Analyzing the Impact of e-Caduc

ee, a Serious Game in Pharmacy on

Students’ Professional Skills over Multiple Years

Katia Oliver-Quelennec

1,2,3 a

, Franc¸ois Bouchet

1 b

, Thibault Carron

1 c

and Claire Pinc¸on

2 d

Sorbonne Universit

e, CNRS, LIP6, F-75005 Paris, France

Univ. Lille, CHU Lille, ULR 2694, METRICS:

Evaluation des Technologies de Sant

e et des Pratiques M

edicales,

F-59000 Lille, France

Univ. Lille, GIVRE, DIP, France

Keywords:

Learning Game, Game-based Learning, Pharmacy, Simulation, Evaluation, Higher education, Dashboard.

Abstract:

In an academic program of our faculty of pharmacy, we tried to improve the training of future pharmacists

aiming at their professionalization. We proposed a learning game called e-Caduc

ee, which allows students

to train during 3 semesters with about one hundred clinical cases. We investigated the consistency between

skills worked in the game with those deﬁned by the pedagogical team as well as the impact of the game and

of the embedded dashboard on students’ skills. We collected data from the game (activity traces), from the

faculty (academic results) and from the students (opinion about the game). To answer our research questions,

we used both multiple linear regressions as well as classical statistical inference. Results reveal that the score

predictions based on the use of e-Caduc

ee correspond with the deﬁnition of the teachers. We also found clues

that the use of e-Caduc

ee helped with learning some professional skills but the result was not conﬁrmed with

statistical analysis. Finally, we found a link between the use of the dashboard in the game and one particular

professional skill’s academic results (prescription). Our future work aims at developing an adaptive learning

dashboard for the game and analyzing its possible impact.

1 INTRODUCTION

When training health professionals, a saying is often

heard: “never practice the ﬁrst time on a patient”. To

meet this need, simulation is a reliable solution to en-

able future health professionals to train without real

consequences. Thus, learners can practice without ex-

posing patients to risk, making mistake and learning

serenely. Simulation associated with a learning game

allows leveraging the beneﬁts of game-based learning

in training. Alvarez et al. (2012) deﬁne “a Serious

Game (as) an artifact, digital or otherwise, for which

the original intention is to combine with consistency,

both serious aspects such as non-exhaustive and non-

exclusive, teaching, learning, communication, or the

information, with playful elements from the game”.

It is sometimes referred to as a learning game in the

education context (George, 2019), and “[their] main

beneﬁt (...) is the user’s motivation linked to the in-

https://orcid.org/0000-0002-7318-7449

https://orcid.org/0000-0001-9436-1250

https://orcid.org/0000-0001-6982-7055

https://orcid.org/0000-0002-5509-6199

herent goals of the game whose fulﬁllment is a source

of satisfaction and rewards”. Some studies have also

shown a positive impact of learning games on aca-

demic results (Pacheco-Velazquez et al., 2019; Pa-

padakis et al., 2020).

The impact of learning games on knowledge

and skills acquisition is usually evaluated with a

pre-test/post-test, but this method can be “expen-

sive, time-consuming and provides limited informa-

tion about the user in the learning process” (Alonso-

Fernandez et al., 2017). Collecting traces of activi-

ties and their analysis can be a complementary way

to evaluate the impact of a learning game, that may

also help to understand the learning process in order

to improve it (Alonso-Fern

andez et al., 2019). We

developed a learning game called e-Caduc

ee, to sim-

ulate professional life in a pharmacy, used by students

over multiple years. Using various learning analytics

methods, we have investigated the possible impacts

on students’ learning process and tried to answer the

following research questions (RQ):

• RQ1. Are the professional skills worked in the

game consistent with those deﬁned by the peda-

Oliver-Quelennec, K., Bouchet, F., Carron, T. and Pinçon, C.

Analyzing the Impact of e-Caducée, a Serious Game in Pharmacy on Students’ Professional Skills over Multiple Years.

DOI: 10.5220/0010453903310338

In Proceedings of the 13th International Conference on Computer Supported Education (CSEDU 2021) - Volume 1, pages 331-338

ISBN: 978-989-758-502-9

331

gogical team?

