See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/328803658
A comparison between thermostat and thermostatic radiator valve setpoint
temperatures in UK social housing
Conference Paper · September 2018
CITATION READS
1 184
3 authors, including:
Adorkor Bruce Rory Jones
University of Plymouth University of Reading
8 PUBLICATIONS   79 CITATIONS    48 PUBLICATIONS   1,340 CITATIONS   
SEE PROFILE SEE PROFILE
All content following this page was uploaded by Rory Jones on 08 November 2018.
The user has requested enhancement of the downloaded file.
A comparison between thermostat and thermostatic radiator valve setpoint 
temperatures in UK social housing 
Adorkor Bruce-Konuah, Rory V. Jones, Alba Fuertes 
Department of Architecture and Built Environment, University of Plymouth, Drake Circus, Plymouth, 
Devon, PL4 8AA, UK 
Keywords: Space heating behaviour, thermostat, thermostatic radiator valves, heating setpoint temperatures 
Abstract 
In the UK, in centrally heated dwellings, space heating is commonly controlled by a whole 
house thermostat as well as thermostatic radiator valves (TRVs) fitted on individual radiators. 
TRV settings define a setpoint temperature at which the radiator is switched off, in order to 
regulate zonal temperatures. This paper presents an analysis of the TRV setpoint 
temperatures which occupants’ select in living rooms and main bedrooms and provides a 
comparison between these and the whole house thermostat setting. The work capitalises on 
primary data from a socio-technical household survey undertaken in a sample of social 
housing in Plymouth, UK during 2015. The mean reported TRV setpoint temperature in the 
living rooms (n = 144) and bedrooms (n = 120) were 23.4°C and 22.1°C respectively. This result 
confirms that occupants prefer cooler conditions in their bedrooms and also suggests that 
occupants are actively using their TRVs to zonally control their heating at home to maintain 
comfortable thermal conditions and reduce their heating energy demand. The results also 
indicate that occupants’ thermostat and TRV setpoint temperatures vary according to their 
household and motivation, behaviour and perception characteristics. The mean reported 
thermostat setpoint temperature was 20.7°C for those who reported a living room TRV setting 
and 20.9°C for those who reported a bedroom TRV setting. This result suggests that there 
may be a misunderstanding of the purposes of the whole house thermostat and the individual 
TRVs within a central heating system. Variations in occupant heating control behaviour have 
an impact on occupant comfort and household energy use. The results of this study have 
significant implications for the planning and implementation of energy efficiency measures, 
behaviour change interventions as well as the design of heating controls. 
INTRODUCTION 
Energy use in domestic buildings accounts for 29% of total UK energy consumption and 
around two thirds is used for space heating (Department for Energy and Climate Change 
(DECC), 2013). Domestic space heating accounts for 11% of the nation’s greenhouse gas 
emissions (Department for Energy and Climate Change (DECC), 2012). Reducing heating 
energy use in homes is therefore essential if the UK is to achieve its commitment to reduce 
national carbon emissions by 80% of 1990 levels by 2050 (HM Government, 2008). The three 
key avenues that are being explored in order to achieve this target are (1) the refurbishment 
or replacement of the existing housing stock (Hamilton et al., 2016), (2) decarbonisation of 
domestic heating supply (Energy Technology Institute, 2015) and (3) social interventions, i.e. 
occupant behaviour change, to encourage more efficient use of energy (Lopes, Antunes and 
Martins, 2012). 
In the UK, over 75% of the current UK building stock will still be in use in 2050 and the stock 
is only being expanded at a rate of 1 – 2% per year (Ravetz, 2008). The main issues with the 
existing housing stock are poorly performing solid walls, single glazed windows and 
uninsulated roofs and floors and are responsible for a significant amount of wasted heat 
(Loveday and Vadodaria, 2013). In response to this and in line with the commitment to meet 
carbon reduction targets, the UK social housing sector in recent years has embarked on a 
large scale programme of thermal upgrades, as well as the installation of more efficient 
heating systems and controls. Regarding social interventions, it has been demonstrated that 
there is a considerable variation in energy consumption between “identical” dwellings and 
this is due to how the dwellings are used (Andersen, 2012). Occupant behaviour has been 
very well noted to significantly affect a building’s energy consumption (Hoes et al., 2009; 
Yoshino, Hong and Nord, 2017). 
Space heating is an important aspect of household energy consumption and occupant 
comfort. Central heating, which allows households to simultaneously heat all the spaces in 
their dwelling is now found in over 90% of UK homes (Department for Energy and Climate 
Change (DECC), 2013). The predominant fuel is gas, which is more efficient than solid fuels, 
and has resulted in greater carbon efficiency for heating. A basic central heating system 
consists of a central boiler, a pump and individual radiators located in multiple spaces 
throughout the dwelling. Most central heating systems will also have some level of controls – 
a full set of central heating controls consist of a central timer, a whole house thermostat and 
thermostatic radiator valves (TRVs). Since 2010, two zone heating has been mandatory for all 
new dwellings which are not open plan (HM Government, 2010), however this is not 
obligatory in existing dwellings. A central heating system that complies with Building 
Regulation Part L1B will have the full set of controls. Even with the widespread ownership of 
central heating systems in UK homes, it is reported that about 70% of the housing stock do 
not have the full set of heating controls specified in the building regulations and 4% do not 
have any controls at all (Heating and Hot Water Task Force, 2010). A dwelling with no 
thermostat may result in excessive room temperatures and with no TRVs a lack of zonal 
temperature control. Where heating controls are available, they will have a significant 
influence on a dwelling’s space heating energy demand (Shipworth et al., 2010; Fabi, 
Andersen and Corgnati, 2013; Huebner et al., 2013; Beizaee et al., 2015; Jones et al., 2016; 
Cockroft et al., 2017). 
