EosJ:

It wasn't a slip of any kind, and I don't share that attitude that all are guilty until polygraphed innocent. That saying, "In God we trust and all others we polygraph" is kind of cocky, I know. I think it's a stupid saying that someone thought was smart.

Thanks for the compliments. Likewise.

Regards.

LBCB,

No face saving needed here. But, the common attitudes from polygraphers, quoting the Air Force OSI, In God we trust all others we polygraph !", is an underlying and reacurring theme. Or in other words, Guilty until polygraphed innocent !!. And what I say is never a personal thing, and yes I do treat your chosen vocation with disdain.  Its not personal with me, never has been, with the exception of one polygrapher. And I do like your posts, for the most part, as you seem to be a polygrapher with some balls and are willing to debate in open forum.  Unlike over at pro site. You are more than welcome to express your views, just as I am in speaking mine.  As you have never done we wrong, then let the slip from your previous post be just that, a slip.

Regards ...

EosJ, 

I like you. Really. You have usually treated me with respect despite treating the polygraph I use with disdain. That's ok, though, because I hate the polygraph too.

But that sentence you highlighted in red can in no way be construed as saying that every examinee or every person is a criminal. I would never say that. All it says is that I have not come across anyone--or at least I didn't know it--who was determined and prepared and who had committed criminal behavior. Most examinees who follow George's countermeasures advice, in my experience, are determined, but they aren't prepared. They just end up looking stupid and feeling embarrassed, as I said.

Because I know you are an intelligent person, I think perhaps you simply tried to save face with that last post about the highlighted part. Don't worry, I sometimes try to save face too. But I don't want anyone who comes to this website to assume something that you want others to assume that I meant when, clearly, that's not what I said.

Regards.

LieBabyCryBaby wrote on Dec 8^th, 2006 at 9:44pm:


LBCB,

Highlighted in red above ... A slip it may be, but still an underlying thought process and assumption.

Regards ...

EosJupiter wrote on Dec 9^th, 2006 at 1:52am:


EosJ,

I'm not sure I follow you. When did I ever say that everyone has something criminal to hide? I think that everyone has made some mistakes in their life, but criminal? Please explain your misunderstanding of my reasoning.

LBCB,

The flaw with your reasoning that everyone has something criminal to hide, is just wrong. I understand paranoia and caution where positions of authority are concerned, and yes I do not want bad fish swimming in the pool. But the screening net does more damage than its worth. I still seethe from the experience, and won't long forget. From my perspective it doesn't pay to be a boyscout anymore, if your going to be falsely accused and then barbecued by an interrogator.  Then only later proven you were telling the truth to begin with. Vindicated I was, Still pissed off I remain.

Regards ....

Bill,

I have to admit, that is hilarious, even if I am the target of the joke.  

EosJ,

10% collateral damage is not acceptable, you're right. At least not when you are in that 10%. Sometimes I find it unacceptable too, I must confess. Whether passing polygraphs or administrating them, I've never had to face it from the side of the person falsely accused. Regardless of my ego, which I gladly admit, and despite Digithead's contrary opinion, I do feel sympathy for those treated with injustice.

That said, though, the agencies most polygraphers work for all want to get rid of the bad fish in the net, even if it means killing some of the good fish at the same time. When there are thousands of qualified applicants, and you have a process that you believe--rightly or wrongly--is correct most of the time, you use what you have. Is that unfair? Not to those hired, but it is unfair to those falsely accused, as well as to the rest of us when a truly bad fish escapes the net and swims in our waters. I just don't believe that either the false positive or the false negative happens as often as some of you would like to believe.  Just because something may have happened to you doesn't make it a very common occurrence.

You are right about something else, I think: Maybe I haven't come across a truly prepared, determined person in an exam--a truly prepared person determined to beat the test despite past criminal behavior. And if I have, I didn't know it. What I have come across are people who are determined but not prepared, even though they've come here and read all of George's advice. Those people fail, after looking really stupid and being very embarrassed.

Finally, Digithead,

Your math makes sense in a purely mathematical world.  Your skill with numbers is undeniable and impressive. My esteem for you has risen lately. But there are variables involved in the polygraph that you just can't figure out, no matter how you manipulate the statistics. Why does the polygraph work so well in the real world? I don't have all the answers, but from my experience I have seen time and time again that it does.

