Available via license: CC BY 4.0
Content may be subject to copyright.
sustainability
Concept Paper
Rethinking Performance Gaps: A Regenerative
Sustainability Approach to Built Environment
Performance Assessment
Sylvia Coleman 1, *, Marianne F. Touchie 2, John B. Robinson 3,4 and Terri Peters 5
1Sustainable Built Environment Performance Assessment Network, The John H. Daniels Faculty of
Architecture, Landscape and Design, University of Toronto, Toronto, ON M5S 2J5, Canada
2Civil & Mineral Engineering and Mechanical & Industrial Engineering, University of Toronto, Toronto,
ON M5S 1A4, Canada; marianne.touchie@utoronto.ca
3Munk School of Global Affairs & Public Policy, University of Toronto, Toronto, ON M5S 3K7, Canada;
johnb.robinson@utoronto.ca
4School of the Environment, University of Toronto, Toronto M5S 3K7, Canada
5Department of Architectural Science, Ryerson University, Toronto, ON M5S 1A4, Canada;
Terri.Peters@Ryerson.ca
*Correspondence: sylvia.coleman@utoronto.ca
Received: 7 November 2018; Accepted: 14 December 2018; Published: 18 December 2018
Abstract:
Globally, there are significant challenges to meeting built environment performance targets.
The gaps found between the predicted performance of new or retrofit buildings and their actual
performance impede an understanding of how to achieve these targets. This paper points to
the importance of reliable and informative building performance assessments. We argue that if
we are to make progress in achieving our climate goals, we need to reframe built environment
performance with a shift to net positive goals, while recognising the equal importance of human and
environmental outcomes. This paper presents a simple conceptual framework for built environment
performance assessment and identifies three performance gaps: (i) Prediction Gap (e.g., modelled
and measured energy, water consumption); (ii) Expectations Gap (e.g., occupant expectations in pre-
and post-occupancy evaluations); and, (iii) Outcomes Gap (e.g., thermal comfort measurements and
survey results). We question which of measured or experienced performance is the ‘true’ performance
of the built environment. We further identify a “Prediction Paradox”, indicating that it may not
be possible to achieve more accurate predictions of building performance at the early design stage.
Instead, we propose that Performance Gaps be seen as creative resources, used to improve the
resilience of design strategies through continuous monitoring.
Keywords:
performance gap; gap analysis; regenerative buildings; post-occupancy evaluation
(POE); pre-occupancy evaluation; qualitative assessment; quantitative assessment; occupant-centred
approach; continuous monitoring; interactive adaptivity
1. Introduction
The climate change goals being adopted by national and subnational jurisdictions around the
world imply the need for substantial reductions in energy use in the buildings sector [
1
–
3
]. At the
COP23 conference in Bonn, Germany in November 2017, mayors from 25 cities around the world,
representing 150 million citizens, pledged to cut their carbon emissions to net zero by 2050 [
4
]. At the
same meeting, the Global Covenant of Mayors for Climate and Energy (representing 7494 cities
worldwide) released a report indicating that the potential for reduction by 2030 by these Global
Sustainability 2018,10, 4829; doi:10.3390/su10124829 www.mdpi.com/journal/sustainability
Sustainability 2018,10, 4829 2 of 22
Covenant cities is nearly 1.3 billion tons of CO2e emissions per year. (Current emissions in those cities
are about 3.5 billion tons.) [5]
The implications of these and other targets for the buildings sector are immense, and have perhaps
not been fully recognized in the building community. As one illustrative example, in 2007, the City
of Toronto, Canada’s largest city, with a population of 2.8 million people, adopted a target of an
80% reduction in GHG emissions from 1992 levels by 2050 (City of Toronto, 2007). In July 2017, City
Council approved more specific targets as part of the new TransformTO climate policy [
6
,
7
], including:
100 percent of new buildings are to be designed and built to be near zero greenhouse gas emissions
by 2030; and 100 percent of existing buildings are to be retrofitted to the highest emission reduction
technically feasible, on average achieving a 40 percent energy performance improvement over 2017
levels, while limiting affordability impacts to residents, by 2050. In the modelling work accompanying
the TransformTO program, the number of residential retrofits required over the period between now
and 2030 to achieve these targets is about 24,000 units per year [8].
There are two challenges in achieving the kinds of results mandated in the City of Toronto and
similar climate plans: achieving the scale of activity required; and ensuring that that activity actually
achieves the required savings or performance. On the question of scale, we need to increase, by one or
two orders of magnitude, the proportion of new buildings and of retrofits. This will be a major task, as
these proportions are currently modest in most jurisdictions.
On the question of performance, there is a growing literature about the existence and causes of a
significant gap between the predicted and actual performance of new and retrofit buildings (usually
defined in terms of energy use), reviewed in detail in Section 3.1 below. Here, we simply wish to
highlight the critical importance of reliable and informative building performance assessments if we
are to address that gap and make significant progress in achieving our climate goals.
This paper concerns itself with performance assessment, though we believe our arguments
also have implications for the problem of scaling up the number of new builds and retrofits that
adopt ambitious sustainability goals. We propose a two-pronged approach to address building
performance issues.
The first part of our proposed approach to performance assessment involves reframing the goals
of building performance. We reframe these goals in terms of broadening the focus of performance
assessment: (i) to move beyond a sole emphasis on energy performance to a broader sustainability
focus, recognizing the equal importance of both human and environmental outcomes; (ii) to move
beyond net zero to net positive approaches; and (iii) to move beyond a focus on individual buildings,
by incorporating neighbourhood-scale built environment systems. (While we consider the third,
neighbourhood-scale component of our proposed reframing essential, it will not be explored in
this paper).
The second part of our proposed approach follows from this reframing, and involves integrating
the qualitative and quantitative dimensions of performance assessment over time in order to address
both human and environmental outcomes.
This dual approach—reframing the goals of building performance, and integrating the
environmental and human dimensions of performance assessment over time—gives rise to a simple
conceptual framework, which is described below in Section 3. The conceptual framework then allows
the identification of three performance gaps, which we further discuss in Section 3as well:
•Predicted versus actual resource use (e.g., modelled and measured energy, water consumption);
•
Expectations regarding the performance of sustainable buildings versus the actual lived experience
of the building occupants (e.g., pre- and post-occupancy evaluations);
•
Measured performance versus lived experience (e.g., thermal comfort measurements and
survey results).
Sustainability 2018,10, 4829 3 of 22
2. Reframing the Performance of the Built Environment: A Regenerative Sustainability Approach
Performance assessment has typically been approached by the building industry in terms of
the goals of minimizing the negative environmental impacts of the buildings (e.g., by reducing
energy use or emissions), and to a much lesser degree, of improving the comfort and wellbeing
of building occupants. Where these two goals are pursued on a single project, they have tended
to be addressed somewhat separately, with the former being addressed through modeling and the
quantitative measurement of environmental performance, and the latter having mainly had to do with
the qualitative post-occupancy assessment of the views of building occupants.
Both the public identity and the marketing of high performance buildings has strongly centred on
environmental measures (e.g., ‘green buildings’), emphasizing reducing their impacts (e.g., ‘net zero
energy’ or ‘near zero emission’ buildings). The human performance side is rarely emphasized in these
materials, and the expressed goal is to avoid negative environmental impacts.
To date, this approach has not been very successful: although the numbers are growing, green
buildings remain a small fraction of new builds and retrofits in both the Residential and Commercial
sectors. For example, as of 2014, the number of LEED-certified buildings across building types amount
to 10% of new buildings in Canada, up from 0.8% from 2004–2009 [9].
What might be more useful is to modify both approaches as follows:
•The initial focus should be on human wellbeing (e.g., health, productivity and happiness)
•
To the extent possible, buildings should be designed to be net positive in both human and
environmental terms.
We call this approach regenerative sustainability [10].
While adopting a net positive approach to building design raises a number of conceptual and
practical difficulties [
10
,
11
], it is starting to be applied in practice [
12
,
13
]. The Living Building Challenge
of the International Living Future Institute [
14
] is an example of a building certification scheme based
on net positive principles. One regenerative sustainability building that has been the subject of much
study is the Centre for Interactive Sustainability at the University of British Columbia in Canada.
Research indicates that the net positive environmental goals are less easily achieved than the human
goals [
15
–
17
]. A crucial consideration is that regenerative approaches are systems-based and are
characterized by inherently unpredictable emergent properties, thus exhibiting levels of complexity
that are difficult to measure and incorporate in practice. This suggests a need to emphasize process
outcomes over performance outcomes [10].
A critical advantage of a regenerative sustainability approach to the built environment is that
many of the design strategies that address human wellbeing (e.g., natural light, air quality, thermal
comfort, natural materials) are essentially the same strategies that deliver environmental performance
and climate goals. In other words, a focus on human wellbeing brings many aspects of environmental
performance along for the ride. A critical question becomes where the two sets of goals overlap and
reinforce each other, where they are independent, and where they might be in conflict.
A second advantage of this approach is that a focus on improving both human and environmental
wellbeing is much more interesting to purchasers, leasers, developers and perhaps even designers
than simply reducing damage.
For example, while much literature on the advantages of sustainable buildings focuses on potential
energy savings, labour costs per square metre of office buildings are much higher than the energy cost
per square metre. As a result, the economic savings from a small improvement in labour productivity
for most occupants of office buildings would negate the savings from even significantly improved
energy efficiency. Since labour costs are a familiar and important part of the economic calculus of
virtually all companies, if such productivity improvements could be reliably calculated, the resultant
savings would likely be much more influential in determining office space lease rates than less familiar
and smaller energy savings. More generally, a building that will make people healthier, happier
and more productive, is likely to be more marketable than one that simply does less environmental
Sustainability 2018,10, 4829 4 of 22
damage than typical buildings. However, as discussed in more detail below, this depends critically on
developing strong metrics of human wellbeing that can be tied to conditions in the built environment.
