The advantage of Personal Learning Environments (PLEs) is to empower a learner in taking control over his/her own learning process. The shift from just being controlled by a teacher towards taking control by oneself in a selfregulated learning (SRL) way can be basically initialised by providing learning environments that can be personalised and individually adapted or created instead of using 'one size fits all' learning environments. A lot of research and development on this subject has been done in the EU-Project ROLE (role-project.eu). In this context extensive experiments have been conducted with widget-based PLEs. Scenarios have been created, implemented, tested and evaluated in real world settings. The contribution of this paper is the presentation of a) three widget-based PLE scenarios, b) evaluation results on comparing the value of the presented PLE scenarios and c) evaluation results on comparing students and teachers point of views against the presented PLE scenarios including SRL aspects.
INTRODUCTION
Responsive Open Learning Environments (ROLE) are based on the idea of Personal Learning Environments (PLEs) by exploiting Cloud Computing Technology (examples are presented in chapter III). Instead of using traditional learning environments which provide tools and content by one single provider and are often owned by one specific educational organization ROLE exploits all existing and developing open educational sources including all popular Web2.0 resources such as Wikipedia, YouTube or Flickr. Historically the idea of PLEs is based on the fact that most learning takes place informally, in different contexts and scenarios, and that content is not provided by one single provider. Following this idea ROLE provides a framework essentially consisting of “enabler spaces” on the one hand and tools, content, services on the other hand [1]. Using this equipment everyone is invited to individually create his/her PLE. In PLE research it is seen as essential to have a learner challenged by offering him/her to create their individually controlled and preferred learning environment in order to trigger and motivate more selfregulated learning. Moreover this approach has the potential to enable and facilitate both informal and formal learning.
The paper presents three PLE scenarios which have been developed in the ROLE project. In real world testbeds learners are confronted with new ways of learning by working with the provided PLE scenarios. While the use of any PLE should trigger self-regulated learning it is especially the third and last PLE scenario which has been implemented a consequent mechanism to support SRL.
This paper investigates the attitudes and reasons for acceptance of PLE technology by students and teachers.
II.
In the field of self-regulated learning (SRL) research it is often pointed to the important role of learners’ strategic use of cognitive and metacognitive strategies to regulate their learning [2], [3], [4]. Still many learners show difficulties in applying concrete metacognitive strategies such as planning, goal setting, monitoring, evaluating and as a result perform less successful [5]. For this reason, much work has been focused on the assessment of students’ SRL strategies to support the learning behaviour accordingly. This work is usually bound to highly controlled learning environment such as intelligent (tutoring) systems [5], [6], [7]. However, understanding, scaffolding or/and facilitating students’ SRL skills is especially important in (responsive) open learning environments. In such open environments goals are less clear and obvious; therefore students might not necessarily be able to predict the outcome of the learning activity or the optimal learning path.
Nevertheless, it could be found that PLEs provide opportunities to enhance SRL skills, especially metacognitive skills, but learners need additional help and guidance [5] during the learning process. In this regard the concept of freedom and guidance comes into play. The concept of freedom and guidance is important, because highly motivated learners attain a better learning performance if they have more control over their learning, but lower motivated learners attain better learning performance if they get more guidance [8]. Issing noted that this is also applicable to hypermedia learning environments.
In this regard it should be envisioned to develop services and learning environments that can be adapted to the individually degree of guidance and freedom according to the learner’s needs and therefore offer the learner an optimal balanced level of control and responsibility for his or her learning environment [9].
B.
In PLEs learners are in the position to create their own
learning environment and shape it to their personal needs
and learning objectives. In order to provide support in such
an open learning approach an underlying and
psychopedagogical sound model which represents the theoretical
backbone of open environment learning has been defined,
the Self-Regulated Learning Process Model (SRL PM).
The SRL PM builds on the cyclic self-regulated learning
model proposed by Zimmerman [
This extension leads to the four phase SRL PM including the phases of: (1) learner profile information is defined or revised, (2) learner finds and selects learning resources, (3) learner works on selected resources, and (4) learner reflects and reacts on learning strategies, achievements and usefulness (see Figure 1) [9]. According to this model, especially meta-cognitive activities are supported by focusing on the recommendation of learning activities which can be performed through the usage of learning resources and therefore enhance self-regulated learning.
ROLE services such as the Mash-Up Recommender
Widget (see Figure 7) offer guidance and help learners by
presenting recommendations and according explanations,
without limiting the degree of freedom, as the learner can
freely choose between the recommendations made by
ROLE services or other alternatives. This concept is
based on an ontology that builds on a connection of
learning phases of a SRL PM to learning strategies,
techniques and activities [
This chapter describes three widget-based PLE scenarios which were evaluated (see chapter IV to VI).