Bill & Digithead,

LBCB actions and responses are typical of those that have no other options but to support their group or organizations. This dedication to supporting fallacy is very similar to those that support cults. The similarities to this mindset are discussed in a book called, 

"Combatting Cult Mind Control", 
by Steve Hassan.

Link: http://www.freedomofmind.com/

A great read by the way ... alot of insight into countering mind control and counter interrogation techniques.

Even though he has freedom of thought and viable alternate options, LBCB still sticks steadfast to his polygraph cult. 

LBCB,

A 10% Collateral Damage rate is not acceptable under any circumstances, if we use your (highly doubtful) 90% accuracy rate. Especially when peoples lifes are being destroyed. But again its not the tool but the technician that must accept the consequences of their action. I still believe you have never run up against a dedicated and prepared person, and if you have you were beaten. And you never even knew it.

Regards ...

LBCB's predicted response of 90% is 90% reminds me of a scene from an old movie called "This is Spinal Tap"

The guitar player for a band is trying to explain how their amps are louder because their volume knob is marked 1-11 instead of the usual 1-10 so they can crank it up an extra notch. When he is told that having 11 increments instead of 10 doesnt mean your amps are any louder, the guitar player goes blank with a dumb look on his face, pauses for a moment and says....

"these go to 11." 

http://members.aol.com/chiprowe/gotoeleven.wav
http://members.aol.com/chiprowe/gotoeleven2.wav

One final thing before I completely discharge this topic. 

Lost in this discussion of hypothetical accuracy and arguments about the true base rate of deception are these indisputable facts:

If the base rate of true positives is less than 50%, your false positive rate goes up. In other words, the rarer true positives are, the greater the number of false positives.

Conversely, if the base rate of true positives is greater than 50%, your false negative rate goes up. In other words, the more common true positives are, the greater the number of false negatives.

Simply put in our context, the rarer the rate of the deception, the greater the number of falsely accused. The more common the rate of the deception, the greater the number of falsely exonerated.

And unless someone here has the skill to rewrite the laws of probability, these are indisputable facts.

digithead wrote on Dec 8^th, 2006 at 10:49am:


So you're going to try and tell us dowsing isn't real, either?  Now you've gone too far!

LieBabyCryBaby wrote on Dec 8^th, 2006 at 3:19am:


And my retort to people who don't understand the difference between the field of Statistics (the science of describing uncertainty) and a statistic (a scalar resulting from a function) is this: Tell me why is lying with numbers is worse than lying with words?

LieBabyCryBaby wrote on Dec 8^th, 2006 at 3:19am:


Simple question - if you don't know how many people are deceptive in your population, how can you estimate any accuracy?

LieBabyCryBaby wrote on Dec 8^th, 2006 at 3:19am:


You can maintain it all you want, but to paraphrase Richard Feynman - nature cannot be fooled... 

LieBabyCryBaby wrote on Dec 8^th, 2006 at 3:19am:


Given your prior postings regarding others' honesty on this board, you're being disengenuous...

LieBabyCryBaby wrote on Dec 8^th, 2006 at 3:19am:


Even if another agency does not know about a prior failed polygraph, the person does and is hopelessly compromised for future polygraphs. You cannot deny this...

LieBabyCryBaby wrote on Dec 8^th, 2006 at 3:19am:


And I've explained to you the concept of wishful thinking yet you ignore it. You also fail to recognize many of your cognitive biases including: 

• Confirmation bias - you only include information that supports your preconceived notions.

• Selective perception - you let your expectations dictate what you perceive. 

• Self-serving bias - you claim more responsibility for your successes than your failures.

LBCB, no amount of evidence can persuade you of the folly of CQT because you are so entrenched in the field of polygraphy that you cannot admit that you're wrong. Similar stubborness exists in believers in other pseudosciences such as dowsers, psychic detectives, and homeopaths. Yet these people still find followers despite the evidence against it. Unfortunately, some of our species have a desperate willingness to embrace wishful and magical thinking. So I'll politely end our discussion because you're hopeless. I only hope that your willing credulity regarding the polygraph does not extend into other parts of your life... 

Regards...