The combination of these two factors: (i) broadening beyond environmental performance to a
wider focus on both human and environmental wellbeing; and (ii) moving beyond harm reduction
and net zero goals to net positive approaches, offers the potential for developing a regenerative
sustainability approach that may contribute to a sustainable building industry that can be much
more successful than in the past in achieving the scale and effectiveness required to meet society’s
ever-more-ambitious sustainability goals.
3. Proposed Integration of Quantitative and Qualitative Performance Assessment over Time
In order to achieve the potential advantages of a regenerative sustainability approach to the built
environment, we need to develop systems of performance assessment that allow us to evaluate the
potential for net positive outcomes in both human and environmental terms. We propose an approach
to built environment performance assessment based on the following simple conceptual framework.
The four quadrants of Figure 1show the kinds of analysis typically undertaken for new or
retrofitted buildings. The top two quadrants are the realm of mostly quantitative analysis, typically of
environmental systems such as energy or water, through instrumented measurements. At the design
stage, use is frequently made of building modeling processes that predict the expected performance of
the building, or the retrofit, while once the building is built, monitoring processes provide assessments
of actual performance. The gap between the two is the well-known performance gap referenced above.
Sustainability 2018, 10, x FOR PEER REVIEW 4 of 21
depends critically on developing strong metrics of human wellbeing that can be tied to conditions in
the built environment.
The combination of these two factors: (i) broadening beyond environmental performance to a
wider focus on both human and environmental wellbeing; and (ii) moving beyond harm reduction
and net zero goals to net positive approaches, offers the potential for developing a regenerative
sustainability approach that may contribute to a sustainable building industry that can be much more
successful than in the past in achieving the scale and effectiveness required to meet society’s ever-
more-ambitious sustainability goals.
3. Proposed Integration of Quantitative and Qualitative Performance Assessment over Time
In order to achieve the potential advantages of a regenerative sustainability approach to the built
environment, we need to develop systems of performance assessment that allow us to evaluate the
potential for net positive outcomes in both human and environmental terms. We propose an
approach to built environment performance assessment based on the following simple conceptual
framework.
The four quadrants of Figure 1 show the kinds of analysis typically undertaken for new or
retrofitted buildings. The top two quadrants are the realm of mostly quantitative analysis, typically
of environmental systems such as energy or water, through instrumented measurements. At the
design stage, use is frequently made of building modeling processes that predict the expected
performance of the building, or the retrofit, while once the building is built, monitoring processes
provide assessments of actual performance. The gap between the two is the well-known performance
gap referenced above.
Figure 1. Sustainable Built Environment Performance Assessment Framework
In contrast to environmental performance assessment, the assessment of human systems in
buildings is usually conducted qualitatively (bottom two quadrants), through methods focused on
occupants using input and feedback methods that include surveys and interviews. It is much less
frequent, consisting mainly of episodic post-occupancy or post-retrofit evaluations.
A Post-Occupancy Evaluation (POE) provides information on how a building functions once it’s
been built, in terms of if and how the building meets stakeholder and user expectations, often with
Figure 1. Sustainable Built Environment Performance Assessment Framework.
In contrast to environmental performance assessment, the assessment of human systems in
buildings is usually conducted qualitatively (bottom two quadrants), through methods focused on
occupants using input and feedback methods that include surveys and interviews. It is much less
frequent, consisting mainly of episodic post-occupancy or post-retrofit evaluations.
A Post-Occupancy Evaluation (POE) provides information on how a building functions once it’s
been built, in terms of if and how the building meets stakeholder and user expectations, often with
regards to user satisfaction with the building environment [
18
]. There is substantial literature about
POE and POE cases, and researchers have developed a range of strategies for their use [19–22].
Sustainability 2018,10, 4829 5 of 22
In retrofits, Pre-OEs are not uncommon [
23
], but for new builds, Pre-OEs are not typically
carried out, meaning that there is usually no baseline for post-occupancy evaluations. Moreover, the
quantitative performance of environmental systems is rarely compared to occupant experience of
these systems, e.g., measured air quality and experienced air quality (For two recent exceptions, see
Chang and Touchie (2017) and Touchie et al. (2016)). However, there is growing interest in these,
with some published studies [
24
–
29
]. Researchers have acknowledged the need for pre-occupancy
evaluations in health care environments, for example, as there is an increasing focus on evidence-based
design and a desire for more scientific bases for design decisions [
30
]. In their manifesto for “building
human agency”, Cole, Brown and MacKay propose a directive that addresses pre- and post-occupancy
evaluations: “Pre- and post- occupancy evaluations in new and existing buildings should become
mandatory steps within the integrated design process to accelerate our understanding of the systemic
inhabitants-architecture interactions” [
31
]. Section 3.2 discusses some relevant survey and interview
Pre-OE and POE methods and questions in more depth.
Figure 1provides a basis for articulating three performance gaps in the built environment,
introduced above and indicated by circled letters on the diagram:
A.
Prediction Gap: Predicted versus actual resource use (e.g., modelled and measured energy,
water consumption);
B.
Expectations Gap: Expectations regarding the performance of sustainable buildings versus the
actual lived experience of the building occupants (e.g., pre- and post-occupancy evaluations);
C.
Outcomes Gap: Measured performance versus lived experience (e.g., thermal comfort
measurements and survey results).
The following sections of this paper discuss these gaps in more detail.
3.1. Prediction Gap: Predicted vs. Actual Performance
The most well-documented performance gap in sustainable buildings is the quantitative
discrepancy between the predicted and the actual performance of Environmental Systems such as
energy, water, or carbon emissions. We are calling this discrepancy the Prediction Gap, and distinguish
it from the other two Gaps that we explore below: it is typically measured in terms of energy use, and
it is straightforward to quantitatively measure and compare calculations at the design or pre-retrofit
stage to energy bills or measured performance on site.
Many published studies have shown that sustainable buildings do not perform as expected [
32
–
34
].
Typically, they perform worse than expected, but sometimes they do better than their targets for
designed energy use intensity [
35
]. Newsham et al. [
36
] found in a re-analysis of LEED certified
office buildings that, generally, they perform better than non-LEED certified buildings, but that at an
individual level results varied widely between designed and measured performance. In a study of
66 Canadian university buildings, Storey [
37
] found no correlation between LEED certification and
energy performance. And in a study of nine high performing buildings in Canada, Bartlett et al. [
38
]
found that three of the nine had actual energy use significantly higher than predicted.
3.1.1. Assessing the Prediction Gap
This section outlines findings from recent studies of the cause of the Prediction Gap, in particular
modeling inaccuracies, assumptions about user behaviour, organizational and disciplinary silos and
lack of comprehensive studies.
Modelling inaccuracy and, therefore, incorrect building assumptions are often blamed for the
Prediction Gap. Common reasons are:
•
simulation tools tend to be incomplete in their representation of energy loads related to specific
areas or systems in the building, especially when a project uses newer design strategies such as
natural ventilation or advanced renewable energy and water systems which present challenges
for modeling software;
Sustainability 2018,10, 4829 6 of 22
•
energy models struggle to account for the actual/future usage of a building and seldom account
for occupancy schedules and levels;
•accurate representations of occupant behaviour are not part of the typical modelling practices;
•
weather files provide historic and therefore inaccurate weather data used in the
simulations [39]; and
•
studies have shown design-assist and compliance energy models prepared at the design stage
are rarely verified or calibrated through as-built models, which provide predictions of energy
performance based on what is actually built [16] (pp. 12–14).
Another source of modeling inaccuracy is the incorporation of rated equipment efficiency rather
than system/plant efficiency and operating strategies [
40
]. Bartlett et al. [
38
] cite the above, as well as
quality issues, occupancy changes, commissioning, and operational issues, which can lead to additional
costs for building owners, reduced occupant productivity, and buildings that fail to live up to their
potential. Also, there is a need to properly model the system components and their control algorithms.
To do this, the design must incorporate real performance curves, not just single point efficiencies. The
perceived failure of a building due to these discrepancies has been termed “the credibility gap”, and it
has created cynicism in the building industry, where people find it hard to ‘trust’ green design [41].
A special issue of BRI Journal (2018) dedicated to the energy performance gap and its causes
suggests the importance of improving behavioural assumptions in energy modeling and simulation,
and delves further into other causes of the gap relating to occupant and other stakeholder practices [
42
].
Studies showed that occupants are a more diverse group than assumed [
43
], and similarly, that data
on occupancy should be grouped by age, income and other variables, rather than considering users as
a homogenous group [44].
In a similar vein, studies have shown that people use their buildings differently, even in the case
of studies comparing similar physical spaces, making accurate predictions is difficult. For example,
Gram-Hanssen [
45
] analyzed quantitative and qualitative data from different households living in
similar homes in a suburb of Copenhagen and found that energy consumption due to usage patterns
varied significantly, with some families using three times the energy of another family in a similar
house. As a result, a challenge to better understanding the gap between predicted versus actual
performance is in better identifying qualities and characteristics of the building users or occupants: to
know how to design for what people want and will do, the users and their wishes must be known [
46
].
Organizational structures and the typical disciplinary silos of the building process also contribute
to the Prediction Gap. In such cases, the gap arises from institutional and cultural barriers in
communication. Fedoruk et al. [
15
] found that institutional issues arise from the way the various
stages of the building life cycle were specified, contracted and implemented. These had the greatest
impact on the discrepancy between anticipated and achieved building energy performance.
Another important issue is the nature of disciplinary roles on a typical project, where the person
making the assumptions that define building performance goals is not the same person carrying out
ongoing monitoring or taking lessons from this project to another one. This illustrates the need to
have a commissioning agent on the project team from the beginning. Fedoruk et al. emphasized
”the importance of having meaningful and effective building energy monitoring capabilities, an
understanding of energy system boundaries in design and analysis, crossing the gaps between different
stages of a building life cycle, and feedback processes throughput design and operation” [
15
] (p. 752).
To this we would add the importance of an integrated design process, as well as an integrated approach
to project delivery that considers the whole building life cycle throughout design, construction,
commissioning and operation. “Beyond technical considerations or simply injecting new information,
a rethink is required of how buildings are planned, designed, constructed, commissioned and operated
in order to close the performance gap” [15] (p. 751).
