceptance of blended learning. The study
adopted Actor-Network Theory (ANT) which took students as active actors and implementors
of blended learning. A semi-structured interview was done with 271 engineering students from
the universities and a step-by-step thematic analysis of collected data (Evenhouse, Lee, Berger,
Rhoads, & DeBoer, 2023). The sample was good enough. The fact that the study used a self-
British Journal of Education, Learning and Development Psychology
ISSN: 2682-6704
Volume 7, Issue 3, 2024 (pp. 40-57)
47 Article DOI: 10.52589/BJELDP-CXGDO8CI
DOI URL: https://doi.org/10.52589/BJELDP-CXGDO8CI
www.abjournals.org
reporting method, weakened the study with subjectivity and bias. Secondly, thematic analysis
weakened the study with limitations of new insights at saturation.
To find key conceptual and theoretical features that facilitated success in implementing blended
learning in higher education, a desk review approach was used to systematically analyze 11
studies using Google Scholar and Scopus as search strategies. University policy was identified
as a core feature affecting blended learning. Other policy-related features identified by the
studies were vision, goal, infrastructure, faculty, strategy, professional development, and
support systems (Bekele, Karkouti, & Amponsah, 2022). Even though the findings are
evidence-based, they are neither public university-specific nor Subsaharan Africa-specific.
In Pakistan, an exploratory qualitative study, involving 30 faculty members and 60
undergraduates enrolled in social sciences, arts, and humanities, was done. The research aimed
to identify the practices and issues affecting blended learning in Islamia University of
Bahawalpur. Lack of policy guidelines was a key finding (Hussain, Shahzad, & Ali, 2019). In
addition, the research found that the university did not support the adoption of online and
blended learning, and lacked sophisticated technology, time management, authentic learning
resources, and information. Weaknesses found in this study are methodic; which is skewed
towards the qualitative strand alone. Secondly, the study setup is in Pakistan which is different
from Africa.
Infrastructure was identified as barrier number one in an exploratory qualitative study
investigating the inhibitors of faculty blended learning in Ghana. The study purposively
sampled 22 teaching staff from four faculties of a university in Ghana. Data collected was
subjected to coding, comparative, and thematic analysis. Other barriers identified were
institutional issues, faculty concerns, and technical support (Anturi-Boampong, 2021). The
findings of the research showed a picture of challenges an African university in matters of
implementing blended learning. However, the methodology is only qualitative with a very
small sample that was purposively sampled. These make the findings weak and biased, hence
the need for a study grounded on mixed methods and a bigger sample.
Across Africa, the adoption of BL was still at an embryonic stage. Kizito (2016) found that
institutional factors such as organizational culture, paucity of trained and motivated staff,
limited technological support, and absence of records of success to build on hampered the
application of BL for teaching and learning by universities in Africa. A summative evaluation
of blended learning in universities in East Africa revealed that blended learning was highly
relevant. Most universities (80%) used blended learning. However, the students and lecturers
experienced inadequacy in ICT infrastructure, a lack of supportive policies, overloaded
teaching staff, unmotivated staff, and inconsistency in the application of blended learning for
teaching and learning (Young et al., 2021).
At Kenyatta University, there were inconsistent efforts to build supervisors' capacity and the
university lacked resources to effectively implement blended learning which affected the
completion rate (Miheso-O'Connor, Bwire & Mwangisi, 2020). Specifically, training,
planning, and legislation were found to be critical in the effective application of a blended
learning model and in creating a favorable educational environment (Masadeh, 2021).
British Journal of Education, Learning and Development Psychology
ISSN: 2682-6704
Volume 7, Issue 3, 2024 (pp. 40-57)
48 Article DOI: 10.52589/BJELDP-CXGDO8CI
DOI URL: https://doi.org/10.52589/BJELDP-CXGDO8CI
www.abjournals.org
Among public universities in Kenya, a mixed method study of one hundred and forty-eight
faculty members in 3 Kenyan public universities using blended learning revealed that the GoK
had an elaborate institutional and policy framework to increase broadband internet and
interconnectivity through Kenya Education Network Trust (KENET) for teaching, learning,
and research in universities (Tarus, Gichoya, & Muumbo, 2015). Although the GoK had
successfully interconnected the universities, only 11% of the students in public universities in
Kenya used the blended learning approach. The barriers to the use of the blended learning
approach were inadequate ICT infrastructure, finance, policies technical skills, assurance
among faculty members, and enough time to create E-learning content.