Sergeant1107 wrote on Dec 8^th, 2006 at 3:51am:


Sergeant,

Good point.  I mis-stated.  I meant accused child molesters, not proven child molesters. Sorry, my mistake.

LieBabyCryBaby wrote on Dec 8^th, 2006 at 3:19am:

If you believe this you have never had to interview any child molesters.  All of them lie.  Without exception.

Some of them mix in a very few truthful statements with their lies, and others mix in a great many truthful statements with their lies.  But all of them lie.

I have a very simple question:

Where do polygraph examiners get their percentages from when they claim, "The polygraph is 90% accurate"?

If an examinee fails the polygraph and there is no overwhelming evidence to prove they were actually being truthful does that result become part of the 90% figure?  Is the assumption in such a case that the polygraph was accurate since no one can prove otherwise?

Or is that the assumption all the time?  That the polygraph is always accurate unless there is overwhelming proof that it wasn't accurate?

I suppose there are rare cases where, after an individual is scored as "DI" on a polygraph, some persuasive evidence comes to light to prove that he was actually telling the truth during the test that resulted in the "DI" score.   

I would imagine that there are far more cases similar to my first three polygraph exams, where the examiner says it is crystal clear to them that I am lying about a particular subject, and it is virtually impossible for me to prove otherwise.  How can a person prove that they never used cocaine or sold cocaine?  How can you prove that something didn't happen?  And why should you have to if the only "proof" it did happen is the guesstimate of a polygraph examiner?

This well-known saying is part of a phrase attributed to Benjamin Disraeli and popularized in the U.S. by Mark Twain:
"There are three kinds of lies: lies, damn lies, and statistics."
The semi-ironic statement refers to the persuasive power of numbers, and succinctly describes how even accurate statistics can be used to bolster inaccurate arguments.

Digithead,

I took this from Wikipedia because I think it is applies here.

Your statistics aren't that difficult to understand, so don't flatter yourself. But you are indeed making a simple concept difficult.

If I have a process that is correct 90% of the time, I am not looking at PPVs or NPVs or MVPs or VIPs.  I'm not assuming anything about the sample of examinees we're talking about--not how many are truly deceptive or how many are not. What I do assume, based on studies used by "pro-polygraph" people and that support my own experience, is that the polygraph is correct almost all of the time. Set that "almost" at 90% or 80% or even 70%, and we can manipulate the statistics, playing with the theoretical base rates ad infinitum. But if I'm talking about 100 examinees and throwing out the inconclusives that we can't count as anything, what we have left is a 90% accuracy rate for all of those examinees, regardless of how many are actually false positives or false negatives.  Why make it more complicated? Whether the examinees are all truthful or they're all a bunch of liars, I'm right 9 out of 10 times.  Ooooh, I just converted that 90% to 9 out of 10, follow me? I can't assume that 99% of all child molesters are liars any more than I can assume that only 1% of job applicants are liars. If I start to make those assumptions, I can manipulate the statistics in . . . well, you figure how many ways.  

Simply stated, if I test 100 examinees, throwing out any inconclusives, what I'm left with is 9 out of 10 correct.

Throw up all the smoke and mirrors you want, but underneath and behind it all, things are what they are, and I maintain that the polygraph, while imperfect, is almost always right.

I do sympathize with those people who are truly false positives. I know they must be out there even if I haven't come across many, if any, in actual exams. And I also agree that a failed polygraph with one agency should not follow a person around. If agencies are so confident in the process that is claimed to be 90% accurate when conducted by a competent examiner, I say let them run their own exam without prejudice.

 

LieBabyCryBaby wrote on Dec 7^th, 2006 at 8:35pm:


Good catch, my mistake and I fixed the error, it now reads "right only 1 out of 12 times." We all have our shortcomings. Perhaps you should consider a career in technical editing...

And positive predictive value is a simple concept. You just don't like the results...

LieBabyCryBaby wrote on Dec 7^th, 2006 at 8:35pm:


Except that people don't have a recourse from which to recover from a failed employment polygraph except to find work outside of LE. It would be fairer to run a lottery, at least then prospective employees would know it was really based on chance rather than some illusion of accuracy...

LieBabyCryBaby wrote on Dec 7^th, 2006 at 8:35pm:


That's because agencies are staffed by humans who can be fooled and can also fool themselves into magical thinking. The lure of Pinnochio's Nose is great except nature didn't equip us with some specific physical lie response. No one has ever proven otherwise...

