Harriet Robles, discussing employer satisfaction with the students they
hire, asks:
>Any ideas about how to collect the data?
I know this sounds obvious, but how about asking them?
I think many people get too hung up about "objective" measurement, and
even more people are mislead by analogies from the simpler physical
sciences into believing that a measure of something must be one procedure
that produces a single number, like laying a ruler next to something and
reading off its length.
In more complex, especially social science, situations one is much better
off to adopt a standard of "evidence" rather than "single number measure."
By that I mean, use triangulation: get evidence of how good (or
successful or cost-effective or ... whatever is important to you) by at
least three different (independent of each other) ways. If all three
"measures" indicate about the same result, feel confident that you know
something about the thing being measured. If they differ, look for other
ways of discovering what is going on.
Doing this can be very freeing -- you can accept "subjective" data, for
instance, as long as it is only one of several measures you're looking at.
You don't have to get everyone to agree that a single measure really
measures "it" (whatever you're trying to measure), because that's often
impossible to do. You only have to create enough evidence that captures
most of what people believe it is important to capture. You can be
inclusive of different perspectives on what's important, rather than
exclusive (settling on measuring just a single aspect.)
And the triangulation results (several independent measures converging on
the same story) are often more convincing and helpful than any single
measure could be.
--"John Gunkler" <jgunkler@sprintmail.com>
Learning-org -- Hosted by Rick Karash <rkarash@karash.com> Public Dialog on Learning Organizations -- <http://www.learning-org.com>