Maturity Model for Assessment of Personalization of
Higher Education
Mariia Rizun
a
and Małgorzata Pańkowska
b
Department of Informatics, University of Economics in Katowice, Katowice, Poland
Keywords: Maturity Model, Higher Education, Personalization of Education, Maturity Level.
Abstract: The paper presents the Education Personalization Maturity Model, developed to provide higher education
institutions with a tool for assessment of the level of their personalized approach to their students. These days
students tend to be much more engaged in the education process; they want their preferences to be considered
in not only in the learning process but in all aspects connected with educational institutions. The presented
model covers four major key process areas of practices at higher education institutions: students’ online
platform/website, courses and fields of study, research activity, and other extracurricular activities. The
Education Personalization Maturity Model is used to assess the personalization of 51 higher education
institutions in 25 countries. The results of this assessment are presented and analyzed in the paper.
1 INTRODUCTION
There is no doubt that higher education worldwide
has undergone significant changes not only within the
last century but even within the last decade.
Governments have been changing their educational
policies, higher education institutions (HEIs) have
been adapting to these novelties as well as
introducing their internal regulations to stay
competitive and attractive for students.
Moreover, the attitude of students towards
education has changed. They are now perceived as
customers and active players in establishing their
learning path (Orîndaru, 2015). These days, when
talking about the quality of higher education, we talk
about a significantly increasing engagement of
students. Such engagement is considered a measure
of the quality of an educational institution: students
of a good institution are supposed to be actively
involved in educationally purposeful activities
(Quaye & Harper, 2014). It is seen as the premise of
students’ happiness. Researchers state that
guaranteeing students’ happiness as a result of their
development is much more crucial than just satisfying
students’ needs as consumers. Students who are
“happy” are more content with their engagement in
a
https://orcid.org/0000-0002-9646-7638
b
https://orcid.org/0000-0001-8660-606X
educational experiences, while those who are just
“satisfied” are more concerned with how education
services were delivered rather than in their
involvement with the process (Dean & Gibbs, 2015).
One of the ways to make students happier about
their educational path is to provide them with a
personalized approach of their HEIs towards their
preferences and aspirations. Developing personalized
education for students means, among others: allowing
them to tailor their study program as they desire (at
least to some extent) (Rollande & Grundspenkis,
2016); providing them with tutors or mentors who
help students define their educational and
professional path (Rollande, 2015); compromising
with students and adapting study plans to their
preferences (as much as it is possible) so that students
could combine studies with work or other important
activities (Grundspenkis, 2010); increasing the
number of workshops and other practical activities to
make students familiar with the business environment
(Zhu, 2016); motivating them to be curious, to
conduct research; developing good infrastructure
with all necessary facilities (e.g., sports, computers,
internet, library, etc.) (Kabak & Dagdeviren, 2014),
and many others.
When students study, participate in research
projects, take part in internships and exchange
Rizun, M. and Pa
´
nkowska, M.
Maturity Model for Assessment of Personalization of Higher Education.
DOI: 10.5220/0011537900003335
In Proceedings of the 14th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2022) - Volume 3: KMIS, pages 43-53
ISBN: 978-989-758-614-9; ISSN: 2184-3228
Copyright
c
2022 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
43
programs, realize projects, or participate in any other
activities offered by their educational institutions,
they form some kind of a portfolio of all their
experience, knowledge, skills, and abilities. In other
words, it can be called a profile or, specifically, a
student’s Individual Higher Education Profile (also –
Individual Profile). This Individual Profile of a
student is most effectively formed with the
personalized approach of educational institutions
towards their students when HEIs provide students
with the elements of personalization discussed in the
previous paragraph (and many more). To make sure
the students are content with their educational path
realization, HEIs need to be able to answer such
questions as: what are the first steps towards
education personalization development; what rights
and privileges the students may have, and in what
activities there should be restrictions; how to evaluate
whether the approach towards students is
personalized, and to what extent; and many more.
The objective of this paper is to present a tool for
assessment of the level of personalization of
education at higher education institutions, developed
by the authors, which is the Education
Personalization Maturity Model (EPMM).
1.1 Maturity Models in HEIs
To examine the maturity models for educational
institutions, developed by researcher in the last
decade, the authors conducted the Systematic
Literature Review (SLR). In the process of selection
by title, abstract and paper content (in accordance
with the PRISMA guidelines
3
), the authors received
43 papers eligible for further review. In these works,
the authors revealed 13 maturity models created for
higher education institutions (Table 1). These models
are dedicated to the maturity of e-learning,
Information and Communication Technology (ICT),
planning and assessment of learning processes, and
some other processes that run at higher education
institutions.
Due to the fact that the literature review did not
reveal any maturity model dedicated to education
personalization or student’s Individual Profile
development, the authors see a research gap which
may be filled with the suggested EPMM model,
presented further in this paper.
3
“Preferred Reporting Items for Systematic Reviews and
Meta-Analyses”. https://prisma-statement.org. [Accessed:
20.08.2022.
Table 1: Maturity models for HEIs: literature review
results.
