CoChat
Sign Up
Cited by 1,901Psychometric Methodologies and TestingStatistical Methods and ApplicationsAdvanced Statistical Methods and Models

The DOI matches its Crossref record; 82 of the 84 references it cites were found in a registry or an index.

Partial Least Squares (PLS) Structural Equation Modeling (SEM) for Building and Testing Behavioral Causal Theory: When to Choose It and How to Use It

Open access
Lowry, Paul Benjamin, Gaskin, James·2014·IEEE Transactions on Professional Communication
DOI 10.1109/tpc.2014.2312452
Open this paper in CoChat — free account

Abstract

Problem: Partial least squares (PLS), a form of structural equation modeling (SEM), can provide much value for causal inquiry in communication-related and behavioral research fields. Despite the wide availability of technical information on PLS, many behavioral and communication researchers often do not use PLS in situations in which it could provide unique theoretical insights. Moreover, complex models comprising formative (causal) and reflective (consequent) constructs are now common in behavioral research, but they are often misspecified in statistical models, resulting in erroneous tests. Key concepts: First-generation (1G) techniques, such as correlations, regressions, or difference of means tests (such as ANOVA or${\rm t}$-tests), offer limited modeling capabilities, particularly in terms of causal modeling. In contrast, second-generation techniques (such as covariance-based SEM or PLS) offer extensive, scalable, and flexible causal-modeling capabilities. Second-generation (2G) techniques do not invalidate the need for 1G techniques however. The key point of 2G techniques is that they are superior for the complex causal modeling that dominates recent communication and behavioral research. Key lessons: For exploratory work, or for studies that include formative constructs, PLS should be selected. For confirmatory work, either covariance-based SEM or PLS may be used. Despite claims that lower sampling requirements exist for PLS, inadequate sample sizes result in the same problems for either technique. Implications: SEM's strength is in modeling. In particular, SEM allows for complex models that include latent (unobserved) variables, formative variables, chains of effects (mediation), and multiple group comparisons of these more complex relationships.

Overview

Integrity signals

Registry

Crossref · 10.1109/tpc.2014.2312452

Cites

82 of 82 references verified0 citation errors · 0 not checkable

Statistics

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

Verification

Claims the paper makes

We couldn’t retrieve this paper’s full text, 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?

Citations

Cites

Flagged first, then by citations

Show all 84 references

2 of the 84 references couldn’t be matched to a known paper.

Research works printed but not linked: 17

  1. gaskin (2012), 0, SmartPLS factor analysis Gaskination's Statistics
  2. gaskin (2012), 0, SmartPLS formative second-order constructs Gaskination's Statistics
  3. schneider, 0, Estimating causal effects using experimental and observational designs, The Governing Board of the American Educational Research Association Grants Program
  4. gaskin (2011), 0, Detecting multicollinearity in SPSS Gaskination's Statistics
  5. ringle, 0, SmartPLS 2 0 (M3) Beta
  6. gaskin (2012), 0, SmartPLS new project load and troubleshoot data Gaskination's Statistics
  7. lohmoller, 1989, Theoretical Empiricism, 1
  8. 2012, WarpPLS 3 0 User Manual
  9. gaskin (2013), 0, Post-hoc power analysis in SmartPLS and AMOS Gaskination's Statistics
  10. wold, 1982, Systems Under Indirect Observation II, 1
  11. gaskin (2013), 0, SmartPLS Multigroup moderation and moderated mediation Gaskination's Statistics
  12. barclay, 1995, The partial least squares approach to causal modeling: Personal computer adoption and use as an illustration, Technol Studies, 2, 285
  13. gaskin (2012), 0, SmartPLS interaction moderation Gaskination's Statistics
  14. cook, 1979, Quasi-Experimentation Design and Analysis for Field Settings
  15. gaskin (2012), 0, SmartPLS mediation bootstrap and Sobel test Gaskination's Statistics
  16. hage, 1987, Social Causality
  17. chin, 1995, The holistic approach to construct validation in IS research: Examples of the interplay between theory and measurement, Proc ASA, 33

Cited by

Sorted by citations

See all 1,901

Related

Metadata

Authors
Lowry, P. B., & Gaskin, J.
Affiliations
  • City University of Hong Kong (HK)
  • Brigham Young University (US)
ORCID
Venue
IEEE Transactions on Professional Communication
Publisher
Institute of Electrical and Electronics Engineers (IEEE)
Published
2014-06
Volume
57
Issue
2
Pages
123-146
ISSN
  • 0361-1434 (ISSN-L)
  • 1558-1500
DOI
10.1109/tpc.2014.2312452
OpenAlex
W2065771650
Cited by
1,901
References
84
Citations by year
2026: 131 · 2025: 198 · 2024: 196 · 2023: 211 · 2022: 229 · 2021: 201 · 2020: 197 · 2019: 157 · 2018: 151 · 2017: 96 · 2016: 69 · 2015: 53 · 2014: 12
FWCI
96.56
Type
journal article
Language
en
Open access
Licence
ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html (vor)
Locations
  • IEEE Transactions on Professional Communication (publishedVersion)
  • ScholarsArchive (Brigham Young University) (submittedVersion)
  • SSRN Electronic Journal (submittedVersion)
  • The HKU Scholars Hub (University of Hong Kong) (submittedVersion)
Topics
  • Social Sciences › Decision Sciences › Management Science and Operations Research › Psychometric Methodologies and Testing
  • Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Applications
  • Physical Sciences › Mathematics › Statistics and Probability › Advanced Statistical Methods and Models
Keywords
partial least squares structural equation modeling, formative constructs, causal models, behavioral research, communication research, model misspecification, sample size requirements, mediation, latent variables, covariance-based SEM

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