At the recent public consultation event on the proposed site, there was a curious display – one that on the face of it should undermine the whole concept of the siting of a park & ride scheme anywhere to the west of the city.
It was this one:
Let’s take a closer look at the pie-chart – taken at face value it is suggesting that 60% of the traffic currently using the Alphington Road corridor – and thereby potential users of the new P&R site by the Ide Interchange.
But the chart also raises a number of questions – questions that a statistician would be likely to ask.
Percentages in such diagrams hide a number of sins – where is the total number of vehicles surveyed?
Was it 100? Was it 1000? Or 10,000? and remember that 10,000 would only be about the number of vehicles using Alphington Road in a day.
So was the survey taken over an hour? A day? Or a week?
And how was the data collected?
Was it the result of roadside interviews? Or analysis of data collected by some sort of automatic numberplate recognition?
Earlier data collection has been well documented.
In August 2009, Parsons Brinckerhoff prepared the Alphington Junction Part and Ride Transport Assessment.
The report contained a table of a survey was conducted by Parsons Brinckerhoff at Matford P&R in June 2003. Users of the P&R were asked the origin of their journey and the results are summarised in the table below.
In 2004, roadside interviews [RSI] were undertaken at several sites, detailed in the map below.
These RSI identified the total market for potential Park & Ride users was identified using the origin and destination locations of cars.
This data was used in the Devon County Council’s Alphington Interchange Park and Ride Transport Report published in March 2011.
So is that new pie-chart an outlier? I believe so, and I think the problem stems from WHEN the data was collected.
I’ve already suggested there was a problem presenting undated data, so f my guess is correct, the figures stem from 2014 when Junction 29 was being remodelled and unusual traffic movements would mean drivers were seeking new and unusual ways to get into the city centre.
Whatever the reason for the errors, DCC must stop using this chart if they want to convince people that a P&R scheme is necessary on this site!
UPDATE [24 July 2015]
I have heard from DCC over their definition of East and West.
It appears that those coming from the East aren’t following a star or anything, but from A30 East or from Teignbridge and Plymouth direction. Which in turn means come from the (south) west, via A38 not the east and M5
The data was collected from from Roadside Interviews (RSIs) which are the most accurate surveying method for being able to establish where drivers are starting and ending their journey and therefore what the potential market for Park and Ride may be.
Due to the significant disruption and cost associated with organising these surveys, RSIs are rarely undertaken to support Transport Assessments for planning applications (so in this respect, having the data is a luxury).
There is no intention to undertake any further surveys of this type but I am told that the Transport Assessment will use Automatic Traffic Count 2014 data.
Generally, the Department for Transport considers traffic data collected within 5 years of the submitted analysis to be suitably representative of conditions