Content uploaded by Martin Ebner
Author content
All content in this area was uploaded by Martin Ebner on Jul 15, 2020
Content may be subject to copyright.
Draft – finally published in: Burazer, M., Ebner, M. & Ebner, M. (2020). Implementation of Interactive Learning Objects for
German Language Acquisition in Primary School based on Learning Analytics Measurements. In Proceedings of EdMedia +
Innovate Learning (pp. 672-679). Online, The Netherlands: Association for the Advancement of Computing in Education
(AACE).
Implementation of Interactive Learning Objects for German Language
Acquisition in Primary School based on Learning Analytics Measurements
Marko Burazer
Educational Technology, Graz University of Technology, Austria
mburazer@student.tugraz.at
Markus Ebner
Educational Technology, Graz University of Technology, Austria
markus.ebner@tugraz.at
Martin Ebner
Educational Technology, Graz University of Technology, Austria
martin.ebner@tugraz.at
Abstract: Obviously, reading and writing are important qualities nowadays, likely more so than ever
before. Whether that be in school, work or everyday life, it is a skill set that is omnipresent. This is
also evident by the countless contributions that are created and published on various online platforms
such as Facebook, Twitter, YouTube or WhatsApp. In order to avoid being misunderstood, it is
crucial to have the ability to express one’s written thoughts in a structured and error-free manner. To
help children in the early age with their spelling skills, the IDeRBlog platform provides a possibility
to reach their goals and support their German spelling learning process. On this platform children
can create own blog entries which are then corrected by teachers and an intelligent dictionary before
they can finally publish it. Mistakes made by the kids are evaluated and on basis of these mistakes,
exercises can be recommended so that the kids can improve their spelling. This paper will present
these exercises (also called learning objects), which should help children to practice writing, reading
and also listening carefully. It focuses not only on the evaluation setup and process but also results
will be explained in the end.
Introduction
Mastering orthography is more important than ever in today’s world. More and more people are
communicating online through various applications; Facebook, Twitter, YouTube, WhatsApp, and many more. It is
necessary to learn how to spell since nobody wants to be publicly criticized or even laughed at for spelling mistakes
(Aspalter, Edtstadler, and Martich 2017). According to (Lehky, 2016), correct spelling is important and should not
to be underestimated in the workforce nowadays, considering how spelling errors can be the cause for not being
hired for a position. The digital age allows almost unlimited possibilities for learning more productively: On a PC or
on mobile devices, individually or together in a group, at home, on the way to work or at school.
This paper gives also an introduction to the IDeRBlog platform, where young children have the opportunity
to create their own texts which might be corrected by teachers and an intelligent dictionary before publishing them.
It also gives a short overview of the current state of art and provides answers to not only the question what learning
objects are and what can be done with them but gives also a short insight about learning analytics.
IDeRBlog
IDeRBlog is a German acronym for “Individuell Differenziert Richtig schreiben mit Blogs“ (Aspalter,
Edtstadler, and Martich, 2017), which means: „Individually differentiated spelling with blogs“ (Ebner, Edstadler,
and Ebner, 2018). It is part of an Erasmus+ project and consists of a team of three countries: Germany, Belgium and
Austria (Aspalter, Edtstadler, and Martich, 2017, p.13). As mentioned by (Gros et al., 2015) writing, reading and
calculating are not only important social qualities, but also a requirement in the work force nowadays. Writing can
be split into three following parts:
Draft – finally published in: Burazer, M., Ebner, M. & Ebner, M. (2020). Implementation of Interactive Learning Objects for
German Language Acquisition in Primary School based on Learning Analytics Measurements. In Proceedings of EdMedia +
Innovate Learning (pp. 672-679). Online, The Netherlands: Association for the Advancement of Computing in Education
(AACE).
1. having writing skills
2. creating texts
3. writing correctly
All of the mentioned parts are acknowledged on the IDeRBlog website
1
. On the platform, the children are
given an opportunity where they can create and publish their own texts in the form of blog entries. The resulting
blogs cannot just be created on computers, but also on tablets and other mobile devices. Pupils can also create class
diaries. As a result, the motivation for writing increases (Gros et al., 2015).
