A Proposal for an Educational Well-Being Index (EWI) for

Undergraduate Course Design*

Maria J. Blesa

, Amalia Duch

, Joaquim Gabarr

and Maria Serna

Departament de Ci

encies de la Computaci

o, Universitat Polit

ecnica de Catalunya, BarcelonaTech, Spain

Keywords:

Social Progress Index, Well-Being, Educational Model, Educational Well-Being Index.

Abstract:

Every day it is more common to hear around us about the publication of studies, surveys or statistical results

about the well-being of people, workers, women in a given country. Indeed, as university professors, our work

cannot be independent of the level of well-being of our students. So, in this work, we propose a methodology to

asses the students well-being inside a course implementation by what we call the educational well-being index

(EWI). We start with a survey that gathers those factors that computing courses’ students at our university –of

two different levels and majors– consider most important. Our second step is the evaluation –by a group of

teachers– of the presence of those factors in different educational models of implementation of the courses.

We use principal component analysis to extract, from the student data, the valuations that they expressed

in the survey: the principal component of their own measurements on well-being. We work only with the

coefﬁcients of the ﬁrst dimension of the principal component. The third step is a (subjective) valuation of the

topics addressed in the survey when considering a particular educational model. Finally, we gather everything

together to obtain a well-being index of an educational model that allows their comparison. Besides the

methodology, we present and analyze the values obtained from our case study.

1 INTRODUCTION

Traditionally, the gross domestic product (GDP) of

a country has been calculated and studied as a mea-

sure of its economic “health” and of its “growth”

rate. Moreover, the economic development of a coun-

try has been an indicator of the “progress” of this

country and it is usually measured by the GDP. The

GDP measures, in money, the goods and services pro-

duced in a year and in order to compare different

countries among them, we usually use the GDP per

capita (Wikipedia, 2024) which has been traditionally

a measure of country’s development.

However, in recent times, this way of measur-

ing the “growth” or “hegemony” of countries and

institutions has been severely criticized. The mea-

sure does not take into account people individually

but just count them collectively, as just another num-

ber. In consequence, the GDP, as a measure of well-

https://orcid.org/0000-0001-8246-9926

https://orcid.org/0000-0003-4371-1286

https://orcid.org/0000-0003-3771-2813

https://orcid.org/0000-0001-9729-8648

∗

Supported by MCIN/AEI/10.13039/501100011033

under grant PID2020-112581GB-C21 (MOTION).

being, has been strongly criticized during the last

years (Skidelsky and Skidelsky, 2013) and, for in-

stance, the term growth has been replaced in some

economy schools by postgrowth or degrowth (Jack-

son, 2021; Paulson et al., 2020). In order to avoid

problems with the term growth, the current tenden-

cies replace it by an assessment of progress. Al-

though the study of Welfare economics is not new,

it dates back to the 1920s with Pigou’s famous book

(Pigou, 2013), we can say that it has now become

not only fashionable but mandatory to measure the

well-being in countries and institutions. One of such

measures is the Social Progress Index (SPI), that pro-

vides a better measure to look at if one wants to com-

pare the progress of countries and/or institutions in

terms of their welfare (Social Progress Org, 2024; The

Economist, 2023) .

Universities have not been an exception where dif-

ferent general indicators are calculated as measures of

the success or failure of the university education pro-

cess. For instance, if we are interested in how a uni-

versity is seen in the world, we can look at one of the

many available rankings, as for example the ranking

of the universities (July 2023) (Webometrics, 2023)

(see Table 1).

626

Blesa, M., Duch, A., Gabarró, J. and Serna, M.

A Proposal for an Educational Well-Being Index (EWI) for Undergraduate Course Design.

DOI: 10.5220/0012737500003693

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 16th International Conference on Computer Supported Education (CSEDU 2024) - Volume 2, pages 626-633

ISBN: 978-989-758-697-2; ISSN: 2184-5026

Table 1: Rank of some institutions in the European Ranking (Webometrics, 2023).

European rank World rank Institution

115 314 University Duisburg Essen

116 320 University Wurzburg

117 322 Universitat Polit

ecnica de Catalunya

118 322 Technion Israel Institute of Technology

118 326 London School of Hygiene & Tropical Med.