Author(s) Maturity model
E-learnin
g
maturit
y
(Nsamba, 2019)
Maturity Assessment
Framework for Open
Distance E-Learning
(MAFODeL)
(Marshall & Mitchell,
2004
)
,
(
Marshall, 2010
)
E-Learning Maturity Model
(
eMM
)
(Penafiel et al., 2017)
Contribution to the eMM
with the inclusion of a new
Key Process Area –
Accessibilit
y
(Hong & Xinyi, 2019)
E-learning Process Capability
Maturity Model (EPCMM)
ICT maturit
y
(Durek, Kadoic, &
Begicevic Redep, 2018)
Digital Maturity Framework
for Higher Education
Institutions
(
DMFHEI
)
(Aliyu et al., 2020)
Holistic Cybersecurity
Maturity Assessment
Framework
HCYMAF
L
earning process planning and assessment maturity
(Thong, Yusmadi, Rusli,
& Nor Hayati, 2012),
(Thong, Jusoh, Abdullah,
& Alwi, 2013
)
Curriculum Design Maturity
Model (CDMM)
(Reçi & Bollin, 2017),
(
Reçi & Bollin, 2019
)
Teaching Maturity Model
(
TeaM
)
(Enke, Glass, &
Metternich, 2017),
(
Enke et al., 2017
)
Maturity Model for Learning
Factories
Other
p
rocesses maturit
y
(Carvalho, Pereira, &
Rocha, 2019
)
Higher Education Institutions
Maturit
y
Model
(
HEIMM
)
(Secundo, Elena-Perez,
Martinaitis, & Leitner,
2015
)
Intellectual Capital Maturity
Model (ICMM) for HEIs
(Matkovic, Pavlicevic, &
Tumbas, 2017)
Business Process Modeling
Maturity and Adoption
Model for HEIs
(Boughzala & de Vreede,
2015
)
Collaboration Maturity
Model
(
Col-MM
)
The structure of the paper is as follows: Section 2
contains the methodology of the EPMM
development; in Section 3 structure of the EPMM is
presented; in Section 4, the authors briefly introduce
the results of verification of the developed Model at
51 HEIs; in Discussion the authors conclude on the
obtained results of EPMM verification and
distinguish the contribution of this work, its
limitations, and potential further research.
KMIS 2022 - 14th International Conference on Knowledge Management and Information Systems
44
2 EDUCATION
PERSONALIZATION
MATURITY MODEL
DEVELOPMENT
METHODOLOGY
2.1 Design Decisions when Developing
the Education Personalization
Maturity Model
The section presents the methodology of building the
Education Personalization Maturity Model developed
by the authors. The primary objective of the EPMM
Model development is to identify the practices related
to the personalization of education at HEIs and to
create the methodology of assessment of the quality
of these practices’ realization.
In 2011, (Mettler, 2011) suggested a framework
for maturity model design process, which consists of
five iterative steps: identify need or new opportunity,
define scope, design model, evaluate design, and
reflect evolution. The authors used this methodology
when making decisions for the EPMM Model (Rizun,
2021).
At the stage “Identify need or new opportunity”,
the authors see the EPMM Model as the emerging
and “new one – because no maturity model for
personalization of higher education has been
developed so far. The “Define scope” decisions are
supposed to set the outer boundaries for the EPMM
Model application and use (de Bruin et al., 2005). The
authors’ choices are: specific issue(applied only for
HEIs), inter-organizational (covers internal
processes of HEIs and their cooperation with other
organizations), and both the Management- and
Technology-oriented staff of HEIs. In the “Design
model” activity, the EPMM Model is characterized as
follows: a process-oriented and multi-
dimensional(focuses on several objectives) model,
where the design process is a combinationof theory
(e.g., literature review) and practice (the experts’
knowledge). The design product of the EPMM Model
is acombinationof textual description and software
instantiation. The EPMM Model is supposed to be
implemented by HEIs with no third parties engaged,
so the application method is self-assessment”.
Finally, the “combination” of management, staff, and
business partners, as respondents of the EPMM
Model, was selected. The subject of evaluation in the
“Evaluate design” stage is “design product”. The
evaluation and verification of the EPMM Model are
to be conducted before it is implemented, so the
option of “ex-ante” evaluation is selected.
Additionally, the evaluation is to be performed with
the naturalisticmethod, i.e., it should be based on
the experience and reflection of real users (Carvalho
et al., 2019). In the “Reflect evolution” activity, the
authors have selected continuous evolution. The
authors also believe that the modifications in the
EPMM Model can be implemented by its users,
which leads to the “external/open” structure of
change.
2.2 Information Basis of the EPMM
Model Development
In the process of developing the Education
Personalization Maturity Model, the authors used the
following primary sources of information:
1. Literature on maturity models in general and
those developed for educational institutions. It
provided the authors with knowledge on how the
models should be constructed and what are the
obligatory and optional constructs of a maturity
model.
2. Design decisions developed in (Mettler,
2011), discussed above. They allowed the authors to
define the issues of particular importance in the
process of maturity model design and to organize and
document this process.
3. Resolutions, ordinances, regulations, and
other official documents, issued by the authorities of
the University of Economics in Katowice (UEKat),
which is the authors’ affiliation. On the basis of these
documents, the authors built a picture of UEKat
policy as for personalized approach toward students.
Good practices were analyzed and used in the EPMM
Model as examples of high levels of personalization
maturity; practices that might require improvement
were put to the lower levels of maturity.
4. Opinions of students of a few Polish HEIs,
obtained through a questionnaire survey.
