We have a new website go to gov.scot

Paper 14th May 2003 - LGRAS(03)4

LGRAS (03) 4 - THE SUBMISSION AND THE USE OF THE "STATS 19" DATA

1. Introduction

1.1 This paper covers various points to do with the submission and the use of the "STATS 19" returns for Scotland. Members of the Group are invited to comment on these matters at the meeting, and to contact the Scottish Executive Transport Statistics branch regarding any subsequent developments or difficulties.

1.2 To facilitate communication in future, Members of the Group (especially those who are "STATS 19" data suppliers) are invited to provide Transport Statistics with their email addresses by emailing transtat@scotland.gsi.gov.uk (if they have not already done so).

2. Submission of "STATS 19" returns

2.1 At the meeting on 24 April 1998, Members of the Group representing data suppliers agreed that every return (apart from the "December" return) should be made within 6 weeks of the end of the month to which the return relates. For example, the "January" return should be made by around 15th March, unless there were major unforeseen problems (such as those that arise from the conversion of computer systems). Because of the backlog which builds up over Christmas / New Year, it was agreed that the "December" return was an exception, and should be sent to the Scottish Executive by the end of February.

2.2 At the time of writing (2 April 2003), the Executive should have received all the returns for January 2003 (on the basis of the "six weeks" agreement, they should all have been made by the middle of March). The latest returns which we have received are for the following months:

December 2002 - three areas

January 2003 - two areas

February 2003 - three areas

The position is much better than it was in previous years. A Police Force which, last year, had a backlog of overdue returns has caught up, for which we are grateful. Some Police Forces have been making regular returns within two weeks of the end of the month, for which they are congratulated . We appreciate the efforts made by all the data suppliers, which have resulted in a more timely submission of data. This is reflected in the latest DfT's quarterly estimates for Great Britain as a whole (see section 5.2 of this paper). Every Scottish Police Force is shown, in the final table of DfT's publication, as having made complete returns for the relevant quarter. We are grateful to Scottish data suppliers for their good work.

2.3 Members of the Group representing those data suppliers are asked to ensure that henceforth they make their returns within the agreed timescale (see paragraph 2.1).

3. Points from data quality checks

3.1 The Scottish Executive's checking of the STATS 19 data received from data suppliers has, for some returns, highlighted:

  • entries which look suspicious, although they are valid codes;
  • entries that are invalid but, as yet, are not identified by the SE validity checks; or
  • entries where the data is missing.

These are detailed below. 3.2 Speed Limits (STATS 19 variable reference 1.15)

The speed limit of the road is used in calculating the built-up (up to 40 mph) / non built-up (over 40 mph) split for injury road accidents. Some returns are showing accidents on motorways with speed limits of 40 mph or less. However, after checking with SE Road Network Maintenance and Management colleagues, we understand there to be no such permanent speed limits. It is therefore likely that officers are mistakenly reporting temporary speed limits for these stretches of roads. STATS 20 Note C states that the normal speed limit should be entered, not the temporary one. A validity check is to be added to the errors report when the Quality Review changes are made to the Executive's computer system.

3.3 Place Accident Reported (ref. 1.26)

STATS 20 Note A states that the code used should be 1, for 'At scene', in cases where the reporting officer attends the scene of the accident. The use of code 1 is not dependant on the officer having witnessed the accident. Despite this definition, some Police Forces report a very high percentage of accidents coded 2 for 'Elsewhere'.

Members of the Group will see (from paper LGRAS[03] 5) that, following the Quality Review, this variable is to be renamed as "Did A Police Officer Attend Accident", with effect from January 2005. This should reduce the likelihood of such errors thereafter.

3.4 Vehicle Movement Compass Point (ref. 2.8)

Some returns are reporting invalid codes i.e. 01, 02, 03, 04, 05, 06, 07, 08. This seems to be due to a mis-reading of the instructions. If a vehicle is parked at the kerb then the digit in the 'from' box should be from 1 to 8 (i.e. an indication of the direction it would have come from to park). In some cases it may simply be that the digits are being entered the wrong way round. A validity check is to be added to the errors report when the Quality Review changes are made to the Executive's computer system.

Members of the Group will see (from paper LGRAS[03] 5) that, following the Quality Review, the coding of parked vehicles will be simplified with effect from January 2005.

3.5 Vehicle Location at time of Accident - Restricted Lane / Away from Main Carriageway (ref. 2.9b)

Some returns are using code 2, implying that accidents took place on 'Tram/Light rail track'. Our understanding is that there are no such tracks in Scotland at present.

3.6 Driver/Casualty Age (ref. 2.22 and 3.8)

It is appreciated that it is not always possible to obtain the exact age for drivers and casualties. However, in cases that do not involve parked cars or hit and run accidents, we would be grateful if an estimate could be submitted.

