PRECODEM, AN EXAMPLE OF TAO IN SERVICE OF
EMPLOYMENT
Jadoul Raynald
Centre de Recherche Public Henri Tudor
29 JFK, L-1855 Luxembourg

Mizohata Sachié
Université de Paris V, René Descartes, Sorbonne
12 rue de l’Ecole de Médecine, 75270 Paris Cedex 06
ABSTRACT
In the present context of fast evolving technology, a number of computer scientists and engineers stay unemployed
because their competencies get outdated and they face difficulty in meeting competency requirements in the workplace.
To provide appropriate job trainings can be a solution. But, a few question come up immediately. "How can we assess
what kind of competencies an individual lacks?" "What kind of training is needed to develop a specific competence?"
This paper describes an initiative that tries to answer these questions by employing a special computer-based assessment
platform called TAO.
KEYWORDS
TAO, PRECODEM, employability, CBA, testing, competence.
1. INTRODUCTION
The information technology sector is vast filled with potentials. Although there have been some stagnations
in recent years, the job market in this industry stays high and the hiring outlook is bright worldwide (Gille
and Marti, 2000; French Ministry of Education, 2006).
Nevertheless, some computer scientists and engineers stay jobless. As Amartya Sen, Nobel Prize winning
economist argues, beyond the financial problems, being unemployed has numerous implications such as loss
of work motivation and marketable skills (Sen, 2000). Also, for the job seekers who usually see
unemployment as unacceptable, especially when the job demand is high, not knowing how to escape this
dead-end situation is even more difficult.
Based on this observation, first, we shall ask why motivated job seekers cannot fit the job vacancies, and
what the solutions can be.
In this paper, we start discussing the problems of unemployment in IT (Information Technology) sector.
We then explore several elements as potential solution. As an example, we present the PRECODEM project,
an initiative of the Labor Agency of Luxembourg (ADEM), financed by the European Union. Afterwards, we
turn to the internals of the TAO* platform, a computer-based assessment (CBA) system, and see how it meets
the ambitions and purposes of the project. Finally, we conclude this paper with some recommendations for
the deployment of the TAO tool kit as a support to a reactive and lasting policy for employability.

* TAO: in French, Testing Assisté par Ordinateur that can be translated by Computer Aided Testing or Computer-Based Assessment.
IADIS International Conference on Cognition and Exploratory Learning in Digital Age (CELDA 2006)
259

2. EMPLOYABILITY AND COMPETENCE
2.1 Why is there a crisis of employability in IT sector?
There are many reasons that can hinder an individual from getting a job, namely social, cultural, structural,
and psychological factors (Sen, 2000). Nevertheless, in IT, previous studies show that the dramatically fast
evolution of knowledge, know-how, and tools in regard with this sector are main factors of job loss.
The power of computers doubles nearly every two years (Tuomi, 2002). Thus, IT systems can afford to be
steadily more complex, and consequently IT professionals try to upgrade their skills and knowledge about
hardware and software architectures, new development paradigms and languages, work reorganization and
the like (Ilkka, 2002).
This rapid evolution, therefore, has a direct impact on the employability of the individual. Note that the
employability can be defined as the set of prerequisites that should have a minimum level to enable the
individual to get a job or to keep a job (Bolton, 1981). Bolton’s definition modified by the recent studies
shows that the employability carries eight factors (INTEREF, 2001).
In the next paragraphs, we will focus on the three first factors of this employability graph (see “Figure 1.
Employability graph”) since they are particularly pertinent for the PRECODEM project.
First, (1) "employment of tomorrow and demanded
profiles" mean that an economic and technology watch
should be held in order to identify the growth markets and
the job trainings that match the demanded profiles.
Unfortunately, this information is not often accessible by
job seekers.
Second, (2) "competence and adaptability" is related to
new knowledge, know-how, organizational modes, and
tools. Earlier studies show that the "semi-professional life"
(or necessary time for an individual to lose half of her or
his employability and technical skills according to the
evolution of knowledge and technologies) was between 7
and 14 years in the 1960s, and now it’s between 3 and 5
years (Kearsley, 1989; Casey, 1998; Guillemard, 2003).
Similarly, in conjunction with the instability and de-
standardization of life course transitions (labor market
entry, career changes, divorce, going back to school,
maternity leave, etc), when the workers hit 40 year-old they meet an age barrier that many companies
consider to be an upper age limit on hiring; after 40, the employees are seen as "too old" to be competitive
and employable and their future trainings would not pay back enough on investment (Guillemard, 2003;
Zanardelli and Leduc, 2006). However, the European Commission opposes this ageism and emphasizes that
the employability mainly depends on the third factor (3) "training and competence acquirement" (European
Commission, 2002) and that the employees should be able to access to life-long learning without
discrimination (Guillemard, 2003; Wannell and Ali, 2002).
Paradoxically, the individuals with high qualifications face less discrimination in terms of trainings access
after 40 and less competency depreciation (Wannell and Ali, 2002). This can be partly explained because the
companies want to protect the heavy investment related to the workers with high degrees. Another
explanation can be found in the nature of universities' programs themselves that are more oriented to meta-
knowledge, helping the individual to progress through self-learning culture. The worker's capacity to gain
and develop new skills during the professional life reinforces her or his usefulness for a company.

