This document is awaiting IRB approval. I have included only the rational for this study, until further approval is granted.
The most current research indicates that Intelligent tutoring systems (ITS) have a measurable and significant positive impact on students’ learning (Barnett, 2003; Bain & Smith, 2000; Carnegie Learning, 2009; D’Aspice & Manzo, 2005; Frahm, 2010; Guzick, 2010; Hagerty & Smith, 2005; Hopson et al., 2002; Judge, 2005; Knighton, 2002; Kozma, 2002; LaManque, 2009; Thierry, 2004; Tiene & Luft, 2002; Witkowsky, 2008; Ysseldyle & Bolt, 2007). In numerous cases, ITS’ have guided students to higher levels of achievement than classroom teaching (Willia, 2014), particularly in cases involving learners who struggle with motivation (D’Apice & Manzo, 2005). These student achievements are possible because ITS provide more engaging and personalized instructions with more immediate feedback (Pane et al, 2013).
However, immediate feedback from a tutor – ITS or human – is not an ideal teaching method for all learners. For example, immediate feedback can cause distracting prompts. The explanations given might not make sense to the learner because they are not scaffolded to meeting the learners’ level of ability. These distractions impede learning and can be counterproductive to, what constructivists claim to be, developing deeper conceptual understanding of the work (Howely, 2012). Another commonly encountered problem with both tutors and ITS is that their pedagogies are structured to guide students through the learning process linearly. “They do not put the learners in control of their learning; in fact, they actually minimize active participation of the learner” (Halverson & Smith, 2010).
For these reasons, collaborative learning has been considered as an alternative methodology to tutoring. When working in collaborative learning experiences, students tend to seek new insights and
perspectives, ask questions openly, and explain difficult concepts, thereby gaining a better understanding of the domain (Doise et al. 1975). Working with others often increases task efficiency and accuracy (Tagard, 1997). Groups sometimes out perform the best individual in the group (Ellis et al., 1994; Schwartz, 1999); however, this is not always the case. Collaborative groups inconsistently achieving these accomplishments, which suggest that effective collaboration that is more efficient, accurate, and innovative depends upon a particular alignment of group dynamics, without which collaborative learners have a difficult time establishing a shared understanding and constructing new knowledge. These difficulties ultimately lead to poor learning outcomes (Jeong, 1998; Winquist & Larson, 1998).
The questions explored in this study center on identifying whether a particular type of collaborative method will reliably produce a particular type of result. The two types of collaboration explored here are (1) collaborations between tutors and tutees, referred to here as asymmetrical collaboration, and (2) collaborations between peers, referred to here as symmetrical collaboration. The results of particular interest in this study are accurate results, time efficient results, and innovative results. Depending on the goal of the project, one of these types of results may be more desirable than the others. Can collaboration be designed to meet a particular target outcome? If this is the case, then choosing an appropriate collaborative method maybe a crucial determinant of a group’s success. Knowing what results can be expected, might cause a tutor or ITS to use a different collaborative method depending on the purpose for learning. An ITS might greatly benefit from the ability to choose the type of collaboration method it employs depending on the desired learning results.