Research Plan

A Framework for Computational Thinking and Collaboration among Second Graders in Finland

Introduction

Computational Thinking (CT) is a basic skill for everyone that is centered on problem-solving, designing systems, and understanding human behavior (Wing, 2008). It draws on a wide range of cognitive tools such as abstraction, decomposition, pattern recognition, and algorithms to assess and identify a problem, assess the problem, and design solutions to the problem (Wing, 2008). According to Wing (2008), CT should be taught as reading and writing. CT has been integrated into many curricular around the world through subjects such as mathematics, informatics, and information technology (Bocconi & Chioccariello, 2016) but not as a standalone subject. Specifically, in Finland, CT is taught as part of Math and Crafts. CT can also be learned through online platforms. Learning CT occurs through formal education, informal education, through self-directed learning or community-based learning. It can be deduced that learning occurs through individual, co-operation, and collaborative efforts. The existing literature shows that CT includes abstraction, algorithmic thinking and decomposition, and these set of skills are transferable across disciplines (Selby & Woollard, 2014).

Methods of teaching computational thinking

A review of existing literature in the context of CT shows that CT has to be learned or taught with block-based programming, unplugged activities, educational robots, teaching scripts, games and animations, and computer simulation (Atmatzidou & Demetriadis, 2014). Unplugged activities typically mean teaching or learning computer programming and science without a computer. It focuses on understanding and applying algorithms that make computers work as they are designed. Knowledge of these algorithms empowers learners to develop and create their own artifacts. Several computational thinking activities allow learners to learn through collaboration, cooperation, and self-direction. CT focuses on problem-solving, operational understanding of CT concepts, it requires no prior knowledge and more importantly no computers are required hence learners everywhere in the world can easily acquire CT knowledge learning.

Computer simulation and modeling is a way of teaching CT. Learners learn to design and implement models and run simulations of the models created (Moursund, 2015). Computer simulation and modeling were used in “Growing up Thinking Scientifically” project to teach CT (Lee et al., 2011). In this project, the students used abstraction to narrow down a real-world problem, followed by creating a model with the features they selected from the real world problem. A model was created to predict how a disease will spread in a school. The school population, layout, number and movement of students were among the features selected for this model. The model constructed was tested with a robot and evaluated to see if it reflected the real-world problem (Lee et.al, 2011).

Game design and creation are another means through which learners learn CT skills. Learners who participated in the iGame program created their own game through storytelling by developing scenes using Alice (Lee et al., 2011). Alice is a 3D animation program. Wenger in 2009, evaluated stories created by learners in Alice. He identified the following CT skills: algorithmic thinking, programming, modeling, and abstraction (Werner, Denner, Bliesner, & Rex, 2009). Similarly, Scratch is a block-based programming environment for creating animations using a community of learners approach. Also, in games such as Minecraft, learners can create their own artifacts with the game.

Educational Robots are used in educational contexts as objects of learning and as the tool for learning. When a robot is an object of learning, the learner constructs a robot. On the other hand, when a robot is a learning tool, it is used to teach subjects such as Mathematics, Science, and CT (Burbaitė, Damaševičius, & Štuikys, 2013). Robots were used by (Denis & Hubert, 2001) to teach computational thinking using a  collaborative problem-solving approach. The learners were given roles such as analyst, algorithm designer, programmer and debugger, which were alternated during each activity. In their research, they discovered that learners understood and adopted CT concepts as the projects progressed and recommended more authentic and engaging problems as it enables learners to grasp CT concepts.

A common pedagogical theory used for teaching CT includes constructivism, constructionism, and collaborative learning. Underpinning computational thinking is Constructivism (Chowdhury, 2015; Denis & Hubert, 2001). Also, collaborative learning approach was suggested by (Chowdhury, 2015) as a way of equipping learners with CT.  Chowdhury suggested that CT is taught or learned through collaborative learning due to the fact that CT concepts are difficult to comprehend (Chowdhury, 2015). Collaborative learning involves the active participation of a group of learners to solve a problem, create or design an artifact. Key to collaborative learning is, communication, and interaction to create shared meaning (Brindley, Walti, & Blaschke, 2009). This implies that learners may succeed or fail in collaborative learning. Collaborative scripts are scaffolds for structuring the interaction between learners (Kollar, Fischer, & Hesse, 2006). Scaffolding from the “Vygotsky's Zone of Proximal Development” (Vygotsky, 1992) enables learners to complete tasks or handle challenges they cannot do on their own. Teaching with scripts (Wood, Bruner, & Ross, 1976) supports learning. (Atmatzidou & Demetriadis, 2012) researched the impact of collaborative scripts on Educational Robot activities. Using Think Aloud Pair Problem Solving, JigSaw and Send a Problem scripting methods, they concluded that collaborative scripts were essential for collaborative learning. These collaborative scripts employ role-based learning approach. Existing tools and platforms for teaching and learning CT includes online platforms such as Scratch and Microbit. CT materials such as Computer Science (CS) unplugged are available online but requires neither computers or internet to implement them.  Also, software like Lego Edu programming is available online for download. It requires a computer. There are text and block-based programming environments for learning CT.

