Analysis of the SERVQUAL Based Mixed Teaching Model in

Colleges

Yinghong Li

Shandong Institute of Commerce and Technology, Shandong, China

Keywords: Hybrid Teaching Model, SERVQUAL Model, Quality Assessment Information.

Abstract: Teachers' information technology teaching level and teaching quality can be improved by combining online

and offline teaching. This paper constructs the SERVQUAL model to measure the service quality of college

professional teaching from the perspective of service quality management. Using factor analysis and entropy

weight method, this paper determines the dimension coefficient of the model, analyzes each index, and dis-

cusses the priority order of index action. The purpose of this paper is to provide a reference for improving the

quality of professional teaching services in colleges and universities.

1 INTRODUCTION

The teaching model is built around student-centered

learning and development, designing teaching situa-

tions according to real job scenarios, further refining

teaching tasks according to the teaching situations,

and turning students into workplace people, playing

different roles and completing various work tasks.

With the TTR mechanism as the guarantee, the mo-

bile classroom and the physical classroom, the pre

class and post class double track and three links run

alternately, and finally achieve the teaching goal. The

student-centered, context-led task-driven hybrid

teaching model is shown in Figure 1.

Figure 1 Context-led task-driven hybrid teaching model

The use of data mining technology in educational

research is a common method of exploring the rela-

tionship between evaluation object and evaluation in-

dex. The development trend in teaching evaluation

and teaching management is to improve the quality of

https://orcid.org/0000-0002-1854-2751

teaching through effective evaluation standards and

methods. The teaching model emphasizes student-

centered learning and development. Making the

teaching environment and experience as realistic as

possible, refine the teaching tasks accordingly, and

Li, Y.

Analysis of the SERVQUAL Based Mixed Teaching Model in Colleges.

DOI: 10.5220/0011910800003613

In Proceedings of the 2nd International Conference on New Media Development and Modernized Education (NMDME 2022), pages 295-300

ISBN: 978-989-758-630-9

 2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)

295

convert students into professionals. Using the TTR

mechanism, the mobile classroom and physical class-

room, operating before and after class, in a double

track, three-way alternating pattern, finally achieve

their educational objectives (Luo 2022).

SERVQUAL model is based on the difference

theory. It believes that service quality has five dimen-

sions, namely, tangibility, reliability, responsiveness,

assurance and empathy. Evaluators' perception of ser-

vice quality is evaluated from these five dimensions.

SERVQUAL model is to set 22 secondary indicators

(questions) based on five dimensions. Evaluators

score the psychological expectations and actual feel-

ings of secondary indicator questions according to

their own experiences and actual needs. The differ-

ence between the two scores is the evaluator's evalu-

ation of service quality. The smaller the difference,

the higher the evaluation.

2 THE PURPOSE OF THIS

STUDY IS TO CONSTRUCT AN

EVALUATION AND ANALYSIS

MODEL OF MIXED TEACHING

QUALITY IN COLLEGES.

Each index score is evaluated based on the difference

between psychological expectation and actual experi-

ence, As shown in Formula (1), the difference is the

students' evaluation of the teaching quality.

𝑆𝑄=𝑊







𝑃



−𝐸







1

1

（1）

Where, 𝑅is the number of second-level indicators to

which each attribute belongs; 𝑃



is students' psycho-

logical expectation score for the 𝑖 th question; 𝐸



the score of students' actual feelings about the ith

question, and 𝑊



is the weight of each dimension.

Because different attributes have different degrees

of influence on the evaluation of teaching quality, us-

ing the same weight assignment will result in large

deviation. The research uses entropy weight method

to determine the weight of each secondary indicator

of the model. Because the secondary indicators of

each attribute have strong relevance (factor analysis

can be used to extract factors for verification), the

weight of each attribute is the accumulation of the

weight values of the secondary indicators.

As a result of the different properties of the teach-

ing quality evaluation, the weight of the same assign-

ment will produce large deviations. To determine the

weights of each secondary index in the model, the re-

search uses the entropy weight method. Because sec-

ondary indicators have strong correlations, the weight

of each attribute is part of the secondary index

weights.

