posted 01-24-2008 09:02 AM            

Part I

I
SR
C
R
C
I
R
C
R
C
R
C

Part II
I
SR
C
R
C
I
R
C
R
C
R
C

We were using the following R/I with at least two iterations:

I
SR
C
R
R
I
R
R
I
R
R
I
R
R
C
R

The benefit of the R/I is that it takes less time to administer, and covers more relevant areas. The downside is that there were too many relevant areas for the examinee to focus on and react to, potentially diminishing reactions to the point of not reacting significantly enough on areas where information was being withheld. Additionally, the question spacing was only 15 seconds from answer to the beginning of the next question, which at times would not give the physiological reactions enough time to fully come back to homeostasis. Because of the discomfort of the blood pressure cuff, and the tremendous amount of questions on one iteration using this R/I format it was necessary to shorten this spacing to 15 seconds.

Ben

[This message has been edited by blalock (edited 01-24-2008).]

posted 01-24-2008 09:17 AM            

Appropriate Question spacing (15 sec, 20 sec, 25 sec)

Formats (R/I, LEPET, Others)

Relevant Questions Used-
If we are limited to the number of relevant questions/charts run, which areas are the most important relevant questions to ask?

Comparison Questions Used

Maximum Number Of Relevant Questions in One Iteration (3, 4, 5, 9)

Is the maximum number of questions for an Iteration affected by blood pressure cuff discomfort, diminishment in psychological salience, or diminishing physiological responses?

posted 01-24-2008 09:26 AM            

I personally use the LEAT. I, SC, C, R, R, C, R, R, C and if you need another R you can add it to the end. CQ's I like to use are: lying to supervisors, lying to cover up something, doing something dishonest or unethical - on current LE officers: considering using excessive force, thoughts of not turning in evidence, falsifying anything in a report.

posted 01-24-2008 10:49 AM            

Why the Neutral in position 6? You can call me stubborn or oppositional, but I don't like the idea of doing something "just because," and I'm not a fan of overly-psychologized assumptions that are not grounded in data or well-established constructs.

Even five RQs seems excessive to me. There is no way of escaping the combinatorics of multi-facet and mixed issues exams, unless you want retreat to fiction. The combined accuracy rate will be estimated accuracy rate per question raised to the power of the number of questions. If you're talking about a test with an error rate of 10 percent, that gives you about .59, which is still better than chance, but...

I just don't think its worth it.

Same with INC rates. The combined INC rate will be the inverse of the resolution rate (which is the inverse of the estimated INC rate per question), raised to the N of RQs. So, if we think of a test with a 10% INC rate , we're looking at an estimated INC rate of .41. If we're optimistic and use 5%, then we still get INC=.23. That's a lot. So what will happen is that field examiners are at risk to cave-in to administrative judgement and modify their practices (read: fudge the number or change the rules arbitrarily = not science).

Ideally, I'd go for three or four questions, targeted around behavioral issues of actuarially supported predictive validity regarding some future risk, such as breeches of professional integrity, training failure, or poor job performance (things that matter).

I have no real problem with the RI test, as long as we disseminate the correct understanding that it is presently scored impressionistically and not mathematically. So, I have no problem with the 13 question test, and with waiting 25 or 30 seconds. That's only 7.5 minutes. Nobody's arm has ever fallen off. At 65mm of pressure, that will never mechanically cut off or completely overpower the circulation of a person who BP is 120/80. Plus we're talking about people who want to be LE officers. I'd guess there are about 1000 or more inconvenient or uncomfortable things about police work. So, if someone can't take 7.5 min of BP cuff at 65mm, see-ya.

Now this. Falsifying a questionnaire seems at first glance to be exceeding broad and lacking in behavioral description and specificity. I dunno though, maybe those things are specific and serious enough for it to work. However, we can explore the principles by putting the idea into another context. If I submitted a report on a PCSOT maintenance exam with a question about lying on a weekly probation questionnaire (written report from the offender), it would never fly. The questionnaires are so extensive that failing provides little guidance about what behaviors to inquire further of (is that grammatically correct?). Passing is also minimally informative, because it leads us to the possibly naive assumption that the questionnaire is 100% honest-to-goodness complete and every single thing there is to know, and that the subject fudged on absolutely nothing. There is always something. We'll never know everything. What matters is whether we know enough of the things that are actually meaningful and actually add incremental validity to our decision processes. Think about this, what will produce the greater reaction potential - lying to a piece of paper or lying about a behavior. From my recollection, it is a fundamental principle of question formulation that the question describe a behavior. Lying on paper is simply indirect and weak, and depends far too heavily on some form of verbal Venn-diagram logic to hold any assumption of validity. Human verbiage is inherently far fuzzier than Venn logic. If the principles or constructs work, they should work in another context. This one doesn't.