• RQ2. Does the learning game allow students to

strengthen their professional skills?

• RQ3. Does the use of the embedded dashboard

have a positive impact on learning?

The outline of this paper is as follows: section 2 re-

views previous works about simulation and learning

games, with a focus on the pharmacy profession, and

introduces the concept of Learning Analytics Dash-

boards (LAD); section 3 describes the learning game

e-Caduc

ee, section 4 details materials and methods of

this study, section 5 shows its main results, section 6

discusses these results and section 7 gives a conclu-

sion and opens the ﬁrst perspectives of our work.

2 PREVIOUS WORK

Simulation and learning game are common tools in

health education and pharmacy curriculum. Cain and

Piascik (2015) identify serious games as a real oppor-

tunity in pharmacy education: “Advantages of seri-

ous gaming in pharmacy education include authen-

tic, situated learning without risk of patient conse-

quences, collaborative learning, ability to challenge

students of all performance levels, high student mo-

tivation with increased time on task, immediate feed-

back, ability to learn from mistakes without becoming

discouraged, and potential for behavior and attitude

change.” A review of several educational games in

pharmacy concludes on a lack of evidence that edu-

cational games foster learning in pharmacy schools,

even if they may impact students’ motivation (Abu-

rahma and Mohamed, 2015). An imbalance between

educational and entertainment components could be

the cause. However when coupled with simulation,

learning games seem to have a positive effect on the

learning process as Berger et al. (2018) experimented

with one clinical case, or as Abed et al. (2016) stud-

ied for vocational training in a professional context.

This result is still uncertain and could be context-

dependent, as some recent studies (Blani

e et al., 2020)

cast doubts over the positive impact of learning with

simulation compared with health traditional teaching

method based on clinical cases and debrieﬁng. It is

therefore critical to keep on investigating this issue in

multiple contexts. Moreover, a potential limit to the

aforementioned works is that they tend to focus on

simulation used over a relatively short period of time

(from a few hours for Blani

e et al. (2020) to several

weeks at home for Pacheco-Velazquez et al. (2019) or

Berger et al. (2018)). We hypothesize that these vari-

ous time of use could be one of the reasons explaining

the discrepancy in the results observed so far. To the

best of our knowledge, we did not ﬁnd study about the

learning impact of a learning game used over several

years by the same cohort of students.

In a learning game, a Learning Analytics Dash-

board (LAD) is a helpful tool which supports online

learning. A LAD can be deﬁned as “a single display

that aggregates different indicators about learner(s),

learning process(es) and/or learning context(s) into

one or multiple visualizations” (Schwendimann et al.,

2017). Thus, the user can use all this information to

make a decision. By summarizing learning data for

the student with an appropriate visualization, LAD

supports awareness of learning process (Jivet et al.,

2017) and self-assessment. Its functions mirror those

of the dashboards in video games, that show to the

player which goals are achieved and that guide them

to the next level. Today, LADs in learning games are

increasingly being studied (Scheneider and Lemos,

2020) and are evolving from a static to an interac-

tive form. Alonso-Fernandez et al. (2017) explained

that ”not all classical gamiﬁcation metrics are suit-

able or useful for learning and training”, by exam-

ple, peer comparison can discourage students or speed

constraint can have negative effects. We can suppose

that data deﬁned for a LAD of a serious game by

Alonso-Fernandez et al. (2017) (proportion of correct

answers, ﬁnal score) are adapted to LAD in a learning

game with simulation in the speciﬁc context of health.

3 LEARNING GAME E-CADUC

3.1 Overview of the Game

The learning game e-Caduc

ee scenario introduces the

player to a little town with several key places (cf. ﬁg-

ure 1), where students start a pharmacy internship.

