- Welcome to AntiPolygraph.org Message Board.
A forum for free and open discussion of polygraph-related topics
Posted by EosJupiter
- Apr 13, 2006, 05:19 PMQuote from: 187Dick on Apr 10, 2006, 06:52 PMPS
Eos,
Sorry about the false accusation. It was later explained to me why my original post had disappeared from the ten most recent posts page. There was no censorship. I was relieved and humbled. Please accept my apology.
187Dick,
Apology accepted .... Keep on debating !!
Regards ....
Posted by Onesimus
- Apr 10, 2006, 09:25 PMQuote from: 187Dick on Apr 10, 2006, 06:52 PMOnesimus,
While my mathematical approach was rather simplistic, it's still relatively accurate.
Here is a more detailed approach I took to a similar problem:
numbers game
Posted by 8675309
- Apr 10, 2006, 09:18 PMQuote from: 187Dick on Apr 07, 2006, 01:11 AMFortunately, the countermeasures you keep advising people to use are not too difficult for most experienced examiners to detect.
Care to elaborate on how you can tell the difference in a readout between a high reading due to lying/withholding information, and someone artificially causing those results?
I would throughlly enjoy hearing your explanation. Examiners say that if you are lying, you will react by breathing heavier, sweating and/or elevated pulse rate. If the examinee were to produce these on their own, how can you tell the difference?
If you choose to answer, please be specific.
Posted by 187Dick
- Apr 10, 2006, 06:52 PMOnesimus,
While my mathematical approach was rather simplistic, it's still relatively accurate. If the polygraph worked on 50% of the population, that would be the same as not working. 50% would represent an equal guess right or wrong on yes or no answers. (The average.) If the polygraph were only that accurate, I would be siding wih almost everyone esle on this site in regards to the elimination of the polygraph exam as part of a security background check, pre-employment screening and even criminal investigations. I can tell you that my opinion of the CVSA is exactly that. It's a piece of junk and does not work. The polygraph as used by a well trained and ethical examiner is a different story.
If the polygraph is working 90% of the time, then it stands to reason that 9 out of 10 people who are administered an exam would be detected by a good examiner. The failures would be divided between false positives and false negatives, hopefully heavily weighted toward false negatives. Now if the polygraph is used properly as an investigative tool (which I will admit is not always the case), then in all cases where the other investigative steps failed, (whether to clear or target subjects), the polygraph would be a useful tool in putting the investigation back on the right course.
Of course the argument to that is that it would also add investigative scrutiny to completed investigations where the subject had passed the investigative process, but had failed the polygraph exam.
Another factor that I don't like to stress, but admit is something to consider, is the deterrent factor. I know that many years ago when I was in college, I was tempted to participate in illegal drug use with many of my friends. Even if I had the attitude back then that I would quit when I finished college, I also knew that if I wanted a career in law enforcement, I would most likely have to submit to a polygraph examination during the background investigation. That piece of knowledge kept me away from even minor drug use in college.
I bring that up because I may have gone down the same slippery slope that so many do when they use illegal drugs. They don't intend to end up where they do, but so many of them do. I believe it's the same with most cases of espionage in this country. Most cases start out small where some entity exploits a weakness in someone who already has the security clearance. Once that step is taken, the person becomes a slave to that entity with no chance of turning back or saying no. Hopefully, the possibility of an annual polygraph exam will keep many from taking that first step in the wrong direction. One thing about deterrence though, and why I don't like to use it as an argument in favor of polygraph exams, it can never be measured. We don't know if it worked one time or a thousand times. Other than my own experience, I can not offer a single specific incident where I can say it absolutely had that effect. But the logic of why I believe it does work is still worth mentioning.
We will talk again.
PS
Eos,
Sorry about the false accusation. It was later explained to me why my original post had disappeared from the ten most recent posts page. There was no censorship. I was relieved and humbled. Please accept my apology.
