Validity of group marks in computing projects vs. individual attainment

The demand for graduate computing professionals has steadily increased over the past numbers of years with the IT sector regularly requiring 140,000 entrants each year in the UK (eskills, 2013). However, according to the Higher Education Statistics Agency (2016) there are only 16,000 computing graduates each year, leaving a shortfall of 120,000 per annum. Accordingly UK universities have responded by increasing their intake in computing degrees. Despite the increase in student numbers there has been not been a receptive increase in teaching staff levels. Consequently, this has placed increased demands on the staff in terms of workload and resulted in changes in teaching and assessment structures. Traditionally with smaller cohort numbers the majority of assessment of programming in computing pathways involves the use of continual assessment, including a variety of individual summative practical assessments and a final summative exam. This is a well-established but not necessarily good practice, with employers generally very critical that it produces graduates that lack productive collaboration skills due to the individual nature of the learning structures and assessments. Moreover, the increased student numbers has led to an evolved change, with many academic staff changing assessment strategies. While final individual exams have for the most part been retained, there has been an increase in group projects in an attempt to manage large scale assessment. Anecdotally it has been observed that some students that gain high marks in group based projects often fail to achieve the same standard in the associated individual assessments and final exam, despite the learning outcomes for both being comparable. The validity, reliability, opportunities for formative assessment and effect on teaching associated with the group programming projects continues to be a concern for students, module owners and external examiners and is a common source of debate at exam board level.
This research seeks to provide a critical evaluation of group based project assignments commonly used as a measure of academic achievement in university computing degree programmes. It statistically analyses a number of module results from different cohorts that have an element of assessment via group work but also individual assessment. It focuses on the relationship between the individually attained marks and that attributed per individual via the group projects and reports on the outcomes of this analysis.