Their main objective is to handle patients (cf. ﬁgure

2), by dispensing medications and/or providing med-

ical advice. Each visit of a patient corresponds to a

clinical case, built with the following general frame:

context deﬁnition with a 3D video (cf. ﬁgure 2), ques-

tions relative to the pathophysiology of the disease,

therapeutic strategies, pharmacology of medication,

treatment optimization, and education and counseling

of the patient. By following this framework, the clin-

ical cases written for the game allow students to work

on the different skills planned by the teaching team:

prescription, prescription dispensing and advice. Two

endings are possible, positive or mildly negative (pa-

tients never die), depending on the student’s score.

Clinical cases have been written by experts from dif-

ferent universities in France and validated by an ed-

CSEDU 2021 - 13th International Conference on Computer Supported Education

332

itorial committee. They have formative evaluation

purposes and contain multiple explanations of the an-

swers to the exercises and links to ofﬁcial references

(i.e. immediate feedback is provided after a case to

explain why the selected answer was correct or incor-

rect).

In e-Caduc

ee, as in a real pharmacy, patients arrive

randomly and the player never knows which topic will

be next. Some clinical cases are related to each other,

with a previously seen patient that may randomly

come back to the pharmacy with questions about the

evolution and the treatment of their pathology. If stu-

dents are successful (more than 60% of good answers

in several clinical cases), they receive some in-game

rewards from the internship supervisor which can be

extradiegetic (e.g. badges) or intradiegetic (e.g. the

virtual character is invited to restaurant). After a ﬁxed

number of successful clinical cases, students can be

promoted as assistant, and then as pharmacist asso-

ciate, the level of the clinical cases increasing with

the status of the student in the game. Although a

scenario leads the student through the game, he or

she can at any time review previously seen clinical

cases and make the choice to deal with lower level

untreated cases. This allows them to revisit previous

topics if needed. When the player becomes assistant,

he can also visit different places of the town, such as

the faculty which allows him or her to review courses

or to train about a speciﬁc pathology. Finally, the tro-

phy room provides the student with a static dashboard

(cf. ﬁgure 3) which gives the obtained scores, the

opinion of the internship supervisor, the correspond-

ing rewards and the rewards to be earned. The game

ends with the most complicated case as the ﬁnal chal-

lenge. The learning game was developed with open

source software Scenari Opale to write clinical cases,

and Scenari Topaze to develop the game engine. This

choice allows teachers to update clinical cases easily

by themselves. 2d illustrations and 3d videos have

been made by external companies.

Figure 1: Illustration of the ﬁctitious town Berdeghem and

its virtual pharmacy.

Figure 2: A patient in the pharmacy to introduce each clin-

ical case, in this example a pregnant woman coming with a

prescription.

3.2 Game Implementation

e-Caduc

ee has been used in several public universi-

ties in France, with different modalities (distributed

over a single semester as an option or spread over

3 years with a validation requirement) and different

game conﬁgurations (clinical cases by theme or by

level of difﬁculty). In the current study context, the

game has been an optional part of a mandatory teach-

ing unit PROFFIteROLE (PRatiques OFFIcinales et

jeux de ROLEs), for the last three semesters of the

academic program of the Faculty of Pharmacy of Lille

(France) intended to future dispensary pharmacists.

In this teaching unit, students alternate practicing on-

line with the game e-Caduc

ee and in physical class-

rooms. Students begin with a physical classroom

where teachers introduce the teaching unit by present-

ing its global organization, learning objectives, evalu-

ation methods, as well as the e-Caduc

ee game. Dur-

ing this course, students can discover the game and

ask questions. Then, they are free to play online, as

many times as they want, in group or alone, or even

to not play at all. Meanwhile, during each semester,

3 sessions of face-to-face simulations are organized

with 3 clinical cases to be treated. Students are given

the topics in advance and asked to prepare them, pos-

sibly with the serious game. Sessions are then or-

ganized in three sequences: a prebrieﬁng introduces

the aim and the contents of the session; the simula-

tion time gives students the opportunity to alternate

playing as a pharmacist, a patient or an observer; a

debrieﬁng summarizes the skills that have been used

properly and those that need to improve. Although

the main focus in on training the pharmacist skills,

taking on the role of patient helps to develop empa-

thy for patients, and playing as an observer helps in

developing a critical and objective look while ﬁlling a

criterion-based evaluation grid of the pharmacist.