Multiple factors have been found to influence space heating preferences (setpoint 
temperature and heating duration) and a detailed international review and discussion of 
these factors have been presented by Wei et al. (2014). Amongst these factors is the type of 
heating controls installed in the dwelling (Guerra Santin, Itard and Visscher, 2009; Guerra-
Santin and Itard, 2010; Shipworth et al., 2010; Consumer Focus, 2012). Heating controls such 
as TRVs have the potential to reduce space heating energy use, as heating demand 
temperatures in less frequently or unoccupied rooms or rooms requiring cooler temperatures 
can be reduced or turned off completely. However, it has been noted that simply providing 
central heating controls does not necessarily result in dwellings being heated in ways that 
reduce energy consumption and carbon emissions (Shipworth et al., 2010). To support 
decisions to help reduce space heating energy demand in social housing, it is important to 
understand how social housing tenants use their available heating controls. There is currently 
a lack of empirical data underpinning the recommendations for space heating energy 
reduction policies. 
This paper presents an analysis of social housing tenant’s choice of TRV setpoint temperatures 
in living rooms and main bedrooms and provides a comparison between these and the chosen 
whole house thermostat setting. The work capitalises on primary data from a socio-technical 
household survey undertaken in Plymouth, UK during 2015. 
The role of TRVs for energy demand reduction 
A TRV controls a single radiator and it is used to keep a room at a different temperature to 
the rest of the dwelling. It offers a cheap and easy way of providing zoned temperature 
control. They usually have a dial marked with a * and numbers from 0 to 5 or 0 to 6. The * 
represents a minimum temperature which is usually 6.9°C for frost protection and the 
number settings correspond to setpoint temperatures from 0 to 28°C. Where the TRVs have 
settings up to 6, the maximum temperature for each setting is lower compared to TRVs with 
settings up to 5 only. When the central heating boiler is in operation, TRVs sense the air 
temperature and regulate the flow of hot water to the radiator, allowing for zonal 
temperature control. They do not control the boiler operation and will only control zonal 
temperature if a lower temperature setting, compared to the whole house thermostat setting 
is selected. Allowing rooms that are not often or are unoccupied to be heated to cooler 
temperatures or not heated at all reduces the difference between internal and external 
temperatures, thus reducing the rate of heat loss and heating energy demand. 
The energy saving potential of zonal temperature control has been demonstrated in several 
previous studies (Meyers et al., 2010; Beizaee et al., 2015; Cockroft et al., 2017). In the US, 
Meyers et al. (2010) showed that 6.2% of total primary energy is wasted from heating or 
cooling living rooms during the night and 9.7% is wasted from heating or cooling bedrooms 
during the day when the spaces are unoccupied. In a modelling study of a pair of identical 
1930s dwellings, one equipped with simple TRVs and the other equipped with programmable 
TRVs, Beizaee et al. (2015) showed that the dwelling with programmable TRVs used 11.8% 
less gas compared to the dwelling with the simple TRVs. With the programmable TRVs, the 
rooms were heated only when occupied and with the simple TRVs, the rooms were heated 
whenever the boiler was on. Furthermore, a 0.6°C reduction in mean indoor temperature was 
observed when programmable TRVs were used. In another modelling study, Cockroft et al. 
(2017) investigated the potential energy savings between non-zoned (heating controlled by 
whole house thermostat only) and zoned conditions (heating controlled by thermostat and 
programmable TRVs). The study demonstrated that significant energy savings in the order of 
8% to 37% can be achieved by adopting a multi-zonal control strategy where both time and 
temperature in individual rooms are controlled. 
This paper aims to provide an insight in to the use of whole house central thermostats and 
TRVs in UK social housing. The paper responds to a gap identified in the literature, the lack of 
evidence as to how occupants are using TRVs to regulate their thermal comfort as well as 
their heating energy demand. The analysis could enable social housing providers and the 
government to target energy efficiency measures, particularly social interventions (i.e. 
behaviour change) at those dwellings and households where their impact may be most 
beneficial, as well as to inform the design of future domestic heating controls. 
METHODS 
The data analysed in this paper are derived from a socio-technical household survey 
undertaken as part of the European Horizon 2020 research project: Energy Game for 
Awareness of energy efficiency in social housing communities (EnerGAware) which was 
conducted in the city of Plymouth, UK (EnerGAware, 2016). The social housing investigated in 
this study are managed by the housing association DCH (formerly Devon and Cornwall 
Housing) who are also a partner of the project. Plymouth was the case study city chosen as 
social housing accounts for 20.1% of the city’s housing stock, one of the largest proportions 
in the UK (Office of National Statistics, 2012). A detailed description of the socio-technical 
household survey is provided in Jones et al. (2016). In summary, the socio-technical survey 
was administered to 2,772 social houses (social rented and shared ownership) in Plymouth. 
