Changes to data inputs

The April 2020 release (R34) incorporates a range of significant developments to the methodology. These affect both the definitions and calculations made and, in turn, the audience numbers produced.

Here we summarise the changes applied to the data inputs – namely the inventory, sample, traffic intensity model and the population counts.

Inventory

The total inventory that Route measures has been updated. In total Route now provides audience measures for 392,432 posters and screens.

Sample

  1. The sample has been refreshed and reduced from a five-year to three-year database, improving the contemporaneity of the data. Route R33 data was based on Travel Survey years 6-10, with a sample size of 29,750.  Years 6-8 have been removed from R34 and the data are based on Year 9-11, with a sample size of 21,994.
  2. Although the number of participants has gone down, the total number of recorded days has remained reasonably stable at 345,127. This is because we have moved from a 9-day to a 14-day survey period
  3. With all years of sample based on 9-day data collection now removed, a 14-day probability model has been introduced. This means that reach is now benchmarked to a 14-day rather than 9-day measure of real behaviour.  Having this measure over a longer period means Route has improved its estimates of how reach builds over time.

Travel Survey – Multi-Sensor Tracker Effects

  1. More MST data are available and applied to an increased number of locations. We are now using the entry and exit data for a larger selection of points of interest. These are based on actual observations from the travel survey. Previously they were allocated randomly according to the behavioural models.

Where the MST data is taking effect for the first time, it is likely to result in changes at frame level. This is because the entry and exit data influences the proportions using each route through the location.  The effects are likely to be less noticeable at more aggregated campaign levels.

The number of locations that benefit from the MST measure is as follows: rail increases from 50 to 59, indoor shopping centres from 57 to 90 and Underground/DLR stations from 44 to 80.  For the first time there is sufficient MST sample for three airports to have MST entry/exit data (Gatwick South, Stansted and Edinburgh).


Traffic Intensity Model, Populations and Counts

The universe estimate of adults aged 15+ has been updated from 52,771m to 53,045m; a 0.5% increase. In addition there are population updates for the following environments:

  1. A 0.7% increase in rail populations, varying by line/region
  2. An 8% increase in London Underground populations
  3. A population update for airports which sees increases for Stansted Luton and Edinburgh in particular, as well as small decreases for some of the other airports

The traffic intensity model (which determines the volume and speed of vehicle traffic and pedestrians) has been enhanced. The updates to the model include:

  1. A geometry update. HERE street Links 2018 q1 replaces 2017q1. There are some new links and changes in traffic rules
  2. An update to the HERE Traffic Patterns – which governs the speed at which people travel
  3. An update of the AADF (road traffic) counts from 2015 to 2017. This shows a top line increase of 2% in traffic passing count points common to both data sets
  4. Updates to the pedestrian counts for London Underground, rail, airport and motorway service stations.
  5. In addition, new PMRS and Local Data Company counts have been applied to high streets and shopping centres. Overall, our test data predict a 2.1% increase in pedestrian counts for roadside frames

 

 

Leave a Reply

Your email address will not be published. Required fields are marked *