Since the primary focus of the paper is the final
version of the EPMM Model, not each step of its
development, the detailed questionnaire results are
not presented. As stated above, they found reflection
in the EPMM Model structure.
3 EDUCATION
PERSONALIZATION
MATURITY MODEL
STRUCTURE
Researchers distinguish three major components that
Maturity Model for Assessment of Personalization of Higher Education
45
define a maturity model (Nelson et al., 2014):
content, its quality, and the indicators of maturity
status. The content is formed by practices, processes,
and categories. Under the term “practices”, the
authors understand the policies and activities of a HEI
on specific issues, which are the focus of a particular
maturity model or framework. In the case of the
EPMM, the focus is on the practices connected with
personalization development. Practices of a similar
kind could be synthesized into broader process
categories or key process areas (KPAs).
The basic structure of the EPMM is presented in
Figure 1.
Source: own
Figure 1: Basic structure of the EPMM Model.
The authors distinguish 34 practices connected
with the personalization of education. That is, with
the realization of these practices, personalization of
students’ education at HEIs is formed and/or
improved. These 34 practices are grouped into four
KPAs. Each of the practices within each of the four
KPAs contains descriptions of five cases, each
referring to a certain maturity level (from 0 to 4, as
shown in Table 2). Each case is a situation suggested
to occur at the analyzed HEI; the higher the level, the
more expanded the description of the case is, i.e., the
more advanced personalization of education is
observed at the selected HEI.
3.1 Levels of Maturity
As stated in the literature, organizations may be
characterized by levels of maturity or by dimensions
(Marshall, 2010). In dimensions, a five-point
adequacy scale is used to evaluate the quality of
performed practices (Anicic & Divjak, 2020). The
scale of maturity levels starts with level 1, which
characterizes the maturity of an organization as the
one that exists at an initial level. It is supposed that at
this level, some practices expected to be realized, are
present, but there is no system, and the realization is
somewhat chaotic and poorly controlled.
However, the authors consider that it might be
reasonable to define a level of maturity characterized
as zero maturity for an organization that has not
made a single step toward developing a particular
practice analyzed. The authors’ suggestion is to name
this level as not assessed (0)” – referring to the first
of the values of maturity assessment criteria (not
assessed, initial, partially adequate, largely adequate,
fully adequate). Therefore, the authors suggest
modifying a more standard scale of maturity levels,
introducing a “zero” or “nor assessed” level of
maturity of personalization of education at higher
education institutions.
The other amendment to the maturity levels,
suggested by the authors, is merging the levels
Table 2: Maturity levels in the Education Personalization Maturity Model.
Levels of
p
ersonalization maturit
y
Description of the levels
Not assessed (0) There is no evidence of personalization of education at the selected HEI.
Initial / ad hoc (1)
Few processes connected with personalization are defined, but much depends on individual effort
and will of a participant in the processes involved. Realization of practices is not systematic, and
there is no centralized control over them.
Repeatable (2)
Basic management is established. Some processes are consistent, but there is still no discipline
for all the processes and sub-processes that might contribute to personalization development.
Defined &
managed (3)
Development of personalization is standard, consistent, and predictable. Practices are
documented and integrated into standard processes. However, the selected HEI lacks suggestions
for potential improvement of the practices connected with personalization. Realization of
p
ractices might not be connected with the external environment.
Optimizing (4)
The process of personalization is being constantly improved. Personalization practices are
formally defined. All practices are controlled and documented. External environment of the
selected HEI is activel
y
en
g
a
g
ed in
p
ersonalization develo
p
ment.
Source: own, based on (Anicic & Divjak, 2020)
KMIS 2022 - 14th International Conference on Knowledge Management and Information Systems
46
“defined” and “managed” into one. The authors give
it a simple name “defined and managed”. It is
suggested that at this level of personalization, all the
processes, sub-processes, and different activities, are
already clearly defined and precisely controlled, yet
the selected educational institution is supposed to
remain at the same stage of personalization
development, and there is no improvement observed.
Table 2 contains the authors’ version of the five
levels of maturity applied in the EPMM.
3.2 Key Process Areas and Practices
Researchers distinguish three major components that
define a maturity model (Nelson et al., 2014): content,
its quality, and the indicators of maturity status. The
content is formed by practices, processes, and
categories. Under the term “practices”, the authors
understand the policies and activities of a HEI on
specific issues, which are the focus of a particular
maturity model or framework. In the case of the
EPMM Model, the focus is on the practices connected
with personalization development. Practices of a
similar kind could be synthesized into broader
process categories or key process areas.
The authors distinguished 34 practices connected
with the personalization of education. That is, with
the realization of these practices, personalization of
students’ education at HEIs is formed and/or
improved. These 34 practices are grouped into four
key process areas. Each of the practices within each
of the four KPAs contains descriptions of five cases,
each referring to a certain level of maturity (from 0 to
4, as shown in Table 2). Each case is a situation
suggested to occur at the analyzed HEI; the higher the
level, the more expanded the description of the case
is, i.e., the more advanced personalization of
education is observed at the selected HEI.
A list of the defined key process areas is presented
in Table 3. In Tables 4-7, practices for each KPA are
enumerated. In addition, these tables include ratings
or, better say, weights of each of the KPAs and
practices. The higher the weight, the more important
a certain KPA or practice is supposed to be, and the
greater its role is in the development of
personalization of education at HEIs.