Among 114 students at Tom Mboya University College (TMUC), an exploratory study was
done on taking advantage of informal education for the expansion of participation in Kenyan
university education. The study used survey methods to collect data. The findings showed BL
in Kenyan public universities was not at the desired level because of infrastructure. specifically,
there was a lack of computing resources that facilitated BL for teaching and learning (Hawi,
Heinrich, & Lai, 2021). Because of the self-reporting method's weaknesses, the findings needed
to be confirmed by a mixed-method study.
In a scoping review of challenges that faced e-learning in universities in Kenya, deficiency of
Information, Communication, and Technology (ICT) infrastructure was cited as a major
barrier. Other challenges were inadequate e-learning policies, fast change in technologies,
technical and pedagogical incompetencies among e-tutors and e-learners, and the absence of e-
learning theory to support the e-learning exercise (Kibuku, Ochieng & Wausi, 2020). In
addition, Kibuku, Ochieng and Wausi (2020) discovered that universities faced budgetary and
sustainability challenges. The investigators also observed undesirable attitudes about e-
learning, quality challenges, the dominance of technology and market forces in e-learning, and
inadequate partnership among the e-learning participants. In as much as the study gives
insightful knowledge on the barriers to the application of blended learning on campuses in
Kenya, it is purely based on literature. No current feelings and views of actual participants are
captured to validate the findings. The study also assumes that e-learning is equivalent to
blended learning. Therefore, there is a need for a mixed-method study with a focus on how
infrastructure hinders the use of blended learning in Kenyan public universities.
Infrastructure and unreliable technology were also found as a barrier to the sustainable
upscaling of ABRACADABRA; an online platform for teaching and learning English and
French in Kenya. These findings were a product of an exploratory qualitative study that
involved 40 respondents whose findings were descriptively analyzed. Other hindrances to the
widespread use of ABRACADABRA were a lack of technical support at school, inadequate
policies, negative student attitudes, and lack of professional development (Lysenko, Abrami,
& Wade, 2022). The weaknesses of this study rest in the small sample and exclusive use of
qualitative methods. Another research that includes quantitative and robust inferential analysis
of data is needed.
Theoretical Underpinnings
Social Learning Theory (SLT) gave the theoretical framework that underpinned this research.
The framework was constructed from SLT as a tested and validated theory to guide the
researcher in thinking and planning as well as giving the foundation on which all knowledge
was constructed (Grant & Osanloo, 2014). The Social Learning Theory (SLT) was propounded
British Journal of Education, Learning and Development Psychology
ISSN: 2682-6704
Volume 7, Issue 3, 2024 (pp. 40-57)
49 Article DOI: 10.52589/BJELDP-CXGDO8CI
DOI URL: https://doi.org/10.52589/BJELDP-CXGDO8CI
www.abjournals.org
by Albert Bandura in 1971. Albert Bandura was a Canadian psychologist (Bandura, 1971).
According to Bandura, observation or perception, self-efficacy, vicarious experiences or past
experiences, motivation or modeling, reinforcement, and reward predicted learning
(Koutroubas & Galanakis, 2022). In addition, the environment plays a critical role in learning,
too (Nabavi, 2014). The theory emphasizes the role of personal factors such as beliefs,
attitudes, and knowledge acquired out of previous experiences that influenced one's
expectations and goals (Koutroubas & Galanakis, 2022).
Of interest to this study are perception, self-efficacy, and vicarious/previous experience as
predictors of blended learning in a public university environment. The researcher tested these
theoretical concepts to establish their applicability and relevance in designing a working model
for students to embrace blended learning sustainably. University preparedness is an addition to
the theoretical underpinnings because of the high dependency of BL on ICT infrastructure,
policy, and competencies.
MATERIALS AND METHODS
The study adopted the pragmatism philosophy because it was after practical solutions that
would make blended learning work for public universities in Kenya. The design was
exploratory and the data collection and analysis methods were both qualitative and quantitative.
The study was done in Kenya.