Each semester, the content of the teaching unit

evolves, students deal with 9 new clinical cases in ad-

Analyzing the Impact of e-Caducée, a Serious Game in Pharmacy on Students’ Professional Skills over Multiple Years

333

Figure 3: Dashboard of the game.

dition to the game in face-to-face simulation and can

work on about 30 new clinical cases in the game on-

line. This progressive releasing of new cases encour-

age students to go back to the learning game for a sus-

tained use over multiple semesters. Clinical cases in-

cluded in the game change and became more complex

with cases involving multiple pathologies. Face-to-

face simulations aim at developing other skills which

cannot be directly targeted in the learning game,

such as communication between health profession-

als by including medical and dental students to simu-

late the direct exchanges between health professionals

needed to solve complex clinical cases. Student as-

sessment is also adapted to the pedagogical goals of

each semester and takes several forms: students have

to solve 2 clinical cases from the serious game in the

ﬁrst semester, to write in group a new clinical case

in the second semester, and to manage a face-to-face

simulation with a teacher for the last semester.

4 MATERIALS AND METHODS

4.1 Materials

In order to answer to the aforementioned three re-

search questions, we collected data from several

sources:

• From the game, for each of the 3 semesters: num-

ber of game attempts, overall average grade, num-

ber of clinical cases completed for each of the 3

levels of difﬁculty, access to the online training

space and access to the trophy room. All these

data are exported in csv format from the Learn-

ing Management System (Moodle) in which e-

Caduc

ee is embedded.

• From the faculty: overall average grade, whether

the year was repeated or not, grade in this teaching

unit, grade obtained in the mandatory internship

each student has to do after the last semester (with

detailed assessments for 6 different skills: posol-

ogy [POS], good preparation practices [GPP],

prescription [PRE], work in public health [WPH],

prescription dispensing [PDI] and advice [ADV]).

All these data are exported in csv format from the

university management software (Apog

ee).

• From a questionnaire to the students: whether

they worked in a pharmacy during their stud-

ies for each semester (independently from the ﬁ-

nal mandatory internship), their opinion about the

game and its current dashboard (trophy room) and

their expectations to improve them. These data

are exported in csv format from an optional on-

line questionnaire (Lime Survey) sent to students

during the third semester.

We worked with data coming from two cohorts of stu-

dents of a faculty of pharmacy, with n=94 for 2017-

2018 and n=100 for 2018-2019.

4.2 Methods

Prior to data collection, we worked with our data pro-

tection ofﬁcer to ensure compliance with the Gen-

eral Data Protection Regulation (GDPR), the Euro-

CSEDU 2021 - 13th International Conference on Computer Supported Education

334

pean Union’s data protection law. Thus, teachers have

informed students about our research and why we col-

lect data. Then, students could choose to access a ver-

sion of the game with or without data collecting.

Initially, we considered comparing the results of

students who played and did not play the game by di-

viding the cohort in two groups, one with the game

and one with just clinical cases. But this method can-

not be used with the actual game design. As explained

in 3.1, to simulate a real pharmacy, patients arrive ran-

domly and students do not have to do all clinical cases

to succeed in the game. We used this method to evalu-

ate the game design and to help us improve it when the

game was created, and we observed that each student

of each group took a different path (different clinical

cases studied and different number of attempts). We

therefore considered other methods to assess the im-

pact of the game on learning: machine learning-based

one and classical statistical inference summarized in

ﬁgure 4.

To examine the ﬁrst and second research ques-

tions, our approach consisted in training multiple lin-

ear regression models in order to predict the grade as-

sociated to each of the 6 skills evaluated in the in-

ternship (as well as the overall internship grade), us-

ing features coming from the game data. The idea

is that if the game trains a particular skill, features

representing how much the students use the game

should help in predicting the grade they obtained on

the skill during the internship. On the opposite, if a

skill is not particularly trained in the game, the corre-

lation between the game usage and the grade on that

skill should be low and therefore the regression model

should perform poorly for it. This analysis was con-

ducted using RapidMiner (version 9.8) with 10-fold

cross-validation to estimate the correlation (Pearson

r) and mean absolute error (MAE).

