Page 8 - Security Today, November/December 2020

IDENTIFICATION
The Use of Technology Proper software, solutions have help identify potentially bad hires
BCy Todd Mickelsen
ustomers and the public trust security firms to safe- guard people, places and information. This trust is quickly destroyed when security staff commit a crime, which is often due to inadequately prescreen- ing job candidates during the interview process.
“Since guards are trusted with customer access codes and/or keys to their facility, it’s imperative to have an efficient and thor- ough screening process,” said Brett Magleby of Salt Lake City- based Panther Security & Investigations. “It is a huge liability for a security company to have someone knowingly dishonest in a customer building or secure site. They can quickly lose a security firm’s contracts by doing something dishonest.”
BAD PRESS FOR SECURITY FIRMS
This liability has reared its ugly head and repeatedly document- ed over the year in news stories. For example, last April an off-duty security guard was arrested and charged with assault. In March, a story broke that more than 600 weapons from the world’s largest security company were reported lost or stolen since 2009. And in January, a security guard was convicted of falsification of evidence as a police officer, disorderly conduct and theft.
In one of the more extreme examples of a bad hire, in 2017 a New Jersey security guard was accused of stealing $100,000 on his first (and last) day at work from his new employer’s armored truck.
SCREENING OPTIONS FOR NEW HIRES
According to recently published research, the U.S. security services market is estimated to reach $39.1 billion this year. With this market growing an average of 2.8% annually between 2015 and 2020, it’s more important than ever for security firms to have a system in place for hiring honest, trustworthy staff.
Unfortunately, there may be disqualifying activities in a job candidate’s history that an interview, personality test, integrity test, background investigation, fingerprint check or drug test may not uncover. Many criminals are never caught. And screen- ing tests or background checking systems are poor at detecting deception or simply aren’t designed to do so.
For example, a personal interview is only 54% accurate at de- tecting deception1, and an integrity test is only 53-64% accurate2. Personality or integrity tests aren’t capable of measuring or de- tecting deception.
POLYGRAPH
The Employee Polygraph Protection Act of 1988 (EPPA) prohib- its any U.S. company in the private sector from using a lie detector on employees, except for three exemptions. One of those exemptions includes security companies. Per the statute, if the technology meets the functional definition of a polygraph, then it can be used to pre- screen “prospective employees of armored car, security alarm, and security guard firms who protect facilities, materials or operations....”
However, the lie detection technology used must meet the gov- ernment’s functional definition of a polygraph. It is required to capture the following physiological data: cardiovascular, respiration
and electrodermal activity. In simple terms, these are breathing ac- tivity, heart rate, blood pressure and sweat/skin conductance. Mea- surements are recorded and analyzed to detect these physiological changes and correlate them to statements made about questions.
During a polygraph test, the testing equipment (2 corrugated tubes, blood pressure cuff, 2 finger sensors, and motion sensor mat) is placed on the chest, abdomen, upper arm and fingers of the ex- aminee. The examiner sits in close proximately to the examinee. One examiner can conduct an average of 3 tests per day, and each test ranges from 1.5 to 5 hours. Preparation of the test results can take hours. Polygraph screening exam costs can cost up to $250.
The first modern polygraph was invented in 1921. The FBI began using it in 1939. A polygraph’s average accuracy in detect- ing deception is 87%3.
EYEDETECT
A new lie detection method, EyeDetect by Converus, was in- troduced in 2014. It also referred to as an ocular-motor test, and is an automated computer-based test that uses a high-definition eye tracker to measure involuntary changes in the eyes and read- ing behaviors while a person answers true/false questions during a 30-minute screening test or 15-minute test on specific alleged crimes. Involuntary eye behaviors include pupil dilation, blink rate and other eye movements. These measurements have been shown to be linked with increases in cognitive load, which have been associated with deception.
Because the test is automated and the results are computed by al- gorithms in the cloud, no examiner is used, eliminating the possibil- ity of human bias. The automation, along with the ability to moni- tor tests remotely, allows for safe social distancing during testing.
Results are available in less than 5 minutes. Research studies show the solution is 88% accurate4.
The standard test doesn’t meet the functional definition of a polygraph and therefore cannot be used by U.S. security compa- nies. But that changed last June when Converus released its new- est version, called “Physio Tracker,” which gathers similar physi- ological data as captured by a standard polygraph instrument: cardiovascular, respiration and electrodermal activity.
THE TRUTH ABOUT LIE DETECTORS
Keep in mind the perfect lie detector test doesn’t exist. All tests have a margin of error. They can give false-positive results and false- negative results. EyeDetect’s automated test minimizes false positives (when an innocent person is determined by the test to be guilty).
Todd Mickelsen is the president and CEO of Converus.
1. Bond & DePaulo, 2006
2. The Criterion-Related Validity of Integrity Tests,
Journal of Applied Psychology, 2012
3.American Polygraph Association Meta-analytic
survey, Table 2, 2011
4. A. Potts Dissertation, University of Utah, 2020
8
NOVEMBER/DECEMBER 2020 | SECURITY TODAY