The survey was occupant self-reported through either a paper-based postal survey or an 
online survey administered through the online survey software, SurveyMonkey and was 
conducted in May 2015. The survey contained 68 standardised closed questions. Out of all 
the surveys administered, 537 responses were received, giving an overall response rate of 
19.4%. The socio-technical survey provided occupant reported winter living room and 
bedroom TRV settings and the whole house thermostat setpoint temperature as well as 
household characteristics (e.g. household size and composition, health of HRP1 and 
households with disabled members) and motivation, behaviour and perception 
characteristics (e.g. affordability of energy bills, worry about energy bills, understanding and 
perceived control of energy use at home, perceived ability to save energy at home, heating 
related behaviours and dwelling occupancy pattern). 
The occupant reported TRV settings were converted to their corresponding setpoint 
temperatures by referring to the manufacturer’s specifications. The survey responses along 
with the whole house thermostat setpoint temperatures and the converted TRV setpoint 
temperatures were input, cleaned and organised in an IBM SPSS Statistics 24 database for 
analysis. 
RESULTS 
Of the 537 households responding to the survey, 29 reported bedroom TRV settings of 0 and 
one reported a * setting. These were excluded from the analysis as they indicate that the 
radiators were not in use. 144 provided a living room TRV setting as well as a thermostat 
setting and 120 provided a bedroom TRV setting as well as a thermostat setting. 
Table 1 shows the overall mean living room and bedroom TRV and thermostat setpoint 
temperatures. The upper and lower 95% confidence intervals (95% CIs) for the data are 
presented to demonstrate the distributions of setpoint temperatures reported, as well as the 
                                                          
1 The Household Representative Person (HRP) is the individual that is taken to represent that household. In this 
study it describes the person that completed the survey. 
extreme values reported in the coldest and warmest homes. The standard deviations (SD) are 
presented to demonstrate how much the reported setpoint temperatures differ from the 
mean value. 
The mean reported TRV setpoint temperature was 23.4°C in the living room and 22.1°C in the 
bedroom and the difference in these mean temperatures was significant (p < 0.01). This 
implies that there is a preference for cooler conditions in bedrooms and shows that social 
housing tenants use their TRVs to zonally control temperatures in different rooms. The 30 
households who reported turning their bedroom TRVs off (0 or * setting) further supports this 
finding. The mean whole house thermostat setting was 20.7°C from those who reported a 
living room TRV setting and 20.9°C from those who reported a bedroom TRV setting. These 
thermostat setpoint temperatures are consistent with the 21°C recommended by the World 
Health Organization (WHO) as a comfortable indoor temperature, and to prevent potential 
health effects (World Health Organization, 1987). 
Table 1 Reported mean TRV and whole house thermostat setpoint temperatures 
  Whole house thermostat TRV setpoint temperatures 
setpoint temperatures (°C) (°C) 
 n Mean (95% CI) SD Mean (95% CI) SD 
Living room 144 20.7 (20.2, 21.2) 2.8 23.4 (22.8, 24.0) 3.6 
Bedroom 120 20.9 (20.4, 21.4) 2.7 22.1 (21.1, 22.7) 4.4 
In both the living rooms and bedrooms, the mean TRV setpoint temperatures were higher 
than the mean thermostat setpoint temperatures. From the 144 households that reported 
living room TRV settings, 94 (65%) had a TRV setting higher than their thermostat setting, 16 
(11%) households had the same setting for their TRV and thermostat and 34 (24%) households 
had TRV setpoints lower than their thermostat. Where the TRV setpoint was higher than the 
thermostat, the TRVs were on average set to 5°C higher than the thermostat setting, with the 
average TRV set to 25°C and thermostat to 20°C. In the households where the TRV setpoint 
was the same as the thermostat setpoint, the setpoint temperature was set to 20°C. In cases 
where the TRV setting was lower than the thermostat, the average difference was 3°C, the 
average TRV setpoint was 20°C and thermostat was 23°C. From the 120 households that 
provided bedroom TRV settings, 56 (47%) had higher TRV setpoint temperatures, 16 (13%) 
had the same and 48 (40%) had TRVs set lower. Where the thermostat setting was higher 
than the TRV setting, the average setpoint temperatures were 20°C and 26°C respectively. In 
households where both settings were identical, the setpoint temperatures were 21°C and in 
households with lower bedroom TRV settings, there was an average 4°C difference, with the 
average thermostat setpoint temperature set to 22°C and the bedroom TRV set to 18°C. 
Table 1 in the Appendix presents the variations in reported mean thermostat setpoint 
temperature and living room and bedroom TRV setpoint temperatures in relation to 
household and motivation, behaviour and perception characteristics. In most of the groups, 
the TRV setpoint temperatures were higher than the thermostat setpoint temperatures. 
However, the differences between the mean thermostat setpoint temperatures and the 
mean TRV setpoint temperatures were not significant. 