** “Capability” refers the Capability Approach developed by A. Sen, M. Nussbaum and others, that studies the functioning of a person
(states and actions) and her capabilities (freedom and opportunities to realize these functioning) (Sen, 2000)
*** This graph does not aim to define a reference such as the Employability Skills Framework (Allen Consulting Group, 2004)

Figure 1. Employability graph***
ISBN: 972-8924-22-4 © 2006 IADIS
260

2.2 Do we really want to escape the employability crisis?
It seems that we do. There are the will as well as the political, financial, organizational and technological
means and resources. The life-long education has indeed become a reality in certain countries such as
England, Denmark, and Sweden and there are numerous initiatives that slowly bring about results (European
Commission, 2006).
For example, in Europe, the Internet has now a high penetration rate in enterprises and in the households
where it has become a well-used pedagogical and semi-structured educational tool (European Commission,
2001; Statec, 2006; Dryburgh, 2002).
Testing and formative assessments have also become important fields of research and technology
development because the initiatives in these sectors are not only reshaping the training contents, but are also
focusing on the improvement of the involved processes in terms of pedagogy, coordination, and suitability of
the contents towards the capabilities, experiences, and expectations of the tested persons.
2.3 PRECODEM, a promising initiative
2.3.1 The PRECODEM project in brief
PRECODEM ("PRÉparation aux COmpétences de DEMain" or Preparation for the Competencies of
Tomorrow) is a project sponsored by the European Union and has been conducted by the Research Centre
Henri Tudor in partnership with the ADEM.
The purpose of this project is to design, to experiment, and to evaluate the new approaches in terms of
training and prevention of unemployment (Szelest and Reiter, 2006). In this context, three work packages
were defined that were solely related to the phases of the project.
During the first phase, the project team built a reference framework eliciting the relevant competencies
for two occupations of IT sector. These two activities were selected by the ADEM (based on a business
watch done by the Centre de Veille Technologique of the Centre Henri Tudor). These occupations were:
development (Analyst-Programmer or AP) and security administration (Security Infrastructure or SI). In
order to develop an appropriate framework, the project team opted for a really high level of details in the
definition of the competencies; so, the maturation of the framework was a long and complex process; the
expected level of granularity was able to be reached through a tight collaboration with a group of
professionals, experts in specific IT domains. The work package resulted in a framework integrating the best
practices and knowledge demanded on the employment market so that, later on, the test contents and courses
could be based on it.
In the second phase, the trainers (most of them were experts in the first work package) designed and
developed ten innovative teaching modules (4 in AP and 6 in SI) related to the practices and the advanced
technologies demanded on the labor market. For example, the modules themed UML analysis, advanced
architectures, Java enterprise development, Microsoft platform security, and Web Services.
The final work package was the most crucial for the project and the most original for the project team. It
consisted, by using the computer-based assessment (CBA) system developed by the Centre Henri Tudor
(a.k.a. TAO platform), in the evaluation of the improvement regarding the quantity and the quality of the
professional competence of the job seekers. The results of the assessment process (pre-tests and post-tests),
enriched with the feedbacks of all the actors of the project (job seekers, experts, trainers, support
structures,...) also had to validate or invalidate the methodology (competence framework, trainings, and
testing) and clearly point out the gains that can be expected by ADEM and by the job seekers' supporting
structures as well as the limitations met using this methodology.
IADIS International Conference on Cognition and Exploratory Learning in Digital Age (CELDA 2006)
261

