[IPAC-List] Potential Testing Process Change

White, Don DWhite1 at Columbus.gov
Fri Jul 20 13:52:11 EDT 2012


Thanks for the input.

One thing I need to be clear on is that the performance test is strictly pass/fail.

Candidates who pass the performance test are put on the list in the first method or stay on the list in the second.

Candidates who fail are either never placed on the list or removed.

Again, thanks for the input.


-----Original Message-----
From: Ronald Clare [mailto:rpclare at aol.com]
Sent: Friday, July 20, 2012 1:06 PM
To: White, Don
Cc: IPAC-List at ipacweb.org
Subject: Re: [IPAC-List] Potential Testing Process Change

I believe there are two approaches that would be viable. First, clearly announce that the test is multiple hurdle and that the top x candidates would participate in the next phase. "x" being the maximum # of candidates "reachable" based on list usage history. The second would not be multiple hurdle but would eliminate those candidates whose written score would not fall into the reachable group even if they max the performance portion. With this, the greater the proportional value of the performance aspect, the fewer candidates are disqualified from the first phase. The first option is more effective in reducing the numbers in the performance phase.

Sent from my iPad

On Jul 20, 2012, at 11:27 AM, "White, Don" <DWhite1 at Columbus.gov> wrote:

> For our non-uniformed classifications (and entry-level Police Officer and Firefighter) our city Charter designates that there will be no fewer than three bands for each eligible list. We Z-score and convert the Z-scores to standard scores, where the distribution mean and standard deviation are pre-set to put the cut-off score at ½ standard deviation below the mean. Our bands are then static (once you're in the band, you stay there) and we move to the next lower band only when there are less than five candidates in the highest band.


> Our Commission rule states that "an open eligible list shall be established for all applicants who attained a final passing grade."


> We have an exam that includes a multiple-choice test and a performance test. The multiple-choice test determines position on the eligible list. We administer and grade the multiple-choice test and invite everyone who passed to the performance test. No one's score changes - the bands are set. For the last administration of this test, we ended up with 28 candidates in the 90 band, 55 in the 80 band, and 53 in the 70 band. Over the two year life of this list the department will typically hire 8 to 10 individuals and rarely gets beyond the 90 band. We can run about 80 candidates per day through the performance test, which is administered outside.


> So here's the question: Anyone see a problem with inviting to the performance test only those candidates reachable on the list? That is, start with the 90 band, run them through the performance test, when we get close to moving into the 80 band, run those candidates through the test, and only test the 70 band if we see we are going to get that deep into the list.


> As you might imagine, we have 2 schools of thought: 1) we have to administer the performance test to everyone who passed the multiple-choice test because the language of the rule indicates that we are required to complete all testing for all candidates before creating the eligible list, and 2) we should invite only those in the reachable bands to the performance test, which is the more practical and cost effective, and doesn't waste staff and candidate's time.


> I'd like to hear your perspective. I'd also like to know if there are any jurisdictions that have a similar testing set-up and do a performance test on an "as needed/as we work through the list" basis.


> _______________________________________________________

> IPAC-List

> IPAC-List at ipacweb.org

> http://www.ipacweb.org/mailman/listinfo/ipac-list

More information about the IPAC-List mailing list