ABSTRACT
This chapter discusses the real issues must instill a measure of caution when determining statistical evaluation of anything else other than a DNA profile. In the absence of whole population databases, sampling statistics are used to extrapolate frequencies of alleles used in the statistical evaluation of the DNA profile from a small database to a population at large. The interpretation of DNA profiles when there is a “match” between profiles requires the determination of a probability: the chance of observing that DNA profile again, under various conditions, and observed in a certain population, and with regard to case information. Calculating a likelihood ratio requires a probability model to describe the real-world processes underlying a DNA profile. Performance testing using a random simulated profile has been suggested as a way to further evaluate or qualify the estimated likelihood ratio, evaluating the specificity of the statistic. Specificity of the interpretation method is indicated by the measure of the likelihood ratios for non-contributors.
