The FBI obfuscates and evades in a written response to questions posed by Senator Charles Grassley (R-IA) regarding the Bureau’s polygraph screening program. Senator Grassley’s questions, and the FBI’s responses, appear at pp. 47-50 of a document submitted to the Senate Judiciary Committee on 30 November 2006 and made available by the Federation of American Scientists here (7 mb PDF):
Senator Grassley’s questions (in bold text), the FBI’s responses, and AntiPolygraph.org’s observations appear below:
48. According to a May 1, 2006, Washington Post article:
[“]Many researchers and defense attorneys say [polygraph] technology is prone to a high number of false results that have stalled or derailed hundreds of careers and have prevented many qualified applicants from joining the fight against terrorism. At the FBI, for example, about 25 percent of applicants fail a polygraph exam each year, according to the bureau’s security director.”
The article also cites “a comprehensive 2002 review by a federal panel of distinguished scientists” which found that “if polygraphs were administered to a group of 10,000 people that included 10 spies, nearly 1,600 innocent people would fail the test[.]”
a. Has the FBI conducted, commissioned, or reviewed scientific studies of the accuracy and effectiveness of polygraph examinations? If so, please describe them in detail. If not, why not?
For clarification, the FBI’s Assistant Director for Security’s comments to the reporter indicated that about 25% of applicants are disqualified as a result of the polygraph test. These results usually include admissions of information or activities that lead to a disqualification decision.
The FBI does not independently conduct or specifically commission polygraph research but it works with other federal agencies to improve polygraph techniques and has participated in research studies with the DoD Polygraph Insitute (DoDPI) which is charged with conducting research for the federal polygraph community. All DoDPI research is available directly from DoDPI.
Although 25% of FBI applicants may be disqualified as a result of the polygraph, this does not necessarily mean that only 25% of those polygraphed fail, as many applicants are disqualified before the polygraph stage. Past statements by senior FBI officials indicate that the FBI pre-employment polygraph failure rate is on the order of 50%.And while the FBI would have the Senate Judiciary Committee believe that most FBI applicants who fail the polygraph also make disqualifying admissions, the Bureau has never documented this assertion. It should be noted that in the past (and likely still) FBI polygraphers’ performance was evaluated based upon the rate of confessions obtained from those who “fail.” This gives polygraphers a strong incentive to inflate or embellish any admissions made to boost their performance ratings. Worse still, the FBI’s deliberate practice of not recording polygraph examinations puts FBI polygraphers at liberty even to fabricate admissions with impunity.
Senator Grassley asked not only whether the FBI had conducted or commissioned any scientific studies of the accuracy and effectiveness of polygraph examinations, but also whether it had reviewed any such studies. Conspicuously absent from the FBI’s response is any mention of any review of the findings of the National Academy of Sciences (NAS), which in its report, The Polygraph and Lie Detection, concluded that “[polygraph testing’s] accuracy in distinguishing actual or potential security violators from innocent test takers is insufficient to justify reliance on its use in employee security screening in federal agencies.” (original emphasis)
It should be noted that when the NAS stated, as mentioned in the Washington Post article, that if polygraphs were administered to a group of 10,000 people that included 10 spies, nearly 1,600 innocent people would fail the test, it was assuming, for the sake of argument, an overall accuracy rate of 90%. The NAS did not conclude that polygraph screening is anywhere near this accurate, but rather used the 90% figure in order to better illustrate the shortcomings of polygraph screening “even when an accuracy is assumed that is greater than can be expected of polygraph testing on the basis of available research.”
b. What is the FBI’s estimated rate of false results on polygraphs used for employment screening?
Because scientists are unable to conduct field studies under ideal (laboratory) conditions, and the absolute truth is not always available to validate the results of polygraph examinations in actual cases, known error rates remain elusive. Although error rates can be estimated, the estimates depend upon the testing situation, the issues being tested, and the persons being tested. Empirical studies cannot be used to generalize rates of error because different polygraph examiners and examination situations will produce different error rates. A major reason why scientific debate over polygraph validity yields conflicting conclusions is that the validity of such a complex procedure is very difficult to assess and may vary widely from one application to another. The accuracy obtained in one situation or research study may not generalize to different situations or to different types of persons being tested. Scientifically accepted research on polygraph testing is hard to design and conduct as evidenced by the depth of studies conducted by academic laboratories. The FBI would welcome and encourage broader research in this area.
We would offer a noteworthy data point concerning FBI internal testing of employees. Since the inception of the PSP Program in 2001, approximately 7500 counterintelligence-focused examinations have been conducted with a Deception Indicated rate of less than 1%. This result is significantly lower than the Washington Post‘s predicted 16% failure rate.
