Available via license: CC BY 4.0
Content may be subject to copyright.
A review of air pollution impact on subjective well-being: Survey
versus visual psychophysics
Yuan Li
a
,
b
, Dabo Guan
b
,
*
, Shu Tao
c
, Xuejun Wang
c
, Kebin He
d
a
Institute of Resource, Environment and Sustainable Development, Jinan University, Guangzhou, 510632, China
b
School of International Development, University of East Anglia, Norwich NR4 7TJ, UK
c
Laboratory for Earth Surface Processes, College of Urban and Environmental Sciences, Peking University, Beijing, 10080, China
d
State Key Joint Laboratory of Environment Simulation and Pollution Control, School of Environment, Tsinghua University, Beijing 100084, China
article info
Article history:
Received 28 November 2017
Received in revised form
24 February 2018
Accepted 26 February 2018
Available online 27 February 2018
Keywords:
Air pollution
Psychophysics
Subjective well-being
abstract
Air pollution is a worldwide environmental and health issue, especially in major developing countries. A
recent World Health Organization report shows about 3 million deaths in the world in 2012 are due to
ambient air pollution and China and India are the countries with the most severe challenge. Air pollution
influences people's thought and experience of their lives directly by visual perceptions. This reduces
people's subjective well-being (SWB) to a significant degree. Empirical researchers have made efforts to
examine how self-reported well-being varies with air quality typically by survey method - matching SWB
data with monitored air pollution data. The ir findings show NO
2
, particles, lead, SO
2
and O
3
have significant
negative impact on SWB. However, it is very hard to match air pollution characteristics from monitor
stations with each respondent's state of SWB at the moment a survey is conducted. Also it is very hard to find
the detailed trend impact from only air pollution factor on SWB. This review illustrates the features and
limitations of previous survey studies on quantifying the effects of air pollution on subjective well-being.
This review further displays the progress of psychophysics and its application in landscape and air qual-
ity research. We propose using psychophysics application to quantify air pollution impact on SWB.
©2018 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/by/4.0/).
1. Introduction
Air pollution is a worldwide environmental and health issue,
especially in major developing countries. According to World
Health Organization (WHO), in the year of 2012, ambient air
pollution (AAP) caused 3 million deaths within the world and about
87% of these deaths occur in low- and middle-income countries
(WHO, 2016). The most deaths came from the low- and middle-
income countries of WHO Western Pacific and South East Asian
with 1.1 and 0.79 million respectively. China and India are the
countries with the most contribution to the figures and also they
are the two countries with the most deaths globally (Liu and Liu,
2011; WHO, 2016). The other low- and middle-income countries
and regions share the burden with 0.68 million. The remaining
deaths occur in high-income countries of Europe, the Americas,
Western Pacific, and Eastern Mediterranean, which are about 0.38
million (WHO, 2016). As China and India are planning for and
experiencing rapid urbanization, the air pollution situation will
continue to deteriorate.
Smoggy days can impact people's visual perception directly.
Subjective well-being (SWB) belongs to a perceptual domain and
involves how people think about and experience their lives. In
addition, SWB includes different evaluations that individuals' make
regarding their lives (Diener, 1984, 2006) that cover the events
happening to their bodies and minds and the circumstances in
which they live. Rather than conforming to external standards,
assessments of SWB are based on an individual's own chosen
criteria. Policy-makers are likely to consider SWB in planning and
assessing the impact of policy decisions. As a unique example, the
Asian nation of Bhutan officially established the Gross National
Happiness (GNH) measure by law and replaced the traditional
economic policy goal of increased GDP with increased GNH.
1
Sub-
sequently, many western governments have officially introduced or
*Corresponding author.
E-mail address: dabo.guan@uea.ac.uk (D. Guan).
1
Further information can be found under the following link: http://ophi.org.uk/
policy/national-policy/gross-national-happiness-index/.
Contents lists available at ScienceDirect
Journal of Cleaner Production
journal homepage: www.elsevier.com/locate/jclepro
https://doi.org/10.1016/j.jclepro.2018.02.296
0959-6526/©2018 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Journal of Cleaner Production 184 (2018) 959e968
initiated measurements of national happiness or life satisfaction,
2,3
(Schmitt, 2013).
Traditionally, a survey instrument is the tool used to examine air
pollution's impact on SWB (one type of people's perception), and it
has been widely used by both economists and sociologists. Mean-
while, environmental psychologists have started adopted psycho-
physical methods to solve environmental issues, including
landscape attractiveness and acceptable visual air quality standard.
However, psychophysical procedure has never applied to evalu-
ating air pollution impact on SWB. This review work aims to pro-
vide a new perspective in measurement of air pollution impact on
SWB. Fig. 1 shows the structure of this work. This paper evaluates
the performance and limitations of empirical surveys in quantifying
air pollution's impact on SWB; reviews certain developments in
environmental psychophysics and their application in visual air
quality research; proposes to use psychophysical methods to
quality air pollution impact on SWB. The remainder of this paper is
organized as follows. Section 2provides a brief introduction of SWB
and air pollution. Section 3illustrates the features and limitations
of previous survey studies on quantifying the effects of air pollution
on subjective well-being. Section 4displays the progress of psy-
chophysics and its application in landscape and air quality research.
Section 5summaries the strength and weakness of survey method
and proposes an application of visual psychophysical experiments
to explain the relation between air pollution and SWB.
2. 2Subjective well-being
For decades, unhappiness has been explored deeply by scientific
research, while positive subjective well-being was largely ignored
by social scientists prior to the 1970s. In 1973, Psychological Ab-
stracts International began to include happiness as an index term.
Since the journal of Social Indicators Research was founded in 1974,
many articles have been published that have focused on SWB.
Diener (1984) comprehensively reviewed SWB studies, including
definitions and measurements.
2.1. Definitions and measurements of SWB
The definitions of well-being can be grouped in three categories.
Firstly, researchers represented by Coan (1977) and Tatarkiewicz
(1976) have defined well-being by external standard such as vir-
tue or holiness rather than subjective state. They thought that well-
being can be obtained by leading a virtuous life. Secondly, social
scientists have tried to define well-being by focusing on the factors
which leads people believe they are in positive state (Andrews and
Withey, 1976; Chekola, 1975; Shin and Johnson, 1978). Thus,
happiness is defined as the harmonious state where people's de-
sires and goals are satisfied. The third meaning of happiness em-
phasizes positive emotional experiences as gaining a superiority of
positive affect over negative affect (Bradburn, 1969). SWB empha-
sizes people's own judgements about themselves and belongs to
the second or third well-being definitions categories. The mea-
surement of SWB is often restricted to the measurement of
happiness (OECD, 2013). This may be because the notion of SWB is
used in research literature as a replace of the term ‘happiness’.A
definition given by Diener et al. (2006) is commonly identified by
research in this field, which is ‘Good mental states, including all of
the various evaluations, positive and negative, that people make of
their lives and the affective reactions of people to their
experiences.’
Single-item survey questions frequently constitute measures
used by early social scientists. Such measures tend to fall in
happiness categories from which scales cannot address all aspects
of SWB (Andrews and Withey, 1974; Cantril, 1965; Gurin et al.,
1960). Later, multi-item surveys were integrated into scholarly
research. During this period, lots of survey scales merged for
different purposes of studies and different backgrounds of in-
terviewees. For examples, Lawton (1975) used 17-item scale to
measure lonely dissatisfaction, agitation and attitude toward one's
aging. Kozma and Stones (1980) used 24 items to measure positive
and negative affect and experiences. Researchers have been trying
to find general components of SWB, which can be used universally
and three general components of SWB were proposed, including
life satisfaction judgement, positive affect and negative affect
(Andrews and Withey, 1976; Bradburn, 1969; Bryant and Veroff,
1982; Harding, 1982; Zevon and Tellegen, 1982). Bradburn (1969)
proposed his global happiness judgement standard Affect Balance
Scale by making comparison of people's negative affect with their
positive affect. The positive affect may including the proud or
pleasant feel and the negative affect may ask about their upset,
unhappy or depressed feelings. In a recent study by Kahneman et al.
