flEECe, an energy use and occupant behavior dataset for net-zero energy affordable senior residential buildings

Abstract

The behaviors of building occupants have continued to perplex scholars for years in our attempts to develop models for energy efficient housing. Building simulations, project delivery approaches, policies, and more have fell short of their optimistic goals due to the complexity of human behavior. As a part of a multiphase longitudinal affordable housing study, this dataset represents energy and occupant behavior attributes for 6 affordable housing units over nine months in Virginia, USA which are not performing to the net-zero energy standard they were designed for. This dataset provides researchers the ability to analyze the following variables: energy performance, occupant behaviors, energy literacy, and ecological perceptions. Energy data is provided at a 1 Hz sampling rate for four circuits: main, hot water heater, dryer, and HVAC. Building specifications, occupancy, weather data, and neighboring building energy use data are provided to add depth to the dataset. This dataset can be used to update building energy use models, predictive maintenance, policy frameworks, construction risk models, economic models, and more.

Full text

1
SCIENTIFIC Data | (2019) 6:291 | https://doi.org/10.1038/s41597-019-0275-3
www.nature.com/scientificdata




 














Building occupant education is an important research direction providing an opportunity to maximize the e-
cacy of energy eciency investments in the housing industry1. Despite increasing eorts by property managers
to educate tenants on technology in housing units, many residents lack an understanding of energy ecient
technologies. As technology becomes further integrated into housing units, user education will become more
of a factor in the optimization of residential energy use. For example, McCoy et. al. reported that residents that
received education on their apartments had a lower average energy usage monthly and annually (over 3 years) by
almost 15% (14.8%) and a lower energy bill by $10.56 per month2.
Providing an education to residents is not only the appropriate ethical choice, but also it can be a powerful
business strategy for builder-developers. In the context of this dataset, electricity pricing in Virginia is trending
upwards with United States and global rates. Virginia electricity prices have increased by an average of 1.5% per
year over 25 years and 3% annually over the last ten years3. Traditionally, utilities provide education in the form
of energy usage feedback with a “lag” which becomes more detrimental as prices rise. Human factors researchers
have reported that people are generally poor at managing systems with lags in information and delayed feedback
loops4,5. For energy ecient housing systems, energy monitoring systems exist which can report energy use in
real-time using an Energy Feedback Display (EFD) for occupants (meetnexi.com).
Using the same hardware, building managers can utilize cloud-based soware platforms to monitor and create
performance reporting, instantly being aware of system failures or ineciencies. Studies have shown that EFDs
have improved behavior towards a more energy ecient lifestyle and resulted in 10–15% monthly energy use
reductions given: (a) frequent feedback, (b) provided over long periods of time, (c) with appliances characterized
individually, (d) presented in clear, appealing ways, and (e) utilizing computerized, interactive tools6–9.
In this paper, we release our dataset EECe10. is dataset is the outcome of a unique case study within a longi-
tudinal multiphase study to explore the impacts of feedback and education on energy-use behaviors of seniors in
1Charles E. Via Jr. Department of Civil and Environmental Engineering, Virginia Tech, Blacksburg VA, 24061,
United States. 2Myers-Lawson School of Construction, Virginia Tech, Blacksburg VA, 24061, United States. *email:
freddyp@vt.edu; jazizade@vt.edu