In relation to household characteristics, the thermostat setpoints were always lower than the 
living room and bedroom TRV setpoints regardless of the education level of the HRP, the 
presence of disabled members in the household, whether households were in receipt of 
welfare benefits or their satisfaction with life in general. Regarding motivation, behaviour and 
perception characteristics, again thermostat setpoints were always lower than living room 
and bedroom TRV setpoints regardless of occupants’ perception of their affordability of 
energy bills, their understanding of how their home uses energy, and their heating behaviour 
(i.e. their use of doors and thermostats during the winter). 
There were some instances where the TRV setpoint temperatures were lower than the 
thermostat setpoint temperatures. In households where the HRP was unemployed, the living 
room (21.6°C) and bedroom (21.0°C) TRV setpoints were lower than the thermostat settings 
(23.0°C). Where the HRP reported bad health and visiting the GP 7-12 times per year, the 
bedroom TRV setpoints were lower (Health: 21.3°C; GP visits: 20.0°C) than the thermostat 
setpoints (Health: 21.8°C; GP visits: 21.2°C). The analysis showed that households that 
indicated that they do not worry about their energy bills and they do not think about how 
they can save energy had lower bedroom TRV setpoint temperatures than their thermostat. 
Households that strongly agreed to having control over how much energy they used and those 
who disagreed to not being able to save any more energy also had lower TRV setpoints in the 
bedroom. In relation to not being able to save any more energy, households that tended to 
disagree also had lower TRV setpoint temperatures in the living room. Regarding heating 
behaviours, households that indicated that they sometimes close windows when the heating 
is on and they very occasionally turn the heating off when no one was at home, set lower TRV 
setpoint temperatures in the living room and bedrooms than on their whole house 
thermostat. 
DISCUSSION 
The findings reported in this paper show that social housing tenants’ space heating behaviour 
(i.e. use of TRVs) varies according to the room within their dwelling. The current study 
indicates that bedrooms are generally cooler than living rooms and not all bedrooms are 
heated; 29 respondents reported a 0 TRV setting and 1 bedroom was only heated when the 
bedroom temperature falls below 6.9°C (frost protection setting). By comparison, all the 
respondents who provided a living room TRV setting indicated that their living rooms were 
heated. 
In general, the mean TRV setpoint temperatures reported in this paper (living room - 23.4°C, 
bedroom – 22.1°C) are higher, than what is assumed in BREDEM-based models where the 
temperature in the living area is set at 21°C and in the rest of the dwelling (including 
bedrooms) is set to 18°C (Anderson et al., 2002). The results are however consist with the 
BREDEM assumption that living room temperatures are higher than that in bedrooms. Also, 
comparing the current results from the whole house thermostat (20.7°C and 20.9°C) with the 
21°C used by BREDEM, suggests that the value is appropriate for living areas. 
Regarding the mean thermostat setpoint temperature selected by these social housing 
tenants, the results obtained (20.7°C and 20.9°C) are similar to the 21.0°C recommended by 
the WHO as a comfortable indoor temperature and to prevent potential negative health 
effects (World Health Organization, 1987). It is also similar to the whole house demand 
temperatures reported by Huebner et al. (2013) (20.6°C), Shipworth et al. (2010) (21.1°C) and 
Kane et al. (2015) (20.9°C). Overall, the high level of agreement between the findings of the 
different studies is noteworthy given the different methods (temperature monitoring and 
self-reported) and different samples (owner-occupied, privately rented and social rented). 
The thermostat demand temperatures are within 0.5°C. 
The work presented here shows that there is a variation in how occupants use heating 
controls in their homes. The whole house thermostat controls the overall heating system and 
the TRVs control temperatures in individual rooms. In rooms where the TRV setpoint 
temperature is higher than the thermostat setting, the TRV setpoint temperature becomes 
redundant as it will not be reached before the heating is turned off by the thermostat 
setpoint. Using TRVs to set cooler thermal conditions in different rooms has the potential to 
reduce space heating energy demand. From the sample presented in this paper, 65% of the 
households reported higher living room TRV settings than their thermostat setting. This 
observation was evident regardless of most household and motivation, behaviour and 
perception characteristics. This finding suggests that: (1) occupants may prefer warmer 
conditions than what the overriding thermostat permits, (2) perceived thermal comfort may 
be more important to occupants than actual thermal conditions, and (3) residents may not 
understand the role of TRVs as part of the home heating system and thus their energy saving 
potential. 
Regarding household characteristics, households with unemployed members and households 
with couples with dependent children had lower TRV settings compared to the whole house 
thermostat settings. This was seen in both the living rooms and the bedrooms. In these 
households, members may be trying to save money by adjusting their TRVs. In the households 
with unemployed members, this finding also gives an indication of a possible impact of 
household income on space heating preferences. The lowest thermostat settings were in 
homes where the HRPs considered their general health in the last 12 months as very bad. Bad 
health may limit their potential to work, hence reducing their household income. 
Alternatively, low thermostat setting indicates cooler thermal conditions in the homes, which 
may contribute to the bad health of the HRP. The effect of household income on TRV setpoint 
temperatures was not directly investigated, as the survey did not ask respondents to report 
their annual household income. Previous studies have identified significant effects of income 
on space heating behaviour (Hunt and Gidman, 1982; Sardianou, 2008) The question of 
household income was considered too sensitive as the study focussed on social housing 
residents who typically are a low-income group. 
