One of the main articles of faith among members of the Church of the Polygraph is that an organization, such as a police department, that polygraphs it's employees and potential employees will have a better work force.  I know of no evidence for this, nor have any poly that I've ever spoken with ever so much as implied that there is any evidence that this is correct; it is simply a tenant of their faith accepted as dogma.

However, it would be very easy to test this central claim.  Here's how:

Locate a number of law enforcement agencies that, within the last 10 years, either started polygraphing potential employees where they didn't previously or that stopped polygraphing potential employees where they did previously.  Hopefully there will be a number of departments, large and small, from different areas of the country in this group.  Between state, county, city, and other law enforcement agencies (i.e. Natural Resources Police, Corrections Officers, et cetera) there are at least hundreds of departments.

Then take a look at the officers hired in the few years before they started or stopped using the polygraph and the officers hired afterwards.  Look at the following:

• Number of official complaints lodged against them;
• How many were fired for cause;
• How many had internal investigations launched against them;
• Scores on any assessments or reviews;
• Number of reprimands issued against officers;
• Number of crimes committed by such officers;
• And other similar metrics that can be quantified.

If the claims of the polygraphers is true, then one would expect that officers hired in the five years before polygraph screening was instituted would have more complaints against them and poorer reviews, among other things, when compared with officers hired by the same department in the five years after instituting the polygraph.  Likewise, departments that did away with polygraph screening (if any exist) should, if the claims of the polygraphers are true, have a higher termination rate for officers hired after they did away with the polygraph than before.

There are a number of variables, but all can be easily corrected for.  By comparing people hired by the same department, you're comparing apples to apples as much as possible.  Also, hopefully the change in polygraph policy didn't all happen at the same time for all the departments (maybe, due to macroeconomic conditions, better candidates were available to hire before 1997 than just after, for example--if the dates are all different, this shouldn't be an issue).

Of course, polygraphers, being the evil bastards that they are, figure that they already have got theirs, so why risk upsetting things?  They'd never countenance this sort of research being done; they'd rather just sit around getting fatter all the time, leaching off of the society that they claim the polygraph protects.  If they were really in it for society, they'd support a study such as I suggest.

Predicted polygrapher reactions (beyond ignoring that which they can't respond to): 

• Claiming it's not their job to do research;
• Claiming that they know the polygraph helps people, so why bother with "research" and "data"
• Claiming it'd be impossible to do such a study, for some B.S. reason or other
• Talking about sex offenders raping little children up the ass and violating all of our women.  Repeatedly.
• Total silence, since they'll realize they just got their asses kicked in this argument and will just make it worse if they try to reply with anything except the truth, which will totally damn them.

The data I am suggesting we look at (1) exists; (2) can be reviewed; (3) has direct bearing on the quality of law enforcement officers; and (4) should definitely be effected if polygraph claims are true.  And if such claims are true, we should know.  And if they're not, we should know.  (Keeping the information from the public, as polygraphers want to do, is undemocratic and worthy of, among other places, North Korea.)

So, polies, why don't you have any evidence that the public is getting anything for the pre-employment screening that they're spending millions of dollars on?  Anyone?

pailryder wrote on Feb 7^th, 2009 at 12:05pm:


It's funny how the burden of proof isn't on the people demanding millions of dollars of public money on the premise that they're doing lots of good for public safety.

It's also funny that the closest you can come to admitting that you have no evidence for your claims is "yeah, it'd be nice if we had some evidence that what we're claiming is correct."  There's no evidence that this shit works to actually improve public safety--which is the central claim, even moreso than "it can detect lies"--and yet you haven't got the slightest problem getting fat off the public dole and ruining the lives of good people.  No problem at all.  You've got yours and that's all that matters.

Putting your own narrow interests above those of others and of society at large is the very definition of evil.  The marital status of your parents is, of course, outside of my knowledge.

pailryder,

Do you admit that there is no evidence that pre-employment screening exams increase the quality of a department or agency's workforce?  A simple question.  Yes or no please (and then, if you'd like to elaborate, please do so).

Lethe wrote on Feb 7^th, 2009 at 6:19am:


Lethe
Your project looks like excellent research.  However, the data would not have any relationship to pre-employment polygraph screening.  Lets just take a look at your impracticable research one point at a time.  
•      Number of official complaints lodged against them; 
o      Polygraph screening is not predictive on future behavior. 
As such, this data, if it could be obtained, does not relate to pre-employment screening. 

•      How many were fired for cause; 
o      This has nothing to do with polygraph screening.  Polygraph screening is not predictive on future behavior.  A perfect applicant and subsequent employee could be terminated for cause for acts committed AFTER they have been employed. As such, this data does not relate to pre-employment screening.  You do understand this fundamental principle, yes? 

•      How many had internal investigations launched against them;
o       Pre-employment screening is not predictive of future behavior.  As such, this data does not relate to pre-employment screening.

