On 9 Dec 1999 15:54:09 -0800, [EMAIL PROTECTED] (Mike Wogan)
wrote:
> Forgive me if I'm tuning in here at the end of the discussion, but I'm
> wondering:
- thanks! for tuning in - I have been reading it, and hoping
someone would say these things.
>
> Who says locus of control is a pure, narrow-band concept? The Stanford
> Binet intelligence test items yeild a low coefficient alpha. So do the
> aggregated items from the Wechsler scales. But that's appropriate, if the
> underlying dimension being measured is heterogeneous. cf Lee Cronbach's
> discussion of bandwidth-fidelity in his books on tests & measures.
>
> Doesn't it also make a difference what the original item pool was that was
> factored? What scale are we talking about?
>
> If the population being tested is different from the one the factor
> analysis was based on, it might account for the drop in coef. alpha.
> Don't you have to be concerned with why the drop? (I'm assuming in the
> original scale the coef. alpha was relatively high, even if, in the
> original group, you only looked at these four items.)
- so, if those items were picked to represent different aspects of
LOC, their low intercorrelations might be consistent with the
original reliability data; they were supposed to be independent, and
they are. Of course, if they are *supposed to* show internal
consistency (high correlations), which is all you have looked at,
and if the sample has as much natural variation in LOC as the original
one where they tested as consistent, then an alpha of .30 is
indication that something went wrong.
--
Rich Ulrich, [EMAIL PROTECTED]
http://www.pitt.edu/~wpilib/index.html