Regarding motivation characteristics, only occupants who reported that they strongly 
disagree to often thinking about how their home uses energy and those who reported that 
they tend to disagree with not being able to save anymore energy had lower TRV settings 
compared to the thermostat setting in both the living rooms and the bedrooms. In the 
remainder of the characteristics, TRV settings were higher than thermostat settings in all the 
groups. These findings suggest that there may be a lack of understanding of the use of these 
heating controls, particularly as a potential to save energy and reduce energy bills. Heating is 
used to provide a comfortable thermal environment, hence achieving thermal comfort may 
be more important to householders than having lower temperatures or shorter heating 
periods in order to save energy. Although it has been shown that the use of TRVs as a heating 
control can decrease heating energy demand (Beizaee et al., 2015; Cockroft et al., 2017), it is 
also noted that the savings are not necessarily achieved unless the user has knowledge about 
the operation of the control mechanism (Shipworth et al., 2010). Perhaps, the householders 
are not aware of how this additional heating behaviour can help them to reduce their heating 
energy demand. 
The usability of heating controls also influences their use (Meier et al., 2010). Although TRVs 
are easy to use, the settings are displayed as numbers ranging from 0 to 5 or 6 with little 
indication of the corresponding temperatures, whereas the settings on the thermostat are 
shown in temperatures. If householders are not able to make the link between TRV setpoint 
temperatures and thermostat setpoint temperatures, the energy saving opportunities could 
be missed. 
The findings suggest that although social housing tenants are using TRVs to zonally control 
temperatures in their homes, they may not be aware of the energy saving potential of these 
heating controls and are currently missing out on reducing their heating energy demand and 
consequently their heating bills. 
The results obtained in this study provide a useful insight in to occupant heating preferences. 
However, there are limitations in the method which may have implications on the study 
findings. The results are based on relatively small sample sizes (living room TRV setpoint 
temperatures and whole house setpoint temperatures for 144 homes and bedroom TRV 
setpoint temperature and whole house setpoint temperatures for 120 homes) from a single 
UK city and therefore extrapolating the results to the wider population is not appropriate. A 
larger national-scale study of TRV and thermostat settings would be a valuable extension to 
the current work and could also be used to validate the findings of the current study. The 
reliability of self-reported data provided by survey participants is most often limitation. 
Without the presence of an interviewer, the respondent may not fully complete the 
questionnaire before submission (as seen in participants providing living room TRV setting 
and not bedroom TRV setting). Also, providing heating settings at one point in time is also a 
limitation as occupants may change the setpoint temperatures may change over time. To the 
author’s knowledge, Andersen’s et al. (2011) study in Denmark which developed a custom 
monitoring device, is the only study to provide direct measurement of TRV setpoint in homes. 
With the rapid development of ‘smart’ internet-connected thermostats and TRVs, which 
allow users to control their heating via a website or on their smart phones, data on space 
heating preferences will become increasing available for further research in this field. This 
could however take many years. 
CONCLUSION 
Based on self-reported thermostat and TRV settings, an analysis of mean setpoint 
temperatures in relation to household and motivation, behaviour and perception 
characteristics have been presented. The findings show that the mean whole house setpoint 
temperature is similar to the WHO recommended comfortable indoor temperature and is also 
in agreement with findings from previous studies. The findings regarding zonal temperatures, 
by the use of TRVs, showed that the social housing tenants in this study preferred different 
thermal conditions depending on the room, i.e. cooler conditions in the bedroom than in 
living rooms. This finding is in agreement with BREDEM-based models and with findings from 
previous studies and suggests that occupants are actively controlling their zonal temperatures 
to ensure their thermal comfort. However, the mean TRV setpoint temperatures obtained in 
this study were found to be higher than the assumed input values typically used for energy 
modelling. The mean TRV setpoint temperatures were also higher than the mean thermostat 
setpoint temperature, implying that the participants in this study are not using these controls 
to reduce their heating energy use. The findings from this study provide further insight into 
social housing tenants heating behaviours and have implications for housing providers, 
heating technology providers, the government and commercial organisations that implement 
energy efficiency measures. The study findings suggest that people may not understand how 
their heating system controls actually work and therefore interventions aimed at reducing 
heating energy use in homes as well as the design of heating controls, should first help 
occupants to understand and operate their heating controls efficiently. 
REFERENCES 
Andersen, R. (2012) ‘The influence of occupants ’ behaviour on energy consumption 
investigated in 290 identical dwellings and in 35 apartments’, in Healthy Buildings 2012. 
Brisbane, pp. 8–12. 
Andersen, R. V., Olesen, B. W. and Toftum, J. (2011) ‘Modelling occupants’ heating set-point 
preferences’, 12th Conference of International Building Performance Simulation Association, 
Sydney, 14-16 November, pp. 151–156. 
Anderson, B. R. et al. (2002) BREDEM-8 Model Description: 2001 Update. Building Research 
Establishment (BRE), Garston, and Department for Environment, Food and Rural Affairs 
(DEFRA), London. 
Beizaee, A. et al. (2015) ‘Measuring the potential of zonal space heating controls to reduce 
energy use in UK homes: The case of un-furbished 1930s dwellings’, Energy and Buildings, 
92, pp. 29–44. 
Cockroft, J. et al. (2017) ‘Potential energy savings achievable by zoned control of individual 
rooms in UK housing compared to standard central heating controls’, Energy and Buildings, 
136, pp. 1–11. 
