The evaluation of an amount of ridge formations in a latent fingerprint impression to determine its usability for purposes of comparison, or the evaluation of an amount of corresponding ridge formations in two impressions in order to determine sufficiency to infer  identification, is based on the total quantitative-qualitative value or T-Value for the amount of ridge features in agreement and the relevant population for the case at hand.  

In terms of T-Values, inference for positive identification is expressed as follows:

If T/RP > 1, then ID

where,

RP = Relevant Population for the case at hand

T = T-Value

ID = Inference for Positive Identification

Therefore, inference for fingerprint identification is established when the T-Value exceeds the Relevant Population for the case at hand.

Identification Depends On Relevant Population

The term "individualization" or "identification" infers the elimination of others in a group leaving one [19].  The key word in this inference is "group".   Group refers to relevant population, which is a variable that impacts the number of friction ridge "look-alikes" that are likely to occur.  A look-alike is defined as “fingerprints from different origin that show an unexpected level of similarity that has the potential danger of a false conclusion about identity [27].”  For example, if the world fingerprint population was 66 quintillion, or 66 billion billions, then the number of look-alikes likely to occur for an amount of ridge details with a given T-Value would be significantly more than if the fingerprint population was only 66 billion, or the current world fingerprint population. 

For example, based on a hypothetical fingerprint population of 66 quintillion, the estimated number of friction ridge look-alikes for an arrangement of 12 excellent ending ridge formations located in a non-pattern force area in excellent agreement, which is and has been generally considered “safe” to individualize [28] [29], would be 16,249,727*.  This number of look-alikes would clearly support invalid basis to establish sufficiency to infer positive identification.

*Note: The estimated number of such look-alikes in this population group can be calculated by the T Model's formulae.

The relationship between an amount of ridge formations found in agreement in two impressions, chance for look-alikes, and fingerprint population is reflected in the following statement by Dr. David Stoney:

“The value of any fingerprint for identification is inversely proportional to the chance of false association.  This chance depends on the number of comparisons, which are attempted.  Each attempt carries a potential for chance correspondence and the greater the number of attempts, the greater the overall chance of false association.  "Attempt" means both the number of possible positioning on one individual and the number of different individuals with which the print is compared.  A fingerprint model should address this issue and provide a means to determine the number of attempted comparisons. [30]”  

The exclusion of all others in a group leaving one means relevant population is a variable that should be defined in order to establish valid basis for sufficiency to individualize.  For example, if a murder is committed in a locked jail cell that contains 20 inmates then the group of people that should be considered suspects is 20.  As a result the relevant fingerprint population is defined as 200.  In this scenario a conservative total world fingerprint population of 66 billion, is not relevant because the only group of people who could have committed the crime at the specific time and place were in the jail cell.  As a result a much smaller aggregate amount of corresponding ridge formations in a latent v. exemplar would be needed to infer positive identification.  A smaller amount is required because the numbers of look-alikes that exist is a function of fingerprint population such that the smaller the fingerprint population, the fewer the number of look-alikes will exist. 

Similarly if a bank robbery is committed in Santa Clara county, then any amount of corresponding ridge formations found in any latent fingerprint impressions recovered from the crime scene should take into consideration the conservative upper-bound number of people who could have committed the crime, i.e. the fingerprint population for that area.  The 2006 estimated population for Santa Clara County based on US Census Bureau figures was 1,731,281 people, which equates to a fingerprint population of 17,312,810.  Based on this fingerprint population, the number of look-alikes that can occur substantially increase.  As a result, a larger and/or better amount of corresponding ridge formations are required in order to establish valid basis for sufficiency to individualize (see Relevant Fingerprint Population).

The Evaluation to Infer Identification

The equation "If T/RP>1, then ID" may also be expressed as follows:

If RP/T<1, then ID

The above equation means that if the relevant population exceeds the T-Value, then there is a likelihood for a duplicate, a twin or friction ridge "look-alike" to be present in that population group, and therefore by definition, inference for identification to the exclusion of all others in the group is not justified. 

In essence, it is the estimated number of close matches or look-alikes for an amount of corresponding ridge features in two impressions based on the relevant population for the case at hand that determines the threshold for sufficiency.  That is what demarcates valid or invalid basis for sufficiency to infer identification, not the examiner. 

If the number of calculated look-alikes is greater than 1, then by definition there is no valid basis to infer identification.  If the number of calculated look-alikes is less than 1, then by definition, there is valid basis to infer identification. That's all.

It is the The T-Model that determines sufficiency or insufficiency to infer identification, not the examiner.  In this framework, the evaluation may be considered completely objective (see Objective Conclusions and Fingerprint Calculator.

NEXT PAGE >>> 

In order to establish individualization to the exclusion of all other sources in any fingerprint population, the number of likely duplicate aggregate amounts of corresponding quantitative-qualitative ridge formations or look-alikes should be less than 1.

In order to define sufficiency to infer positive identification, in a manner that can be most easily understood by the layman, formulae to calculate the estimated number of look-alikes likely to occur for any fingerprint population group and for any combination of corresponding ridge formations found in two impressions was developed (see The Formulae).