With online teaching quality evaluation to illus-

trate the basic principle of entropy weight method,

there are 𝑚 students participating in online teaching

quality evaluation, 𝑛 secondary index of evalua-

tion, the statistical value of student evaluation index

is marked 𝑋





, indicating the evaluation of the JTH

index by the 𝑖 th student. The extreme value method

is used to standardize statistical data to eliminate the

influence of different units among indicators. The

standardized formula is as follows:

𝒚

𝒊𝒋

𝒙

𝒊𝒋

𝟏

𝒎𝒊𝒏



𝒙

𝒊𝒋



𝒎𝒂𝒙

𝒋

𝒙

𝒊𝒋

−𝒎𝒊𝒏𝒙

𝒊𝒋



𝟏

(2)

After quantifying the same degree of standardized

indicators, the proportion of the 𝑖 th student's eval-

uation value of the 𝑗 th indicator is:

𝑷𝒊𝒋=

𝒚

𝒊𝑱



𝒚

𝟐

′

𝒋

𝒎

𝒊𝟏

(3)

Weight of each attribute:

𝑾

𝒋

=𝒘

𝒋

𝑹

𝒊𝟏

(4)

Score gaps between multiple evaluators on a sin-

gle index increase with the larger the gap. The smaller

the entropy and the larger the entropy weight, the

greater the entropy, indicating that the index can pro-

vide more information about the research topic and

play a greater role in sample comparison.

The perceptual difference (psychological expecta-

tion-actual feeling) of 230 samples was calculated to

obtain the perceptual matrix of online teaching eval-

uation, and the entropy weight of 26 second-level in-

dicators was calculated by entropy weight method (as

shown in Table 1).

NMDME 2022 - The International Conference on New Media Development and Modernized Education

296

Table 1. Entropy weight, the second-level index of SERVQUAL Scale for teaching Quality in Colleges

Number A1 A2 A3 A4 A5 A6 A7 A8 A9

Entropy weight 0.0457 0.0547 0.0214 0.0277 0.0369 0.0147 0.0258 0.0369 0.0124

Number B10 B11 C12 C13 C14 C15 C16 D17 D18

Entropy weight 0.0248 0.0325 0.0244 0.0257 0.0235 0.0741 0.0239 0.0258 0.0147

Number D19 D20 D21 E22 E23 E24 E25 E26

Entropy weight 0.0147 0.0741 0.0258 0.0357 0.0159 0.0753 0.0852 0.0458

3 TEACHERS' DIGITAL

TEACHING ABILITY AND

INNOVATION AND

ENTREPRENEURSHIP

ABILITY

Online and offline hybrid teaching puts forward un-

precedented high standards and requirements for

teachers' digital processing ability, information teach-

ing ability and innovation and entrepreneur-ship abil-

ity.

As a result of online and offline hybrid teaching,

teachers are expected to demonstrate unprecedented

levels of digital processing ability, information teach-

ing ability, and innovation.

3.1 Transforming Teachers' Teaching

Philosophy

Traditional teaching has formed a teacher-centered

closed classroom model, teachers guide and act by

themselves, blindly instill knowledge into students,

only to complete teaching tasks, and completely

ignore students' independent learning and

personalized development. If teachers can't adapt to

the development of the times, change the outdated

educational ideas, and change their roles in education,

they will certainly mislead students. So, teachers must

break through the old and create new ones, establish

the information mixed teaching concept, and adopt

the most advanced teaching technology and means

(Liu 2020, Yu 2022).

3.2 Teachers are Encouraged to

Participate in Various IT Teaching

Ability Trainings

The school has created all conditions, invited IT

teaching experts, digital processing experts,

innovation and entrepreneurship entrepreneurs to

give lectures in the school, encouraged teachers to

actively participate in various forms of online and

offline modern teaching ability training, digital

processing ability training, digital quality training and

innovation and entrepreneurship ability training, held

symposiums and exchanges, and discussed and

consulted with each other.

3.3 Teachers Are Encouraged to

Participate in Various Levels of

Teacher Teaching Ability

Competitions

By learning the content of the competition, forming

pairs to participate, preparing for the competition,

participating in the competition, exchanging the

competition and reflecting on the competition, a set of

process, even if they do not win the prize, the teachers'

informatization teaching level must have improved.

3.4 Build an Intelligent Teaching

Platform Centered on Active

Student Learning

Blended online and offline teaching requires teachers

to upload digital resources, students to be able to learn

locally when and where they want, and to be able to

implement interactive online and offline teaching.

This requires teachers to build a platform. To

successfully build a teaching platform in a short

period of time, the following steps are required.

Step 1: Understand the dual interface and dual

ports of the platform. Dual interface refers to the

teacher interface and student interface, and dual ports

refer to the computer side and mobile side.

Step 2: Build the platform. To build a platform,

you have to build a course first, and give all your

courses a good framework and columns. Then upload

digital resources (Liang 2021). Set up before, during

and after class activities.

By building a platform course and implementing

online and offline hybrid teaching, the teacher

uploads digital materials before class, allowing

Analysis of the SERVQUAL Based Mixed Teaching Model in Colleges

297

students to pre-study online and discuss problems

with the teacher and classmates at any time, the

teacher explains the important and difficult problems

offline during class, and through online questions,

discussions, salons, quizzes and quizzes, the online

and offline interleaved operation improves students'

motivation, and the teacher assigns homework and re-

leases extended materials after class to Classroom

knowledge is further enhanced.