There is a need for more detailed studies of the Prediction Gap that examine a range of data
and causes. A good example is a recent study of nine Canadian buildings that provides important
findings that compare their predicted performance based on design stage modeling and green rating
Sustainability 2018,10, 4829 7 of 22
submissions with the actual building performance over two years using metered data for energy
and water from the utility bills and submeters. The energy and water data were also compared
to benchmarks for typical performance of similar buildings. In each of the buildings there was a
significant gap between the measured and predicted performance of at least one of the systems
being examined, and the researchers gained important insights into the difficulties of resolving the
performance gap [
38
,
47
]. An unusual feature of this study was that it compared benchmarks of similar
buildings, metered data, spot site measurements, interview data with the design team and building
manager and occupant survey data.
This comprehensive approach found that rigorous reconciliation between projected and actual
performance was not considered possible because it would mean revising performance projections
made at the design stage to reflect actual building use and occupancy (for example, if more people
used the building than initially assumed, predictions would need to be revised), and they found that in
the nine projects, only one of the building models had an energy model that had been recalibrated [
38
].
Their findings related to the nature of the disciplinary silos, distinct workflows and timetables, lack
of funding, and a lack of interaction, and confirmed many of the reasons for the Prediction Gap
cited above. The lack of model recalibration means that within the current way of working, design
stage assumptions are not corrected. The usual workflows and outputs are not designed to be
paired with findings from after the building is built and inhabited. This leads to obvious difficulties
in comparing and evaluating the success of building performance. Consistency in what is being
evaluated, and communication about what data is collected between design stages and occupancy
stages of a building’s life span are important considerations for this gap.
3.1.2. The Future of the Prediction Gap
The energy performance gap is the easiest to define and measure, and yet it remains an urgent
problem. Buildings are not performing as expected, and energy policies and building regulations are
not adequately acknowledging and planning for this gap. The Prediction Gap has been studied in a
number of disciplines, and many researchers see this gap as a problem relating to assumed building
characteristics that can be solved by better modeling. However as noted above, a number of new
studies argue that importance must be placed on understanding the nuances of institutional rules
governing building design, construction and commissioning, occupant behaviour and buildings in use.
Future research relating to the Prediction Gap will need to acknowledge and delve more deeply
into fundamental questions about the barriers to better understand this gap. There is a need to critically
reflect upon what kinds of tools and methods we use to collect building performance data, what kinds
of data to collect, who collects this data and how is it interpreted.
Relevant to the challenges of this gap are what we have termed the Prediction Paradox. The Paradox
lies in the issue of how to predict building performance accurately, and also, in how to interpret results.
If we want to better understand a building’s overall performance, the issues described above mean we
cannot make accurate predictions; in fact, we argue that more information will not improve predictive
accuracy of building performance as a whole. This is an inherent result of the complexity and resultant
emergent properties of socio-technical systems like buildings, not of lack of knowledge. We can predict
accurately only at the component level (e.g., heat loss through a specific material), where physical
performance is well understood.
Much of the literature on what we are calling the Prediction Gap focuses on ways to narrow or
eliminate it. However, due to the challenges of accurately predicting building performance outlined
above, we argue that it is undesirable to attempt to resolve the complexities of building performance
in terms of a single prediction for quantitative aspects such as energy performance.
Instead, we propose supplementing predictive approaches by moving in the direction of scenario
analysis [
48
,
49
] and backcasting techniques [
50
,
51
] that focus on the range of possible outcomes,
and on the ways that desirable outcomes could be approached. Developing alternative scenarios of
performance based on different assumptions about behavioural and institutional issues would allow
Sustainability 2018,10, 4829 8 of 22
us to get a sense of the range of potential performance outcomes. At the same time, backcasting
from desirable outcomes may help to guide the design process by focussing attention on those design
strategies that work best to achieve design goals across the range of scenarios.
In this way, if the process of predicting performance is understood to be inherently uncertain,
the design team can innovate within the predictive space by use of scenario analysis and backcasting.
There are obvious challenges to instituting such processes for design teams and clients who are used
to working towards specific performance targets. However, scenario analysis may present a good
supplemental activity, as it can be seen as falling within the realm of conceptual design where various
built forms and strategies are discussed, iterated, and analysed.
3.2. Expectations Gap: Official Story vs. Lived Experience
Gap B is the qualitative, social analogue to the quantitative performance gap discussed above, and
occupies the Human Systems bottom half of Figure 1, noted as ‘B’: it is often present when comparing
Pre-Occupancy and Post-Occupancy Evaluation results. The Expectations Gap in the context of
occupancy assessment is defined by Coleman and Robinson [
17
] as the gap in occupant expectations:
it lies between what occupants expect and what they experience in a building, and is expressed in
qualitative feedback through survey and/or interview. The gap, summarized here, therefore describes
differences between an assumed or predicted story about building performance as perceived and
shared by occupants (based on prior experience, and sometimes influenced by information available
about the building) and the dynamic, ongoing lived experience of it [17].
In considering this gap, the role of people in buildings, and their assessment of the building
through survey and/or interview, becomes of central importance. Cole et al. [
52
] argued for the
necessity of involved occupants, in the aim of re-contextualizing comfort requirements under climate
change goals and building energy use [
3
]. To enable this re-contextualization, occupant opinions and
experiences are ideally sought and valued in building design and operations, allowing feedback and
feedforward to design stakeholders [17,52–56].
Outside of ratings of comfort, which can be quantified and analysed statistically, themes and
stories can be discerned from occupant commentary through content analysis and interpretation.
Drawn out by survey and/or interview questions, once coded and interpreted [
57
–
61
], these stories
can illuminate how the building is, how it should have been, or that it has not yet become a space that
enables health, well-being and productivity, for example.
The evidence for a qualitative performance gap can be seen in the form of disappointment and
backlash, (including claims of greenwashing in the case of a green or regenerative building) [
17
]. Thus,
there are two levels at which the qualitative performance gap has potential negative consequences:
at the level of occupants, in which disappointment about building performance may affect their
perception of their employer or the sustainable building movement as a whole; and at the level of
designers, who rely on the good reputation of precedent buildings for continued success.
A particular instance of the Expectations Gap is found in highly sustainable buildings [
62
], where
stakeholders’ expectations (designers, clients, and those with an invested interest in the building’s
performance) become codified in promotional material, which then influences occupant expectations,
a finding supported by other researchers [
53
,
54
]. Conceived of as ‘bids’ for social or normative
alignment [
63
], occupant stories about the qualities of built space constitute a powerful socially binding
force, and have implications for the normalization of current and future building design [17,52,62].
The characterization and content of stories or narratives about building energy efficiency has been
discussed in policy literature [
64
–
66
], in which the concern is to bridge and close the performance gap.
However, Coleman and Robinson argued that it may be more fruitful to understand this qualitative
gap as a generative space for a re-making of the building performance story as a whole [
17
]. In doing
so, the aim is to understand the building context as a continuously adaptable, and therefore, potentially
optimizable setting, requiring ongoing feedback from occupants [55].
Sustainability 2018,10, 4829 9 of 22
In other words, the very existence of the Expectations Gap calls for interactive adaptivity, a concept
developed by Cole et al. [
52
], which asserts that building performance must take into account
more than just the individual and ecological factors; performance must include broader societal
circumstances, over time. Interactive adaptivity integrates “[a] dynamic and complex building system
with a participatory process, interactivity between inhabitants, and between inhabitants and building
elements”, and allows adaptation “to changing conditions (e.g., seasonal temperature change, or, on a
larger scale, global climate change), resulting in a fluid but robust design that is responsive to social,
ecological, and economic conditions over time.” [
52
] (p. 333). In effect, the building engages in a
conversation with its occupants, leading to mutual adaptation over time.
The Jim Pattison Pavilion in an iiSBE case study [
67
] provides a simple but non-trivial example
of interactive adaptivity: windows fitted with red and green lights guide the occupant to open or
close windows in order to maximize cooling. In turn, these lights constitute a tool for awareness
of energy consumption, and the process as a whole engenders continual optimization. The Comfy
app (www.comfy.com) provides another example in which building response can adapt based on
occupant feedback on temperature, lighting, room bookings and other aspects, thereby centering on
the occupant experience with both building systems and occupants continuously learning from it.
3.2.1. Assessing the Expectations Gap
Exploring the Expectations Gap relies on the collection of at least one source of data, taken
from Pre- Occupancy and Post-Occupancy Evaluations, or pre-retrofit and post-retrofit studies. A
Pre-occupancy evaluation (Pre-OE) or Pre-retrofit evaluation is an evaluation used as a baseline against
which future evaluations can be compared. Usually, the same survey is conducted in the Pre-OE and
POE so that data can be statistically compared. Responses in the Pre-OE and POE, with identical or
different populations, can be compared in the sense that the data depicts responses to the “before” and
“after” moving or retrofit conditions.
Some assessment studies have used a control group, in the form of a sub-population of individuals
that remain behind in the prior building: both groups (those remaining behind, and those moving) are
surveyed with the Pre-OE, and later the POE, at the same time [29,68].
The formulation of control populations is challenging, however, since building typologies and
programming, compounded by differing building tenants and populations, constitute incredibly
diverse ‘entities’ with many variables affecting evaluation outcomes. Paired matches (the same
individual is surveyed before and after the move or retrofit) may therefore provide a better way to
control for effects, which we assume are perceived more consistently by a single individual, than they
are perceived between different individuals.
Yet, the goal of surveying pre- and post-occupancy matched pair individuals is itself hampered
by attrition amplified by the sometimes lengthy amount of time between pre- and post-occupancy,
as well as the fact that it is often not known what tenant, let alone individuals, are moving into a
building. In this case, an unmatched control (different individuals in groups before and after the move)
is assumed to be better than none.