KPA1 “Students’ Platform” (SPL), contains
seven practices (Table 4). Under the Students’
Platform, the authors understand a separate website,
or a part of HEI’s website, which is dedicated only to
students’ needs like providing them with information
about fields of study and specializations, elective and
obligatory courses, practices, and internships, exams
and grades, conferences, social events, changes in
schedule or any other organizational issues;
submitting documents or registration forms; and other
issues that might be necessary for students of a
particular educational institution.
Table 3: Key process areas in the EPMM Model.
Key process area Acronym Rating
Practices
inside the
KPA
Students’ Platform
SPL 0,400 SPL1 – 7
Courses and Fields of
Study
CFS 0,300 CFS 1 – 15
Research Activity
RSA 0,100 RSA 1 – 4
Extracurricular Activities
EXA 0,200 EXA 1 – 9
Source: own
Table 4: Students Platform KPA: practices.
Acronym Practice statement Rating
SPL1 Students’ platform availability 0,250
SPL2 Course schedule available online 0,214
SPL3
Schedule of exams and other
evaluation works available online
0,143
SPL4 Grades available online 0,179
SPL5 Course materials availability online 0,036
SPL6 Registration forms availability online 0,107
SPL7
Students’ platform in a mobile
a
pp
lication version
0,071
Source: own
Table 5: Courses and Fields of Study KPA: practices.
Acronym Practice statement Rating
CFS1 Course grades transfer 0,125
CFS2 Changing the field of study 0,117
CFS3 Double diploma programs 0,092
CFS4 Exchange programs courses 0,075
CFS5 Course transfer in exchange programs 0,100
CFS6 Individual study plan 0,033
CFS7 E-learning 2.0 (informal) 0,058
CFS8 E-learning organization 0,067
CFS9 Evaluation works 0,083
CFS10 Course teachers 0,042
CFS11 Elective courses content 0,017
CFS12 Elective courses number 0,008
CFS13 Study plan content 0,025
CFS14 Students’ opinions 0,050
CFS15 Course schedule flexibility 0,108
Source: own
Maturity Model for Assessment of Personalization of Higher Education
47
KPA2 “Courses and Fields of Study” (CFS),
contains 15 practices (Table 5), being the largest key
process area in the model. This is the KPA dedicated
to the issues of selecting courses, transferring
between specializations and fields of study or
between institutions, expressing preferences and
opinions on academic teachers and their teaching
methods, learning online, and others.
KPA3Research Activity (RSA), has only four
practices in it (Table 6). These practices consider
theses of bachelor’s and master's levels, students’
participation in scientific conferences, membership in
scientific organizations, and the overall research
activity of students.
Table 6: Research Activity KPA: practices.
Acronym Practice statement Rating
RSA1 Scientific tutorship 0,200
RSA2 Students’ scientific organizations
4
0,100
RSA3 Scientific conferences 0,300
RSA4 Bachelor’s / master’s thesis 0,400
Source: own
The last key process area (KPA4)
“Extracurricular Activities” (EXA), includes eight
practices (Table 7). It covers, among others: students’
personal development, engagement in various
activities and events, providing students with access
to the internet and information databases, and
integration of students from exchange programs.
As stated above, all practices contain description
of cases for five maturity levels (0-4). Since it is not
possible to fit the tables with cases for all practices
into the paper, the authors chose to provide examples
of case descriptions, one per each key process area.
Table 7: Extracurricular Activities KPA: practices.
Acronym Practice statement Rating
EXA1 Personal development 0,222
EXA2 Students' organizations (non-scientific) 0,056
EXA3 Exchange students’ engagement 0,083
EXA4 Infrastructure 0,194
EXA5
Access to databases and electronic
resources
0,167
EXA6 Students’ decision-making 0,139
EXA7 Practices and internships 0,111
EXA8 Volunteering 0,028
Source: own
4
Scientific circles, research groups, research seminars,
research laboratories, etc.
Practice SPL1 (Students’ platform availability),
level 0 (Not assessed): “There is no platform or
website with information for students at the HEI”.
Practice CFS9 (Evaluation works), level 1
(Initial): The HEI sets crediting formats for all
courses, and they cannot be changed on students'
request. The information is given in the official course
description”.
Practice RSA1 (Scientific tutorship), level 2
(Repeatable): Students can apply for additional
scientific tuition only when they begin working on
bachelor's or master's thesis. Students' grades are not
taken into consideration. The HEI appoints the tutor”.
And finally, practice EXA5 (Access to databases
and electronic resources), level 4 (Optimizing): The
HEI provides students with access to most (or all) of
the largest databases or other electronic resources.
Access is also possible from students’ private
computers (e.g., using VPN connection or a special
account)”.
Procedure of verification of the EPMM model
with all the practices and cases, which was conducted
at 51 higher education institutions, is discussed
further.
4 EDUCATION
PERSONALIZATION
MATURITY MODEL
VERIFICATION
Verification of the Education Personalization
Maturity Model, developed by the authors, was
conducted to assess the maturity of education
personalization at higher education institutions in
Poland and abroad. This assessment was conducted
with the help of a questionnaire that was distributed
among colleagues from different HEIs worldwide.
The sampling for this survey is a non-random
convenience sampling since the respondents were
selected based on their experience, their places of
work, as well as on the convenience of reaching them
out, and their willingness to participate in the survey.
