Creating an audio self-presentation in French

In this report, we describe experiences deploying Open Learning Environments at a higher learning institution in China. It was introduced in established courses at the School of Continuing Education (SOCE) at Shanghai Jiao Tong University (SJTU). Online colleges/universities such as SOCE play a particular role in the Chinese education system. Foreseeing the enormous demand for higher education, the Chinese government decided in 1998 to establish a number of online institutions that were open to those students who did not pass the university entrance exams (in 2007, 10.1 million students applied for 5.67 million places) [People's Daily Online, 2007]. The SOCE test-bed enables us to learn about PLE from the viewpoints of “average” users, that is learners who are not highly technically literate or who have limited time due to jobs and families.

Language courses at School of Continuing Education

The students on the SOCE (the situation is similar at the other online institutions) are mostly adult learners who have a job. They study two to three years for associate or bachelor degree courses. The courses are similar to regular university courses with the difference that teaching takes place in the evenings and weekends. Each class lasts for three hours. Our blended classrooms are based on the Standard Natural Classroom model [Shen, 08], in which students can either attend the classes in person or attend via the live Web broadcast. Lectures can also be downloaded for asynchronous watching. During the lectures, online students can communicate with the teacher via short messages. The SOCE software system records, encodes and broadcasts each lecture in real-time. Target devices/destinations include desktop/laptop PC via ADSL broadband connections, IPTV devices via the Shanghai Educational IPTV channel, universities in western China via two satellite connections and mobile devices via GPRS. SOCE produces about 6GB of content every day. The current SOCE learning management system supports student-student communication via forums. Teachers can create exercises/homework to be completed by the students.

Some of the gaps in the current learning processes are representative for most Chinese educational settings. Students are degree/certificate oriented: the primary goal is to receive a formal acknowledgment rather than the mastery of the subject itself. Group work is hard to perform as it is seldom performed at school. In consequence, the practical application of learned knowledge is difficult for Chinese students. For instance, in language learning, the students have extreme difficulties communicating with native speakers. They are insecure, shy, and make frequent mistakes. Furthermore, in our lectures we over and over make the experience that the students do not know, have difficulties or simply do not use existing tools such as online dictionaries, pronunciation tools or micro-blogging sites. Reasons given include lack of time, interest and motivation. In ROLE, we experiment with different scenarios to address these problems.

We experimented with an Open Learning Environment and a variety of tools (video conferencing, micro-blogging, translation services, text-to-speech, etc.) over two semesters in the courses English Listening and Speaking, German I/II, French I/II and Introduction to Computer Science. About 50-100 students attended the language learning courses, and about 1.200 students took part in the Computer Science class. From our experience we knew that the students at SOCE have limited knowledge about Web tools (RSS is virtually unknown), only limited time at their disposal and limited technical expertise. Furthermore, in the Confucian culture of China learning is still very teacher-centered [Zhang, 07], and students are not used to actively contribute within a class. We therefore decided to build PLE according to the suggestions of the teachers and make these pre-built PLE accessible to the students. The teachers presented the PLE in class and showed example usages. In order to encourage usage, the students were assigned homework that required using those PLE. We then observed the students' usage of the PLE and collected feedback from the teachers and students by interviews and questionnaires.

Scenario for the personal learning environment

For testing the ROLE technologies within the SJTU test-bed we defined the following scenario. The French teacher assembled the Personal Learning Environment (PLE) shown in Figure 1. The widgets on the left hand side consist of tools helpful in a language learning context, namely a translator widget, a spell checker, a text-to-speech engine and a recording widget. In a first experiment, the PLE was introduced to the students during class and their usage was illustrated by examples. Then, the students received the homework to author and record a self-introduction using the specified tools. The intention was that the students, which were beginners in French with almost none French knowledge, would use the translation tool to translate English or Chinese source content into French, fine tune the translation with the help of the spell checker, listen to the pronunciation of the text using the text-to-speech widget, and finally record their own pronunciation using the recording widget an compare it with the given pronunciation of the text-to-speech widget. The outcome of this scenario which was used in a class is reported in the last section.

Specific Technical Realization

For the experiments at the SJTU test-bed, we applied a version which is based on the portal Liferay. While we did not offer advanced features such as inter-widget communication, the basic functionality was the same and enabled us to collect information about the usage of an Open Learning Environment (OLE) in a higher education context.

Liferay is an open-source portal system that enables the creation of groups, group-specific pages, and the population of pages with widgets. In the SJTU setting we created groups for each class (French I, II, computer science) as well as user accounts for each student in class and for each teacher. The teachers then added a number of pages for each of their lectures. Figure 1 shows one page of the PLE that contains several widgets grouped together to facilitate writing and speaking. From top to bottom, left to right, the widgets are a translator, a spell-checker, a text-to-speech tool, a recording tool, the CAM widget (which is minimized and serves to collect data on how the student interacts with all the widgets), a business dictionary and a conjugation tool.

Figure 1: Screenshot of the SJTU prototype

Figure 2 shows exercises that were included in the PLE. These exercises consist of third-party Web pages,

which we transformed into widgets to have them easily accessible to the student.

Figure 3 shows Multiple-Choice-Exercises that our PLE offers. These exercises were played using a IMS QTI compatible Web application. 