(2004), the SWB for an activity can be measured as the net affect of
the average of all positive affects (happy, enjoying themselves) less
the average of all negative affects (Frustrated, depressed, worry)
and the length of time people spending on it. Nowadays, life
satisfaction and affects are both widely studied by researchers.
2.2. Drivers of SWB
Over the last two decades, the importance of SWB has received
considerable attention in various fields. The output of SWB mea-
surements is widely implicated in social and health fields (Cohen
et al., 2003; Danner et al., 2001; Fujiwara and Campbell, 2011;
Ostir et al., 2001; Steptoe et al., 2005). Growing literature in the
field of economics has advanced the definition of the factors
impacting SWB at the individual level. A predetermined level of
happiness unique to each individual's genetics and personality is
proposed (Costa et al., 1987; Cummins et al., 2003; Lucas et al.,
2003; Lykken and Tellegen, 1996). Easterlin (2003) outlined an
improved theory of how life events affect SWB. Life events and
personality interact with one another to shape happiness at the
Fig. 1. The structure of this review.
2
The full article can be downloaded from the web page of the Federal Ministry of
Finance: http://www.bundesfinanzministerium.de.
3
Further information can be found under the following link: https://www.
washingtonpost.com/business/economy/if-youre-happy-and-you-know-it–let-the-
government-know/2012/03/29/gIQAlSL2jS_story.html?utm_term¼.7a170ddc8352.
Y. Li et al. / Journal of Cleaner Production 184 (2018) 959e968960
individual level. Life circumstances such as income (Di Tella et al.,
2003), good health (Gerdtham and Johannesson, 2001; Steptoe
et al., 2005) and family (Lucas et al., 2003) are characterized by
positive relationships with SWB in economic studies. By contrast,
other events such as unemployment (Knabe and Ratzel, 2010;
Theodossiou, 1998) and inflation (Di Tella et al., 2001) negatively
affect SWB.
2.3. Air pollution and SWB
In this context, it is quite important to improve our under-
standing of the elements that influence SWB, particularly in new
lines of research. Air pollution is an environmental and health issue
worldwide, particularly in developing countries, that has been
implicated as a risk factor in health issues ranging from respiratory
diseases, cardiovascular diseases, cancers, and impaired cognitive
function (Brook, 2008; Brunekreef and Holgate, 2002; Weir, 2012).
Early studies suggest that air pollutants with high concentrations
may affect exposed people's emotional state. The effects may range
from subclinical alterations of mood to pronounced psychopatho-
logical symptoms (Rotten and Frey, 1984; Rotton, 1983; Strahilevitz
et al., 1979). Thanks to the psychobiological stress concepts, the
relationships between air pollution and well-being is clarified
(Bullinger, 1989; Frankenhaeuser, 1980; Lazarus and Cohen, 1978).
In line with these concepts, air pollution can be regarded as a
stressor (Campbell, 1983) and its influence including emotional,
behavioral and physical changes can be mediated by cognitive
appraisal (Cohen, 1980; Cohen et al., 1986; Folkman et al., 1979).
Thus, air pollution may also damage people health indirectly
through influencing people's well-being by its nature of disgust.
When exposed people appraising this aversive nature, stress re-
actions of damaged well-being and physiological dysregulation are
induced (Baum et al., 1982; Baum and Singer, 1986). Meertens and
Swaen (1997) even proposed that air pollution's psychological ef-
fects might have a considerable influence on well-being that ex-
ceeds its physical effects. Further refinement of the relationship
between SWB and air pollution has been attempted in more recent
decades. In the following sections, the challenges encountered and
the methods employed in this field of study will be summarized
and then a new approach will be explicated that involves the
application of psychophysical experiments to assess the effects of
air pollution on self-reported well-being.
3. Using surveys to reveal the relationship between air
pollution and SWB
However, air pollution characteristics are not normally collected
in surveys but instead from monitor stations. Generally, air pollu-
tion characteristics can change dramatically based on time, space,
weather and climate (Luechinger, 2009; Schmitt, 2013). It is nearly
impossible to match air pollution characteristics from monitor
stations with each respondent's state of SWB at the moment a
survey is conducted. A conventional challenge to empirical studies
is to seek high quality air pollution data with fine spatial and
temporal disaggregation and to connect this information with a
specific respondent. Previous researchers have typically merged
the average annual air pollution data at the country or regional
level with collected or ready-to-use panel SWB data from surveys
to analyse the regression correlations between the two datasets
(Schmitt, 2013; Welsch, 2007). Socio-economic and demographic
variables are often considered simultaneously in controlling the
influence on SWB as opposed to other potential factors. Weather
and other climate characteristics are also sometimes covered.
Table 1 summaries the findings and contributions of empirical
studies according to data spatial and temporal levels. Sections
3.1e3.3 distinguish the reviews of previous studies at different
spatial levels.
3.1. Air-quality data collected at the country level
Early researchers examining how self-reported well-being var-
ies with air quality typically used cross-sectional air-quality data
collected at the country level. In Welsch's studies, both the SWB
and air pollution datasets are based on average country level. One
of these studies covers data from 54 countries. The SWB data comes
from the World Database of Happiness and air pollution data are
taken from the database of the global Environmental Sustainability
Index. The variables considered include air pollution (SO
2
,NO
2
and
particles), freedom, rationality, income and water pollution, and
the results show that the impact of air pollution on SWB is difficult
to measure (Welsch, 2002, 2007). Another study covers 10 Euro-
pean countries. Again, the SWB data comes from the World Data-
base of Happiness and air pollution data are provided by the
Organisation for Economic Co-operation and Development (OECD),
based on country-wide networks of measurement stations. Pol-
lutants considered in this study include NO
2
, lead and particles. To
avoid overcompensating for unobserved heterogeneity, income is
the only controlled variable. It has been demonstrated that air
pollution plays a significant role in predicting changes in SWB on
both cross-country and inter-temporal bases.
In Luechinger (2010), life satisfaction, SO
2
and household in-
come were considered the main variables, and domestic and socio-
economic variables were also controlled. The data cover 13 Euro-
pean countries over the 1979e1994 period. The air pollution data
were collected from the OECD at the average year level, and
individual-level SWB data came from Eurobarometer. Air pollution
was found to have a statistically significant and robust negative
impact on SWB. Schmitt (2013) used the individual level for life
satisfaction data from the German socio-economic panel (SOEP)
survey and the average of country-level air pollution data from 765
monitor stations in Germany to examine the impact of CO, NO
2
and
O
3
on life satisfaction. He analysed daily air pollution data in
Mecklenburg-West Pomerania in 2005 and found that the density
of pollutants varies with the seasons. To improve the sensibility of
the survey-based life satisfaction data, average daily country-level
air pollution data in Germany were used in his study. Socio-
economic and weather factors were controlled for in the analysis;
among the three assessed pollutants, only O
3
has a significantly
negative impact on life satisfaction.
3.2. Air-quality data collected at the regional level
Most empirical papers using more spatially disaggregated
pollution data at the regional level to illustrate the air pollution
impact on SWB have focused on one country. Smyth et al. (2008)
evaluated environmental features, including SO
2
emissions, using
SWB in 30 cities in urban China. Individual level SWB data were
obtained from China Mainland Marketing Research Company
(CMMRC). The regional average annual air pollution data from the
China Statistical Yearbook were matched to each respondent based
on his or her residence. Socio-economic and demographic variables
were well-controlled and a clear negative impact of SO
2
emission
on SWB was found.
Luechinger (2009) gathered SO
2
concentration data for nearly
two decades, covering 553 monitor stations in Germany for the
German Federal Environmental Agency. Due to the shortage of air
pollution data from some monitor stations in some individual
years, interpolation was applied to estimate the missing data. SWB
data were provided by the SOEP survey for the matching years.
