Kent Fire and Rescue Service
Customer Insight Research Phase 1 2022 - Results and Learnings

7.1. Appendix A – detailed Respondent Demographics

To assess the sample representativeness the KFRS customer insight demographic results were compared with secondary data sources.

District of residence

To assess eligibility, participants across both online and paper surveys (N=688) were asked in which area of Kent or Medway they lived. Whilst it was confirmed with participants before beginning the in-person paper survey that they lived in Kent or Medway, a total of twenty-two respondents opted out of completing the separate paper demographics questionnaire and thus the specific county or district they live is unknown. These responses have been treated as missing data in the below calculations.

Figure 1 shows the proportion of customer insight respondents across the Kent districts and Medway, in comparison to proportions calculated by KFRS from the Census 2021 data based on a population of 18 years and over. The ONS data has been used as a benchmark for comparisons to understand if representation has been achieved in the KFRS customer insight sample. The representation of respondents in the KFRS customer insight survey from Kent (81%) and Medway (16%) was fairly similar to the proportions calculated from the ONS (2022a); see Figure 1. However, there were some disparities when comparing at district level. Exact representation was achieved for Folkestone and Hythe, Sevenoaks, and Tonbridge and Malling districts; but for other districts, representation differed in relation to the ONS proportions. Respondents from Thanet were overrepresented in the sample as were respondents from Maidstone. Dartford, Gravesham and Tunbridge Wells had the lowest representation of respondents from these areas. Thus, greater effort will be needed to reach customers residing in these areas to achieve greater representation in Phase 2 of the customer insight survey. One method to help achieve this would be to collaborate more closely with fire station personnel in these areas, for help in disseminating the survey. Overrepresentation in Thanet may be due to the close liaison between the researchers and Collaboration teams where focused efforts were being made at the time of data collection.

* Census 2021 district proportions calculated by KFRS based on census 2021 singular age Lower Tier Local Authority data set (ONS, 2022a).

Age

Across online and paper surveys, the representation of the different age groups within the sample differed from the benchmark proportions worked out from the Census 2021 data (ONS, 2022b) across all age categories; see Figure 2. There was overrepresentation of the 25 – 34 years and 35 – 44 years age groups in the KFRS sample, which may be due to reaching out to parents at fire station open days. Representation of the age groups from 45 years to 74 years were the most similar to the ONS data proportions. The largest discrepancies were in the youngest age group of 18 – 24 and the oldest age group of 75 years and over. A previous evidence review conducted by KFRS into accidental dwelling fires identified young and old age being associated with fire fatality more greatly in an ADF (KFRS, 2021). This highlights the importance to having representation of younger and older age brackets in the customer insight sample. According to Census 2021 data calculations, the districts of Maidstone and Folkestone and Hythe have the largest proportions of residents aged 75 years and over. Moreover, census 2021 data revealed that in Kent; Canterbury and Maidstone have the highest proportion of residents aged between 18 – 24 years. Therefore, these areas could be tapped into to reach the underrepresented age groups in Phase 2.

*Census 2021 age group proportions calculated by KFRS based on census 2021 singular age data set for Upper Tier Local Authorities (ONS, 2022b).

Sex, sexual orientation and gender

Overall, female representation in the customer insight questionnaire was significantly higher (61%) in relation to the proportion of customer insight respondents who were male (35%) across both online and paper responses. Four percent of respondents either did not answer the question or preferred not to answer. The difference between the proportions of male and female respondents was greater for the paper responses, in comparison to the online responses. According to calculations by KFRS using Census 2021 data (ONS, 2022c); the proportion of females was 52% and males was 48% in the population of Kent and Medway aged 18 years and over. Therefore, it can be concluded that females were overrepresented in the customer insight sample. Greater targeting of males will be needed in the Phase 2 of the research.

Respondents were asked the voluntary question of how they would best describe their sexual orientation. Similarly, to the previous demographic variables, benchmark comparisons were set based on proportions reported by the Annual Population Survey (APS) 2020 (ONS, 2022d). The APS data was used as it was the most recent data set that could be found that reported on sexual orientation. However, it should be noted that the APS values are based at the South East level and include anyone aged 16 years and over. Thus, caution should be taken when interpreting comparisons with the KFRS survey proportions which are at a Kent and Medway level and include anyone aged 18 years and over.