Table 1. Age distribution Table 2. Some statistics about the volunteer job seekers
2.3.2 Some identified factors of success for the project
In order to respond to the needs and wants of the job seekers, the Centre Henri Tudor contacted the
volunteers and interviewed them. Each curriculum vitæ was also analyzed and the relevant information (e.g.
age, gender, number of months of unemployment) were encoded in the TAO platform (see. “Table 1. Age
distribution” and “Table 2. Some statistics about the volunteer job seekers”).
In addition, the trainers were informed about the TAO platform, so that they could better use it.
Moreover, they were allowed to attend improvement trainings for their teachings (group management,
creation of course support, etc) in the FormIT project (Vandenabeele and Giltay, 2005).
2.3.3 The key steps of the project
The TAO evaluation for the volunteers took place twice: before (“pre-test”) and after (“post-test”) the
trainings. The comparison of the results of the tests combined with the profile analyses will allow to evaluate
(in July - August 2006) the progress in terms of professional competence of the volunteers, their facilitation
skills and to assess the quality of both the trainings and methodology.
2.3.4 Final report and recommendations
The final results of this project will be published in the end of summer 2006. By that time, the synthesis
report containing all the recommendation will be addressed to the national competent authorities and, if
possible, PRECODEM methodology will be generalized and available in entire Luxembourg.
2.4 What is TAO?
2.4.1 A description of TAO
For some people, TAO may appear to be another platform dedicated to the computer-based assessment. But,
as we will see it, indeed, TAO does a lot more!
To explain its power and potentialities, we must give a few words about its origin. The TAO project was
initiated by the EMACS research group of the University of Luxembourg and the Research Centre Henri
Tudor became responsible of its development.
ISBN: 972-8924-22-4 © 2006 IADIS
262