Correlations between the two variables were analysed based on
Y. Li et al. / Journal of Cleaner Production 184 (2018) 959e968 961
average annual regional data in Germany. Variables such as socio-
economic factors and particles were controlled and a significant
impact of SO
2
on SWB was found in this study. Ferreira and Moro
(2010) valuate PM
10
with regional data in Ireland. According to
respondents' locations, the average annual pollution data from the
closest monitoring station were linked with respondents’SWB
data. Demographic, socio-economic and climate variables were
controlled during the analysis. The concentration of PM
10
was
shown to have an effect on individual-level SWB.
Levinson (2012) obtained the happiness data from the General
Social Survey (GSS), which provides the date and location of the
survey for each respondent. The EPA's Air Quality System (AQS)
provides the daily air quality data collected from thousands of
monitor stations and also the locations of these monitors so that
most respondents' country or city is identifiable. With the help of
the National Climate Data Center, the daily temperature and rainfall
information were specified. A significant relationship between
happiness and air quality was found in the average at daily and
regional levels. He found that higher levels of particulates are
negatively correlated with well-being in the US. Demographic and
weather variables were well-considered and lower happiness
levels were found to be related to worse local air pollution on the
interview day.
Ferreira et al. (2013) first conducted cross-sectional analyses
with spatially disaggregated data at the region level on SO
2
to
explain individual SWB in Europe. Other spatial controls were also
considered in their study including temperature, precipitation and
regional economic performance. They adopted SWB data provided
by the European Social Survey (ESS) collected between 2002 and
2007 and created the dataset on SO
2
concentrations from 248
regions in Europe for the same period. The annual mean SO
2
data
were interpolated from the Geographic Information Systems (GIS)
in regions in 23 European countries between 2002 and 2007. A
robust negative impact of SO
2
concentrations on self-reported life
satisfaction was found.
All these studies based on multi regions provide comprehensive
interactive information between the significant indicators like re-
lations between air pollutants and economic/climate backgrounds.
However, limited air pollutants are investigated in each of the Eu-
ropean studies. Little predications of SWB state in developing
countries with more serve air pollution can be made.
3.3. Air-quality data collected at individual level
Few papers use spatially disaggregated air pollution data
focused on the individual level. Some of them used subjective self-
reported air pollution levels instead of objective monitor measures
or modelling data. Rehdanz and Maddison (2008) analysed the
relationship between self-reported impact of air pollution and SWB
data, both of which were obtained from the German socio-
economic panel (SOEP) survey. Li et al. (2014) conducted a similar
research based on a mining area in Jinchuan, China. Both studies
provided a mindful control of socio-demographic, economic vari-
ables affecting SWB and showed a negative correlation between
SWB and air pollution. However, their air pollution data were in-
dividual perspectives from survey but not monitor stations and also
pollutants were not specified. Thus, it is difficult to specify the
objective impact from air pollution on SWB.
Thanks to air pollution models, the spatial data can be broken
down to individual level. MacKerron and Mourato (2009)
Table 1
Air pollution and SWB research developments based on survey. ***,**,*denote statistical significance at 1, 3e5 and 10% levels used in empirical studies, respectively. Some
pollutants are associated with different significance levels as different models are considered.
Author Spatial level of air
pollution data
Temporal level of
air pollution data
Air pollutants Air pollution data source Controls of other factors
in SWB
Main findings
Welsch (2002) country level (54
countries)
yearly SO
2,
NO
2,
particles Environmental Sustainability
Index
freedom, rationality,
income, water pollution
impact is hard to
measure
Welsch (2007) country level (10
countries)
yearly NO
2,
particles, lead monitor stations income negative impact with
NO
2
**, particles**, lead**
Luechinger
(2010)
country level (13
countries)
yearly SO
2
monitor stations socio-economic,
demographic, climate
negative impact with
SO
2
*,**
Schmitt (2013) country level
(Germany)
daily CO, NO
2
,O
3
monitor stations socio-economic, weather negative impact with O
3
*
Smyth et al.
(2008)
region level (30 regions
in China)
yearly SO
2
monitor stations socio-economic,
demographic
negative impact with
SO
2
***
Luechinger
(2009)
region level (about 445
regions in Germany)
yearly SO
2
monitor stations and
interpolation
socio-economic, particles negative impact with
SO
2
*
Ferreira and
Moro (2010)
region level (9 regions
in Ireland)
yearly PM
10
monitor stations demographic, socio-
economic, climate
variables
negative impact with
PM
10
*,**
Levinson (2012) region level daily PM
10
monitor stations demographic, weather negative impact with
PM
10
*,**
Ferreira et al.
(2013)
region level (248
regions in Europe)
yearly SO
2
monitor stations weather, economic negative impact with
SO
2
*
Rehdanz and
Maddison
(2008)
individual level
(Germany)
daily not specified self-reported affect levels socio-economic,
neighbourhood
negative impact
MacKerron and
Mourato
(2009)
individual level
(London)
monthly NO
2
,PM
10
estimated from models,
interpolation
socio-economic,
demographic
negative impact with
NO
2
**
Li et al. (2014) individual level
(Jinchuan, Gansu, China)
daily not specified self-reported pollutant levels socio-economic,
demographic
negative impact ***
Ambrey et al.
(2014)
individual level
(Queensland)
monthly PM
10
estimated from models socio-economic,
demographic, weather
negative impact with
PM
10
*
Orru et al. (2016) individual level
(Estonia)
yearly PM
10
estimated from models socio-economic,
demographic
negative impact with
PM
10
**
Zhang et al.
(2017)
Individual Level daily air pollution index
(SO
2,
O
2
,PM
10
)
monitor stations and estimated
from models, interpolation
socio-economic,
demographic, weather
negative impact ***
Y. Li et al. / Journal of Cleaner Production 184 (2018) 959e968962
developed a survey to collect individual SWB data in London. The
air pollution data they used were from the Air Dispersion
Modelling System provided by Cambridge Environmental
Research Consultants Ltd with an average annual level covering an
area of 3260 km
2
in 50 50 m grid squares (cells). Pollutants
examined included NO
2
and PM
10
.Withthismodel,theair
pollution levels for respondents were estimated according to their
postcode. To match the survey time, a linear interpolation was
used to estimate the air pollution level during a specifictimeof
year for each respondent. This method was used under the
assumption of a smooth and continuous air pollution change in
London. Socio-economic and demographic variables were
controlled and only NO
2
was found to have a significant impact on
SWB. Their results indicate that measured air pollution data are
negatively associated with SWB.
In recently years, most studies are based on air pollution
models. In Ambrey's work based on Queensland, multi air pollut-
ants were considered. However, only PM
10
was found to have the
strongest negative association with life satisfaction (Ambrey et al.,
2014). In Orru et al.’s study in Estonia, individual SWB data were
obtained from European social survey (ESS). Air pollution datawere
gotten from Eulerian air quality dispersion model with 1 1km
grid squares covering the whole country (Orru et al., 2016). A sta-
tistical negative influence of PM
10
is found on SWB. However, all
studies only found limited kind of air pollutants negatively related
with SWB, which cannot provide an overall prediction of air
pollution impacts on SWB. This may be caused by the relatively
higher quality of local air conditions. Moreover, the air pollution
data is in monthly or yearly levels, which could due to the avail-
ability of the source of getting air pollution data. In the very new
study by Zhang et al. they successfully valued air quality using
moment-to-moment happiness data in a daily and local level and
found bad daily air quality affect overall life satisfaction not much
but it reduces hedonic happiness and increases the rate of
depressive symptoms (Zhang et al., 2017). The API data in their
research showed that there is no obvious air quality improvement
from 2010. Thus, there is a possibility that long-term air pollution
has already caused impact on people's long-term life satisfaction
and this perception cannot be easily changed by some days expe-
rience of good air condition during a year unless air condition is
improved as a long term. This impact is hard to be found by reality
of stable poor or good air condition.
