Home // eKNOW 2023, The Fifteenth International Conference on Information, Process, and Knowledge Management // View article

Computational Grounded Theory - An Experiment: Human Versus Machine

Authors:
Clint Wolfs
Eric Mantelaers
Martijn Zoet
Rick Reijnders

Keywords: Computational grounded theory; automatic keyword extractors; qualitative research; theory development; coding process; validity; reliability; RAKE; PRE; SRE; TRE; MRE; Yake!; KBERT

Abstract:
Grounded theory has been a fundamental concept within qualitative research for decades. While human creativity forms an important element during the creation of new theories, there have been suggestions in which computers might support this creative process. As a result, the computational grounded theory framework was introduced. Currently, there is a lack of studies that evaluate practical performance implications of computational grounded theory approaches. This paper aims to contribute by evaluating the differences between a manual and an automated keyword extraction process; a process that is considered to be important during the first stage of the open coding process. Results indicate that the outcomes of the automated process are - to some extent - in line with the outcomes of the manual process. Nonetheless, phi coefficients do not exceed 0.21, meaning that the results do not perfectly agree with each other. As a result, some keywords might be left out while other unimportant words may be labeled as being a keyword. Therefore, although automatic keyword extractors can be helpful during the open coding process, results should still be cautiously interpreted. Moreover, the results indicate that elements of the computational grounded theory framework can be implemented in practice, without significant different results.

Pages: 1 to 6

Copyright: Copyright (c) IARIA, 2023

Publication date: April 24, 2023

Published in: conference

ISSN: 2308-4375

ISBN: 978-1-68558-082-7

Location: Venice, Italy

Dates: from April 24, 2023 to April 28, 2023