Collaborative Learning and Anxiety

Collaborative Learning and Anxiety A phenomenographic study of collaborative learning activities Katrina Falkner Nickolas J.G. Falkner Rebecca Vivia...
Author: Heather Lucas
1 downloads 0 Views 546KB Size
Collaborative Learning and Anxiety A phenomenographic study of collaborative learning activities Katrina Falkner

Nickolas J.G. Falkner

Rebecca Vivian

[email protected]

[email protected]

[email protected]

School of Computer Science The University of Adelaide Adelaide, South Australia, Australia, 5005

ABSTRACT

benefits of group discussion and group learning. Collaborative learning techniques have received significant practical and research focus within first year tertiary courses in particular, due to the evidence of increased engagement and the impact of social education structures upon progression and retention [10]. However, when collaborative learning activities place undue cognitive or metacognitive load upon students, they may introduce problems with anxiety and indeed decrease learning. The learning objectives buried within collaborative activities are complex - we frequently intertwine objectives related to discipline experience and skill, with professional skill and metacognitive development. Wiggberg [24] identifies that unclear learning objectives may confound the learning outcome; when participants are not clear on their role within the community, learning effectiveness may actually be decreased (p.129). A self-regulating student will set their goals, marshal their resources and then manage their time effectively [20] - without this fundamental level of metacognition they cannot direct their knowledge in a useful and constructive manner. Unsurprisingly, given that we expect our students to move on and take their place within the community, it is their ability to use their knowledge in accordance with others’ expectations and time frames that will be one of the greatest tests of their self-regulation. Students who are not experienced with collaboration, i.e. they have not as yet been able to extend self-regulation to group self-regulation, frequently break up collaborative activities into parts, assign each part to a participant in the collaboration, continue their work primarily independently, and then undertake a brief integration exercise prior to submission [1]. Wiggberg identifies that stress introduced through unclear learning objectives can lead to a predisposition to this behaviour (p. 114). In these cases, students also have a tendency to adopt those roles within a collaboration that they are most comfortable with, meaning that they may never be exposed to activities that address particular skill development. When contributions are not equally made, or equally respected within the collaboration, knowledge asymmetry can lead to decreases in peer learning and self-confidence [1]. Issues of power inequality and crosscultural concerns [12] may arise that impact upon the ease of adoption of certain roles; where students represent cultures with greatly varying power distances they may be reluctant to participate fully and equally in the collaboration. In this paper we describe a phenomenographic study exploring the variation in student experiences of collaborative learning activities. Our analysis shows that students

Collaborative learning encourages deeper learning, producing significant benefit in learning outcomes. There has been an increasing trend to adopt collaborative activities, due to the expected learning benefits but also because of the expected social benefits and their impact on transition concerns. However, collaborative activities may also introduce additional stress and anxiety for students as they cope with altered participation expectations, and the need to develop collaboration, communication and management skills concurrently with their discipline skills. In this paper we describe a phenomenographic analysis of student’s reflections on collaborative activities, including their perceptions of the purpose of such activities, and corresponding behaviours.

Categories and Subject Descriptors K.3.2 [Computing Milieux]: Computers and EducationComputer and Information Science Education

General Terms Human Factors

Keywords Computer Science Education, Collaborative Learning, Phenomenography

1.

INTRODUCTION

Constructivist pedagogy encourages students to engage with and develop ownership over their learning processes, encouraging reflective and deep learning. There has been significant research into the application of constructivist learning strategies, in both small and large classes, with significant benefit in learning outcome [4, 11]. Constructivist learning techniques are commonly, although not exclusively, collaborative; these forms of social constructivism leverage increased engagement and deeper learning resulting from the

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. SIGCSE’13, March 6–9, 2013, Denver, Colorado, USA. Copyright © 2013 ACM 978-1-4503-1868-6/13/03...$15.00.

227

view collaborative learning activities as exhibiting a range of learning objectives, ranging from task-oriented through to metacognitive, and that the choice of learning objective impacts upon their behaviour and experiences within those activities. When students recognise metacognitive learning objectives they are more readily able to disassociate their individual performance and focus on the metacognitive development of the group. In contrast, students who exhibit a task focus, incorporating an assumption of group selfregulation skill along with the need to demonstrate technical expertise, exhibit degrees of anxiety and stress.