As a result of the survey, the authors obtained 51
responses, i.e., 51 higher education institutions were
assessed using the Education Personalization
Maturity Model (I = 51). These 51 institutions belong
to 25 countries (C = 25) in Europe (68%), Asia (16%),
South America (12%), and Africa (4%). The authors
KMIS 2022 - 14th International Conference on Knowledge Management and Information Systems
48
have applied Alpha-3 codes for countries, and further,
these codes are used to encode educational
institutions from particular countries.
Table 8 provides information about the positions
the respondents occupy at their institutions. The
question about the position was a multiple-choice
one, so there was a possibility for a respondent to
claim to be, for instance, both an academic teacher
and a member of university authorities.
Table 8: Distribution of respondents’ positions at their HEIs
(I = 51).
Positions
Number
of
answers
% of total
number
of HEIs
Academic Teacher 33 64,71%
Academic Teacher, Research
Worke
r
9 17,65%
Academic Teacher, University
Authorities Membe
r
3 5,88%
Academic Teacher, Administrative
Staff Membe
r
2 3,92%
Administrative Staff Member 2 3,92%
University Authorities Member 2 3,92%
Total number of answers = number
of HEIs
51
4.1 Higher Education Institutions
Assessment Results
To assess the maturity of individualization of a HEI,
the respondents had to select one of the five cases
(referring to five maturity levels) for each of the
practices within each of the key process areas.
All the answers from the Google Forms
questionnaire were gathered in Google Sheets, where
the authors then manually connected all the answer
options with the corresponding number of maturity
level (from 0 to 4). Thus, the personalization maturity
levels for each practice within the four KPAs
appeared. In the next step, maturity levels were
calculated for each key process area by calculating
the weighted average with the help of the ratings of
practices (presented in Tables 4-7). Further, the
ratings of the KPAs (presented in Table 3) were
applied in the weighted average to calculate the final
personalization maturity levels for the 51 HEIs. The
results of the described calculations are presented in
Table 10. The aggregated statistics for 51 higher
education institutions are given in Table 9.
5
https://www.coursera.org
6
https://www.udemy.com
Table 9: Personalization maturity assessment: aggregated
results (I=51).
Level of
personalization
maturit
y
Number of HEIs
% of total
number
of HEIs
3 27 52,94%
2 18 35,29%
4 3 5,88%
1 3 5,88%
0 0 0,00%
Source: own
These final results of the personalization maturity
assessment allow the authors to distinguish some
HEIs that had developed personalization for their
students and had put it on a high level, and, on the
contrary that are still at the beginning of
personalization development and have a lot of good
practices to introduce.
Of the 51 HEIs assessed, most (27; 52,94%)
obtained the level of maturity 3 “Defined and
managed”, which means that they had developed
most of the personalization practices considered in
the EPMM, but still have options for improvement.
The fourth place is taken by three HEIs with level 1
“Initial / ad hoc” (5,88%). These institutions are
characterized as those having only a few
personalization practices realized, without any
common system, and supported mainly by individual
efforts of academic and administrative staff.
Within the “Students’ Platform” KPA, most
institutions (24; 47,06%) have the same high level of
personalization (level 4 or close to it) related to the
platform for students. For all institutions, the weakest
point seems to be the mobile application with all
necessary information for students, with frequent
updates and notifications. One more issue, which
appears to be insufficiently developed, is the schedule
of courses and exams available and updated online
(when students do not have to download, for instance,
a PDF file from a website and compare it with
previous versions to reveal changes).
Levels of practices development within the
“Courses and Fields of Study” KPA vary rather
significantly. Most common problems in this KPA,
for all institutions, are: 1) informal e-learning taking
online courses on platforms like Coursera
5
, Udemy
6
,
Edx
7
, Mooc
8
etc., is not forbidden by HEIs, but it is
not (or poorly) motivated, supported, and rewarded;
2) internal e-learning HEIs either do not offer any
courses provided online, or have very few of them,
probably with low possibility to replace traditional
7
https://www.edx.org
8
https://www.mooc.org
Maturity Model for Assessment of Personalization of Higher Education
49
courses with their online version; 3) choosing
teachers students do not have any influence on the
process of assigning teachers to courses, or, probably,
they can choose teachers for very few courses (like
courses that are additionally selected); 4) number of
elective courses from 0% to only 50% of courses
students are offered within their study programs can
be selected by students themselves on the basis of
their preferences, and the list to choose from is quite
small; the rest are set by HEIs authorities; 5) course
schedule flexibility student have zero to low
influence of the schedule of courses they take; they
cannot apply for changes to be able to combine
education with work or other activities efficiently.
The most well-developed practices in this KPA are:
1) exchange programs as host institutions in
programs like Erasmus, HEIs offer a lot of courses
and do not forbid to take more courses than it is set
by the program if students are interested in getting
more knowledge and skills; 2) variety of elective
courses such courses can belong to different
specialization and fields of study at the selected HEI,
they can be taught in foreign languages, and their
number is not limited by the HEI; 3) consideration of
students’ opinions courses evaluation is conducted
every semester, students’ opinions about teachers,
content, learning methods, etc., are gathered;
information is given to Heads of Departments and to
teachers to conduct improvements. Polish HEIs
mostly take levels between 2 and 3 in this KPA.
