Analysing quadratic effects of formative constructs by means of variance-based structural equation modelling

Jörg Henseler, Georg Fassott, Theo K. Dijkstra, Bradley Wilson

Research output: Contribution to journalArticleAcademic

46 Citations (Scopus)
1 Downloads (Pure)

Abstract

Together with the development of information systems research, there has also been increased interest in non-linear relationships between focal constructs. This article presents six Partial Least Squares-based approaches for estimating formative constructs’ quadratic effects. In addition, these approaches’ performance is tested by means of a complex Monte Carlo experiment. The experiment reveals significant and substantial differences between the approaches. In general, the performance of the hybrid approach as suggested by Wold (1982) is most convincing in terms of point estimate accuracy, statistical power, and prediction accuracy. The two-stage approach suggested by Chin et al (1996) showed almost the same performance; differences between it and the hybrid approach – although statistically significant – were unsubstantial. Based on these results, the article provides guidelines for the analysis of non-linear effects by means of variance-based structural equation modelling
Original languageEnglish
Pages (from-to)99-112
JournalEuropean journal of information systems
Volume21
Issue number1
DOIs
Publication statusPublished - 2012
Externally publishedYes

Fingerprint

performance
Information systems
Experiments
experiment
systems research
information system

Keywords

  • Partial least squares (PLS)
  • Path modelling
  • Non-linear effect
  • Quadratic effect

Cite this

@article{c9f5ddd142614eae8927a9f55c03241c,
title = "Analysing quadratic effects of formative constructs by means of variance-based structural equation modelling",
abstract = "Together with the development of information systems research, there has also been increased interest in non-linear relationships between focal constructs. This article presents six Partial Least Squares-based approaches for estimating formative constructs’ quadratic effects. In addition, these approaches’ performance is tested by means of a complex Monte Carlo experiment. The experiment reveals significant and substantial differences between the approaches. In general, the performance of the hybrid approach as suggested by Wold (1982) is most convincing in terms of point estimate accuracy, statistical power, and prediction accuracy. The two-stage approach suggested by Chin et al (1996) showed almost the same performance; differences between it and the hybrid approach – although statistically significant – were unsubstantial. Based on these results, the article provides guidelines for the analysis of non-linear effects by means of variance-based structural equation modelling",
keywords = "Partial least squares (PLS), Path modelling, Non-linear effect, Quadratic effect",
author = "J{\"o}rg Henseler and Georg Fassott and Dijkstra, {Theo K.} and Bradley Wilson",
year = "2012",
doi = "10.1057/ejis.2011.36",
language = "English",
volume = "21",
pages = "99--112",
journal = "European journal of information systems",
issn = "0960-085X",
publisher = "Palgrave Macmillan Ltd.",
number = "1",

}

Analysing quadratic effects of formative constructs by means of variance-based structural equation modelling. / Henseler, Jörg ; Fassott, Georg; Dijkstra, Theo K.; Wilson, Bradley.

In: European journal of information systems, Vol. 21, No. 1, 2012, p. 99-112.

Research output: Contribution to journalArticleAcademic

TY - JOUR

T1 - Analysing quadratic effects of formative constructs by means of variance-based structural equation modelling

AU - Henseler, Jörg

AU - Fassott, Georg

AU - Dijkstra, Theo K.

AU - Wilson, Bradley

PY - 2012

Y1 - 2012

N2 - Together with the development of information systems research, there has also been increased interest in non-linear relationships between focal constructs. This article presents six Partial Least Squares-based approaches for estimating formative constructs’ quadratic effects. In addition, these approaches’ performance is tested by means of a complex Monte Carlo experiment. The experiment reveals significant and substantial differences between the approaches. In general, the performance of the hybrid approach as suggested by Wold (1982) is most convincing in terms of point estimate accuracy, statistical power, and prediction accuracy. The two-stage approach suggested by Chin et al (1996) showed almost the same performance; differences between it and the hybrid approach – although statistically significant – were unsubstantial. Based on these results, the article provides guidelines for the analysis of non-linear effects by means of variance-based structural equation modelling

AB - Together with the development of information systems research, there has also been increased interest in non-linear relationships between focal constructs. This article presents six Partial Least Squares-based approaches for estimating formative constructs’ quadratic effects. In addition, these approaches’ performance is tested by means of a complex Monte Carlo experiment. The experiment reveals significant and substantial differences between the approaches. In general, the performance of the hybrid approach as suggested by Wold (1982) is most convincing in terms of point estimate accuracy, statistical power, and prediction accuracy. The two-stage approach suggested by Chin et al (1996) showed almost the same performance; differences between it and the hybrid approach – although statistically significant – were unsubstantial. Based on these results, the article provides guidelines for the analysis of non-linear effects by means of variance-based structural equation modelling

KW - Partial least squares (PLS)

KW - Path modelling

KW - Non-linear effect

KW - Quadratic effect

U2 - 10.1057/ejis.2011.36

DO - 10.1057/ejis.2011.36

M3 - Article

VL - 21

SP - 99

EP - 112

JO - European journal of information systems

JF - European journal of information systems

SN - 0960-085X

IS - 1

ER -