Research output

Article in Journal ()

Workflows for Quantitative Data Analysis in The Social Sciences

Citation
Turner KJ & Lambert P (2015) Workflows for Quantitative Data Analysis in The Social Sciences, International Journal on Software Tools for Technology Transfer, 17 (3), pp. 321-338.

Abstract
The background is given to how statistical analysis is used by quantitative social scientists. Developing statistical analyses requires substantial effort, yet there are important limitations in current practice. This has motivated the authors to create a more systematic and effective methodology with supporting tools. The approach to modelling quantitative data analysis in the social sciences is presented. Analysis scripts are treated abstractly as mathematical functions and concretely as web services. This allows individual scripts to be combined into high-level workflows. A comprehensive set of tools allows workflows to be defined, automatically validated and verified, and automatically implemented. The workflows expose opportunities for parallel execution, can define support for proper fault handling, and can be realised by non-technical users. Services, workflows and datasets can also be readily shared. The approach is illustrated with a realistic case study that analyses occupational position in relation to health.

Keywords
e-Social Science; Quantitative Data Analysis;d Scientific Workflow; Service-Oriented Architecture; Statistical Analysis

StatusPublished
AuthorsTurner Kenneth J, Lambert Paul
Publication date06/2015
Publication date online27/04/2014
PublisherSpringer
ISSN 1433-2779
LanguageEnglish

Journal
International Journal on Software Tools for Technology Transfer: Volume 17, Issue 3

© University of Stirling FK9 4LA Scotland UK • Telephone +44 1786 473171 • Scottish Charity No SC011159
My Portal