As for theResearch Activity KPA, 49,02% of
assessed HEIs take level 3 (“Defined and managed”)
for the development of practices connected with
supporting their students’ research activity. For
50,98% of the educational institutions, the practice of
scientific tutorship for students is not developed
(levels 0 and 1). This means that either there is no
tutorship at all, or tutors can be appointed only at the
master’s degree, only to students with a high average
grade, and, perhaps, only by HEI authorities with no
option for students to make their own choice. One
more weak point in the personalization of research
activity is the practice of running scientific
organizations. 19,61% of institutions have this
practice at level 0 (“Not assessed”). These HEIs do
not run any students’ scientific organization (e.g.,
scientific circles, laboratories, etc.). Two other
practices remaining in this KPA are well-developed
at 68,63% of educational institutions and poorly
developed at 31,37%. Therefore, 68,63% of HEIs
regularly organize scientific conferences for their
students to participate in, with many urgent topics
covered; HEIs may finance the participation of their
students in conferences organized by other
Table 10: Personalization maturity assessment results.
HEI
KPA1 -
SPL
KPA2 -
CFS
KPA3 -
RSA
KPA4 -
EXA
Level of the
HEI
ALB-1 3 2 2 3 3
ALB-2 3 1 2 2 2
BRA-1 3 2 4 4 3
BGR-1 1 2 2 3 2
CZE-1 4 2 3 4 2
DEU-1 3 3 3 3 3
EGY-1 3 1 2 3 3
FRA-1 4 2 0 2 2
GRC-1 3 2 3 3 3
GRC-2 3 1 1 2 3
HUN-1 3 2 3 3 2
HUN-2 3 2 2 3 3
IRL-1 3 2 1 4 3
IRL-2 3 2 2 3 3
ITA-1 3 2 0 2 3
KAZ-1 3 2 3 3 2
OMN-1 4 3 2 3 3
OMN-2 4 1 1 2 3
PRY-1 2 1 1 2 1
POL-1 3 1 3 3 2
POL-2 4 3 4 3 2
POL-3 2 2 3 3 2
POL-4 4 3 3 3 3
POL-5 4 2 2 3 3
POL-6 1 1 1 1 1
POL-7 2 1 3 2 2
POL-8 3 2 3 3 3
POL-9 4 3 4 4 4
POL-10 4 3 3 4 4
POL-11 4 3 3 4 4
POL-12 2 2 3 4 2
POL-13 2 2 2 2 2
POL-14 3 3 4 4 3
PRT-1 3 2 1 3 2
PRT-2 4 2 4 3 3
ROU-1 4 3 3 3 3
ROU-2 4 3 4 3 3
RUS-1
3 3 3 3 3
RUS-2
3 3 3 3 3
SV
K
-1
2 1 1 1 1
SVN-1
3 2 2 1 2
ESP-1
3 2 1 3 2
ESP-2
4 1 2 3 3
ESP-3
3 1 3 3 2
TUR-1
3 3 2 3 3
UKR-1
3 3 3 2 3
UKR-2 3 3 3 3 3
UKR-3
2 2 3 2 2
UKR-4
3 3 3 3 3
Source: own
KMIS 2022 - 14th International Conference on Knowledge Management and Information Systems
50
institutions and may reward participation with
extra grades for some courses. Also, at 68,63% of
higher education institutions, students have the
freedom of selecting thesis topics themselves (it is not
forced by the authorities), and the scientific area of
these topics is not limited; thesis can be supervised by
business representatives to make the research more
related to practice.
Finally, analyzing the “Extracurricular Activities”
KPA, it can be seen that 70,51% of the assessed
institutions appear to take levels 3 (“Defined and
managed”) and 4 (“Optimizing”) of personalization
maturity. One of the poorly developed practices here
is “Volunteering”. Following the EPMM cases, such
a level means that HEIs do not engage students in any
volunteer programs, nor do they motivate and reward
participation; they only provide students with
information about existing volunteer programs and
may be engaged in some programs. In turn, the
strongest practice for 70,51% of HEIs is “Personal
development”, which means that these institutions
regularly invite speakers from business units to
conduct classes (workshops, lectures, etc.) for their
students, also engaging students in the search of the
most interesting speakers; they support students in
problems of personal development (tutorship,
mentorship), and frequently inform them about the
most attractive career options offered in the region (or
whole country).
5 DISCUSSION
Nowadays, it is crucial to provide students with the
proper approach toward their preferences and
aspirations for education and personal and
professional development to make them feel content
with their experience at higher education institutions.
Therefore, it is necessary for educational institutions
1) to become aware of whether they give students
enough freedom and flexibility for the realization of
their plans and ambitions; 2) to be able to compare
their personalization policy with other educational
institutions; and 3) to learn about ways of making
their personalization policy more effective.
This paper presents the Education Personalization
Maturity Model developed to assist HEIs in the
assessment of the level of their personalized approach
toward their students.
5.1 Contribution of the Research
The results provided by the Education
Personalization Maturity Model are useful, first of all,
for the higher educational institutions that were
engaged in the survey since the conclusions obtained
allow to pay attention to the weak places of the
process of education personalization development.
Moreover, the authors consider that the descriptions
of cases presented for each practice within each key
process area of the EPMM can serve as kinds of
prompts or small guidelines on what should be
changed or what options should be added to provide
students with a higher level of personalized approach
towards their Individual Profile formation.