Identification of the study unit and respondents was done using multistage and proportionate
sampling methods. The research selected the 8 universities using purposive sampling based on
the criteria of availability of education programs, willingness to participate in the study, and
regional balance. Based on the criteria, the eight (8) universities were: Pwani University,
University of Embu, University of Nairobi, Kibabii University, Kirinyaga University, Maseno
University, Laikipia University, and Garissa University. After getting 8 universities, the
researcher used to go for the 3rd-year students enrolled in B. ED programs. The third-year
students enrolled in B. ED were selected because of their long experience and knowledge of
educational concepts. According to Table 1, there were 7385 third-year B. ED students in the
eight selected universities. To get the representative populations of participants from the
universities per each stratum, the researcher used the Nassiuma formula on the students’
population.
Where;
n= the desired sample size;
N = the proportion of the target population
C = Covariance = 0.3; and e = standard error ± 0.02.
e = the margin of error
British Journal of Education, Learning and Development Psychology
ISSN: 2682-6704
Volume 7, Issue 3, 2024 (pp. 40-57)
50 Article DOI: 10.52589/BJELDP-CXGDO8CI
DOI URL: https://doi.org/10.52589/BJELDP-CXGDO8CI
www.abjournals.org
After the establishment of the sample size of the eight universities as n=218, the researcher
determined the sample size for each university. Each university’s sample size was determined
as a fraction of the 218 proportionate to its 3rd-year B. ED enrolment population. For example,
Maseno University and the University of Nairobi had the highest sample sizes as compared to
Garissa University and Kirinyaga University.
Data was collected using Open (qualitative) and closed-ended (quantitative) questionnaires. On
a seven-point Likert Scale, students and lecturers were asked for information to address
research questions. The response range was from 1 -Strongly disagree; 2 – disagree; 3 –
Slightly disagree; 4 - Neither agree nor disagree; 5 - Slightly agree; 6 – Agree; 7 – Strongly
agree. Reliability coefficient alpha was applied to determine if Likert-scaled instruments were
reliable (Huang, 2016). In a similar study to assess student satisfaction with BL, the reliability
coefficient of Cronbach's alpha was found to fit in determining the internal consistency of
research instruments (Naaj, Nachouki, & Ankit, 2012). According to the test, an instrument is
reliable when the alpha is close to 1; meaning that the items in the instrument had high
internally consistent and covariance (Hajjar, 2018). After testing all scale items, the tool passed
the test. On average the reliability was 0.8; above the Cronbach alpha threshold of 0.7 and
closer to 1; meaning that reliability was good (Oluwatayo, 2012 & Balan,2013).
Further, the validity of the research instruments was enhanced by the lecturers’ opinions.
Content validity was observed by the researcher identifying and outlining the domain of interest
in the adoption of blended learning in institutions of higher learning. The construct validity of
research instruments was checked by the use of correlation analysis. Nevertheless, the
researcher borrowed and modified the instrument developed by the University of Trinidad &
Tobago in a mixed method research; Student Blended Learning Experience Questionnaire
(SBLEQ) for students. The instruments were effectively used to gather evidence on students’
experiences of switching to blended learning from the traditional learning method (Jackman,
2018).
Ethical Issues Considered
‘Do no harm,’ seeking informed consent, anonymity and confidentiality, plagiarism check, data
integrity, and approvals were the principles that the researcher observed during the study. For
example, the researcher embraced the ‘do no harm principle.’ Risks entailed possible harm that
may arise from the research. Such harm would be loss of resources such as time, reputation,
physical and emotional (Fleming & Zegwaard, 2018). Second, the respondents were made to
fill and sign the informed consent form before engaging as proof that they were sufficiently
informed, gave voluntarily the information without compulsion, and were free to withdraw at
any point of the research process (Abed, 2015). Third, the researcher upheld the anonymity and
confidentiality of the respondents by not sharing with other participants private information
and concealing the source of information (Bos, 2020). Moi University librarian checked and
issued with non-academic plagiarism certificate to prove the authenticity of the study. The
investigator ensured data integrity by not manipulating respondents’ answers (Bassey, 2019).
Approval from relevant institutions was another ethical issue that was considered important,
especially when human participants were involved (Fleming & Zegwaard, 2018). Keeping in
line with GoK laws, the researcher obtained letters of introduction from Moi University and
approval from NACOSTI before going to the field to collect data.
