direction of arrows in the measurement model

PLS is broadly applied in modern business research. This forum is the right place for discussions on the use of PLS in the fields of Marketing, Strategic Management, Information Technology etc.
Post Reply
yiqingyu
PLS User
Posts: 13
Joined: Tue May 18, 2010 3:03 pm
Real name and title:

direction of arrows in the measurement model

Post by yiqingyu » Sun Dec 03, 2017 8:41 am

Hi PLS users,
Since the direction of arrows in the measurement model does not indicate whether a factor or composite model is estimated, but Mode A/B instead (Henseler et al. 2016, Sarstedt et al. 2016),
1. why does not the direction of arrows change when I set indicator weighting to a specific Mode?
2. why does the direction of arrows change when the measurement model switches between formative/reflective? Whichever measurement models I specify, PLS will estimate a composite model, is it correct?

Thank you for your help!

jmbecker
SmartPLS Developer
Posts: 888
Joined: Tue Mar 28, 2006 11:09 am
Real name and title: Dr. Jan-Michael Becker

Re: direction of arrows in the measurement model

Post by jmbecker » Mon Dec 04, 2017 8:47 am

The arrows specify whether you want to treat a construct as reflective (factor) or formative (composite). By default reflective (factor) is estimated using Mode A and formative (composite) using Mode B. However, by double clicking on the construct you can change this and, for example, also specify formative (composites) to be estimated with Mode A.

So what is the difference between a Mode A reflective (factor) and a Mode A formative (composite)?
1) If you use normal PLS, it is only the interpretation that changes and therefore the results output. For example, with formative constructs the focus is on the outer weights, while with reflective constructs the focus in on the outer loadings. In addition, traditional reliability and validity criteria such as AVE and Composite Reliability don't make sense for formative constructs and are therefore not displayed. Also discriminant validity assessment changes.
2) If you use consistent PLS, also the results change. Reflective (factor) constructs are corrected for attenuation and therefore produce different outgoing and incoming path coefficient estimates and loading estimates as the normal PLS. Formative (composites) will not be corrected, unless you pre-specify a user-determined reliability of the composite by double-clicking on the construct.

To sum up: In normal PLS the direction of arrows only changes the interpretation, in PLSc it also changes the estimates.
Dr. Jan-Michael Becker, University of Cologne, SmartPLS Developer
Researchgate: https://www.researchgate.net/profile/Ja ... v=hdr_xprf
GoogleScholar: http://scholar.google.de/citations?user ... AAAJ&hl=de

yiqingyu
PLS User
Posts: 13
Joined: Tue May 18, 2010 3:03 pm
Real name and title:

Re: direction of arrows in the measurement model

Post by yiqingyu » Tue Dec 05, 2017 2:20 am

Ah! In nature, the mode rather than model specification, determines loadings, weights and path coefficients. Thank you very much for the detailed explanation!

Post Reply