Additionally, the authors expect the EPMM to be of
interest to administrative and management staff of
higher educational institutions in countries of Europe
and beyond; the example of 51 HEIs that already took
part in the assessment of personalization should serve
as proof that the EPMM can effectively assist in
personalization assessment, at least at the initial level.
5.2 Limitations of the Research
One of the limitations of the developed EPMM is that
it may miss some practices that might be considered
important for students of HEIs. As stated, the Model
was developed on the basis of students’ opinions, and
the questionnaire presented to them was quite
extensive. Yet, there is a chance that with more
opinions, some new practices would appear. The
authors also believe that a survey conducted among
academic teachers may have given interesting results.
The other limitation of the EPMM consists in the
necessity of finding experts to use it. An employee of
a HEI, who is going to use the Model to assess
personalization at that particular institution, should
possess knowledge about various activities taking
place there: didactic process and research, sports and
other activities, mobility programs and cooperation,
etc.
The limitation of verification of the EPMM,
presented in this paper, is that only one expert from
each HEI used the EPMM. For a better, complete
picture of each institution, at least a few opinions
about each HEI would be necessary.
5.3 Avenues for Future Research
As mentioned earlier, the EPMM is supposed to have
a continuous process of evolution; it can be modified
by the authors or by other users (like the
administrative staff of a HEI) that would like to apply
the Model. Amendments to the EPMM can be
conducted to adjust it to the specificity of a particular
HEI or to the education policy of a certain country.
The authors also believe that the changes that might
Maturity Model for Assessment of Personalization of Higher Education
51
happen to the EPMM will be caused by the natural
changes in higher education connected with the
flow of time, with the progress of IT, with higher
demands and greater ambitions of students, with the
constants self-development of teachers and
improvement of their teaching techniques, etc.
Therefore, the authors distinguish two directions
for future work. The first one would be the
modification and improvement of the Education
Personalization Maturity Model. The second
direction would be the development of guidelines for
personalization maturity improvement. For now, the
only piece of advice that can be obtained from the
EPMM can be found in the descriptions of particular
practices. Descriptions of the levels higher than the
one defined for the analyzed higher education
institution can serve as small prompts on what
measures to take to improve the level of
personalization and make students of this institution
more content. Thus, the extended guidelines on how
to improve personalization by performing changes in
particular KPAs of institutions’ activity or in
particular practices that they perform would be a
valuable potential contribution to higher education
management.
5.4 Final Remarks
Results of analysis of the selected higher education
institutions with the help of the Education
Personalization Maturity Model, developed by the
authors, lead to a conclusion that HEIs, represented
by their administrative and management staff, may
benefit from the application of the EPMM.
Implementation of the Model enables assessment of
the level (i.e., degree) of the personalized approach
that higher education institutions provide for their
students. It also provides suggestions on possible
ways of improving the current situation with the
personalization of education.
REFERENCES
Aliyu, A., Maglaras, L., He, Y., Yevseyeva, I., Boiten, E.,
Cook, A., & Janicke, H. (2020). A holistic
cybersecurity maturity assessment framework for
higher education institutions in the United Kingdom.
Applied Sciences (Switzerland), 10(10). https://doi.
org/10.3390/app10103660
Anicic, K. P., & Divjak, B. (2020). Maturity Model for
Supporting Graduates’ Early Careers Within Higher
Education Institutions. SAGE OPEN, 10(1). https://doi.
org/10.1177/2158244019898733
Boughzala, I., & de Vreede, G.-J. (2015). Evaluating Team
Collaboration Quality: The Development and Field
Application of a Collaboration Maturity Model.
Journal of Management Information Systems, 32(3),
129–157. https://doi.org/10.1080/07421222.2015.109
5042
Carvalho, J. V., Pereira, R. H., & Rocha, A. (2019).
Development Methodology of a Higher Education
Institutions Maturity Model. In Xhafa, F and Barolli, L
and Gregus, M (Ed.), Advances in Intelligent Networking
and Collaborative Systems (Vol.23, pp. 262–272).
https://doi.org/10.1007/978-3-319-98557-2\_24
de Bruin, T., Rosemann, M., Freeze, R., & Kulkarni, U.
(2005). Understanding the main phases of developing a
maturity assessment model. ACIS 2005 Proceedings -
16th Australasian Conference on Information Systems,
8–19.
Dean, A., & Gibbs, P. (2015). Student satisfaction or
happiness? A preliminary rethink of what is important
in the student experience. Quality Assurance in
Education, 23(1), 5–19. https://doi.org/https:/
/doi.org/10.1108/QAE-10-2013-0044
Durek, V., Kadoic, N., & Begicevic Redep, N. (2018).
Assessing the digital maturity level of higher education
institutions. In 2018 41st International Convention on
Information and Communication Technology,
Electronics and Microelectronics, MIPRO 2018 -
Proceedings (pp. 671–676).
Enke, J., Glass, R., & Metternich, J. (2017). Introducing a
Maturity Model for Learning Factories. Procedia
Manufacturing, 9, 1–8. https://doi.org/10.1016/
j.promfg.2017.04.010
Grundspenkis, J. (2010). MIPITS and IKAS--Two Steps
towards Truly Intelligent Tutoring System Based on
Integration of Knowledge Management and Multiagent
Techniques. International Conference on E-Learning
and the Knowledge Society (e-Learning 2010), Riga,
Latvia, August, 26–27.