4 AN EMPIRICAL ANALYSIS OF

BLENDED TEACHING IN

COLLEGES AND

UNIVERSITIES

4.1 Data Sources and Processing

This paper selects the online teaching quality

evaluation data of business administration majors in

a university in Shandong Province for empirical

analysis. Each student (considering that there are few

theoretical courses for senior students and the sample

size is insufficient, this paper does not do research)

scores the expectation and actual perception of 38

indicators, and uses formula (1) to calculate the

evaluation of individual students on the current online

teaching quality. The online questionnaire

(questionnaire star) was used to eliminate invalid

samples, and a total of 230 samples were determined,

including 85 marketing majors and 145 business

administration majors.

4.2 Construction of Comprehensive In-

dicators of Teaching Evaluation

1. Ensure the reliability of the SERVQUAL scale and

each dimension subscale by conducting a reliability

analysis, it is necessary to conduct reliability, validity

and paired sample 𝑇 -test on the collected question-

naires to determine the designed online teaching qual-

ity scale for colleges and universities. At the same

time, the differences in the scores of students' psycho-

logical expectations and actual feelings about the in-

dicators in the questionnaire were compared. The rel-

evant indexes obtained by SPSS software are shown

in Table 2.

Service quality and gap The SQ value of the over-

all service quality of professional education is 0.358,

and its percentile score is 92.12; The service quality

SQ value of each indicator is 0.07-1.2 (Table 2), in

which the mean values of the "curriculum setting",

"curriculum teaching" and "teaching resources" di-

mensions are 0.0658, -0.0998 and 0.195 respectively.

The paired t-test results showed that there were statis-

tically significant differences between students' ex-

pectations and perceptions of various indicators of

professional education services (P<0.05, Table 2).

Digital resources should not only achieve sharing

with the paper content of the textbook, but also

achieve interaction with the platform. Ping The plat-

form includes resource platform, teaching plat-form,

training platform, etc. The digital content of the

teaching materials should be shared with the platform

content to facilitate teachers and students to access

and learn anytime and anywhere using different web

tools and mobile tools, and to promote more interac-

tive learning between teachers and students. To facil-

itate students' reading and use, loose-leaf textbooks

can be developed according to the characteristics of

the course.

In order to ensure the reliability of SERVQUAL

and its sub scales, it is necessary to conduct reliabil-

ity, validity and paired sample T tests on the collected

questionnaires to determine the designed online

teaching quality scale for colleges and universities,

and compare the differences between the students'

psychological expectations and actual feelings of the

indicators in the questionnaire. Relevant indexes ob-

tained by SPSS software are shown in Table 2

Table 2. SERVQUAL Scale for Teaching Quality in Colleges and Universities: Reliability and Validity

Scale α KMO

Bartlett's test

for sphericity

Factor load-

ings

Factor load-

ings

SERVQUAL Total 0.897 0.997 0.000 - -

Tangibility Quality Scale 0.810 0.784 0.000 0.547 0.000

Reliability Quality Scale 0.904 0.946 0.000 0.577 0.000

Responsiveness Quality Scale 0.987 0.863 0.000 0.641 0.000

Assurance Quality scale 0.901 0.897 0.000 0.487 0.000

Empathy Quality Scale 0.963 0.904 0.000 0.701 0.000

NMDME 2022 - The International Conference on New Media Development and Modernized Education

298

Table 3. KMO and Bartlett spherical tests and Cronbach's coefficient of correlation

Dimension Index number

Cronbach’s α KMO

Expectation

subscale

Perceptual

subscale

Expectation

subscale

Perceptual

subscale

The curriculum 7 0.721 0.856 0.851 0.829

Course teaching 4 0.851 0.744 0.732 0.873

Teaching resources 8 0.744 0.893 0.799 0.805

Toatal 19 0.825 0.862 0.816 0.844

5 CONCLUSION

Combining factor analysis and entropy weight

method, the five-dimension weight coefficients of the

SE-VQUAL model can effectively reflect the im-

portance of each dimension, and the SE-VQUAL

model based on the weight coefficients can reasona-

bly score online teaching quality. The result shows

that the score of online teaching quality is propor-

tional to the grade and major. In view of the curricu-

lum setting in colleges and universities, it shows that

colleges and universities should strengthen the im-

provement of online teaching quality of basic courses;

There is a gap in the teaching quality scores of differ-

ent majors, which indicates that there are differences

between the secondary majors of business administra-

tion. Colleges and universities should improve or

evaluate the secondary majors based on the particu-

larity of the secondary majors.