2.

that disengagement may be invisible, in that a collaboration may dissolve without explicit agreement or discussion. Vartiainen [22] describes a phenomenographic study of the moral conflicts perceived by students within collaborative projects, exploring the conceptions underlying student responses to issues such as differing skill level and communication problems. Vartiainen identifies two dimensions of variation in his analysis, namely whether the student focusses on the impact of issues on the self, or on the other, i.e. the other members of the group. Wiggberg [24] presents arguably the most comprehensive study in this area, exploring students’ experience with collaborative group work and how they correspond to teachers’ expectations. Wiggberg presents three different stories of students’ experiences with a industry-standard software development project, exploring issues such as work allocation practices, responsibility for learning and degree of ownership of project goals. One key finding from this study is that unclear learning objectives, such as whether the purpose of the collaboration is to achieve deeper discipline learning or collaboration skills, may hinder learning outcomes and introduce additional stresses.

RELATED WORK

The crucial role of community within learning has been long established, and forms the basis for many collaborative and cooperative learning techniques. Vygotsky [23] elaborates the ideas of social constructivism through his observation that learning awakens a variety of internal development processes that operate only when one is working with others, i.e. the zone of proximal development, where learning is achieved within a group and as a direct result of the group activities. In this model, the learning of new concepts and processes is achieved by first establishing mastery as a group, and only then, developing mastery as an individual through continued practice and internalisation of knowledge. Liu, Joy and Griffiths [15] describe a national surveybased study on student perceptions on group work, encompassing both online and face-to-face collaborations. They incorporate a significant literature review that identifies three key problems in collaborative group work: poor motivation, lack in individual accountability and negative interdependence, where one or more students within the group complete the majority of the work, decreasing the overall group contribution. Drury, Kay and Losberg [9] describe a cross-sectional evaluation on perceptions of group work, combining analysis of focus group interviews and an extensive questionnaire to monitor changes in students’ perceptions of collaborative group work throughout their studies. They identified the need to model group work within foundation courses, due to the complexity of learning both discipline concepts and metacognitive aspects of group collaboration. The majority of students reported positive experiences, however a significant minority indicated negative experiences within their collaborations. Beven, Warner and McDowell [6] explore successful structuring mechanisms for pair programming, a form of collaborative learning where two students work together on the one program, with one acting as driver and one as navigator. In this study, the authors identify that students at a high skill level prefer not to be paired with a student of a lower skill level, while Thomas et al [21] find that students perform at their best when they are paired with a student of equivalent skill and confidence level. Begel and Nagappan [3] explore industry-level pair programming, identifying consistent issues related to differences in skill and confidence level, and communication issues. In these studies, anxiety is linked to the degree of cognitive and metacognitive dissonance. Plonka et al [17] explore the causes and impact of disengagement in pair programming; they identify five reasons why disengagement may occur, primarily focussed on differences in skill level, and the structuring of tasks, including degree of difficulty and time co-ordination. They identify

3.

RESEARCH METHOD

In this project, we explore student’s experiences with collaborative learning activities and their understanding of the purpose of these activities within their curriculum. Phenomenography [16, 7] is a qualitative and empirical research method that aims to explore how groups of people experience and understand certain events. The basis of phenomenography is the establishment of descriptive categories that describe the variation in the experiences of the participants in the study. The participants are selected in order to represent varied interpretations and understandings rather than to form a statistically representative sample. The research methodology underpinning a phenomenographic study typically consists of qualitative analysis of interview transcriptions, identifying quotes of experiences that reflect categories within one or more dimensions of variation. These dimensions, and corresponding categories, may be pre-determined from related literature and analysis and/or may evolve from the analysis process. Although a perceived weakness of qualitative studies is the inability to generalize from the small sample size, studies such as this capture the complexities of a phenomenon; such detailed observations cannot be captured in surveys or experimental designs [19]. A typical interview cohort consists of 10-15 participants, to support deep analysis of variation of experience and understanding [5].