While my mathematical approach was rather simplistic, it's still relatively accurate. If the polygraph worked on 50% of the population, that would be the same as not working. 50% would represent an equal guess right or wrong on yes or no answers. (The average.) If the polygraph were only that accurate, I would be siding wih almost everyone esle on this site in regards to the elimination of the polygraph exam as part of a security background check, pre-employment screening and even criminal investigations. I can tell you that my opinion of the CVSA is exactly that. It's a piece of junk and does not work. The polygraph as used by a well trained and ethical examiner is a different story.
If the polygraph is working 90% of the time, then it stands to reason that 9 out of 10 people who are administered an exam would be detected by a good examiner. The failures would be divided between false positives and false negatives, hopefully heavily weighted toward false negatives. Now if the polygraph is used properly as an investigative tool (which I will admit is not always the case), then in all cases where the other investigative steps failed, (whether to clear or target subjects), the polygraph would be a useful tool in putting the investigation back on the right course.
Of course the argument to that is that it would also add investigative scrutiny to completed investigations where the subject had passed the investigative process, but had failed the polygraph exam.
Another factor that I don't like to stress, but admit is something to consider, is the deterrent factor. I know that many years ago when I was in college, I was tempted to participate in illegal drug use with many of my friends. Even if I had the attitude back then that I would quit when I finished college, I also knew that if I wanted a career in law enforcement, I would most likely have to submit to a polygraph examination during the background investigation. That piece of knowledge kept me away from even minor drug use in college.
I bring that up because I may have gone down the same slippery slope that so many do when they use illegal drugs. They don't intend to end up where they do, but so many of them do. I believe it's the same with most cases of espionage in this country. Most cases start out small where some entity exploits a weakness in someone who already has the security clearance. Once that step is taken, the person becomes a slave to that entity with no chance of turning back or saying no. Hopefully, the possibility of an annual polygraph exam will keep many from taking that first step in the wrong direction. One thing about deterrence though, and why I don't like to use it as an argument in favor of polygraph exams, it can never be measured. We don't know if it worked one time or a thousand times. Other than my own experience, I can not offer a single specific incident where I can say it absolutely had that effect. But the logic of why I believe it does work is still worth mentioning.
We will talk again.
PS
Eos,
Sorry about the false accusation. It was later explained to me why my original post had disappeared from the ten most recent posts page. There was no censorship. I was relieved and humbled. Please accept my apology.
Posted by Onesimus
- Apr 08, 2006, 03:35 AMQuote from: 187Dick on Apr 07, 2006, 01:11 AMIf the polygraph works accurately on 90% of the population, then added to other investigative measures, you have just reduced 90% of the failures of systems in place.
Care to elaborate on this point? Specific numbers like Drew used in his spy example would work. (hint: If the polygraph were only 50% accurate, you would reduce 0, not 50, percent of the failues of systems in place. You can't just go from 90% accuracy to 90% of failures removed.)
Posted by Fair Chance
- Apr 08, 2006, 02:02 AMGentlemen,
As a soldier on the front lines, I will admit that the polygraph usage is increasing in the Federal sector. I have admitted as much in many of my past postings. I have no delusions that the polygraph is going away.
That said, I assure you that prescreening polygraph usage in today's FBI is killing a huge crop of applicants that would have been accepted in the 1970s and 1980s.
Good scientific minds that can think "out of the box" and "connect the dots" are not even applying because they give no validity to any agency that uses the polygraph for employment decisions.
The private sector unemployment is down and competition for top talent is up. This is the worst case scenario that I have predicted for the last two years.
The FBI is not able to get the people it needs to accomplish its mission and the polygraph is an anchor being dragged on the bottom by a ship that needs to sail at full speed.
187Dick,
Thank you for your civil discussion. Only in extreme cases have I found this site censored in anyway.
Regards.
As a soldier on the front lines, I will admit that the polygraph usage is increasing in the Federal sector. I have admitted as much in many of my past postings. I have no delusions that the polygraph is going away.
That said, I assure you that prescreening polygraph usage in today's FBI is killing a huge crop of applicants that would have been accepted in the 1970s and 1980s.
