Fire Statistics Review

4. Fire Statistics, Scotland

In this section of the report we focus on Fire Statistics, Scotland, reporting on fire and non-fire incidents attended by the Fire and Rescue Service. The data is currently produced annually, with supporting datasets of key data at local authority level and long term trends. The data is collected through a web-based Incident Recording System (IRS) managed by DCLG, which allows data to be compared across the UK.

4.1 How do people use the statistics

4.1.1 Using the publication

63% of respondents said they use the Fire Statistics, Scotland publication, 31% said they do not but may do so in the future and 6% said no, they do not use it.

The next questions in this section were answered by the 22 respondents who said that they use the publication.

We asked how they view the publication - online or in print, and HTML or pdf. Users could select more than one option. Most respondents used an online pdf version of the publication, though they also used the HTML and printed pdf formats. No one printed out the HTML version of the publication.

How do you normally view the publication (please select all that apply)?	Per cent
Online as HTML	32%
Online as PDF	82%
Printed version of HTML	0%
Printed version of PDF	32%
Total respondents	22

Table 2 : Viewing the publication

We also asked whether respondents use the spreadsheets containing tables and charts from the publication. 77% answered that they do, another 14% said that they may do in the future and 9% said no they do not.

We wanted to know approximately how often users referred to the statistics in the publication.

In general, how often do you refer to information within Fire Statistics, Scotland on the following topics?
	Per cent
	Weekly	Monthly	3 or 4 times a year	Twice a year	Once a year	Never
Main Points (highlight report)	5%	9%	36%	18%	23%	9%
Total number of fires	5%	14%	45%	14%	14%	9%
Fire casualties	5%	9%	55%	14%	9%	9%
False alarms	5%	9%	55%	0%	18%	14%
Special service incidents	5%	5%	36%	5%	27%	23%
Building fires	5%	9%	41%	9%	23%	14%
Outdoor fires	5%	5%	41%	9%	18%	23%
Fire rescues	5%	5%	41%	14%	18%	18%
Deliberate & accidental fires	5%	5%	50%	18%	18%	5%
Smoke alarms	5%	5%	59%	0%	18%	14%
Fire cause/source of ignition	5%	5%	55%	0%	18%	18%
Spread of fire and appliances attending	5%	5%	32%	14%	23%	23%
Fires by time of call	5%	5%	27%	9%	32%	23%

Table 3 : Frequency of use

There were 22 respondents to each of these questions. The most common frequency of using the data was 3 or 4 times a year. All topics were used, to some extent, by some respondents. Around a quarter of people who use the data never use data on special service incidents, outdoor fires, spread of fire and appliances attending or fires by time of day. The most used category was deliberate and accidental fire numbers.

4.1.2 Using extra datasets

Additional datasets providing long-term trend data for casualties, fires and false alarms are supplied alongside the Fire Statistics, Scotland bulletin. These datasets are supplied as Excel spreadsheets. We asked the 22 respondents who used the publication whether they used the datasets. 59% had used them, 32% had not but would do so in the future and 9% would not use them.

Further datasets provide data on fires, casualties, false alarms and special services at Local Authority level, and are supplied after publication. These datasets are supplied as Excel spreadsheets. We asked the 22 respondents who used the publication whether they used the datasets. In this case 46% had used the datasets, 27% had not, but would do so in the future and 27% would not use them.

Of the 22 users of the publication, we asked what they used the data for. The distribution of responses is shown below.

Figure 1 : Purpose of the data

General awareness of trends and issues was the most common use of the statistics, followed by aiding research and analysis, reporting at national and local authority levels, benchmarking and performance management.

In support of the response of 'other sub-national areas' the user stated a need for data that covers Scotland, Wales, England and Northern Ireland.

Other reported uses of the data include 'Training and teaching fire safety and in publications', and 'to inform publicity and media campaigns'.

4.2 Changes to Fire Statistics, Scotland

The format of the Fire Statistics, Scotland bulletin has evolved over time and we took this opportunity to propose some changes. The intention is to provide the information in the most useful way, providing an appropriate balance between commentary and tabular data. In the first instance we asked users about levels of detail in the publication and associated data on the web.

