# An Empirical Analysis of Task Allocation in Scrum-based Agile Programming

**Authors:** Jun Lin, Han Yu, Zhiqi Shen  
**arXiv:** [1411.6201](https://arxiv.org/abs/1411.6201)  
**Date:** 2014-11-23  
**Summary page:** https://www.summarizepaper.com/en/arxiv-id/1411.6201v1/

## Key points

- Agile Software Development (ASD) methodology has gained widespread adoption in the industry
- Need to understand challenges faced by software engineering students for effective training methods
- Existing research mostly focused on eXtreme Programming (XP) based ASD methodologies
- Lack of empirical studies on Scrum-based ASD programming, which is the most popular agile methodology among industry practitioners
- Paper presents empirical findings on task allocation decision-making, collaboration, and team morale in Scrum ASD process
- Study involved 125 undergraduate software engineering students from a renowned university working in 21 Scrum teams over a 12-week project
- Fine-grained data obtained through logging students' activities in an online agile project management platform called HASE
- Over 10,000 ASD activities logged during the study
- Results suggest negative correlations between collaboration and both team performance and team morale, challenging previous assumptions about its positive impact on outcomes
- Research contributes to a better understanding of how to effectively train software engineering students in using agile techniques

## Summary

Agile Software Development (ASD) methodology has gained widespread adoption in the industry and there is a need to understand the challenges faced by software engineering students in order to design effective training methods. Existing research has mostly focused on eXtreme Programming (XP) based ASD methodologies, but there is a lack of empirical studies on Scrum-based ASD programming which is the most popular agile methodology among industry practitioners. This paper presents empirical findings on task allocation decision-making, collaboration, and team morale in the Scrum ASD process which have not been well studied previously. The study draws its findings from a 12-week long project involving 125 undergraduate software engineering students from a renowned university working in 21 Scrum teams. Instead of relying on traditional survey or interview methods with their limitations in scale and level of detail, fine-grained data was obtained through logging students' activities in an online agile project management platform called HASE. Over 10,000 ASD activities were logged during the study. Contrary to existing preconceptions, the results suggest negative correlations between collaboration and both team performance and team morale. This finding challenges previous assumptions about the positive impact of collaboration on team outcomes. By providing insight into aspects of Scrum-based ASD programming that have not been extensively studied before this research contributes to a better understanding of how to effectively train software engineering students in using agile techniques.