The question of when to conduct the pre- and post-evaluations is also significant, and depends on
the aim of evaluation. If the desire is to capture deficiencies or the immediate impact of a retrofit, then
earlier in occupancy of the new building or retrofit is best; and if it is to understand how the building
performs under optimized conditions, then later in occupancy is best. The first 6–12 months of an
occupant’s experience in a new build or retrofit is a sensitive adjustment period for both occupants
and building operators. At this time, complaints are high and new habits and routines are being
established by all. For the aim of identifying and fixing deficiencies, this is a good period to run a POE,
but of course, it will likely turn up the largest Expectations Gap, which also may be highly transitory
depending on how soon the deficiencies are rectified. For later POEs, Paevere and Brown found that
the case study green building (CH2) was still being fine-tuned after one year [
69
], while McCunn and
Gifford suggest that 4 years may not be long enough for the positive effects of a green building to be
Sustainability 2018,10, 4829 10 of 22
felt [
70
]. As such, building operators would know best when it is an optimal time to run a POE, or at
least, they would be able to flag ongoing deficiencies during analysis. The season that an assessment is
run is also significant, since if the pre- and post-occupancy evaluations are to be matched with as little
variability as possible, then ideally, the seasons are matched as well, meaning that at least a year’s gap
between the two evaluations is necessary.
In terms of requesting qualitative data, questions in the Pre-OE survey and/or interview [59,71]
can be used to query occupant expectation of future conditions, benefits, and features, such as:
anticipated IEQ performance, support of sustainable activities, the expected impact of building
conditions and features on productivity, well-being and health, and so on. In the absence of interviews,
a section for open comments in surveys usually elicits illuminating and unexpected feedback. As [
72
]
indicates, surveys are often used as repositories for complaints; further, complaints are useful to
investigate because they often indicate a change in what is normal and expected [62].
This data is then analysed through coding, categorization, distillation and interpretation [
57
,
58
,
60
,
61
] for themes that build into narratives (software like NVivo can be useful for this purpose), with
outliers taken into consideration. Thus, pre- and post-commentary is analysed and coded for themes
(e.g., the expression of skepticism, a feeling of forgiveness, the sense of tribe) and narratives (e.g., “I
heard about the user interface we’d get and I was annoyed when we didn’t, but it’s not really a big
deal—work is easier now because all my colleagues are all here”), and the before and after stories are
then compared for similarity and difference.
If data that can be analysed more quantitatively is desired, questions that directly ask about
expectations before the move or retrofit can also provide a picture of difference or similarity in
expectations, before and after a retrofit or move. For example, in the Pre-OE, a set of Likert scale
questions could ask, “Rate your level of agreement with the statement: ‘I expect that air quality
will be excellent at all times.’” Following up on the POE after the move or retrofit, the mirroring
question would ask, “Rate your level of agreement with the statement, ‘My expectations were met
regarding air quality.’” This data can be statistically analysed, and it can also be considered alongside
the prevalent themes, narratives, and outliers produced by the analysis of interview questions and
survey commentary, for further informal comparison.
Traditional POE data (e.g., Likert scale data on workplace and IEQ satisfactions) can also be
inspected for how it connects to, supports or contradicts the data on expectations, or commentary
before and after.
Interpretations can be further connected to other data where expectations are unmet (e.g., in the
Pre-OE, an occupant thought there would be art installed; in the POE, it turns out that the lack of it
is perceived to hamper well-being, thereby connecting the two), or where expectations are exceeded
(the occupant expected plentiful fresh air; they experienced more fresh air than expected, and this is
perceived to directly enhance health). In all cases, a coding session of the same qualitative data could
be conducted independently by another researcher to corroborate interpretations [57].
As mentioned, adding another layer of complexity to the assessment of the Expectations Gap is the
particular case of highly novel, innovative, and/or green buildings, which are usually touted for their
difference in publicly-available and distributed promotional materials. According to Coleman and
Robinson, for buildings that have been publicly marketed the promotional material itself (brochures,
building tour script, external and internal signage, online building manual, media coverage, etc.) will
significantly influence occupant expectations. These materials can be collected and analysed as well,
and interpreted to constitute a collective “Official Story” [
17
]. The same authors found that the Official
Story derived from promotional materials and a media analysis was clearly reflected in occupant
stories about the building [
17
,
62
]. In terms of analysis of the Expectations Gap, in the case of highly
promoted buildings, designer and stakeholder aspirations may dominate occupant expectations, and
the lived experience assessed in the POE is then positioned as a reaction to that Official Story [17].
Sustainability 2018,10, 4829 11 of 22
3.2.2. The Future of the Expectations Gap
The Expectations Gap exists in green building performance as it does in other fields, because
the complexities of reality—often of an institutional and social nature—intrude upon expectations.
However, the Expectations Gap is unique in that it relies entirely on subjective qualitative data, which
is based on the interplay, feedback and feedforward between prior, and later, occupant experience.
When the stakeholders’ collective vision for the building is available at or even before moving in, the
occupant experience is ultimately an evaluation of that vision.
It is worth noting that in the special case of sustainable buildings where promotional Official
Stories influence occupant expectations, this promotion is almost always focused on innovative and
novel building features, rather than on occupant social and well-being-related opportunities. This
is significant because researchers [
17
,
73
] have shown that occupants are more readily disappointed
with the anticipated physical benefits of a building than they are with the intangible benefits (such as
social and creative opportunities, well-being and productivity, etc.) afforded by a building context.
As a result, we suggest that the existence of a significant Expectations Gap in a sustainable building
indicates that building stakeholders should focus more on promoting the social and productive aspects
of inhabitation, rather than on innovative building features [17,53].
This focus on the role of the occupant provides a rationale for their involvement in design and
operations, as is suggested by the Soft Landings approach [
56
] and a variety of other authors [
55
,
74
,
75
].
Cole et al. noted that building occupants may be considered ‘inhabitants’ when they play an active role
in the maintenance and performance of their buildings, as opposed to ‘occupants’, who are passive
recipients of pre-determined comfort conditions [52].
The logical complement to the involvement of engaged occupants is a continuous form of POE,
allowing continuous feedback, feedforward and optimization of the building environment in tandem
with occupants’ satisfaction, well-being, and productivity. This is furthermore in keeping with a
regenerative approach, seeking to balance human and environmental well-being, and ultimately
provides interactive adaptivity [
52
,
53
], allowing for an active process of mutual accommodation
between the building and its occupants.
The concept of forgiveness illustrates the opportunity of the recommended notion of interactive
adaptivity. A phenomenon noted by various building researchers, occupants of green buildings
appear to be more forgiving of conditions affecting their comfort than are occupants of conventional
buildings [
53
,
72
,
76
–
79
]. (Leaman and Bordass calculate a ‘forgiveness factor’ as the ratio of overall
comfort, to the mean of the individual indoor environmental quality (IEQ) variables scores. The factor
ranges from 0.8-1.0. A factor higher than 1.0 indicates more tolerant occupants, who are willing to
tolerate insufficient conditions in spite of expressed dissatisfactions (Leaman and Bordass, 2007)).
Importantly, greater forgiveness is also associated with greater occupant control and feedback [
62
,
77
,
80
].
The message is that a green building, with greater control and feedback allowances for occupants, is
more likely to be forgiven its failings. At the very least, adaptation through interactivity, in the gap
between what was expected and what was delivered, allows for new expectations [17].
Last, a general challenge related to instituting interactive adaptivity and feedback processes entails
a change in the ecosystem of building design and facilities management, for occupants, managers,
designers, and clients. These changes have significant implications for the required skills and training
of building operators. Further, occupants would be faced with typically unfamiliar expectations of
engagement, and building and construction stakeholders would need to learn how to work with
negative feedback (which must be seen as part and parcel of instituting dialogue in a regenerative
sustainability context).
3.3. Outcomes Gap: Measured vs. Experienced
The Outcomes Gap C describes the difference between measured performance and the lived
experience—between Environmental and Human System data. Specifically, this measured performance
typically relates to indoor environmental quality parameters that are indicators of visual, acoustical
Sustainability 2018,10, 4829 12 of 22
and thermal comfort and indoor air quality. These parameters may include light levels, air temperature,
mean radiant temperature, relative humidity, and carbon dioxide concentration, among others. The
measured data is often compared to standards or accepted models which are used to determine
if the measurements indicate a satisfactory environment. Data on the lived experience is typically
captured through occupant surveys or interviews, both structured and unstructured. Guerra-Santina
and Aidan [
81
] provide a comprehensive overview of data collection methods to evaluate building
performance energy and thermal comfort using both qualitative and quantitative methods. Then,
these results can be compared to determine if the monitored conditions and the reported conditions
agree with one another. Examination of the Outcomes Gap is a critical first step to contextualizing the
sometimes misleading quantitative metrics like energy performance or temperature and getting to
the root whether the building occupants are satisfied or not and, perhaps more importantly, why they
perceive the building in the way they do [82].
Numerous studies have gathered both Environmental and Human System data that have been
used to examine the Outcomes Gap. Studies related to this gap are typically conducted in office
environments and have largely focused on issues of personal environmental control and perceptions
versus measured conditions relating to daylight and ventilation. While all of the studies presented
here involved the collection and analysis of both qualitative and quantitative data, only some studies
have collected these various data types attempting to correlate qualitative responses to the quantitative
monitoring data in various ways, either directly for certain parameters like lighting levels, or indirectly,
through translational models, such as in the case of thermal comfort.
For example, studies that simply examine both data types include Altomonte and Schiavon [
83
],
Geng et al. [
84
] and Newsham et al. [
85
]. Altomonte and Schiavon [
83
] studied the difference between
BREEAM and non-BREEAM office buildings using survey data on lighting, acoustic and thermal
comfort which were collected while spot measurements of various IEQ parameters were collected at the
respondent’s workstation. The findings related to lighting led the authors to hypothesize that personal
control over lighting available in non-BREEAM buildings, which allowed occupants to intervene
during periods of visual discomfort, resulting in greater satisfaction with the indoor environment
and a tolerance for greater fluctuation in lighting levels. Geng et al. [
84
] compared environmental
measurements, survey results and productivity tests under various air temperature conditions from
16
◦
C to 28
◦
C. Interestingly, they found that thermal dissatisfaction appeared to override awareness of
other IEQ parameters like lighting, noise and IAQ, but when occupants were thermally comfortable,
they became more aware of these parameters, specifically, noise and lighting, demonstrating the
interconnectedness of the IEQ perceptions and primacy of thermal comfort. A laboratory-based study
by Newsham et al. [
85
] used occupant surveys to evaluate occupant attitudes to having personal
control over the changing environmental conditions that would occur during a demand response event.
