Confusion with convergent validity with formative measure

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
Priyankafox
PLS Senior User
Posts: 21
Joined: Fri Sep 18, 2015 6:40 pm
Real name and title: Ms Priyanka

Confusion with convergent validity with formative measure

Post by Priyankafox » Thu Sep 24, 2015 12:28 pm

I understand the example used in Chp 5 of A primer on partial least squares structural equation modeling.

Does it mean we cannot do a test on convergent validity if we dont have a single item global measure that is reflective for our formative construct.

Say my formative construct is safety concern.

However this is no single item global measure for safety concern available in the literature (in fact safety perception is a scale I have developed, lets say) then what do I do?

Does this mean I can't conduct the test??

In this situation, does the research stop because i chose to construct a formative construct that has no single item global measure available in the literature.
Thanks,
Priyanka

User avatar
cringle
SmartPLS Developer
Posts: 812
Joined: Tue Sep 20, 2005 9:13 am
Real name and title: Prof. Dr. Christian M. Ringle
Location: Hamburg (Germany)
Contact:

Re: Confusion with convergent validity with formative measur

Post by cringle » Sun Sep 27, 2015 1:23 pm

Well,

the best option is using a reflective measurement model of the same model. But usually you do not have indicators for both types of measurment models. The second best option - as a workaround and good approximation - is using the single indicator that represents the scale. If you don't have this one, the third best option is using a very similar construct that you have in the measurment model. However, the latter options may only allows you determining if the sign of the relationship meet your expectations. Otherise, you cannot do the test.

Best
CR

Priyankafox
PLS Senior User
Posts: 21
Joined: Fri Sep 18, 2015 6:40 pm
Real name and title: Ms Priyanka

Re: Confusion with convergent validity with formative measur

Post by Priyankafox » Sat Oct 03, 2015 7:33 pm

Thank you Dr Ringle.

I did the first option and I got a path coefficient of 0.675.

Does this mean my formative construct does not have convergent validity?

I did:
VIF (no issue, all below 5),
t-test of outer weights (some were NS after bootstrapping for a two tail test with 0.05 sig level)
t test of outer loading(all significant)


So now my redundancy analysis indicates a value below 0.8 (Hair et al, 2013).

My questions are the following:
(1) Does this mean my formative construct items do not have convergent validity?
(2) Does this mean I have to delete items (I only have four items)?

Look forward to your reply/and communities. I really need some support on this. I've read other material on formative measurement model but it is not as clear cut to understand as reflective measurement model.

Beraza
PLS Junior User
Posts: 3
Joined: Wed Dec 09, 2015 6:12 pm
Real name and title: Ana Beraza. Professor in the University of The Basque Country (Spain)

Re: Confusion with convergent validity with formative measur

Post by Beraza » Thu Dec 10, 2015 2:52 pm

Dear Dr. Ringle,

Idon't have reflective indicators nor a single indicator that represents the scale. If I use the third option -using a very similar construct that I have in the measurment model-:

1) what do you mean with "a very similar construct"?
2) This similar construct, does it have to be a reflective construct?

All the constructs included in my model are formative, three of them exogenous and another three endogenous constructs.

3) Is it possible to estimate this kind of model in Smart PLS 3 (3.2.3.)?

Thanks,

jmbecker
SmartPLS Developer
Posts: 1059
Joined: Tue Mar 28, 2006 11:09 am
Real name and title: Dr. Jan-Michael Becker

Re: Confusion with convergent validity with formative measur

Post by jmbecker » Thu Dec 10, 2015 4:22 pm

Yes, it is possible to estimate such a model in SmartPLS. However, without any reflective or single item measures it is hard to assess the convergent validity in a reasonable way.
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

janschreier
PLS Expert User
Posts: 116
Joined: Fri Sep 12, 2014 2:12 pm
Real name and title: Jan Schreier

Re: Confusion with convergent validity with formative measur

Post by janschreier » Fri Dec 11, 2015 7:55 am

Dear Jan,

in such cases, would you consider a short questionnaire where both reflective and the formative "questions" are asked only for the purpose of assessing convergent validity acceptable if the sample is comparable? I'm not seeking absolution here but just wonder if you would give this a shot, if you find out that you missed some questions in the original survey.

Thanks!
Jan

jmbecker
SmartPLS Developer
Posts: 1059
Joined: Tue Mar 28, 2006 11:09 am
Real name and title: Dr. Jan-Michael Becker

Re: Confusion with convergent validity with formative measur

Post by jmbecker » Fri Dec 11, 2015 9:22 am

Yes, I think that this is a possible solution. Not optimal, but better than nothing. I as a reviewer would definitely value such an approach.
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

Beraza
PLS Junior User
Posts: 3
Joined: Wed Dec 09, 2015 6:12 pm
Real name and title: Ana Beraza. Professor in the University of The Basque Country (Spain)

Re: Confusion with convergent validity with formative measur

Post by Beraza » Fri Dec 11, 2015 3:27 pm

Thanks Jan Michael.
Would you accept as convergentes validity test a Principal Components Analysis in SPSS?

janschreier
PLS Expert User
Posts: 116
Joined: Fri Sep 12, 2014 2:12 pm
Real name and title: Jan Schreier

Re: Confusion with convergent validity with formative measur

Post by janschreier » Fri Dec 11, 2015 7:27 pm

thanks for your opinion on that issue, Jan!

Priyaah23
PLS Junior User
Posts: 1
Joined: Mon Sep 09, 2019 8:24 pm
Real name and title: Ms Priya (PhD scholar)

Re: Confusion with convergent validity with formative measure

Post by Priyaah23 » Mon Sep 09, 2019 8:27 pm

Greetings,

If in case the redundancy value is lower than 0.7, may I know what should I do next?

Thank you.
PR

User avatar
cringle
SmartPLS Developer
Posts: 812
Joined: Tue Sep 20, 2005 9:13 am
Real name and title: Prof. Dr. Christian M. Ringle
Location: Hamburg (Germany)
Contact:

Re: Confusion with convergent validity with formative measure

Post by cringle » Tue Sep 10, 2019 8:58 am

In that case you may need an additional formative indicator that helps to better cover the domain of formative construct.

Post Reply