Understanding the issues and factors of crowdsourcing tasks in a call centre environment

Fardell, Chloe (2015) Understanding the issues and factors of crowdsourcing tasks in a call centre environment. [Dissertation (University of Nottingham only)]

[thumbnail of Chloe Fardell.pdf] PDF - Registered users only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (1MB)

Abstract

This research project aims to gain a deeper understanding of what it would take to meet the opportunity of converting the wasted time waiting to speak to call centre agents into a productive and worthwhile activity.

The primary ambition is to understand the factors and issues that will firstly affect the task structure, such as instructions and length of play, and secondly the motivations and perceptions of the users taking part in such a task. The effect of context is also investigated to understand the difference in outputs of standalone audio-only tasks and tasks in the call centre environment.

As a landline telephone is the lowest common denominator for contacting call centres it restricts the type of task to only audio or dual-tone multi-frequency (DTMF) input; a unique angle for investigation in both crowdsourcing and human computation fields where tasks are usually conducted online.

The study conducts twelve semi-structured interviews under laboratory conditions to understand the issues and factors. These include use of a prototype audio-only task to stimulate reaction to factors that were believed would affect task success such as interruptions and content of instructions. The tasks created were music tagging games called Label the music and Tag it, both based around previous research of a similar online task by Turnball et al. (2007). In the interviews participants interact with the tagging games through placing three phone calls; removing any chance of visual stimuli in the game. They placed two calls to a theoretical call centre where they were invited to play the game instead of listening to hold music and a further third call to the game on its own to measure the differing task effects.

It successfully concludes some principle factors that will affect audio-only tasks. The users’ background including preferences, health, propensity to learn and memory skills are key considerations for recruitment. It also finds a sense of task entrapment for a certain period of time can hinder the process, user perception of time compare with the actual length of time participating can vary greatly. Conversely to tasks which have both audio and visual input if these music tagging tasks were interrupted and instructions were not simple enough there was a negative effect on users’ memory and enjoyment. The research finds that this type of task has great potential, although there is much scope for future research into the viability of such a task before it could become an activity presented to general society for human computation or crowdsourcing purposes.

Item Type: Dissertation (University of Nottingham only)
Depositing User: Gonzalez-Orbegoso, Mrs Carolina
Date Deposited: 09 Dec 2015 15:33
Last Modified: 27 Dec 2017 22:05
URI: https://eprints.nottingham.ac.uk/id/eprint/30803

Actions (Archive Staff Only)

Edit View Edit View