•      Scores on any assessments or reviews; 
o      Pre-employment screening is not predictive of future behavior.  As such, this data does not relate to pre-employment screening.

•      Number of reprimands issued against officers;
o      Pre-employment screening is not predictive of future behavior.  As such, this data does not relate to pre-employment screening.

•      Number of crimes committed by such officers; 
o      Criminal offenses committed after being hired have nothing to do with the previously administered pre-employment polygraph.  Again, the polygraph is not a predictive device.  As such, this data does not relate to pre-employment screening.

•      And other similar metrics that can be quantified.
o      Without identifying similar and specific metrics, I can’t comment, but I suspect my response would be similar since your criteria, thus far, does not relate to pre-employment polygraph screening.

Your premise appears to be that pre-employment screening does not weed out applicants who will at some future time have problems.  Your use of the polygraph in this fashion would be useless because, that’s right………… Pre-employment screening is not predictive of future behavior.  

Now, if you really want to get an idea what, if anything, pre-employment screening is doing.  You need to look at data that is actually verifiable and readily available.

•      Number of pre-employment exams conducted during a specific time period.

•      Number of “DI” tests

•      Number of verified “DI” tests based upon confessions/admissions/or disqualifying admissions after a “DI” test.  If there is no confession/admission/or disqualifying information, it is not verified and should not be used as the sole reason to disqualify.

•      Number of disqualifying admissions prior to the administration of the test and obtained during the pre-test interview

•      Number of “NDI” test in which information later surfaced that disqualified or resulted in the termination of the applicant/employee.  This would be information in which the applicant lied to a specific question on the pre-employment polygraph test and was scored as “NDI” (This would be the most significant finding and I, for one, would be interested in this statistic) 

•      And just for the sake of detailed information, the number of “NDI” tests in which the applicant was hired, and then subsequently counseled and/or terminated, and the reason for the counseling and or/termination.  Was the counseling/termination for activity that occurred prior to the pre-employment test, or was it post-pre-employment test?

So, Lethe, that is your assignment.  Do let us know the result of your research.

There was a time when I was skeptical of pre-employment screening, until I actually started running these tests.  I was concerned that they were not effective.  It did not take long for me to realize it is an invaluable process that weeds out a remarkable number of verifiable and confirmed unqualified applicants.  The scary part is this……..how many unqualified applicants were easily able to slip through the hiring process before the tests were implemented?  That number will never be known. 
  

pailryder wrote on Feb 7^th, 2009 at 6:29pm:

I think if that were the sum total of all data then one could certainly say that the polygraph successfully weeded out less-than-desirable applicants.

However, one would also have to take into account how many excellent applicants were disqualified despite not making any admissions, not witholding any information, and not being deceptive.

I have written on this board many times that if the polygraph were used soley in an attempt to elicit a disqualifying confession I don't think I'd have any issue with the polygraph.  When it is used to determine truth or deception is when it moves into the realm of junk science and that is also when its utility in pre-employment testing is worthless.

nopolycop wrote on Feb 7^th, 2009 at 9:46pm:


That would produce interesting data, however I think there are too many variables.  It could well be the case that a small, racially homogeneous, northeastern state that is politically liberal, like Vermont, attracts a different sort of person to be a police officer than a large, racially and geographically diverse, politically conservative state, like Texas.  While the experiment you suggest would be interesting, I think more reliable data could be obtained by comparing apples to apples, instead of oranges.

Now, I am curious about yankeedog's comments.  He seems to say that there is no measurable way in which polygraphing applicants improves the quality of the resulting workforce.  He even admits that he was initially skeptical, but then when he started earning a paycheck by using the polygraph he saw the light in the wavy lines.   He now says that "it is an invaluable process that weeds out a remarkable number of verifiable and confirmed unqualified applicants."   

So, he says it weeds out unqualified people, but seems to imply that unqualified people make just as good police officers as unqualified ones, since there is no way in which a police force with few or no unqualified people differs from one with many unqualified people.

I think one problem in this discussion is that polygraphers and I have a different idea of what a qualified or proficient police force is.  It appears the poly's definition is simply "a workforce that has been polygraphed."  Or, maybe, "a workforce that has few or no members with factors that have been declared disqualifying."  However, I define a qualified police force, for purposes of this discussion, at least, as one that solves crimes, obeys the law, is efficient, protects the community, et cetera.  Police officers that smoked dope in adolescence might make worse police officers, but that should be demonstrable through actual evidence, if so.

I am proposing actual metrics that we can use to measure how effective the polygraph is in doing what you claim it can do.  All you are doing is trying to obfuscate the issue and contradicting yourselves over whether it actually does any good or not.

Anyway, how many polies here agree that a study, along the basic lines that I've suggested, should be done or--at the very least--looked into to see how easy and costly it would be to do?  Can I get a show of hands, of the internet forum equivalent thereof?  Polygraphers only, please.  Who thinks it's a good idea to base public policy that spends taxpayer money on sound data as opposed to mere speculation?  Anyone?