In the ROLE project the basic equipment for creating PLEs has been developed according to the idea of an easy drag and drop system of widgets. Browser-based prototypes have been developed like sketched in Figure 2.
On the one hand a repository (widget store) is necessary to store and administrate useful widgets. On the other hand an enabler space (widget space) is necessary to have learners their individually preferred widgets integrated, used and managed in their personal style.
Starting from this provided prototype essentially consisting of Widget Store and Widget Space the creation of PLEs has been tested in real world use cases and scenarios which are described in the following sections.
In the first scenario learners were provided with the
ROLE Widget Store [
In the following the ROLE widget store is described as
well as an example how ROLE widgets have been
integrated and used in iGoogle.
The ROLE Widget Store (Figure 3) is a living system
and repository of open educational resources. It hosts and
offers all kind of learning widgets. For registered
developers and users it is possible to “add a new widget”
(see icon on the upper navigation in Figure 3) whenever
they have found or created a useful widget with
pedagogical value. Everyone interested in these kinds of
open educational resources can make use of it.
Figure 4 [
This scenario had already been tested by students in
2011 at an early stage of development. Results of this
evaluation were already presented and discussed at
PLE2011 conference [
B.
The following use case is not an implemented
prototype, but a mock-up which has been created as a
consequence of early stage evaluations [
The mock-up scenarios presented in Figure 5 and Figure 6 have been used to discuss and evaluate taking into account teachers’ and students’ perspectives (see chapters IV to VI). Both mock-up scenarios are designed with the idea to be not restricted to use the widgets within a browser-based widget space like iGoogle. Moreover instead of using a collection of widgets at the same place it should also be possible to select and use only one very specific widget.
Thus, choosing between several means of (personalised) integrating and using the offered widgets should be one distinctive added value of all widgets in the ROLE Widget Store.
Figure 5 presents the use of the ROLE translator
widget which accesses and displays the results of different
popular resources such as LEO.org, dict.cc, Wikipedia,
Google translator all at the same time for comparisons of
translations [
This kind of PLE is created to efficiently work on a text document. While reading or writing a text in a foreign language the ROLE translator widget is always visible and usable in the desktop-sidebar. A click on the sidebar-widget-icon will open the widget like sketched in Figure 5. The widget will stay in the front while copying a term from the document in the background to transfer this term to the translator widget. The translation is shown including the resource of translation (dict.cc, Wikipedia, Google, etc.). This mean of widget integration should ensure a very efficient way of learning and working. It enables the user to learn new terms by using the widget but without losing sight of the text document. Moreover, using several resources of Web2.0 based translations stimulates the user to have a more critical reflection of the offered translations.
Figure 6 presents the use of a vocabulary trainer widget which can be opened in the browser sidebar right next to the text a user is working on. While reading the text in a foreign language terms might appear a user is not familiar with and wants to systematically train them. Then the terms can be added to the vocabulary trainer widget.
The widget has been implemented a slightly modified
Leitner system [
Important for the presented evaluation is the fact that these mock-up scenarios give ideas of some other ways how to use the offered widgets from the ROLE Widget Store.
C.
Scenario III presents an implemented prototype to mash-up PLEs which is called “Mash-Up Recommender” (MR, see Figure 7).
The unique aspect of the MR is the fact that it services as a gate and a guide to access the large number of widgets and gadgets available on the web in a reasonable self-regulated way. For this purpose the MR templates are based on learning activities related to the SRL Process Model described in chapter II.
The main purpose of the MR is to support the
selfregulation of learners in mashing up their learning
environments. Therefore, psycho-pedagogical information
is transferred into applicable recommendation by using the
MR widget. The MR widget can be seen as a filtering
system that provides more or less widgets that can be
added to the PLE depending on the used template. The
MR contains a predefined template called SRL template.