The recent COVID pandemic has made clear that

it is necessary to look beyond the classic global in-

dicators –especially as a result of the multiple men-

tal health problems it has generated in society. It is

needed to look closely at the people individually and

in particular pay attention to their well-being. Many

studies of well-being at the university or campus level

have appeared in the last years; see, e.g., (Dodd et al.,

2021; Wang et al., 2022; Kanonire et al., 2022).

At a lower level, the success of a course is usu-

ally measured by general indicators such as the pass

rate although there are discussions on the teaching

methodology that a traditional face-to-face institu-

tion should adopt in the future and we believe that

such evaluations need to take into account a mea-

sure of well-being. In this work, we propose the

Educational well-being index (EWI), inspired in the

SPI (Social Progress Org, 2024), to assess the well-

being of the design of a course at university level. We

exemplify our proposal with a case study made on a

relatively small sample of students and several educa-

tional models.

To do so, we prepared a well-being question-

naire formed by 20 well-being related questions to

be ranked between 1 and 5. We poll the students

of two different courses at the Universitat Polit

ecnica

de Catalunya (UPC): Programming I (ﬁrst year Math

studies) taught at the Facultat de Matem

atiques

(FME) and Algorithms (third year CS studies) at the

Facultat d’Inform

atica de Barcelona (FIB)to deter-

mine the elements that conform –according to them–

their well-being at university courses. We perform a

Principal component analysis (PCA) on the gathered

data to extract the ﬁrst principal component (as it is

done to compute the SPI) which provides a quantiﬁ-

cation of the value of each question.

Aside, we consider the three dimensions of teach-

ing that we believe have impact on well-being: lec-

turing, transportation, and school models. The ﬁrst

dimension considers four different models to teach

the above-mentioned courses from totally in person

to as much online as possible in our university. Let

us observe that we focus on a university that offers

only synchronous education, meaning that in class

hours students and lecturers coincide either on-line

or in person. The transportation dimension consid-

ers the type of mobility needed to reach the university

campus, for example, cheap/expensive or long/short.

The school dimension takes into account the organi-

zational model which has particular trends in different

schools. Each combination of one model from each

dimension provides a description of a potential edu-

cational model.

For each educational model, we evaluate the

amount of the well-being components that conforms

it by grading –following our own opinion– the ques-

tions relative to the model in the survey. Finally, we

use the results of the PCA to provide a comparable

rank value. We name this value as EWI as an equiv-

alent of the SPI for courses. We study the trends and

components of our proposed index analyzing the ten-

dencies and singularities of our case study.

The paper is organized as follows: in Section 2 we

give the required preliminaries on SPI. In Section 3

we present the design of the student’s poll. Then, in

Section 4, we present the different course implemen-

tation models and our evaluation of them with respect

to the well-being elements. In Section 5, we provide

the EWI ranking of the course implementations, ac-

cording to our data, and analyze the tendencies by

component. Finally, we provide some conclusions

and lines of future work in Section 6.

2 WELL-BEING: THE SOCIAL

PROGRESS INDEX

The social progress index (SPI) is informally deﬁned

in (Social Progress Org, 2024) as

The capacity of a society to meet the basic hu-

man needs of its citizens, establish the build-

ing blocks that allow citizens and communi-

ties to enhance and sustain the quality of their

lives, and create the conditions for all individ-

uals to reach their full potential.

The SPI considers only non-economic aspects of the

countries. The indicators (a total of 60) are organized

in three dimensions: 1) Basic Human Needs, 2) Foun-

dations of Well-being, and 3) Opportunities. Each

dimension contains four components. Our interest is

in the second dimension, that considers the following

components:

• Access to Basic Knowledge

A Proposal for an Educational Well-Being Index (EWI) for Undergraduate Course Design

627

Table 2: Countries and rankings; four indicators: (1) Social Progress Index (SPI) in 2022; (2) Gross Domestic Product (GDP)

according to FMI 2013; (3) CO

emissions in 2021; and (4) Most highly ranked university in the country in 2023.

