Instant Nationwide Criminal Database Searches
January 4, 2008
Instant nationwide criminal database searches are extremely tempting. From the day we are born we appreciate “instant” gratification. “Statewide” and “nationwide” are powerful words and imply something extremely comprehensive. “Low-cost” and “inexpensive” are also words that trigger a Pavlovian response in most of us. Put all of these words in the same sentence and you’ve overcome the arguments of all but the most determined skeptics!
Few hard statistics are provided by those proffering these tools and it is virtually impossible to know exactly what is missed. Not many of us can realistically put perspective around numbers like “millions” and “billions” (frequently used to describe the number of records housed in a typical commercial database) except to say “gosh, that’s a lot.” A million is certainly a lot of SOME things, but if you have 500 million records in your database and there are 5 billion records in existence across the country, would you still believe that to be an accurate and reliable criminal background check?
How big is the universe?
How many criminal records actually exist in the United States? Few of us can declare the exact number of counties that comprise the state we live in or certainly across the entire United States. There is no feasible way to measure or reliably estimate the total number of criminal records in existence, due to terminology and archiving disparities, widely varying records retention policies, and the simple geography of thousands of court jurisdictions (there are over 10,000 court jurisdictions spread among 3,127 counties nationwide). In attempting to gauge the reliability of these databases, it is important to understand what they consist of.
One way to illustrate the risks is to look at the “whole” as the sum of its parts. The “millions” and “billions” of records claimed by database vendors do not adequately describe the sources or usefulness of those records. Take the resource list from one of the most prominent nationwide database vendors as an example. Using a font size of 8.5 (a bit small for comfortable reading), this resource list fills 26 standard-size pages. A pessimist might consider this seemingly endless list of minutiae an attempt to overwhelm the curious and encourage them to simply rely on the fact that “millions” of records must mean it is comprehensive and accurate.
Dissecting this resource list provides a clearer picture of what this database contains. The “millions” of records all fall into one of the following four categories:
- County Felony / Misdemeanor Criminal Court Records – County-based felony/misdemeanor records are the primary information needed for employment screening. These records require the most effort to obtain and are the least likely to make it into the nationwide databases. County court records are obtained by database vendors either from a statewide court repository, where available, or purchased from individual counties. Individual counties have no mandate whatsoever to provide data to commercial database vendors. Similarly, those that choose to provide data have no responsibility to provide regular or consistent updates. In this particular nationwide criminal database, there are 36 states for which the database either contains no county court records at all, has extremely significant coverage gaps, or the state has suspended updates altogether. Based strictly on this database’s own disclaimers, less than 42% of counties nationwide are covered by this database. Even 42% is an especially generous estimate because for the states without a disclaimer, it assumes full participation by every county in the state, which is widely known to not occur.
- State Department of Corrections Records (DOC) – DOC databases contain records ONLY on persons who have served time in state correctional facilities. Most would consider this important information, but it contains significant gaps that can only be avoided with county court research. Any conviction that did not result in prison time, resulted in a suspended sentence or credit for time served, or resulted in serving time in a municipal, county, or federal correctional facility will be missed. There are 14 states, including California, that provide no DOC information to this database. The probability of identifying an existing criminal record is far greater through direct county research.
- State Sex Offender Registries (SOR) – each state maintains a Sex Offender Registry, virtually all of which are accessible via the internet. This too represents important information, and several industries, including those involving child- and elderly-care, mandate it. While the nationwide database search can be considered an effective tool for checking SORs, the registries themselves are a database tool and subject to the same limitations. Ernie Allen, President and CEO of the National Center for Missing & Exploited Children, recently stated “Today, there are more than a half million sex offenders that are “supposed” to be registered in the United States; however at least 100,000 of these offenders are actually “missing” from the system.”1 Sex offenders are required to register when they move to a new location, but if they fail to do so, county criminal history research based on their address history is the only reliable way to identify this criminal past.
- Proprietary Criminal Data – “proprietary” is another word commonly used to create the illusion of quality. This simply means that if, by chance, a particular individual has previously been screened by this specific database vendor and a record was found, that record will be included in the report. This process ignores the FCRA requirement that a reasonable effort be made to ensure any records reported are accurate, complete, and current.
Faithful readers of The Verifier™ may recall the test2 EmployeeScreenIQ recently conducted on one of the most popular nationwide criminal databases. In that test we submitted 651 randomly selected criminal “hits,” found in county court searches by our professional researchers, to one of the most prominent instant nationwide criminal databases. 451 of these known records were disqualified from the test by the database vendor as being from county jurisdictions that were not covered by this database. Speaking ONLY of the 200 records that were accepted for the test, more than 35% were missed completely by this database, and of those that were found, 50% contained incomplete or inconsistent information compared with what our researchers found at the primary source, the county courthouse. Extrapolating these numbers back to the original sample of 651 records, this “nationwide” criminal record database identified less than 20% of the known criminal records, with half of those containing discrepancies.
Nationwide databases do serve a purpose
Industry best practices dictate conducting physical research in each county in which your candidate has lived, worked, or gone to school over the past 7-10 years. There is always a small chance that someone may have committed a crime far from the specific counties in which they have lived, worked, or studied. A nationwide criminal database can be used to attempt to identify “vacation” criminal records, understanding that if such a record does exist there is no guarantee it will be found in a nationwide criminal database. Once identified through a database, the record will also need to be validated by the local court jurisdiction to satisfy FCRA requirements.
While some databases are better than others, the key takeaway is that the very best criminal database search falls FAR short of the accuracy and reliability of conducting primary research in county courthouses. Nationwide criminal databases can be considered a useful supplement to criminal research in county courts by professional researchers, but as a standalone product cannot be relied upon to identify threats to the safety and security of your customers, employees, and corporate assets.
Please note that the database vendors used for these analyses remain anonymous, as the intent is not to detract from a particular vendor’s product, but rather to educate on the information gaps rampant within all of these tools. The nationwide databases used in these analyses are considered to be (and sell themselves as) among the most comprehensive with the most sources of database records in the industry.
1Allen, Ernie, “NATIONAL CENTER FOR MISSING & EXPLOITED CHILDREN JOINS WITH CONGRESSMAN FOLEY, SENATOR HATCH AND VICTIM FAMILIES TO ANNOUNCE TOUGHER SEX OFFENDER LEGISLATION,” press release, Washington, D.C., May 18, 2005
2Bachman, Kevin, “BIS Research Exposes Holes in Instant Criminal Records Databases,” The Verifier IV, 3rd Quarter 2004
Latest posts by admin (see all)
- SterlingBackcheck Acquires EmployeeScreenIQ, a Global Provider of Background Screening Services - November 3, 2015
- Many Employers Not in Compliance With EEOC Criminal Records Guidance, EmployeeScreenIQ Research Shows - October 21, 2015
- The EEOC’s Criminal Background Screening Guidance 3 Years Later - October 8, 2015