Content courtesy of Springer Nature, terms of use apply. Rights reserved

2
SCIENTIFIC Data | (2019) 6:291 | https://doi.org/10.1038/s41597-019-0275-3
www.nature.com/scientificdata
www.nature.com/scientificdata/
aordable housing units. e units were designed to be net-zero energy and followed the EarthCra Multifamily
design guidelines11, adhering to the Low-Income Housing Tax Credit requirements12. e participants, in this
study, participated in an educational workshop and were provided with a budget-based (giving feedback based
on remaining budget) energy monitoring system to receive feedback on their electricity usage in real-time. e
following two questions motivated the collection of the presented data.
• How does targeted energy education and energy consumption feedback impact energy literacy of seniors in
net-zero energy apartment units?
• How does targeted energy education and energy consumption feedback impact energy consumption of sen-
iors in net-zero energy apartment units?
e use of “how” in the questions reects the qualitative portion of this study which is not presented in this
dataset. e dataset described here is solely the quantitative portion of this mixed methods case study. e quan-
titative data collected can be analyzed in multiple ways to expand beyond the original objectives of our study.
How occupants shi their energy use behaviors is linked to their environmental perceptions, previous energy
behaviors, and level of education13,14. Targeted occupant education has shown potential to be an eective method
for reducing energy consumption15,16 and in a previous phase of this study, education strongly correlated with
reduced energy consumption2. A unique aspect in our study is the focus on senior residents who are not nan-
cially incentivized to conserve energy – a topic which needs further investigation17 and is critical in the context
of aordable housing. Furthermore, this dataset reects a case study framed according to the concepts in the U.S.
Department of Energy (DOE) Energy Literacy guide18 through coded survey responses.
Scholars may use the EECe dataset to investigate:
• Energy use behaviors in the context of the US DOE Energy literacy concepts
• Long-term resident consumption patterns
• Evaluate the performance of energy efficient water-heaters, mini-splits, and dryers in affordable senior
housing
• Understand the impact of climate on consumption
• Non-intrusive energy monitoring techniques for data with dierent resolution

 is dataset was collected in the context of a singular case study in continuation of a
multiphase state-wide longitudinal study in Virginia1. Building upon ndings in the last phase of the longitudinal
study, this case study utilizes interviews, eld observations, surveys, and energy consumption data to explore the
impact of targeted education on residents’ energy consumption. e dataset presented in this paper has detailed
information on participants energy consumption, perceived behavior, beliefs, and energy literacy. is dataset is
an addition to the increased amount of building monitoring data we need for improving residential energy use
through advanced metering and monitoring19,20.
e property under examination is a senior living Low Income Housing Tax Credit (LIHTC) project in
Richmond, Virginia that has been being monitored since 2013. is property was selected for its unique occu-
pant energy consumption patterns and design features. A unique feature of this study pertains to that fact that
the occupants are not nancially incentivized to conserve energy given that the energy bills in these buildings
are paid by the property managers. While the property was performing better than average building nationally
and in Virginia, the property was falling short of its net-zero energy goal2. As reected in Tables1 and 2, the
one-bedroom housing units were designed to be grid-tied net-zero housing units with all electric energy ecient
appliances, tight building envelope, energy ecient glazing, and solar PV arrays on site. e windows and their
shadings are manually operable and there is no known outside factors (such as noise or air pollution) to prevent
occupants from opening the windows. Building characteristics can be found within the dataset in the tabular data
excel le. A color coded illustration of the building and unit layout, material properties, window to wall ratio, and
heating and cooling specications can be found on the “Architectural Data” tab of the dataset10. e 6 units which
are the focal point of this study come together to form a 7-unit building. Although all 7 units were monitored, due
to unreliable data from one unit, the EECe dataset only includes circuit level data from 6 units. Also provided in
the dataset is monthly data for the adjacent 32-unit building which is a part of the same housing property.
 The study protocol used for this study was approved by the Virginia Tech
Institutional Review Board. Participants of this study were not subject to any known risks and informed consent
was provided from all subjects.
 is dataset includes energy use data, collected by advanced circuit level
energy monitors and standard utility energy meters. e NEXI energy feedback device6 captured second level
data (i.e., @1 Hz) for 6 units in this study on four circuits: main, hot-water heater, dryer, and HVAC (mini-split)
system using 100A non-invasive split core current transformer clamps with ±1% accuracy. NEXI uses a proprie-
tary data collection and aggregation process designed to work with their unique color-wheel display that provided
users with simplied feedback. e detailed energy data from NEXIs including power, voltage, current, and phase
angle for four circuits are included in the dataset as raw CSV les. e calibration process for the NEXI data is
provided in the dataset as well. Each unit was sub-metered and energy use data was collected per month through
a utility benchmark tracker, WegoWise (wegowise.com), which logged ve years of monthly energy consumption
Content courtesy of Springer Nature, terms of use apply. Rights reserved