3.1

Data Collection

In our project, we undertook a pilot stage refining an initial set of interview questions derived from the literature, before conducting a series of in-depth semi-structured interviews with our participant cohort. Participants were sought from those students who were currently studying, or who had recently completed, undergraduate Computer Science courses, with 10 students agreeing to participate in interviews of approximately 45 minutes duration. The participants represented a diverse cross section of our cohort: of our cohort 3 were female, and 7 were male; 4 were international students, while 6 were domestic; and 3 were enrolled in first year studies, 2 in second year, and the remainder in ad-

228

vanced level courses. All students had been exposed to a variety of collaborative learning activities within their courses, ranging from formative, informal pair programming, peer collaboration and group discussions within tutorials and lectures, through to more formal, structured group assignments and industry-related projects. Each interview commenced with a series of key positioning questions, designing to orient the participants towards the area of interest, following by a series of followup questions. The purpose of the unstructured followup questions, specific to the context of the individual interview, was to explore further each student’s experiences and construct a common understanding of the student’s experiences and understanding. The key positioning questions used within the interviews were formed from three question areas (QAs):

Some comments relate the pressure of collaborative activities to a lack of control over group decisions, such as time management and goal setting: P10: I still don’t enjoy the pressure of having to do the work at someone else’s pace; I try to do a good job anyway, but I have to do a certain degree of, like a good job, and I have to have done that good job by a certain time so that I am not disappointing the other person. (q2) In contrast, some students identified the degree of pressure and anxiety as having a positive impact upon their studies: P4: if we are doing the group project, we have a schedule in the group, we have our own schedule, so that can push you so that you don’t get behind. You are feeling that someone is pushing you, you have to do it by tomorrow, or something, so it is good. (q3)

• QA1: How do students understand the purpose of collaborative learning activities? • QA2: Do students view collaborative learning activities as positive or negative experiences?

In our discussion, we explored potential causes of this anxiety, including group assessment, and alternatives such as individual and peer assessment. Interestingly, students also exhibited variation in response here, with some comments focussing on the impact upon their individual achievement, while others adopted responsibility for the group, defining their individual achievement through that of the group.

• QA3: How do students perceive their relationship with their group? Each interview was transcribed verbatim and analysed in an iterative process using NVivo (Version 9) to refine both the dimensions of variation and the internal categories, identifying logical relationships between the qualitatively different categories. Within phenomenographic research, individual interviews are not categorised, rather, individual expressions and experiences are categorised according to the identified dimensions of variations.

4.

P8: It is still important, as the rest of the assignment obviously has to get completed and a lot of the time, if you are trying to drag the rest of the group through you end up leaving your own section to the very end and that can suffer because you have to put the rest of the group through (q4)

ANALYSIS

4.2

The results of a phenomenographic study consist of a description of the categories of understanding that have been isolated within the interview transcripts, and the relationships between those categories. We have identified five categories relating to how collaborative learning activities are experienced in the outcome space, illustrated in Table 1.

4.1

Category 2: Achievement

In Category 2, students experience collaborative learning activities as a way of producing a substantial achievement. The focus is less on the learning outcomes, either discipline or higher order, that are achieved but on the completion of a task where the outcome is greater than what an individual could achieve on their own. Some students identified the expected efficiencies to be gained by having multiple people contributing and collaborating. This contrasts with (q4), where responsibility for the group achievement is taken on board by the individual, whereas in this experience, we have a student who appreciates only having to adopt part of the total task load.

Category 1: Anxiety

In Category 1, the focus of the student is upon the emotional pressure and anxiety that they experience as a result of the multifaceted skill requirement of collaborative activities. The student does not acknowledge group or selfregulation as a learning objective, but instead assumes it as a required skill, which places additional pressure upon both themselves and the group. They view the purpose of the activity as task-oriented, but their focus is not so much on the achievement itself, but upon any potential compromises to their individual achievement that may occur. Some of the comments focussed on the structuring of groups, identifying differing skill levels and familiarity as the source of anxiety and pressure (identified by participant and quote):

P5: I like it. and it sort of splits the work in half a bit, so it’s sort of easier? (q5) While others recognise the synergy available in such activities: P4: We were working in groups, me and my friend, well, my partner, we’ve actually come up with two totally separate solutions, and we sort of joined them together to make one awesome solution. (q6)

P3: I hated it, especially if it was randomised groups where we didn’t get to pick our groups, that was terrible. I didn’t like it because if you got a bad group, it really destroyed your mark for the subject, or you have to do a lot more work. (q1).