SWB involves people's evaluations regarding their mind and
body. Of course, the traditional survey measurements address the
perceptions of both mind and body aspects. However, the chal-
lenges in empirical studies are clear. From a psychological
perspective, it is known that the perspective at the time of the
interview cannot adequately explain the current SWB state because
a respondent's answer to the question of happiness level is strongly
dependent on his or her current mood. The air pollution data must
be disaggregated at the individual level from both spatial and
temporal aspects to explain individual SWB data. Few of the
empirical studies achieved this standard, shown in Table 1. Typi-
cally, limited air pollutants are examined in difficult locations.
However, various air pollutants always appear together, floating in
the atmosphere and working together to change people's percep-
tion, including affecting SWB. The same levels of air pollutants in
different weather and at various sun angles can generate very
different sensations. In seldom studies individual's idiosyncrasies
were considered from all the perspectives of socio-economic, de-
mographic, climate and weather together although all the consid-
ered perspectives can still not explain all the possibilities of
individual's idiosyncrasies. In attempting to provide a full picture of
the combined effects of air pollutants on SWB, the development of
psychophysical applications for landscape attractiveness and visual
air quality is reviewed and the possibility of applying psycho-
physical methods for quantifying the impact of air pollution on
SWB is discussed in following sections.
4. The possibility of using psychophysics to assess air
pollution impact on SWB
In modern psychophysics research, the term of psychophysics
refers to the relationship between external stimuli in the physical
domain and mental events in the psychological domain (like peo-
ple's sensations and perceptions) and the method used to deal with
the relation between them (Marks and Gescheider, 2002). Percep-
tion is a process of representing and understanding the physical
environment by the organization, identification and interpretation
of sensory information. All perception involves physical or chemical
stimulation of the sense organs (Kail and Cavanaugh, 2015). Psy-
chophysical scaling addresses the quantification of people's sen-
sations and perceptions in psychological domains obtained from
external stimuli in the physical domain (Engeldrum, 2000; Marks
and Gescheider, 2002). As discussed above, SWB belongs to peo-
ple's perception domain related to their feelings about life satis-
faction. As a type of physical stimuli, changes in air pollution can
result in different responses in people's SWB directly through visual
experience, and this response can be quantified by psychophysical
scaling methods.
It's hard to find any empirical work using psychophysical
method to measure the impact of air pollution on SWB. However,
lots of empirical studies proofed that visual psychophysical method
is a valid and reliable way to estimate the relationship between
environmental features and human perceptions. In this section, the
developments in this field will be displayed and using photo slides
as stimuli to quantify air pollution impact on SWB is proposed.
4.1. The preliminary application of psychophysics in environmental
research
Theoretically, visual perception may not entirely represent the
entire perception that people process from air pollution, especially
for chemical simulations. However, because sight dominates the
way we ‘see’the world and about one-third of our brain is dedi-
cated to processing visual experience (Gilbert and Walsh, 2004),
researchers have suggested that visual input plays the primary role
in human brain development (Kirk, 2006), and visual air quality is
therefore one aspect of the principal information that individuals
use to judge air pollution (Barker, 1976; Hyslop, 2009).
Thanks to Craik and Zuhe's (1976) milestone text, convincing
arguments have been presented to propose the use of “Perceived
Environmental Quality Indices”as an assistant to existing physi-
cally based systems for estimating various environmentally rele-
vant aspects. Perceptual judgement procedures have been widely
involved in the quantification of environmental features. Scenic
quality assessment is an important research area where perceptual
approaches are applied in early stage. Compared with other, such as
descriptive inventories, questionnaires or opinion surveys,
perceptual preference approaches are believed to represent the
landscape more intuitively than verbal surveys (Daniel and Boster,
1976 ). Studies have shown that colour slides or photographs can
represent actual landscapes quite well (Boster and Daniel, 1972;
Daniel and Boster, 1976; Zube, 1974).
However, professional psychophysical data collection and
analysis method are not properly applied at the same time. Among
a number of procedures for obtaining observers' judgments, an
individual rating approach is quickly used extensively not only
because of its efficiency but also because it can provide relative
differences between samples. However, to appropriately analyse
Y. Li et al. / Journal of Cleaner Production 184 (2018) 959e968 963
the data from ratings, researchers must address two potential
problems in the perceptual approach. One is the difference in ob-
servers' judgment criteria, which depends on the nature of their
past experiences. The other is the difference of scale units that
observers tend to use, although the same length of scale is pro-
vided. In Boster and Daniel's (1972) study, estimations were made
of the empirical analysis methods, including the standardization of
ratings to adjust an observer's idiosyncratic use of the response
scale. This standardization method is efficient for eliminating the
arbitrary differences between observers' use of the response scale.
Even results from different scales can be directly compared. How-
ever, it tends to hide the real discrimination between individual
observers' judgments. Based on some classic psychophysical the-
ories and methods (Green and Swetts, 1966; Swetts, 1973;
Thurstone, 1927; Torgerson, 1958), an appropriately standardized
estimation method, Scenic Beauty Estimates, was proposed (Daniel
and Boster, 1976). This method provides an approach to evaluate
different landscapes with each landscape containing a number of
different scenic photographs/slides.
These studies provide proof that visual psychophysical method
can be used to solve perception and environmental problems.
Moreover, the psychophysical analysis procedure can also adjust
observers' idiosyncratic automatically without collecting observers’
social background data. All these works have established a foun-
dation for further visual air quality research (Balling and Falk, 1982;
Buhyoff et al., 1983; Peterson, 1967; Peterson and Neumann, 1969;
Propst and Buhyoff, 1980). All the developments of psychophysics
applied in air quality research are illustrated in Table 2.
4.2. Reliability and validity of slide observation of air quality
compared with field studies
Studies by Malm et al. (1980) have shown that observers’visual
perceptions of air quality are reliable, which is consistent and re-
veals substantial sensitivity to measured variables of air quality by
optical instruments. Efforts have also been made by previous re-
searchers to determine the relationship between judgments of vi-
sual air quality and the field.
In Malm et al. (1981), a total of 40 slides and three-dimensional
corresponding scenes were judged. Student's t-test and f-test were
used to compare the means from the field and slide observations.
Results indicate no significant difference between the means and
also suggest that colour slides/photographs can be acceptable as
surrogates for actual scenes for perceptual judgements.
Stewart et al. (1983) published their paper based on a five-year
visual air quality study in Denver. A pilot study and a main study
were conducted with a large number of observations. The observers
made both field and photographic observations. A strong correla-
tion between field and photographic judgments of visual air quality
was found, indicating that non-visual cues such as smell or im-
pressions of pollution from earlier have little influence on visual air
quality. The paper also suggests that this procedure, with high
reliability and validity, can be used to examine the relation between
visual air quality and other variables, such as well-being or life
quality and pollutant concentrations or sources.
Later, Stewart et al. (1984) improved the methodologies used by
the empirical studies on the validity of photographic judgment. In
the new study, rather than averages of group ratings, individual
observers’judgments were analysed. Moreover, various judgments
were collected to compare the relationships between judgments of
photographs and the field, and systematic components of the
variation in judgments between two observation procedures were
examined. The results of this study enhanced previous findings and
encouraged the use of photographs to investigate visual air quality
issues.
After a broad review of various approaches on visual air quality
management, including human perceptual judgment, physical and
chemical measurements, Middleton et al. (1985) summarised that
photograph judgment has been highly recommended as a feasible,
cost-effective substitute for field judgments in evaluating the
relationship between other variables and visual air quality. Thus,
visual perceptual experiments with photograph judgments are
widely used in later research of air quality perceptions (BBC
Research and Consulting, 2003; Fajardo et al., 2013; Pryor, 1996).