To explore the second and third research ques-

tions, our approach used continuous variables ex-

pressed as mean ± SD or median [25th – 75th per-

centile], as appropriate. Categorical variables are pre-

sented as absolute numbers and percentages. Linear

regression analysis was used to study the relationship

between academic results and the use of the learn-

ing game. Models were adjusted on potential con-

founding factors. The linearity assumption for con-

tinuous covariates was assessed by testing the addi-

tion of a quadratic component in the model. Multi-

variate models were built by ﬁrst including all predic-

tors and then using a manual backward selection to re-

duce the model. Regression underlying assumptions

were visually inspected with residual plots. Longitu-

dinal data of academic results for the three semesters

were analyzed using repeated-measures analysis of

covariance (PROC MIXED) with a REPEATED state-

ment for within student correlation over the three

semesters. The main covariate was the number of vis-

its of the LAD, and potential confounding variables

were included in the analyses. Models were built as

described above for the linear regression models. The

two-sided type I error was set at 5%. Analyses were

conducted using SAS software (SAS version 9.3, SAS

Institute Inc., Cary, NC, USA).

5 RESULTS

To answer our ﬁrst and second questions, the table 1

presents our results to predict the grade associated to

each of the 6 skills evaluated in the internship (posol-

ogy [POS], good preparation practices [GPP], pre-

scription [PRE], work in public health [WPH], pre-

scription dispensing [PDI] and advice [ADV]). Ac-

cording to the teaching team, e-Caduc

ee allows to

work mainly on prescription (PRE), prescription dis-

pensing (PDI) and advice (ADV) skills.

Table 1: Performance of the linear regression models

trained on each skill and on the overall internship grade.

Skill Correl. (r) MAE Best

M SD M SD model

POS .200 .314 1.871 1.123 Lin. reg.

GPP .337 .439 2.900 1.369 GLM

PRE .425 .374 2.575 1.087 Lin. reg.

WPH .330 .398 1.763 0.993 Lin. reg.

PDI .306 .363 2.306 0.971 GLM

ADV .156 .307 2.830 1.093 Lin. reg.

Total .271 .369 1.816 0.861 GLM

To answer our second question, our analyses have

identiﬁed links between the learning game and aca-

demic results. A link between the use of the game

and the teaching unit score (p = 0.02) is highlighted.

Playing 3 times on e-Caduc

ee increases the teaching

unit score by an average of 1.29 points (IC95% [0.40;

2.18]). But we did not identify any link between the

use of the game and the internship results. The stu-

dents who responded to the game evaluation survey

(n = 102 for both cohorts) felt that: the game was sat-

isfying at 99%, the distribution of clinical cases by

level were appropriate at 86%, the number of cases

to be treated was adapted at 92%, the gaming was

adapted at 92%, the game allowed a synthesis of

knowledge at 85%, the game allowed to learn new

knowledge at 86%, the game helped to improve one’s

professional practice at 83%.

Analyzing the Impact of e-Caducée, a Serious Game in Pharmacy on Students’ Professional Skills over Multiple Years

335

Figure 4: Data and analyses used to investigate the three research questions.

And last but not least, we did a focus on players only,

to observe if the use of the LAD had consequences

on their results. We analyzed the use of the game by

each student for each of the 3 semesters (n = 431)

and we identiﬁed a link between the number of visits

and the results in the game (p = 0.003). Visiting vol-

untarily the trophy room at least twice increased the

results in the game EC by an average of 6.41 points

(IC95% [2.82; 10.]). Then, we tried to search a link

between the use of the LAD and the internship re-

sults, with adjustment factors on gender, overall av-

erage, teaching units results and game results of each

semester. We could demonstrate a link between the

use of the LAD during the last semester and prescrip-

tion skill evaluation (p = 0.04). Visiting the trophy

room once increased the average prescription score by

2.12 points (IC95% [0.41; 3.82]). A link between the

number of visits to the trophy room during the second

semester and the prescription dispensing skill score

(p = 0.02) was found. Visiting the trophy room once

increases the average prescription dispensing score by

1.82 points (95% IC [0.08; 3.56]).

