Rationale on Data-Driven Learning

“New technologies are the perfect aid to assist teachers in their "need to broaden their scope for creative pedagogical initiatives." (Little et al.)

"In teaching a second language we must design forms of work in which the student's attention shall be directed to the subject matter and away from the form in which it is expressed". (Harold Palmer)

It has often been pointed out that the advent of technology enhanced learning materials requires a re-thinking of the methodological framework of language learning. Furthermore, it is now widely recognized that computer tools can facilitate the implementation of a methodology for language learning that focuses more than in the past on authenticity in contents, context, and task. New technologies may solve a large number of practical problems, particularly in the area of exploiting authentic resources. After all, authenticity in content, task, and classroom interaction is "a crucial issue" in language learning methodology. (cf. van Lier, 1996:123)

This is, in fact, the basic idea of data-driven learning. It can be summed up in the following bullet points:

Before going any further into the background of DDL, we invite those readers more familiar with the German language to visit the GrazVOLL website and look at a text in German describing the êrationale behind data-driven learning.

The ideas underlying data-driven language learning as well as those guiding some of the other sections of the GrazVOLL resource are, in fact, firmly rooted in some of the current guiding paradigms. Most obviously, concepts described as task-based learning (TBL) form a relevant backbone of aspects explored and exemplified in the DDL-section. Let us briefly discuss task-based learning and see how it fits into our concept.

Task-based language learning is an approach which was developed in the 1980's by the Indian language teaching specialist N.S. Prabhu. Although TBL was mainly used in learning environments in which second and not foreign languages are learned (e.g. India), the most important concepts of this approach have been discussed recently within the language teaching profession. Jane Willis, for example, has made the task-based cycle an important part of foreign language classes. There can be no doubt that TBL has some relevance when discussing the exploitation of new technologies for language teaching purposes in general as well as DDL in particular.

TBL is based on the idea that the acquisition of language and linguistic competence as well as language and language learning awareness can best be realised through tasks which encourage the learner not to focus explicitly on the structure and the rules of the new language. Learners will acquire the form of the foreign language because they are engaged in exploring aspects of the target language on the basis of authentic content. The types of activities developed in the course of the workshop series based on êusing concordances are perfect examples of this approach. Again, a similar set of information and accompanying examples dealing with êKonkordanzen is available in German.

The principal aim of all teaching activities must, therefore, be the creation of a learning environment in which learners are asked to carry out authentic tasks. To be able to do so they are equipped with materials which are interesting for them and which motivate them to communicate with each other. TBL methodologists define the learning aims of their approach in a way similar to the communicative approach. But for them competence is developed entirely through meaningful interaction. Although they do not deny the necessity for the learner to acquire grammatical knowledge, they would rather have this knowledge develop subconsciously.

There can be no doubt that TBL, while clearly learner-centred, is seen by its earlier advocates such as Prabhu as strongly teacher-directed. Teachers put the materials together, they prepare the activities, they organise the learning environment. Group work as a student-oriented form of learning is not seen very positively. In this context, Prabhu makes the following very interesting remark: "The principle that interaction between the teacher and the learner, or between a text/task on paper [on a computer screen] and the learner, …” This quote sums up perfectly the principle approach of the DDL-related activities within the GrazVOLL workshop series; in addition, it also hints at the subtle difference between this section of our resource and the other, no doubt equally task-based type of activities dealt with in the sections on êWeb Literacy and êClassroom Extensions.

Consequently, we should make as much use as possible of genuine materials in the language classroom and enable learners to do as many authentic things with such materials as feasible. Genuine materials are, as Widdowson pointed out in in1979, language samples not constructed for the purpose of language learning. (cf. Widdowson, 1979:80) Authentic tasks would then be tasks and learning projects as well as activities of knowledge construction which truly enable learners to explore the target language in it structure and functionality when working with such genuine "texts". David Little describes this approach to authenticity in language learning as creating opportunities for the learner to "psychologically interact" with the target language, "by which we mean the psychological processing of target language input in such a way that it interlocks with and modifies the learner’s existing knowledge." (Little, 1989: 5) He continues by pointing out that such a process is only possible if the learner regards the learning material as having "personal significance"; it can, therefore, be reasonably assumed that authentic texts, preferably taken from sources which relate to the learners’ real world and everyday life, have such a quality.

A short summary of what DDL is all about was put together by participants at the final workshop of the series. So, if readers are looking for a quick answer to the question êWhat is DDL? as well as pointers towards additional information on the internet they are invited to follow this link. Similar information on êFremdsprachenerwerb als data driven learning (DDL) was compiled by participants at the Graz 2002 workshop.

Finally, those who want something on the didactic and methodological principles underlying data-driven learning in a Slavonic language are invited to check êIrina’s input on the part of the GrazVOLL resource created at the Moscow workshop.

