The defense may argue that the statistical interpretation of the test results is misleading and thus unreliable because the calculations were derived from a flawed database. The following are the three most common defense challenges to a database and examples of how to rebut these challenges.
 

1. The database was erroneously created because the persons who collected the blood from the donors did not do a thorough ancestry analysis to determine if someone was of one race or a racial mixture. 
   
   First, the prosecutor should review with the DNA analyst how the database was created and that it was done in accordance with NRC recommendations. Second, the prosecutor should have the analyst explain that the NRC II has studied the issue and concluded that there is no reason to suspect that persons who contribute to blood banks differ from a random sample of the population.
    
2. The database is comprised of too small a number of samples from which a reliable statistical calculation can be derived.
   
   First, determine from your analyst the size of the database used and ask the analyst to compare it with the size of the database used by the FBI.  Second, have the analyst explain that a statistically sufficient database requires a minimum of only 100 samples and that the FBI DNA database is also comprised of only a few hundred samples. Third, state databases were typically reviewed and validated by population genetic experts at the time of their creation. If that occurred, have the analyst testify that this external review occurred.
    
3. The database does not reflect all of the subpopulations, e.g., Haitian, Chinese, Jamaican, Mickosukee Indian, in the jurisdiction.
   
   First, have the analyst explain that the database includes the largest subpopulations in a jurisdiction merely to assist in interpreting the DNA results. Second, have the analyst review with the jury that considerable empirical data from published and peer reviewed research studies indicates that calculations made from different databases throughout the nation and the world are quite similar. The finding of this research is that allele frequency differences between subpopulations are considered to be statistically insignificant. The NRC II recommends the use of a theta correction in the calculation to account for these minor differences between subpopulations which was included in the statistical calculation used.

Back Forward