To complete this study, we asked the students’

opinions from the cohort 2017-2018 about the LAD.

For students who responded (n = 47), 85% of the

players remembered visiting the trophy room. Some

elements are considered useful by the majority, such

as overall results and results by pathology. Other el-

ements do not meet unanimous approval, such as the

opinion of the internship supervisor. The 15% who

did not remember having visited the trophy room ex-

plained this for 3 reasons: not being aware of it, ﬁnd-

ing it useless or not taking the time. Some students’

comments are encouraging, as for example “Thanks

to these results, I could see the areas in which I had

already achieved good results, but also and above all,

I could see the gaps that I had not necessarily noticed

when I had completed the questionnaires. When other

clinical cases dealing with subjects in which I did not

have a lot of points, I took more time to answer and

possibly do further research on the internet or on my

courses.” The students then expressed some expecta-

tions for improving the LAD in the future, with dif-

ferent opinions on the types of data and visualizations

desired.

6 DISCUSSION

The main goal of the game is to prepare students

before real professional practice. Teachers have de-

ﬁned more precisely the skills concerned within the

game, which are prescription, prescription dispensing

and advice. The results of the score predictions cor-

respond partially to the teachers’ opinions. We can

answer to the ﬁrst question, some professional skills

worked in the game are consistent with those deﬁned

by the teaching team, as prescription and prescrip-

tion dispensing, but not the advice skill. To under-

stand, it would be necessary to work with the peda-

gogical team to identify how this skill is evaluated.

We can conclude the pedagogical alignment between

the game, the assessment and the skills of prescription

to be acquired is correct.

In response to our second question, which sup-

poses that the learning game allows to reinforce pro-

fessional skills, the ﬁrst analysis with a machine

learning type of approach suggests links between the

game and some professional skills. But with the clas-

sical statistical inference method, we establish a link

between using the game and academics results in the

teaching unit, but not with the professional skills eval-

uated during the internship. We compared students

who played a little, a lot and those who did not. But

the deﬁnition of non-player may be inaccurate, as we

simply deﬁned non-players as students with no trace

in the game. We can assume that these data can be

reﬁned, considering that some students have chosen

CSEDU 2021 - 13th International Conference on Computer Supported Education

336

to use a game with no record, but they were counted

as non-players. Then, as the game is optional, they

could play in groups, and in this case, we have no

data about it. Informal interaction with students sug-

gest that this type of behavior does exist, but we have

not evaluated their prevalence to know if it is high

enough to potentially signiﬁcantly affect our observa-

tions. We should identify students who played with

no data record to have some more accurate results.

According to the students’ opinions, the majority be-

lieves that the game e-Caduc

ee improves their profes-

sional practice (83%). It might contribute to build up

the self-conﬁdence in their professional practice that

is necessary for caring for real patients.

Some results on the impact of the dashboard on

learning allow us to answer partially our last question,

suggesting that its use can improve learning. We have

observed that the use of the LAD improves the result

in the game. We can assume that the proposed data

are relevant for some players. The use of the LAD

also shows an increase in the scores of the internship

evaluation. If we look at the semestrial results more

precisely, it appears that its use in the second semester

has a stronger impact on the prescription note. Most

of the clinical cases of that semester involve a pre-

scription, which could explain this result.

Debrieﬁng has an important place in learning

games as Crookall (2010) explains “Some learning

often occurs while a game is being played, but deeper

lessons are drawn and drawn out in a debrieﬁng ses-

sion”. He says that with “the use of computers, we

have a powerful tool indeed for debrieﬁng. . . . The

data can then be processed to provide material for

feedback during play, as in-game debrieﬁng”. This

tool has some similarity with the deﬁnition of a LAD.