In order to enhance the language learner's role as an experimentor and researcher in the classroom, êauthoring software for the creation of tutorial and exploratory exercises as well as cognitive tools, such as êconcordancers, are very important. With such materials, teachers can prepare tailor-made exercises for their classes and put together worksheets and create tasks which permit the learners to actively and often consciously explore the target language. Such tasks and learning projects will help to develop learners’ language awareness and understanding of the structure and functionality of the target language. Providing the appropriate data, organized into suitable units of information and supported by relevant processes must be the foremost task of educational design of any kind of language learning and teaching resource in order to contribute to the success of any learning situation.

It must be taken into consideration, though, that a number of language teachers consider the use of authoring tools either too time-consuming or potentially difficult. However, innovative tools for text and data-processing and for the creation of multimedia hypertexts have now reached a level of sophistication and user-friendliness that the didactic manipulation (as Edelhoff, 1989) puts it, and adaptation of authentic materials for classroom use should soon become daily routine of grassroots teaching, comparable to the use of the use of the blackboard or the photocopier today. In addition, a vast amount of learner relevant materials is now becoming available in digital format, either in the format of local resources or globally accessible data on the worldwide web. In the following we present a few examples of tools and multimedia authoring facilities which we consider as useful to assist teachers in drawing on such resources.

The first step towards integrating more authentic, real-life-based materials into the language curriculum is the use of authoring tools, such as êWIDA Software’s authoring suite or êHot Potatoes. A number of resources for teacher training on the internet provide an in-depth introduction to authoring tools in addition to the information available on the GrazVOLL resource, most notably the section on êauthoring tools of the ICT4LT project.

Such tools provide teachers with ready-made templates for most of the exercise types and interactions commonly used in self-study packages. These templates can then be filled with content and the authoring tool automatically "creates" an interactive exercise using this input. It is important to note that the creation of self-study exercises specifically geared towards a particular target group can be achieved without any knowledge in programming or script-writing. Basic computing skills are sufficient even to be able to integrate multimedia features into exercises.

Throughout the workshop series, participants created sample exercises and put together a list of resources for those who want to find out more about the potential of these tools. The following links provide a first insight into the kind of examples of Hot Potatoes exercises created by participants of ICT in VOLL workshops at an early stage of the series:

More êSites & Exercises built with Hot Potatoes around the world can be accessed directly on the Hot Potatoes Homepage, which also provides a download feature for this free authoring tool. This list is regularly updated and added to.

The easiness of using such tools is probably best reflected by a contribution in German put together by a group of computer novices at the final workshop in Graz 2002, entitled êEin "Beginners" Report (mit Infos zu Hot Potatoes). Further examples are available on the websites created at êMoscow 2001 and êEssen 2002 workshops.

When it comes to finding innovative examples of the use of new technologies in language learning, any tool that allows for the creation of discovery-based and exploratory learning materials must rank very highly within a typology of TELL software. One such tool is concordancing software, originally developed as a device to assist research in corpus linguistics. Such a tool can be used with any textual corpus, i.e. a potentially unlimited number of texts compiled into a database. Its basic function is to extract lists with sample contexts of any word or structure entered into the search option:

Such lists can then be used as a basis for what Tim Johns (1994) refers to as data-driven learning. Considering the example above, a learners' task linked with such a selection of concordances would be for them to deduce themselves the exact difference in meaning, connotation, and grammatical features with regard to the verbs look, see, and watch. Grammatical rules can be acquired in such a discovery-based or exploratory mode, e.g. on the basis of lists with concordances of adverbs, offering learners the opportunity to discover rather than to be taught a rule concerning adverbs and word order in English sentences. Tim Johns provides a complete website with samples and links on data-driven learning (êhttp://web.bham.ac.uk/johnstf/timconc.htm). In addition, Chris Tribble (1997) has published an interesting manual on "Using Concordances in the Classroom".

A detailed description of concordancing and its principles is available on the GrazVOLL website in the sections on data-driven learning. The latest is the page on êCONCORDANCING: information – practice – resources compiled at the final workshop.

A similar resource for êKONKORDANZEN and their use in German is also available on the GrazVOLL website.

When looking at the examples quoted in Tribble's publication or on êTim Johns' website, it becomes apparent that the use of concordancing tools in language learning comes very close to putting into practice some of the theoretical principles discussed at the outset of this paper. Grammatical rules and the meanings of words are not simply learned but constructed by the learners themselves, thus enabling learners to develop language awareness in addition to a structural knowledge of sets of meanings.

