posted 02-28-2009 03:02 PM
I've had some offline requests for a little more information on the differences between 2/19/09 draft and the 1/11/09 draft.You can find em here:
http://www.pcsot.info/model_policy/PCSOT_MODEL_POLICY_alternate_2-19-09.html
http://www.pcsot.info/model_policy/PCSOT_MODEL_POLICY_alternate_1-11-09.html
Here are the discussion points of major difference, through which the 2/19/09 draft is able to more closely resemble the committee's position on certain things.
------------------------------------------------------
1. Exploratory vs. Screening (language) - want to minimize exposure resulting from the damaged-merchandise language of screening, resulting from examiners having told consumers that sex offender tests are not screening tests. also want to minimize exposure to scientific criticism for failing to account for the fact that PCSOT tests are in fact screening tests. declare once that PCSOT tests are screening, and use "exploratory" language throughout the remainder of the document.
A lot of us never use the term screening in public. However, avoidance of an obvious fact – that PCSOT tests are screening tests – is likely to be troublesome in the long run. It is obvious to any scientific minded persons that these are screening exams. By definition, a screening test is intended to be an expedient and cost effective test that can be given to anyone regardless of whether a know incident/allegation/concern is identified. Screening exams are fishing trips. Some screening exams are broad multi-issue (mixed issue) tests – such as the variety of blood screens or the MMPI used in police applicant screening. Some screening tests are narrowly focused (single issue) tests such as the good-'old rubber-glove prostate test. It is not the number of issues that makes a screening test, but the presence or absence of a known concern (allegation/incident/conflict).
New York recently published their research review (Jan 2009), after receiving training and instruction from Eric Holden, and stated clearly that PCSOT exams are screening exams. The fact that PCSOT exams are screening exams is a Humpty that has fallen off the wall and broken and ain't going back together again.
A polemic view of any effort to state that PCSOT exams are not screening exams would be that we are either stupid or dishonest for attempting to make such an assertion. An apologeticist's view might recognize that an attempt to state that PCSOT exams are not screening exams may be an attempt to differentiate a structured and numerically scored CQT approach from the loosely structured and unregulated practices of pre-employment and employee-periodic screening exams of the pre-EPPA era.
The “screening” nature of the test is not determined by the breadth of the RQ. For example a behaviorally descriptive RQ “physically forced anyone/else for sexual contact” is just as much a screening question as “left anything out of his written sex history" or "done anything he shouldn't have in the last 6 months,” though the second two examples are clearly broader and not acceptable.
Screening and diagnostic tests are selected and defined, in medicine and psychology, by the presence or absence of a known concern (symptoms). If we want the polygraph to be accepted as science, then we should adopt the same scientific language and concepts as other sciences and not engage in idiosyncratic polybabble dialog that makes us vulnerable to pseudoscience accusations by setting ourselves apart from basic science and other sciences.
2. POSE / PAE (language) - agreed to change the language. It is the concept that is most important and not the label. "offense" is a legal determination, "allegation" is unequivocal
3. Reoffense RQ in ME - the draft has been edited to reflect the committee's position on this - Rick and I agree (for different reasons) that it is not the best, but the committee has apparently achieved a solution to this difficult issue
4. Conflict/Allegation - emphasize the language of allegation and not conflict (attempt to avoid exacerbating perceived conflict surrounding polygraph
5. TOR/FOR as a definition - keep requirement to be separated by either TOR or FOR - RQs have FOR and TOR that is reflected in the yet unresolved discussion about the TOR = pre-conviction/pre-treatment or TOR = date of sex history polygraph.
6. Principle 5 - similar potential consequences (biggest issue of internal inconsistency) - SOLVED - the principle is inconsistent with reoffense RQs in the ME. Have a principle and stick to it or don't have a principle. Do not have a principle and then violate it.
7. CQ TOR - probably not an issue - CQs will keep requirement to be separated by either TOR or FOR - is a commonly accepted requirement
8. mandatory posttest interrogation re CQs - will include language in 10.6 to state "examiner should ask the examinee to explain any significant reaction to any test question" - does not telegraph CQ directly, and is better suited to post on the WEB for public consumption
9. Restriction to evidentiary testing techniques for screening purposes - will include no requirement, therefore validated screening techniques can be used in screening situations.
----------------------------------------------------------
The remaining issue for which I have a proposed solution is the committee's attempt to “define” a sexual history RQ as having a TOR prior to the date of the current supervision.
Firstly, matters of science are not defined in a declarative sense the way the committee would like to do; they are discovered. Moreover, matters of science are articulated in a hypothesis (fancy idea) and then supported by scientific experiments that fail to reject the null-hypothesis that the fancy-idea makes no difference.
In our discussions so far we all agreed that we would probably not really want an offender to pass the sex history re other offenses prior to beginning treatment a year ago after raping and murdering a child last month - it would look to the community and media like a false-negative polygraph failure.
If the TOR actually worked an offender could stuff a child's body in the trunk of his car, then drive to the sex history polygraph and pass questions re other offenses prior to beginning treatment. we all seemed to agree that someone who had reoffended while in treatment would probably not pass a sex history exam with a TOR prior to beginning treatment.
The TOR sex history concern is therefore NOT a matter of empirical validity or science. It IS a matter of systemic function, and the need to ensure some form of rights protection to the examinee. More importantly is it useful to comfort bleeding-heart rights advocates that the offender's rights against self-incrimination will not be violated (the TOR doesn't really do this, but simply argues the consequences are likely to be different for pre-treatment and post-treatment offenses).
This is, of course, a lot of wrangling, but the TOR an effective solution and mimetic device to help rights-advocacy types and reluctant professionals to endorse the requirements for polygraph testing. This is three layers deep. It becomes interpreted as a validity issue because the perceived potential consequences are different for pre-conviction/pre-treatment offenses compared to post-conviction/post-treatment offenses.
To the extent that consequences and fear of consequences are believed to be the underlying basis for reaction, it is MISINTERPRETED as a validity issue. Honts et al, and the Utah researchers tell us that an element of jeopardy contributes to polygraph accuracy. Those researchers also state that there is no linear relationship regarding the degree of jeopardy and accuracy - meaning that while some jeopardy mean more accuracy, even more jeopardy does NOT mean even more accuracy. This is the old correlation-causation thing (there is a correlation between cars that are fast and cars that are shiny, but even shinier cars are not necessarily even faster). If there were a linear and causal relationship between consequences/fear/jeopardy and polygraph accuracy/validity then we should expect to get the most accurate results while holding a gun to someone's head.
The TOR = pre-conviction/pre-treatment model is a solution to the right/advocacy/resistance issue. It is not the only solution.
The TOR is NOT a matter of empirical validity, and flies in the face of what we'd expect in terms of polygraph outcomes. I do understand how and why it gets discussed as a validity concern.
My proposed possible solution is to honor the TOR as an acceptable, though not the only, solution to the systemic problem. Permitting this solution is wise. Requiring it is not. It is solved in a different way in Colorado and the Northwest states. See sections 8.3.2.2 and 8.3.3.3.
--------------------------------------------
We are approximately six months past the initial completion deadline of the APA 2008 AGM conference, and nearing 3 months past the second completion deadline of the January 2009 BOD meeting. I think we would all agree that the wisest and best thing to do would be to put egos and personalities aside and seek the best common solution.
------------------
"Gentlemen, you can't fight in here. This is the war room."
--(Stanley Kubrick/Peter Sellers - Dr. Strangelove, 1964)