4.3. Looking for the most satisfactory indicators of visual air quality
Another interesting and relevant study area defines the ele-
ments that can influence visual air quality. Researchers attempt to
build models to predict human perceptions of air quality according
to emissions or relevant perceptual cues like sun angle, clarify,
colour and border.
Malm et al. (1980) applied slide and field observation proced-
ures to explore the relations between visual air quality and colour
contrast. High correlations were detected between both variables
and this relation was found to be independent of the demographic
background of observers. In the study by Latimer and Hogo, 1981,
the slide observation procedure was adopted to collect perceptual
data. Visual range and inherent scenic characteristics were found to
be sensitive to visual air quality. Malm et al. (1981) found that sun
angle, colour contrast, inherent scenic beauty and the distance to
each of the scenic elements are sensitive to observers' perceived
visual air quality. However, observers’demographic background
affects visual air quality ratings very little.
To develop a physically based index of visual air quality,
Mumpower et al. (1981) collected field perceptual data and phys-
ical environmental data from various local locations and used factor
analysis and multiple regression analysis to investigate the re-
lationships among them. The perceptual data included visual air
quality and relevant perceptual cues (distance, clarify, colour and
border). The environmental data included aerosol scattering and
absorption measures as well as meteorological measures and
pollutant data. The results suggest that multiple considerations of
perceptual cues can describe visual air quality judgments. However,
the collected environmental measurements are not sufficient pre-
dictors of perceptual judgments.
Middleton et al. (1983) attempted to build a model to predict
human judgments of air quality influenced by emissions. Field
research was conducted to obtain perceptual data. Only directly
emitted fine particles were included in this model. The model also
included the distance between the observer and the target, the sun
angle, and the intensities of the background and target. Results
indicate that refined concentrations can increase the validity of the
model between target clarity and emissions, but sky colour and
borders are not easily explained by emissions.
To address the indicators of judgments of visual air quality, in
the study of Middleton et al. (1984), physical/chemical environ-
mental data and perceptual data were collected over various time
frames, observation locations and atmospheric conditions. The
correlation coefficient was used for analysing the single indicator,
which can provide the best overall prediction of visual air quality
judgments. Light scattering extinction (b
ext
or b
xp
) by a tele-
photometer is found to be the most satisfactory and direct indicator
of visual air quality. Multiple regressions were applied to compare
the prediction performance of different combinations of variables.
It is indicated that combinations of physical/chemical measures
together seem to provide better predictive ability than the single
measurement of b
xp
. Fine particle 4-h averaged S and 12-h averaged
S, sulfate, nitrate and ammonium are all highly correlated with
visual air quality judgments. Later, the deciview (dv) scale was
Y. Li et al. / Journal of Cleaner Production 184 (2018) 959e968964
created to describe the total light extinction capability of all haze
species in the ambient air at a given time at a given location.
From then on, Light scattering extinction (b
ext
or b
xp
)/deciview
(dv) has been used as a primary index to describe air quality in later
research of the evaluation human acceptable air quality (BBC
Research and Consulting, 2003; Fajardo et al., 2013; Pryor, 1996).
4.3.1. Perceptual procedure used in the establishment of visual air
quality standards
As the reliability and validity of photographs judgment has been
proved and widely accepted and also the Light scattering extinction
has been taken as a universal indictor to predict human perceptions
of air quality, since the 1990s the photographs/slide perceptual
procedure has been widely applied to collect observers’judgment
data for socio-politically relevant purposes.
To establish a visibility standard in the Denver metropolitan
area, a study was conducted in 1989 to measure the acceptable
point of visual air quality and its corresponding visual range (Ely
et al., 1991). Slides were taken by camera and, at the same time,
light extinction was measured by a transmissometer. Slides were
selected based on the following criteria: available hourly average
transmissometer value; no extremes of over-exposure or under-
exposure; no snowstorms; no off-centre extremes; and humidity
below 70%. Seventeen groups with a total of 214 observers esti-
mated a total of 160 slides. The report recommends that the group
average violation standard should be at an atmospheric extinction
level (b
ext
) of 0.076/km, which is equivalent to a visual range of
approximately 50 km.
Pryor (1996) set visibility standards in the Lower Fraser Valley
area were set. Again, a camera was used to capture scenic charac-
teristics and open-chamber nephelometers were used to measure
atmospheric clarity. Availability of optical clarity data, humidity,
cloud cover, and sun angle were considered in selecting experi-
mental slides. About 200 university students evaluated the 26
slides showing the same scenes with differing levels of optical
clarity. The results indicate that local rather than regional visibility
standards should be set. In the Lower Fraser Valley, the criteria of
visibility standard (b
ext
) for Chilliwack and Abbotsford should be
0.096e0.105/km (a visual range of approximately 40 km) and
0.039/km (a visual range of approximately 60 km), respectively.
Actual photographs of the same view taken at different times
were used in both the Denver and Fraser Valley studies. Thus, these
Table 2
Developments of perceptual approach in air pollution research.
Authors Developments Main findings
Craik and Zuhe (1976) Psychophysics introduced into landscape research Propose the use of “Perceived Environmental Quality Indices”as an auxiliary to existing
physically based systems for the estimation of various environments.
Boster and Daniel (1972) Developed the method called Scenic Beauty Estimates based on traditional
psychophysical theory.
Balling and Falk (1982) Perceptual approach used in landscape research Results provide limited support for the hypothesis that humans have an innate
preference for savannah-like settings that arises from their long evolutionary history
based on the savannahs of East Africa.
Buhyoff et al. (1983) Hypotheses regarding cultural influences on landscape preference are extended
Malm et al. (1980) Reliability and validity evaluation of visual
perceptions of air quality
Visual perception of air quality is consistent and substantially sensitive to measured air
quality.
Malm et al. (1981) No significant difference between the means of field and slide observations. Colour slide/
photographs can be acceptable as surrogates for actual scenes for perceptual judgments.
Stewart et al. (1983, 1984) Photographic procedure with high reliability and validity can be used to examine the
relation between visual air quality and other variables such as well-being or quality of life
and pollutant concentrations or sources.
Middleton et al. (1985) Judgments of photographs have been highly recommended as a feasible, reduced-cost
substitute for field judgments.
Malm et al. (1980) Modelling of variables influencing visual air quality
from different perceptual cues
High correlations are found between visual air quality and colour contrast and this
relationship is independent of the demographic background.
Latimer and Hogo, 1981 Visual range and inherent scenic character are found to be sensitive to visual air quality.
Malm et al. (1981) Sun angle, colour contrast, inherent scenic beauty and the distance to each of the scenic
elements are sensitive to observers' perceived visual air quality.
Mumpower et al. (1981) Perceptual cues (distance, clarify, colour and border) can describe visual air quality but
not for environmental data, including aerosol scattering and absorption measures,
meteorological measures and pollutant data.
Middleton et al. (1983) Sky colour and border are difficult to explain by emissions. However, refined
concentrations can increase the validity of the model between target clarity and
emissions.
Middleton et al. (1984) Light scattering extinction (b
ext
or b
xp
) by a telephotometer is found to be the most
satisfactory direct indicator of visual air quality.
Ely et al. (1991) Perceptual approach (actual photographs) used to
establish visual air quality standards
The group average violation standard in the Denver area should be at atmospheric
extinction level (b
ext
) of 0.076/km, which is equal to a visual range of approximately
50 km.
Pryor (1996) In Lower Fraser Valley, criteria of visibility standard (b
ext
) for Chilliwack and Abbotsford
should be 0.096e0.105/km (a visual range of approximately 40 km) and 0.039/km (a
visual range of approximately 60 km), respectively.
Malm et al. (1983) Modelling air pollution images Simplified atmosphere model and radiative transfer model are used to de visual air
quality simulation techniques.
Molenar et al. (1994) Using atmosphere aerosol and radiative transfer models, advanced simulation model is
developed.
BBC Research and
Consulting (2003)
Perceptual approach (modelled photographs) used
in establishment of visual air quality standards
24 deciviews is an acceptable level of visual range standard in Phoenix area. Gender and
age are sensitive.