Research Problem

Computer programming is a 21st-century skill that the European Union and other countries in the world have stated as important and must be taught in schools. The new Finnish curricula demand that all students be taught computer programming. However, it is difficult to teach computer programming to people with no prior knowledge. Computational thinking has been found to prepare people to code. A major challenge Finnish teachers face is that they do not have the skill or knowledge to teach computational thinking.

Aim and Research Questions

The aim of this research is to investigate how second (2nd) graders can be introduced to computer programming through computational thinking.

The specific objectives of this research are as follows: 

1.  To examine existing computational thinking methods in existing literature. (e.g., individual, teacher-centered, etc)
2. To extend existing knowledge of computational thinking methods/approaches and their application.
3. To design and test a computational thinking program aimed at introducing 2nd graders to computer programming.

Research Questions

1. What are the existing computational thinking methods in literature?
2. Can the use of computational thinking methods equip learners with computational thinking skills?
3. Does the computational thinking program developed to enable learners to grasp computer-programming concepts?

Framework for Computational Thinking through Collaborative Learning

Designing a computational thinking program

In designing a CT program, core concepts in computational thinking such as decomposition, pattern recognition, abstraction, and algorithms will be emphasized. Also, collaborative learning scripts may be employed to initiate interaction among learners and subsequently making meaning of the CT concepts. The use of collaborative learning will enable learners to understand CT and construct their knowledge and artifacts. The diagram below shows the detailed steps of how the CT program will be developed. DBR, the methodology employed will be further explained under the Data Collection Methods.

Figure 1. Framework for Developing Collaborative CT Activity

Data Collection Methods

Design-Based Research (DBR ) is a systematic methodology that seeks to bridge the gap between theory and practice through iterative analysis, design, development, and implementation, emphasizing on collaboration between researchers and practitioners (Wang & Hannafin, 2005). The DBR methodology will be used to develop and test the computational thinking program. This method is chosen because it is the most appropriate method of designing and testing a new program (Kennedy-Clark, 2013). Unlike quantitative and qualitative research methods that answer research questions descriptively, the aim of DBR is to give insights as to how some ways of teaching and learning can be improved using existing theories (Barab & Squire, 2004). DBL is a cyclical methodology with 3 phases and evolves through iterations. A typical DBR research consists of the preparation and design phase, the teaching experiment phase and the retrospective analysis phase. DBR methodologies include Hypothetical Learning Trajectory (HLT) among others. According to Barab and Squire (2004), HLT is useful for bridging the gap between instructional theories and actual teaching experiments by providing information on how teaching experiments should be conducted. It also guides researchers to evaluate teaching methods employed against the outcome of teaching experiment. DBL has been by used (Denis & Hubert, 2001). HLT is made up of the learning goal, the learning activity and a hypothetical learning process (Bakker & van Eerde, 2015).

Preparation and Design

In this research, we begin with the preparation and design phases of DBR by formulating our HLT which is composed of the learning goal ( thus to equip 2^nd graders with computational thinking skills), prior knowledge, and assumptions of the learning process. Based on this HLT, we collect and invent a collaborative learning activity.

Teaching Experiment

During the teaching experiment phase, lessons will be conducted using the designed CT program. The HLT will serve as a guide when conducting the lessons. As a result, the CT may undergo smaller modifications due to unforeseen learning conditions. These smaller modifications will be documented for further analysis. Data collected in this phase includes video and audio recording of lessons and discussions, interviews with students, students’ learning product, pre and post-test, and field notes.

Retrospective Analysis

In the retrospective analysis phase, data will be collected during the teaching experiment session. It will include student work, tests conducted before and after the experiment, field notes, video recording of the experiment, and mini-interviews with the students lasting for about 20 seconds. Assumptions made concerning the learning process of students will be assessed against the observations made during the teaching experiment to reveal the learning process of students in this context through a task-oriented analysis using the table below.

Hypothetical Learning Trajectory			Actual Learning Trajectory
Task number	Formulation of task	Conjecture of how students would respond	Transcript excerpt	Clarification	Match between HLT and ALT (quantitative impression of )

This table will reveal inefficiencies of the program providing information on the possible reasons for the inefficiency as well as information on what can be improved. In addition, a comparative method will be used to examine the learning process based on the existing theory. In using the comparative method, the interview transcripts and videotapes recorded will be watched in a chronological order and compared with other data collected to confirm findings and show counterexamples.

Data Analysis Methods

Qualitative data analysis methods will be used to analyze data using NVivo (a qualitative data analysis tool) to code the interview transcripts and video data. A major point of interest is how students reason and solve problems hence their actual work process will be analyzed with Fuxion Disco software. The outcome will be triangulated with the students’ actual working process.