Consumer Focus (2012) Consumers and domestic heating controls: a literature review. 
Available at: 
http://webarchive.nationalarchives.gov.uk/20130103084529/http://www.consumerfocus.o
rg.uk/files/2012/01/Consumers-and-domestic-heating-controls-a-literature-review.pdf 
(Accessed: 23 December 2016). 
Department for Energy and Climate Change (DECC) (2012) Emissions from Heat: Statistical 
Summary. London. Available at: 
https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_
data/file/140095/4093-emissions-heat-statistical-summary.pdf (Accessed: 19 March 2018). 
Department for Energy and Climate Change (DECC) (2013) United Kingdom housing energy 
fact file, Publication URN: 13D/276. London. Available at: 
https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_
data/file/345141/uk_housing_fact_file_2013.pdf (Accessed: 20 February 2018). 
EnerGAware (2016). Available at: http://www.energaware.eu/ (Accessed: 17 October 2017). 
Energy Technology Institute (2015) Smart Systems and Heat: Decarbonising Heat for UK 
Homes. Available at: https://s3-eu-west-
1.amazonaws.com/assets.eti.co.uk/legacyUploads/2015/03/Smart-Systems-and-Heat-
Decarbonising-Heat-for-UK-Homes-.pdf (Accessed: 13 March 2018). 
Fabi, V., Andersen, R. V. and Corgnati, S. P. (2013) ‘Influence of occupant’s heating set-point 
preferences on indoor environmental quality and heating demand in residential buildings’, 
HVAC and R Research, 19(5), pp. 635–645. 
Guerra-Santin, O. and Itard, L. (2010) ‘Occupants’ behaviour: Determinants and effects on 
residential heating consumption’, Building Research and Information, 38(3), pp. 318–338. 
Guerra Santin, O., Itard, L. and Visscher, H. (2009) ‘The effect of occupancy and building 
characteristics on energy use for space and water heating in Dutch residential stock’, Energy 
and Buildings, 41(11), pp. 1223–1232. 
Hamilton, I. G. et al. (2016) ‘Energy efficiency uptake and energy savings in English houses: A 
cohort study’, Energy and Buildings, 118, pp. 259–276. 
Heating and Hot Water Task Force (2010) Heating and Hot Water Pathways to 2020. 
Available at: https://www.hotwater.org.uk/uploads/559534E788187.pdf (Accessed: 20 
January 2018). 
HM Government (2008) Climate Change Act 2008, HM Government. London. Available at: 
http://www.legislation.gov.uk/ukpga/2008/27/pdfs/ukpga_20080027_en.pdf. 
HM Government (2010) Domestic Building Services Compliance Guide. 
Hoes, P. et al. (2009) ‘User behavior in whole building simulation’, Energy and Buildings, 
41(3), pp. 295–302. 
Huebner, G. M. et al. (2013) ‘Heating patterns in English homes: Comparing results from a 
national survey against common model assumptions’, Building and Environment, 70, pp. 
298–305. 
Hunt, D. R. G. and Gidman, M. I. (1982) ‘A national field survey of house temperatures’, 
Building and Environment, 17(2), pp. 107–124. 
Jones, R. V. et al. (2016) ‘Space heating preferences in UK social housing: A socio-technical 
household survey combined with building audits’, Energy and Buildings, 127, pp. 382–398. 
Kane, T., Firth, S. K. and Lomas, K. J. (2015) ‘How are UK homes heated? A city-wide, socio-
technical survey and implications for energy modelling’, Energy and Buildings, 86, pp. 817–
832. 
Lopes, M. A. R., Antunes, C. H. and Martins, N. (2012) ‘Energy behaviours as promoters of 
energy efficiency: A 21st century review’, Renewable and Sustainable Energy Reviews, 16(6), 
pp. 4095–4104. 
Loveday, D. L. and Vadodaria, K. (2013) Project CALEBRE: Consumer Appealing Low Energy 
technologies for Building REtrofitting - a summary of the project and its findings. 
Loughborough. 
Meier, A. et al. (2010) ‘How People Actually Use Thermostats’, Controls and Information 
Technology, 2, pp. 193–206. 
Meyers, R. J., Williams, E. D. and Matthews, H. S. (2010) ‘Scoping the potential of monitoring 
and control technologies to reduce energy use in homes’, Energy and Buildings. Elsevier 
B.V., 42(5), pp. 563–569. 
Office of National Statistics (ONS) (2012) Census 2011: Key Statistics for local authorities in 
England and Wales. London. 
Ravetz, J. (2008) ‘State of the stock-What do we know about existing buildings and their 
future prospects?’, Energy Policy, 36(12), pp. 4462–4470. 
Sardianou, E. (2008) ‘Estimating space heating determinants: An analysis of Greek 
households’, Energy and Buildings, 40(6), pp. 1084–1093. 
Shipworth, M. et al. (2010) ‘Central heating thermostat settings and timing: Building 
demographics’, Building Research and Information, 38(1), pp. 50–69. 
Wei, S., Jones, R. and De Wilde, P. (2014) ‘Driving factors for occupant-controlled space 
heating in residential buildings’, Energy and Buildings, 70, pp. 36–44. 