Country SPI Rank GDP (k$) Rank CO

t per cap. University Rank

Sweden 89.42 6 58.014 7 3.82 Lund Universtiy 123

Germany 88.72 8 44.999 18 8.06 TU M

unchen 86

UK 86.13 19 39.372 23 4.95 Oxford University 4

France 86.04 20 44.099 20 4.58 Sorbonne University 227

Spain 85.35 21 29.150 27 4.99 Universitat de Barcelona 108

Italy 85.23 22 34.715 26 5.41 Universit

a di Bologna 105

Greece 82.44 33 21.857 35 4.82 Patras University 571

Canada 88.17 10 52.037 10 14.86 Toronto University 16

USA 84.65 25 53.001 9 14.24 Harvard University 1

Israel 83.17 31 36.926 25 6.74 Tel Aviv University 115

Chile 80.78 36 15.776 45 4.61 Universidad de Chile 300

Argentina 78.64 41 14.709 50 4.12 Universidad de Buenos Aires 382

Mexico 70.84 66 10.650 64 3.09 Universidad Nacional Aut

onoma 108

• Accent to Information & Communication

• Health & Wellness

• Environmental Quality

It seems that, at least in developed countries wealth

and well-being are strongly correlated but not always

coincide as we can see in the SPI. Concerning En-

vironmental Quality, the main interest is on climate

change due to CO

emissions which is, as for today,

an unavoidable problem (Nordhaus, 2013).

In order to quantify some speciﬁc trends on the

different approaches, in Table 2, we give the SPI,

GDP, CO

emissions and the best University rank for

some countries. The different values of the parame-

ters show an up-and-down across countries and num-

bers. Some countries demonstrate strengths in certain

topics while displaying weaknesses in others.

For instance, if we compare Spain and EEUU,

we can observe that Harvard is the best world and

EEUU university, while the best Spanish university,

Barcelona U., is ranked 26. The GDP per capita in

EEUU (53k$) is much larger than in Spain (29k$).

However, in some aspects, Spain seems to be bet-

ter than EEUU; Spain is slightly better ranked in SPI

than EEUU. The difference becomes greather when

we look at CO

, 4.99t in Spain and 14.24t EEUU.

We could summarize (perhaps simplifying too much)

saying that people in EEUU are more rich and have

better universities than in Spain, but people in Spain

have a better well-being and pollute less.

3 QUESTIONNAIRE DESIGN

The ﬁrst thing that we did was to identify which of

the topics listed in the SPI report were applicable to

university education on computing (and to university

education in general). Once these topics were identi-

ﬁed, we created a list of 20 statements. For each of

them we add a question of the form: From 1 to 5, how

important is the topic T for your well-being in the sub-

ject, school, university, etc. with the idea of surveying

the largest number of students possible.

As we have already mentioned, we concentrate in

three dimensions:

• Lecturing

• Transportation

• School

The ﬁrst dimension consider question related to the

ways of teaching. In particular the relevance from

the well-being perspective of having face-to-face or

on-line interactions. The second dimension addresses

one natural complaints, especially when reaching the

university requires some time, which has a clear im-

pact in well-being. On the other hand, all of us have

a clear idea that public transportation is better than

private but here we are using a point of view of CO

production. Our questionnaire focus on the comfort

of traveling and not on the division public/private mo-

bility. The third dimension, consider the institution

that is organizing the teaching. In our context, the

school is the agent that controls the use of resources.

The school decides the time tables, the class rooms,

the placement of the exams, the sizes of the groups,

etc.

4 EDUCATIONAL MODELS

As we already mention we focus only in what we

call “synchronous education”: at class time stu-

dents and lecturers are both present either in-person

CSEDU 2024 - 16th International Conference on Computer Supported Education

628

or in streaming. In order to identify which educa-

tional models ﬁt better with those well-being topics

that our students consider more important, we con-

sidered three dimensions:

• Type of lecturing

• Transportation

• Resources and organization

4.1 Type of Lecturing Classes

We consider four different ways of teaching the

classes of the two courses on which we based our

study. These follow the different combination of face-

to-face and on-line interaction that we experimented

during the pandemics. Each type is set by ﬁxing the

percentage of in-person requirements for each of the

following parts of the course: a) theory classes, b)

problem classes, c) laboratory classes, d) video taped

classes and e) assessment. Table 3 shows the percent-

age of in-person attendance of each of the four ways

of lecturing.