Thus, a LAD may have a positive impact on learn-

ing games because it can take the role of the in-

game debrieﬁng. LAD’s elements deﬁned as useful

by some students, e.g. the results by categories, seem

to provide useful feedback for this in-game debrief-

ing, and thus enable awareness of acquired and under-

developed skills. Finally, we may wonder why some

players have not used LAD. We can assume that this

very simple LAD was not adapted to their needs as

demonstrated by the work of Roberts et al. (2017)

and Teasley (2017). Indeed, adaptive LADs seem

to better meet the expectations of all users (Dabbebi

et al., 2017), which is supported by the expectations

expressed by our students, and this is one of the direc-

tions we intend to work on in the future.

The positive results of our study are certainly in

line with several existing works on the acquisition of

professional skills. But replicating in different con-

texts is important to conﬁrm a result, as some studies

shew different results in health domain (Blani

e et al.,

2020). Then, our research is based on an moderate

sample size, but with the use of a game spanning over

multiple years, which is a modality little studied ac-

cording to our investigations. Finally, our approach

presents an original analysis using machine learning

methods in addition to traditional analysis methods.

7 CONCLUSIONS AND NEXT

STEPS

e-Caduc

ee offers students in pharmacy the opportu-

nity to work on their professional skills by treating

almost a hundred of clinical cases. In our context,

we ﬁrst identiﬁed that the professional skills worked

in the game were consistent with those deﬁned by the

teaching team, e-Caduc

ee seemed to be adapted to the

pedagogical objectives. According to the students’

opinion, the learning game may reinforce professional

skills, but with a partial conﬁrmation of a link be-

tween the game and internship results using several

analysis methods as learning machine and classical

statistical inference. Finally, we observed that the

use of the student’s dashboard in the game enhanced

learning, with a link between the use of the dashboard

and prescription skills results. To conclude, our work

deals with a widely addressed issue but with an origi-

nal context with a multi-year game and several analy-

sis methods. Although we do not demonstrate it here

and further studies would be needed, the fact we ﬁnd

a positive impact of the game is in line with our ini-

tial hypothesis that it may be important to consider

long-term use of learning game simulations to mea-

sure positive impact.

Our future work will focus on the LAD by ex-

ploring adaptive approaches. First, we will im-

prove the LAD design using co-design methods such

as PADDLE (Oliver-Quelennec, 2020) adapted from

Dabbebi et al. (2019), and then observe possible im-

pacts of these adapted and adaptive LAD.

ACKNOWLEDGEMENTS

The research published in this article is based on re-

sults of teaching unit PROFFIteROLE and we thank

all the actors involved: teachers of GIVRE and Ben-

jamin Hourdouillie to extract data. This work is in-

cluded in the P3 project, developed by Universit

e de

Lille and co-ﬁnanced by the iSite Universit

e Lille

Nord-Europe.

Analyzing the Impact of e-Caducée, a Serious Game in Pharmacy on Students’ Professional Skills over Multiple Years

337

REFERENCES

Abed, H., Pernelle, P., Ben Amar, C., and Carron, T. (2016).

Serious Games for vocational training: a compared

approach. In AICCSA 2016 - 13th ACS/IEEE Inter-

national Conference on Computer Systems and Appli-

cations, Agadir, Morocco.

Aburahma, M. H. and Mohamed, H. M. (2015). Educa-

tional Games as a Teaching Tool in Pharmacy Curricu-

lum. American Journal of Pharmaceutical Education,

79(4):59.

Alonso-Fernandez, C., Calvo, A., Freire, M., Martinez-

Ortiz, I., and Fernandez-Manjon, B. (2017). Sys-

tematizing game learning analytics for serious games.

In 2017 IEEE Global Engineering Education Confer-

ence (EDUCON), pages 1111–1118.

Alonso-Fern

andez, C., Cano, A. R., Calvo-Morata, A.,

Freire, M., Mart

ınez-Ortiz, I., and Fern

andez-Manj

on,

B. (2019). Lessons learned applying learning analyt-

ics to assess serious games. Computers in Human Be-

havior, 99:301–309.

Alvarez, J., Djaouti, D., and Rampnoux, O. (2012). In-

troduction au Serious Game - Serious Game an In-

troduction (French and English). L>P. Questions

eoriques, Paris, 2e

edition edition.