The EMACS group (mainly made of pedagogues, education researchers, and teachers) had developed a
global and integrated approach to the assessment process; this vision was not only focused on the testing
phase but also was including the modeling of all the resources involved in a test (profiles of the people taking
the test, contents of the questions, etc.), as well as the definition and exploitation of the results. Thus, this
holistic approach quickly revealed itself as a challenge that could not be managed without the organizational
capabilities of the computer science as a tool (Martin, Busana et al., 2004).
As they could not find the tool that could instrument their vision of the assessment process, the members
of the EMACS group asked the Centre Henri Tudor to develop an open, flexible, scalable, distributed,
collaborative, and yet integrated and coherent testing platform (Martin, Latour et al., 2005).
During the analysis process, the interviewed people were asked to give their own definitions of the
concepts of student, test, question, and the like, and none of each definition was strictly same as the one
given by other colleagues, thus it was obvious that the analysts of the Centre encountered a dilemma: either
they had to model "large" – it means that they had to define each concept, for example, a subject including all
the features and properties enumerated by the interviewees, or they had to model "small" – only retaining the
commonly accepted and pertinent properties for each concept (while secretly hoping that no new concept or
property would ever emerge). Between these main and opposite directions, the analysts found a path: this was
called "semantic web" and the project team decided to give it a chance (Berners-Lee, Hendler et al., 2001).
In short, "semantic web", in the particular case of TAO, means that instead of defining all the concepts
manipulated by the platform, the developers model only the strict minimum of the concepts (called TAO
basic ontology) and let the users of the platform develop this basic model so that it maps their own reality.
Starting from this idea, most of the efforts were turned to the development of an ontology engine
designed to be generic and versatile. It was named GeneriSSSS (Generic Information System to Shape, Share
and Store the knowledge) or Generis4 (Plichart, Jadoul et al., 2004). It was elaborated on the top of the W3C's
RDFS standard (Brickley and Guha, 2004).
Using this ontology engine, the engineers of the Centre Henri Tudor simply created an ontology dedicated
to the CBA and a few plugins (self-adaptive HTML forms) offering a graphic user interface (GUI) less
generic and more specialized to the assessment purpose; this was called TAO (Latour, Martin et al., 2005).
So, Each concept of the CBA (e.g. a subject, a group of subjects or a test) is articulated inside an action
unit named module. For example, a subjects-module can manage subjects and, in the context of
PRECODEM, an instance of subjects-module was holding and managing the data related to the job seekers.
The architecture of Generis4 is entirely oriented to the web paradigm and technologies, and since TAO
inherits all the capabilities of Generis4, the TAO modules can be spread all around the Internet.
Each installed Generis4 system can manage one or more modules of different types but can also share its
resources with other modules living on other instances of Generis4 anywhere on the Internet. That makes an
expanding peer-to-peer network of collaborative resources, completely decentralized like the web itself.
Generis4 supports different languages as it is using UTF8 character set. It means that the GUI can be
translated in different languages but it also means that the resources — e.g. the items (or questions) and the
tests — managed by the modules can exist in different languages; such a feature enables multi-language tests.
As already mentioned earlier, the modules are able to manage any type of resource and, thus extra details,
for example, the number of months of inactivity of a job seeker, can easily be attached to her or his profile in
the subjects-module; this action not only enriches the profile of the job seeker but also helps to explain,
interpret the collected results available in a results-module after this particular job seeker takes a test. It also
can help experts to infer behaviors and socio-cognitive profiles, in this way enabling the experts to emit
recommendations towards the training contents, the training process, and so on.
In brief, TAO (via Generis4) allows people to collaborate anywhere, in different languages to the building
of an effective CBA program. In addition, this platform, when it is used to some specific extents (e.g. the
management of the employability in the case of the PRECODEM project) can even help to produce tactical
strategies and policies towards the training of specific profiles of subjects.
2.4.2 How did TAO address the goals of the PRECODEM project?
In the “Figure 2. Overall schema of the TAO/Generis4 platform specialized for the PRECODEM project”,
you can see (in the top part of the schema) that Generis4, amongst other ontologies (set of vocabulary and
rules), owns a specialization dedicated to the testing purpose and named TAO. This TAO model has been
instantiated for the purpose of the PRECODEM project. So, in the middle of the figure, you see a
PRECODEM specialization that inherits from the TAO ontology.
IADIS International Conference on Cognition and Exploratory Learning in Digital Age (CELDA 2006)
263

It means that each module of the TAO ontology is adapted for the employment domain; for example, the
Subjects Module (standard construct of a TAO model) is specialized as a Job Seekers module, the Group
module is derived in Occupations (AP/SI) as mentioned in the definition of the PRECODEM project, and so
on. The bottom part of the figure is made of particular instances of the PRECODEM specialization
considered as a model. These are, for example, instances of job seekers (for example, Miss x), items or
questions, tests (e.g. AP_Test_xx delivered on July, the 12th ), of occupations (Group AP, Group SI).