Learning Objects
Wayne Hodgins already mentioned this topic in the year 1994 (Polsani, 2006). Many different persons and
organizations offer several definitions on what learning objects really are. The committee, 2019 describes the
learning object metadata standard as: “Learning Objects are defined here as any entity, digital or nondigital, which
can be used, re-used or referenced during technology supported learning.” They also mention examples of
technology supported learning such as computer-based training systems, interactive learning environments,
intelligent computer-aided instruction systems, etc. and examples of learning objects include multimedia content,
instructional content, learning objectives and so on.
The author of the book The Instructional Use of Learning Objects (Wiley et al., 2002) describes that the
idea of learning objects is based on the object-oriented development in computer science. Learning objects should be
seen as digital entities which are supplied only over the internet, because small objects can be easily created and
then put together, and also reused, to form a much larger and more complex system. The use of learning objects is
also not limited to a specific place at a certain time like traditional instructional media (e.g. Overhead, video tape,
etc.). It can be used all over the word by several people concurrently at the same time. An important difference is
that it allows people to benefit immediately if there are newer versions of learning objects, which also improves the
existing teaching aids a lot.
Learning Analytics
The term learning analytics can be traced back to the twentieth century, to the 1970’s even; a time before
online learning and big data have become an essential part of today’s world (Ferguson, 2012). The evolution of
learning analytics is speeded up by three elements:
1. Big Data
2. Online Learning
3. Political Concerns
Big Data
This term was first used by John Mashey in the 1990s (Kitchin and McArdle, 2016). In 2001 Doug Laney
described it according to the following three features: volume, velocity and variety, which are also known as the
three Vs and show clearly that big data is not only about data volume (Russom et al., 2011).
Online Learning
According to Ferguson, 2012 learning online does not only have benefits. One of the biggest obstacles is
that online learning isolates students, because they do not have physical proximity to their teachers. Students may
get demotivated easily due to technical problems or disorientation in the online world.
Political Concerns
1
http://iderblog.eu/ (last visited on January 08, 2020)
Draft – finally published in: Burazer, M., Ebner, M. & Ebner, M. (2020). Implementation of Interactive Learning Objects for
German Language Acquisition in Primary School based on Learning Analytics Measurements. In Proceedings of EdMedia +
Innovate Learning (pp. 672-679). Online, The Netherlands: Association for the Advancement of Computing in Education
(AACE).
Every economy in this world needs a well-educated population, therefore many countries are reforming
their educational systems (Campbell and Oblinger, 2007). Ferguson, 2012 sees the third driver for speeding up the
evolution of learning analytics as a political/economic challenge.
Implementation of Interactive Learning Objects
Five exercises have been created for the IDeRBlog platform in total which are split into two different
categories: “standard exercises” and “database-oriented exercises”.
1. Einsetzen (insert)
2. Fehlerstellen (fault locations)
3. Merkwörter (cue words)
4. Zuordnen (assign)
5. Glücksrad (wheel of fortune)
The first and second exercise are part of the standard exercises, the other three are database-oriented
exercises. This means that the standard exercises are general exercises, which do not focus on user specific errors.
The database-oriented exercises focus on single words, which are provided by IDeRBlog’s intelligent dictionary. As
one also can see, the names of all the exercises are German words, which is due to the fact that the IDeRBlog
exercises are about German language acquisition.
Exercise - Einsetzen (Insert)
This is a typical cloze text exercise where students are presented with several sentences with gaps. Gaps
have to be filled in with correct words or letters. It is unrelated to the specific spelling mistakes a user makes.
Several sentences can be displayed at once for this exercise. Depending on the settings, which can be managed by an
administrator, or a teacher with rights to edit the exercise settings, there is also a tooltip for every gap (see Figure 1).
Figure 1. Einsetzen (insert) exercise with cloze text example.
Exercise – Fehlerstellen (Fault Locations)
Comparable to the first exercise when looking at its basic structure, it also does not entail user-specific
spelling mistakes and contains several sentences with errors too. Fault locations must be located in this exercise and
each sentence can contain one or more errors. Errors can be found by clicking on a word. If a fault location was
identified correctly by the user, a small pop-up dialog appears and the user has to justify the decision (see example
in Figure 2). Justifying a decision is important, because this is how the teaching staff can verify whether a student
completed the exercise based on their own knowledge or by cheating.