As far as tools for data and information processing and êdata-driven learning are concerned, concordancing tools are probably the most widely usable tools if this kind used in language learning. These tools represent a special kind of application as their use does not necessarily require the use of computers with the learners themselves. Rather, it is more feasible to create innovative worksheets to be discussed in class with the use of computers rather than have learners, particularly at lower levels, use the software hands-on. Both uses have been described in great detail in a book on the subject of concordancers in language learning by Chris Tribble [Tribble, Chris (1989): Using concordances in the language classroom. Longman]

To make the use of concordances clearer to the user of the GrazVOLL resource, we have placed a number of sample exercises and concordances on the websites. These were developed at the various workshops by participants and can, for example, be found at the following locations:

In addition, participants have put together links into the major resources concerning concordancing available on the internet. These include links into online concordancers, access to online corpora and texts as well as links into further sample tasks making use of concordance lists. Check out the following link to access this information:

Participants at the various workshops of our series worked in great detail with concordancing tools and continued to do so in between workshops. It is felt that quoting a few Comments on the use of concordancing software & data-driven learning from participants at Workshop 8/2000 might provide further motivation to explore the potential of this aspect of using authentic materials in the context of more authentic tasks in the language classroom. Here are some of the comments:

“Just a week ago when I was preparing for my English classes, I was trying to do the same thing that this software can do so much more effectively than me. I was looking up into a number of dictionaries for differences between angry, cross and furious, or security and safety, trip, journey and travel, etc. writing down all the different definitions, then pouring all this information into my students.

And now, with the help of Concordancing, I can prepare loooots of examples on vocabulary/grammmar,etc features, so that students themselves can make the difference. I can prepare exercises to check how they learned the material.

Of course, there is some technical magic that teacher has to learn to be able to produce her worksheets, but it's not so complicated.

Conclusion: dear colleagues do learn to use Concordancing. Satisfaction guaranteed both for you and your students!"

2) “Concordance is a very powerful tool for the enhancement of language
teaching and learning.”

“The advantages are obvious. It can help teachers make a lot of practice on words that are difficult to be understood, on grammatical meanings that are confused. And most important, it is fast and easy to use. However I can see some disadvantages. The words are out of context and the students may find difficulties in understanding the new words they face. The teacher can’t just give the concordance list without careful consideration of what he has in front of him.”

3) “Concordancers are the tools I have been looking for since I started
teaching many years ago.”

“They take you a lot of time to prepare your own text base, but after that they save a lot of your precious energy.”

Finally, in addition to the materials contained on our own webpage, we would like to draw the readers’ attention to a few resources which became available only after the completion of the ICT in VOLL workshop series.

This site contains DDL materials and teaching units designed to draw the learners' attention to certain vocabulary and linguistics features by providing the students with concordance data and guiding them to make a generalisation. All the items presented in his DDL teaching units were taken from a sub-component of the Bank of English Corpus, known as 'CobuildDirect', using its retrieving software 'Lookup'. References into this corpus und the software are available on the GrazVOLL website at

A second and equally important aspect of data-driven learning is presented on a website dedicated to êresource-assisted learning. This slight shift in terminology emphasises quite nicely the special focus of the approach described and elaborated in this section of the GrazVOLL website and this accompanying publication. In its own words, the site describes the approach as follows:

« Resource-assisted learning is a strategy and model for virtual learning via the WWW which involves the integration of vocabulary, concordancing, dictionary reference and multimedia to provide on-demand support for the learner. Pronunciation, dictionary explanation and concordancing examples are available as needed, providing a learning environment in which the computer acts as a sort of “expert reading partner and linguistic consultant” (Cobb, Greaves & Horst, 2000)

This is a concept of integrating local as well as online resources into scenarios for language learning with the clear aim of exploiting the potential of ICT to provide a platform or tool for ensuring more flexibility in the content and organisation of learning. A further aspect is the creation of more authenticity in content and task for language learning. This is where data-driven learning and all the other aspects dealt with within the ICT in VOLL workshop series are working towards identical aims. Let us briefly summarise these in conclusion to this section.

The ultimate aim of teaching and learning in this "knowledge society" is to assist learners in their need to develop strategies of knowledge processing. Therefore, the traditional transmission model of learning must be replaced by models which emphasise information processing and knowledge construction as acts of learning most suited to the acquisition of the kind of skills needed for the knowledge society. Education and teaching in the knowledge society can no longer be reduced to "the act, process, or art of imparting knowledge and skill" as Roget’s Thesaurus proposes, but learning must be recognised as an act in which a learner plays the role of an active constructor of knowledge. Criteria based on such principles need to be considered when evaluating the effectiveness and value of technology enhanced materials for language learning. The question remains, however, as to how the principle of "learning without being taught" as proposed by Piaget (cf. Papert, 1980: 7) can be put into practice and integrated into a technology enhanced learning environment of the future.

This is where data-driven learning, resource-assisted learning as well as the use of the internet as an extension of the classroom and the acquisition of web literacy come together. All this is taking place in a context, where, over the past decade, language learning theory has seen a shift from a highly guided to a more open learning environment, with constructivism as a new and very much learner-centred paradigm for learning. Learning is now perceived as a self-structured and self-motivated process of knowledge construction and the learner is regarded as a self-governed creator of knowledge.

In addition to the undeniable need to achieve instructional goals, the development of cognitive and strategic abilities suitable for the knowledge society is defined as one of the principle aims of a learning process based on knowledge construction and discovery learning. As far as new technologies and their use in language learning are concerned, the ICT in VOLL project and its section dealing with data-driven learning has shown that technology enhanced materials do have a lot of potential to assist the process of innovation which is needed in this fields as much as in any other area of education.