They found that the ability to control one’s lighting and ventilation levels both increased satisfaction
and decreased energy use.
Examples of studies that attempted to find a correlation between survey responses and
environmental measurements include Leder et al. [
86
] and Choi et al. [
87
]. A field-based study
by Leder et al. [
86
] measured perceived conditions versus measurements of IEQ parameters in an office
space. Stepwise regression was used to compare field measurements of environmental conditions at
individual workstations to simultaneous online questionnaires about the respondent’s environment
to determine which conditions most impacted occupant satisfaction. The study findings support the
idea that green buildings can provide better perceived indoor environmental quality compared with
conventional buildings. Choi et al. [
87
] compared 15-min IEQ measurements at various workstations
in the perimeter and interior of an office with survey data yielding numerical ratings of various IEQ
parameters. They found that IEQ guidelines did not necessarily yield comfortable conditions, and
thus, made a series of recommendations about temperature, lighting and air flow rates to improve
occupant satisfaction.
Sustainability 2018,10, 4829 13 of 22
This type of research is less common in the residential sector, but a handful of studies have begun
to examine the Outcomes Gap in this sector. Similarly, in the residential sector, some studies only
examine these two datasets, such as Liu et al. [
23
] and Dascalaki and Sermpetzoglou [
88
]. For example,
Liu et al. (2015) examined the performance of two Swedish apartment buildings: one that had
undergone an energy retrofit and one that had not. They used indoor temperature data and energy
data to calibrate a building energy model which was then used to generate PMV and PPD data based
on the model described in ASHRAE Standard 55. These pre- and post-retrofit modeled thermal comfort
indicators were compared with reported thermal comfort issues from pre-retrofit and post-retrofit
occupant surveys. A correlation was found, but not quantified given the different metrics output from
the model and the surveys. The surveys also found improvements in many other areas of indoor
environmental quality, including noise and air quality, but site measurements for these parameters
were not collected for comparison with these perceived improvements. A study of Hellenic schools
included IEQ measurements for one week as well as occupant surveys of teachers and some pupils [
88
].
Approximately 60% of the time the monitored conditions were considered unacceptable relative to
standards such as ASHRAE. The survey data, which included a numerical rating of various IEQ
parameters, was reported separately without a comparison to the monitored data. The most frequent
survey complaints related to insufficient ventilation, noise disturbance, glare and thermal discomfort.
Only one residential study directly examined the gap between the qualitative and quantitative
data. A three-year study of social housing buildings in Toronto showed a significant difference in
the resident-reported thermal comfort and the calculated thermal comfort levels based on ASHRAE
Standard 55 and in-situ environmental measurements in both summer and winter [
89
,
90
]. Generally,
occupants were less satisfied with wintertime conditions than the comfort standard would suggest,
but interestingly, the particular combination of the building type (e.g., low or high rise) and occupancy
type (e.g., senior or family) appeared to have an impact on the level of agreement between reported
and calculated thermal comfort. The mid-rise buildings (which had a combination of senior and single
occupancy types) saw the closest agreement between measured and perceived thermal comfort. While
the measured data collected from high-rise buildings occupied by families suggested a higher level of
comfort than the survey responses indicated, the data collected from the low-rise buildings occupied
by seniors suggested a lower level of comfort, despite occupants reporting satisfactory conditions.
There are some obvious limitations to comparing monitored data with results from a survey.
Comparing qualitative survey responses with quantitative parameter measurements leads to questions
about what constitutes an agreement between these two data sets. Are we measuring the right
parameters and asking the right questions to allow for a direct comparison between these qualitative
and quantitative data? How do we compare continuously monitored parameter data to episodic
survey data? Furthermore, should average or extreme conditions be compared? The next section
explores these aspects related to assessing the Outcomes Gap.
3.3.1. Assessing the Outcomes Gap
(1) Comparison of qualitative and quantitative data
Of the three performance gaps, the Outcomes gap is perhaps the most challenging to assess and
quantify, primarily because different metrics must be compared directly. The first two performance
gaps involve comparisons of the same data type (e.g., degrees Celsius or ekWh/m
2
for the Prediction
Gap or the results of a ranking survey question for the Expectations Gap). To assess the Outcomes
Gap, on the other hand, survey results must be compared with monitored data.
This ‘translation’ of data to enable the comparison between these two data types often occurs via
an existing model. The model serves as a means of converting one data set into a form where it can be
directly compared with the other. The challenge with this comparison, and a likely contribution to the
existence of the Outcomes Gap, lies in the model assumptions. While it may be possible to reduce the
influence of inaccurate assumptions by gathering more performance data, there is a practical limit to
Sustainability 2018,10, 4829 14 of 22
the number, type and placement of sensors in occupied spaces. Furthermore, there is a question as to
whether we are even measuring the correct parameters to assess occupant satisfaction.
For example, the thermal comfort model in ASHRAE Standard 55: Thermal Environmental
Conditions for Human Occupancy includes inputs such as air temperature, mean radiant temperature,
relative humidity, air velocity, clothing level and metabolic rate to determine whether the majority of
occupants would find the particular set of conditions comfortable. These values are used to determine
the Predicted Mean Vote (PMV) or the Predicted Percent Dissatisfied (PPD), which can be directly
compared to the results from a survey question which asks the occupant to rate their thermal comfort
on a Likert scale of the ASHRAE Thermal Comfort Index, or state whether they are satisfied or
dissatisfied with their thermal environment. The empirical relationship between these model inputs
and outputs may be quite strong when the inputs are fully characterized at the occupant level in a
laboratory setting and the testing is conducted on a sufficiently large sample. However, gathering the
model inputs in an actual building is significantly more challenging. This can be seen by examining
how we would collect data to input into a thermal comfort model in order to allow for a comparison
with the survey data.
Devices such as smart thermostats and sensors connected to building automation systems can
easily collect data on air temperature and relative humidity; however, both of these parameters are
collected locally at the sensor location and do not necessarily reflect the variation in these conditions
throughout the zone of interest, and specifically, where the occupant is currently located. Data on
the other parameters are significantly more challenging to collect. Mean radiant temperature (MRT)
can be measured using a globe thermometer located in the geometric centre of the zone, which is
obviously impractical for an occupied space, and only reflects MRT at the location of the thermometer.
Alternatively, infrared imaging can be used to determine interior surface temperatures, and then shape
factors can be used to determine the resulting mean radiant field at the location of the occupant.
Once again, the imaging and shape factor calculations are impractical for long term monitoring as
all surface temperatures in the zone must be collected and the occupant(s) location must be known.
To characterize draught sensation and air speed throughout a zone, a matrix of air velocity sensors
distributed throughout the zone volume would be required, but this would be impossible to instrument
in an occupied space. Clothing levels and metabolic rate depend on occupant preferences and activities.
All of these parameters vary throughout the zone or from person to person and through the day and
year. Therefore, based on current and commonly-available sensor technology, assumptions about
many of these monitored parameters are often required to use these ‘translational’ models, such as
those described in ASHRAE Standard 55.
Even if it were possible to collect all of these environmental data in an occupied space, some
occupant perceptions and preferences will extend beyond the upper and lower acceptability limit
as dictated by the model. So, while the model might provide an indication of the conditions under
which most occupants would be satisfied, most current models cannot precisely predict how a given
occupant will perceive certain interior conditions. Therefore, perhaps our efforts are better spent using
survey data to interpret quantitative data collected through monitoring, rather than try to directly
compare them.
(2) Comparison of continuous and episodic data collection
Inexpensive data storage means data on IEQ parameters can be collected at short intervals
(e.g., every few minutes) for long monitoring periods (e.g., months or years); however, surveys and
interviews are generally expensive and time consuming to conduct, and so, for a given study, are
often only conducted at a single point in time or, at best, a handful of times throughout the study
period. Even with surveys conducted via smartphone, there is a practical limit to the number of times
one can survey a study participant before annoyance or impatience sets in. This difference in data
collection frequency presents challenges for comparing the qualitative survey data and the quantitative
monitored data.
Sustainability 2018,10, 4829 15 of 22
These occupant surveys may reflect real-time observations or retrospective ones [
82
]; however,
the ability to recollect their experiences and perceptions varies between individuals and the time span
which they are asked to consider. For example, an occupant may easily be able to recall the quality of
their experience in a space over the last 15 min or hour, but may find it more challenging to characterize
this experience over a few months or a year. This can be a result of trying to recall past experiences
under different current conditions or after significant passage of time (e.g., considering wintertime
thermal comfort in response to a survey question posed during the middle of summer) or difficulty
aggregating a range of conditions experienced (e.g., will the occupant report on the extreme periods
of discomfort or their impressions of the average conditions in the space). The wording of survey
questions, as well as the motivation of the occupant, can influence how these questions are interpreted
and responded to.
With respect to the monitored data, we need to determine what time span should be used for
comparison with the survey responses. This decision is relatively simple if the survey question asks an
occupant to recall conditions over a 15-min period, but if queried about seasonal differences in terms of
the level of satisfaction with the space, how should the monitored data be processed for comparison?
One option is to examine the average or extreme values over the time period and compare these to the
survey responses. Alternatively, a translational model can be used to determine if the conditions would
be considered satisfactory over the given time period, and then the proportion of time with satisfactory
conditions in the given time period can be compared to the survey response. Regardless, we must
make assumptions about how the respondents are interpreting the questions before processing and
comparing the data.
3.3.2. The Future of the Outcomes Gap
There are numerous challenges that must be addressed with respect to the Outcomes Gap.
