Self-managing teams currently prevail in all software development organizations. Small and large software houses often handover projects to various self-managing teams with the trust that they can manage the whole operation without much supervision – which is the most advantageous factor of self-managing teams. This training on selfmanagement is often provided at a cost to the employees, thus the employee turnover is considered a loss in technological field. The personnel working on these software development projects were once students in a degree program, where they did worked as self-managing teams on their capstone projects but without much knowledge of how to handle its different facets such as team formation, organizing team building activities, managing the project at hand, documentation etc. Capstone projects being an instance of Software Engineering is the right place where students can be taught about how to work as self-managing teams. This research is a step in that direction. Among many aspects of self-management, this research focused on team formation (alternatively referred to as team composition or team building). Team composition has been hypothesized as a multifaceted concept that influences the team cohesion, team effectiveness, longevity etc. (da Silva et al., 2013). Research has shown that team composition is driven by the amalgamation of characteristics of individual team members and their team roles. The work on team building process begins with the identification of individual characteristics that are required to assemble the desired composition for a particular context (da Silva et al., 2011). Not much research is done to identify these characteristics or the criterions to select an individual in the context of computer science capstone students’ teams. Da Silva et al., 2013, noted that the process of individual team member selection for building software teams is not studied much. They found only nineteen studies that particularly addresses the necessity and identification of criteria to select individuals to compose software teams; only two of these researches were conducted with student teams in academic settings. This research therefore chose to work with students in the academic settings where clearly very little work is done. Research conducted with students have implications in the industry as well, as evident from the literature. Those software development organizations that heavily depend on self-managing teams working on various projects may use the propositions made in this research. This thesis identified the criteria (to be referred now as: Team Building Criteria or TBC) for selecting software engineering students when building a self-managing team for capstone projects in undergraduate computer science degree program. The main research of identifying TBC is guided by the Knowledge, Skills, Abilities and Other factors’ (to be referred now as: KSAOs) framework for student teams that is developed as part of this research.The thesis also looked at the effects of the use of TBC on team cohesion. There are several KSAO frameworks for professional teams available in the literature. Many of them were proposed for non-technological personnel in mind however some of them were proposed for technological personnel as well. However a common limitation of all these frameworks is that none of them proposed KSAO factors for taskwork skills. All of them describe only KSAO factors for teamwork skills. The KSAO framework proposed in this research have not only the teamwork skills component but the taskwork skills component as well. Moreover the framework proposed in this research is student focused. Therefore, this research has filled a very genuine gap in the literature on KSAO frameworks and for self-managing computer science Page 11 of 269 capstone project teams. The factors enlisted in KSAO frameworks are historically termed as higher-order factors. Based on formal names and conception from the literature, the researcher have identified second order factors as well, called the team building criterion. A second order formulation is necessary when the underlying measures can be viewed as a reflection or an expression of the higher-level construct. Moreover this research has also answered to the literature’s complain that when students are supposed to build own teams, they are usually not provided any “basis to critic and select appropriate team members” (Connerly et al., 2001). Henceforth, at least the students of computer science degree program shall have a set of criterion to judge applicants on before letting them become a self-managing capstone project’s team members. This research has also focused on the factor on which team effectiveness depends, i.e. team cohesion. Team cohesion is necessary in self-managing computer science capstone project teams because many student teams develop serious differences within the group midway through the project, although they know each other before hand. This can be attributed to the faulty selection process that is not guided by any criterions and often depends solely on relational biases and personal agendas. A project completed in academic environment with desired team cohesion has many long term benefits such as awareness of benefits of cohesion, stability and longevity of the team; the students shall carry this experience with them into their professional lives as well. The criterion proposed in this research are introduced into the students through adventure-based team building intervention. Adventure in English means “undertaking”, “venture”, “quest” or “exploration”. The capstone projects too are ventures, quests, explorations or undertakings which students are the sole caretaker of. As this research has resulted in identifying and subsequently introducing the team building criterion in the students so as to improve and enhance the team cohesion therefore these criterions can be rightfully called intervention. And the form of choice for introducing this intervention is the adventure based experience. The methodology used in this research is mixed methods. Qualitative content analysis is done on the psychographic self-evaluation questionnaire and modified Group Environment Questionnaire (mGEQ) and quantitative descriptive statistics (using frequency tables, chart and graphs) is used to describe and analyze the data obtained from psychographic self-evaluation questionnaire and the mGEQ. This strategy is defended as mix methods research and is adopted in several researches including Chen and Meindl (1991). There are two main instruments used in this research. The first main instrument is the psychographic self-evaluation questionnaire that has actually introduced the criterions to the students. The other main instrument is the mGEQ. Two hundred and twenty computer science students from three major public and semi-private sector universities of Karachi participated in this research. Out of these two hundred and twenty, one hundred and ten students were those that were unaware of any criterion for team building and had already passed out before this research gave its results. These students were asked to fill two questionnaires; one is the major instrument i.e. the modified GEQ; this was for assessing the cohesion in the team of their capstone project that had not used any proposed criterion for its origination. These students also helped in proposing the criterion as well, when they were asked to fill a questionnaire (appendix B) meant for asking past capstone project students about how they selected a team member and which criterion they think should assist in this selection. The team Page 12 of 269 building criterion proposed in this research were proposed through the students filling appendix B questionnaire as well as (and largely) through literature on team building in technology student teams. The other one hundred and ten students filled the psychographic self-evaluation questionnaire (appendix C) which was meant for introducing the TBC to them. Students enrolled in one course were allowed to look at the responses of all other students (of that course) on the questionnaire. However, a working online prototype software is now available and is made as part of this research which facilitates the comparison of answers to the psychographic questions across any number of students enrolled in the software as a member of a class. This will not only safe the students from tedious comparison but will also enable collection of data for a long term. The prototype is developed after the completion of this research and it has not played any role in data collection. Those one hundred and ten students that had filled the psychographic self-evaluation questionnaire, also filled the mGEQ (appendix A) questionnaire. There is several months of space between the filling of self-evaluation questionnaire and mGEQ. To assess if the students really used the proposed team building criterion, the students were asked to fill a team building criterion usage questionnaire that asked the students about which of the criterion were low, medium or highly used in selecting team members. The software is developed to show groupings not only on all the questions (criterions) of the psychographic questionnaire but also on the basis of those criterions that students opted as medium or high formalization criterions. All hundred and ten students filled the whole questionnaire because proposing a student to join a team is different from whether the student shall accept that offer or not. Therefore all students must assess others for their abilities using these criterion. The improvement in team cohesion was assessed through qualitatively analyzing and comparing the responses of those students to mGEQ questionnaire that had used the team building criterion proposed here and those that didn’t used any proposed criterion. The analysis has shown that the team cohesion in self-managing computer science capstone student teams improved tremendously as compared to those that didn’t used any criterion. The students were of the opinion that they selected skilled friends and not just friends on the team using these criterions. The mGEQ data has shown that the individual’s attraction to the group on both social and task level has improved much like group integration on social and task level. The implications of this research in criterion centric team building, psychographic and student segmentation, big data formulation and automated self-managing team selection in software industry is also discussed. This research is as intensive in terms of identifying team building criterion for computer science capstone project teams as team climate inventory theory proposed initially by West (1990) was for professional teams. As long as it gets the attention of the larger research community, positive results can be drawn by using these criterions in capstone project team building." xml:lang="en_US