LieBabyCryBaby wrote on Dec 7^th, 2006 at 8:35pm:


Except GKT is grounded in science and CQT is grounded in wishful thinking...

LieBabyCryBaby wrote on Dec 7^th, 2006 at 8:35pm:


And astrologers, dowsers, and mediums are all experts in their respective fields. It doesn't mean that I'd go to any of them for advice... 

Plus ballistics and fingerprints have science behind them. And eyewitness testimony has been proven unreliable unless its corroborated with other evidence...

LieBabyCryBaby wrote on Dec 7^th, 2006 at 8:35pm:


Duly noted, I'm working on identifying it. GKT and Event-related potential seem promising...

LieBabyCryBaby wrote on Dec 7^th, 2006 at 8:35pm:


I never failed the polygraph. Indeed, I've never taken a polygraph. However, I do have solid background (B.A. in Statistics, M.A. in Biometrics, and finishing a Ph.D. in Criminal Justice) in criminology, statistics, program evaluation, and research so that does make me an expert at evaluating statistical accuracy of criminological instruments...

I see that since you can't seem to rebut the statistics other than framing it as "opportunity costs", you'll resort to your ad hominem attacks...

Quote:


Actually, that would result in 82% accuracy.  .9 * .9 + .1 *.1 = .82
You'd have to accuse them all of lying to get 90% accuracy.

wow. I am kind of shocked that you dont understand statistics. 

what he is saying is that the higher % of people you bring in to polygraph that are actually lying, the less impressive a 90% accuracy rate is.  Do you realize that if you bring me 10 people and 9 of them are in fact, lying about the relevant issue, I could put a beanie on their head, tell them its a mind reader, send 1 guy home at random and accuse the other 9 of lying because of my magic beanie, and i'd be right 90% of the time. Its easy to look good with a biased sample in other words.

If you are calculating accuracy from anecdotal evidence of 90 out of 100 people you accuse of lying confess, then really the compliments go to the investigators who almost always bring you the correct person to interview, not to your machine. Now, I will not argue that the poly is a fabulous interrogation prop to help get confessions. It clearly is that. 

If you are talking about specific incidents, its probable  that by the time a person is asked to come in for a polygraph, they are suspected based on other evidence and you know this fact. I'll bet if people were chosen randomly off the street and you didn't know who the suspected person was, your accuracy would be much lower.

I'd also bet that if I were an experienced LEO and interrogator, that I could pick up on "lying cues" without a polygraph much of the time and simply skew the test results or outright lie about them to get a guy to confess based onthe polygrpah busting him when in fact, it was my skill as an interrogator.

Right. Good one, Twoblock.

Digithead, I don't have "math phobia." But I do dislike math when the whole point of it seems to be to make a simple concept more difficult for the average person to understand. First you say the process is accurate just 1 out of 12 times, and now you say it is wrong one out of 12 times, thereby inferring that being wrong that often is necessarily a bad thing.

I know it's difficult for you to accept when I talk about the "cost of doing business." If I were one of those true false positives, the cost would definitely be harder to swallow. But I've seen it from both sides--as an examinee and as an examiner. Now that I can see it from both sides, it doesn't make me happy to see people fail the polygraph. Fortunately for me, when they've failed it's almost always been easy to figure out due to admissions, and due to the fact that when you are right, people don't put up much of an argument when they fail. It's not that hard for an experienced polygrapher or investigator to figure out.

The point I will again try to make here is that the agencies consider the polygraph a good screening tool. Some might ignorantly go to far, as I have witnessed myself, considering the polygraph to be completely infallible. But most intelligent people involved with the hiring process, including polygraphers, know that the polygraph is not perfect, but simply right most of the time.

I agree that the GKT is more accurate, generally, than the CQT. Neither is 100%, though. But they are excellent tools.

Fingerprinting, ballistics, eyewitness accounts, etc. are all dependent on data collection and a judgment call by an expert. Polygraphers are experts in their field, whether polygraphy is a 100% process or not. You may be an expert in mathematics, or so it seems. I am an expert in polygraphy. If I were the casual reader, I would be very interested in your theories, but I would be more interested in knowing polygraphy from the mind of a polygrapher.

