One in 73 million: wrong (silly) in good faith

[lenny]

if an expert witness gives a ludicrous expert opinion which results in a conviction, then should they be held professionally accountable for the statement merely because they said it in good faith?

does the fact that a conviction resulted play any role?

these questions inspired by case in the UK where (according to a radio report) Prof Sir Expert said there was a 1 in 73 million chance the death of two children in the same family from X, which lead to the conviction of the mother, later overturned. i am more interested in the general question than the facts of that case, which may have been mis-reported. i am told that there are similar cases in DNA testing where statistics are similarly abused.

 

[Mr. Gruff]

Absolutely. The fact is that the expert in question wasn't very good at statistics, therefore when he used statistics to propose his "law", he was doing so in a field in which he was not comfortable. He should have been capable of recognising his own limitations, and should not have presented evidence to a court which was outside his own boundaries.

 

[Mojo]

There's a short article about this here.

 

[lenny]

There's a short article about this here.

Thanks for the link; it seems statistics was not only a problem for the expert witness, but also for the appeals court:

The appeal court ruling accepted flaws with the figure, but said it established “a very broad point, namely the rarity of double Sids”. Scream now.

presumably an appeals court has access to (someone who can interpret) freshman statistics...

But what chance does a "expert statistician" have against an "expert doctor" in front of a jury? Or better, what chance does justice have?

and should the prosecution be held liable for presenting a case (they should have) known to be flawed? or is their job merely to get a conviction?

 

[Ben Best]

But what chance does a "expert statistician" have against an "expert doctor" in front of a jury? Or better, what chance does justice have?

What are the motivations of these "experts"? Can a lack of criminal motive or interest be protection?

 

[neil]

If we jail or fine expert witnesses who make rather obvious errors, most will plead ignorence to avoid testifing (unless we pay $100 per minute) Unless we have evidence that they deliberately falsified, the publicity that they erred is sufficient punishment. Neil

 

[CapelDodger]

and should the prosecution be held liable for presenting a case (they should have) known to be flawed? or is their job merely to get a conviction?

A very important question. In the adversarial system truth is not the issue, even for prosecutors. Some experts become known as prosecution-friendly, which brings them unaccustomed attention - who knew from Meadows before all this? - and a (subconscious) symbiosis forms. The same thing can happen on the defense side.

There's a glaring mismatch between the worldviews of scientists and of lawyers. Lawyers - like professional politicians, and note the overlap - are taught that truth is malleable. Scientists are taught exactly the opposite. Law and science are immiscible.