Figure 2. Overall schema of the TAO/Generis4 platform specialized for the PRECODEM project
ISBN: 972-8924-22-4 © 2006 IADIS
264

3. CONCLUSION
Besides the political will to support the life-long trainings initiatives, European societies need to deploy and
experiment the set of appropriate tools in order to sustain the effort of the employability (European
Commission, 2006). Thanks to economical, business, technical watch, such tools will first allow to identify
the competency and profile standards to accurately reflect IT industry requirements.
In the perspective of the implementation of an active and reactive employment policy, the life-long
learning as well as the structured and semi-structured trainings should be promoted, not only inside of the
companies but everywhere. This policy should use effective tools that permit professionals to evaluate the
quality of the trainings and the improvements of job seekers' employability as well as the suitability of the
trainings towards the learners’ profiles (Latour, Martin et al., 2005; Jadoul and Mizohata, 2006).
These tools must also provide a better analysis of the expectations and preferences of the individuals who
want to find or change a job. These tools also should show the competency gaps that have to be filled by job
seekers, and thus help and guide them whatever their initial academic or vocational backgrounds are and
whatever sex and age they have.
ACKNOWLEDGEMENT
We thank all the persons who contributed to this project including the job seekers who volunteered for the
PRECODEM project, the experts who, at the genesis of the project, made the strategic watch to choose the
two occupations (AP and SI) and to elicit the related marketable competencies. We are also grateful to the
people who gave the training sessions: Gaëlle Labarane, Stéphanie Aubertin, Alexandre Baudet, Yannick
Naudet, Christophe Incoul, Damien Nicolas, Evren Bulut, Michael Hamm, Stéphane Plehiers, Jean-François
Merche, Mikael Andres, Stefan Leidner, Décosse Céline, Chrystel Petiteville, Linda Szelest, Hugues Henriot,
Thibaud Latour, and Patrick Plichart (all our apologies for the people we may have forgotten to mention their
names).
REFERENCES
Allen Consulting Group. 2004. Development of a Strategy to Support the Universal Recognition and Recording of
Employability Skills. A Skills Portfolio Approach. Employability skills, final report. Australian Government
Department of Education, Science and Training, Camberra, Australia.
Berners-Lee, et al. 2001. The semantic web: A new form of web that is meaningful to computers will unleash a
revolution of new possibilities. Scientific American 284 (2001)
Bolton, B. 1981. Assessing Employability of Handicapped Person : The Vocational Rehabilitation Perspective. Journal
of Applied Rehabilitation Counseling, 12, (1).
Brickley, D., Guha, R.: Rdf vocabulary description language 1.0: Rdf schema. W3C Recommandation (2004)
Casey, B. 1998. Retraite anticipée ou tardive : incitations et désincitations. Document de travail, OCDE, AWP 3.3F,
projet « Préserver la prospérité dans une société vieillissante ». OCDE, Paris, France.
Dryburgh, H. 2002. L'acquisition de compétences en informatique. Tendances sociales canadiennes, Statistique Canada,
Ministre de l'Industrie, Ottawa, Canada.
European Commission. 2001. Direction générale des Affaires Economiques et Financières. 2001. Economie Européenne :
l’économie de l’Union, bilan 2000. IP/06/618. Communautés européennes, Bruxelles, Belgium.
European Commission. 2002. Evaluation de la mise en oeuvre des lignes directrices pour l'emploi 2002. Document
d'appui pour le rapport conjoint sur l'emploi 2002. COM(2002)
http://ec.europa.eu/employment_social/employment_strategy/report_2002/supp/supp_fr.pdf Bruxelles, Belgium.
European Commission. 2006. Education et formation en Europe : de nouveaux efforts sont nécessaires pour atteindre les
objectifs de Lisbonne. IP/06/618.
http://ec.europa.eu/education/policies/2010/objectives_en.html, Bruxelles, Belgium.
French Ministry of Education. 2006. Prospective emploi-formation à l'horizon 2015. Note d'information No. 06.03
février. Ministère de l'Éducation nationale, Paris, France.
IADIS International Conference on Cognition and Exploratory Learning in Digital Age (CELDA 2006)
265