World Health Organization (1987) Health impact of low indoor temperatures: Report on a 
WHO meeting. Copenhagen, Denmark. 
Yoshino, H., Hong, T. and Nord, N. (2017) ‘IEA EBC Annex 53: Total energy use in buildings—
Analysis and evaluation methods’, Energy and Buildings, 152(March 2013), pp. 124–136. 
 
Appendix 
Table 1 Reported mean TRV and whole house thermostat setpoint temperatures according to household and motivation, behaviour and 
perception characteristics 
 Living room Bedroom 
Household characteristics n Mean thermostat Mean TRV n Mean thermostat Mean TRV 
temp. (°C) temp. (°C) temp. (°C) temp. (°C) 
Duration of tenancy       
<3 years 44 20.8 23.3 41 21.0 22.3 
3-5 years 20 21.3 23.7 18 21.7 22.6 
6-10 years 27 20.5 24.0 21 20.8 21.9 
11-20 years 34 20.4 23.5 26 20.6 22.2 
20+ years 19 20.7 22.2 14 20.1 19.4* 
Household size       
1 75 20.7 23.2 62 20.8 21.7 
2 45 20.7 23.8 37 20.8 22.1 
3 12 21.3 23.5 8 23.0 22.0* 
4 5 20.2 22.0 5 20.2 21.6 
5+ 7 19.7 24.3 8 20.4 22.3 
Household composition       
One person 68 20.6 23.4 62 20.8 21.7 
Couple, no dependent children 44 21.1 23.6 32 20.8 21.7 
Couple, dependent child(ren) 4 20.1 20.0* 5 23.2 22.4* 
Lone parent, dependent child(ren) 5 21.6 22.4 5 19.8 20.8 
Highest qualification of HRP       
O’Level, GCSE, NVQ Level 2 or equiv. 25 21.1 24.2 24 21.1 21.3 
A’Level, NVQ Level 3 or equiv. 28 20.8 24.6 28 20.9 22.6 
Degree level or above 22 20.8 22.5 18 21.0 22.1 
 Living room Bedroom 
Household characteristics n Mean thermostat Mean TRV n Mean thermostat Mean TRV 
temp. (°C) temp. (°C) temp. (°C) temp. (°C) 
Another kind of qualification 12 22.0 23.5 10 22.1 23.0 
No qualification 28 20.1 22.9 20 20.5 21.9 
Employment structure       
Employed 26 20.5 23.9 19 20.5 21.8 
Unemployed 5 23.0 21.6* 4 23.0 21.0* 
Retired 33 21.0 23.1 27 20.9 21.6 
Student 2 15.5 27.0 1 19.0 16.0* 
Household in receipt of welfare benefits     
Yes 59 21.0 24.0 48 21.1 22.8 
No 72 20.4 23.1 65 20.7 21.6 
Health of HRP       
Very good 26 20.6 22.9 26 21.1 21.7 
Good 41 20.6 23.1 33 20.7 21.2 
Fair 33 20.6 24.6 28 20.6 23.1 
Bad 20 21.9 24.0 17 21.8 21.3* 
Very bad 17 19.5 22.9 11 20.1 23.5 
Number of GP visits in a year       
0-1 32 20.3 23.2 30 20.7 22.7 
2-4 35 21.2 24.1 31 21.5 22.3 
5-6 14 20.9 24.1 12 20.8 22.7 
7-12 14 20.7 22.9 10 21.2 20.0* 
12+ 9 22.4 23.6 7 22.3 24.3 
Household with disabled members       
Yes 53 21.1 21.9 53 21.1 22.0 
No 67 20.7 22.1 67 20.7 21.8 
 Living room Bedroom 
Household characteristics n Mean thermostat Mean TRV n Mean thermostat Mean TRV 
temp. (°C) temp. (°C) temp. (°C) temp. (°C) 
Satisfaction with life       
0-3 (Dissatisfied) 20 21.1 22.4 11 21.6 23.1 
4-6 (Neither dissatisfied nor satisfied) 47 20.9 24.2 44 20.9 22.5 
7-10 (Satisfied) 76 20.4 23.2 64 20.6 21.3 
Motivation, behaviour and perception characteristics 
Affordability of energy bills    
Very easy 14 20.5 23.6 13 20.9 21.7 
Fairly easy 35 20.9 23.1 28 20.6 20.6 
Neither easy nor difficult 53 20.7 23.7 46 21.1 22.6 
Fairly difficult 29 20.2 23.3 23 20.3 21.7 
Very difficult 11 21.9 22.7 8 21.6 22.3 
I am worried about my energy bills      
Strongly agree 18 20.6 24.0 15 20.9 23.3 
Tend to agree 53 20.8 23.7 44 20.7 21.8 
Neither agree nor disagree 26 19.8 23.1 19 20.3 22.8 
Tend to disagree 19 21.2 23.3 17 21.1 20.9* 
Strongly disagree 21 21.8 22.8 20 22.0 20.7* 
I don’t understand how my home uses energy     
Strongly agree 11 21.7 24.6 9 22.3 23.8 
Tend to agree 37 21.1 23.5 30 20.6 21.6 
Neither agree nor disagree 31 20.1 24.0 25 20.7 22.6 
Tend to disagree 23 20.9 23.8 19 21.1 21.4 