4.2 Transportation

One of the factors inﬂuencing well-being is the kind

of journeys needed to reach a university campus. In

our university, we were able to identify three kind of

campuses: a) campuses in which the duration of the

average journey is less than 30 minutes, b) campuses

in which the duration of the average journey is greater

than an hour and a half and c) campuses in which the

average journey duration is in between the other two.

For the ﬁrst kind of campus, we observed that the

students transportation is mainly by means of public

transportation. It is questionable but we considered –

arbitrarily– that public transportation is less comfort-

able than private one but cheaper. Therefore, we have

considered that the journeys on public transportation

only were pretty uncomfortable (assigning a 1 value

in the corresponding question) and inexpensive (as-

signing a 5 value in the corresponding question). As

distances are higher, the use of public transportation

decrease, so for the second one we assume the per-

centage of public transportation to be around the 30%.

Thus, we assigned values of 5 in comfort (completely

comfortable journeys) and very expensive (assigning

a 1 value in the corresponding question). For the third,

it is likely to be 50% public transportation and the rest

private. Consequently all the assigned values were in

the middle (assigning them a value 3). The reason

to these choices was to include in the kinds of trans-

portation all the possible contrast of values.

Table 3: Percentage of in-person attendance according to

the type of classes.

Type L

Theory 100% 0% 0% 100%

Problems 100% 0% 0% 50%

Laboratory 100% 0% 100% 0%

Videos 0% 100% 50% 50%

Assessment 100% 50% 75% 75%

Table 4: Percentage of use of public transportation in stu-

dents journeys to university campuses.

Type Campus Campus Campus

Public 100% 30% 50%

Private 0% 70% 50%

4.3 School Types

The FIB offers a Computer Science degree with four

different specialties. Each generation consists of ap-

proximately 400 students who study subjects from a

common two-year core and then choose a specialty

from among four possible ones: computing, software

engineering, information systems and hardware. The

theory classes are divided into groups of 60 students

while the laboratories are conformed by 20 students.

There are lectures in the morning and afternoon for all

subjects and an enrollment order is imposed in which

the students choose the lecture times that they want

according to their average grade. In general, the stu-

dents do not feel that they are part of a group –since

it can change semester to semester and from one sub-

ject to another, even from theory classes to labora-

tory classes. In summary, we consider this school as

a large one in which students do not group together

in a single group but are grouped into small groups

according to friendships and afﬁnities.

For its part, the FME offers the degree in Mathe-

matics without independent specializations. Each co-

hort consists of approximately 75 students. The the-

ory classes are divided into groups of 35-40 students

while the laboratories have a maximum of 30 stu-

dents. There are classes only in the morning for all

subjects and students are in the same group for both

theory classes and problem classes throughout their

entire degree. For this reason, a lot of cohesion is cre-

ated among all the students of this school. In contrast

to the FIB, the FME is a small school in which all the

students of the same generation (and even from dif-

ferent generations) work closely together and form a

kind of big family.

Both schools also have students’ associations, al-

though for the reasons we explained before, FME’s

students participate in them more actively than FIB’s

students. The same goes for the social and academic

A Proposal for an Educational Well-Being Index (EWI) for Undergraduate Course Design

629

Figure 1: Ranked EWI of the different models for the con-

sidered populations.

activities that both schools offer to their students.

In order to provide support to students, it is also

possible in both schools (for those students who re-

quest it) to have a tutor (a lecturer who can advise

them on academic issues) and a mentor (a student of

more advanced courses who guide them in the ﬁrst

years of studies). As expected, FME’s students tend to

ask for mentors more frequently than FIB’s students

and although they do not formally request them, the

cohesion between students in FME is so strong that

students from more advanced courses usually act as

mentors for newly arrived students.

4.4 Educational Models and Well-Being

Finally, we deﬁne educational models as 3-tuples se-

lecting the types of lecturing (4), transportation (3)

Table 5: Coefﬁcients of the 1st PCA component for the con-

sidered populations.

