Tranche two - about the data

Data in this report is from a range of sources, including:

  • Home Office;
  • Office for National Statistics (ONS);
  • Chartered Institute of Public Finance and Accountancy (CIPFA);
  • public perception survey;
  • our inspection fieldwork; and
  • data we collected directly from all 45 fire and rescue services in England.

Where we collected data directly from fire and rescue services (FRS), we took reasonable steps to agree the design of the data collection with services and with other interested parties such as the Home Office. This was primarily through the FRS Technical Advisory Group, which brings together representatives from FRSs and the Home Office to support the inspection’s design and development, including data collection. We gave services several opportunities to validate the data they gave us and to ensure the accuracy of the evidence presented. For instance:

  • We asked all services to check the data they submitted to us via an online application.
  • We asked all services to check the final data used in the report and correct any errors identified.

We set out the source of Service in numbers data below.

Methodology

Population

For all uses of population as a denominator in our calculations, unless otherwise noted, we use ONS mid-2017 population estimates (Spreadsheet). This is the most recent data available at the time of inspection.

BMG survey of public perception of the fire and rescue service

We commissioned BMG to survey attitudes towards fire and rescue services in June and July 2018. This consisted of 17,976 surveys across 44 local fire and rescue service areas. This survey didn’t include the Isles of Scilly, due to its small population. Most interviews were conducted online, with online research panels.

However, a minority of the interviews (757) were conducted via face-to-face interviews with trained interviewers in respondents’ homes. A small number of respondents were also interviewed online via postal invitations to the survey.

These face-to-face interviews were specifically targeted at groups traditionally under-represented on online panels, and so ensure that survey respondents are as representative as possible of the total adult population of England. The sampling method used isn’t a statistical random sample. The sample size was small, varying between 400 and 446 individuals in each service area. So any results provided are only an indication of satisfaction rather than an absolute.

Survey findings are available on BMG’s website.

Staff survey

We conducted a staff survey open to all members of FRS workforces across England. We received 2,905 responses between 1 October 2018 and 15 February 2019 from across 16 FRSs during this period in Tranche 2.

The staff survey is an important tool in understanding the views of staff who we may not have spoken to, for a variety of reasons, during fieldwork.

However, you should consider several points when interpreting the findings from the staff survey.

The results are not representative of the opinions and attitudes of a service’s whole workforce. The survey was self-selecting, and the response rate ranged from 8 percent to 31 percent of a service’s workforce. So any findings should be considered alongside the service’s overall response rate, which is cited in the report.

To protect respondents’ anonymity and allow completion on shared devices, it was not possible to limit responses to one per person. So it is possible that a single person could have completed the survey multiple times. It is also possible that the survey could have been shared and completed by people other than its intended respondents.

We have provided percentages when presenting the staff survey findings throughout the report. When a service has a low number of responses (less than 100), these figures should be treated with additional caution.

Due to the limitations set out above, the results from the staff survey should only be used to provide an indicative measure of service performance. 

Service in numbers

A dash in this graphic indicates that a service couldn’t give data to us or the Home Office.

Perceived effectiveness of service

We took this data from the following question of the public perceptions survey:

How confident are you, if at all, that the fire and rescue service in your local area provides an effective service overall?

The figure provided is a sum of respondents who stated they were either ‘very confident’ or ‘fairly confident’. Respondents could have also stated ‘not very confident’, ‘not at all confident’ or ‘don’t know’. The percentage of ‘don’t know’ responses varied between services (ranging from 5 percent to 14 percent).

Due to its small residential population, we didn’t include the Isles of Scilly in the survey.

Incidents attended per 1,000 population

We took this data from the Home Office fire statistics, ‘Incidents attended by fire and rescue services in England, by incident type and fire and rescue authority’ (Spreadsheet) for the period from 1 October 2017 to 31 September 2018.

Please consider the following points when interpreting outcomes from this data.

  • There are seven worksheets in this file. The ‘FIRE0102’ worksheet shows the number of incidents attended by type of incident and fire and rescue authority (FRA) for each financial year. The ‘FIRE0102 Quarterly’ worksheet shows the number of incidents attended by type of incident and FRA for each quarter. The worksheet ‘Data’ provides the raw data for the two main data tables (from 2009/10). The ‘Incidents chart – front page’, ‘Chart 1’ and ‘Chart 2’ worksheets provide the data for the corresponding charts in the statistical commentary. The ‘FRS geographical categories’ worksheet shows how FRAs are categorised.
  • Fire data, covering all incidents that FRSs attend, is collected by the Incident Recording System (IRS). For several reasons some records take longer than others for FRSs to upload to the IRS. Totals are constantly being amended (by relatively small numbers).
  • We took data for Service in numbers from the February 2019 incident publication. So figures may not directly match more recent publications due to data updates.
  • Before 2017/18, Hampshire FRS did not record medical co-responding incidents in the IRS. It is currently undertaking a project to upload this data for 2017/18 and 2018/19. This was not completed in time for publication on 14 February 2019. 