Reliability and validity of research

Internal validity will be tested with data triangulation of episodes of interest and other data such as field notes. The iteration of teaching experiments may confirm and explain findings. Also, theoretical insights will be used to support confirmations and counterexamples. It is expected that improvements made in the first teaching experiments will be successful in successive experiments to show that the program can be transferred to other similar contexts.

The reliability of this experiment will be ensured by peer examination of critical episodes. In addition, the Cohen’s kappa (Fleiss & Cohen, 1973), a method of evaluating the inter-rater agreement data will be used to measure agreement of coded videos and audios. Also, the consistency of responses given by students in questionnaires will be evaluated using the Cronbach's alpha (Tavakol & Dennick, 2011). With this method, we seek to ensure internal consistency of the data. To ensure that others can replicate this research, the research process will be documented well.

Conclusion

In this research, a CT program based on collaborative learning will be designed and developed This CT program is expected to equip learners with computational thinking skills such as abstraction, and pattern recognition as well as extends the literature on tools, methods, and pedagogical approaches. The new Finnish curriculum specifies that all students are taught computational thinking and later programming as part of the set of transversal skills. Major challenge teachers face is how to it. This research seeks to develop a program that teachers can use to teach computational thinking in their classrooms.

References

Atmatzidou, S., & Demetriadis, S. (n.d.). How to Support Students’ Computational Thinking Skills in Educational Robotics Activities.

Atmatzidou, S., & Demetriadis, S. (2012). Evaluating the Role of Collaboration Scripts as Group Guiding Tools in Activities of Educational Robotics: Conclusions from Three Case Studies. Proceedings of 2012 IEEE 12th International Conference on Advanced Learning Technologies (ICALT). https://doi.org/refwid:14996

Barab, S., & Squire, K. (2004). Design-Based Research: Putting a Stake in the Ground. Journal of the Learning Sciences, 13(1), 1–14. https://doi.org/10.1207/s15327809jls1301_1

Bocconi, S., & Chioccariello, A. (n.d.). Developing Computational Thinking in Compulsory Education. https://doi.org/10.2791/792158

Brindley, J. E., Walti, C., & Blaschke, L. M. (2009). Creating Effective Collaborative Learning Groups in an Online Environment Introduction: The Challenge of Creating Effective Study Groups. International Review of Research in Open and Distance Learning, 10(3). Retrieved from https://pdfs.semanticscholar.org/2c12/011c069cc7db1a41440fdad25c54b50cbc4e.pdf

Burbaitė, R., Damaševičius, R., & Štuikys, V. (2013). Using Robots as Learning Objects for Teaching Computer Science. Retrieved from https://www.researchgate.net/profile/Robertas_Damasevicius/publication/320111593_Using_Robots_as_Learning_Objects_for_Teaching_Computer_Science/links/59ceae79a6fdcc181abb61e1/Using-Robots-as-Learning-Objects-for-Teaching-Computer-Science.pdf

Chowdhury, B. (2015). Understanding Collaborative Computational Thinking. In Proceedings of the eleventh annual International Conference on International Computing Education Research - ICER ’15. https://doi.org/10.1145/2787622.2787736

Kollar, I., Fischer, F., & Hesse, F. W. (2006). Collaboration Scripts – A Conceptual Analysis. Educational Psychology Review, 18(2), 159–185. https://doi.org/10.1007/s10648-006-9007-2

Lee, I., Martin, F., Denner, J., Coulter, B., Allan, W., Erickson, J., … Werner, L. (2011). Computational thinking for youth in practice. ACM Inroads. Retrieved from https://users.soe.ucsc.edu/~linda/pubs/ACMInroads.pdf

Moursund, D. (2015). Technology and Problem Solving : PreK-12 Education for Adult Life, Careers, and Further Education.

Vygotsky, L. S. (1992). Thought and language (rev. ed.) Peer interaction and learning in small groups. International Journal of Educational Research, 13(1), 21–39. https://doi.org/10.1016/0883-0355(89)90014-1

Werner, L., Denner, J., Bliesner, M., & Rex, P. (2009). Can middle-schoolers use Storytelling Alice to make games? In Proceedings of the 4th International Conference on Foundations of Digital Games - FDG ’09 (p. 207). New York, New York, USA: ACM Press. https://doi.org/10.1145/1536513.1536552

Wing, J. M. (2008). Computational thinking and thinking about computing. Philosophical Transactions. Series A, Mathematical, Physical, and Engineering Sciences, 366(1881), 3717–25. https://doi.org/10.1098/rsta.2008.0118

Wood, D., Bruner, J. S., & Ross, G. (1976). THE ROLE OF TUTORING IN PROBLEM SOLVING. Journal of Child Psychology and Psychiatry, 17(2), 89–100. https://doi.org/10.1111/j.1469-7610.1976.tb00381.x