| Institutions of higher education are being called upon to provide a more robust pathway to a college degree and improve upon the advanced workforce for the needs of the 21st century. As 21st century skills call for an employee to successfully work collaboratively in groups, an increase in technology adoption, globalization and increased competition are among the factors that make collaboration one of the most important skills that employers insist that individuals obtain today. An active learning environment through collaborative learning techniques has been encouraged in higher education as a means of improving student engagement (Freeman, Eddy, McDonough, Smith, Okoroafor, Jordt, & Wenderoth, 2014; Slavich & Zimbardo, 2012; Prince, 2004), but there is a gap in the literature when it comes to connecting the two research areas of collaborative learning and student intention to persist. Continued research is warranted to further understand the factors that may contribute to improving the situation of attrition, and to suggest ways that institutions can enhance engagement and ultimately improve student success. The purpose of this study is to create a model that will measure the factors that significantly influence a student's persistence in a computer supported collaborative learning environment. Based on prior theoretical research, the model is developed to analyze how collaborative learning is mediated by campus connectedness and a sense of community, and subsequently how it impacts student persistence utilizing affective organizational commitment and turnover intention measures. A survey instrument was developed based on the factors of the research model, and was tailored to the terminology used for communities and academia. To test the model, a cohort of students across multiple institutions was invited to participate in a virtual learning community, and a total of 103 students participated. To test the entire model, partial least squares structural equation modeling (PLS-SEM) was used. In testing the design of the overall model utilizing structural equation modeling, the relationship between all factors but one were found to be statistically significant. In further analysis, the model design was also able to discern between two separate groups, adding to its versatility. Implications for research in this area include an expansion of student attrition research through turnover intention, scalability with the addition of more constructs, and ultimately a new model that contributes to future research that is not limited to a higher education domain. |
