Importance-Performance map analysis (IPMA)

Frequently asked questions about PLS path modeling.
Post Reply
lilyv
PLS User
Posts: 11
Joined: Thu Apr 04, 2019 5:07 pm
Real name and title: Lily Vu

Importance-Performance map analysis (IPMA)

Post by lilyv »

Hello,

I have some questions about Importance-Performance map analysis.

1- For assessing map, we use the results of I-P map constructs standardized or unstandardized effects ? By reading the instruction in the book of Hair et al. (2018) Advanced issues in PLS Modeling, I think to use the map of unstandardized effects to assess but I am not really sure.

2- Whether IPMA calculates automatically mean value of Target construct ? I don't find it out in the report, it shows just mean values of predecessors.

Could anyone help me to confirm these information ?

Thank you in advance.
jmbecker
SmartPLS Developer
Posts: 1129
Joined: Tue Mar 28, 2006 11:09 am
Real name and title: Dr. Jan-Michael Becker

Re: Importance-Performance map analysis (IPMA)

Post by jmbecker »

1. It depends. Using unstandardized effects make sense if all your predictors are on the same scale (otherwise you cannot compare the effects in their importance) and if the scale is somehow meaningful (i.e., if you want to interpret effects as "a one unit increase, increases the outcome by p units" where p is the coefficient size). If one of the before mentioned criteria is not given you may want to use standardized coefficients as they usually better reflect the relative importance that you want to illustrate.

2. IPMA should give you the mean for all constructs (LV Index Values)
Dr. Jan-Michael Becker, University of Cologne, SmartPLS Developer
Researchgate: https://www.researchgate.net/profile/Jan_Michael_Becker
GoogleScholar: http://scholar.google.de/citations?user ... AAAJ&hl=de
mous
PLS Junior User
Posts: 2
Joined: Wed Nov 04, 2020 11:29 am
Real name and title: Mr. Mous

Re: Importance-Performance map analysis (IPMA)

Post by mous »

Hello
In my model, I have three constructs. Y1 is a formative multi-dimensional, Y2 is a reflective multi-dimensional, and Y3 is a uni-dimensional that acts as a moderator on the relationship between Y1 and Y2. Would IPMA be a valid procedure for such a model? Y1 positively impacts Y2 and Y3 positively moderates the relationship.
Thanks in advance.
Post Reply