Draft – finally published in: Burazer, M., Ebner, M. & Ebner, M. (2020). Implementation of Interactive Learning Objects for
German Language Acquisition in Primary School based on Learning Analytics Measurements. In Proceedings of EdMedia +
Innovate Learning (pp. 672-679). Online, The Netherlands: Association for the Advancement of Computing in Education
(AACE).
Figure 2. Example when user has to justify decision in the Fehlerstellen (fault locations) exercise.
Exercise – Merkwörter (cue words)
Unlike the previous two exercises, the main focus for the Merkwörter (cue words) exercises are the user-
specific spelling errors. The focus is not on sentences but single words only. First a word is displayed and the user
then has to take a closer look at that word and think about possible problems that could arise when using it. Single
letters can also be clicked on to mark those problems (see Figure 3). An example could be that a word is always
written in lowercase instead of uppercase. But also, specific letters of a word can be forgotten or swapped with
another letter. This exercise does not only help the student, it is also beneficial for teachers to know the weaknesses
of their students.
Figure 3. Example for marking letters/graphemes of a word in the Merkwörter (cue words) exercise.
Exercise – Zuordnen (assign)
This exercise is split into two variants: Drag&Drop and Writing. Also, in this exercise the user is presented
a word which has to be analyzed first and then be assigned to a certain category. In the Drag&Drop variant (see
Figure 4) the word is displayed together with all possible categories which the user then has to assign to the correct
category (or categories). But in the Writing variant (see Figure 5) the word is not visible; the user can listen to it by
clicking on a speaker icon. Through Text-To-Speech technology, the word is the spoken by the computer.
Figure 4. Example of the Zuordnen (assign) exercise with Drag&Drop variant.
Draft – finally published in: Burazer, M., Ebner, M. & Ebner, M. (2020). Implementation of Interactive Learning Objects for
German Language Acquisition in Primary School based on Learning Analytics Measurements. In Proceedings of EdMedia +
Innovate Learning (pp. 672-679). Online, The Netherlands: Association for the Advancement of Computing in Education
(AACE).
Figure 5. Example of the Zuordnen (assign) exercise with Writing variant.
Exercise – Glücksrad (Wheel of Fortune)
The last exercise is very similar to the Merkwörter (cue words) exercise but with the difference that the
user doesn’t see the word. Just like in the Zuordnen (assign) exercise, it is spoken to the user with Text-To-Speech
technology by clicking on a speaker icon (see Figure 6).
Figure 6. Example of the Glücksrad (wheel of fortune) exercise
Evaluation
The evaluation took place in an Austrian elementary school called “Praxisvolksschule der Pädagogischen
Hochschule Steiermark
2
”. The participants were children of a 3rd grade school class with 21 pupils in total with a
distribution of eleven girls and ten boys. There were no distinctions made between the sexes. In total there were only
15 computers available in the computer room of the school. Therefore, some students had to work together in groups
of two. All exercises have been evaluated in the Mozilla Firefox browser with the exception of the Zuordnen
(assign) exercise (the Drag&Drop variant) which was evaluated in the Microsoft Edge browser, because of some
compatibility issues. To test the different levels of difficulty two instances have been prepared with different settings
of the Merkwörter (cue words) exercise. On the one hand an easy mode without any time limitations and therefore
no pressure for the students and on the other hand a hard mode with 20 seconds time for reading and 40 seconds
time for writing.
All exercises needed to be done in the narrow time window of 1.5 hours to ensure a smooth evaluation
process. Also, because of an odd number of computers, it was agreed upon to run most exercises on three computers
simultaneously and one, the Merkwörter (cue words), on four. After completing each exercise, the children had to
move to another device, because it would have taken too much time to explain to them how to switch to another
exercise on every single computer.
Finding out how the exercises have been received by the students was achieved with a questionnaire which
was prepared in advance. It had to be completed after each exercise and contained following five questions:
1. I knew what the task was.
Ich wusste, was die Aufgabenstellung war.