.02

(now back to obsessing about Zelicoff)

r

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

posted 01-24-2008 11:08 AM            

Also, in HR, it is legally required that every facet of applicant screening tests/ prerequisites must be 100% relevant to the tasks required to perform a position. I recall a factory that required each emloyee be able to lift 65lbs above ther head. The job at no time required such lifting and so it was a wink and a smile prerequisite to weed out female applicants-----many of which take greater sick days as many have children that they care for. EEOC doesn'tlike those kinds of rules, and beleive me, when the EEOC gets involved, HR reps get fired without cause.

When I view applicant screening tests as a whole, I see a future weakness as many of the questions are too loosely interpreted, and even more so, passing a polygraph test itself is not required ON the job. In time, such cases will be made even in law enforcement regarding polygraph testing.

posted 01-24-2008 11:28 AM            

As far as the discomfort of the BP cuff, I agree that if they can't handle that discomfort, then see ya. Most of my applicant's who do complain about it (which is not very many actually) seem to be those who are withholding information anyway. I wonder if this is a conscious or subconscious appeal for sympathy... Anyway, that is another topic. I used the "Have you falsified the application (or the application process) in any way?" question, quite a bit, and decided that it is just to broad of a question, and it seems that it may cause more false positives than other questions, since it is such a searching question. Comparison questions that are vague, cause similar reactions...

In coming up with the relevant questions to ask, I listed out all of the areas of concern, including possible RQ questions to ask, then prioritized them placing the most important questions first. Then, I took the top eight questions, and divided them into two iterations, in order to keep the maximum amount of RQ's to four per iteration.

Here are the RQs settled on:

**********

• Have you intentionally withheld any job from this application?

• Have you ever been fired from a job?

• Have you ever taken money or merchandise of value from places you have worked? (could be valued more than $25)

• Have you ever committed any serious crime?

• Have you ever taken part in any of the sexual behaviors reviewed today? (prostitution, child porn, sexual contact with minors, rape)

• Other than ____ , have you used any other illegal drug?

• Have you used any illegal drugs since _____ ?

• Have you ever sold or transported any illegal drug for a profit?
  

Other possible questions include:

• Are you withholding information about your job history? (more vague, includes being fired, disciplined, withholding jobs, etc.)

• Are you intentionally withholding any information about your involvement with illegal drugs? (once again vague, includes selling, using, distributing, manufacturing, buying, possessing, etc.)
  

posted 01-24-2008 12:51 PM         

Why does your drug question end with "for a profit"? It seems to me that somone would have lots of room to rationalize on that one.

posted 01-24-2008 12:57 PM            

This is like handing the homeless guy a dry piece of cardboard, a cheesburger, and 40.

Life is good.

Thanks Blalock.

quote:

---

• Have you intentionally withheld any job from this application?

---

OK. Are these in order of preference or importance? How common is this, and is it not possible to investigate this concern more accurately through the BI.

quote:

---

• Have you ever been fired from a job?

---

How would you word this if they had been fired?

quote:

---

• Have you ever taken money or merchandise of value from places you have worked? (could be valued more than $25)

---

I like this one. Too may “haves” in the words, sounds odd to me. Why not simply “places that you worked”

Also, what about other types of thefts? Do they inform us at all about future risk at work?

quote:

---

• Have you ever committed any serious crime?

---

I'd have to assume that “serious crime” is defined carefully in the pretest. Its probably better than the legal jargon “felony,” but I don't know.

quote:

---

• Have you ever taken part in any of the sexual behaviors reviewed today? (prostitution, child porn, sexual contact with minors, rape)

---

This is a bit broad for me, and I think it might be clearer to simply state “any unlawful sexual behaviors” or “unlawful sexual activities,” have reviewed the list during the pretest.