Figure 3 shows that most respondents (88.8%) reported being Straight/Heterosexual; a slightly lower proportion in comparison to the APS value (93.6%). Representation of respondents who identified being gay, lesbian or bisexual was achieved. However, the proportion of respondents who preferred not to give their sexual orientation was higher in the KFRS survey, in comparison to those who refused or stated not knowing their sexual orientation in the APS. Respondents who reported an ‘other’ sexual orientation such as ‘Asexual’ were lower in the KFRS survey as compared to the APS.

Thirty-four per cent respondents across online and paper surveys, identified as a man (including Trans-man). A higher proportion (60%) identified as a woman (including Trans-woman). Six percent of respondents either did not answer the question or preferred not to say. Upon comparing gender identity with sex, it was observed that a very small minority of four participants potentially identified as transgender, as their sex at birth differed to the gender they identified with.

Ethnicity

According to the Census 2021 results on ethnicity; ethnic diversity has increased in England and Wales since the 2011 census (ONS, 2022e). Whilst the majority of the population still classified themselves as being from a ‘White’ ethnic background, those who identified as Asian, Asian British or Asian Welsh saw the highest percentage point increase since 2011 (ONS, 2022e).

Representation of minority ethnic groups in the KFRS customer insight survey sample was low in comparison to Census 2021 Kent and Medway ethnicity proportions (ONs, 2022f), as calculated by KFRS (See Table 1). It should be noted that caution should be taken when comparing KFRS customer insight to Census 2021 ethnicity proportions. The Census 2021 ethnicity results are inclusive of all ages, whilst KFRS customer insight survey population is those aged 18 years and over. Also, KFRS customer insight ethnicity proportions have been calculated to include the percentage of those respondents who preferred not to answer the question on ethnicity or who did not provide an answer at all, whereas the Census 2021 proportions are calculated only on the population that answered the question. Five per cent of respondents either did not provide an answer or preferred not to say their ethnicity in the KFRS customer insight survey.

Nonetheless, it can be said that those who identified as having a ‘White: English, Welsh, Scottish, Northern Irish or British’ ethnic background was overrepresented in the KFRS sample (88.4%). Except for ‘White: Irish’; all other White ethnic backgrounds were underrepresented in the sample. In accordance with the Census 2021 ethnicity proportions in the population, all Asian, Black and Mixed or Multiple ethnicities were underrepresented in the KFRS customer insight sample. There was no representation of the following ethnic categories: ‘Mixed/Multiple Ethnic Groups - White and Black African’, ‘Asian/Asian British – Pakistani’, ‘Asian/Asian British –  Bangladeshi’, ‘Black/African/Caribbean/Black British – Caribbean’, ‘Other Ethnic Group – Arab’, ‘Black/African/Caribbean/Black British: Other’.

Additional efforts will need to be made in the Phase 2 customer insight survey to ensure more responses are obtained from ethnic minority groups to ensure sample representativeness.

*Kent and Medway combined percentage calculated by KFRS based upon census 2021 Upper Tier Local Authority ethnicity data set (ONS,2022f).

Religion

The representation of different religions within the customer insight sample differed across all categories in comparison to proportions calculated by KFRS on Census 2021 data when looking at the Kent and Medway population (ONS, 2022g). Caution should be taken when making comparisons between the data sets as the Census 2021 calculations are based on the Kent and Medway population of all ages, rather than the population of those aged 18 years and over. Despite this, it appears that all religions (except for the Jewish religion) were underrepresented in the KFRS customer insight sample (Table 2). There was no representation of the Sikh religion. According to Census 2021 results of religion in England and Wales, the proportion of the population who reported having ‘No religion’ increased from the 2011 census (ONS, 2022h). Most respondents (48.0%) reported having ‘No religion’ in the KFRS customer insight survey, however this group was still overrepresented in the sample in comparison to the Census 2021 Kent and Medway proportion (41.2%).

Additionally, the Census 2021 results found that the percentage of the population who reported themselves as a ‘Christian’ decreased by 13.1 percentage points from the Census 2011 results in England and Wales. Similarly to the census 2021 results, less than half of the KFRS customer insight survey reported being ‘Christian’, however this group is still considered underrepresented in comparison to the census 2021 Kent and Medway Christian population.

Other religions reported in the customer insight survey included, spiritualism (N=4); Catholic (N=2); Jedi (N=1); Humanist (N=1) and Jehovah’s Witness (N=1).

