Hi folks! We did a large survey and planned to use 3 indicators to represent a latent construct (Sense of Place/SoP) in a latent regression model (structural equation modeling). However, one of the indicators did not work well (too low model fit) and I was left with only two indicators. I fear that reviewers who are used with formative scales will think that two indicators yields a very weak scale.
As reflective items are interchangeable, the latent construct should not change whether you use few or many indicators. I can argue for the face & construct validity of the indicators/scale, but there are many other suggested indicators for SoP if one were to develop a formative scale. Is there some good references for scale with few indicators? By the way: the model fit was excellent. (see enclosed structural model, the SoP construct had only two indicators).
John-Kåre,
generally, people run into trouble with small numbers of indicators because their model is under-identified. Obviously, that is not the case with your model, and it also doesn't seem to be what you are concerned about.
From a measurement perspective, the question is: Do your indicators adequately capture the scope of your construct? Do they reliably and validly measure your construct? For a relatively simple construct, two indicators might be just fine. Of course, it always helps if you can provide independent evidence.
Why exactly did you omit one indicator? And was this a previously validated instrument?
thanks, Heiko Breitsohl No, it was not a previously validated instrument. Typical measures for "sense of place" (SoP) are quite lengthy. Elements of SoP is for example, "belonging", "place satisfaction", and "emotional attachment" and I had indicators for these. Unfortunately, the "thrive" indicator to represent "place satisfaction" was measured with different response categories than the other two indicators (4-point ordinal scales vs 10-point scales) and the combination of these items did not work.
I believe I will be able to defend that the remaining indicators adequately capture the scope of the construct, but was concerned that I had only two! Several authors underscore the interchangeability of indicators in reflective models, thus, one can probably say that it is not important how many observed indicators I have, as long as the ones I have are valid, and the model is identified?
Best
John-Kåre
In line with what Heiko suggested, as long as your indicators are qualified in all reliability and validity criteria, which is the case in your measures, two items should be "fine". This is specifically the case for reflective measures, which you already mentioned. As you know, there are few rules of thumps for SEM, suggesting a minimum of 3 indicators per construct with a reason that you should have minimum 2 in case you may need to drop one. This is also what you did and the dropped item was not previously validated, which is important.
Regarding reviewers' comments, we never know as different reviewers have different "eyes" and different mindset. I have experience in two occasions where reviewers wanted me not be too parsimonious with the number of items (=3) to start with. But we always have our own justification as well!
Best,
Khan
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Structural Equation Modeling