The teacher and the students used the widgets in a number of ways. For the widgets in Figure 1, the teacher showed how to use them in order to achieve tasks such as introducing yourself to a new colleague, writing a report, etc. The students then received homework that required them to use the PLE. The exercises were solved jointly in-class, but also worked-on individually from at home. 

Evaluation

Here, we will briefly report on our experiences with using the PLE in the SJTU test-bed for language learning support, namely the classes French I and French II over the three semesters Spring 09 (S1, June 2009–August 2009), Summer 09 (S2, September–November 2009), and Spring 10 (S3, starting March–June 2010). The classes were taught in French and English. Numbers about students inscribed for the French lectures are difficult to give. The school administration has figures about the total numbers of students in each tier, but no information about the selected second language as this can be freely chosen (students have the choice between German, French and Japanese). The S1 tier (in which French I was taught) counted 128 students, S2 474 students (347 French I tier, 130 French II tier) and S3 369 students (178 French I, 191 French II). The students progress from French I to French II. This means that the S2 French II students are mostly identical to the S1 French I students, etc. Similarly, concrete numbers about attendance are not available: students do not attend regularly due to their busy schedule. They can come to class in person, watch via their browser live or later download the video of each class. In average, about 20–30 students attended the French I lectures in the classroom and about 10–15 students the French II lectures. An indication of participation is given by the numbers of students who took the final exams: 33 for S1 (French I), 106 (French I: 72, French II: 34) for S2, 95 (French I: 37, French II: 56) for S3. This is about 23% of the total numbers in each tier. These numbers are typical for the second language (which is deemed as rather unimportant by the students). The lecturer was popular among the students and received good grades in students’ evaluations.

We employed several approaches to capture qualitative and quantitative data. The research team observed teacher-student interactions during the lectures to learn about the general interactions. Additional information was obtained by collecting feedback from the teacher. A questionnaire and a video-taped Q&A session served to collect information about the impressions of the students. Furthermore, we analyzed the log-files of the PLE server in order to gain quantitative data about students’ usage of the PLE. While the PLE was introduced and used in class, the students were expected and encouraged to use it outside of the class. The student accounts were automatically created using the list of students inscribed in each tier. An analysis of the PLE logging data showed that for S1 and S2, 49 students logged in at least once which is about half of the 106 students who took the exam in S2 (recall that this number includes the S1 students). For S3, 38 students logged in at least once. These numbers are comparable to the amount of students who handed in homework.

We believe that this rather low uptake rate of the PLE was mainly due to how the PLE was introduced in class. For the students it is not sufficient to present a single example of how to use the OLE in class. In itself, this will not enable and motivate students sufficiently to work with the provided tools. We hope to improve these figures by increasing the incentive to use the PLE. Measures will include extra points for the final grade but also better communication of the value of a PLE. In particular, students need to understand how the tasks and services work which can help them to achieve their goals. Each OLE usage needs to be broken down into individual steps. For instance the task of doing a spoken self-introduction can involve the steps of writing the introduction in the native tongue, translating it, polishing it, using a text-to-speech tool to listen to it, a recording to practice one’s own pronunciation and finally recording and publishing it. Demonstrating this whole sequence only once or twice overtaxes the student. Each single step needs to be shown and done by the students several times. The single steps as well as the combination of services should be assigned as homework giving the students an initiative for practicing.  Breaking down the usage of an OLE helps students to understand how to use it. Even more important is that the students understand why they should use a PLE. Students need to see the value of performing additional tasks which are not directly related to language learning.

One particular feature frequently mentioned in favour of PLE is that users (students and teachers) can configure and assemble them depending on their needs and with tools they are already using. We investigated whether our students actually have a need for these features and what kind of tools they are using. The data was collected using a Question and Answer session which was videotaped and by a questionnaire. 65 students completed the questionnaire which was completely voluntary. According to the data, our students are only using few of the available tools. 20% use blogs, 6% forums, 15% social networks, 6% RSS, and 85% chats. None of the students uses Wikipedia (which is not blocked in China) and only 3% use translation tools. 63% do not access the Internet with their mobile phone.

For western students the access to the internet is almost as common as watching TV. Applying a web based PLE to other countries requires taking into account country-specific restrictions. For example as SJTU is located in China the access to quite a large number of Web sites including social networks such as Twitter, Facebook and Friendfeed as well as many Web services is blocked. Furthermore, many institutions such as schools or companies restrict the access of sites deemed to be inappropriate for their scholars. 

Furthermore, we experienced new directions for technical research which should be taken into account for the development of the next ROLE prototypes. This includes the necessity to create accounts for some services which require a login or the necessity to have a single sign on feature since logging in several times leads to frustration of the students. Finally, even though logging data was collected in the SJTU test-bed, students did not raise privacy concerns. On the contrary, repeatedly, students uttered concerns that their contributions in the employed tools might not be noticed by the teacher.

References

[People’s Daily Online, 07]  People's Daily Online: More than 9.5 Mio. Chinese compete in world's largest exam. http://english.peopledaily.com.cn/200706/07/eng20070607_381831.html (last access: Oct 2010).

[Zhang, 07] Zhang, J. A cultural look at information and communication technologies in Eastern education, In Educational Technology Research and Development, 2007, 55, 301-314