I am currently working on a third study that aims to further establish validity of a scale that I have developed which assesses individuals' propensities to believe in unconventional ideas (e.g., conspiracies and the paranormal). I am considering adding another dimension to this validation package that compares people's scores on my scale to their
(
Read more... )
Conformity to mainstream thought, I think (pun intended?), is a difficult concept to measure for this purpose. When I think "conformity" to any kind of thought, obviously I think of Sherif's autokinetic experiments (among others), but I think what you're getting at is perhaps an invariant or inherent tendency to conform to "traditional" modes of thought, rather than situationally influenced conformity to some thought(s)? It seems to be quite an ambitious endeavor to parse a stable tendency to endorse mainstream beliefs from contextually salient demands conform to ostensible consensus. To my knowledge, no such measures exist. And, if that is what you're driving at, can you be certain that holding mainstream beliefs is mutually exclusive of holding unconventional beliefs? If not, such a scale would not aid in discriminant validity. Or is that not why you want such a measure? Then again, it depends on how you're thinking of operationally defining conformity in thought. I'd like to hear more about what you were considering; I think it would be an interesting addition if you could define "mainstream beliefs". How did the creators of the older paranormal/unconventional beliefs address this issue (I assume they did not)?
I really like the idea of a measure of locus of control. Have you looked at any that are out there? I've only ever seen Rotter's, and have no idea how widely it is used to date. I like the idea of forced-choice format for such a scale. I would think an external locus of control would be a related concept to some unconventional beliefs (e.g., belief in fate & the paranormal, perhaps) but not others (e.g., conspiracies, perhaps). However, Rotter's scale (I'm not sure about some others) is a global measure of locus of control. Wouldn't it be more useful to determine the locus of control with respect to each of the various situations/beliefs/domains addressed in your scale? I'm sure there's nothing out there for that purpose(obviously). And given that you want to determine the relationship between locus of control and a variety of unconventional beliefs, I think Weiner's stable-unstable concepts would also be applicable.
Personally, I think you should give everyone who takes your unconventional beliefs scale the MMPI-2. It would only cost a lot of money and take 2 extra hours per participant... But, with over 100 subscales, you could really have some fun with that. You'll have to ask me why, if you're interested (even though it's not at all feasible); it's too involved to include in a comment and I don't feel like compromising ethics today.
Reply
Also, this would not be assessing discriminant validity. I'm not trying to show that unconventional thinking is unique from conformist tendencies. Few would say that the two concepts are identical. This measure would add an aspect of construct/divergent validity and aim to clarify the murky mess that is the state of affairs in the current literature.
Rotter's scale is the primary measure of locus of control and the one that I think is most appropriate for this purpose (it would hold constant the measure used for locus of control used in previous paranormal studies). Further, I can assess the relationships between global locus and my subscales, I need not go off searching for another locus measure.
I am not familiar with Oscar Meyer's stable-unstable concepts. Call me a bad student.
The MMPI is silly. The people who maintain the scale aren't even sure which items belong on which subscale.
Reply
Leave a comment