4.2.1 Publication format

We are considering reducing the level of detailed information in the Commentary section of Fire Statistics, Scotland and providing this more detailed information in the format of Excel tables, published alongside the main report. This would result in a more concise "Commentary" section summarising key points and statistics with the underlying data being accessible for users to carry out their own analysis.

In section 4.1, we explored users' preference for viewing the Fire Statistics, Scotland publication. Here we consider how they use the data tables and charts. These are currently published in two main formats:

i) The main HTML/PDF document, containing written commentary on the main statistical findings with an appendix of statistical tables, also in HTML/PDF format; and

ii) A spreadsheet containing all tables and charts from the publication in Excel format

We are aware that many users choose to access the tables and charts directly from Excel, possibly because they find it easier to analyse the information in this format. For this reason, we are proposing that in future Fire Statistics, Scotland publications we will:

i) remove the HTML/PDF copies of tables and charts from the Appendix

ii) publish a spreadsheet containing Excel copies of all tables and charts

We started by asking respondents whether this proposed new publication format would meet their requirements and whether they have any concerns about whether the publication will still meet their needs.

88% of respondents said they had no concerns and 12% did have concerns.

Of the respondents with concerns, one respondent cited technical difficulties they have had in the past with Excel spreadsheets on a website. Another appreciates a single document which incorporates tables for quick reference, supported by the Excel tables for deeper analysis. Without the tables in the publication, users would have to open the spreadsheets even for a single number. Another user cites the need for information at local authority level.

4.2.2 Detail in the Commentary section

We believe that the Commentary section of Fire Statistics, Scotland may contain more detailed statistical information than many users require. As a result, we intend to streamline this section of the publication so that it provides users with a more concise summary of key findings and statistics, leaving the more detailed data easily accessible in the format of Excel sheets for those users who require it.

Only one respondent had any concerns that this change would prevent the publication from meeting their needs. These concerns were tentative, expressing the possibility that statistics without commentary may fail to provide the necessary detail required to complete analysis of the data.

4.3 Priorities of topics

So that we could better decide on the content of future Fire Statistics, Scotland publications, we asked respondents to select the three most important information topics of those currently provided.

THREE topic areas which are of MOST value to users	Per cent
Total number of fires	61%
Fire casualties	55%
Main points (highlight report)	45%
Fire cause/source of ignition	30%
False alarms	27%
Building fires	24%
Deliberate and accidental fire numbers	21%
Smoke alarms	18%
Fire rescues	6%
Spread of fire and appliances attending	6%
Special service incidents	3%
Fires by time of call	3%
Outdoor fires	0%
Total responses	33

Table 4 : Users' priority of the topics

Unsurprisingly, fires and fire casualties were most often cited in the three most important topics, with no one selecting outdoor fires amongst their top three.

4.4 Changes to data presentation 4.4.1 Geographic levels of data

As we noted at the start of this report, it is no longer meaningful to present data for the 8 legacy Scottish Fire and Rescue Service areas. This means we have to decide the geographic level at which we provide the information in Fire Statistics, Scotland. Having asked respondents about the topics they find most useful, we then asked about the geographic level they would like to see.

While two thirds of respondents would like to see reporting at Scotland and Local Authority level data, 12% requested other geographic breakdowns.

What level(s) would you find the key fire information useful at (please select all that apply)?
Scotland	66.7%
Local Authority	66.7%
Other (please specify)	12.1%
Total	33

Table 5 : Geographic breakdowns

Respondents additionally requested data in the following geographic levels.

• Prison Fire Statistics
• LSO area
• IMG zones or Scottish data zones
• Wards

In the table below a short definition of the geographic areas, and the number of the areas within Scotland are detailed to provide more context for deciding about reporting levels.