2. I could solve the task myself.
2
https://pvs.phst.at/ (last visited on January 08 2020)
Draft – finally published in: Burazer, M., Ebner, M. & Ebner, M. (2020). Implementation of Interactive Learning Objects for
German Language Acquisition in Primary School based on Learning Analytics Measurements. In Proceedings of EdMedia +
Innovate Learning (pp. 672-679). Online, The Netherlands: Association for the Advancement of Computing in Education
(AACE).
Ich konnte die Aufgabe selber lösen.
3. It was fun to practice.
Es hat mir Spaß gemacht zu üben.
4. I also want to do the exercise at home.
Ich möchte die Übung auch Zuhause machen.
Each of these questions (also referred to as statements in the following chapter) should have been rated
from one to five after completing each exercise. Smileys were also allowed to be used for rating (see Figure 7).
Figure 7. Smileys used for ratings (Agarwal, 2019)
Results
A summary from all questionnaires can be seen in Figure 8 which shows the differences for all exercises.
As one can see, the Zuordnen (assign) exercise was the most fun for all kids who all rated this exercise with the
grade 1.00. The least fun was the Merkwörter (cue words) exercise with the hard mode, where kids had a time
limitation of 20 seconds reading and 40 seconds for writing and was rated with an average of 1.89.
Figure 8. Results of the questionnaire
During the evaluation in the class room, it became clear that usability is the key factor for the success of
these implemented learning objects. Kids perform better and enjoy an exercise when they know exactly what they
have to do to complete a task.
Einsetzen (insert)
The most mistakes were done because of an inaccurate description and lots of typing errors. Some children
thought that it was about a specific grammar exercise which can be seen in the analyzed data where a mistake was
for example “Bieren” (beers) instead of “Bären” (bears).
Draft – finally published in: Burazer, M., Ebner, M. & Ebner, M. (2020). Implementation of Interactive Learning Objects for
German Language Acquisition in Primary School based on Learning Analytics Measurements. In Proceedings of EdMedia +
Innovate Learning (pp. 672-679). Online, The Netherlands: Association for the Advancement of Computing in Education
(AACE).
Typical typing errors can also be seen in the data. Some kids knew how to write a word but added a white
space character or entered a character twice where it should not be. Someone also used the past tense instead of
present tense. Most of these errors are understandable and shows that the description should have been more
detailed.
Fehlerstellen (fault locations)
For most kids it was not clear enough what the purpose of this exercise was and whether all mistakes were
found to complete the task. Some thought there was one mistake per sentence and were surprised when they realized
there are not finished yet, although there was a message at the top of the page hinting how many mistakes are left to
be found.
Furthermore, from the data it also can be seen that there were kids who didn’t take this exercise serious
enough (or struggled very hard with this at least). This is because a few kids clicked on every single word in this
exercise until all errors were found. For example, the word “Blüte” (blossom) was selected 21 times, 20 of those
being from the same user.
Merkwörter - Leicht (cue words - easy)
Although it was not clear enough what to do in the beginning, there were only a few typical typing errors in
this exercise. Most kids seem to enjoy it and some are also willing to do this exercise at home again. It was received
quite well by the students. This is probably because of the simplicity behind the task and no time pressure.
Merkwörter - Schwer (cue words - hard)
One can read in the questionnaires that this exercise was not been received as well as the others. In both
exercises (the easy mode and hard mode) spelling errors and typos were made. The number of errors differs
remarkably with an average error rate of 4.29 out of 6 words (71.5%). It shows clearly how difficult this exercise
was for the children. Obviously, 40 seconds for writing was too short because many input fields were left blank or
words were not spelled out completely. Not a single session was error free.
Zuordnen (assign) - Drag & Drop
They key factor for the success of this exercise was probably simplicity and user interaction. The exercise
was very well received which also had an average error rate of 2.11 out of 10 total words (21.1%).
Summary and Future Work
On the evaluation day it became clear that comprehensive descriptions are an essential part for certain
exercises, considering how some children didn’t know immediately what the given task for each exercise was. This
can be also seen in Figure 8 where some children stated that they didn’t need help although they did not know what
the task was in the beginning. Perhaps the motivation for doing the exercise at home would also be higher if the task
had been clear from the beginning. Thorough feedback is also important during an exercise, as can be seen in the
Fehlerstellen (fault locations) exercise where some kids wondered why the exercise was not finished yet, although
they found in every sentence an error.