Note that the FBI studiously avoided answering Senator Grassley’s question: What is the FBI’s estimated rate of false results on polygraphs used for employment screening? Former Attorney General John Ashcroft himself acknowledged a 15% false positive rate in a 2001 press conference.
The <1% failure rate for the FBI’s internal polygraph screening program for employees can only be achieved by arbitrarily shifting downward the threshold for a reading of “deception indicated.” It would be bureaucratically untenable for the FBI to have a significant percentage of its employees failing the polygraph, and thus outcomes are manipulated to ensure that almost everyone passes. In July 2001, journalist Lenny Savino reported that some 25 of 500 FBI employees thus far polygraphed, that is, five percent, had failed to pass, a rate that a senior FBI official at the time described as “surprisingly low.” Apparently, it wasn’t surprisingly low enough, however, and “less than 1%” seems to be the prevailing officially acceptable polygraph failure rate for FBI employees. This has no bearing on the failure rate associated with applicant screening, where a fifty percent failure rate is evidently deemed acceptable.
c. Given the high rate of false results, should a “failed” polygraph alone be the basis for a negative employment decision or personnel action? How many times per year is a polygraph result the primary reason for a negative employment decision or personnel action?
We do not believe that the FBI is experiencing a high rate of false positive results. Throughout the Federal polygraph community, the polygraph is considered to be an effective and acceptable screening tool and is a strong contributor in conjunction with the entire applicant process which examines the prospective employee from several standpoints. These include field investigations, records checks and polygraph examinations. As noted earlier, polygraph results, including statements and admissions, account for about 25% of applicant disapprovals. With regard to on-board employees, a “failed” polygraph is never used as the sole basis for an adverse personnel decision. Anomalies are addressed through additional interviews and investigative work. The polygraph program does not make determinations on negative employment issues or personnel actions.
When Senator Grassley earlier directly asked the FBI for its estimated false positive rate, it dodged the question and declined to provide any estimate. But in the next breath, the FBI claims that it does “not believe that the FBI is experiencing a high rate of false positive results.” What rate then does the FBI believe it is experiencing? On what basis?
Why should a failed polygraph alone reason be enough not to hire an applicant, but not reason enough to take personnel action against an employee? For applicants who wrongly fail the polygraph, there are no field examinations or record checks. They are summarily barred from FBI employment. For life.
Note that the FBI failed to answer Senator Grassley’s question: “How many times per year is a polygraph result the primary reason for a negative employment decision or personnel action?” How many times a year is a polygraph result the primary reason an applicant is not hired? How many times a year is a polygraph result the primary reason for adverse personnel action against a current employee? The FBI’s not saying.
d. What steps has the FBI taken to identify more reliable alternatives to polygraph tests for ensuring the trustworthiness of current and prospective employees?
The FBI supports DoDPI research through a cooperative agreement and currently has two SAs assigned to DoDPI. Later this year, DoDPI will host a summit sponsored by the interagency Technical Support Working Group and DoD’s Counterintelligence Field Activity. The purpose of assembling these experts is to develop a research plan for the next 5-10 years for means to assist in determining truth of statement [sic].
Given the National Academy of Sciences’ finding that polygraph screening is completely invalid, we need not wait until a working lie detector is invented before scrapping one (the polygraph) that does not work.
49. In response to a previous question for the record regarding the New York Police Department (NYPD), you indicated that during a meeting to explore cooperation with the NYPD’s translation and analysis program, the NYPD indicated that it did not want its officers and translation staff to undergo FBI polygraph testing as a condition of being granted access to “FBI information.” The response further stated, “we understand that the CIA and Pentagon have found a means of ensuring trustworthiness without the use of polygraph examinations.”
a. Please describe the alternative method of ensuring trustworthiness to which that response refers.
b. The previous response also stated, “We will work with both organizations to learn more about this process and will evaluate our ability to do the same.” Please explain what progress has been made toward implementing this polygraph alternative.
Response to subparts a and b:
We have established a program where NYPD translators work on unclassified IC materials through the National Virtual Translation Center (NVTC). The FBI is also providing the NYPD with romanization training, teaching the IC’s standard for transliterating foreign scripts into the Roman alphabet. Although we contacted our sister agencies to discuss their internal policies in this regard, we were pleased to find the NVTC to be a suitable vehicle through which we could fully us the NYPD’s available translator resources.
Note that the FBI fails to provide the requested description of the “means of ensuring trustworthiness without the use of polygraph examinations” of which FBI Director Mueller had spoken. What are these means? Why won’t the FBI say? If means of ensuring trustworthiness without the polygraph exist, why use the polygraph at all?