ANOVA Model for the Effectiveness of Blended Teaching Model
Jing Zhang and Yinan Guo
*
Jiaying University, Meizhou, China
Yinan Guo’s Email: gyncn@163.com
Keywords: Analysis of Variance Model, Blended Teaching Model, Hypothetical Test.
Abstract: Based on information technology and guided by innovative teaching concepts, the blended teaching mode
combines online network learning and offline classroom learning, making students the main body of teach-
ing. Is this king of teaching model effective? This paper proposes an ANOVA model for analyzing the effec-
tiveness of the blended teaching model, introduces the formulae and methods for parameter estimation and
hypothesis testing, and presents a linear model for evaluating the effectiveness of blended learning. The
conclusion that blended teaching mode has a significant effect on test scores was drawn.
1 INTRODUCTION
In the age of information technology, the change of
teaching paradigm is unstoppable. In order to
achieve deeper reform in education, we need to
deeply integrate information technology with educa-
tion. The main direction of development is the im-
plementation of blended teaching, which is an or-
ganic combination of "online independent learning"
and "face-to-face teacher teaching" (He 2014)
Taking Comprehensive English course as an
example, this paper conducts an ANOVA on blended
teaching to test the significance of its effect on Eng-
lish performance and to provide theoretical support
for the wide application of the blended teaching
model.
2 BLENDED TEACHING MODEL
The blended teaching mode supported by infor-
mation technology breaks the one-way integration of
traditional information technology and classroom
teaching, and builds a smart teaching platform
against the background of information technology,
so as to achieve precise guidance for teaching, in-
cluding clear sorting of teaching objectives and
overall design of teaching process, with the core
goal of cultivating students' independent learning
ability and higher-order thinking
(Mathur, R. & Oli-
ver, L. 2007
). It integrates online and offline teach-
ing methods, actively integrates "cooperation" and
"discussion" teaching design, realizes teaching and
learning as one, learning and doing as one, and uses
real-time monitoring and multi-dimensional evalua-
tion to maximize the formation of a clear under-
standing of blended learning. (Bloom 1978)
As can be seen from Table 1, the mastery of
knowledge in the blended teaching mode supported
by information technology is mainly completed
before and during class. Before class, through inde-
pendent learning, learners understand and criticize
the acquired information, integrate it with existing
knowledge, and construct the knowledge system
independently; during class, through classroom
teaching activities, the knowledge learned before
class is transferred and applied, and the construction
of the knowledge system is improved from class-
room activities. (Macdonald 2006) Capacity devel-
opment is reflected in the whole blended learning
process, and learners use resources to study inde-
pendently before class, which helps to develop in-
dependent learning ability. During the class, collab-
orative inquiry activities, online discussions and
post-class question and answer sessions help to im-
prove communication and collaboration skills.
Learning is a process from problem identification to
problem solving
(Merrill M D, 2002), and the
"learner-centered" teaching model is conducive to
the development of problem solving skills. The en-
tire blended learning process is focused on the emo-
tional experience of the learners, trying to give full
play to their motivation and create a positive learn-
492
Zhang, J. and Guo, Y.
ANOVA Model for the Effectiveness of Blended Teaching Model.
DOI: 10.5220/0011914300003613
In Proceedings of the 2nd International Conference on New Media Development and Modernized Education (NMDME 2022), pages 492-495
ISBN: 978-989-758-630-9
Copyright
c
2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)
ing environment. (Macdonald, C 1991) Evaluation
feedback is used throughout the whole blended
learning process, combining process evaluation and
summative evaluation, integrating intra-group, in-
ter-group and teacher evaluation, and multi-faceted
evaluation methods
(Lipponen, L. 2002).
In order to compare the teaching effects of
blended teaching mode and traditional classroom
teaching, the author selected two administrative
classes of the second year of computer science ma-
jors in a university in eastern Guangdong for the
teaching experiment. The experimental group was
Class 1 of Grade 20 with 41 students, and the con-
trol group was Class 2 of Grade 20 with 43 students.
84 students did not differ much in their overall level
and ability. The control group adopted the traditional
classroom teaching method, while the experimental
group adopted the blended learning mode supported
by information technology, dividing the teaching
process into three parts: before, during and after the
class. (Hofmann 2001)
Table 1 Blended teaching mode
3 ANALYSIS OF VARIANCE
MODEL
For some products in production life, there are many
factors that affect their evaluation index. To know
which factors have an impact on the product, we
need to conduct tests and analyze them according to
the test results to find out the factors that have a
significant effect is called ANOVA.
3.1 Example
Let the English scores of the students of the m clas-
ses participating in the experiment be x and the gen-
eral examination scores of the first semester final
examination be y. Where Xij denotes the English
scores of the jth student in class i and Yij denotes the
general examination scores of the jth student in class
i in the first semester final examination, (i=1,2,---,
m; j=1,2,---n). The effectiveness of the blended
teaching model is evaluated for m teaching classes.