However, as with the Expectations Gap, instead of trying to close this gap, we may wish to use the
existence of it as a creative resource to inform the development of new, more meaningful ways of
thinking about performance, and thereby better performance indicators of occupant satisfaction. The
critical question we must ask ourselves is which data type, Environmental System or Human System,
is indicative of ‘real’ performance? In other words, we can measure indoor conditions and compare
them to a standard, but if the occupants are not satisfied, does it really matter that the quantitative
measurements meet the standard? This section discusses some of the challenges and opportunities
related to collecting and comparing these two data types.
To improve occupant satisfaction, we can consider increasing the frequency of occupant feedback,
allowing for adequate control over the space either centrally or individually in response to occupant
discomfort and, perhaps most importantly, finding a way to communicate the consequences of this
gap to building designers and owners.
With respect to data collection, as sensors become smaller and less expensive, there are a number
of wearable options that might be able to provide real-time, local-to-the occupant environmental
feedback to building control systems. Mobile devices, desktop apps or wireless polling stations
throughout the building can provide easier, less-intrusive ways to collect real-time data on occupant
perceptions of their space, and provide these data directly to building operators. Alternatively, image
analytics of facial expressions from security footage or assessment of mood via social media could
yield occupant feedback without requiring their explicit participation in the process. In all cases,
establishing best practices around privacy considerations is a challenge not limited to methods of
building performance assessment, and insight may be drawn from fields where monitoring is common.
Another challenge emerges where satisfying the desire for an individual to have control over their
space may be at odds with the satisfaction of the larger group occupying the space, which must be
carefully considered and addressed at the design stage. Furthermore, gathering and processing this
individual versus group feedback must be managed by building operators in order to make use of
these data. Applications, such as the Comfy App, that can aggregate feedback and provide guidance
Sustainability 2018,10, 4829 16 of 22
to operators, are necessary to quickly process these vast amounts of real time data to make the findings
actionable. However, even with these advances in data collection and processing, we are often limited
by the controls and zoning associated with our current heating, cooling, ventilation, lighting, window
and blind control systems, just to name a few.
Finally, the findings from exploration of this gap must be communicated beyond academic
circles. Best practices and common challenges should be incorporated into guidelines, standards and
regulations so that minimization or elimination of this gap will be considered from even the earliest
stages of schematic design. Only through consideration of this performance gap throughout the design,
construction and operation of a building can we put occupant wellbeing first and foremost in our list
of performance objectives for the built environment.
4. Conclusions
Achieving the kinds of climate change and sustainability goals increasingly adopted by various
jurisdictions around the world will require unprecedented improvements in the performance of the
built environment. A crucial component of achieving such goals will be the implementation and
evaluation of performance assessment approaches that will allow us to evaluate the sustainability
performance of the built environment in accurate and meaningful ways.
Adopting a regenerative approach to sustainability assessments suggests that such assessments
must be able to evaluate both human and environmental performance, and also to assess the degree to
which net positive outcomes in both of these areas have been achieved. This paper has provided a
conceptual framework in terms of which net positive outcomes in both environmental and human
terms can be assessed. The framework quadrants (Figure 1) indicate human and environmental system
performance assessment activities, which include predicting building performance; using pre- and
post-occupancy evaluation to understand occupant and possibly stakeholder expectations and actual
experience; and measuring actual environmental performance. These activities taken as a whole, along
with innovations on these activities as discussed above and summarized below, could provide the
creative material for new pathways towards net positive design and sustainability.
Articulation of this framework led us to posit three important performance gaps:
•
Prediction Gap: Predicted versus actual resource use (e.g., modelled and measured energy,
water consumption);
•
Expectations Gap: Expectations regarding the performance of sustainable buildings versus the
actual lived experience of the building occupants (e.g., pre- and post-occupancy evaluations);
•
Outcomes Gap: Measured performance versus lived experience (e.g., thermal comfort measurements
and survey results).
The Prediction Gap between the predicted and actual performance of environmental systems in
the built environment is the best known of these performance gaps, and has been the subject of much
research and scholarly discussion. However, in the main, the approach taken to this gap is to look for
ways for it to be narrowed or eliminated. In other words, the goal has been to make the predictions of
building performance more accurate.
In contrast to this approach, we identify a Prediction Paradox which suggests that because of the
complex set of behavioral and institutional factors that give rise to the Prediction Gap, trying to achieve
accurate predictions of actual whole building performance at the design stage is perhaps the wrong
goal. Instead, we should supplement predictive analysis of building components and technologies
with scenario analysis and backcasting approaches. These are intended to identify design strategies
that may be resilient to the inherent uncertainty about how the building will actually perform once
built and occupied. In this sense, the Prediction Gap can be seen as a creative resource to be explored
and used to improve the resilience of design strategies.
There is increasing interest in the literature in the experience and behaviour of building occupants,
and how this affects the performance of buildings in both human and environmental terms. The
Sustainability 2018,10, 4829 17 of 22
Expectations Gap is the difference between the expectations and the actual experience of building
occupants. As suggested here and in the original analysis [
17
], it may be useful to think of this gap as a
source of creative tension and opportunity for interplay between the building systems and the building
occupants. Such ‘interactive adaptivity’ may enable passive building ‘occupants’ to instead become
active ‘inhabitants’ of the building, with a sense of place in, and engagement with, the building itself.
The Expectations Gap becomes the basis for a conversation between the building and its inhabitants,
with a goal of improving conditions in both directions over time. As buildings become smarter, this
conversation can be expected to become more meaningful, and potentially, much more effective.
The Outcomes Gap between the measured performance of the built environment and the human
experience of it, is, in some ways, the most difficult to assess, combining both quantitative performance
measurement and qualitative human responses. Moreover, the existence of such a gap raises an
important philosophical question: what are the real or true performance measures of the built
environment in question? Are they the measured values of performance, such as temperature or
humidity, or the experiences of comfort and ease?
This philosophical question has important practical implications: do we try and bring the
experience of the environment in line with the measured performance (e.g., use our understanding
of people’s experience to help them better interpret the meaning of the measured data), or do we
adapt performance measurements to better reflect actual experience? Our suggestion, finally, is to
do both: use qualitative data to better interpret performance measurement, and also measure more
experientially-meaningful outcomes. In so doing, we can explore the interplay between Environmental
and Human Systems data.
For all three performance gaps, we propose using the existence of such gaps as creative resources,
offering the potential to better understand how to achieve net positive sustainability outcomes in both
human and environmental terms. If our built environment is to become significantly more sustainable,
we need to harness the creative energy of designers, operators and building inhabitants, in developing
new ways not only to design and build or retrofit our built environment, but also to engage in processes
of continuous improvement over time. Treating the three performance gaps we have identified as
evidence of opportunities to pursue such improvements is one important way to contribute to this goal.
Author Contributions:
All authors contributed to the writing and editing of this manuscript, and each has
approved the submitted version.
Funding: This research received no external funding.
Acknowledgments:
We would like to thank the members of the Sustainable Built Environment Performance
Assessment group at the University of Toronto for their feedback on drafts, particularly Fiona Miller, Bryan Karney
and Frances Silverman. We would like to acknowledge the following University of Toronto Faculties for funding
the Sustainable Built Environment Performance Assessment network: Faculty of Architecture, Landscape and
Design; Faculty of Arts and Science; Faculty of Applied Science and Engineering; and The Dalla Lana School of
Public Health.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
De Wilde, P.; Coley, D. The implications of a changing climate for buildings. Build. Environ.
2012
,55, 1–7.
[CrossRef]
2.
Wan, K.K.; Li, D.H.; Pan, W.; Lam, J.C. Impact of climate change on building energy use in different climate
zones and mitigation and adaptation implications. Appl. Energy 2012,97, 274–282. [CrossRef]
3.
Lucon, O.; Ürge-Vorsatz, D.; Zain Ahmed, A.; Akbari, H.; Bertoldi, P.; Cabeza, L.; Eyre, N.; Gadgil, A.;
Harvey, L.; Jiang, Y.; et al. Buildings. In Climate Change 2014: Mitigation of Climate Change. Contribution of
Working Group III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change; Edenhofer, O.,
Pichs-Madruga, R., Sokona, Y., Farahani, E., Kadner, S., Seyboth, K., Adler, A., Baum, I., Brunner, S.,
Eickemeier, P., et al., Eds.; IPCC: Geneva, Switzerland, 2014; Chapter 9.
Sustainability 2018,10, 4829 18 of 22
4.
Euractiv Climate Change News. Cities Take up Climate Baton at COP23, Make Ambitious Emission Pledges.
Available online: https://www.euractiv.com/section/climate-environment/news/cities-take-up-climate-
baton-at-cop23-make-ambitious-emission-pledges/ (accessed on 30 November 2018).
5.
Global Covenant of Mayors for Climate and Energy. At COP23, Global Covenant of
Mayors Raises the Level of Ambition at the Subnational Level and Local Leaders Showcase
their Commitment to Accelerating Global Progress on Climate Change. Available online:
http://www.globalcovenantofmayors.org/press/cop-23-gcom-reports-collective-impact-committed-
cities-announces-new-ghg-emissions-inventory-standard-cities-local-governments/ (accessed on 30
November 2018).
6.
City of Toronto. Agenda Item PE 19.4.: Transform TO: Climate Action for a Healthy, Equitable and Prosperous
Toronto, Report 2: The Pathway to a Low Carbon Future; Sustainable Solutions Group: Vancouver, BC,
Canada, 2017.
7.
City of Toronto. City of Toronto. 2050: Pathway to a Low-Carbon Toronto, Report 2, Highlights of the City of Toronto
Staff Report; Sustainable Solutions Group: Vancouver, BC, Canada, 2017.
8.
City of Toronto. TransformTO: Sustainability Solutions Group and what if? Climate Action for a Healthy, Equitable,
Prosperous Toronto, Results of Modelling Greenhouse Gas Emissions to 2050; Sustainable Solutions Group:
Vancouver, BC, Canada, 2017; p. 42.
