NotesFAQContact Us
Search Tips
Peer reviewed Peer reviewed
Direct linkDirect link
ERIC Number: EJ1115727
Record Type: Journal
Publication Date: 2016
Pages: 5
Abstractor: As Provided
ISSN: ISSN-1536-6367
Pride and Prejudice and Causal Indicators
Lee, Nick; Chamberlain, Laura
Measurement: Interdisciplinary Research and Perspectives, v14 n3 p105-109 2016
Aguirre-Urreta, Rönkkö, and Marakas' (2016) paper in "Measurement: Interdisciplinary Research and Perspectives" (hereafter referred to as ARM2016) is an important and timely piece of scholarship, in that it provides strong analytic support to the growing theoretical literature that questions the underlying ideas behind causal and formative indicators (e.g. Cadogan & Lee, 2012; Edwards, 2011; Hardin, Chang, Fuller, & Torkzadeh, 2011; Howell, Breivik, & Wilcox, 2007; Lee, Cadogan, & Chamberlain, 2014, 2013; Rhemtulla, van Bork, & Borsboom, 2015). Such literature provides in our view compelling reasoning to avoid, or at best be extremely cautious in using, formative/causal indicators. However, the theoretical arguments presented in such work seem to have had little impact on either the common use of causal/formative indicators in practice or the continuing proliferation of methodological articles defending their use (e.g., Bollen, 2007, 2011; Bollen & Bauldry, 2011; Bollen & Diamantopoulos, 2015; Diamantopoulos, 2011; Diamantopoulos, Riefler, & Roth, 2008). It should be no surprise that we hope that the approach used in ARM2016 proves to be more-convincing evidence to scholars that there are significant dangers in unthinkingly applying the causal/formative approach to measurement. In this commentary, we hope to supplement and add clarity to a small number of areas of ARM2016. In doing so, we hope both to add support to the main conclusions of ARM2016 and to open up the potential for causal/formative indicators to provide some useful function in future work rather than continue in the rather confused and contradictory place they seem to occupy at present. Specifically, we explain that, while we support ARM2016 strongly, there really should have been no need for such a demonstration because basic understanding of the principles of measurement leads to exactly the same conclusions. In doing so, we first focus on what "measurement" actually means and demonstrate where literature on formative /causal indicators makes important missteps, leading to erroneous conclusions. We are hardly the first to point this out (e.g., Borsboom, 2005), yet such lessons continue to go unheeded. Second, we diverge from ARM2016 in recommending a distinct nomenclature that distinguishes between formative and causal indicators, which follows from our earlier work (e.g., Lee, 2010; Lee et al., 2013) and again remains generally unheeded in current literature. Finally, we briefly suggest how separating formative from causal indicators allows each to have its distinct use in empirical research, even though neither is a measurement model.
Psychology Press. Available from: Taylor & Francis, Ltd. 325 Chestnut Street Suite 800, Philadelphia, PA 19106. Tel: 800-354-1420; Fax: 215-625-2940; Web site:
Publication Type: Journal Articles; Opinion Papers; Reports - Evaluative
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A
Grant or Contract Numbers: N/A