Cited by 38Creativity in Education and NeuroscienceComputational and Text Analysis MethodsDesign Education and Practice

Four Text‐Mining Methods for Measuring Elaboration

Dumas, Denis, Organisciak, Peter, Maio, Shannon, Doherty, Michael·2021·The Journal of Creative Behavior

Abstract

When individuals engage in divergent thinking, they vary on their Elaboration, or the degree to which they explain and embellish their responses. Although Elaboration has been considered relevant to creativity research for decades, its measurement has remained under‐developed. Here, we leverage technical and methodological perspectives from the text‐mining literature to posit four methods for quantifying elaboration: Unweighted Word Count, Stoplisted Inclusion, Part of Speech Inclusion, and Inverse Frequency Weighting. Although the Unweighted Word Count method is becoming typical in the field, more complex weighting methods appear to better fit the conceptualization of Elaboration. We explain the benefits of each of the included methods and demonstrate their application to responses from the Alternate Uses Task: showing that all four of these text‐mining methods produced Elaboration scores with high levels of reliability, but the Stoplisted Inclusion method appeared to maximize the score validity both in terms of criteria correlations and power to discriminate among creative experts and non‐experts. We offer an open‐access module for creativity researchers to apply these methods to their own data via our laboratory website [https://openscoring.du.edu/].

Integrity signals

Registry

Crossref · 10.1002/jocb.471

Cites

68 of 68 references verified0 citation errors · 0 not checkable

Statistics

Recomputing p-values needs the paper’s full text.

Claims the paper makes

This paper’s full text isn’t openly available, so its claims and self-check weren’t run. The registry and citation checks still apply.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Among several potential computational means of gauging idea elaboration, stoplisted word count demonstrates the strongest correlation with human judgments of elaboration.

    “Among several potential computational means of gauging idea elaboration (which is roughly synonymous with complexity or level of detail), stoplisted word count (i.e., counting the words used to express the idea while excluding a 'stoplist' of common low-information filler words) demonstrates the strongest correlation with human judgments of elaboration [27].”

Cites

Flagged first, then by citations

1 of the 69 references couldn’t be matched to a known paper.

Cited by

Sorted by citations

Authors
Dumas, D., Organisciak, P., Maio, S., & Doherty, M.
Affiliations
  • University of Denver (US)
  • Actors' Equity Association (US)
Corresponding author
Dumas, Denis
Venue
The Journal of Creative Behavior
Publisher
Wiley
Published
2020-09-28
Volume
55
Issue
2
Pages
517-531
ISSN
  • 0022-0175 (ISSN-L)
  • 2162-6057
OpenAlex
W3088387423
Cited by
38
References
69
Citations by year
2026: 7 · 2025: 6 · 2024: 9 · 2023: 8 · 2022: 5 · 2021: 3
FWCI
3.87
Type
journal article
Language
en
Open access
closed
Locations
The Journal of Creative Behavior (publishedVersion)
Topics
  • Social Sciences › Psychology › Experimental and Cognitive Psychology › Creativity in Education and Neuroscience
  • Social Sciences › General Social Sciences › Computational and Text Analysis Methods
  • Physical Sciences › Engineering › Mechanical Engineering › Design Education and Practice
Keywords
elaboration, divergent thinking, creativity assessment, word count, alternative uses task, scoring reliability, scoring validity

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review