There is a tension here between achievement and cognition: students may have a focus on achieving a substantial goal, but also want to access all of the available skill development. This represents a cognitive disconnect:

229

Category Anxiety

Achievement Professionalism Cognition Metacognition

Table 1: How Collaborative Learning Activities are experienced. Collaboration is experienced as... Focus a stressful experience that introduces emotional potential compromise to individual responses and pressure without substantial learning achievement benefit. an opportunity to complete a larger, more complex task. discipline achievement an opportunity to develop professional skills and aware- professional skill development ness. an opportunity to develop new skills, techniques and dis- individual knowledge development cipline knowledge. an opportunity to share and develop cultural awareness Higher order development and selfand higher order thinking skills. regulation

P8: If you can break the topic up into small elements, you can just divide it up amongst the group and work on isolated bits and that may mean that you don’t actually ever work on particular pieces in a subject ... you get an imbalance between which parts of the group know which parts. (q7)

4.3

Some students identified this as a move from independent study and self reliance to a more community-focussed model, identifying the need to work with others to achieve the zone of proximal development and, hence, deeper learning: P8: I like to work on my own, I need to work with others... so I like working by myself but I know I need the help, and it is better for me to work with other people. (q11)

Category 3: Professionalism

For some students, the development of professional behaviours is their primary understanding of the purpose of collaborative activities. Students’ conceptions of professional behaviour were quite mature, exhibiting a deeper understanding than the simplistic development of generic skills, extending to the importance of those skills to achieving something greater than yourself. One student identifies the scale of output expected of a professional, and the ability to take on larger and more complex projects as a group.

One student describes this as the “a ha!” moment, where the group works in a united fashion to achieve knowledge construction: P8: But I think it is that a ha! moment, where someone says something and you are like, there you go, and then you just go and do it.... and it falls into place. (q12) The impact of different learning objectives can be seen here, where competing learning objectives of cognition and achievement introduce group coordination problems and block learning outcomes. The following student identifies the issue at hand, but considers this mismatch just part of the standard practice – the roulette of group construction:

P5: Definitely teamwork is very important in society and even if you are quite smart you are just a single person, you can not make all of the things by yourself. (q8) Another student focusses on the mentorship and leadership aspects, identifying the role of an ICT professional in managing others and helping develop skills in less experiences colleagues.

P10: I never really learnt a lot because I was never really lucky with groups ... The problems were that people in my groups either were just interested in getting the assignment done and not really learning anything from it... (q13)

P9: Most of the time you are going to have a group or a team of people where you are going to have to work with them and even if you are not working with them you’ll have meetings with them or that sort of thing, and I think it represents what will happen once you leave university, rather than just, yeah, you might be able to do it, but you might not be able to work with other people to help them do it ... (q9)

4.4

4.5

Category 5: Metacognition

In Category 5, the student’s understanding of collaborative activities is focussed on the ability to develop and share higher order problem solving and analysis skills, and in learning to develop both self-regulation and group selfregulation skills. This comment makes reference to the ability to develop these skills within a multi-cultural context, incorporating multiple understandings of processes and technical content:

Category 4: Cognition

In this category, the student’s understanding of collaborative activity is a means to achieve new knowledge or skills, and a means to share knowledge and techniques with others. Many of the comments expressed the benefit to self, i.e. the ability to gain knowledge or skills from working with others.

P6: Yeah, I learn a lot of things, actually, even if I just pass in a group project, I think I learn that what I am thinking about is not always right, because other people come from different hierarchies, different cultural background, they have their own ideas on the course work, even when we are focusing on the same topic they have different experiences of it... we collect all of the

P3: I can do something by myself , but sometimes I need some very, very good idea to enlighten me because i need to enlighten myself to solve the problem. (q10)

230

ideas, and use our judgement to improve - it is interesting learning process. (q14)