British Journal of Education, Learning and Development Psychology
ISSN: 2682-6704
Volume 7, Issue 3, 2024 (pp. 40-57)
51 Article DOI: 10.52589/BJELDP-CXGDO8CI
DOI URL: https://doi.org/10.52589/BJELDP-CXGDO8CI
www.abjournals.org
Ethical considerations were upheld by the researcher because they promoted collective work
standards such as fairness, mutual respect, compliance with the legal framework, social
responsibility, and human rights (Natade, Murunga & Kabesa, 2023).
DATA ANALYSIS
After data collection, data analysis was done by descriptive and inferential statistics.
Descriptive statistics involved percentages, mean and standard deviation, and reliability using
SPSS. Inferential statistics used Structural Equation Modelling (SEM) that entailed factor
loadings analysis, and confirmatory fit indices to determine a good fit model for adopting
blended learning in public universities in Kenya. The development of a pedagogical model
explaining institutional influence on the use of blended learning was done using structural
equation modeling with the aid of AMOS version 21 software. Results of items relating to
learners’/students’ perception, self-efficacy and previous experience were grouped and
transformed into indices under each variable as exogenous variables. The Same was done to
the outcome variable; that is institutional factors and effective teaching and learning.
RESULTS AND DISCUSSIONS
The study developed a pedagogical model that explained institutional characteristics that
influenced the usage of blended learning for teaching and learning among B. ED students in
public universities in Kenya. The development of a pedagogical model explaining institutional
influence on the use of blended learning was done using structural equation modeling with the
aid of AMOS version 24 software loaded on SPSS. The results are as per the unstandardized
and standardized models below. The unstandardized model gave the covariates of exogenous
variables and factorial loading of all the variables in the model depicted in figure 2 below.
Figure 2: Model 1: Unstandardized
According to Model 1 in figure 2 and table 1, there was a significantly strong correlation
between student perception and previous ICT experience, student perception and their self-
British Journal of Education, Learning and Development Psychology
ISSN: 2682-6704
Volume 7, Issue 3, 2024 (pp. 40-57)
52 Article DOI: 10.52589/BJELDP-CXGDO8CI
DOI URL: https://doi.org/10.52589/BJELDP-CXGDO8CI
www.abjournals.org
efficacy, and previous ICT experience and Self-efficacy. The correlation between students'
perception and self-efficacy was the strongest 0.036. The correlation between self-efficacy and
previous experience, students' perception, and previous perception were the same 0.034.
Table 1: Covariance of exogenous variables
Covariances: (Group number 1 - Default model)
Estimate
S.E.
C.R.
P
Label
SELF_EFFICACY_X2
<-->
PREVIOUS_EXPERIENCE_X3
.034
.005
7.121
***
PERCEPTIONS_X1
<-->
SELF_EFFICACY_X2
.036
.005
7.341
***
PERCEPTIONS_X1
<-->
PREVIOUS_EXPERIENCE_X3
.034
.005
7.082
***
According to the findings in Table 1, one unit change in previous experience resulted in 0.034
change in the student’s self-efficacy; one unit variance in self-efficacy resulted in 0.036
positive change in student perception; and one unit change in previous ICT experience affected
change in 0.34 in students’ perception. Implied by the result was that self-efficacy components
influenced BED students’ positive perception towards BL than previous experience Therefore
public universities needed to invest more in training students on setting up LMS, downloading
and organizing learning materials, using LMS for group work, doing and uploading
assignments. In addition, the university management needed to train the students on using
digital devices to access and use LMS.
A standardized model was used to establish the regression estimates or estimate predictor
relations. The model established six (6) predictor pathways. They include self-efficacy
predicting university preparedness; students’ perception influencing university preparedness;
previous ICT experience influencing university preparedness; university preparedness
influencing predicting BL adoption; student perception influencing BL adoption and previous
ICT experience influencing BL adoption.
Figure 3: Model 2: standardized model
British Journal of Education, Learning and Development Psychology
ISSN: 2682-6704
Volume 7, Issue 3, 2024 (pp. 40-57)
53 Article DOI: 10.52589/BJELDP-CXGDO8CI
DOI URL: https://doi.org/10.52589/BJELDP-CXGDO8CI
www.abjournals.org
Apart from university preparedness having an inverse relationship with previous experience,
the rest of the pathways had positive variances; meaning that one unit change in exogenous
variables caused a positive change in Blended learning adoption as shown in Table 2.