9.
Canada Green Building Council. Green Building in Canada: Assessing the Market Impacts & Opportunities.
Available online: https://www.cagbc.org/cagbcdocs/advocacy/Green_Building_in_Canada_CaGBC_and_
Delphi_Report_Executive_Summary.pdf (accessed on 30 November 2018).
10.
Robinson, J.; Cole, R. Theoretical underpinnings of regenerative sustainability. Build. Res. Inf.
2015
,43,
133–143. [CrossRef]
11. Cole, R.J. Net-zero and net-positive design. Build. Res. Inf. 2015,43, 1–6. [CrossRef]
12.
Chu, A.-M.; Cayuela, A.; Robinson, J. Visions and Strategies for Sustainable Buildings and Neighbourhoods—An
International Scan of Highly Sustainable Building and Neighbourhood Projects around the World. Report Prepared for
Copenhagen Business School, Centre for Interactive Research on Sustainability; The University of British Columbia:
Vancouver, BC, Canada, 2015.
13.
Gorgolewski, M.; Brown, C.; Chu, A.-M.; Turcato, A.; Barlett, K.; Ebrahimi, G.; Hodgson, M.; Mallory-Hill, S.;
Ouf, M.; Scannell, L. Performance of Sustainable Buildings in Colder Climates. J. Green Build.
2016
,11,
131–153. [CrossRef]
14.
International Living Future Institute. Available online: https://living-future.org/lbc/ (accessed on
30 November 2018).
15.
Fedoruk, L.; Cole, R.; Robinson, J.; Cayuela, A. Learning from failure: Understanding the
anticipated-achieved building energy performance gap. Build. Res. Inf. 2015,43, 750–763. [CrossRef]
16.
Chu, A.-M. Understanding the Performance Gap: An Evaluation of the Energy Efficiency of Three
High-Performance Buildings in British Columbia. Master’s Thesis, University of British Columbia,
Vancouver, BC, Canada, 2016.
17.
Coleman, S.; Robinson, J. Introducing the qualitative performance gap: Stories about a sustainable building.
Build. Res. Inf. 2017,46, 485–500. [CrossRef]
18.
Vischer, J. Post-Occupancy Evaluation: A Multifaceted Tool for Building Improvement. In Learning from Our
Buildings: A State-of-the-Practice Summary of Post-Occupancy Evaluation; Federal Facilities Council & National
Research Council, National Academies Press: Washington, DC, USA, 2001; pp. 23–34.
19.
Preiser, W.F.; White, E.; Rabinowitz, H. Post-Occupancy Evaluation (Routledge Revivals); Routledge:
Abingdon-on-Thames, UK, 2015.
20.
Leaman, A.; Bordass, W.; Cohen, M.; Standeven, M. The Probe Occupant Surveys, Buildings in Use ‘97: How
Buildings Really Work; Commonwealth Institute: London, UK, 1997.
21.
Heerwagen, J.; Zagreus, L. The Human Factors of Sustainable Building Design: Post Occupancy Evaluation of the
Philip Merrill Environmental Center; UC Berkeley Center for the Built Environment: Berkeley, CA, USA, 2005.
Available online: https://escholarship.org/uc/item/67j1418w (accessed on 17 December 2018).
22.
Mallory-Hill, S.; Preiser, W.F.E.; Watson, C.G. (Eds.) Enhancing Building Performance; Blackwell Publishing
Ltd.: Oxford, UK, 2012; ISBN 978-0-470-65759-1.
Sustainability 2018,10, 4829 19 of 22
23.
Liu, L.; Rohdin, P.; Moshfegh, B. Evaluating Indoor Environment of a Retrofitted Multi-Family Building with
Improved Energy Performance in Sweden. Energy Build. 2015,102, 32–44. [CrossRef]
24.
Shin, S.; Jeong, S.; Lee, J.; Wan Hong, S.; Jung, S. Pre-Occupancy Evaluation based on user behavior prediction
in 3D virtual simulation. Autom. Constr. 2017,74, 55–65. [CrossRef]
25.
Alzoubi, H.; Al-Rqaibat, S.; Bataineh, R.F. Pre-versus post-occupancy evaluation of daylight quality in
hospitals. Build. Environ. 2010,45, 2652–2665. [CrossRef]
26.
Reckermann, J. CIRS pre-Occupancy Evaluation: Inhabitant Feedback Processes and Possibilities for a
Regenerative Place. Master’s Thesis, University of British Columbia, Vancouver, BC, Canada, 2014.
27.
Newsham, G.; Birt, B.; Arsenault, C.; Thompson, L.; Veitch, J.; Mancini, S.; Galasiu, A.; Gover, B.;
Macdonald, I.; Burns, G. Do Green Buildings Outperform Conventional Buildings? Indoor Environment
and Energy Performance in North American Offices; National Research Council Canada: Vancouver, BC,
Canada, 2012. Available online: http://nparc.nrc-cnrc.gc.ca/eng/view/object/?id=1714b57c-88c2-4dec-
953a-0e640b7db12b (accessed on 30 November 2018).
28.
Singh, A.; Syal, M.; Grady, S.C.; Korkmaz, S. Effects of green buildings on employee health and productivity.
Am. J. Public Health 2010,100, 1665–1668. [CrossRef] [PubMed]
29.
Schreuder, E.; van Heel, L.; Goedhart, R.; Dusseldorp, E.; Schraagen, J.M.; Burdorf, A. Effects of newly
designed hospital buildings on staff perceptions: A pre-post study to validate design decisions. HERD-Health
Environ. Res. 2015,8, 77–97. [CrossRef] [PubMed]
30.
van der Zwart, J.; van der Voordt, T.J. Adding value by hospital real estate: An exploration of dutch practice.
HERD-Health Environ. Res. 2015,9, 52–68. [CrossRef] [PubMed]
31.
Cole, R.J.; Brown, Z.; McKay, S. Building human agency: A timely manifesto. Build. Res. Inf.
2010
,38,
339–350. [CrossRef]
32.
Birt, B.; Newsham, G.R. Post-occupancy Evaluation of Energy and Indoor Environment Quality in Green
Buildings: A Review. In Proceedings of the 3rd International Conference on Smart and Sustainable Built
Environments, Delft, The Netherlands, 15 June 2009; pp. 1–7.
33.
Diamond, R. Evaluating the Energy Performance of the First Generation of LEED-Certified Commercial Buildings;
Lawrence Berkeley National Laboratory: Berkeley, CA, USA, 2011.
34.
Oates, D.; Sullivan, K. Postoccupancy energy consumption survey of Arizona’s LEED new construction
population. J. Constr. Eng. Manag. 2012,138. [CrossRef]
35.
Turner, C.; Frankel, M. Energy Performance of LEED for New Construction Buildings; Final Report; US Green
Building Council: Washington, DC, USA, 2008.
36.
Newsham, G.; Mancini, S.; Birt, B.J. Do LEED-certified buildings save energy? Yes, but. Energy Build.
2009
,
41, 897–905. [CrossRef]
37.
Storey, S. Application of Life-Cycle Approaches for the Evaluation of High Performance Buildings. Ph.D.
Thesis, University of British Columbia, Vancouver, BC, Canada, 2014.
38.
Bartlett, K.; Brown, C.; Chu, A.-M.; Ebrahimi, G.; Gorgolewski, M.; Hodgson, M.; Mallory-Hill, S.; Ouf, M.;
Scannell, L.; Turcato, A. Do our green buildings perform as intended. In Proceedings of the 2014 Sustainable
Building Challenge (SB14), Barcelona, Spain, 28–30 October 2014. Available online: http://iisbecanada.ca/
umedia/cms_files/Conference_Paper_1.pdf (accessed on 30 July 2018).
39.
Touchie, M.F.; Pressnail, K.D. Using suite energy-use and interior condition data to improve energy modeling
of a 1960s MURB. Energy Build. 2014,80, 184–194. [CrossRef]
40.
Touchie, M.F.; Binkley, C.; Pressnail, K.D. Correlating Energy Consumption with Multi-Unit Residential
Building Characteristics in the City of Toronto. Energy Build. 2013,66, 648–656. [CrossRef]
41.
Bordass, W.; Cohen, R.; Field, J. Energy Performance of Non-Domestic Buildings: Closing the Credibility Gap;
Building Performance Congress: Frankfurt, Germany, 2004. Available online: http://www.usablebuildings.
co.uk/Pages/Unprotected/EnPerfNDBuildings.pdf. (accessed on 30 November 2018).
42.
Gram-Hanssen, K.; Georg, S. Energy performance gaps: Promises, people, practices. Build. Res. Inf.
2018
,46,
1–9. [CrossRef]
43.
Sunikka-Blank, M.; Galvin, R.; Behar, C. Harnessing social class, taste and gender for more effective policies.
Build. Res. Inf. 2018,46, 114–126. [CrossRef]
Sustainability 2018,10, 4829 20 of 22
44.
van den Brom, P.; Meijer, A.; Visscher, H. Performance gaps in energy consumption: Household groups and
building characteristics. Build. Res. Inf. 2018,46, 54–70. [CrossRef]
45.
Gram-Hanssen, K. Residential heat comfort practices: Understanding users. Build. Res. Inf.
2010
,38, 175–186.
[CrossRef]
46.
Robinson, J.F.; Foxon, T.J.; Taylor, P.G. Performance gap analysis case study of a non-domestic building.
Proc. Inst. Civ. Eng. Eng. Sustain. 2016,169, 31–38. [CrossRef]
47.
Bartlett, K.; Brown, C.; Chu, A.-M.; Ebrahimi, G.; Gorgolewski, M.; Hodgson, M.; Issa, M.; Mallory-Hill, S.;
Ouf, M.; Scannell, L.; et al. Poster for Canadian Building Performance Evaluation Project. In Proceedings
of the 2014 Sustainable Building Challenge (SB14) Conference, Barcelona, Spain, 28–30 October 2014.
Available online: http://iisbecanada.ca/umedia/cms_files/Canada_Overview_Poster_V5.pdf (accessed on
17 December 2018).
48.
