Re: key performance indicator
- From: Richard Ulrich <Rich.Ulrich@xxxxxxxxxxx>
- Date: Fri, 25 Nov 2005 21:08:54 -0500
On 24 Nov 2005 11:43:57 -0800, "Nick" <nickdw2002@xxxxxxxxx> wrote:
> Hi there,
>
> i have a scale of 10 items( 5 point likert scale)and a good cronbach's
> alfa 0.85. A statisticus made a key performance indicator graph
>
> Y axes: mean score (min0, max 5))
> X axes: impact regression
>
> So every item in the scale is plotted by its mean via the Y axes and by
> its impact via the X axes;
>
> Can anyone tell me what they mean with regression and how it is
> calculated
I googled for <"impact regression" scaling> to see if there
was something by that name associated with a scaling
procedure. And, apparently not, or not in a big way.
So you seem to have something about the importance
(impact) of an item, plotted against its mean. How did
your statistician measure the importance of an item?
There are a lot of ways; you'll have to ask him. If it is
"regression" between two variables, it is mainly the same
as correlation.
The mean of an item is not usually of primary
importance in determining how well it works, unless
it is too extreme to have variation. What was the
purpose of the plot? Was it showing that the
importance does not increase with the mean?
--
Rich Ulrich, wpilib@xxxxxxxx
http://www.pitt.edu/~wpilib/index.html
.
- References:
- key performance indicator
- From: Nick
- key performance indicator
- Prev by Date: Re: Comparing different scales
- Next by Date: Re: Comparing different scales
- Previous by thread: key performance indicator
- Next by thread: key performance indicator
- Index(es):
Relevant Pages
|