Unless you can come up with something better than the imperfect investigative tools that we have now, including the polygraph, you're not doing anybody much good in your proclaimed life's pursuit.

I advise the casual reader of this forum to take any advice given here, even by polygraphers, with a healthy skepticism. But I also caution casual readers to avoid playing mindgames with themselves simply because a few people who failed the polygraph get on this website and pose as experts.

Digithead and LieBabyCryBaby

I have enjoyed your debates although it's hard to understand your high -powered mathmatical calculations. Us dummies are stuck with calculating things like the diffusivity of carbon atoms per minute through each unit cell of a steel bar  at 1000 degrees C. 

And the answer is (drum roll):

  3 x 10 to the -11th power meters squared per second!

See. I can't even figure out how to use the tool bar for scientific notations.

LieBabyCryBaby wrote on Dec 7^th, 2006 at 4:02am:


Wow, I must've hit a nerve. 

I guess you suffer from math phobia but I was unaware that addition, subtraction, multiplication and division were considered "complicated math."

LieBabyCryBaby wrote on Dec 7^th, 2006 at 4:02am:


Except you are making the base rate rate fallacy. In the real world, we never know the truth beforehand, so the real measure of polygraph employment screening accuracy is whether the prospective employee is deceptive given that the test is positive. With your mythical world of 90% accuracy, I have clearly shown you that when the base rate is low the majority of positives will be false. This will result in too many people wrongly accused of lying with no recourse to clear their names. How can you defend this as a cost of doing business? How can you continue to claim that something is accurate when it's right only 1 out of 12 times for the very thing it was designed to detect? And given that we're dealing in hypotheticals, I contend that the base rate of deception is even lower and that polygraph accuracy is nowhere near 90%, causing the false positive rate to be even more dramatic than this example. 

And when this is spun into the context of sex offender screening, you have no defense against the 10% false negative rate. They're simply a danger to the community rather than "a cost of doing business" or "an acceptable loss."

LieBabyCryBaby wrote on Dec 7^th, 2006 at 4:02am:


Sorry, you're still making the base rate fallacy. But continue your handwaving it eases your cognitive dissonance...

LieBabyCryBaby wrote on Dec 7^th, 2006 at 4:02am:


You're using misdirection again. Fingerprinting, ballistics and eyewitness accounts are examples of pattern recognition. Similarities in patterns, false memories, or fuzzy data account for their mistakes. These can also be verified with other evidence. CQT Polygraph is testing for the presence of deception based on a comparison to a probable lie. Not similarities in swirls, ridges, or descriptions. It requires a different standard in which to evaluate its accuracy.

LieBabyCryBaby wrote on Dec 7^th, 2006 at 4:02am:


I concur that law enforcement is a very tough job. Which is why I moved from the field into academia because I saw all of the junk being used and felt that I could do a better job at getting rid of it via good research and evaluation.

And somebody (Lykken, RIP) did invent something better than CQT polygraph. It's called the guilty knowledge test and it rests on the scientific bedrock of cognitive response rather than CQT's emotional response.

Trust me, LBCB, I'm doing all I can to improve law enforcement and crime control policies. It may take me my whole career, but I hope to be able to dump CQT polygraph onto the trash heap of phrenology and other pseudosciences once used in law enforcement and criminology.

ETA: Change "wrong" to "right only". My mistake...

Digithead,

You can be condescending and spout all the complicated math that you want, but it makes no difference. It's all very basic, not complicated. If we test 100 people, and we are correct 90% of the time, we can be wrong 10% of the time and it doesn't matter. From the viewpoint of the powers that be in most agencies, a screening process that gets it right 90% of the time, or even 80% of the time, does what it is supposed to do--it screens. 

So, let's assume that out of that 100 people we get 90 right and 10 wrong. Of course any agency is going to hope that all of those 10 wrong are false positives, not false negatives. Why? Because that is simply 10 people who don't get the job. But if 10 false negatives get the job, then an agency has 10 out of 100 people on the job who got through the process despite having various relevant issues to hide--the issues the agency cares about most. Screening out 10 false positives is simply the cost of doing business, you see. What they don't want are criminals working for them. Where those 10 false positives came from, there are plenty of other qualified candidates to choose from. If you want to raise those 10 false positives to 20, then it's not as attractive, but still an agency will have 80% of its employees who are the type of employees the agency wants.