Hong, Y., & Xinyi, Z. (2019). Mapping algorithm design
and maturity model construction of online learning
process goals. International Journal of Emerging
Technologies in Learning, 14(4), 31–43.
Kabak, M., & Dagdeviren, M. (2014). A hybrid MCDM
approach to assess the sustainability of students’
preferences for university selection. Technological and
Economic Development of Economy, 20(3), 391–418.
https://doi.org/10.3846/20294913.2014.883340
Marshall, S. (2010). A Quality Framework for Continuous
Improvement of E-learning: The E-learning Maturity
Model. Journal of Distance Education Revue De
L’Éducation À Distance, 24(1), 143–166.
Marshall, S., & Mitchell, G. (2004). Applying SPICE to e-
Learning: An e-Learning maturity model? Sixth
Australasian Computing Education Conference
(ACE2004), 30(Ims 2003), 185–191.
Matkovic, P., Pavlicevic, V., & Tumbas, P. (2017).
Assessment of Business Process Maturity in Higher
Education. INTED2017 Proceedings, 1(March), 6891–
6898. https://doi.org/10.21125/inted.2017.1600
KMIS 2022 - 14th International Conference on Knowledge Management and Information Systems
52
Mettler, T. (2011). Thinking in Terms of Design Decisions
When Developing Maturity Models. International
Journal of Strategic Decision Sciences, 1(4), 76–87.
https://doi.org/10.4018/jsds.2010100105
Nelson, K., Clarke, J., & Stoodley, I. (2014). An
exploration of the Maturity Model concept as a vehicle
for higher education institutions to assess their
capability to address student engagement. A work in
progress. Ergo, 3(1), 29–36.
Nsamba, A. (2019). Maturity levels of student Support E-
services within an open distance E-learning University.
International Review of Research in Open and Distance
Learning, 20(4), 61–78.
Orîndaru, A. (2015). Changing Perspectives on Students in
Higher Education. Procedia Economics and Finance.
https://doi.org/10.1016/s2212-5671(15)01049-7
Penafiel, M., Lujan-Mora, S., Stefanie Vasquez, M.,
Zaldumbide, J., Cevallos, A., & Vasquez, D. (2017).
Application of E-learning Maturity Model in Higher
Education. In I. Chova, LG and Martinez, AL and
Torres (Ed.), 9th International Conference on
Education and New Learning Technologies
(EDULEARN17) (pp. 4396–4404). Lauri Volpi 6,
Valenica, Burjassot 46100, Spain: IATED-INT Assoc
Technology Education & Development.
Quaye, S. J., & Harper, S. R. (Eds.). (2014). Student
Engagement in Higher Education: Theoretical
Perspectives and Practical Approaches for Diverse
Populations (2nd ed.). Routledge. https://doi.org/
10.4324/9780203810163
Reçi, E., & Bollin, A. (2017). Managing the Quality of
Teaching in Computer Science Education. In
Proceedings of the 6th Computer Science Education
Research Conference (pp. 38–47). New York, NY,
USA: Association for Computing Machinery.
https://doi.org/10.1145/3162087.3162097
Reçi, E., & Bollin, A. (2019). A Teaching Process Oriented
Model for Quality Assurance in Education - Usability
and Acceptability. IFIP Advances in Information and
Communication Technology, 524, 128–137. https://
doi.org/10.1007/978-3-030-23513-0_13
Rizun, M. (2021). Assessing the Personalization of Higher
Education: Maturity Framework Development. In B. Z.
Janusz Nesterak (Ed.), Knowledge Economy Society.
Business Development in Digital Economy and
COVID-19 Crisis (pp. 195–206). Institute of
Economics Polish Academy of Science.
Rollande, R. (2015). Research and Implementation of
Personalized Study Planning as a Component of
Pedagogical Module. Doctoral Thesis. Riga: RTU.
Rollande, R., & Grundspenkis, J. (2016). Personalized
Planning of Study Course Structure Using Concept
Maps and Their Analysis. Procedia Computer Science.
https://doi.org/10.1016/j.procs.2017.01.093
Secundo, G., Elena-Perez, S., Martinaitis, Ž., & Leitner, K.
H. (2015). An intellectual capital maturity model
(ICMM) to improve strategic management in European
universities: A dynamic approach. Journal of
Intellectual Capital, 16(2), 419–442. https://doi.org/
10.1108/JIC-06-2014-0072
Thong, C. L., Jusoh, Y. Y., Abdullah, R., & Alwi, N. H.
(2013). Application of curriculum design maturity
model at private institution of higher learning in
Malaysia: A case study. Lecture Notes in Electrical
Engineering, 229 LNEE, 579–590.
Thong, C. L., Yusmadi, Y. J., Rusli, A., & Nor Hayati, A.
(2012). Applying capability maturity model to
curriculum design: A case study at private institution of
higher learning in Malaysia. In Lecture Notes in
Engineering and Computer Science (Vol. 2198, pp.
1070–1075).
Zhu, Y. (2016). Research on Personalized Education in
Chinese Universities. 2016 2nd International
Conference on Social Science and Development
(ICSSD 2016), 138–144.
Maturity Model for Assessment of Personalization of Higher Education
53