Geography	Definition	Number in Scotland
Current options
Scotland		1
Local Authority		32
Consultation responses
LSO area	The Scottish Fire and Rescue Service is organised into aggregations of local authorities, each headed by a Local Senior Officer. Currently there are 17 LSO areas, though fluctuation is possible at the discretion of the Scottish Fire and Rescue Service.	17
Intermediate geographies (IMG)	IMG are geographies used in Scottish Neighbourhood Statistics (SNS). They are aggregations of datazones within local authorities with each zone containing between 2,500 and 6,000 people. The intermediate geography was introduced because not all statistics are suitable for release at the data zone level due to the sensitive nature of the statistics or for reasons of reliability.	1,235
Scottish data zones	Data zones were created by combining 2001 Census output areas. The data zone geography covers the whole of Scotland and nests within local authority boundaries. Data zones have populations of between 500 and 1,000 household residents, and some effort has been made to respect physical boundaries.	6,505
Wards	Multi member electoral wards replaced the 1,222 existing wards, each electing one councillor, which were in use from 1999. Within SNS the 353 wards are defined in terms of 2001 data zones and statistics relating to these areas are 'best-fit'.	353

Table 6 : Geographic area definitions and numbers

4.4.2 Presenting long term trends

We asked users about ways of presenting additional datasets containing long-term trend data. These have previously provided casualties, fires and false alarms data at Fire and Rescue Service level. We wanted to know if the data would be used at Scotland level if no FRS level data was available. Most respondents would use Scotland level data, either often or occasionally, with just under 10% not using Scotland level data.

4.4.3 Publication date

The changes being made to Fire Statistics, Scotland mean the 2013-14 publication will take additional time to prepare. Whilst Fire Statistics, Scotland has previously been published annually in October, it is likely that the 2013-14 publication will not be ready until November or December 2014. We asked users if this later publication date would meet their requirements. Of 33 responses, one said that it would be helpful to have the data available in August to feed into the Single outcome agreement (SOA) for the local authority.

4.4.4 More or less data

We asked users whether they would like to see extra data included that was not already there or whether they would like some of the existing data removed.

Almost half the respondents (45%) said they would like to see new fields, many of which were specific to their particular interests.

The table below contains all of the comments. These have been grouped into categories for analysis purposes, but have not been aggregated in any way (i.e. if two respondents proposed a change there are two bullet points in the table). These will be examined more closely in the Discussion section of this report.

Premises information	Information about dwelling fires by housing tenure Accidental dwelling fires, details of where the fires took place i.e. tenure breakdown Incidents in listed buildings - causes, spread of fire, etc Incidents in Scotland's statutorily protected heritage buildings Fires involving timber framed residential and commercial buildings. Details of construction type and age of buildings involved in fires.
Cause	Detailed causes of fires. Cause factors More information on identified contributory/underlying factors More detailed information concerning causes of false alarms
Detection and suppression systems	Fires involving automatic fire suppression systems Kind of commercial fire system in operation at a site and if it worked effectively More information about the type and nature of the fire detection and warning system/devices Information about fires controlled by automatic fire suppression (sprinkler) systems Information about incidents involving the actuation of carbon monoxide alarms and/or suspected carbon monoxide poisoning.
Research projects	Information about seasonal variations/peaks in fire demand A breakdown of fires involving rubbish/refuse by type More about the trend in the severity of dwelling fires and dwelling fire casualties Had portable extinguishers been deployed? Damage to equipment and injuries to staff as a result of antisocial behaviour
Specialist topics	Notification of the percentage for each category of IRS incident type has been quality assured for accuracy before publication Detail on the number of fire incidents and fire false alarms at prison establishments Where fires originate from commercial kitchens we would like it to be recorded when the fire spreads to other parts of the building via the grease extract duct system.
FRS data - not incident data	Had the premises a proper fire risk assessment in place? - to see impact of Fire (Scotland) Act conformance. Policing of Fire (Scotland) Act, notices issued and audits carried out. Home Safety Visits Electric Blanket Safety Testing Proactive Preventative Work

Table 7 : Extra data fields requested in the Fire Statistics, Scotland publication

Users did not want to see anything excluded from the publication.

4.5 General Comments about Fire Statistics, Scotland

We asked respondents to provide any additional comments about Fire Statistics, Scotland. Of those who did, two respondents said that annual reports were suitable, two felt quarterly updates would be useful and one emphasised the need for timely data, published as soon as possible after it is collected.

The long term trend data was noted as important, particularly the ability to look back at the trends and consider the transition to a single service.

There were detailed comments on technical terms (e.g. 'smoke alarms' should be reviewed to comprise both smoke and heat alarms) and the provision of rates in addition to numbers. One comment expressed general appreciation for keeping the statistics within the publication, as it helps to put the content into context.