Fajardo et al. (2013) 19 to 33 deciviews is an acceptable range of visual range standard in Beijing for young
people.
Smith (2013) Different people living in different air quality conditions may vary in their opinions of
acceptable levels of visual air quality
Y. Li et al. / Journal of Cleaner Production 184 (2018) 959e968 965
photographs varied not only in light extinction but also in lighting
and cloud conditions. Computer-imaging techniques made it
possible for light extinction to become the sole varying factor in the
photographs rather than other factors, such as cloud cover, sun
angle, precipitation, vista colour, birds, or jet trails. Using atmo-
sphere aerosol and radiative transfer models, Molenar et al. (1994)
developed visual air quality simulation techniques based on the
pioneering work of Malm et al. (1983). From then on, modelled
images have become the major impetus for investigating visual air
quality procedures.
In 2003, BBC Research &Consulting was retained by The Arizona
Department of Environmental Quality to conduct the Phoenix Area
Visibility Survey to assist in the development of a visibility index for
the Phoenix area (BBC Research and Consulting, 2003). Altogether,
385 observers at six separate locations in the Phoenix area evalu-
ated 21 different images with the same scene but with varying
visibility levels. The slides with various visibility levels were
modelled by WinHaze software and ranged from 15 to 35 deciviews
(dv), with “15”and “35”representing the clearest and the least
clear visual air quality. The results indicate that 24 deciviews is an
acceptable level for the local visual range standard. Gender and age
are sensitive to the 50% acceptable level of visual air quality.
In 2013, a similar study was conducted in the Beijing area
(Fajardo et al., 2013). Eighty-five young people between 15 and 18
years old participated in this experimental survey. Twenty photo-
graphs were simulated by WinHaze software with a range of 15e51
deciviews. The investigated haziness range was designed based on
the local air quality state. The 50% acceptable level of visual air
quality for young people in Beijing is 19e33 deciviews.
Smith (2013) evaluated previously established visual air quality
standards by adopting the previous methods under various speci-
fied conditions. He argues that the different acceptable standards
recommended by studies may result from the images shown in
judgments with different visual range, which indicates that
different people living in different air quality conditions may vary
in their expectations and that local, rather than regional, visibility
standards should be set.
Therefore, photographs judgments based on the perceptual
approach should be valid to examine the relationship between SWB
and air pollution emissions. This psychophysical procedural has
been successfully used to solve other issues between environment
and people's perceptions, like landscape beauty evaluation and
establishment of visual air quality standard. Thus, traditional psy-
chophysical experiments and analysis methods can be used to
collect observers' judgments and quantify the observers' percep-
tion on the SWB continuum.
5. Discussion and conclusion
This paper evaluates the advantages and limitations of tradi-
tional surveys and psychophysics in quantifying the effects of air
pollution's impacts. Survey data fully addresses SWB perception
from all senses, and it is a sort of field survey to a certain degree.
The shortages of survey data are obvious. Firstly, it takes long time
and is hard to match air pollution characteristics from monitor
stations with each respondent's state of SWB at the moment a
survey is conducted. A conventional challenge to empirical studies
is to seek high quality air pollution data with fine spatial and
temporal disaggregation and to connect this information with a
specific respondent. Most of previous researchers have typically
merged the average annual air pollution data at the country or
regional level with collected or ready-to-use panel SWB data from
surveys to analyse the regression correlations between the two
datasets. Secondly, the survey method is under the constraint of
local air quality. Neither an averagely comparative good nor poor air
condition is enough to explain a full picture of air quality impact on
SWB. Thirdly, huge data collection is needed to control the influ-
ence on SWB from socio-economic, demographic and weather or
climate variables. To control observers' idiosyncrasies, empirical
researchers made efforts to take as many as possible variables,
which are potential to influence SWB in to consideration. Fourthly,
the air pollution data used in empirical studies are in a relatively
narrow range with European environmental data so that it is quite
hard to predict the SWB states of people who often expose in more
serious air pollution environment in developing countries like India
and China.
Of course, photographic psychophysical procedures only pro-
vide measures in a visual sense and it is not sufficient to describe
the SWB state of a person. However, it is a good tool to examine the
impact of visual air pollution on SWB and worth to try. Sight
dominates the way we ‘see’the world and visual input has a pri-
mary role in brain information processing. Photograph judgment
has been highly recommended as a feasible, cost-effective substi-
tute for field judgments. Compared with surveys, it costs less and
removes observers' idiosyncrasies easily without collecting socio-
economic or demographic data. It even generates high quality re-
sults without a large number of observers. For example, just over
ten observers were used in Smith (2013). Most importantly, it
makes it possible to specify spatial and temporal conditions by
examining observers' SWB states when viewing the various pho-
tographs and obtaining the environmental data from when the
photographs were taken.
Taking psychophysical procedures can improve people's un-
derstanding of air pollution effect on SWB. It is essential to build an
air concentrations image index with high-resolution images in
good format. This should be achievable in countries like China,
where air quality shows a wide range from excellent to hazardous
and air pollutants data are more reliable. With hourly environ-
mental data, the corresponding details such as various air pollutant
concentrations, sun angle, humidity, temperature, wind angle and
speed for each photograph can be obtained. Modelling the rela-
tionship between hourly environmental data and pixel colours in
photographs is the beginning step. With the model and given
weather variables, an image-based index with weather conditions
control explaining various air pollution levels can be built. Using
this image-based index, lots of psychophysical research on air
pollution and SWB can be conducted with ease. Studies could be
quantifying the impacts of different levels of air pollution on SWB.
Studies could evaluate the sensitivity of SWB of different social
groups under the impact of air pollution. The air conditions of
different cities can also be modelled and shown in images so that
air condition impact of different cities can be estimated. With help
of colour imaging technology, the psychophysical procedure can
also be adopted to solve the other environment and perception
issues.
Acknowledgements
This work was supported by the National Key R&D Program of
China (2016YFA0602604), the Natural Science Foundation of China
(41629501), the UK Economic and Social Research Council (ES/
L016028/1), Natural Environment Research Council (NE/N00714X/
1), British Academy Grant (AF150310), and the Philip Leverhulme
Prize.
References
Ambrey, C.L., Fleming, C.M., Chan, A.Y.C., 2014. Estimating the cost of air pollution in
South East Queensland: an application of the life satisfaction non-market
valuation approach. Ecol. Econ. 97 (1), 172e181.
Y. Li et al. / Journal of Cleaner Production 184 (2018) 959e968966
Andrews, F.M., Withey, S.B., 1974. Developing measures of perceived life quality:
results from several national surverys. Soc. Indic. Res. 1 (1), 1e26.
Andrews, F.M., Withey, S.B., 1976. Social Indicators of Well-being: America's
Perception of Life Quality. Plenum Press, New York.
Balling, J.D., Falk, J.H., 1982. Development of visual preference for natural envi-
ronments. Environ. Behav. 14 (1), 5e28.
Barker, M.L., 1976. Planning for environmental indices: observer appraisals of air
quality. In: Craik, K.H., Zube, E.H. (Eds.), Perceiving Environmental Quality:
Research and Applications. Plenum, New York.
Baum, A., Deckel, A.W., Gatchel, R.J., 1982. Environmental stress and health: is there
a relationship? In: Sanders, G.S., Suls, I. (Eds.), Social Psychology of Health and
Illness. Lawrence Erlbaum Associates Publishers, Hillsdale, New Jersey.
Baum, A., Singer, I.E., 1986. Advances in Environmental Psychology. Lawrence Erl-
baum and Associates Publishers, Hillsdale New Jersey.
BBC Research &Consulting, 2003. Phoenix area Visibili Ty Survey. Arizona
department of environmental quality &BBC research &consulting.
Boster, R.S., Daniel, T.C., 1972. Measuring public responses to vegetative manage-
ment. In: 16 Ann. Arizona Watershed Symp. Arizona Water Commission,
Phoenix, pp. 38e43.
