Biology Reference
In-Depth Information
Because with DNA evidence the probability under the prosecution's hypothesis is
equal to 1, the value is the same as when calculating a random match probability;
therefore, if we use the example from Chapter 8 we would have a likelihood ratio
of over 4 trillion. The way that the value is expressed differs from a random match
probability. The likelihood ratio would normally be expressed in a report with a
statement, for example as in the likelihood ratio statement; the following statement
is based on a likelihood ratio of 1 billion.
Likelihood ratio statement
DNA analysis of the bloodstain from the crime scene gave a full DNA profile
that matched that of the suspect. If this blood did not come from the suspect
then the STR profile must match by chance. The results of the DNA analysis
are approximately 1 billion times more likely if the DNA came from the suspect
than if the DNA came from a random unrelated male in the population (a billion
is a thousand million).
The figure being quoted in a likelihood ratio is the odds in favour of the proposition
put forward by the prosecution. When the statement reads that 'it is 1 million times
more likely that the DNA came from the accused than if it came from any unrelated
male', the figure of 1 million is not a probability but is an odds value, that is
how many times more likely it is that the DNA matched the crime scene stain if
it originates from the suspect, compared to coming from any other unrelated male.
The frequency approach has a problem when the chance of a match, or the match
probability, exceeds the total sample size. This is not the case when quoting odds
in favour, as odds can reach near infinity. In a horse race with only five horses, one
horse may have odds against of 10 to 1 and here the odds outweigh the number of
possibilities.
The disadvantage with this statement is that it can seem to be cumbersome when
presented to a jury. It is easy to make an error and state the probability that the
evidential material came from the suspect instead of the probability that the DNA
profile obtained from the evidential material matches that of the suspect.
Using the current multiplex STR kits that analyse between 10 and 15 STR loci,
the match probabilities and therefore likelihood ratios are extremely high. In order to
avoid some of these complications in presenting huge numbers, and also because it
is easier for the scientist to express verbally the weight of the evidence, it has been
suggested [3, 8] that verbal scales might be used for likelihood ratios as presented
in Table 9.1.
The use of verbal equivalents is itself a contentious issue because of its subjective
nature and also because some believe that it is encroaching on the role of the jury.