Schwartz, P. The Art of the Long View: Planning for the Future in an Uncertain World; Doubleday: New York, NY,
USA, 1991; ISBN 978-0385267328.
49.
Wiebe, K.; Zurek, M.; Lord, S.; Brzezina, N.; Gabrielyan, G.; Libertini, J.; Westhoek, H. Scenario development
and foresight analysis: Exploring options to inform choices. Annu. Rev. Environ. Resour.
2018
,43, 1.1–1.26.
[CrossRef]
50.
Robinson, J. Energy backcasting A proposed method of policy analysis. Energy Policy
1982
,10, 337–344.
[CrossRef]
51.
Robinson, J.; Burch, S.; Talwar, S.; O’Shea, M.; Walsh, M. Envisioning sustainability: Recent progress in the
use of participatory backcasting approaches for sustainability research. Technol. Forecast. Soc. Chang.
2011
,
78, 756–768. [CrossRef]
52.
Cole, R.J.; Robinson, J.; Brown, Z.; O’Shea, M. Re-contextualizing the notion of comfort. Build. Res. Inf.
2008
,
36, 323–336. [CrossRef]
53.
Brown, Z.; Cole, R.J. Influence of occupants’ knowledge on comfort expectations and behaviour. Build. Res.
Inf. 2009,37, 227–245. [CrossRef]
54.
Brown, Z. Occupant Comfort and Engagement in Green Buildings: Examining the Effects of Knowledge,
Feedback and Workplace Culture. Ph.D. Thesis, University of British Columbia, Vancouver, BC,
Canada, 2009.
55.
Brown, Z.; Dowlatabadi, H.; Cole, R. Feedback and adaptive behaviour in green buildings. Intell. Build. Int.
2009,1, 296–315. [CrossRef]
56.
Way, M.; Bordass, B. Making feedback and post-occupancy evaluation routine 2: Soft landings–involving
design and building teams in improving performance. Build. Res. Inf 2005,33, 353–360. [CrossRef]
57. Grbich, C. Qualitative Data Analysis: An Introduction, 2nd ed.; Sage: Abingdon-on-Thames, UK, 2013.
58. Saldaña, J. The Coding Manual for Qualitative Researchers, 2nd ed.; Sage: Abingdon-on-Thames, UK, 2013.
59.
Kvale, S.; Brinkmann, S. Interviews: Learning the Craft of Qualitative Research, 2nd ed.; Sage:
Abingdon-on-Thames, UK, 2009.
60.
Bazeley, P. Master Class: Theory and Practice of Qualitative Data Analysis in NVivo, NVivo Symposium
with Pat Bazeley. 2015. SFU Harbour Centre, Vancouver: Symposium conducted by SFU Library’s Research
Commons. Available online: https://scarp.ubc.ca/school/weekly-digest/2015/03/nvivo-symposium-and-
advanced-workshop-qualitative-data-analysis-april (accessed on 17 December 2018).
61. Bazeley, P. Analysing qualitative data: More than ‘identifying themes’. Malays. J. Qual. Res. 2009,2, 6–22.
62.
Coleman, S. Normalizing Sustainability in a Regenerative Building: The Social Practice of Being at CIRS.
Ph.D. Thesis, University of British Columbia, Vancouver, BC, Canada, 2016.
63.
Berkhout, F. Normative expectations in systems innovation. Technol. Anal. Strateg.
2006
,18, 299–311.
[CrossRef]
64.
Janda, K.B.; Topouzi, M. Telling tales: Using stories to remake energy policy. Build. Res. Inf.
2015
,43, 516–533.
[CrossRef]
65.
Janda, K.B.; Topouzi, M. Closing the loop: Using hero stories and learning stories to remake energy policy.
In Rethink, Renew, Restart, Proceedings of the ECEEE Summer Study, Belambra Les Criques, Toulon/Hyères,
France, 3–8 June 2013; European Council for an Energy-Efficient Economy: Brussels, Belgium, 2013;
ISBN 978-91-980482-3-0.
Sustainability 2018,10, 4829 21 of 22
66.
Rotmann, S.; Mourik, R.; Goodchild, B. Once Upon a Time... How to tell a good energy efficiency story that
‘sticks’. In First Fuel Now, Proceedings of the ECEEE 2015 Summer Study, Stockholm, Sweden, 1–6 June 2015;
European Council for an Energy-Efficient Economy: Brussels, Belgium, 2015; ISBN 978-91-980482-7-8.
67.
Chu, A.-M.; Ebrahimi, G.; Scannell, L.; Save, P.; Hodgson, M.; Bartlett, K.; Gorgolewski, M. Building
Performance Evaluation for Jim Pattison Centre of Excellence in Sustainable Building Technologies and
Renewable Energy Conservation. 2015. Available online: http://iisbecanada.ca/umedia/cms_files/Report_
-_JPCOE_Final_Feb2015.pdf (accessed on 17 December 2018).
68.
Thatcher, A.; Milner, K. Changes in productivity, psychological wellbeing and physical wellbeing from
working in a ‘green’ building. Work 2014,49, 381–393. [CrossRef] [PubMed]
69.
Paevere, P.; Brown, S. Indoor Environment Quality and Occupant Productivity in the CH2 Building:
Post-Occupancy Summary. Report No. USP2007/23. Available online: http://dro.deakin.edu.au/eserv/DU:
30018085/luther-indoorenvironment-2008.pdf (accessed on 30 November 2018).
70.
McCunn, L.J.; Gifford, R. Do green offices affect employee engagement and environmental attitudes?
Architect. Sci. Rev. 2012,55, 128–134. [CrossRef]
71.
Gray, G.; Guppy, N. Successful Surveys: Research Methods and Practice, 4th ed.; Thomson Nelson: Nashville,
TN, USA, 2008.
72.
Deuble, M.P.; de Dear, R.J. Is it hot in here or is it just me? Validating the post-occupancy evaluation. Intell.
Build. Int. 2014,6, 112–134. [CrossRef]
73.
Dunn, E.W.; Wilson, T.D.; Gilbert, D.T. Location, location, location: The misprediction of satisfaction in
housing lotteries. Pers. Soc. Psychol. Bull. 2003,29, 1421–1432. [CrossRef] [PubMed]
74.
Leaman, A.; Bordass, B. Assessing building performance in use 4: The Probe occupant surveys and their
implications. Build. Res. Inf. 2001,29, 129–143. [CrossRef]
75.
Mang, P.; Reed, B. Designing from place: A regenerative framework and methodology. Build. Res. Inf.
2012
,
40, 23–38. [CrossRef]
76.
Cranz, G.; Lindsay, G.; Morhayim, L.; Lin, A. Communicating sustainability: A postoccupancy evaluation of
the david brower center. Environ. Behav. 2014,46, 826–847. [CrossRef]
77.
Deuble, M.P.; de Dear, R.J. Green occupants for green buildings: The missing link? Build. Environ.
2012
,56,
21–27. [CrossRef]
78.
Leaman, A.; Bordass, B. Are users more tolerant of ‘green’ buildings? Build. Res. Inf.
2007
,35, 662–673.
[CrossRef]
79.
Rapoport, A. Human Aspects of Urban form–Towards a Man-Environment Approach to Urban form and Design;
Pergamon Press: New York, NY, USA, 1977; ISBN 9781483156828.
80. Leaman, A. Dissatisfaction and office productivity. Facilities 1995,13, 13–19. [CrossRef]
81.
Guerra-Santina, O.; Aidan, C.A. In-use monitoring of buildings: An overview of data collection methods.
Energy Build. 2015,93, 189–207. [CrossRef]
82.
Day, J.K.; O’Brien, L. Oh behave! Survey stories and lessons learned from building occupants in
high-performance buildings. Energy Res. Soc. Sci. 2011,31, 11–20. [CrossRef]
83.
Altomonte, S.; Schiavon, S. Occupant Satisfaction in LEED and BREEAM-certified office buildings.
In Proceedings of the 36th International Conference on Passive and Low Energy Architecture, Cities,
Buildings, People: Towards Regenerative Environments, Los Angeles, CA, USA, 11–13 July 2016.
84.
Geng, Y.; Ji, W.; Zhu, Y. The impact of thermal environment of occupant IEQ perception and productivity.
Build. Environ. 2017,121, 158–167. [CrossRef]
85.
Newsham, G.; Mancini, S.; Veitch, J.; Marchand, R.; Lei, W.; Charles, K. Control strategies for lighting and
ventilation in offices: Effects on energy and occupants. Intell. Build. Int. 2009,1, 101–121. [CrossRef]
86.
Leder, S.; Newsham, G.R.; Veitch, J.A.; Mancini, S.; Charles, K.E. Effects of office environment on employee
satisfaction: A. new analysis. Build. Res. Inf. 2016,44, 34–50. [CrossRef]
87.
Choi, J.; Loftness, V.; Aziz, A. Post-occupancy evaluation of 20 office buildings as basis for future IEQ
standards and guidelines. Energy Build. 2012,46, 167–175. [CrossRef]
88.
Dascalaki, E.; Sermpetzoglou, V. Energy performance and indoor environmental quality in Hellenic schools.
Energ Build. 2011,43, 718–727. [CrossRef]
Sustainability 2018,10, 4829 22 of 22
89.
Chang, C.; Touchie, M.F. Investigating Wintertime Thermal Comfort of Post-War Multi-Unit Residential
Buildings using Surveys and In-suite Monitoring. In Proceedings of the 15th Canadian Conference on
Building Science and Technology, CCBST, Vancouver, BC, Canada, 6–8 November 2017.
90.
Touchie, M.F.; Tzekova, E.S.; Siegel, J.A.; Purcell, B.; Morier, J. Evaluating Summertime Overheating in
Multi-Unit Residential Buildings using Surveys and In-Suite Monitoring. In Proceedings of the 13th
International Conference on Thermal Performance of the Exterior Envelopes of Whole Buildings XIII,
Clearwater, FL, USA, 5–8 December 2016.
©
2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).