Good scientific minds that can think "out of the box" and "connect the dots" are not even applying because they give no validity to any agency that uses the polygraph for employment decisions.
The private sector unemployment is down and competition for top talent is up. This is the worst case scenario that I have predicted for the last two years.
The FBI is not able to get the people it needs to accomplish its mission and the polygraph is an anchor being dragged on the bottom by a ship that needs to sail at full speed.
187Dick,
Thank you for your civil discussion. Only in extreme cases have I found this site censored in anyway.
Regards.
Posted by Drew Richardson
- Apr 07, 2006, 06:06 PM187Dick,
In order for screening exam/confirmatory exam combination to be effective, several things have to occur: (1) the screening exam has to be highly sensitive (i.e. will detect espionage, etc. in the case of a national security polygraph exam), (2) the confirmatory exam has to be highly specific (will never lead to an innocent person being left a suspect or accused of espionage, etc. in the case of a follow-up investigation following national security polygraph exam, (3) the screening exam should always be paired/confirmed by a confirmatory exam, and (4) the confirmatory exam should be orthogonal (independent) of the screening exam and should be able to be performed in a reasonable and timely manner so as to quickly confirm or deny the screening result.
Polygraph exams and follow-up investigation if viewed as such a pair fail on every count, but it is the fourth point I would like to elaborate on for a second. Far from being independent, the investigation is frequently driven by the polygraph result (e.g., in the Dr. Thomas Butler case (https://antipolygraph.org/forum/index.php?topic=1400.msg11116#msg11116) how much effort was devoted to looking at other sources for the cause of the missing vials of plague following the polygraph exam results?? --the vials were never recovered and Butler was exonerated on most of the plague charges) or the (changing) investigative theory will often affect the polygraph exam outcome (e.g., public reporting of the Wen Ho Lee case (https://antipolygraph.org/forum/index.php?topic=429.msg1989#msg1989)). If you believe that you have the background to discuss screening/confirmatory testing as appropriately used in the physical sciences, I would be happy to compare and contrast these and the aforementioned polygraph/further investigation model with you.
Because of such problems being left unaddressed by the (at first blush reasonable and sufficient) pairing of a polygraph result with further investigation , the innocent will often be left hanging in the wind for a long if not indefinite period of time and the guilty may be able to continue in whatever nefarious activity (e.g., Aldrich Ames, Gary Leon Ridgway a.k.a. Green River Killer) with impunity.
In order for screening exam/confirmatory exam combination to be effective, several things have to occur: (1) the screening exam has to be highly sensitive (i.e. will detect espionage, etc. in the case of a national security polygraph exam), (2) the confirmatory exam has to be highly specific (will never lead to an innocent person being left a suspect or accused of espionage, etc. in the case of a follow-up investigation following national security polygraph exam, (3) the screening exam should always be paired/confirmed by a confirmatory exam, and (4) the confirmatory exam should be orthogonal (independent) of the screening exam and should be able to be performed in a reasonable and timely manner so as to quickly confirm or deny the screening result.
Polygraph exams and follow-up investigation if viewed as such a pair fail on every count, but it is the fourth point I would like to elaborate on for a second. Far from being independent, the investigation is frequently driven by the polygraph result (e.g., in the Dr. Thomas Butler case (https://antipolygraph.org/forum/index.php?topic=1400.msg11116#msg11116) how much effort was devoted to looking at other sources for the cause of the missing vials of plague following the polygraph exam results?? --the vials were never recovered and Butler was exonerated on most of the plague charges) or the (changing) investigative theory will often affect the polygraph exam outcome (e.g., public reporting of the Wen Ho Lee case (https://antipolygraph.org/forum/index.php?topic=429.msg1989#msg1989)). If you believe that you have the background to discuss screening/confirmatory testing as appropriately used in the physical sciences, I would be happy to compare and contrast these and the aforementioned polygraph/further investigation model with you.