Bradburn, N.M., 1969. The Structure of Psychological Well-being. Aldine, Chicago.
Brook, R.D., 2008. Cardiovascular effects of air pollution. Clin. Sci. 115 (6), 175e187.
Brunekreef, B., Holgate, S.T., 2002. Air pollution and health. Lancet 360 (9341),
1233e1242.
Bryant, F.B., Veroff, J., 1982. The structure of psychological well-beling: a sociohis-
torical analysis. J. personality Soc. Psychol. 43, 653e673.
Buhyoff, G.J., Wellman, J.D., Koch, E.E., Gauthier, L., Hultman, S., 1983. Landscape
preference metrics: an international comparison. J. Environ. Manag. 16 (2),
181 e190.
Bullinger, M., 1989. Psychological effects of air pollution on healthy residents-a
time-series approach. J. Environ. Psychol. 9 (2), 103e118 .
Campbell, J.M., 1983. Ambient stressors. Environ. Behav. 15 (3), 355e380.
Cantril, H., 1965. The Pattern of Human Concerns. NJ: Rutgers University Press, New
Brunswick.
Chekola, M.G., 1975. The Concept of Happiness. University of Michigan.
Coan, R., 1977. Hero, Artist, Sage, or Saint?, first ed. Columbia University, New York.
Cohen, S., 1980. Cognitive processes as determinants of environmental stress. In: I.S,
Spielberger, C. (Eds.), Stress and Anxiety. Hemisphere Press, Washington D.C.
Cohen, S., Doyle, W.J., Turner, R.B., Alper, C.M., Skoner, D.P., 2003. Emotional style
and susceptibility to the common cold. Psychosom. Med. 65 (4), 652e657.
Cohen, S., Evans, G.W., Stokols, D., Krantz, D.S., 1986. Behavior, Health, and Envi-
ronmental Stress. Plenum, New York.
Costa, P.T.J., Zonderman, A.B., McCrae, R.R., Cornoni-Huntley, J., Locke, B.Z.,
Barbano, H.E., 1987. Longitudinal analyses of psychological well-being in a na-
tional sample: stability of mean levels. J. Gerontol 42 (1), 50e55.
Craik, K.H., Zuhe, E.H., 1976. Perceiving Environmental Quality: Research and Ap-
plications. Plenum Press, New York.
Cummins, R.A., Eckersley, R., Pallant, J., Van Vugt, J., Misajon, R., 2003. Developing a
national index of subjective wellbeing: the Australian unity wellbeing index.
Soc. Indic. Res. 64 (2), 159e190 .
Daniel, T.C., Boster, R.S., 1976. Measuring Landscape Esthetics: the Scenic Beauty
Estimation Method. USDA Forest Service, U.S. Department of Agriculture.
Danner, D., Snowdon, D., Friesen, W., 2001. Positive emotions in early life and
longevity: findings from the Nun Study. J. Personality Soc. Psychol. 80, 804e813.
Di Tella, R., MacCulloch, R., Oswald, A., 2001. Preferences over in ation and unem-
ployment: evidence from surveys of happiness. Am. Econ. Rev. 91 (1), 335e341.
Di Tella, R., MacCulloch, R., Oswald, A., 2003. The macroeconomics of happiness.
Rev. Econ. Statistics 85 (4), 809e827.
Diener, E., 1984. Subjective well-being. Psychol. Bull. 95 (3), 542e57 5.
Diener, E., 2006. Guidelines for national indicators of subjective well-being and ill-
being. Res. Qual. Life 20 06 (1), 151e157.
Diener, E., Lucas, R.E., Scollon, C.N., 2006. Beyond the hedonic treadmill: revising
the adaptation theory of well-being. Am. Psychol. 61 (4), 305e314.
Easterlin, R., 2003. Explaining happiness. PNAS 100 (19), 11176e1118 3.
Ely, D.W., Leary, J.T., Stewart, T.R., Ross, D.M., 1991. The establishment of the Denver
visibility standard. In: 84th Annual Meeting &Exhibition. Air &Waste Man-
agement Association, Vancouver.
Engeldrum, P.G., 2000. Psychometric Scaling a Toolkit Imaging Systems Develop-
ment, first ed. Imcotek Press, Winchester.
Fajardo, O.A., Jiang, J., Hao, J., 2013. Assessing young people's preferences in urban
visibility in Beijing. Aerosol air Qual. Res. 13 (5), 1536e1543.
Ferreira, S., Akay, A., Brereton, F., Cu~
nado, J., Martinsson, P., Moro, M., Ningal, T.F.,
2013. Life satisfaction and air quality in Europe. Ecol. Econ. 88, 1e10.
Ferreira, S., Moro, M., 2010. On the use of subjective well-being data for environ-
mental valuation. Environ. Resour. Econ. 46 (3), 249e273.
Folkman, S., Schaefer, C., Lazarus, R.S., 1979. Cognitive processes as mediators of
stress and coping. In: Hamilton, V., Warburton, D.M. (Eds.), Human Stress and
Cognition: an Information-processing Approach. Wiley, London.
Frankenhaeuser, M., 1980. Psychoneuroendocrine approaches to the study of
stressful person-environment transaction. In: Selye, H. (Ed.), Selye's Guide to
Stress Research. Van Nostrand Reinhold, New York.
Fujiwara, D., Campbell, R., 2011. In: Pensions, U.H.a.D.f.a. (Ed.), Valuation Techniques
for Social Cost-benefit Analysis: Stated Preference,revealed Preference and
Subjective Well-being Approaches.
Gerdtham, U., Johannesson, M., 2001. The relationship between happiness, health,
and socioe- conomic factors: results based on swedish microdata, 30 J. Socio-
Economics 30 (6), 553e557.
Gilbert, J.G., Walsh, V., 2004. Vision: the versatile ‘visual’cortex. Curr. Biol. 14,
1056e1057.
Green, D.M., Swetts, J.A ., 1966. Signal Detection Theory and Psychophysics. John
Wiley, New York.
Gurin, G., Veroff, J., Feld, S., 1960. Americans Review Their Mental Health. Basic
Books, New York.
Harding, S.D., 1982. Psychological well-being in great Britain: an evaluation of the
Bradburn affect balance scale. Personality Individ. Differ. 3 (2), 167e175.
Hyslop, N.P., 2009. Impaired visibility: the air pollution people see. Atmos. Environ.
43 (1), 182e195.
Kahneman, D., Kreuger, A.B., Schkade, D., Schwartz, N., Stone, A., 2004. Towards
national well-being accounts. Am. Econ. Rev. 94 (2), 429e434.
Kail, R.V., Cavanaugh, J.C., 2015. Essentials of Human Development: a Life-span
View, second ed. Cengage Learning, Boston.
Kirk, E.C., 2006. Visual influences on primate encephalization. J. Hum. Evol. 51 (1),
76e90.
Knabe, A., Ratzel, S., 2010. Better an insecure job than no job at all? unemployment,
job insecurity and subjective wellbeing. Econ. Bull. 30 (3), 2486e2494.
Kozma, A., Stones, M.J., 1980. The measurement of happiness: development of the
memorial university of newfoundland scale of happiness (MUNSH). J. Gerontol
35 (6), 906e912.
Latimer, D.A., Hogo, H., 1981. The effects of a tmospheric optical conditions on
perceived scenic beauty. Atmos. Environ. 15, 10e11.
Lawton, M.P., 1975. The philadelphia geriatric center morale scale. J. Gerontol 30 (1),
85e89.
Lazarus, R.S., Cohen, I., 1978. Environmental Stress. In: Altman, I., Wohlwill, I. (Eds.),
Human Behavior and Environment. Plenum, New York.
Levinson, A., 2012. Valuing public goods using happiness data: the case of air
quality. J. Public Econ. 96 (9e10), 869e880.