Question PCA1 coefﬁcients

CS + Math CS Math

Q1 -0.085275 0.105952 -0.419600

-0.009672 0.017823 0.209514

0.044217 0.079005 -0.049599

0.029739 -0.016453 0.194054

-0.068040 -0.200985 0.239976

-0.154504 -0.198831 0.096388

-0.196589 -0.159725 -0.260992

-0.174137 -0.170388 -0.276614

-0.084349 -0.030297 0.222551

-0.103875 -0.134508 0.056949

-0.159900 -0.115195 -0.021422

-0.313385 -0.349402 -0.157717

-0.409164 -0.379408 -0.281585

-0.314704 -0.371641 0.037266

-0.198096 -0.145104 -0.405415

-0.096386 -0.150898 0.122577

-0.147899 -0.135657 -0.023578

-0.403038 -0.367978 -0.196697

-0.322291 -0.283208 -0.301314

-0.384016 -0.364963 -0.245365

and school (2). This gives a total of 24 combinations

deﬁning the educational models considered in this pa-

per. For each of them, we calculate how present were

–in the model– each of the 42 topics of the survey.

Our topic assessment is personal, following from our

experience in teaching and our perception of the rele-

vance of the topics according to the selected parame-

ters.

5 THE EDUCATIONAL

WELL-BEING INDEX

Here we relate the components of Sections 3 and 4 to

obtain the EWI. Our index allows us to evaluate the

24 educational models considered in the previous sec-

tion in relation to what the surveyed students consider

important for their well-being.

The ﬁrst step is to use the data gathered in the poll

to extract the relevance of each of the topics included

CSEDU 2024 - 16th International Conference on Computer Supported Education

630

in the survey. For doing so, we use the principal com-

ponents analysis (PCA) statistical method, that re-

duces categorized data to their essential features, the

so-called principal components. The principal com-

ponents are linear combinations of the original vari-

ables. The coefﬁcients in the linear combination are

obtained so that the direction maximally explain the

variance of all the variables. In general, the method

provides an approximation of the original data table

using only these few major components. Following

the SPI deﬁnition, we only consider the ﬁrst compo-

nent of the PCA. This approach allows us to rank by

its relative importance each one of the well-being fac-

tors of the survey. This rank is a projection on the ﬁrst

component of the PCA.

On the other hand, in Section 4, we provide an

assessment (subjective to our view) on how present

are each of the factors considered in the survey within

each of the proposed educational models. We ob-

tained this assessment by subjectively grading the

same survey that we had administered to the students

for each of the 24 possible models.

To obtain the EWI of an educational model, we

use the coefﬁcients of the ﬁrst component of the PCA

to assign an index value to each educational model.

We multiply the value (duly scaled) given to each

question (or factor) by the coefﬁcient of the question

in the ﬁrst PCA component of the students responses.

This allows us to position each model on the principal

component axis in such a way that the higher the value

of the corresponding model on that axis, the more cor-

related the model is with the principal component, or

in other words, the higher the value obtained for each

model, the better the model adapts to what students

consider important for their well-being.

In our case study, we wanted also to analyze the

sensibility of the model to a population. For doing

so, we performed the PCA analysis on three data sets:

the CS, the Math, and the aggregated CS + Math. In

Figure 1, we depict the sorted values of the EWI, for

the three data sets. In Table 5, we present the co-

efﬁcients of the PCA. As the two considered popu-

lation appear to have different sensibilities towards

well-being, none of the three rankings coincide. Let

us look to the PCA coefﬁcients for questions in the

lecturing dimensions. Having in-person classes (Q

)

gets positive weight in EWI from CS and negative in

the other cases, while having them broadcasted live

) gets negative weight in the global data but posi-

tive in the others. Having ofﬂine classes prerecorded

) gets negative weight only from Math and broad-

casting platforms (Q

) only from CS.

It is also worth mentioning the weighting of the

questions in the transportation dimension. Travel-

Figure 2: The EWI per dimension taking the PCA of the

poll from both CS and Math students.

ing time (Q

) and comfort (Q

) get positive weight

in EWI from Math and negative in the other cases.

This might be explained because there are more CS

students and usually their schedule is quite chaotic

while Math students are less and usually have com-

pact schedules.

In Figures 2, 3 and 4 we depict the results of the

EWI rankings per dimension. Under this point of

view, we can visualize better the variations of the EWI

when ﬁxing a dimension. Analyzing the school di-

mension, we can see that the EWI of the educational

models follow the same tendencies in the three cases.