Also, I think I prefer “Did you ever” vs. “Have you ever.” I'm no linguist, but “did you” is a direct, active verb and might be more stimulating than the more passive “have you.” “have” and “had are verbs indicating possession, while “did” and “do” indicate action. I dunno. I'm sure there are other arguments. I generally use “did you” in RQs and “have you” in CQs, but that's probably more a comfortable habit that a necessity. Sometimes in PCSOT “did you” doesn't work (i.e, “have you been completely alone with a child”), because the verb is a form of “be” not “do”

quote:

---

• Other than ____ , have you used any other illegal drug?

---

This one bothers me for reasons both empirical and practical. To begin with, this question assumes that we are “testing the limits” of an admitted behavior. That is a very different measurement problem than testing “any involvement” in a behavior which is completely denied. If we want the polygraph to be regarded as a form of science, then we have to pay more careful attention to the the psychological and physiological constructs that support why test question would work or don't work (not just psychological set).

Nearly every description of construct or description of convenience is built around the idea that involvement in a behavior or event is a fundamental component of a subject's reaction potential to the target stimulus question. Reaction to the stimulus is interpreted as indicative of involvement, while lack of reaction is interpreted as indicative of non-involvement in the behavior or event. Our job as examiners is to conduct an examination such that nothing other than involvement in the behavior or event would cause the examinee to react, and such that nothing other than non-involvement would become the cause of non-reaction.

If a person reacts to “other than...” and we then ask ourselves what that means... one conclusion might be “he has some involvement other than...,” but its always prudent to also consider what other obvious factors might cause a reaction. Is is likely or possible likely that his admitted involvement could cause a reaction, even if reported accurately? To put it differently, if we expect that behavioral involvement can reliable produce a reaction potential to a test stimulus question, do we also believe that say the words “other than...” can reliably negate the subjects expected reaction potential?

I'm bothered less when the numbers are small (really small), but then I've favor using the number “besides those five times...” When when the number of admitted incidents gets beyond a very small number I don't believe they know the answer. So the results become difficult to interpret beyond a small number of admissions. Think about this: if the subject passes “other than...” after admitting 12 times or 20 times or 200 times, do we really believe the results mean the number isn't actually 13 or 15, or 25 or 205 or 225. And what do we really know about the contribution of these issues to job related risk. Now if someone reacts to the question after admitting 20 times or 12 times or 200 times, do we really think we are any better informed than if we had simply taken the admissions and scratched the question from the test?

I think the recency of drug use might be a more informative concern to questions about potential LE officer training, integrity, and job performance. I don't really know but, I'd guess that some contact with illegal drug is more common that we want to know about among our college students. What is probably more informative is whether we think someone has a recent or historic pattern of excessive use or misuse, including frequent recreational use, medicinal use, dependency, addiction, cross-addiction, or poly-substance abuse. If I were thinking in a forensic risk management capacity, I be concerned about recent (past several years) recreational use, self-medicinal use (or chronic patterns), or perhaps guidelines around dependency, addiction, cross-addiction (changing from one to another substance) or poly-substance abuse. - not whether it was 20, 25 or 200 times during college.

quote:

---

• Have you used any illegal drugs since _____ ?

---

I like this one.

quote:

---

• Have you ever sold or transported any illegal drug for a profit?

---

Seems good to me.

quote:

---

**********

Other possible questions include:

---

quote:

---

• Are you withholding information about your job history? (more vague, includes being fired, disciplined, withholding jobs, etc.)

---

Holy smokes that's a broad question!

OK, so if we're using an RI test, and know that we're not numerically scoring – that the test is really a form of “polygraph interview” - and the goal is to stress, challenge, explore everything, then this could be a lot of fun. You'd could just attach components and then rock-n-roll, 'splainin.' as we go 'splorin.' We'd have to ignore the pneumos for garbage, and cardio would probably be uninterpretable. We'd need some potasium-choloride-super-glue to get EDA sensors attached for maximum stability and minimal movement noise. If you had someone by the short-hairs you could probably shake 'em down for all kinds of embarassing things. You could even maybe make a network or syndicated game show...

quote:

---

• Are you intentionally withholding any information about your involvement with illegal drugs? (once again vague, includes selling, using, distributing, manufacturing, buying, possessing, etc.)

---

Broad again. And again, a truly valid principle or construct with be valid in a number of testing contexts. That word “intentionally” would not be accepted in a PCSOT question in Colorado. Sex offenders would love to mitigate their culpability with some version of “I didn't mean it that way...” Also, wadaya think we'd get from DA's office and LE examiners in any random jurisdiction from anywhere in the country if a bunch of private practice examiners started submitting polygraph reports with questions about whether an accused suspect “intentionally” withheld information from police, or intentionally beat their wives, or intentionally shot someone.