3.2 Analysis of Variance Model
Xij is closely related to Yij, and there are two factors
here, one is the teacher, which is a qualitative factor
and is called the variance variable. The second is
Xij, which is a quantitative variable called covariate.
y
=μ+α
+γx
+ε
,（i=1,2,…，m; j=1,2,…，n
）
Teaching
Design
Teaching
Preparation
Teaching
evaluation
Analysis of the learning situation
Network Resource Selection
Instructional Desi
g
n
Before
Class
Self-directed Learning Web-based learning resources
Group assignment Group study and discussion
During
Class
Pre-learning feedback Review key points and answer
questions
Share and exchange Personalized analysis of key points
Presentation of results
Communication, discussion and
inter-group evaluation
Class Summary Wrap-up Reflection
After Class Assignment Reinforcement of learning content
Discussion Online Question and Answer
Formative evaluation + Summative evaluation
Online self-directed learning perfor-
mance
Unit Test
Teamwor
k
Results Show
Classroom Discussion Assignment grades
ANOVA Model for the Effectiveness of Blended Teaching Model
493
where α is the effectiveness of class i and satis-
fies ∑_(i=1)^m▒〖αi=0〗, εij is the random error, and
γ is the regression coefficient.
3.3 Parameter Estimation of the Model
The least squares estimation (LSE) of the unknown
parameters in (1) is μ=γ，α
= γ
γ,γ=
3.4 Hypothesis Testing of the Model
Now let's test whether there is a significant differ-
ence in teaching effectiveness α_i among m teaching
classes in model (1), and only when there is a sig-
nificant difference in teaching effectiveness among
m teaching classes, the evaluation of teaching effec-
tiveness is carried out. The hypothesis test of model
(1) is noted as.：
Test the statistic of H_0:α_i=0,(i=1,2,...,m)
F=
（）
⁄
（）
⁄
~F（m1,nm1）,
For the convenience of calculation, the data are
presented in the form of Table 2 for data analysis
Table 2 Test table for the analysis of covariance model
Source of
variance
Modified Sum
of Squares
Correction of
degrees of
freedom
Mean Square F ratio Threshold Significance
between
classes
R
=
R
R
m1
V
=
R
m1
F
=
V
V
F
= （m
1，nm
1）
In Class
R
=S
S
S
nm1
V
=
R
nm1
Total
R
=S
S
S
n2
Of which n =
∑
n
, x
=
∑
x
, y
=
∑
y
, x=
∑∑
x
,
y=
∑∑
y
, S
=
∑
x
x
,
S
=
∑
y
y
,
S
=
∑
x
x
y
y
, S
=
∑∑
（x
x）
,
S
=
∑∑
（y
y）
, S
=
∑∑
x
x
y
y
,
S
=
∑
S
, S
=
∑
S
, S
=
∑
S
,
3.5 Linear Model for Effectiveness of
Blended Instruction
When there is a significant difference in α_i, the
linear model for evaluating teaching effectiveness is
constructed using the regression parameter γ in
model (1), and because there is an effect of the co-
variate x_ij, the covariates need to be taken at the
same level at the same time, that is, the mean of the
English Advanced
Placement scores of each class
(y_i ) ̅ minus γ ̂ = (x_i ) ̅ - x ̅, that is, the estimated
value of the regression coefficient γ in (2) to re-
move. The linear model (Yang Wenli 1998) of the
effectiveness of blended instruction after ranking the
mean scores after the effect of achievement x is
x
=y
γ（x
x）， （γ=
,i=
1, … ，m）
4 CONCLUSION
The constructivist view of knowledge suggests that
effective teaching and learning emphasizes the time-
liness and output effectiveness of instruction.
Teachers use both theoretical foundations to bring
intelligent technology into teaching and learning,
achieving a higher level of integration between in-
formation technology and classroom teaching (Oli-
ver, M., & Trigwell, K.2005).
NMDME 2022 - The International Conference on New Media Development and Modernized Education
494
This paper proposes an ANOVA model for ana-
lyzing the effectiveness of the blended teaching
model, introduces the formulae and methods for
parameter estimation and hypothesis testing, and
presents a linear model for evaluating the effective-
ness of blended learning. The conclusion that
blended teaching mode has a significant effect on
test scores was drawn.
Of course, there is still much room for improve-
ment in the blended learning model, and teachers
should gradually explore teaching strategies that are
appropriate to it, and students should also enhance
their sense of autonomy and collaboration.
REFERENCES
Bloom. New views of the leader: Implications for Instruc-
tion and Curriculum. Published at Educational leader-
ship of America .1978:3～8
Douglas, C. Montgomery, Design and Analysis of Exper-
iments [M] John Wiley & Sons Inc., U.S.A.,
1991:369-385
He Kekang. How to realize the “deep integration" of in-
formation technology and education [J]. Curriculum -
Teaching Materials - Teaching Method, 2014 (2).
Hofmann, J. (2001). Blended Learning Case Study.
Learning Circuits, 5(3), 26.
Lipponen, L. (2002). Exploring Foundations for Comput-
er-Supported Collaborative Learning. In G. Stahl
(Ed.), Computer Support for Collaborative Learning
(pp.72-81).
Macdonald, J. (2006). Blended Learning and Online Tu-
toring. Burlington: Gower Publishing Limited
Mathur, R. & Oliver, L. (2007). Developing an Interna-
tional Distance Education Program: A Blended Learn-
ing Approach, Online Journal of Distance Learning
Administration, 10(4)
Merrill M D. First principles of instruction[J]. Educational
Technology: Research and Development,2002,
(3):43-59
Oliver, M., & Trigwell, K. (2005). Can “Blended Learn-
ing” Be Redeemed? E–Learning, 2(1), 17-26.
Yang Wenli. Introduction to Linear Models [M] Beijing:
Bejing Normal University Press 1998.166-211
ANOVA Model for the Effectiveness of Blended Teaching Model
495