Using CS data, models with S

get higher values of

EWI than models with S

, the situation is reversed for

the Math data. When dealing with the complete data

set, the tendency seems to be dominated by the FIB

population. This is in accordance with the fact that

A Proposal for an Educational Well-Being Index (EWI) for Undergraduate Course Design

631

Figure 3: The EWI per dimension taking only the answers

of the CS students in the PCA.

when weighting school dimension, we have in mind

most trends of CS for S

and more from Math for S

A further interpretation is that in both populations, the

EWI increases in models considering a school model

similar to the chosen one.

From the point of view of the transportation one

can observe variability in the three models, while in

the lecturing dimension, models L

and L

present

the most different inﬂuences in the index. We do not

have any explanation of the changes of the EWI due

to these components. We suspect that there are some

relevant correlations between the lecturing and trans-

portation models that require further study to be ex-

plained.

Figure 4: The EWI per dimension taking only the answers

of the Math students in the PCA.

6 CONCLUSIONS AND FUTURE

WORK

At present, where the economic growth has been

widely criticised, the well-being appears as a funda-

mental issue. Recently, the importance of measuring

well-being has been recognized as a relevant tool in

the evaluation of countries, institutions, communities

and individuals in addition to the classic indicators

used in economics. We can say that, as individuals,

the well-being of every one is a personal concept that

depends at least in part on our worldview (weltan-

schauung). Therefore, the well-being has also many

components and can be analyzed at different levels.

We have proposed and study the EWI as a measure of

well-being for educational models to be used as an a

CSEDU 2024 - 16th International Conference on Computer Supported Education

632

priori appraisal when designing a course implemen-

tation. The EWI relies on parameters extracted from

a student valuation of some well-being topics. As the

student were taken from sectors with diverse interests,

we have been able to see that the index is sensible to

the trends of the selected population. In the coming

semesters, we plan to run a similar poll but asking for

the valuation of the topics in the context of the course.

This will help us to validate the result and understand

better the applicability of the index.

The student’s well-being is a multidimensional

concept, in this paper we have focused only in three

of the many possible dimensions, namely lecturing,

transportation and school facilities, but many other

could be incorporated in the index. Note that, some

of these aspects, like facilities or public transporta-

tion, are external to the pedagogical aspects of course

organization. Our proposal could be adapted to other

environments or interests, by redeﬁning dimensions

and the corresponding topics of interest. We plan to

extend the survey on well-being to a bigger popula-

tion of students from CS and Math and look into more

than one dimension of the PCA. It will be also worth

to perform a factor analysis to identify the relevant

topics in every considered dimension.

REFERENCES

Dodd, A. L., Priestley, M., Tyrrell, K., Cygan, S., Newell,

C., and Byrom, N. C. (2021). University student well-

being in the United Kingdom: a scoping review of its

conceptualisation and measurement. J. Ment. Health,

30(3):375–387.

Jackson, T. (2021). Post Growth: Life after Capitalism.

Polity Press, Cambridge.

Kanonire, T., Kulikova, A., and Orel, E. (2022). Measur-

ing well-being in school and university: Approaches

and challenges. In Irtelli, F. and Gabrielli, F., edi-

tors, Happiness and Wellness - Biopsychosocial and

Anthropological Perspectives. IntechOpen.

Nordhaus, W. (2013). The Climate Casino: Risk, Uncer-

tainty, and Economics for a Warming World. Yale

University Press.

Paulson, S., D’Alisa, G., Demaria, F., and Kallis, G. (2020).

The Case for Degrowth. Polity Press, Cambridge.

Pigou, A. C. (2013). The Economics of Welfare. Palgrave

Classics in Economics. Springer. Originally published

in 1920.

Skidelsky, E. and Skidelsky, R. (2013). How Much is

Enough?: Money and the Good Life. Penguin.

Social Progress Org (2024). 2024 Social Progress Index -

Executive Summary.

The Economist (2023). How well does your country provide

for its citizens?

Wang, R.-J., Lin, S., Tseng, M., Tsai, M.-H., and Chang, T.-

H. (2022). Measuring college campus well-being with

multidimensional indices: Sustainability of higher ed-

ucation in Taiwan. Sustainability, 14(8260).

Webometrics (2023). Ranking Web of Universities.

Wikipedia (2024). List of countries by GDP (nominal) per

capita.

A Proposal for an Educational Well-Being Index (EWI) for Undergraduate Course Design

633