overload occurs. We view cognition and metacognition as elements of process-orientation, where metacognition subsumes cognition in terms of higher order development. We see professionalism as representing a separate point within our outcome space. Where students are able to focus on the development of the group, and the development of group self-regulation skills, they are able to move beyond the focus on self achievement. Correspondingly, where students have unclear metacognitive learning objectives, or where students are overwhelmed by the metacognitive or cognitive load, associated anxieties and pressures modify their behavioural responses. Interestingly, one category that we had expected to identify, that of social development, was not one that was explicitly identified by our participants. As academics, we frequently utilise collaborative learning activities as a way of promoting social bonds to help address transition concerns, however this conceptualisation was not readily recognised by our participants. Several students identified friendships and relationships that had developed as part of their activities, but hesitated to articulate the connection between the activity itself and their social development. Rather, they viewed these relationships as resulting from their own initiation, coincidentally placed within the structure of the activity. So what can we say about how we should structure collaborative learning activities? We conclude our discussion by connecting what we have observed through our analysis with support from the literature. Collaborations are more likely to break up into disjoint, and unequal activities if the learning objectives are not clear. It is therefore imperative to make the learning conditions of the activity clear, enabling students to adapt their assumptions and behaviours from the outset. However, in some cases, the learning objectives of our collaborative activities are primarily task-oriented and are therefore open to metacognitive overload. Ideally, at this point, we will have provided sufficient scaffolding for our students so that they are not overwhelmed by the metacognitive requirements. Unfortunately, with the realistic knowledge that this is unlikely to be the case for all of our students we need to identify strategies for minimising anxiety. Cohen [8] identifies that open-ended ill-structured collaboration activities may be better positioned to encourage participation by reluctant students, as they tend to avoid potential causes of stress. We concur, in that these kinds of problems promote problem solving discussion and move away from an achievement or task-oriented focus. Rick and Guzdial [18] further identify that this form of structuring is more important for STEM students, due to their tendency to initially adopt a dualist view towards their activities, making them reluctant to contribute collaboratively. Although variably experienced by our participants, incorporation of individual and peer assessment components is also recognised as alleviating stress [13]. Perhaps the most successful measures for alleviating stress are through the explicit development of team management, group decision making and problem solving skills. Kavanagh et al [13] describe a detailed process of activity design and team management skill development, which utilises peer and self-assessment, in part, for their assistance in self-reflection and support for metacognitive development. Leeper [14] adopts a progressive project process designed to build student self-confidence in stages as they approach increasingly

One student identifies the ability to learn not only technical content from others, but also their learning processes: P10: When people work together to share ideas and share the processes that they use to get to solutions, so that other people can learn from what their peers do. (q15) Some students recognised the tension inherent here between anxiety and metacognition - many students identified group structure and the need to develop relationships and communication networks with new people as a significant source of anxiety, however some students also recognised the embedded benefits of such actions: P4: It is easy to stick with the old friends, it is easier, but it is not good in the future... you can get more different opinions, and you can... because everyone has their way to do things, maybe you never thought of the thing that they are doing, the way that they are doing, that can surprise you. (q16)

5.

DISCUSSION

Collaborative activity asks a lot of our students; while they are learning technical and professional skills, they are also having to master self-regulation skills, and then move those into the space of group self-regulation, coordinating their own self-regulation with that of other group members and learning how to integrate group self-regulation into their skill set. These so-called soft skills, including monitoring of group progress, group decision making, and evaluating and providing feedback on other’s contributions, are not given high priority within university pedagogy. In contrast, our experiences with industry, as a community, identify that our students are inadequately prepared, and more consideration needs to be given to these areas [2]. In their analysis of industry-level collaboration, Begel and Nagappan [3] identify that the issues involved in their form of collaborative activity can be traced back to sources of anxiety. However, our analysis identifies that in an educational setting, additional complexities related to learning objective may introduce further issues and problems for collaborative groups. We identify that the conceptualisation of purpose of the activity is closely tied to the way that the student approaches the activity and the development of relationships with their group. Students have a complex understanding of the benefits and costs of collaborative learning, and adopt a range of approaches based upon assumptions of purpose. When learning objectives are unclear, and students must determine them for themselves, this introduces a degree of tension and impacts upon the behavioural choices of the students involved in the collaborative activity [24]. The five categories in our outcome space correspond to the different ways that students experience collaborative learning activities. We do not view these categories in a strict hierarchy or spectrum, instead identifying individual relationships as follows. In our analysis, we view anxiety and achievement as being closely related categories, representing two ends of a spectrum. Both exhibit task-orientation, moving towards anxiety when cognitive or metacognitive

231

difficult tasks. Subsequent tasks can be customised in degree of difficulty to support varying skill levels. Wiggberg [24] identifies that students are more likely to value other’s contributions when they perceive that the other student is competent, either through the demonstration of skill or the presumption of skill. This supports structured activities where each student is provided with opportunities, relative to their experience, to demonstrate skill within any collaboration.

6.

[9]

[10]

[11]

CONCLUSIONS

Collaborative learning presents many opportunities for educators and students, supporting increased engagement, deeper learning and opportunities to develop and reflect upon group communication and self-regulation. However, sources of anxiety may be introduced that impact upon the success of the collaboration where students are unclear on, or have conflicting perceptions of, expected learning outcomes. In this study, we have explored students’ experiences of collaborative learning activities, identifying themes in terms of their understanding of the purpose of said activities and how that impacts upon their behaviour and expectations of other students. We have adopted phenomenography as the research method for this study, identifying five categories of variation associated with task- and process-oriented views. Our analysis extends and improves our understanding of how to structure collaborative activities, supporting moves to explicitly address group management and decision making skills, and viewing the development of these skills not only as a professional requirement but as essential to group success, and discipline learning. The connection between metacognitive overload and anxiety appears promising, and warrants further investigation. We intend to expand our analysis through the development of a broader survey study to gather statistically representative data reflecting students’ experiences with collaborative learning activities and learning outcomes.

