Factors Affecting Students’ Satisfaction in Blended Learning Courses:

A Case Study in Thai Nguyen University

NGUYEN THI HONG MINH

, NGO THI BICH NGOC

, NGUYEN THI HONG CHUYEN

in teaching and studying methods amid the accelerating digitalisation and the fourth industrial revolution, especially

after three years of the COVID-19 pandemic. This study showed that the student’s awareness of the ease of use of

blended learning had a positive influence on their motivation and engagement, which affected their satisfaction with

blended learning courses. These results acted as useful references for universities in general, and Thai Nguyen

University in particular, to proactively adjust and implement more practical solutions to assist their students with better

study results, and to help their lecturers find specific measures to provoke the students’ proactiveness, enthusiasm and

creativity.

Key-Words:

-

satisfaction, student, motivation, blended learning,

perceived usefulness (PU), perceived ease of use

(PEU), perceived engagement (POE), learners’ satisfaction (SAT)

In education, technological advancements in teaching

and learning have been applied more widely, especially

during the COVID-19 pandemic. Technologies have

formed a new trend in education, which is “stop

schooling but not stop studying”. In addition, this is

also the main factor promoting changes in teaching and

learning methods in the fourth industrial revolution.

According to, [1], blended learning is a combination of

methods. A blended learning environment is the

However, in blended learning courses, it is vital to pay

attention to various key determinants of the students’

satisfaction as this is among the factors taken into

consideration in assessing the effectiveness of the

courses. Moreover, clearly understanding these

determinants also has a direct impact on blended

learning quality improvement and the relation between

them and the student’s satisfaction, which partly

maximises the effectiveness of blended learning

courses.

In this research, blended learning refers to any official

course in which learners partly or fully work with

online contents or guidelines under certain control of

time, place and remote monitoring, [4]. Numerous

lecturers have recently adopted different technology-

teaching methods, [5].

Blended learning is considered a promising land

for teaching and learning at the university level as it

promotes the achievements, of course, learning

outcomes by skilfully adopting information

technological applications to maximise the

effectiveness of learning and delivery of knowledge

2.1.2 Theoretical Framework

To have successful blended learning courses, there are

a wide range of factors affecting their quality.

According to, [8], theories related to TAM

(Technology Acceptance Model), especially those

involving perceived usefulness, perceived ease of use,

perception of engagement and learners’ satisfaction,

have great impacts on the success of these courses.

Fig. 1: Recommended research model

and perceived ease of use are initial testing tools for

students’ satisfaction towards the courses. Perceived

usefulness can easily be observed and considered one

of the biggest factors in TAM. It refers to the level at

which one can use technologies to improve

PU reflects the belief that using technologies in

teaching likely leads to more effective teaching than

other methods.

the convenience of using technologies in teaching

without any user difficulty, [10]. Therefore, the PEU of

blended learning courses possibly affects learners’

engagement in learning.

learners’ positive attitudes when attending and

In this research model, the independent variable

was Perceived engagement (POE). It included two

factors: Perceived usefulness (PU) and Perceived ease

of use (PEU). The dependent one was Learners’

satisfaction (SAT). The identification of such variables

contributed to the survey in the later stages.

positive impact on learners’ engagement in the study;

this is intended to test the correlation between

Perceived ease of use and students’ satisfaction;

positive impacts on learners’ engagement in the study;

this aims to test the correlation between the Perceived

usefulness and Perception of engagement;

positive impact on learners’ satisfaction with a blended

2.2.1 Research Data

In this research, the author surveyed students of Thai

Nguyen University from January 2021 to December

2022 with random sampling as the main sampling

method. To get the necessary data, a questionnaire is

selected as the major data-collecting tool. The

questionnaire was designed based on a four-factor

measure the research concepts with five-level Likert-

scale items: 1 –disagree, 2 – disagree, 3 – neutral, 4 –

agree, and 5 –agree.

The data analysis methods were as follows.

Alpha was used to test the reliability of the variables in

the research model. Those that cannot meet the

required level of reliability were excluded and not

0.5; the Factor loading of at least 0.5 indicates the close

correlation between the variables and factors; and the

observed variables with factor loading of less than 0.5

are eliminated.

the first test aimed to identify how a model fits to the

market data. According to, [14], a model is considered

fit to the market data when it has good indices

consisting of CMIN/df not higher than 2, or 3 in some

cases, the Goodness of Fit Index (GFI), Tuker-Lewis

Moreover, the convergent value of the model was also

classified.

- Structural equation model (SEM): SEM was

used to test the research model. If a model has such

indices as CMIN/df not higher than 2, or 3 in some

cases, GFI, TLI and CFI not lower than 0.9 and

RMSEA not higher than 0.05, it is considered of good

quality and to fit with the market data, [14]. Besides, a

model with TLI and CFI ≥ 0,9, CMIN/df ≤ 2, and

Table 2 (Appendix) illustrated that the Chronbach’s

Alpha of PEU factor was 0.828, higher than 0.6, and all

variables had a corrected item-total correlation higher

than 0.3. This marked the end of the initial step of

reliability testing. As a result, 23 observed variables

were reduced to 21 with PEU1 and PEU2 excluded. All

of these 21 variables in the four-factor groups in the

official research model entered the next step of EFA.

was 0.854, which was higher than 0.5 and the Sig.

value of the Barlett test was 0.000, lower than 0.05,

which meant all 21 observed variables were correlated

and appropriate for fact analysis.

In terms of Rotated component matrix, the

author used Promax produce to minimise the number

of observed variables in one factor. This also aimed to

as only those with a factor loading of 0.5 or more were

valid for explaining a factor. In other words, after the

that were able to explain 61.88% of data fluctuations.

were clearly distinguished and all 21 observed

variables in the four groups met the requirements for

further analysis.

After implementing Cronbach’s Alpha test for

reliability of the measurements and EFA, the author

carried out CFA via AMOS to test the appropriateness

testing and identifying the appropriateness of the

SEM analysis results showed that Perceived ease

of use (PEU) had influences on Perceived usefulness

(PU), PU had impacts on Perception of engagement

(POE), and POE affected learners’ satisfaction as their

P values were all below 0.05. However, there was no

relation between PEU and POE.

In more detail, PU had positive impacts on the

POE of a blended learning course with the estimated