Because of such problems being left unaddressed by the (at first blush reasonable and sufficient) pairing of a polygraph result with further investigation , the innocent will often be left hanging in the wind for a long if not indefinite period of time and the guilty may be able to continue in whatever nefarious activity (e.g., Aldrich Ames, Gary Leon Ridgway a.k.a. Green River Killer) with impunity.
Posted by George W. Maschke
- Apr 07, 2006, 03:48 AMQuote from: 187Dick on Apr 07, 2006, 01:51 AMEos,
I noticed the posting you took the quote from disappeared from the last ten posts on the site. I did not realized this site censored opposing opinions. It shows you for what you really are.
Your post was not censored. When you view the "most recent posts" link, what is displayed is not the last 10 posts to the message board, but rather the last post in each of the 10 most recently updated discussion threads. Mr. Mystery's post automatically replaced yours as the most recent post for this discussion thread.
Posted by EosJupiter
- Apr 07, 2006, 03:43 AMQuote from: 187Dick on Apr 07, 2006, 01:51 AMEos,
I noticed the posting you took the quote from disappeared from the last ten posts on the site. I did not realized this site censored opposing opinions. It shows you for what you really are.
187Dick,
As I have no idea of what your talking about ? I can't or won't delete posts unless they are my own. Sorry I am not an administrator. And if you got something to say, then say it straight out. As I do know what I am ....
If someone is deleting your posts, check elsewhere, as anyone else on this board knows me much better than that. I do use snippets of posts in my posts, but no more.
So enough of this nonsense ...
Regards
Posted by 187Dick
- Apr 07, 2006, 01:52 AMQuote from: 187Dick on Apr 07, 2006, 01:11 AMDrew,
Your numbers are a little low compared to my research on valid and reliable studies of single issue tests; however, let's take your numbers and look at them a different way. If the polygraph works accurately on 90% of the population, then added to other investigative measures, you have just reduced 90% of the failures of systems in place. Additionally, when it comes to espionage cases, I think a blind system where the subject did not know if he passed or not, or was told he passed when he failed, and then monitored, I think the approach would be very good at screening for weaknesses in the system.
Any of these issues can be addressed, but unfortunately, the Federal Government management systems are poster examples of broken systems. There is no effort to rate effectiveness of managment approaches to various problems, which is in itself a major problem. But that's an arguement for another web site.
My point is this, any background or criminal investigation can greatly benefit by the use of a polygraph instrument with an ethical and well trained examiner. It is not a replacement for any part of a background investigation, and needs to be used as a tool to make the investigation more accurate. I have seen it also help people get into sensitive careers. (If it were not for the polygraph exam, most people with ex-spouses may not get hired.) I use that example to illustrate the fact that first hand accounts of incidents or a person's character are far less accurate than polygraph exams, and yet the reliance on them is rarely questioned.
Until a better system is created, the polygraph exam is here to stay. People will continue to come to this site who are scheduled for a polygraph exam. As more states adopt mandatory polygraph exams for child molestors who are released from prison, I'm sure the number of visitors to this site will increase. Fortunately, the countermeasures you keep advising people to use are not too difficult for most experienced examiners to detect.
As I have stated before, it's the foolish person who has never taken a polygraph examination, and takes your advice, causing them to fail the polygraph, and gets denied a job because of it that I hope to influence. (Run on sentence, I know, but I couldn't help it.) The best way to pass the polygraph examination is to tell the truth. Hundreds of thousands of people have used that tactic and have been successful.
We will talk again,
Take care.
Posted by 187Dick
- Apr 07, 2006, 01:51 AMEos,
I noticed the posting you took the quote from disappeared from the last ten posts on the site. I did not realized this site censored opposing opinions. It shows you for what you really are.
I noticed the posting you took the quote from disappeared from the last ten posts on the site. I did not realized this site censored opposing opinions. It shows you for what you really are.
Posted by Mr. Mystery
- Apr 07, 2006, 01:35 AMQuote from: 187Dick on Apr 07, 2006, 01:11 AMAs more states adopt mandatory polygraph exams for child molestors who are released from prison, I'm sure the number of visitors to this site will increase. Fortunately, the countermeasures you keep advising people to use are not too difficult for most experienced examiners to detect.
It has come full circle now. We have had a polygrapher blow in here and claim that those who dislike polygraphs are a bunch of child molesting spies. Here is my all time favorite, it is the 2nd post down.
http://antipolygraph.org/cgi-bin/forums/YaBB.pl?board=trashbin;action=display;num=1090549423;start=1#1
Posted by 187Dick
- Apr 07, 2006, 01:11 AMDrew,
Your numbers are a little low compared to my research on valid and reliable studies of single issue tests; however, let's take your numbers and look at them a different way. If the polygraph works accurately on 90% of the population, then added to other investigative measures, you have just reduced 90% of the failures of systems in place. Additionally, when it comes to espionage cases, I think a blind system where the subject did not know if he passed or not, or was told he passed when he failed, and then monitored, I think the approach would be very good at screening for weaknesses in the system.
Any of these issues can be addressed, but unfortunately, the Federal Government management systems are poster examples of broken systems. There is no effort to rate effectiveness of managment approaches to various problems, which is in itself a major problem. But that's an arguement for another web site.
My point is this, any background or criminal investigation can greatly benefit by the use of a polygraph instrument with an ethical and well trained examiner. It is not a replacement for any part of a background investigation, and needs to be used as a tool to make the investigation more accurate. I have seen it also help people get into sensitive careers. (If it were not for the polygraph exam, most people with ex-spouses may not get hired.) I use that example to illustrate the fact that first hand accounts of incidents or a person's character are far less accurate than polygraph exams, and yet the reliance on them is rarely questioned.
Until a better system is created, the polygraph exam is here to stay. People will continue to come to this site who are scheduled for a polygraph exam. As more states adopt mandatory polygraph exams for child molestors who are released from prison, I'm sure the number of visitors to this site will increase. Fortunately, the countermeasures you keep advising people to use are not too difficult for most experienced examiners to detect.
As I have stated before, it's the foolish person who has never taken a polygraph examination, and takes your advice, causing them to fail the polygraph, and gets denied a job because of it that I hope to influence. (Run on sentence, I know, but I couldn't help it.) The best way to pass the polygraph examination is to tell the truth. Hundreds of thousands of people have used that tactic and have been successful.
We will talk again,
Take care.
Your numbers are a little low compared to my research on valid and reliable studies of single issue tests; however, let's take your numbers and look at them a different way. If the polygraph works accurately on 90% of the population, then added to other investigative measures, you have just reduced 90% of the failures of systems in place. Additionally, when it comes to espionage cases, I think a blind system where the subject did not know if he passed or not, or was told he passed when he failed, and then monitored, I think the approach would be very good at screening for weaknesses in the system.
Any of these issues can be addressed, but unfortunately, the Federal Government management systems are poster examples of broken systems. There is no effort to rate effectiveness of managment approaches to various problems, which is in itself a major problem. But that's an arguement for another web site.
My point is this, any background or criminal investigation can greatly benefit by the use of a polygraph instrument with an ethical and well trained examiner. It is not a replacement for any part of a background investigation, and needs to be used as a tool to make the investigation more accurate. I have seen it also help people get into sensitive careers. (If it were not for the polygraph exam, most people with ex-spouses may not get hired.) I use that example to illustrate the fact that first hand accounts of incidents or a person's character are far less accurate than polygraph exams, and yet the reliance on them is rarely questioned.
Until a better system is created, the polygraph exam is here to stay. People will continue to come to this site who are scheduled for a polygraph exam. As more states adopt mandatory polygraph exams for child molestors who are released from prison, I'm sure the number of visitors to this site will increase. Fortunately, the countermeasures you keep advising people to use are not too difficult for most experienced examiners to detect.
As I have stated before, it's the foolish person who has never taken a polygraph examination, and takes your advice, causing them to fail the polygraph, and gets denied a job because of it that I hope to influence. (Run on sentence, I know, but I couldn't help it.) The best way to pass the polygraph examination is to tell the truth. Hundreds of thousands of people have used that tactic and have been successful.
We will talk again,
Take care.
Posted by Drew Richardson
- Apr 06, 2006, 06:14 PMPolywantahcracker,
Please excuse me if I use my last post to address a separate audience. I hope I addressed your comment satisfactorily. I would like to address any National Security officials who might peruse this site from time to time and would like to pick up where I left off:
I'd like to look at the aforementioned from the point of view of our fictional next spy. The picture I have painted is considerably more favorable towards catching our spy than really exists, but even using the previously stated assumptions our spy finds himself sitting quite pretty. At worst he is lumped into a population of a thousand and his chances of being caught are considerably reduced due to this 1 in a 1000 statistic. Both he and you (national security official) know that this population is virtually all composed of false positive polygraph results. All he has to do is to take the Aldrich Ames approach and deny deny deny. He is virtually guaranteed to be cleared based on your (National Security official) known and reasonable predisposition in the matter. What if he wants to make the game a little more difficult and put himself in a bigger haystack and one that won't even be further looked at. As I have indicated many times, a child can be taught to successfully employ countermeasure techniques. Our spy is more motivated and, no doubt more knowledgeable than most, so I would suggest to you that he could easily put himself into the bigger grouping of the 9000 true negatives (innocent employees who were found to be non-deceptive) and be the only direct false negative in this case. I hope it should be clear to you that you are playing a losing game if this is the strategy and game plan you have and currently employ. If any in your community would like to contact me directly I would be happy to discuss some ideas regarding how you might do a better job than that which has been done in the past. Regards....
Please excuse me if I use my last post to address a separate audience. I hope I addressed your comment satisfactorily. I would like to address any National Security officials who might peruse this site from time to time and would like to pick up where I left off:
Quote
Precisely. Control Question Test (CQT) polygraph exams produce a completely unacceptable level of false positive results with specific issue testing and (due to this effect) indirectly a large number of false negative (i.e., calling a spy non-deceptive in a polygraph exam) results in a national security screening exam. Let me explain. For the sake of this conversation let's say an individual polygraph exam were 90 per cent accurate (needless to say it's not and the results of what I am going to explain would be far worse than the dismal situation I will subsequently describe if we were to use realistic accuracy rates).
Let's assume we have reason to believe the FBI has its next Robert Hanssen in its group of (to the first approximation) 10,000 FBI agents. If the polygraph exam were 90 per cent accurate we would have a 90 per cent chance of having a deceptive chart for our next spy. Because it is 10 per cent inaccurate we would also falsely accuse roughly 1000 innocent individuals out of our 10,000-person population. We would (and do) have the intolerable situation of falsely accusing 1000 innocent employees while at the same time trying to find the proverbial needle in a haystack, i.e., 1 true positive (our spy) to every 1000 false positives. We would never (and have never) identify our spy based on this nonsense and would throw the routine comradery, trust, and ability to work effectively in the FBI into chaos. Because of this phenomenon, the FBI (and other similar agencies) will virtually ignore all their positive test results (avoiding the immediate threat of falsely accusing an employee--their are exceptions (false positives that is, e.g., Mark Mallah) but they will never catch a spy with this algorithm. If this whole exercise appears to you to produce a lack of national security, a serious risk for innocent employees and applicants, and a large waste of precious taxpayer-funded resources, you are as perceptive as your quoted comment would indicate. Regards...
I'd like to look at the aforementioned from the point of view of our fictional next spy. The picture I have painted is considerably more favorable towards catching our spy than really exists, but even using the previously stated assumptions our spy finds himself sitting quite pretty. At worst he is lumped into a population of a thousand and his chances of being caught are considerably reduced due to this 1 in a 1000 statistic. Both he and you (national security official) know that this population is virtually all composed of false positive polygraph results. All he has to do is to take the Aldrich Ames approach and deny deny deny. He is virtually guaranteed to be cleared based on your (National Security official) known and reasonable predisposition in the matter. What if he wants to make the game a little more difficult and put himself in a bigger haystack and one that won't even be further looked at. As I have indicated many times, a child can be taught to successfully employ countermeasure techniques. Our spy is more motivated and, no doubt more knowledgeable than most, so I would suggest to you that he could easily put himself into the bigger grouping of the 9000 true negatives (innocent employees who were found to be non-deceptive) and be the only direct false negative in this case. I hope it should be clear to you that you are playing a losing game if this is the strategy and game plan you have and currently employ. If any in your community would like to contact me directly I would be happy to discuss some ideas regarding how you might do a better job than that which has been done in the past. Regards....
