single-item measure

Questions about the implementation and application of the PLS-SEM method, that are not related to the usage of the SmartPLS software.
Post Reply
nazila1366
PLS Junior User
Posts: 1
Joined: Sat Oct 30, 2010 7:23 pm
Real name and title:

single-item measure

Post by nazila1366 »

hi every one:
my quesition is regarding to the problme of single-item measure versus mulriple-item. i am trying to find a justification for using single-item meausre instead of multiple-item measue. i know that multiple-item is preferable to single-item, however, can any one help to to find a reliable reason to use single-item measure? because my quaestionnaire was designed and my date has already been gathered. i have no choice than finding a way to justify it
thank you
Naizla
User avatar
Diogenes
PLS Super-Expert
Posts: 899
Joined: Sat Oct 15, 2005 5:13 pm
Real name and title:
Location: São Paulo - BRAZIL
Contact:

Post by Diogenes »

Hi,

With just one indicator: LV = indicator.

In fact, we do not have a LV.

If you modeled the hypothesis in terms of LV, it will be very difficult to “defend” the use of single indicator, because:
- the assessment of reliability will be not possible
- the domain of the definition of the construct will be under-represented by single indicator (content validity issues).

Inverting everything…
If you have just a single indicator, how to think in a model (hypothesis)?
- Write the hypothesis in terms of the indicator (not in terms of the construct or LV).
- Conclude and comment the results in terms of the indicator, and not in terms of the LV (that has a broader definition than the indicator).

Best regards,

Bido
Post Reply