Welcome to the State Personnel Board

Hello again Job Analysis Colleagues - yes, I am playing catch up today!
:)

In response to the following very important and oft-asked question,
here is the latest Board position:

What do you do when KSAs identified in the job analysis aren't
in the class specification? Can we test for them?

Well, the answer is a two-parter, so here goes:

1. If the KSA statements from the job analysis differ from those KSA
statements on the class spec. but remain within the intent and scope of
those KSAs included in the class spec., then, the KSA statements
resulting from the job analysis process should be used for exam
development purposes and as the basis for the exam bulletin verbiage.
For example: The class spec. KSA reads "Skill to tie miscellaneous
knots." The job analysis indicates that only bow knots are used; thus
the statement from the job analysis reads "Skill to tie bow knots."
The job analysis KSA is within the scope and intent of the class spec.
statement; however, it better indicates that which the job currently
requires and should be used over the class spec. KSA.

2. If the KSA statements from a job analysis differ from those KSA
statements on the class spec. and either no longer remain within the
intent and scope of those KSAs included in the class spec. or are not
addressed by the class spec., then the class spec. should be revised
prior to exam development/administration. It wouldn't make sense to use
out-dated, no-longer-applicable KSAs from the class spec. for exam
development purposes when we have data/evidence not only deeming those
said KSAs inapplicable but also identifying those KSAs that are
applicable. To have job analysis data and then to not use it would
certainly compromise the exam process in terms of quality, soundness,
and legal defensibility. (Translated to: Who knows if those on the
list could even do the job!)

I hope this clarifies the current state of the job analysis vs. class
spec. dilemma.


Shelley

 
 
Print this page.Print this page.
Last modified: 10/19/2007
 
 
spacer for ie6 width bug