Additional effort should be made to collaborate with different religious groups during Phase 2 of the customer insight survey, to increase representation.

*Kent and Medway combined percentage calculated by KFRS based upon census 2021 count data.

Online respondents (N=517) were additionally asked what their main language was to speak and write in at home. Paper respondents did not receive these questions. Nearly all respondents who answered the questionnaire online reported English as their main language for speaking (98%) and writing in at home (98%). Other languages respondents mainly used to speak in at home included Polish, Taglog Slovakian (N = 4.1%). The majority of whom reported speaking in English ‘Often’ or ‘Sometimes’. Other languages respondents reported writing in at home included Scottish (0.4%; N=2). Most respondents reported ‘Always’ writing in English at home (96%), but others reported writing in English ‘Often’ (1.5%); ‘Sometimes’ (0.6%) or ‘Rarely’ (0.2%).

Long term health conditions and illnesses

Most respondents in both the online and paper survey did not have a long-term health condition or illness (Figure 4). Physical or mobility related health conditions were most common amongst respondents. Mental health conditions were less common across online and paper responses at 11% and 6% respectively. Few respondents reported having a different health condition or illness. Four per cent of online respondents had 2 or more long term health conditions/illnesses; in comparison to only 1% of paper respondents.

Those who reported having a long-term health condition or illness in the online questionnaire (N=159), were asked the follow up question of whether their illness or condition reduced their ability to carry out day to day activities. Most responded with ‘Yes a little’ (44%) or ‘Not at all’ (37%). A minority responded ‘Yes, a lot’ (16%).

Those who reported having a long-term condition or illness in the online survey (N=159), were asked whether they had a speech, hearing or sight impairment; the majority of whom did not (75%). A minority of respondents (1%) reported having a speech impairment. Hearing and sight impairments were prominent amongst 16% and 11% of respondents respectively. When asked, the majority of those who had a speech or hearing impairment had not previously heard of Relay UK; a service which people with speech or hearing impairments can use to make contact with KFRS during an emergency. Therefore, more awareness may be needed around Relay UK to ensure people with hearing or speech impairments do not experience barriers in making urgent contact with KFRS.

Household composition and income

Respondents in both the online and paper survey were asked how many people were there in their household, including themselves. Additionally, respondents were asked to specify how many people there were in the following age bands: adults aged 18-64; adults aged 65 years and over; and children under 18 years. Figure 5 below displays different household compositions amongst both online and paper respondents.

When considering the combined percentage across online and paper surveys, the most common household composition were households with children aged 18 years and under (42%). Figure 5 shows that most respondents from households with children completed the paper survey at events (74%), which is to be expected due to most of the events being family open days, involving activities for children. Mixed  aged, adult only households were the most common amongst respondents in the online survey (36%). All other household compositions had a lower representation, with the lowest being single person households of someone aged 65 years and over across paper and online responses. This highlights the importance of using different channels for survey collection to target different demographics.

Additional information was asked from respondents in the online survey, including average annual household income and accommodation type of their permanent address. The majority of respondents reported an annual household income in the lower three income bands (57%); see Figure 6. A minority reported an annual household income in the top three income bands (22%). A further 21% of respondents either preferred not to answer the question or did not provide an answer at all.

Most online respondents reported living in a ‘Whole House or Bungalow’ (82%; N=422). Of which most either lived in a detached (34%), or semi-detached house/bungalow (37%); with slightly less living in a terraced accommodation (28%). A minority of respondents reported living in a flat, maisonette, or apartment (12%); although those that did mostly reported living in a purpose-built block of flats or house divided into separate flats (77%). A lower proportion reported living in part of a converted or shared house including bedsits (16%); or in a commercial building, such as above a shop (8%). No respondent reported living in a Caravan or other mobile temporary structure. Therefore, the representation of this group could be improved.

Respondents across online and paper surveys were also asked for the first half of their post code. Of the 688 participants, 8.9% of responses were treated as missing data, either due to them not answering the question; the format of the first half of the post code was not met, or the post code provided did not match the district they gave in a previous question. There was at least one respondent across all 75 post codes in Kent and Medway. Figure 7 below displays the proportion of respondents across BR, CT, DA, ME and TN post codes. Post codes beginning with CT and ME were the highest represented post codes in the sample; whereas BR and DA post codes had the smallest representation. However, this may be expected as the number of post codes starting with BR and DA were smaller in comparison to all others.