The SRL template can consist of the four basic SRL
phases “Planning”, “Searching”, “Learning” and
“Reflecting” which are displayed in the upper navigation
of the MR (see Figure 7). Each category contains a
number of relevant widgets, e.g. the category “Reflecting”
contains widgets such as recording tools, writing tools,
mind map tools etc. To have the SRL template adequately
working according these four SRL phases a ROLE
ontology [
The MR can be used to provide guidance on different
levels and for different stakeholders (e.g. teachers,
workplace learners, students, beginners, and advanced
students or experts). A high level of guidance is necessary
for instance for beginners and can be prepared by a
complete predefined PLEs based on a specific template by
a teacher or tutor. Later the tutor can share this PLE with
her students who can use it or modify. A lower level of
guidance can be provided if the teacher just shares the
template with the students, so that they have to create
their own PLE. For example, a teacher could select the
SRL entities goal setting, resource searching, note taking,
and reflecting for a template. Teachers or learners using
this template could easily search these SRL entities for
widgets and include them in a PLE. In this way the PLE
consists of widgets for each SRL entity. Learning
strategies are on a higher abstraction level, which results
in an increased number of widgets that can be
recommended. Learning techniques are on a lower
abstraction level, which leads to a smaller number of
related widgets that can be recommended. While in the
first case the learner gets more widgets recommended and
thus less guidance, in the second case the level of
guidance is higher because of the smaller number of
recommended widgets. For a detailed description of the
MR and its technical background see [
The evaluation took place equally in two focus groups:
• Teachers: The three scenarios were presented, tested
and evaluated in a teacher workshop taking place at
the Aha-Conference 2012 in Vienna [
Quantitative data were essentially collected by a short questionnaire in the end of testing and discussing the three scenarios. To investigate the main research question if and why these PLE scenarios will be accepted or rejected by students and teachers two more concrete questions were ask to think about while testing and discussing each of the three scenarios: • The first question was on worsening / improvement of learning outcome; • the second question on the technical including cognitive and time-wise burden / ease of personal learning process.
The answer categories ranged on a six-point-Likertscale from 1: worsening to 6: improvement resp. 1: burden to 6: ease, which means: the higher the value the better the acceptance of the respective scenario.
The graphic shows the mean values and the standard deviation (in brackets) for the three scenarios. Each of the scenarios was rated by eight teachers according to the two evaluation criteria described in chapter IV. Due to the small number of participants no inference statistical analyses were conducted.
The question regarding a possible improvement in learning was answered most positive in scenario 3: Mean value of improvement of learning increased from 4.14 in scenario 1 to 4.67 in scenario 2 up to 5.14 in scenario 3. The standard deviations show that respondents do not differ very much in the assessment of the three scenarios concerning improvement in learning; it ranges from 1.03 to 1.07. It tends to be consensus in this question.
The question regarding a possible ease of the personal learning process was altogether also rated most positive in scenario 3: The mean value is 4.50. But at the same time there is also the highest standard deviation of 1.41 revealing a wider disagreement among the respondents in this question. In contrast to scenario 3 the worst result is displayed for scenario 1 with a mean value of 2.50. Moreover in this case respondents do agree most indicated by the lowest standard deviation of 0.93. In other words: While the teachers come to the agreement that scenario 1 will tend to be an additional burden instead of easing the personal learning process scenario 3 is rated much better by teachers but with a broader variance of opinions.
Altogether the results in both questions show a coherent picture for the three evaluated scenarios: While scenario 1 can be assumed to be potentially rejected by teachers scenario 3 tends to be accepted.
The question regarding a possible improvement in learning was again rated best in scenario 3: The mean value increased from 3.55 in scenario 1 to 4.11 in scenario 2 up to 4.68 in scenario 3. The standard deviation (sd) shows that the respondents differ most in rating scenario 1 (sd=1.34) followed by scenario 2 (sd=1.20) and scenario 3 (sd=1.11). In other words: Students not only rated scenario 3 best but also agreed in the answers of this question in scenario 3 most.
The question regarding a possible ease of the personal learning process was also rated best in scenario 3 with a mean value of 4.29. Students also agreed in the answers of scenario 3 most (sd=1.28) while they had the broadest variance of opinions in scenario 1 (sd=1.60) which was rated lowest with the mean value of 3.00. Considering a significance test scenario 3 is significantly better than scenario 1 in both questions (Improvement: F2,36 =5.48, p=0.008; Ease: F2,36 =4.52, p=0.018). Due to the small sample this can be randomly and thus is not further discussed.
Altogether the results in both questions show again a coherent picture for the three evaluated scenarios: While the results of scenario 1 neither show a clear tendency to be rejected nor to be accepted scenario 3 clearly tends to be accepted by students in this comparison of PLE scenarios.
The use of widgets within a widget space such as iGoogle was evaluated positive in its easy technical handling but negative in the challenge to efficiently support daily learning activities. Thus there is neither acceptance nor a clear rejection of scenario 1.
Better accepted was the use of single widgets wherever and whenever learners wants them to use (e.g. in a desktop-sidebar or browser-sidebar, online and offline) sketched in scenario 2
Best accepted was the idea to support self-regulated learning (SRL) by using a four-phases activity model while learners are challenged to select widgets from a wide variety (scenario 3). The idea to connect different stages of SRL (Planning, Searching, Learning, Reflecting) with corresponding widgets was seen most needed and most useful.
The research leading to these results has received funding from the European Community’s Seventh Framework Programme (FP7/2007-2013) under grant agreement no 231396 (ROLE project).
Available: