posted 05-13-2008 08:12 PM            

The statement sounds overly simplistic, and again, there is new evidence - from what I hear - that polygraph is "working" depending on what you're looking at.

posted 05-13-2008 10:45 PM            

There are some non-dumb and educated people who don't like the polygraph. They think themselves better equipped to "know."

Digithead seem to me to write like a criminal justice researcher. He, like Sergeant1107, seems to have a deontological (vs utilitarian) ethical slant. That's a little odd for people in CJ, but less so with more academic folks.

He's wrong about Colorado moving towards testing DV offenders. There may be some small-scale testing. After evaluating the issues, the DV board is not moving towards polygraph requirements. Probation might want to, but DV cases are misdemeanors in CO. What this really means is that we are on the short side of the cost-benefit equation for the offender. If it costs less to kill 2 years in jail, they simply do the time.

What digithead is really showing us is the eagerness of ATSA folks to chip away at the polygraph. They know that all it will really take is a steady stream of minor and mildly discouraging findings to add up to a reasonable conclusion that the polygraph is not worth the trouble.

The less hear is that we should be doing more study. Not utility/disclosure study, because they don't really care about that. Most of what we get in terms of disclosure holds no real signal value for risk assessment or risk management. We pursue issues like increased #s primarily to satisfy to ourselves that the test works. The real signal value is in a couple of key issues that are defined by the content of actuarial risk prediction measures. Our real challenge involves validating the polygraph for PCSOT and mixed issues exams - and that means validating the polygraph can access data-points salient to risk predication and risk management. Not data-points which we think are important, but data-points which risk prediction researchers have identified as predictive.

The other thing we have to get right is to learn to refrain from making up things just so we have an answer, just to sound knowledgeable, or just to ensure ourselves of an authoritative voice in our profession. Making up things without construct validity, sound theory, coherent statistical measurement, and empirical validity will not win the long game. Doing our homework will help. That means we be integrating polygraph theory with cognitive and neurophysiological theories, testing theories, and validating our techniques with inferential statistics and empirical data that will be endorsable by, educated experts in the related fields of psychology, physiology, statistics and measurement, criminology, forensics, and human development.

We have to learn to refrain from building our reputations on methods that are not defensible in front of educated opponents.

.02

r

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)

posted 05-14-2008 08:15 AM            

quote:

---

Our real challenge involves validating the polygraph for PCSOT and mixed issues exams - and that means validating the polygraph can access data-points salient to risk predication and risk management.

---

posted 05-14-2008 10:02 AM         

This is the citation for the applicable study:

McGrath, R., Cumming, G., Hoke, S., & Bonn-Miller, M. (2007, December). Outcomes in a Community Sex Offender Treatment Program: A Comparison Between Polygraphed and Matched Non-polygraphed Offenders. Sexual Abuse: A Journal of Research & Treatment, 19(4), 381-393.

Here's a link to the abstract.
http://www.ncbi.nlm.nih.gov/pubmed/17914673

posted 05-14-2008 10:17 AM            

There is currently a kabal within ATSA that is holding outward some research that suggests a whole lot of odd data regarding sex offender treatment and risk. One such project (Abrams?) suggests that Offenders do not need to even admit their offense for treatment to "work." We field workers find this to be completely counterintuative----but why? Such a lack of responsibility for the offense(s) seems to run against the module of "victom impathy" exercises and the like. Data seems to suggest that such treatment is a collosul waste of time and money. Really? One of the goals of pcsot examiners is to break through denials so that offenders can "plug in" to treatment modules. Research is strongly suggesting otherwise. For examiple protection plans for denyers have been guided by the aversion of simply "being arrested/ revoked" rather than the stronger suggestion of "endangering children." I can see the sense in this, but the data elludes me at this time.

My point(s) is/are that there exists a host of polygraph opposition based on data that suggests that pcsot testing isn't the panaceas we once hoped for. Sure we catch plenty of bad behavior from the riskier folks, but what of the offenders that would otherwise be of a lower risk but have acute dynamic risk factors that are CAUSED by the current treatment programs that result in so much financial, emotional, and social strife? Treatment providers are keenly interested in this topic of polygraph worthwhiledness (my word) regarding testing ALL sex offenders as a right of passage.

I dunno, but the word on the street with the more keyed-in treatment providers is that polygraph has been "alleged" to cause great problems for a great many offenders who were of a certain make-up---that is low risk/ static 99 type "rated pg" sex offenders. A wise man (Ray) once remarked to me that he disdained such labels, but like categorizing snakes into various venom types--poisonous to small animals, poisonous to humans, and non-venomous---the behavioral spectrum begs such labeling. I agree with such labeling and I disagree at the same time. The problem is that we examiners don't have the credentials to engage in any sort of behavioral labeling whatsoever. My business degree doesn't allow me such epistemic allowances. But I digress.

Ultimately, Dan Sosnowski is directly on point to be very vigilent over/with ATSA and just which way the treatment winds are blowing. It would take very little to make polygraph into an unorthodox treatment tool if the winds blew against us much more. I predict that the treatment of sex offenders will be nearly unrecognizable in 5 years or so----as the field is changing very rapidly relative to the other treatment modalities.

my 2 cents

posted 05-14-2008 09:18 PM            

The therapists I work with find the polygraph very useful and there are numerous examinees that have been found to be engaging in re-offending type behaviors and/or re-offended. What other tool is gonna be able to assist in this area? I think the conversations on this forum neglect to mention the value of the interview process and how much valuable information is gained from conducting an effective interview. The interview along with the polygraph is a great screening tool in PCSOT testing and I sure hope it that it continues to be.

------------------
"Truth Seekers"

posted 05-14-2008 09:28 PM            

What therapists "think" isn't all that helpful. That's why these people did the research - to see if what people "felt" or "thought" was accurate. It's no different from a polygraph examiner saying his technique works because he gets confessions. The same was true with the Xerox machine.

posted 05-14-2008 09:50 PM            

As far as therapists and polygraph exams, at least in this state, we know it works because we find out valuable information via the polygraph on an almost daily basis. I, nor them, need a research project to tell us that. It works, it really works!!

posted 05-14-2008 10:29 PM            

No one knows the amazing value of pcsot as much as yours truly. I don't want to get in a horn blowing song here, but I've seen and shown the value repeatedly for my over 2ooo tests and subsequent admissions. Unfortunately, you seem to miss or not give due, the underlying point of this discussion. That we examiners and our fans don't make the rules. That there is a core group of researchers that see trends in the macro-study of post conviction sex offender testing. This core group will release their data regardless of the strongly held values of polygraph examiners and our containment team cohorts.

My point of the 5 year future changes should not be confused with the anecdotal value of polygraph utility, but instead, the polygraph effect----and there is a difference.

Look at how atsa figure heads heralded the Penal Plethiesmograph (PPG) about 7 years ago. They saw it as a panaceas (cure) for the challenge of sexual interests and/or possible sexual fixation. Turns out that PPG was very scammy---no two test results looked the same from 2 different examiners----and once more there was/is NO BASELINE studies on what gives minute amounts of groin-wood to"normal" men. Now the PPG is a pariah--a back-door test with a small, quirky following. Let's remember, the damn PPG was invented by Chech. Army to weed out suspected homosexuals---a rather demented history if ever there was such for a would- be "clinical test."
Did the PPG result in confessions? Hell yes man. Did the PPG help determine "rated pg" offenders from some "rated XXX" type of whacko paraphiliacs? Defenitely. Did the PPG stand the test of time. No---not by a long shot. Research showed that PPG results had NOTHING to do with whether a sex offender is more or less likely to reoffend. Hell, we now know that a pedophile can go an entire lifetime without touching a child, nor viewing illegal pornography. Crazy. Remember, that is the point of sex offender management---risk of recidivism and or precipitous behaviors.

On an aside, I believe that despite the immunity to the craziness we see in pcsot---and that we do become outwardly callous to the hurt----there exists a very secret hatred we have for these men. To be blunt, we sometimes reall enjoy f____ with them --even though we do it legally, it is no less sadistic. This is no secret----people like to catch bad guys----we all love that, eh? But a byproduct of that fever to nail some of these characters is that we are blinded to the true differences between punitive treatment and therapeutic treatment---the former being an oxymoron. We can sanitize pcsot all we want as polygraph examiners, we still are pushing their asses around---and it is too fun. The figure heads in ATSA are fully aware of that fact, and they further know that such sanitized bullying has limited (if any) resulting decrease in risk. I am for pcsot. I know that it isn't always productive for the offenders themselves, but it sure makes for a good dog and pony show. 5 years is a long time for ATSA CK.

posted 05-14-2008 10:48 PM            

CK remarked of the "value" of the information. Well, the people who are writing the books on treatment are the ones defining what is and what is not valuable. Sometimes what we see as valuable, is in fact just a point of humiliation for the clients----or worse yet, just trivial but dramatic information about some mixed up goofball that the therapist already assumed was untrustworthy to begin with. Sure, we uncover some real stones sometimes---revealing gems----but the upper echelon of atsa see things differently. It's not about liberal or conservative----it's about data and science. The uppers see polygraph utility as being a better deterant than it is as a test or even an information harvest (disclosures). That hurts, and those atsa higher-ups have secret and even sinister disrespect for us. The initial hurdle is to not tell ourselves that they are jealous of us. They think we are collectively, a dangerous but sometimes useful joke.

posted 05-14-2008 11:15 PM            

...re polygraph....It really does appear that the threat of a polygraph is far more potent than our amazing abilities when we are called to actually work. If that perceived threat were to diminish over time, than conventional wisdom might suggest that our true data strength grows a little weaker by the day. Look at the offenders taking their 10th or even 50th tests. They are becoming immunized to polygraph in some anecdotal examples.
The industry is slowly moving toward/recommending less polygraph than what has been previously used---just like PPG. The romance apex is nearly over, except in the slower states to adopt pcsot.
We need to change the game plan, and use a different kind of testing altogether to survive that modality.

ahem, nuero-lock

posted 05-15-2008 01:16 PM            

One thing which I believe we should learn to do is to re-evaluate the term "utility."

Just go to AP and look around, and you'll notice that our detractors smirk when we raise the term "utility." We may regard the term as a satisfactory response to questions and challenges about polygraph testing, but a lot of others do not.

A better term would be "incremental validity." It's a terms that is recognizable to scientists, and guides our thinking and discussion towards questions about whether the information we develop does or does not actually add increased validity to risk assessment and risk management activities.

While "utility" is a bit vague, "incremental validity" get right to the heart of whether it does or does not make any bit of difference, in terms of risk assessment or risk management, whether an offender has exposed himself publicly 100 times or 200 times. This is the same concern in LEPET testing; does it really make a difference, in terms of good or bad hiring decisions, whether someone used marijuana 15 times or 25 times. I think I'd be more interested in how recent that mj use was - whether the person stopped only recently, now that he's decided to try to pursue a career in LE.

I staffed a case last week, on a 25 year water dept employee who had a two year relationship with a 13-14 year old who now has his child. His history includes all kinds of questionable behavior during college: mooning, tea-bagging, and chicken-hearts - over a period of a couple of years. So we already know he doesn't alway keep his junk in his pants. The PO and therapist want to know about other exposing - whether he's reported everything. No, he has not. There is always more. Now that he's told us about 100 or so incidents during college, failing or passing a polygraph question doesn't add much of anything to what we already know, and probably doesn't mean anything anyway. If he passes, does that mean there isn't reason 125 incidents. If he fails, is that a big surprise? He's already told us (in so many words) that he's got a conditioned response potential to the exposing question or that he's going to react to that question, because he's done it a bunch of times.

He's already told us about his exposing issues. Our concern should be around possible sexual compulsivity, because some exposers tend to expose a lot. And sexual compulsivity is a data-point that you'll find in actuarial risk prediction measures.

The really informative concern under the circumstances is the range of his compulsivity, poor boundardies, and poor judgement. So, I suggested we become more interested in the things he has not yet told us about: peeping/voyeurism, public masturbation, and stealing underwear/undergarments. Involvement or non-involvement in these other sometimes compulsive behavior does improve our understanding about whether or not all the tea-bagging with the college fraternity is really an expression of a broader pattern of pathology.

To simply test him on the number of victims would add nothing. Specific victim selection behavior mightadd a lot. Does he offend against sleeping/unconscious/intoxicated/passed-out persons?

We already know that he had an ongoing relationship with his victim's whole family. Their two families worked together, played on a community softball team, and generally endorsed the relationship until the victim became pregnant.

So, we know he has the capacity to engage in and appreciate relationships with friends and family, and to pursue recreational activities with those persons.

We also know that he worked for the water dept, good solid trade, and after loosing that job due to his legal circumstances got an entry-level union job in sheet-metal HVAC. So, he's not completely dysfunctional.

We also know that he has a history of severe anxiety and depression, and he knows that he functions much better while taking medications. His family members have similar histories, several of them also take medications.

What might be more informative, in a phenomenological sense, is whether he's ever engaged in sexual contact with other underage persons (<15) since becoming an adult. As it turns out, he has. We still don't know about other sexual compulsivity behaviors.

Anyhow, all of this seemed somehow important. Perhaps the point is that the concept of "full disclosure" is increasingly viewed as silly, and doesn't seem to add much in terms of incremental validity. However, interest in some rather specific actuarial or phenomenological concern does seem capable of adding incremental validity to our risk assessment and risk management goals.

r

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)

posted 06-01-2008 07:34 PM            

Part of the reason is the sheer nonsense that a lot of treatment providers want us to test. I have had people make pre-test admissions to over 500 exposures....then the treatment provider wanted to know if there were more! (Of course, I didn't test for that)

I worry that we are stepping so far outside of our bounds that ATSA or some other organization will move away from using polygraph.

I feel strongly that the utility value of polygraph cannoty be overstated. I am results oriented and all of the research seems to be "means" oriented.

posted 06-01-2008 09:23 PM            

I think what you and Bill are saying is that the utility of polygraphy CAN be overstated. That is the problem. Investigation targets need to be realistic and meaningful (useful to people other than the examiner) whether the offender passes or fails. Part of the criticism against PCSOT is that the information and results may not add as much as we think. Or, we haven't studied the problem enough to know or prove our assertions. Others are busy trying to prove assertions that it makes no difference or even causes more problems.

The challenge will be to fit PCSOT into the information matrix so that other professionals clearly see the incremental validity that results from using the polygraph. It is "incremental validity" not utility. "Utility" only causes our oponents eyes to roll, because they've seen that is the meaningless answer we retreat to when we don't have an answer. "Incremental validity" speaks directly to the notion that information and test results can be additive to good decision processes. If it is not additive (does not lead to better decisions), there is no need to do it. So testing the limits of 500-odd incidents provides no incremental validity (no utility). The admission of 500-odd incidents does provide incremental validity - because it alerts us to a possibly compulsive offender. Sexual compulsivity is a data-point in some actuarially derived risk prediction schemes.

Sex offender treatment folks are impressed by science and data. Data for them, is like Jesus for Christians.

If you want to impress sex offender treatment folks speak the language of science and data.

r

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)

posted 06-02-2008 12:18 AM            

I don't commonly do PCSOT testing,but on the face of it DigitHead is wrong ! Why ?

1. If No PCSOT Testing was done there would MOST CERTAINLY be more sex offenses that would occur prior to the perpetrator's second conviction.

2. The option of revoking the offender is always on the table from either the treatment provider or probation officer, therefore that is a clear deterant to at least some of the offenders.

3. The very conviction itself and social labeling of the person as a new legal class of registrant sex offender (Once out of the denial stage) is a deterant to at least some of the offenders, this given how they are treated in prison should they have to serve any sentence at all.

I'm sure there are other arguments that could be made, but I think Digithead's main argument is that the craving for sex is not reduced--------- but that is NOT OUR JOB, but rather that of the treatment provider. We can only validate how they have been spending their time, and if they are working their way up to the next victim in the cycle of deviance. Clearly then, NOT ALL will act out who would have otherwise, but certainly some will. The utilitarian use alone justifies the use of the polygraph, let alone the diagnostic value(s) derived.

Most posters who are anti-polygraph lead us into the trap of all or nothing in their arguments which is where we spend most of our time correcting them. It is further the case that newspapers sell on BAD news vs. good news. There is no way to calculate the negative in what might / could have been the case had we made no effort at all.

posted 06-02-2008 10:48 AM            

Its possible that Digithead, and perhaps the authors, fail to understand the meaning of their own study.

From memory, so check my details...

PCSOT did not reduce sexual recidivism (though that could be a type-2 problem), but did reduce non-sexual/general-criminal recidivism.

That is the point! The goal of PCSOT is not in fact to catch people reoffending, but to prevent reoffending and other non-compliance that leads to reoffending and other problems.

Another goal of PCSOT should be to improve probation and supervision outcomes - whatever that means. In my view good outcomes include moving some people into highly structured and secured settings, while re-normalizing the lifestyles of others to the extent possible. Longer sentences and tough-on-crime laws mean that probation rolls are increasing and sex offenders are stacking up in prison. The economic forces alone will cause policy makers and program managers to begin to re-evaluate and ask tough questions about what we are really accomplishing.

In the past, PCSOT programs have been used as a sophisticated form of cudgel to gain compliance. Or, as a bag of sand, to slow the process through treatment down to a nice slow and safe crawl. I've heard therapists and POs say things like "I hope he fails this polygraph." Think about that carefully. I don't believe these people are thinking about what they are are saying, because the statement implies that we 1) hope the offender is out doing dangerous, abusive or unhealthy things, or 2) we hope he fails anyway. Which means that either treatment and supervision don't work, or the polygraph doesn't work. By default, we should want everyone to follow the rules, tell the truth and pass.

If we think of offenses as occurring in some kind of cycle (an unproven assumption) or series of events (more likely), then our goal is to detect the EARLY warning signs of an escalating risk level. We would rather make a big deal out of some less-serious behavior problem. Fixing small problems is easier than fixing big problems (new victims), and can often be accomplished through treatment and probation - without a revocation or court involvement. If offenders are motivated to improve their compliance, as a result of the increased likelihood of getting caught, that is a good thing.

We have played the fear-card for a long time, and the folks at ATSA don't seem much impressed. They do know a lot about sex offenders. They also sometimes have other agenda, and seem to me to sometimes be moving in a direction of professional denial (regarding risk, dangerousness, deviancy, and victim impact), but that's another matter.

.02

r

------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)