7.

[12]

[13]

[14]

[15]

[16] [17]

[18]

REFERENCES

[1] L. J. Barker. When do group projects widen the student experience gap? In In Proceedings of iTiCSE’05, pages 276–280, June 2006. [2] A. Begel and S. B. Novice software developers, all over again. In Proceedings of ICER’08, pages 3–14, 2008. [3] A. Begel and N. Nagappan. Pair programming: What’s in it for me? In Proceedings of ESEM’08, pages 120–128, 2008. [4] M. Ben-Ari. Constructivism in computer science education. In Proceedings of SIGCSE’98, pages 257–261, 1998. [5] A. Berglund and M. Wiggberg. Students learn cs in different ways: Insights from a empirical study. In Proceedings of iTiCSE’06, pages 265–269, 2006. [6] J. Bevan, L. Werner, and C. McDowell. Guidelines for the user of pair programming in a freshman programming class. In Proceedings of the 15th Conference Software Engineering Education and Training, pages 100–107, 2002. [7] S. Booth. On phenomenography, learning and teaching. Higher Education Research & Development, 16(2):135–158, 1997. [8] E. Cohen. Restructuring the classroom: Conditions for

[19] [20]

[21]

[22]

[23]

[24]

232

production small groups. Review of Educational Research, 64(1):1–35, 1994. H. Drury, J. Kay, and W. Losberg. Student satisfaction with group work in undergraduate computer science: do things get better? In Proceedings of ACE’03, pages 77–85, 2002. K. Falkner and D. Munro. Easing the transition: A collaborative learning approach. In Proceedings of ACE’09), volume 95, pages 65–74. ACS, 2009. U. Geer and D. Rudge. A review of research on constructivist-based strategies for large lecture science classes. Journal of Science Education, 7(2), December 2002. G. Hofstede. National cultures in four dimensions: A research-based theory of cultural differences. International Studies of Management & Organisation, 13(1/2):46–74, 1983. L. Kavanagh, D. Neil, and J. Cockley. Developing and disseminating team skills capacities using interactive online tools for team formation, learning, assessment and mentoring. Technical report, The Australian Learning and Teaching Council, 2011. R. Leeper. Progressive project assignments in computer courses. In Proceedings of the 20th Technical Symposium on Computer Science Education (SIGCSE’89), pages 88–92, 1989. S. Liu, M. Joy, and N. Griffths. Students’ perceptions of the factors leading to unsuccessful group collaboration. In Proceedings of the 10th IEEE International Conference on Advanced Learning Technologies, pages 565–569, 2010. F. Marton and S. Booth. Learning and awareness. New Jersey: Lawrence Erlbaum Associates, 1997. L. Plonka, H. Sharp, and J. van der Linden. Disengagement in pair programming: does it matter? In Proceedings of the 2012 International Conference on Software Engineering, pages 496–506, 2012. J. Rick and M. Guzdial. Situating coweb: a scholarship of application. Journal of Computer-Supported Collaborative Learning, 1(1):89–115, 2006. M. Sandelowski. Sample size in qualitative research. Research in Nursing & Health, 18(2):179–183, 1995. G. Schraw, K. Crippen, and K. Hartley. Promoting self-regulation in science education: Metacognition as part of a broader perspective on learning. Research in Science Education, 36(1-2):111–139, 2006. L. Thomas, M. Ratcliffe, and A. Robertson. Code warriors and code-a-phobes: a study in attitude and pair programming. In Proceedings of SIGCSE’03, pages 363–367, 2003. T. Vartiainen. Moral conflicts perceived by students of a project course. In Proceedings of the 6th Baltic Sea Conference on Computing Education Research (Koli Calling 2006), pages 77–84, 2006. L. Vygotsky. Mind in Society: The Development of Higher Psychological Processes. Harvard University Press, 1978. M. Wiggberg. Computer Science Project Courses: Contrasting Students’ Experiences with Teachers’ Expectations. Doctor of Philosophy, Acta Universitatis Upsaliensis, 2010.