FUZZIER CONSTRUCTS IN CONTENT ANALYSIS: RATINGS FROM THE CROWD VS. TRADITIONAL EXPERTS

Search for this publication on Google Scholar
Moffit, M. (2020). FUZZIER CONSTRUCTS IN CONTENT ANALYSIS: RATINGS FROM THE CROWD VS. TRADITIONAL EXPERTS. Unc Charlotte Electronic Theses And Dissertations.
Download PDF

Analytics

16 views ◎
8 downloads ⇓


Abstract

The task of rating (or "coding") text data as a component of qualitative content analysis remains a time-consuming process. Advances in crowdsourcing platforms have presented novel opportunities to outsource this work away from the desks of academics, graduate students, and other "traditional" Subject Matter Experts. Recent studies suggest that the crowd could produce reliable ratings of qualitative constructs more exclusive to the social and organizational sciences, which could dramatically alter the way content analyses are conducted, largely by reducing the time requirements of those analyses. This is particularly true for what might be called "fuzzy" constructs; constructs without logical boundaries and traditionally considered less accessible to non-specialist raters. This study makes use of extant fuzzy construct data from an archival study (n = 177), ratings of subject matter experts (SMEs) from the same study (n = 6), and newly collected crowd ratings (n = 96) of the same data rated by the SMEs. Comparisons of the crowd’s ratings relative to the graduate-level researchers’ (i.e. "traditional") ratings revealed a high level of similarity. Crowd-based groups of as few as six randomly selected raters were similarly reliable to groups of three traditional experts, while not significantly more or less accurate. When specific selection criteria were used instead of random selection, as few as three crowd-based raters were likewise similarly reliable and accurate. These and other data on hand support a set of future recommendations for qualitative researchers, as well as for further empirical investigation into the use of crowd-based raters for traditional research and content ratings tasks.

Details
Author

Moffit, Miles

Title

FUZZIER CONSTRUCTS IN CONTENT ANALYSIS: RATINGS FROM THE CROWD VS. TRADITIONAL EXPERTS

Physical Description

1 online resource (87 pages) : PDF

Date

2020

Degree Granting Institution

University of North Carolina at Charlotte

Abstract

The task of rating (or "coding") text data as a component of qualitative content analysis remains a time-consuming process. Advances in crowdsourcing platforms have presented novel opportunities to outsource this work away from the desks of academics, graduate students, and other "traditional" Subject Matter Experts. Recent studies suggest that the crowd could produce reliable ratings of qualitative constructs more exclusive to the social and organizational sciences, which could dramatically alter the way content analyses are conducted, largely by reducing the time requirements of those analyses. This is particularly true for what might be called "fuzzy" constructs; constructs without logical boundaries and traditionally considered less accessible to non-specialist raters. This study makes use of extant fuzzy construct data from an archival study (n = 177), ratings of subject matter experts (SMEs) from the same study (n = 6), and newly collected crowd ratings (n = 96) of the same data rated by the SMEs. Comparisons of the crowd’s ratings relative to the graduate-level researchers’ (i.e. "traditional") ratings revealed a high level of similarity. Crowd-based groups of as few as six randomly selected raters were similarly reliable to groups of three traditional experts, while not significantly more or less accurate. When specific selection criteria were used instead of random selection, as few as three crowd-based raters were likewise similarly reliable and accurate. These and other data on hand support a set of future recommendations for qualitative researchers, as well as for further empirical investigation into the use of crowd-based raters for traditional research and content ratings tasks.

Genre

masters theses

Subjects--Topics

Communication
Psychology

Degree

M.A.

Keywords

Crowdsourcing
Fuzzy Constructs
Mechanical Turk

Subject Area

Psychology Industrial/Organizational

Advisor(s)

Rogelberg, Steven

Committee Members

Scott, Clifton
Heggestad, Eric

Degree Note

Thesis (M.A.)--University of North Carolina at Charlotte, 2020.

Rights Statement

This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s). For additional information, see http://rightsstatements.org/page/InC/1.0/.

Rights Holder Information

Copyright is held by the author unless otherwise indicated.

Identifier

Moffit_uncc_0694N_12479

Permalink

http://hdl.handle.net/20.500.13093/etd:1838