Premium
Trawling Genetic Databases: When a DNA Match is Just a Naked Statistic
Author(s) -
Scurich Nicholas,
John Richard S.
Publication year - 2011
Publication title -
journal of empirical legal studies
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.529
H-Index - 24
eISSN - 1740-1461
pISSN - 1740-1453
DOI - 10.1111/j.1740-1461.2011.01231.x
Subject(s) - frequentist inference , conviction , argument (complex analysis) , bayesian probability , statistic , psychology , statistics , bayesian inference , econometrics , law , political science , economics , mathematics , biochemistry , chemistry
Genetic databases are highly controversial. Significant controversy followed a report by the National Research Council (1996) concluding that a DNA match resulting from a database trawl is less probative than when only a single test is conducted. Legal scholars and statisticians have demonstrated why this conclusion is specious, but there is no empirical research examining what impact a trawl match has on jurors. The current experiment demonstrates that mock jurors are less likely to convict when a DNA match arises from a database trawl compared to a conventional confirmation case. Interestingly, however, the probability judgments of guilt did not differ between the cases. The Wells effect (Wells 1992) is consistent with this disjunction, where a trawl match is more likely to be viewed as naked statistical evidence that influences the perception of guilt but not guilty verdicts. A second study examined mock jurors' perceptions of the arguments advanced by the Bayesian and frequentist camps over the probative value of trawl matches. The frequentist argument led to fewer convictions, while the Bayesian argument increased the conviction rate, as expected. The Bayesian argument also attenuated the Wells effect. Implications for the introduction of trawl matches are considered in light of these findings.