Li, Z., Folmer, H., Xue, J., 2014. To what extent does air pollution affect happiness?
The case of the Jinchuan mining area, China. Ecol. Econ. 99 (1), 88e99.
Liu, G., Liu, H., 2011. Using insulation in China's buildings: potential for significant
energy savings and carbon emission reductions. Low. Carbon Econ. 2 (4),
220e223.
Lucas, R.E., Clark, A.E., Georgellis, Y., Diener, E., 2003. Reexamining adaptation and
the set point model of happiness: reactions to changes in marital status.
J. Personality Soc. Psychol. 84 (3), 527e539.
Luechinger, S., 2009. Valuing air quality using the life satisfaction approach. Econ. J.
119 (536), 482e515.
Luechinger, S., 2010. Life satisfaction and transboundary air pollution. Econ. Lett.
107 (1), 4e6.
Lykken, D., Tellegen, A.,1996. Happiness is a stochastic phenomenon. Psychol. Sci. 7
(3), 180e189.
MacKerron, G., Mourato, S., 2009. Life satisfaction and air quality in London. Ecol.
Econ. 68 (5), 1441e1453.
Malm, W., Kelley, K., Molenar, J., Daniel, T., 1981. Human perception of visual air
quality (uniform haze). Atmos. Environ. 15 (10), 1875e1890.
Malm, W., Molenar, J., Chan, L.L., 1983. Photographic simulation techniques for
visualizing the effect of uniform haze on a scenic resource. J. air Pollut. control
Assoc. 33 (2), 126e129.
Malm, W.C., Leiker, K.K., Molenar, J.V., 1980. Human perception of visual air quality.
J. air Pollut. control Assoc. 30 (2), 122e131.
Marks, L.E., Gescheider, G.A., 2002. Phychophysical scaling. In: Pashler, H.E.,
Atkinson, R.D. (Eds.), The Stevens' handbook of experimental psychology. Wiley,
Chichester.
Meertens, R., Swaen, G., 1997. Psychological Factors in Air Pollution Epidemiology,
Air Pollution Epidemiology Report Series, pp. 105e112.
Middleton, P., Stewart, T.R., Ely, D., 1984. Physical and chemical indicators of urban
vair quality judgments. Atmos. Environ. 18 (4), 861e870.
Middleton, P., Stewart, T.R., Leary, J., 1985. On the use of human judgment and
physical/chemical measurements in visual air quality management. J. air Pollut.
control Assoc. 35 (1), 11e18.
Molenar, J.V., Malm, W.C., Johnson, C.E., 1994. Visual air quality simulation tech-
niques. Atmos. Environ. 28 (5), 1055e1063.
Mumpower, J., Middleton, P., Dennis, R.L., Stewart, T.R., Veirs, V., 1981. Visual air
quality assessment: Denver case study. Atmos. Environ. 15 (12), 2433e2441.
OECD, 2013. Guidelines on Measuring Subjective Well-being.
Orru, K., Orru, H., Maasikmets, M., Hendrikson, R., Ainsaar, M., 2016. Well-being and
environmental quality: does pollution affect life satisfaction? Qual. Life Res. 25
(3), 699e705.
Ostir, G.V., Markides, K.S., Peek, M.K., Goodwin, J.S., 2001. The association between
emotional well-being and the incidence of stroke in older adults. Psychosom.
Med. 63, 210e215.
Middleton, Paulette, S, T.R., Dennis, Robin L., 1983. Modeling human judgments of
urban visual air quality. Atmos. Environ. 17 (5), 1015e1021.
Peterson, G.L., 1967. A model of preference: quantitative analysis of the perception
of the visual appearance of residential neighborhoods. J. Regional Sci. 7 (1),
19e31.
Peterson, G.L., Neumann, E.S., 1969. Modeling and predicting human response to
the visual recreation environment. J. Leis. Res. 1, 219e237.
Propst, D.B., Buhyoff, G.J., 1980. Policy capturing and landscape preference quanti-
fication: a methodological study. J. Environ. Manag. 11 (1), 45e49.
Pryor, S.C., 1996. Assessing public perception of visibility for standard setting ex-
ercises. Atmos. Environ. 30 (15), 2705e2716.
Y. Li et al. / Journal of Cleaner Production 184 (2018) 959e968 967
Rehdanz, K., Maddison, D., 2008. Local environmental quality and life-satisfaction in
Germany. Ecol. Econ. 64 (4), 787e797.
Rotten, J., Frey, J., 1984. Psychological costs of air pollution: atmospheric conditions,
seasonal trends, and psychiatric emergencies. Popul. Environ. Behav. Soc. Issues
7 (1), 3e16.
Rotton, J., 1983. Affective and cognitive consequences of malodorous pollution.
Basic Appl. Soc. Psychol. 4 (2), 171e193.
Schmitt, M., 2013. Subjective well-being and air quality in Germany. Schmollers
Jahrb. 133 (2), 275e286.
Shin, D.C., Johnson, D.M., 1978. Avowed happiness as an overall assessment of the
quality of life. Soc. Indic. Res. 5 (1), 475e492.
Smith, A.E., 2013. An evaluation of the robustness of the visual air quality “pref-
erence study”method. J. air &waste Manag. Assoc. 63 (4), 405e417.
Smyth, R., Mishra, V., Qian, X., 2008. The environment and well-being in urban
China. Ecol. Econ. 68 (1e2), 547e555.
Steptoe, A., Wardle, J., Marmot, M., 2005. Positive affect and health-related
neuroendocrine, cardiovascular, and inflammatory processes. In: Proceedings
of the National Academy of Sciences of the United States of America,
pp. 6508e6512.
Stewart, T.R., Middleton, P., Downton, M., Ely, D., 1984. Judgments of photographs
vs. field observations in studies of erception and judgment of the visual envi-
ronment. J. Environ. Psychol. 4 (4), 283e302.
Stewart, T.R., Middleton, P., Ely, D., 1983. Urban visual air quality judgments: reli-
ability and validity. J. Environ. Psychol. 3 (2), 129e145 .
Strahilevitz, M., Strahilevitz, A., Miller, I.E., 1979. Air pollutants and the admission
rate of psychiatric patients. Am. J. Psychiatry 136 (2), 205e207.
Swetts, J.A., 1973. The relative operating characteristic in psychology. Science 182
(4116), 990e1000.
Tatarkiewicz, W., 1976. Analysis of Happiness. Martinus Nijhoff, The Hague,
Netherlands.
Theodossiou, I., 1998. The effects of low-pay and unemployment on psychological
well-being: a logistic regression approach. J. Health Econ. 17 (1), 85e104.
Thurstone, L.L., 1927. A law of comparative judgement. Psychol. Rev. 34, 273e286.
Torgerson, W.S., 1958. Theory and Methods of Scaling. John Wiley, New York.
Weir, K., 2012. Smog in our brains. American Psychol. Assoc. 43 (7), 32e35.
Welsch, H., 2002. Preferences over prosperity and pollution: environmental valu-
ation based on happiness surveys. Kyklos 55 (4), 473e494.
Welsch, H., 2007. Environmental welfare analysis: a life satisfaction approach. Ecol.
Econ. 62 (3e4), 544e551.
WHO, 2016. Ambient Air Pollution: a Global Assessment of Exposure and Burden of
Disease. World Health Organization, pp. 40e42.
Zevon, M.A., Tellegen, A., 1982. The structure of mood change: an idiographic/
nomothetic analysis. J. personality Soc. Psychol. 43, 111e122.
Zhang, X., Zhang, X., Chen, X., 2017. Happiness in the air: how does a dirty sky affect
mental health and subjective well-being? J. Environ. Econ. Manag. 85, 81e94.
Zube, E.H., 1974. Cross-disciplinary and intermode agreement on the description
and evaluation of landscape resources. Environ. Behav. 6 (1), 69e89.
Y. Li et al. / Journal of Cleaner Production 184 (2018) 959e968968
