Dear everyone,
Imagine I have an instrument called ET Belief Scale. There are 10 statements that indicate the respondent believes in Extra-Terrestrial life, and 10 indicating the opposite (reverse coded): 1 = very unlikely and 5 = very likely. I have collected data from a pretty large sample of participants. Then, both EFA and CFA showed that all 10 anti-ET items formed a separate dimension and all 10 pro-ET items formed another. This is actually what one of my teammates found, though the information about the instrument is hypothetical.
I did do some literature search and there's a similar published example, where control and autonomy granting each formed a dimension:
Article Parental Psychological Control and Autonomy Granting: Distin...
The authors offered the following interpretation: "Barber and colleagues (e.g., Barber et al., 2002; Barber et al., 2005) have called for a re-examination of the association between psychological control and autonomy granting, arguing that important distinctions between these constructs may be lost if they are combined into a single scale. Conceptualizing these constructs as opposite ends of a continuum assumes that parents who are high on one of the dimensions must be low on the other (Silk et al., 2003). "
Yet this could not be of sufficient help for my question, because (as I suppose) while it makes some sense to separate control and autonomy as two continua, it seems the only way to conceptualize belief and disbelief about something as opposite ends of a continuum. How should I interprete such results?
Thanks in advance.
Meng
Hi Meng,
There is a research about this topic and the link is below.
"Based on these results it can be said that the scale categories in reverse scored items could not be discriminated by responders at all type of rating, and that reverse scored items did not measure the same latent structure as straightforward items did."
So i think reverse coded items should not be used. But if your scale has this type of items you can report this. Actually in almost all studies have same results beacuse of reverse coded items.The article can be helpful.
Article The Number of Response Categories and the Reverse Scored Ite...
I agree with Rimantas Vosylis. The results are very common. You should read and interpret it as opposed to your initial consideration.
Often, the inverse result can be a "new vein of gold." Try and consider it under "the light of logic".
Best regards
Dirceu
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size