1. Cronbach's α coefficient was used to test the re-

liability of SERVQUAL model. The overall

Cronbach's α coefficient of the scale was greater than

0.900, and each dimension was between 0.802 and

0.859 (Table 1). Based on KMO and Bartlett spheri-

cal tests, all dimensions had KMO values greater than

0.5 (Table 1), and the differences were statistically

significant (P＜0.05). The results of factor analysis of

SERVQUAL model validity showed that the factor

matrix was orthogonal rotated with maximum vari-

ance, and the three factors with characteristic root

greater than 1 accounted for 94.930% and 64.304%

of the perceived and expected variation.

2. Using the SERVQUAL model scale, the empir-

ical study builds an online teaching quality evaluation

scale for colleges and universities., and the Reliability

and validity are tested. In each dimension index, there

is a significant difference between psychological ex-

pectation and actual perception. This shows that the

SERVQUAL model can be used to evaluate colleges'

and universities' teaching quality. Additionally, it

combines the generality of service quality manage-

ment theory with the specificity of university teaching

and learning quality management. Provide new ideas

for studying teaching quality management in colleges

and universities, and improve the theory of teaching

quality management.

By building a platform course and implementing

online and offline hybrid teaching, the teacher up-

loads digital materials before class, allowing students

to pre-study online and discuss problems with the

teacher and classmates at any time, the teacher ex-

plains the important and difficult problems offline

during class, and through online questions, discus-

sions, salons, quizzes and quizzes, the online and of-

fline interleaved operation improves students' moti-

vation, and the teacher assigns homework and re-

leases extended materials after class to Classroom

knowledge is further enhanced.

REFERENCES

Dado, Taborecka–Petrovicova, J. Riznic, D, & Rajic, An

Empirical Investigation into the Construct of Higher

Education Service.

Foropon, Seiple, Kerbache, Using SERVQUAL to Exam-

ine Service in the Classroom: Analyses of Undergradu-

ate and Executive Education Operations Management

Courses [J]. International Journal of Business & Man-

agement, 2013(20): 105 - 116

Guo Lijun Current Situation and Trend of Research on Uni-

versity Teaching Evaluation since the 21st Century [J].

Modern University Education, 2019 (6)

Guruler H, Istanbullu A. Modeling student performance in

higher education using data mining [J]. Educational

Data Mining, 2014 (1) :105 - 124.

Int. Conf. on Information Technology Interfaces. Cavtat:

IEEE, 2012. 207-212.

Jia Yanhong, Zhao Jun, Zhao Chuanyan, Wang Shengli.

Evaluation of Grassland Ecological Security based on

entropy weight Method: A Case study of Gansu Pasto-

ral areas [J]. Chinese Journal of Ecology, 2006(8):

1003-1008.

Krpan D. , Stankov S. Educational data mining for grouping

students in E-learning system [C]. Proceedings of the

ITI 2012 34th

Analysis of the SERVQUAL Based Mixed Teaching Model in Colleges

299

Liang Shan. Discussion on the construction of digital re-

sources for accounting majors in higher education insti-

tutions -taking financial accounting course as an exam-

ple. Journal of Yan'an Vocational and Technical Col-

lege, August 2021: Vol. 35, No. 4.

Liu Jian, Huang Yuying, Yan Lichao. Data Mining and

Analysis of Classroom Teaching Evaluation [J]. Journal

of Education Science, Hunan Normal University,

2019(2): 118-124.

Liu Xiao. Exploring the improvement of teachers' ability in

blended teaching mode. Journal of State Grid Technical

College,2020 05): Vol. 23, No. 5.

Liu Xiaozhe. Some Thoughts on Promoting the Construc-

tion of Undergraduate Teaching Evaluation Index Sys-

tem [J]. China University Teaching, 2014 (7)

Luo Zhiyong, Gan Mengyu, Luo Yiwen. Research on the

construction of a blended teaching model of university

chemistry based on flipped classroom. University Edu-

cation, July, 2022.

Quality [J]. Serbian Journal of Management, 2012(2): 203

－218.

Shan Wengui. College Teaching Evaluation Index System

and Information Processing [J]. Work Viewpoint, 2004

(10)

Wu Zhenggang, Yan Ming, Zhang Ruihong. The Construc-

tion of Student centered University Quality Evaluation

System [J]. Heilongjiang Education (Higher Education

Research and Evaluation), 2019 (4)

Yu Haijiao. Exploration and practice research of "flipped

classroom" in electronic information class based on

"online+offline" hybrid teaching mode. Computer

Knowledge and Technology, July 2022: Vol.18, No.21.

NMDME 2022 - The International Conference on New Media Development and Modernized Education

300