In an educational project aimed at introducing reflective pedagogical practice to improve the situation of preservice teachers in computer science at the University of Stuttgart, the need to restructure the introductory 'Programming and Software Engineering' course for all computer science students arose. So, the reflective pedagogical practice - by the lecturers - led to changes in the course. The first steps of the educational re-design - following a design-based research approach - are describe here.
Pre-service teachers1 in computer science at the University of Stuttgart have to take the introductory “Programming and Software Engineering” course (PSE) in their 1st semester together with all the other computer science students. For most of these students, especially the ones from BadenWürttemberg, this is the first contact with concepts and strategies for programming.2
So, we wanted to use this opportunity, when these future
teachers make their own learning experiences with
programming, to introduce them to a reflective pedagogical
practice [
We believe that the pre-service teachers’ problems with their first experiences using – formal – programming and software engineering concepts and strategies will be similar to the ones their students will face in the future when they teach at school. So, this point in time at the beginning of the university program is a good opportunity for the pre-service teachers to reflect on their own learning experiences. When they will have finished their university degree in Computer Science Education and will start their work in schools, they will be advanced computer scientists themselves. So, it will probably be much harder for them to remember their first steps into programming including the problems they had.
In a project funded by the Ministry for Science, Research,
and Art, Baden-Württemberg the group combined of a
software engineer, a computer science educator, a
1 ‚Pre-service teacher‘ means students in pre-graduate or
graduate teacher programs compared to
‚in-serviceteachers‘, who are already working in schools.
2 In the state of Baden-Württemberg computer science as a
compulsory subject in secondary schools only started
recently and with only one hour a week in 7th grade.
mathematics educator and an experienced PSE tutor, who is
also an advanced pre-service teacher, works in a design-based
research approach [
Reflective pedagogical practice was the most important of
the concepts, we used to design our intervention. Reflective
practice in teacher education means on one hand that
preservice teachers reflect on their in-school-experience in an
internship [
There is a similar problem for pre-service mathematics
teachers called the “double discontinuity” [
Some of the recommendations of the above-mentioned
initiatives to deal with the problems arising out of the double
discontinuity are to group the pre-service teachers in special
groups for the tutorials 3 and to substitute one of the
mathematics problems on the weekly worksheets by special
learning or teaching related problems [
Fig 1: Example of pre-service-specific math problem [
The aim of the project is to improve the situation of the pre-service computer science teachers in the PSE course. So, we planned these measures: • • • all pre-service teachers are gathered in two extra tutorial groups4 these tutorials are conducted by an advanced preservice teacher in each worksheet there is one dedicated extra task asking for a reflective pedagogical practice
In the discussions while preparing these tasks, the project group realized that it is not enough to change one task per worksheet, but that the structure of the course and the content of the different lectures were also not very supportive to learning programming and software engineering regardless whether the students were pre-service teachers or not. So, we went through all the lectures to identify and name the learning aims, to sort the prerequisites and the necessary definitions for the students to understand the topics.
Examples of these changes include that we added explicit learning objectives at the beginning of each lecture focusing on the programming related concepts and definitions. These learning objectives play the role of a table of contents of the lecture which makes it easier for the students to get an idea of the extend of concepts taught in the lecture. Following that information, we now provide insights into the didactical5 idea of the lecture in order to make it transparent to the students. For example, we use the Hamster simulator by D. Bohles (http://boles.de/hamster/simulator.html) to start with a miniature language and a miniature world to mitigate the effect of being overwhelmed by the sheer number of concepts used in programming. Also, it adds a gamification aspect to the exercises on the worksheets to increase the students’ motivation to explore programming on their own.
We also apply the objects-first teaching [
However, already in the previous year we learnt that the students do not see why teaching objects-first is of relevance. In the lecture feedback forms they said the course is not teaching what they believed is of most importance. Explaining in detail why objects are taught first and why this has some advantages aims at increasing the acceptance within the students. As a first effect, it has already increased the acceptance of this didactical method among the tutors which start to favor it now. In the meantime, also the student evaluation of the second instance of the lecture has taken place. In the feedback forms we can see a significant trend that the students understood much better why teaching objects first is an interesting and useful didactical approach. The number of students complaining about object firsts has decreased to outliers (less than 5 students in a group of almost 200). Their main reason for rejecting the approach is often based on insufficient tooling support (BlueJ as IDE in contrast to Eclipse or IntelliJ). We consider this not as fundamental issues with the objects first approach, but rather with its implementation.
Finally, the lectures have been enhanced with a competence-oriented list of skills and capabilities the students should have once they fully understood the contents of the lecture. In the last semester, we observed that a large number of students learned a concept’s definition but were unsure about what level of application they should be able to do in
attending PSE and about 400 students of computer science
plus 100-200 students taking PSE as imported subject.