Something, that couldn’t be seen during the evaluation in the class room but is clear through analysis of the
logged data, can be seen for the Merkwörter (cue words) exercise. For testing the different difficulty levels, a
version with a time limit and another without time limit was created in advance. After analyzing the data, it became
clear that choosing the right timing is difficult. The kids had more troubles completing all tasks when there was a
time limitation. This is probably because children, at this age, are not used to the keyboard which might be the
reason for the higher amount of errors.
Draft – finally published in: Burazer, M., Ebner, M. & Ebner, M. (2020). Implementation of Interactive Learning Objects for
German Language Acquisition in Primary School based on Learning Analytics Measurements. In Proceedings of EdMedia +
Innovate Learning (pp. 672-679). Online, The Netherlands: Association for the Advancement of Computing in Education
(AACE).
Some kids demanded more interactive tasks. One child stated during the evaluation that the Fehlerstellen
(fault locations) exercise was too “boring”. They would like to do a task where they can type or click on something.
For the future all implemented learning objects can be improved a lot, especially regarding the usability.
More immediate feedback duration and also a small introduction tutorial might be helpful. Feedback messages could
be more positive, instead of displaying how many mistakes were found the users should be told how well they are
progressing so far. This might motivate them to continue and do even better in the future.
Since these learning objects were made for browsers, making them available on mobile devices might also
be of interest for potential users. Expanding the ways in which these exercises could be accessed, the desire and
curiosity for learning at home on one’s smartphone or tablet might be increased. The analyzed user data can then be
used to also help teachers to create more exercises which can be added to this web application.
References
Aspalter, Christian, Konstanze Edtstadler, and Susanne Martich (2017). “Recht-/schreibdidaktische Überlegungen
für den Einsatz der Internetplattform IDeRBlog.” In: R&ESOURCE. Bd, pp. 1–15.
Campbell, John P and Diana G Oblinger (2007). “Academic analytics.” In: EDUCAUSE.
Committee, IEEE Learning Technology Standards (2019). The Learning Object Metadata standard. Retrieved
February 11, 2009, from IEEE Learning Technology Standards Committee Web site. URL:
https://www.ieeeltsc.org/working-groups/wg12LOM/lomDescription (visited on 05/02/2019).
Ebner, Markus, Konstanze Edtstadler, and Martin Ebner (2018). “Tutoring writing spelling skills within a web-
based platform for children.” In: Universal Access in the Information Society 17.2, pp. 305–323. ISSN: 1615- 5297.
DOI: 10.1007/s10209- 017- 0564- 6. URL: https://doi.org/10. 1007/s10209-017-0564-6.
Ferguson, Rebecca (2012). “Learning analytics: drivers, developments and challenges.” In: International Journal of
Technology Enhanced Learning 4.5/6, pp. 304–317.
Gros, Michael et al. (2015). “Schreiben – Rechtschreiben lernen und Lesen mit der Plattform Individuell
Differenziert Rechtschreiben mit Blogs – kurz IDeRBlog (www.iderblog.eu).” In: LA-Multimedia 2015, pp. 22–24.
Kitchin, Rob and Gavin McArdle (2016). “What makes Big Data, Big Data? Exploring the ontological
characteristics of 26 datasets.” In: Big Data & Society 3.1, p. 2053951716631130.
Lehky, Andrea (2016). Bewerbungsfehler, die den Job kosten können. URL:
https://diepresse.com/home/karriere/5112591/Bewerbungsfehler- dieden-Job-kosten-koennen (visited on
08/26/2019).
Polsani, Pithamber (2006). “Use and Abuse of Reusable Learning Objects.” In: Journal of Digital Information 3.4.
ISSN: 1368-7506. url: https://journals.tdl.org/jodi/index.php/jodi/article/view/89.
Russom, Philip et al. (2011). “Big data analytics.” In: TDWI best practices report, fourth quarter 19.4, pp. 1–34.
Wiley, David A et al. (2002). The instructional use of learning objects. Vol. 1. Agency for instructional technology
Bloomington, IN.