Table 2: Regression weights
Regression Weights: (Group number 1 - Default model)
Estimate
S.E.
C.R.
P
Label
University
preparedness
<--
-
Students’ perception
.476
.097
4.912
***
University
preparedness
<--
-
Students’ self-
efficacy
.389
.098
3.951
***
University
preparedness
<--
-
Previous experience
-.060
.094
-.641
.521
BL Adoption
<--
-
University
preparedness
.182
.058
3.153
.002
BL Adoption
<--
-
Previous experience
.077
.069
1.126
.260
BL Adoption
<--
-
Students’ perception
.482
.077
6.218
***
This implied that, to maximize learning outcomes among B. ED students using Blended
learning, public universities needed to focus more on students’ perception, self-efficacy, and
preparedness. Figure 4 shows the best model with the three critical paths identified in Table 2.
Figure 4: Best students’ pedagogical model (Standardized model)
According to the model in Figure 4, there was a significantly strong correlation between student
perception and Self-efficacy. The correlation between students' perception and self-efficacy
improved from 0.036 to a coefficient of 0.68. Based on the standardized regression weights,
British Journal of Education, Learning and Development Psychology
ISSN: 2682-6704
Volume 7, Issue 3, 2024 (pp. 40-57)
54 Article DOI: 10.52589/BJELDP-CXGDO8CI
DOI URL: https://doi.org/10.52589/BJELDP-CXGDO8CI
www.abjournals.org
students’ perception and adoption of blended learning (0.550) was the strongest path, followed
by university preparedness and BL adoption (0.399).
The researcher also determined the intercept value. As an intercept, university preparedness
significantly influenced BE Students’ adoption of blended learning as shown p= 0.04 in Table
3.
Table 3: Intercepts for predicting endogenous variables
Estimate
S.E.
C.R.
P
Label
University preparedness
.112
.055
2.053
.040
BL adoption
.241
.040
6.081
***
It meant that university preparedness significantly moderated students’ perception and self-
efficacy in adopting BL. The results of the study in Table 3 implied that public universities
needed to promote institutional measures to mediate the use of blended learning. Examples of
measures included anchoring BL on policies, structures, and good infrastructure. In addition,
the universities should establish good technical support systems, Q & A systems, strong
bandwidth internet and train students and lecturers thoroughly on BL.
Model tests of fit
The researcher used the Chi-square (CMIN test), Tucker-Lewis Index (TLI), Comparative Fit
Index (CFI), and Root Mean Square Error of Approximation (RMSEA) to determine if the data
fit well in the model. The chi-square results were less than 0.05, p=0.011 as shown in table 4.
Table 4: Chi-square test of model fit
Model
NPAR
CMIN
DF
P
CMIN/DF
Default model
13
6.447
1
.011
6.447
Saturated model
14
.000
0
Independence model
8
328.019
6
.000
54.670
The chi-square results indicated that the data did not fit well in the model because p < 0.05.
The second and third tests were TLI and CFI as in Table 5.
Table 5: Tucker-Lewis Index (TLI) and Comparative Fit Index (CFI)
Model
NFI
Delta1
RFI
rho1
IFI
Delta2
TLI
rho2
CFI
Default model
.980
.882
.983
.899
.983
Saturated model
1.000
1.000
1.000
Independence model
.000
.000
.000
.000
.000
British Journal of Education, Learning and Development Psychology
ISSN: 2682-6704
Volume 7, Issue 3, 2024 (pp. 40-57)
55 Article DOI: 10.52589/BJELDP-CXGDO8CI
DOI URL: https://doi.org/10.52589/BJELDP-CXGDO8CI
www.abjournals.org
Again, TLI was 0.899 which was less than the threshold of 0.95. However, CFI was 1.00,
greater than the 0.90 threshold; which meant that data fitted well in the model. Finally, the
researcher carried out the RMSEA test as shown in Table 6.