3.7 Postcode Information (ref. 2.27 and 3.18)

Although most returns are averaging 85% completion of these variables, it is important that we try to maintain and improve on this. The data may become increasingly important if we are asked to produce indicators based on injury road accident statistics for areas of residence.

3.8 School Pupil Casualty (ref. 3.13)

The number of child casualties on their way to and from school is of significant interest both nationally and locally. It is therefore important that this information is accurately recorded and submitted to the Executive.

Members of the Group representing data suppliers are asked to note the need to ensure the quality of their data in these areas.

4 Deleted or re-classified accident records, and the reconciliation of Scottish Executive and Police Force figures

4.1 Deleted or reclassified accident records

Data suppliers are reminded that they must inform the Scottish Executive whenever (a) they delete any "injury" accident sets from their systems or (b) they reclassify any accidents from "injury" to "damage only". Otherwise, those accidents will continue to be counted as "injury accidents" on the Executive's system, and therefore its figures will be too high.

Similarly, data suppliers must ensure that their systems send to the Executive the records for any accidents which they re-classify from "damage only" to "injury". Otherwise, those accidents will not be counted in the Executive's statistics of injury road accidents.

Members of the Group representing data suppliers are asked to note the importance of the above points, and to change their systems, if necessary, to ensure that the Scottish Executive is henceforth informed of all deleted or reclassified accidents.

4.2 Reconciliation of Scottish Executive and Police Force figures

The quarterly and end-year reconciliation process is aimed at improving the quality of statistics provided in our publications and in answer to enquiries, such as Parliamentary Questions. It also allows the Executive and Police Forces to identify any flaws in their systems whereby data are either not being transferred or not being processed properly.

Some Police Forces may not be doing enough to ensure that the reconciliation is carried out properly on both a quarterly basis and an annual basis. In the course of the reconciliation of the final figures for 2002, it was found that four Police Forces had, between them, records for a total of about 200 accidents which had occurred in 2002, but which had not been sent to the Executive at the appropriate time. The nature of the shortfalls differed between the Forces. For example:

  • one Force had not sent to the Executive records for about 2-4 accidents of the accidents in its area in each month of 2002;
  • most (but not all) of another Force's "missing" records were for accidents which had occurred in October and December 2002.

So, it appears that different Forces' systems are affected by different kinds of problem. For example, it might be that the first Force's system is not "flagging" (as "to be sent to the Executive") the records of the few accidents each month that are reclassified from "damage only" to "injury"; and it might be that the second Force's problem was caused by the late receipt of details of some accidents.

It should normally be possible to to identify, and deal with, such problems at a much earlier stage than the finalisation of the figures for the year. That is why the quarterly reconciliation prints are produced: if a Police Force's system is failing to send some of its accident records to the Executive, that should be discovered when the figures for the first quarter or two are reconciled. It is far better for all concerned if such problems can be identified and dealt with then, rather than as part of a "last minute rush" to finalise the figures for the year.

Members of the Group representing data suppliers are asked to ensure that reconciliation takes place properly on both a quarterly and an end-year basis.

In cases where the reconciliation indicates that there must be problems with their systems, Members of the Group representing data suppliers are asked to deal with them by providing any missing accident records swiftly, and by changing their systems as soon as possible thereafter to ensure that their subsequent "Stats 19" returns are not affected by the same kinds of errors.

5. Publication of the figures

5.1 Provisional totals for Scotland for 2002 will be published in Key 2002 Road Accident Statistics in June. The precise date will be decided nearer the time. Copies of this bulletin will be sent to Members of the Group, and to non-members who receive copies of LGRAS papers, at the time. Please contact the Secretary if you do not receive a copy.

5.2 Members of the Group will recall (from the discussion at the meeting in June 2002 - see paper LGRAS [02] 4) that DfT is publishing quarterly estimates of the total numbers of road accidents and casualties for Great Britain as a whole (and that the advice of the Group was that there was no demand for quarterly estimates for Scotland). The final table in each edition of DfT's "quarterly estimates" publication shows how up-to-date (or otherwise) the figures for each of the 50 Police Force areas in Great Britain were at the time that DfT started to produce those estimates. The first such publication, which provided estimates for the second quarter of 2002, showed that DfT had not received complete returns for that quarter from 11 areas across Great Britain, of which two were in Scotland. The next publication, for the third quarter, showed that DfT had not received complete returns for that quarter for 12 areas across Great Britain, none of which were in Scotland. In other words, at that time, all eight Scottish areas were "up-to-date". Scottish data suppliers are thanked for making their "STATS 19" returns more promptly.

5.3 The "final" detailed figures for Scotland for 2002 should be published by the end of November in Road Accidents Scotland 2002. The precise date will be decided nearer the time. Members of the Group, and non-members who receive copies of LGRAS papers, will receive their copies shortly afterwards.

5.4 Members of the Group are invited to note these points.