Strongly disagree 28 20.8 22.2 27 20.9 21.3 
I often think about how my home uses energy     
Strongly agree 41 21.0 23.5 30 21.4 22.5 
 Living room Bedroom 
Household characteristics n Mean thermostat Mean TRV n Mean thermostat Mean TRV 
temp. (°C) temp. (°C) temp. (°C) temp. (°C) 
Tend to agree 67 20.5 23.8 59 20.7 21.8 
Neither agree nor disagree 16 21.1 23.0 16 20.5 21.6 
Tend to disagree 6 18.3 21.7 3 21.0 24.0 
Strongly disagree 8 22.9 22.5* 7 22.6 18.9* 
I have control over how much energy is used in my home     
Strongly agree 36 21.3 22.1 28 21.3 20.6* 
Tend to agree 54 20.3 24.0 45 20.3 22.5 
Neither agree nor disagree 29 20.8 22.8 23 21.4 21.1* 
Tend to disagree 13 19.8 24.5 12 21.7 22.5 
Strongly disagree 5 22.4 25.2 6 21.7 23.7 
I am not able to save anymore energy     
Strongly agree 15 20.6 23.7 11 21.1 22.7 
Tend to agree 46 20.5 24.1 37 20.7 22.1 
Neither agree nor disagree 38 20.6 23.3 31 20.5 21.7 
Tend to disagree 18 20.6 22.8* 19 21.0 20.7* 
Strongly disagree 11 22.3 22.0 10 22.5 21.4* 
I make sure the curtains/blinds are closed when the heating is on in the evening    
Always 82 20.8 23.3 67 20.9 21.8 
Often 35 20.3 23.8 28 20.8 21.5 
Sometimes 15 20.7 24.4 14 20.8 23.3 
Very occasionally 3 22.0 24.7 3 22.0 22.7 
Never 7 21.4 21.1* 6 21.2 22.0 
I make sure the curtains/blinds are open when the sun is shining in winter    
Always 103 20.9 23.6 87 21.1 22.0 
Often 27 19.8 23.2 20 19.7 21.4 
 Living room Bedroom 
Household characteristics n Mean thermostat Mean TRV n Mean thermostat Mean TRV 
temp. (°C) temp. (°C) temp. (°C) temp. (°C) 
Sometimes 8 21.1 24.8 7 21.1 24.0 
Very occasionally 3 19.7 17.3* 3 19.7 17.3* 
Never 1 26.0 24.0* 1 26.0 28.0 
I make sure the windows are closed when the heating is on    
Always 106 20.7 23.8  87 20.8 22.3 
Often 24 19.7 23.5 21 20.2 21.1 
Sometimes 10 23.0 20.4* 9 22.8 19.8* 
Very occasionally 2 23.0 22.0* 1 26.0 28.0 
I closed the doors between rooms     
Always 49 20.6 22.9 38 20.6 22.0 
Often 24 21.8 23.8 20 21.9 22.0 
Sometimes 26 20.5 23.8 27 20.9 21.4 
Very occasionally 13 19.8 24.6 10 20.1 23.6 
Never 26 20.3 23.6 20 20.7 21.9 
I wear very warm clothes in winter so I keep the heating low or off    
Always 60 20.8 23.4 48 21.1 22.5 
Often 42 20.2 22.8 34 20.1 20.5 
Sometimes 24 19.9 25.1 24 21.7 23.1 
Very occasionally 9 23.5 23.8 5 21.7 21.2* 
Never 8 21.3 21.8 8 20.4 22.0 
I change the temperature on my thermostat    
Always 64 20.8 23.8 55 21.1 21.8 
Often 29 20.7 22.8 26 20.7 21.5 
Sometimes 32 20.8 23.9 24 20.7 22.6 
Very occasionally 6 18.4 23.0 5 20.5 20.8 
 Living room Bedroom 
Household characteristics n Mean thermostat Mean TRV n Mean thermostat Mean TRV 
temp. (°C) temp. (°C) temp. (°C) temp. (°C) 
Never 10 20.9 22.6 7 20.9 24.0 
I turn the heating off when no one is at home     
Always 97 20.8 23.4 81 21.0 22.2 
Often 20 20.2 24.3 18 20.3 21.4 
Sometimes 16 19.9 23.5 13 20.0 22.2 
Very occasionally 4 23.5 22.0* 4 23.5 17.0* 
Never 6 21.3 23.0 3 21.3 24.7 
I turn off the heating in rooms that are not normally used    
Always 56 20.5 22.7 35 20.7 21.9 
Often 25 20.3 23.2 22 20.3 19.7* 
Sometimes 17 20.1 24.7 19 20.6 22.5 
Very occasionally 14 21.9 23.3 13 21.7 21.1* 
Never 23 21.5 24.2 22 21.5 23.5 
I adjust the temperature on my radiators     
Always 40 20.5 23.2 31 21.4 22.0 
Often 30 21.3 22.2 24 21.3 20.3* 
Sometimes 35 20.6 23.4 31 20.5 21.2 
Very occasionally 16 20.3 25.1 13 20.4 22.8 
Never 18 20.9 25.0 16 20.8 24.9 
* TRV setpoint temperature is lower than the whole house thermostat setpoint temperature 
 
View publication stats