Now, of course we want to look for the most accurate screening methods we can find. But looking and finding are two different things. While we are looking, we use what we have. Before DNA, we used what we had, despite whatever inaccuracies, simply because we needed to use something, right? And I still don't think law enforcement agencies are ready to give up on fingerprinting or ballistics or eyewitness accounts, for example, simply because they aren't 100% accurate.  Neither are they willing to give up on the polygraph when it is one of those useful tools and the best thing currently available.

Finally, one million people doing the wrong thing doesn't make it right. But when those one million people are doing a very tough job, they use the best tools they have, and they'll keep using them until someone invents something better. Got any inventions in mind, Digithead?

LieBabyCryBaby wrote on Dec 6^th, 2006 at 7:53pm:


Nope, it's not impressive, it just requires the ability to do arithmetic. Understanding fractions and percentages would help too...

So to reiterate, they are called positive predictive value (PPV) and negative predictive value (NPV).

PPV is the probability that a person has the condition given that the test is positive. 

With your example of 90% accuracy and assuming 1% of the population is deceptive and 99% are truthful. If we have 1000 examinees, this means 10 are deceptive and 990 are truthful. 

That means you will have .9x10=9 true positives and 1 false negative. Notice the sneaky conversion from percents to decimals.

It also also means you will have .9x990=891 true negatives and 990-891=99 false positives. Ah, substraction, addition's tricky friend.

So the total of true and false positives in this example is 9+99=108 total positives. Are you still with me?

That means your PPV=9/108=8.3% probability that the person is deceptive given that the test is positive. Great accuracy if the test is positive.

Warning, division coming up. 

In other words, 99/9=11 people will be falsely accused for every person correctly identified. To put it another way, 1 out of every 12 positives (11 false + 1 true positive) will be correct. 

That's some fancy arithmetic, wouldn't you say? 

If the base rate is 99% instead, the numbers come out the same for false negatives.

LieBabyCryBaby wrote on Dec 6^th, 2006 at 7:53pm:


I absolutely agree. But we can pursue more accurate methods, discard ones that don't work, seek supporting evidence that in totality reduces error and continously improve the system through science. The CQT polygraph is not based on science and will never get more accurate, therefore it should be discarded...

LieBabyCryBaby wrote on Dec 6^th, 2006 at 7:53pm:


If one million people do a foolish thing, it's still a foolish thing...

Digithead,

I simply don't follow your numbers. Sorry. If a process is accurate 90% of the time, that doesn't equate to being correct only 1 out of 12 times. Perhaps you have a theorem that accounts for this. If not, I'm sure you will at least make it sound impressive.

Here's an interesting article about forensic "science."

http://men.msn.com/articlepm.aspx?cp-documentid=808224&GT1=8883

Most people don't realize it, but many of the forensic tools used in police work aren't as accurate as shows like "CSI" would have us believe.  There aren't very many of them that you could stake a case on and be 100% sure of making the right call. As a district attorney is quoted in the article, "Hair analysis, fiber analysis, bite marks--you don't want to base too much of a case on those. Some prosecutors succumb to the temptation to rest their case on a fiber or a hair. But a good case is made up of a bunch of little things."  Even fingerprints are said to be inaccurate a significant percentage of the time.

But would we throw out these methods that are not 100% accurate, and use eyewitness testimony alone?--Which, by the way is also nowhere near 100% accurate.

I will agree with any "anti-" person on this forum that polygraph charts alone should not determine guilt or whether a person should be hired for a job. But knowing from experience that the polygraph is usually right, I would also argue, as many agencies do, that we should keep it as a useful tool, despite the fact that it is not 100% accurate. Remember, those of you who claim to be "false positives": In law enforcement you use the best tools you have until something better comes along. If you pass the screening process and get the law enforcement job you want, you will in fact be using many of those same tools that are not 100% accurate, thereby creating your own "false positive" victims while you're right most of the time, not all of the time. Ironic, then, that many of you who want those law enforcement jobs are sitting here arguing against an imperfect law enforcement tool that is widely accepted, by many, many people and agencies, as one of those good tools.