Dear Dr Bullas
It is difficult to give a complete answer to your question either
based on traditional frequentist statistics or on the more recent
methods of Bayesian risk analysis without more information.
It is a tricky area, since what must be compared is the actual
performance in the period of the intervention with what would have
occurred in the same period (not an earlier period) without the
intervention. Many studies fall into the trap of not dealing with
possibility of "regression to the mean" bias, which bedevils studies
of treatments at accident black spots.
I suggest you read "Observational Before - After Studies in Road
Safety" by Exra Hauer (Pergamon) which deals with the issues in depth
and adopts an Empirical Bayes approach.
Best regards
Robert Cochrane
At 20:03 08/01/2010, Dr John C Bullas wrote:
>As an earth scientist and road surface specialist I am outisdeof my
>comfort zone with accident statistics
>
>I have data for values for KSIs and slights for the months September
>to December for 2008 when an
>intervention was in place and for the same months for 2005-2007 when
>it was not
>
>I believe since I cannot show the data is normally distributed, the
>wilcoxon rank sum test might be
>the best measure of whether 2008 is significantly lower than the other
>years (as a group)
>
>I do not have control data to hand nor traffic flows so will have to
>state an assumption
>
>Is this test (aka MANN -WHITNEY 'U' ?) a good choice?
>
>Dr B
Robert A Cochrane
7 Lawn Terrace
Blackheath
London SE3 9LJ UK
Tel (44) 020 8297 1978
Mob (44) 07764 197 701
|