Navigation
Search
New Comments
Handy Links
Syndication
« Book Review: 150 Things You Need to Know | Main | Internal Stakeholders' Reactions to Testing »
A Primer on Testing Issues
On October 28, 2008 by Jamie Madigan
I got a piece of advertising-slash-content in my inbox the other day that actually turned out to be noteworthy. And so I shall make note of it. A company called Peopleclick (a name we can only assume they settled on after rejecting "Peopledrag" and "Peoplerightclick") put together a white paper entitled "Questions the Government is Asking about your Employment Tests: Do you have the Answers?" You can get it by performing clicking motions here. They'll ask you to input contact information, but if you're so disposed you don't have to make it accurate.
This isn't going to score you any continuing education credits, but the white paper actually does make for a decent primer on adverse impact, validity, and federal requirements on documentation of all the above. It's something that you might give to hiring managers who want a little more information about the regulations and legal landmines around testing. Or as a take-away from a meeting on the same kind of topic. I often have meetings with clients who say "I want to do testing!" and while I'm always (well, almost always) happy to hear this sentiment, sometimes a little eduction and setting the scene is necessary. This paper tells them just enough to let them know that they need the help of an expert in things like this.
My only slight complaint is that there is relatively little consideration given to more cutting edge validation techniques, such as job component validity, validity generalization, and validity transportation. This is probably because not only do those techniques quickly bog down in jargon, statistics, and other detailed considerations, but also documents like the Uniform Guideslines and even SIOP's Standards are a bit out of date in those areas, not to mention case law. Still, that's a whole different debate.
Existing comments:
Post your own comment:
