This document describes Checkr’s process in collecting and determining aliases for candidates when conducting background check screenings, and includes the following information:
- Checkr's sources for aliases
- Checkr's alias types:
- How Checkr uses aliases
- How Checkr evaluates 'partial' names
Aliases are different names or permutations of the same name associated with a candidate. When applicable, Checkr uses aliases to identify records under names that differ from the candidate’s provided name, thus increasing the comprehensiveness of the background check report.
Aliases are identified through names returned from various sources, including candidate creation, SSN Trace records, Motor Vehicle Records check, and document transcriptions. These returned names are stored as “name occurrences”, and evaluated to generate aliases.
- Candidate submission: The information submitted by the candidate to Checkr acts as a starting point for the candidate’s profile.
- SSN Trace: Checkr identifies additional names from SSN Trace data. These names are aggregated into the candidate’s record and stored.
- Motor Vehicle Records: MVR checks often return name variants. These names are aggregated into the candidate’s record and stored.
- Document transcription: Customer and Checkr support team document transcriptions may introduce additional names for the candidate.
- Name normalization: Checkr software identifies potential nicknames for a candidate. For example, if a candidate submits Bob as a first name, our software automatically recognizes that Bob is a nickname for Robert and stores Robert as a first name for that candidate.
Checkr uses name occurrences from all previous reports associated with an individual candidate when computing aliases
Checkr leverages a variety of algorithms to evaluate returned name occurrences, and aggregates this raw data to generate aliases for a candidate. Two types of aliases are generated in this process:
- Request alias: Request aliases are used to initiate screenings for the candidate. Package configuration will determine whether and how many distinct aliases are used in requests.
- Validation alias: Validation aliases are used to confirm that the records returned by our vendors match the candidate. In order to ensure accuracy, alias matches are only one of several identifiers used to match.
Request alias name selection
The selection process of request aliases ensures that distinct name components are represented and therefore searched. When generating request aliases, Checkr evaluates all available first, middle, and last names for each candidate returned in the sources listed above.
First name evaluation
When two different first names are collected, Checkr will select both as request aliases. For example, given Patricia Stewart and Martha Stewart, Checkr will select both Patricia and Martha as distinct request aliases.
Last name evaluation
When two different last names are collected, Checkr selects both as request aliases. For example, given Martha Jones and Martha Stewart, Checkr will select both Jones and Stewart as distinct request aliases.
Middle name evaluation
When evaluating multiple middle names, Checkr selects the most representative name variation as the request alias, allowing our providers to return all data that is likely to be related to the candidate.
- Given Patrick and P, Checkr selects Patrick
- Given Patrick, Pete and Paul, Checkr selects P
- Given Patrick and Neal, Checkr selects neither
Validation alias name selection
While request aliases are selected to broaden search coverage and increase comprehensiveness, validation aliases are generated to ensure all components from various name occurrences are represented when validating records.
The following example is constructed to illustrate these differences.
For example: Checkr receives the following name occurrences for a candidate from various sources:
- JOHN J SMITH
- JOHN SMITH
- JOHN S SMITH
- JOHN SAM SMITH
- JOHN JONES SMITH
Because “JOHN SMITH” is the name that can represent all other name occurrences, Checkr will select the name “JOHN SMITH“ as a request alias.
And, to ensure all name components are represented in record validation, Checkr will use both “JOHN SAM SMITH” and “JOHN JONES SMITH” as validation aliases to ensure maximum possible accuracy.
Checkr enables the use of aliases for the following screenings:
- National Criminal Database Check
- Sex Offender Registry Check
- Global Watchlist Records Check
- Federal Criminal Records Check
- Federal Civil Records Check
- County Criminal Records Check
- County Civil Records Check
- State Criminal Records Check
- Municipal Criminal Records Check
Partial name handling
In some cases, candidates may not have both first and last names. In these cases FNU is used in place of first name to designate First Name Unknown, and LNU is used in place of last name to designate Last Name Unknown. Because the same designations are also used in public records, Checkr accepts FNU and LNU as valid name input, as they will function as regular name values in searches, and provide a basis for matching records to be returned.
While it’s common for candidates to have no “first” name (or FNU) on the record, it is very rare for candidates to have no “last” name (or LNU). Candidates with a single name often submit it as their last, rather than their first for United States documents.
Checkr does not accept single letter last names as valid input, and customer API calls that POST single letter last names will be responded to with an error.
Checkr does accept single character first names as valid input.