Table 6: Root Mean Square Error of Approximation (RMSEA)
Model
RMSEA
LO 90
HI 90
PCLOSE
Default model
.180
.069
.322
.030
Independence model
.564
.513
.616
.000
According to the rule of thumb, data fitted well in the model if the RMSEA was less or equal
to 0.8. In this case, the RMSEA readings are 0.18, far below the threshold; implying data fitted
well with the model. Conclusively, out of the four model fit tests, two proved that data fitted
well in the model as shown in Table 7.
Table 7: Summary of the model fit tests
Test
Chi-square
TLI
CFI
RMSEA
Threshold
P ≥0.05
TLI ≥0.95
CFI≥0.90
RMSEA ≤0.8
Actual
P =0.11
TLI =0.899
CF1=1.00
RMSEA=0.18
Conclusion
Not fit to the
data
Not fit to the data
Good fit to the
data
Good fit to the
data
Though Chi-square and TLI in Table 7 did not find a good fit, the rest of the tests found the
data fit well in the model. This is an indication that the exogenous variable predicted university
preparedness and adoption of blended learning among B.ED students in the public Universities.
CONCLUSION AND RECOMMENDATIONS
Conclusion
The last set of findings addressed, ‘What pedagogical model best explains institutional
characteristics that influence the use of blended learning for teaching and learning among B.
ED students in public universities in Kenya?’ The study found significant pathways in the
model; 1) University preparedness and students’ perception (regression estimate = .399; P<.05;
2) university preparedness and students’ self-efficacy (regression estimates = .389; P<.05); and
3) BL adoption and students’ perception (regression estimates = .55; P<.05). Finally, the
researcher observed that the implementation of blended learning suffered from the weak and
high cost of internet connectivity, poor scheduling of classes, breakdowns of ICT, difficulty in
lecturer-student interaction, lack of digital devices and unsupportive environment.
Recommendations
Recommendations for practice
a) Universities should focus on fostering an ecosystem that focuses on university
preparedness, student self-efficacy, and perception/attitudes.
British Journal of Education, Learning and Development Psychology
ISSN: 2682-6704
Volume 7, Issue 3, 2024 (pp. 40-57)
56 Article DOI: 10.52589/BJELDP-CXGDO8CI
DOI URL: https://doi.org/10.52589/BJELDP-CXGDO8CI
www.abjournals.org
Recommendations for policy
b) Universities should develop policies that focus on improving students’ proficiency,
efficacy, and attitudes towards blended learning.
c) The GoK, through MOEST, should develop policies and guidelines on BL use for
curriculum delivery in universities
Recommendations for further studies
a) Further studies should be done on appropriate BL models for TVET and secondary
schools in Kenya.
REFERENCES
Adamu, I., & Hawamdeh, M. (2020). The flex model of blended learning enabled digital
citizen. Online Pedagogy and Management for Smart Societies.
Adiwisastia, M. F., Mulyani, Y. S., Alawiyah, T., Wibisono, T., Iskandar, I. D., & Purnia, D.
S. (2020). Implementation of the lab rotation model in blended learning based on student
perspectives. Journal of Physics, 1-9.
Alshobramy, H. A. (2019). The effectiveness of Bandura's social learning theory in learning
English speaking skills among secondary school EFL students. International Journal of
Vocational and Technical Education Research, 5(5), 11-23.
Ayere, M. A. (2020). Mainstreaming blended learning in a low-income university. In M. M.
Shohel, E-learning and digital education in the twenty-first century.
Azukas, M. E. (2019). Cultivating a blended community of practices to promote personalized
learning. Journal of Online Learning Research, 5(3)., 275-310.
Bandura, A. (1971). Social learning theory. Open Library.
Bussetto, L., Wick, W., & Gumbinger, C. (2020). How to use and assess qualitative research
methods. Neurological Research and Practice, 2(14)., 1-10.
Chukwuemeka, J. C., Anekwe, J., & Ochuma, O. (2020). Investigation on the influence of A
La Carte model and face-to-face learning on students' performance in phonetics. IOSR
Journal of Research & Methods in Education, 10(2), 8-14.
Edinyang, D. S. (2016). The significance of social learning theories in the teaching of social
studies education. International Journal of Sociology and Anthropology Research, 2(1),
40-45.
Faustino, A., & Kaur, I. (2022). Blended learning models: Perspectives in higher education.
International Journal Conference and Engineering, 29, 185-191.
Grant, C., & Osanloo, A. (2014). Understanding, selecting and integrating a theoretical
framework in dissertation research: Creating the blueprint for your "House.".