I suggest we stick to the basic principles as closely as possible, and that means developing and emphasizing all versions of “did you do it.”

.02

r

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

posted 01-24-2008 01:05 PM            

The reason I put it in there is to distinguish from just sharing or giving someone illegal drugs. Additionally, if someone is driving around and the applicant or another applicant has it in their pocket for use, it is not considered transportion for profit. In the pretest, it is explained that receiving any money or anything else in exchange for drugs is profitting.

With a prior experience law enforcement officer, it distinguishes from transporting evidence, or participating in undercover operations.

Would you suggest "Have you ever sold or transported any illegal drug?"

Thanks for your participation in this discussion!

posted 01-24-2008 01:11 PM            

Great insights so far! Fixin' to do a test. I'll check in later.

posted 01-24-2008 01:12 PM            

IP: Logged

posted 01-24-2008 01:25 PM            

quote:

---

quote:
--------------------------------------------------------------------------------

Are you intentionally withholding any information about your involvement with illegal drugs? (once again vague, includes selling, using, distributing, manufacturing, buying, possessing, etc.)

--------------------------------------------------------------------------------

Broad again. And again, a truly valid principle or construct with be valid in a number of testing contexts. That word “intentionally” would not be accepted in a PCSOT question in Colorado. Sex offenders would love to mitigate their culpability with some version of “I didn't mean it that way...” Also, wadaya think we'd get from DA's office and LE examiners in any random jurisdiction from anywhere in the country if a bunch of private practice examiners started submitting polygraph reports with questions about whether an accused suspect “intentionally” withheld information from police, or intentionally beat their wives, or intentionally shot someone.

I suggest we stick to the basic principles as closely as possible, and that means developing and emphasizing all versions of “did you do it.”

---

And there's the problem with screening exams.

I use "intentionally" because it results in fewer reactions. I pre-test it by saying is means that right now the person thinks or knows he should be disclosing information, but he is failing to do so. If not, you get the examinee thinking he's not sure if there's anything else out there - which means you've set up a great comparison question.

In other words "intentionally" separates the "no's" from the "I don't knows."

Don't forget that the pre-test explanation of the questions in a screening exam is probably more time consuming and difficult than in a single issue CQT. In those we want vague CQs and we want RQs that are either yes or no. We have to work much harder with these vague issues, but anecdotally, most of us have found that we can get to the point where an examinee understands the point of the RQs and can feel confident of his answers.

I use a drug question similar to the one above, and I have the examinee tell me what it means. By the time we're done, the person tells me exactly what it means (usually somewhat annoyed that we've so dissected the question (actually all questions), and they almost universally finish with, "and I've told you everything about... there's nothing on my mind causing me any concern about that question," making the point that they want to get on with the test and end the insanity.

posted 01-24-2008 01:31 PM            

I would hazard a guess---and based on my own experience witnessing various character flaws in some lawman----is that the paramount questions would address moral turpitude and rape/stalking types of behaviors. 2 very pitiful cents.

I also must say that people often are not fired but are given the opportunity to "leave"----in order to spare the expense of unemployment benefits. Not having great experience in app screening, I wonder how examinee's approach the subject when asked about being "fired."

Somehow, I feel like I am going to get clobbered for this post.

posted 01-24-2008 01:40 PM            

quote:

---

I still believe with all of my heart and soul that if an affordable and boob proof physiological drug test could replace polygraph drug questions, than our false positive rates for app screening would be greatly diminished.

---

I've had only one unresolved drug question candidate - and I'm confident that I know what happened there. Most all of the rest resulted in admissions, and the remaining few were "remedied" in a follow-up more focused exam.

To call a positive response on a screening exam a false positive is not correct - unless you are making DI calls at that point. With a successive hurdles approach, positive results on a screening exam are only tentative.

posted 01-24-2008 03:06 PM            

I think the word "intentionally" was forbidden by the old school to be safe ---as it is a word tht can so easily be abused and misused.

As in life, accidents happen;

posted 01-24-2008 03:19 PM            

new;
"SYSP, have you consumed any forbidden substances?"

old;
Since we first met, do you think I am more dashing than BradPitt or George Clooney?

new
"Since we first met, do you think I am more dashing than any man on the planet?"

posted 01-24-2008 04:23 PM            

Do you have more information, beyond the National Polygraph Consultants' Quick Reference Guide on the LEAT?