Home fire safety checks per 1,000 population

We took this data from the Home Office fire statistics, ‘Home Fire Safety Checks carried out by fire and rescue services and partners, by fire and rescue authority’ (Spreadsheet) for the period from 1 April 2017 to 31 March 2018.

Each FRS’s figure is based on the number of checks it carried out and doesn’t include checks carried out by partners.

Please consider the following points when interpreting outcomes from this data.

  • Dorset FRS and Wiltshire FRS merged to form Dorset & Wiltshire FRS on 1 April 2016. All data for Dorset and Wiltshire before 1 April 2016 is excluded from this report.
  • Figures for ‘Fire Risk Checks carried out by Elderly (65+)’, ‘Fire Risk Checks carried out by Disabled’ and ‘Number of Fire Risk Checks carried out by Partners’ don’t include imputed figures because a lot of FRAs can’t supply these figures.
  • The checks included in a home fire safety check can vary between services. You should consider this when making direct comparisons between services.

Home fire safety checks may also be referred to as home fire risk checks or safe and well visits by FRSs.

Fire safety audits per 100 known premises

Fire protection refers to FRSs’ statutory role in ensuring public safety in the wider built environment. It involves auditing and, where necessary, enforcing regulatory compliance, primarily but not exclusively in respect of the provisions of the Regulatory Reform (Fire Safety) Order 2005 (FSO). The number of safety audits in Service in numbers refers to the number of audits FRSs carried out in known premises. According to the Home Office definition, “premises known to FRAs are the FRA’s knowledge, as far as possible, of all relevant premises; for the enforcing authority to establish a risk profile for premises in its area. These refer to all premises except single private dwellings”.

We took this from the Home Office fire statistics, ‘Fire safety audits carried out by fire and rescue services, by fire and rescue authority’ (Spreadsheet) for the period from 1 April 2017 to 31 March 2018.

Please consider the following points when interpreting outcomes from this data.

  • Berkshire FRS didn’t provide figures for premises known between 2014/15 and 2017/18.
  • Dorset FRS and Wiltshire FRS merged to form Dorset & Wiltshire FRS on 1 April 2016. All data for Dorset and Wiltshire before 1 April 2016 is excluded from this report.
  • Several FRAs report ‘Premises known to FRAs’ as estimates based on historical data. 

Firefighter cost per person per year

We took the data used to calculate firefighter cost per person per year from the annual financial data returns that individual FRSs complete and submit to CIPFA, and ONS mid-2017 population estimates (Spreadsheet).

You should consider this data alongside the proportion of firefighters who are wholetime and on-call / retained.

Number of firefighters per 1,000 population, five-year change in workforce and percentage of wholetime firefighters

We took this data from the Home Office fire statistics, ‘Total staff numbers (full-time equivalent) by role and by fire and rescue authority’ (Spreadsheet) as at 31 March 2018.

Table 1102a: Total staff numbers (FTE) by role and fire authority – Wholetime Firefighters and table 1102b: Total staff numbers (FTE) by role and fire authority – Retained Duty System are used to produce the total number of firefighters.

Please consider the following points when interpreting outcomes from this data.

  • We calculate these figures using full-time equivalent (FTE) numbers. FTE is a metric that describes a workload unit. One FTE is equivalent to one full-time worker. But one FTE may also be made up of two or more part-time workers whose calculated hours equal that of a full-time worker. This differs from headcount, which is the actual number of the working population regardless if employees work full or part-time.
  • Some totals may not aggregate due to rounding.
  • Dorset FRS and Wiltshire FRS merged to form Dorset & Wiltshire FRS on 1 April 2016. All data for Dorset and Wiltshire before 1 April 2016 is excluded from this report.

Percentage of female firefighters and black, Asian and minority ethnic (BAME) firefighters

We took this data from the Home Office fire statistics, ‘Staff headcount by gender, fire and rescue authority and role’ (Spreadsheet) and ‘Staff headcount by ethnicity, fire and rescue authority and role’ (Spreadsheet) as at 31 March 2018.

Please consider the following points when interpreting outcomes from this data.

  • We calculate BAME residential population data from ONS 2011 census data.
  • We calculate female residential population data from ONS mid-2017 population estimates.
  • The percentage of BAME firefighters does not include those who opted not to disclose their ethnic origin. There are large variations between services in the number of firefighters who did not state their ethnic origin.
  • Dorset FRS and Wiltshire FRS merged to form Dorset & Wiltshire FRS on 1 April 2016. All data for Dorset and Wiltshire before 1 April 2016 is excluded from this report.