Administrative Issues Journal: Connecting Education, Practice and Research, 4(2), 12-
27.
Hajjar, S. T. (2018). Statistical analysis: Internal consistency reliability and construct validity.
International Journal of Quantitative and Qualitative Research Methods, 6(1)., 27-38.
Harris, L. (2017). Blended learning benefits academic growth. Journal of First-Year Writing,
1(2), 1-8.
Hawi, F., & Sudira, P. (2019). The effect of the blended learning model is to improve the
conceptual understanding of computer and network engineering students. Journal of
Physics: Conference Series, 1413., 1-7.
British Journal of Education, Learning and Development Psychology
ISSN: 2682-6704
Volume 7, Issue 3, 2024 (pp. 40-57)
57 Article DOI: 10.52589/BJELDP-CXGDO8CI
DOI URL: https://doi.org/10.52589/BJELDP-CXGDO8CI
www.abjournals.org
Huang, G. (2016). Learners' perception of blended learning and the roles and interactions of
f2f and online learning. ORTESOL Journal, 33(1)., 14-34.
Hycner, R. H. (1985). Some guidelines for the phenomenological analysis of interview data.
Human Studies, 8., 279-303.
Jackman, W. M. (2018). Switching from traditional to blended learning at university level:
students' and lecturers' experiences. International Journal of Learning, Teaching and
Educational Research, 17(5), 1-14.
Jayanthi, R. (2019). A study about blended learning. Its importance and concept. International
Journal of Scientific Development and Research, 4(4)., 387-398.
Kathula, D. N. (2021). Technology: The classroom of the future for universities in Kenya.
Journal of Education, 4(1), 94-111.
Koutroubas, V., & Galanakis, M. (2022). Bandura's social learning and its importance in the
organizational psychology context. Psychology Research, 12(6)., 315-322.
Kundu, A. (2020). Toward a framework for strengthening participants' self-efficacy in online
education. Asian Association of Open Universities Journal, 15(3)., 351-370.
Martin-Martinez, L., Sainz, V., & Rodriguez-Legendre, F. (2020). Evaluation of a blended
learning model for pre-service teachers. Knowledge Management & e-Learning, 12(2).,
147-164.
Marunic, G., & Glazar, V. (2015). Challenges of blended learning. Scientific Proceedings XXIII
International Scientific-Technical Conference, 3, 61-64.
Naaj, M. A., Nachouki, M., & Ankit, A. (2012). Evaluating student satisfaction with blended
learning in a gender-segregated environment. Journal of Information Technology
Education: Research, 11(1)., 186-203.
Nabavi, R. T. (2014). Bandura's Social learning theory and social cognitive learning theory.
Researchgate, 1-24.
Oduor, H. N., Ayiro, L., & Boit, J. (2018). Teachers' gender influence on adoption and use of
information and communication technology in public secondary schools in Kenya.
African Journal of Education, Science and Technology, 4(3), 236-243.
Raspopovic, M., Cvetanovic, S., Medan, I., & Ljubojevic, D. (2017). The effects of integrating
social learning environment with online learning. International Review of Research in
Open and Distributed Learning, 1(1)., 141-161.
Rusijono, C., & Bachfair, B. (2019). The application of blended learning station rotation
method in elementary school and science education to improve higher order thinking
skills. DINA MIKA Jurnal Ilmiah Pendidika, 11(2)., 79-86.
Saboowala, R., & Mishra, P. M. (2019). Blended Learning: The new normal teaching_learning
pedagogy in post-COVID-19 pandemic. International Journal of Research, 7(8)., 208-
211.
Saragih, M. J., Cristanto, R. M., Effendi, Y., & Zamzami, E. M. (2019). Application of blended
learning supporting digital education 4.0. Journal of Physics, 1-7.
Shamad, A., & Wekke, I. S. (2019). Lecturers' participation in applying blended learning in
Islamic higher education in Indonesia. Universal Journal of Educational Research,
7(12)., 2604-2608.
Staker, H., & Horn, M. B. (2012). Classifying K-12 blended learning. Innosight Institute.
Subedi, D. (2016). Explanatory sequential mixed method design as the third research
community of knowledge chain. American Journal of Education Research, 4(7), 570-
577.
