The Name Game: Feds and States Makes Matching Names an Important Accuracy Issue

By W. Barry Nixon, COO, PreemploymentDirectory.com

For many years background checking professionals have been challenged by common names and doing their best to make sure they correctly identify the right person associated with a name. If an applicant’s name is James Smith or Jose Mendez, they predictably have a lot of other people with the same name. Also, name variations can be problematic as well. Consider a common name such as Cindy can also be spelled Cindy, Cyndi, Cindi, Cyndy, Syndy, Syndi, Sindi and Sindy. 

So what is a Private Investigator to do? How do they take effective steps to follow “reasonable procedures to assure maximum possible accuracy of the information concerning the individual about whom a report relates.” 

Let’s start with the legal requirements and state initiatives that are driving the importance of effective name matching tools and processes for background check firms beyond their own desire to provide accurate information. 

Federal Requirements

The Fair Credit Report Action (FCRA) which is the primary legislation that impacts the practice of conducting background checks “requires you [consumer reporting agencies (CRAs)] to establish and follow “reasonable procedures to assure maximum possible accuracy of the information concerning the individual about whom the report relates. See FCRA 15 U.S.C. § 1681e(b), Section 607.

Certain practices may be indicators that a background screening company isn’t following reasonable procedures. 

For example, if a report includes the following:

  • criminal convictions for people other than the applicant or employee 
  • a person with a middle name or date of birth different from the applicant’s
  • including screening reports with multiple entries for the same offense or that list criminal records that have been expunged or otherwise sealed.”1

The FCRA further provides two choices to CRAs that include public record information in their your background checks for employment purposes: 

  1. Notify the person who is the subject of the report when public record information is being reported; or 
  2. Maintain what the FCRA calls “strict procedures” designed to ensure that reported public record data is complete and up to date.

In addition, the Consumer Financial Protection Bureau (CFPB) issued an advisory opinion on November 4, 2021, stating that “a consumer reporting agency (CRA) that engages in-name-only matching violates the Fair Credit Reporting Act’s (FCRA) reasonable procedures requirement. (15 U.S.C. § 1681e(b).)” Although this was described as an “advisory opinion”, the CFPB made clear that the opinion is considered an “interpretive rule” issued under the CFPB’s authority to interpret the FCRA.

State Requirements

On the state side, the California Court of Appeals ruled in All of Us or None v. Hamrick that a DOB and driver’s license number may not be used to identify an individual when searching a court’s electronic criminal index as these data cannot be used to authenticate the identification. 

In Michigan, a rule that would have redacted dates of birth (DOBs) from court records statewide as of January 1, 2022, has been amended. Upon registration, individuals (such as consumer reporting agency personnel) may access birth dates on criminal records for identity matching purposes with the candidate’s consent under the amended regulation, which goes into effect on April 1, 2022.

Professional Background Screening Association (PBSA) Established Industry Practices

Derek Hinton, President, CRAzoom and member of the Background Screening Credentialing Council (BSCC) which oversees the PBSA accreditation process shared that “the PBSA Accreditation standard requires that a CRA must have and follow procedures requiring reasonable procedures to assure maximum possible accuracy when determining the identity of a consumer who is the subject of a record prior to reporting the information. 

The requirements to meet the standard is to match a minimum of two identifiers where one identifier is first name + middle name/middle initial where available + last name (or reasonable derivative thereof) and the second identifier is one of the following or any reasonable procedures that are demonstrably as effective::

  1. month of birth + day of birth + year of birth,
  2. SSN, 
  3. driver’s license number, 
  4. passport or country identification number, 
  5. current or previous addresses, or
  6. multiple partial identifiers.” 

One point that should be apparent is that address information is more important than ever for use as an identifier. So dust off that file of firms that focus on verifying addresses.

What Are “Reasonable Steps” You Can Take Look Like? 

Your first consideration should be to consider the commonality of the name. Statistically, reporting a name match only—when the name is Shadrach Meshach MacCullah—is more certain than reporting James Smith, DOB 09/15/1980. To help with the commonality of the name issue take a look at https://namecensus.com or check out a system such as www.NameGrades.com that gives a score on the commonality of name by year of birth and comes with a policy and procedure.

The article, What is Name Matching? that appeared on NetOwl.com shared that “an emerging tool that uses a machine learning algorithm and large-scale, real-world, multi-ethnicity name variant data can help you with this process. This approach automatically learns a collection of intelligent, probabilistic name matching rules from the data. Since the rules are automatically leaned from the real data, they are not bound by limitation of humans’ knowledge, but they reflect countless name variants that occur in the real world. It also has automated name ethnicity detection capabilities and applies the most appropriate matching models to names based on their ethnicity values in order to attain high accuracy.

Additionally, this approach handles cross-lingual name matching gracefully where users can search names in multiple languages and scripts even if they know only the name in, say, English.”2

You may also want to consider requiring applicants to provide a middle name, if they have one. We suggest you do not let them simply tab past the middle name entry field. Perhaps this a place for a factoid that states, omission of your middle name may cause a delay in processing your background check which could impact the selection process, as well as implement an automatic system procedure that enforces the rule.

There are also address history tools that you can make available to your case managers who make report/don’t report determinations. Several vendors in our industry have excellent products. 

As I am sure you are aware real typos have a major impact on the quality of name search results. A misspelled name can affect 50% or 100% of what you are looking for when you are performing name searches. This is another area where you may want to use a factoid to emphasize to applicants the importance of checking the spelling of their name to ensure it is correct and explaining how misspelled names can delay the background checking process and impact their selection. In addition, you may want to implement an automatic system procedure that enforces the rule.

Another step is to take advantage of the dynamic quality of AI which enables identification of name variations in real time. One example of this new hybrid intelligence/AI approach to name matching is the Rosette Name Indexer.

Conclusion

Name matching has emerged as an important issue in the background checking process because the FCRA requires “reasonable procedures to assure maximum possible accuracy of the information concerning the individual about whom a report relates” and frowns upon CRAs using name only matching. At the same time you have states suppressing the DOB which means other forms of identifiers are increasing important including address history.

The good news is that there are emerging technologies that can help you deal with the name matching issue:

  • commonality of name tools that can give a score predicating likely match
  • machine learning algorithms and large-scale, real-world, multi-ethnicity name variant data tools
  • AI tools

In addition, there is the need to tighten up your processes and procedures to reduce applicant problematic behaviors such as typos in the spelling of their name, omitting middle name or middle initial and address history is more important than ever.

  1. What employment background screening companies need to know about fair credit reporting act; https://www.ftc.gov/business-guidance/resources/what-employment-background-screening-companies-need-know-about-fair-credit-reporting-act
  2. What is Name Matching?; https://www.netowl.com/what-is-name-matching