Berger, J., Bawab, N., De Mooij, J., Sutter Widmer, D., Szi-

las, N., De Vriese, C., and Bugnon, O. (2018). An

open randomized controlled study comparing an on-

line text-based scenario and a serious game by Belgian

and Swiss pharmacy students. Currents in Pharmacy

Teaching and Learning, 10(3):267–276.

Blani

e, A., Amorim, M.-A., and Benhamou, D. (2020).

Comparative value of a simulation by gaming and a

traditional teaching method to improve clinical rea-

soning skills necessary to detect patient deterioration:

a randomized study in nursing students. BMC Medical

Education, 20(1):53.

Cain, J. and Piascik, P. (2015). Are Serious Games a Good

Strategy for Pharmacy Education? American Journal

of Pharmaceutical Education, 79(4):47.

Crookall, D. (2010). Serious Games, Debrieﬁng, and Simu-

lation/Gaming as a Discipline. Simulation & Gaming,

41(6):898–920.

Dabbebi, I., Gilliot, J.-M., and Iksal, S. (2019). User

Centered Approach for Learning analytics Dashboard

Generation. In CSEDU 2019, volume 2 of CSEDU

2019, pages 260–267, Heraklion, Greece.

Dabbebi, I., Iksal, S., Gilliot, J.-M., May, M., and Gar-

latti, S. (2017). Towards Adaptive Dashboards for

Learning Analytic: An Approach for Conceptual De-

sign and implementation. In 9th International Con-

ference on Computer Supported Education (CSEDU

2017), pages 120–131, Porto, Portugal.

George, S. (2019). Games, Simulations, Immersive Envi-

ronments, and Emerging Technologies. In A, T., edi-

tor, Encyclopedia of Education and Information Tech-

nologies, pages 1–10. Springer International Publish-

ing.

Jivet, I., Scheffel, M., Drachsler, H., and Specht, M. (2017).

Awareness Is Not Enough: Pitfalls of Learning Ana-

lytics Dashboards in the Educational Practice. In Data

Driven Approaches in Digital Education, pages 82–

96, Cham. Springer International Publishing.

Oliver-Quelennec, K. (2020). Vers une conception partic-

ipative de tableaux de bord

etudiants adaptatifs avec

PADDLE. In Actes des 8i

emes RJC-EIAH 2020,

page 8.

Pacheco-Velazquez, E., Palma-Mendoza, J., Arana-Solares,

I., and Cotera-Rivera, T. (2019). LOST: A Serious

Game to Develop a Comprehensive Vision of Logis-

tics - ProQuest. In Proceedings of the 13th European

Conference on Games Based Learning, Denmark.

Papadakis, S., Trampas, A., Barianos, K., Kalogiannakis,

M., and Vidakis, N. (2020). Evaluating the Learning

Process: The “ThimelEdu” Educational Game Case

Study. In CSEDU 2020 Proceedings, pages 290–298.

Roberts, L. D., Howell, J. A., and Seaman, K. (2017). Give

Me a Customizable Dashboard: Personalized Learn-

ing Analytics Dashboards in Higher Education. Tech-

nology, Knowledge and Learning, 22(3):317–333.

Scheneider, T. and Lemos, R. (2020). Use of Learning An-

alytics Interactive Dashboards in Serious Games: A

Review of the Literature. International journal for in-

novation education and research, 8(3):150–174.

Schwendimann, B. A., Rodr

ıguez-Triana, M. J., Vozniuk,

A., Prieto, L. P., Boroujeni, M. S., Holzer, A., Gillet,

D., and Dillenbourg, P. (2017). Perceiving Learning at

a Glance: A Systematic Literature Review of Learning

Dashboard Research. IEEE Transactions on Learning

Technologies, 10(1):30–41.

Teasley, S. D. (2017). Student Facing Dashboards: One

Size Fits All? Technology, Knowledge and Learning,

22(3):377–384.

CSEDU 2021 - 13th International Conference on Computer Supported Education

338