*****

Barry,

The reason for splitting up the drug questions into three is because I want to know specifically what area to question about in more depth on the prohibited drug areas. Occasionally, I will find that the reaction on the test has to do with a different drug question on the test, but this seems to be the exception rather than the rule. In other words, if someone responds to the sale of drugs, it is the sale of drugs that they were lying about, not the use of drugs...

I suspect, however, that perhaps that 3 questions on drugs in a screening exam may cause an unbalanced reaction, and that perhaps I should save the specificity for a breakout exam...

*****

Ray,

I agree about your comments regarding lying to a piece of paper versus lying about a behavior. That is why I dropped asking, "have you lied in the application process" or "...on the application."

I also appreciate your insights on the "Did you" versus "Have you." I try and keep the beginning of the questions similar to each other, either Have you's or Did you's. Your brief comment about using "Have you" on comparison questions, and Did you" on relevant questions is very interesting to me, and I think worthy of more discussion.

Also, your comments on "Other than..." also piqued my interest. I know that some like to use "Are you now withholding information about..." instead of "Other than..." I'd like to hear from other examiners 1) what method they choose, and 2) what their experiences are with "Other than..."

I chose to eliminate a drug question about the number of times XYZ was used, because it seemed too fuzzy and difficult for applicants to commit to a true number, when the number was beyond 1-2 times. The recency of drug use is split up into three main categories at our agency. One is that use of marijuana within the past two years is not allowed. Two, any use of specific types (can't recall at the moment) of drugs in a lifetime, is not allowed. Three, any use of the rest of illegal drugs in the past five years is not allowed. That is why I use the Other than _____ (say marijuana), have you used any illegal drug since _____ (2003)?

*****

Barry,

Your comments about the use of the word intentionally in some contexts goes right along with why I have used it. Many times, I get applicants who say "I don't know" versus, "yes I did, or no I didn't."

I have not been comfortable with doing breakout exams immediately after quizzing my applicants about why they reacted consistently and significantly to relevant areas. I am almost never able to clear them with a breakout exam immediately following, and I am fairly certain it is because I am like a bulldog in questioning them until I get to the root of the SR. I prefer to set up a specific issue exam on another day. Of course, that presents a dilema when the applicant is from out of state or town.

Ben

posted 01-25-2008 09:08 AM            

PART ONE
1 Prior to 2007, did you graduate from high school?
SR2 Regarding your past, do you intend to answer each question truthfully?
C3 Have you ever taken credit for another person's work?
R4 Did you withhold any job from this application?
C5 Are you the kind of person who would betray a friend?
R6 Were you ever fired from a job?
C7 Are you absolutely trustworthy?
R8 Did you ever take money or merchandise of value from places you worked?
C9 Are you a really honest person?
1a Are you sitting in a chair?

PART TWO
21 Prior to 2007, did you live in _____ ?
SR22 Regarding your past, do you intend to answer each question truthfully?
C23 Prior to 2007, did you ever lie to make yourself look better or important?
R24 Did you ever take part in any unlawful sexual behaviors?
C25 Are you the kind of person who would betray a member of your family?
R26 Are you withholding information regarding your involvement with illegal drugs?
C27 Did you ever do anything that could bring shame upon yourself or your family?
R28 Did you ever commit any serious crime?
C29 Did you ever cheat in school, sports, or games?
21a Are the lights on in this room?

Begin with an acquaintance exam. Minimum of 20 seconds between answer and question onset. Run each part a minimum of 3 times. Run a 4th if artifacts render a spot unscorable.

SR = Sacrifice Relevant
##a = extra irrelevant question to ask if needed to establish homeostasis or to identify possible countermeasures

Ben

posted 01-25-2008 09:23 AM            

Serious crime: by using serious I think individuals have wiggle room. I would stick with crime(s) and if they make those minor admissions add 'that you have not already reported' Also, taking part is an unlawful sexual behavior would be a crime. I would think in the pretest to work unlawful sexual acts into 'crimes' resulting in one less question.

CQ's about lying to family members just appears like a CQ to a well read examinee. I guess it is how you pretest the question. I like lying to supervisors with the pretest hype on 'past behaviors predict future behaviors'. Also cheating in sports is kind of out in left field (to me) and the relevance on a pre-emp is questionable to me. Cheating on taxes, cheating to get a BA or in POST, lying to cover up mistakes at past jobs... that is the direction I would go.