Jadoul, R. and Mizohata, S. 2006. PRECODEM, un exemple de tao mis au service de l'emploi. Communication orale à
ADMEE 2006. L’évaluation au 21e siècle. Vers de nouvelles formes, modélisations et pratiques de l’évaluation ? 19e
Colloque International de l'ADMEE-Europe, 11-13 septembre 2006. Abbaye de Neumünster, Luxembourg.
Gille, L. and Marti, R. 2000. Les technologies de l'information et des communications et l'emploi en France. Ministère de
l'Economie, des Finances et de l'Industrie Direction Générale de l'Industrie, des Technologies de l'Information et des
Postes. BIPE, Paris, France.
Guillemard, A. 2003. L'âge de l'emploi. Les sociétés à l'épreuve du vieillissement. Armand Colin Paris, France.
INTEREF. 2001. Résultats du Programme d'étude « Gestion des Ressources Humaines Demain », Glossaire. Espace
d'échange des observatoires emploi-formation. Villeneuve d’Ascq, Pollen Conseil, Paris.
Kearsley, G. 1989. Introducing new technology into the workplace : retraining issues and strategies. Background Paper
No. 8b, Department of Labor, Washington, DC., Commission on Workforce Quality and Labor Market Efficiency
(1989), the US Department of Labor, Washington, DC, USA, pp 457-492.
Latour, T. et al. 2005. TAO, a Modular and Versatile Platform for the Collaborative and Distributed Computer-based
Assessment Management. Awarded, Communication de type poster à EDMEDIA 2005 World Conference on
Educational Multimedia, Hypermedia & Telecommunications, 27 June - 2 July 2005. Montréal, Canada.
Martin, R. et al. 2004. A distributed architecture for internet-based computer-assisted testing. Communication orale à
EDMEDIA 2004. World Conference on Educational Multimedia, Hypermedia & Telecommunications at the
University of Lugano, 21-26 juin 2004, Lugano, Suisse.
Martin, R. et al. 2005. Covering Different Levels of Evaluation Needs by an Internet-Based Computer-Assisted Testing
Framework for Collaborative Distributed Test Development and Delivery. Communication à EDMEDIA 2005.
Montréal, Canada.
Plichart, P. et al. 2004. TAO, a Collaborative Distributed Computer-Based Assessment Framework Built on Semantic
Web Standards. Communication at the Advances in Intelligent Systems – Theory and Application AISTA2004,
Luxembourg, Centre Henri Tudor, Luxembourg.
Sen, A. 2000. Development as Freedom. Anchor Books, New York, USA.
Statec. 2006. L’équipement, la formation et les compétences informatiques des Luxembourgeois. Statnews. No. 28,
Statec, Luxembourg.
Szelest, L. and Reiter, S. 2006. PRECODEM - Préparation aux compétences de demain pour les demandeurs d'emploi -
cadres et ingénieurs. Fiche descriptive. Centre Henri Tudor, Luxembourg.
Vandenabeele, L. and Giltay, D. 2005. FormIT. Fiche descriptive. Centre Henri Tudor, Luxembourg.
Wannell, T. and Ali, J. 2002. Travailler intelligemment : le changement technologique influencé par les compétences.
Tendances sociales canadiennes. Statistique Canada, Ministre de l'Industrie, Ottawa, Canada.
Zanardelli, M. and Leduc, K. 2006. Favoriser le vieillissement actif : les pratiques des entreprises en matière de formation
continue pour les travailleurs âgés. Collection Population & Emploi, N° 15/2006. Differdange, CEPS / Instead,
Luxembourg.
ISBN: 972-8924-22-4 © 2006 IADIS
266