Home Education Learning Analytics as a Competitive Differential

Learning Analytics as a Competitive Differential

564
0
SHARE

Ronaldo Mota – Chancellor of Estácio Group, educational executive, lecturer, member of the board of the Brazilian Association of Owners of Private Institutions of Higher Education, he writes about new technologies in education and innovative educational methodologies.

There was a period when the adoption of more competitive management models alone was sufficient for one educational institution to achieve high results. This was not simple, it was innovative and generated significant results. However, over time, management models have been limited, on the one hand, and transferable and replicable, on the other. The future scenario brings even more complex challenges. Among them is the fact that institutions that know how to appropriately incorporate new technologies and innovative methodologies will be those that will stand out and will be rewarded with the opportunity to combine, with sustainability, scale and quality. Among the technologies with the greatest potential for achievement, in terms of academic results, I highlight Learning Analytics.

Learning Analytics is the methodology that enables educators to make decisions by taking into account systematic and elaborate analyses of learners’ data and the educational contexts in which learning develops. By analyzing the data on how much and how students are learning, a sharper perception of educational realities is possible. Such procedures enable appropriate learning designs (“Learning Designs”) as well as strategies and learning paths to be implemented. At the same time, this methodology contributes to the selection of which resources, including technological and content delivery modes, are the most appropriate for each context and, in the limit, for each student.

In fact, teachers in traditional teaching routinely use data in teaching processes. However, they usually do so in a limited and preliminary version, precursor to what we now call Learning Analytics. For example, final grades, resulting from a few products, have relevant consequences, such as approving students or not. Exceptionally, more dedicated teachers are able, through their sensibilities, to perceive the peculiarities of a student group, to identify typical needs and to change procedures, but these are rare cases and in small in scale. In general, according to the available data, some academic achievement of the students are insufficient to motivate and guide changes in educational pathways. Learning Analytics, in theory, allows and stimulates adaptations, improving learning as it reconfigures, in a timely manner, the educational processes, adapting them to the specific realities and, whenever possible, to the characteristics of each of the actors involved.

In sharp contrast to the few data available until very recently (basically individual test notes), thanks to digital technologies, today we have an abundance of information (“big data”), which allows us to try to understand, in a new and innovative way, complex educational realities. In addition to the usual forms of evaluation, we can explore, among countless other possibilities, data resulting from:

  1. i) level and speed of information assimilation,
  2. ii) the student’s ability to access content and its autonomy in the use of knowledge,
  3. iii) characteristics of “wrong” answers in multiple choice tests,
  4. iv) communication skills through ability to interpret and write complex texts,
  5. v) ability of team collaboration, perceiving and combining frailties and potentialities of each member,
  6. vi) productivity and effectiveness in the manufacture of artifacts,
  7. (vii) competence in solving problems and performing missions,
  8. viii) social-emotional attitudes and behavior in the face of complex challenges,
  9. ix) mathematical literacy, and
  10. x) individualized adaptation to different modes of content delivery.

As an example, let’s look in particular at item iii) above, regarding how we can make use of the non-correct answers (in Learning Analytics they are as relevant as the correct ones) to improve learning. In a simple way, suppose that in a certain multiple-choice test item the correct answer is b). It is possible for a statement, especially designed, that seeks to identify the learner who, even if he dominates the basic concepts involved, because he reads without due attention, choose a). In this case, it is a potential indication that it is a student whose ability to focus deserves our full attention. Now, let us suppose that, although he has studied and understood the subject matter, the student, even if concentrated, has difficulty interpreting slightly more complex texts. It is possible to construct a statement that, probably as a result of this deficiency, is guided to alternative c). in the same way, imagine someone who has weaknesses in mathematical literacy; thus, even with eventual mastery of the content, a mathematical operation in which he/she has difficulty, takes it to alternative d). Finally, a specially prepared statement may try to identify, on a case-by-case basis, more than a fault in the general content domain, we may be deficient in a specifically particular concept involved and, if this is the case, he/she will probably opt for alternative e).

The above recipe is purely illustrative and there is not the slightest chance, from a single item or a single test, to be able to say anything substantive. However, with abundant questions and answers, and with the additional element in the broader set of qualifying data cited above, there are certainly relevant educational features that can be drawn. The great secret is that, from an immense amount of information, some more well-grounded considerations can begin to be weighed on the specific educational contexts and on each individual student.

Once we have more consolidated and properly analyzed material, we can begin to build multiple and personalized educational paths. These are individual tracks that tell or attempt to account for:

  1. i) improved focus,
  2. ii) improved skills in reading and writing more complex texts,
  3. iii) guidance and enabling of basic mathematical operations,
  4. iv) exploring a particular concept on which the student demonstrated fragility, etc.

Curiously, the more students, the greater the number of tests and the more analysts and curators we have, the better elaborated will be the specific paths and approaches that we can propose. For those who have always associated quality with a few and poor quality to many, we have a new paradigm: the scale that generates quality.

We began, therefore, to build intelligent educational algorithms that allow us to get out of the handicrafts and other limitations that characterize traditional teaching, to give qualified answers to the demands and to provide large-scale care. This is a mark of a contemporary education where everyone learns, learns all the time and each one in a unique way. This is the path of a flexible, hybrid, adaptive and personalized education.

In particular, educational businesses and their educational institutions must highlight that the challenges in search of sustainability and profitability, of course, depend on good management models. However, while warfare may indeed be lost in management, on the other hand, in order to have substantive achievements one has to explore innovative methodologies that make appropriate use of new technologies. In this case, there are no miracle solutions or ready recipes. But, of course, Learning Analytics is an indispensable ingredient for successfully addressing contemporary educational challenges.