Posted by Drew Richardson
- Apr 06, 2006, 03:12 PMPolywantahcracker,
You write in part:
Precisely. Control Question Test (CQT) polygraph exams produce a completely unacceptable level of false positive results with specific issue testing and (due to this effect) indirectly a large number of false negative (i.e., calling a spy non-deceptive in a polygraph exam) results in a national security screening exam. Let me explain. For the sake of this conversation let's say an individual polygraph exam were 90 per cent accurate (needless to say it's not and the results of what I am going to explain would be far worse than the dismal situation I will subsequently describe if we were to use realistic accuracy rates).
Let's assume we have reason to believe the FBI has its next Robert Hanssen in its group of (to the first approximation) 10,000 FBI agents. If the polygraph exam were 90 per cent accurate we would have a 90 per cent chance of having a deceptive chart for our next spy. Because it is 10 per cent inaccurate we would also falsely accuse roughly 1000 innocent individuals out of our 10,000-person population. We would (and do) have the intolerable situation of falsely accusing 1000 innocent employees while at the same time trying to find the proverbial needle in a haystack, i.e., 1 true positive (our spy) to every 1000 false positives. We would never (and have never) identify our spy based on this nonsense and would throw the routine comradery, trust, and ability to work effectively in the FBI into chaos. Because of this phenomenon, the FBI (and other similar agencies) will virtually ignore all their positive test results (avoiding the immediate threat of falsely accusing an employee--their are exceptions (false positives that is, e.g., Mark Mallah) but they will never catch a spy with this algorithm. If this whole exercise appears to you to produce a lack of national security, a serious risk for innocent employees and applicants, and a large waste of precious taxpayer-funded resources, you are as perceptive as your quoted comment would indicate. Regards...
You write in part:
Quote
...because naturally we are not a country full of spies....but wouldn't that bias make it so, so difficult to find a spy....
Precisely. Control Question Test (CQT) polygraph exams produce a completely unacceptable level of false positive results with specific issue testing and (due to this effect) indirectly a large number of false negative (i.e., calling a spy non-deceptive in a polygraph exam) results in a national security screening exam. Let me explain. For the sake of this conversation let's say an individual polygraph exam were 90 per cent accurate (needless to say it's not and the results of what I am going to explain would be far worse than the dismal situation I will subsequently describe if we were to use realistic accuracy rates).
Let's assume we have reason to believe the FBI has its next Robert Hanssen in its group of (to the first approximation) 10,000 FBI agents. If the polygraph exam were 90 per cent accurate we would have a 90 per cent chance of having a deceptive chart for our next spy. Because it is 10 per cent inaccurate we would also falsely accuse roughly 1000 innocent individuals out of our 10,000-person population. We would (and do) have the intolerable situation of falsely accusing 1000 innocent employees while at the same time trying to find the proverbial needle in a haystack, i.e., 1 true positive (our spy) to every 1000 false positives. We would never (and have never) identify our spy based on this nonsense and would throw the routine comradery, trust, and ability to work effectively in the FBI into chaos. Because of this phenomenon, the FBI (and other similar agencies) will virtually ignore all their positive test results (avoiding the immediate threat of falsely accusing an employee--their are exceptions (false positives that is, e.g., Mark Mallah) but they will never catch a spy with this algorithm. If this whole exercise appears to you to produce a lack of national security, a serious risk for innocent employees and applicants, and a large waste of precious taxpayer-funded resources, you are as perceptive as your quoted comment would indicate. Regards...