5 ‚Didactical‘ is used here in the European understanding
and is similar to the concept of Shulman’s ‘pedagogical
content knowledge’ [
The following examples give an idea of the of the pre-service teacher specific tasks in the worksheets from wintersemester 2019/20:
As stated above - beside adding information about educational aspects to the lectures - we designed exclusive tasks for pre-service teachers. As this is still an ongoing process we give some highlights of the tasks we added to the worksheets of this lecture’s instance. At the time of writing roughly 60% of all worksheets of the semester have been designed and used. One task focused on our “challenging assignments”. Challenges have been added to all worksheets to offer some interesting tasks also for those students who already have prior programming experience as they complete the worksheet in half the time of the beginners. This is a common challenge in teaching programming as often teachers will face students who learned programming before, e.g., in a self-taught manner. In the pre-service teachers’ worksheet, we make the future teachers aware of this problem by making them reflect on their own different levels of experience and discuss the pros and cons of this approach.
In the following worksheets, the pre-service teachers are supposed to reflect the gamification approach using the Hamster simulator and the objects-first approach. They will have to discuss how they experienced their own learning using these didactical elements of the lecture. They can base their discussions on our explicit information about the designed didactical concept of each lecture. One specific task was to reflect on the differences between the Hamster simulator mini world and Kara the ladybug’s mini world 6 . There are differences in the complexity, the number of supported languages, and also in the objects first vs. algorithms first way of teaching. We expected the pre-service teachers to identify and analyze those differences.
Finally, in even further worksheets we had the pre-service teachers reflect on algorithmic problem solving vs. objectoriented problem solving. They should list the pros and cons of using the objects-first vs. the algorithms-first approach at school and make up their mind which approach they would like most based on their own experiences.
As part of the latter, we had a task for the pre-service teachers to discuss the pros- and cons of using a simplified IDE for objects first (BlueJ in our case). They had to contrast it to using another simple IDE, a more complex IDE or no IDE at all. This task was supposed to make the pre-service teachers explicitly reflect on the role IDEs play in learning to program. The idea was to contrast hiding a lot of technical details in the IDE vs. seeing and potentially understanding them when using no IDE or a complex IDE.
Another task for the pre-service teachers was to compare their experience in learning a computer science topic to the experiences they had in learning other, but related topics (math, physics, etc.). The idea was to identify and utilize the 6 https://www.swisseduc.ch/informatik/karatojava/kara/ positive aspects in teaching computer science and to mitigate the negative aspects of it.
Contracts and design-by-contract are important aspects of the objects first teaching approach. Hence, we made the preservice teachers to reflect on the use of contracts in a (potentially more advanced) course at school. The objective of the task was for them to come up with ideas how to teach a rather complex topic in such an advanced course in school. In detail, the pre-service teachers were asked how they would find an abstraction of the concept or how they would design simple examples which convey the motivation of contracts in programming.
As we used the hamster simulator’s mini world to introduce gamification in our exercises, we asked the preservice teachers to reflect on their experiences in doing gamebased exercises. In addition, they were asked to survey other game-based teaching methods which already exist and can be used in schools (like writing games in Scratch, using teaching related smartphone apps, etc.)
As part of teaching OOP, the pre-service teachers had to learn the concept of inheritance. As this is another more advanced topic, we made them again reflect on their experiences in understanding inheritance. Based on this, we asked them again to come up with ideas whether and how to introduce inheritance in the context of a school’s course. In addition, we asked them to design small tasks and exercises which potentially could be used in such a course.
In the last exercise designed so far, we added a task about how to introduce and illustrate the usefulness of debugging. As debugging is often a task which involves many and often overwhelming features of an IDE, the main objective was how to illustrate debugging to students. The hamster and ladybug mini worlds provide already partial answers in their integrated IDEs. These approaches make the idea and usefulness of debugging much more tangible.
As the course is still ongoing with the redesigns sketched above it is too early for an objective assessment. However, from a subjective point of view based on observations made in the lecture and tutorial groups the implemented improvements have a positive impact. Students are more motivated and seem to be less lost in the large amount of information a lecture at a university is able to provide. The pre-service teachers’ tutorial groups find the additional exercises interesting. Before, they were treated as any other student in the course Programming and Software Development, the question how to teach the subject has not been addressed and remained implicit. Because of this they also did not develop an identity as a group and complained a lot that they were confronted with teaching related questions only from the third semester onwards – and then it was disconnected from their own experiences. Now they get a preservice teacher related view on the topics right from the first lecture in their computer science topic.
V. FURTHER STEPS IN THE DESIGN-BASED RESEARCH PROCESS
The whole project is an example of design-based
educational research [
Some of the theoretical inputs are described above and the
first implantation of the course is taking place during the
winter semester 2019/20. Beside the detailed comparison of
the evaluation results (s. above) we will ask students to give a
formative feedback on the improvements using the Teaching
Analysis Poll method [