3
SCIENTIFIC Data | (2019) 6:291 | https://doi.org/10.1038/s41597-019-0275-3
www.nature.com/scientificdata
www.nature.com/scientificdata/
data for the property. Paper utility bills were used to spot check the data collected by both the NEXI devices and
WegoWise platform.
Given the targeted community for this study, we opted for a hard-copy survey which was later digitized for
data analysis. Participants were surveyed on site in a community room typically used for social events at the
property. e survey combined instruments from literature to measure energy behaviors, perceptions, and lit-
eracy2,21,22 adapted for use in this case. Surveying in person requires a great deal of exibility and eort. Surveys
were designed to be senior-friendly utilizing large fonts, high contrast, and large formatting for recording
responses. Even so, multiple participants needed assistance while taking the survey. A team of two researchers
was present to administer the survey with one researcher reading the questions aloud to the group and the other
researcher assisting participants one on one when necessary. Property managers helped with recruitment and
provided logistics during the education and survey data collection. A $25 gi card was provided by the research
team during the pre and post survey, to participating residents to aid in recruitment. e survey questions were
coded to the energy literacy guide providing descriptive data on the lessons learned by the study participants.
e recruitment eorts resulted in twenty residents attending a one-hour community meetings and receiving
a Tenant Education Intervention (TEI). Of the twenty residents, twenty successful records were developed for
analysis. rough stratied sampling, six residents who received the TEI, also were provided with an in-home
display. It is important to note that the researchers were able to leverage the existing energy use data from pre-
vious work to compare monthly energy usage and the ecacy of resident education eorts. An overview of the
behavioral data sample is provided in Table2.
 ere are two interventions in this study delivered in the following formats: 1)
residents who received a Targeted Energy Intervention (TEI) (see Fig.1a) on March 28th, 2017 and 2) residents
who received a TEI + an energy feedback device (see Fig.1b–d) with the EFD installation occurring on July 6th,
2017. e TEI consisted of the authors guiding residents through seven educational videos and a ten-minute
PowerPoint presentation that featured technologies specic to their apartment unit.
Principles and concepts in the DOE Energy Literacy guide have a great deal of overlap and the most critical
concepts were selected as a scope for this study. Concepts were prioritized in regard to their connection to the TEI
and in-home experiences. Aer the concepts were prioritized, the scope of this study was set to t the limitations
of the data collection and analysis methods utilized. e research team selected ve concepts to bound this study
which link to daily energy decisions and the targeted energy education we provided: (1) human use of energy
is subject to limits and constraints, (2) conservation is one way to manage energy resources, (3) electricity is
Characteristics Building 1 Building 2
Climate zone US 4A US 4A
# of apartments 7 (6 were presented in EECe) 32
# of buildings 1 1
PV-system 16.9 m2, south-facing, 2.43 kWp 166.1 m², south-facing, 2.43 kWp
Living area 668 2 (62.1 m2)708 ² (65.9 m2)
Building volume 6,178 cf (175 m3)6,372 cf (180 m3)
Heating system Air-source heat pump 18 kBtuh, 9 HSPF Air-source heat pump 18 kBtuh, 9 HSPF
Cooling system Air-source heat pump, 18 SEER Air-source heat pump, 18 SEER
Distribution 2, Duc tless air systems 2, Ductless air systems
Water heating Standard electric storage (0.92 EF) Standard electric storage (0.92 EF)
Ventilation Energy recovery ventilator Energy recovery ventilator
Windows U-value SHGC (g-value) 0.30 BTUh/2/°F (0.95/W/m2K) 0.28 0.30 BTUh/2/°F (0.95/W/m2K) 0.28
Wall U-value 0.04 BTUh/2/°F (0.13/W/m2K) 0.04 BTUh/2/°F (0.13/W/m2K)
Roof/Attic U-value 0.03 BTUh/2/°F (0.11/W/m2K) 0.03 BTUh/2/°F (0.11/W/m2K)
Air tightness 4.1 ACH50 6.1 ACH50
Tab le 1. Building specications. All electric 1-bedroom units.
Meta Data Total Units Targeted Education (TEI) TEI + In-home Display Control
Number of
Units 38 20 6 18
Energy Data Main
aggregate data
per month Main aggregate data per month
Main aggregate data, 3 sub-circuits
(dryer, water heater, and air
conditioning) – 1-sec, 1-min, and
15-min resolution
Main aggregate
data per month
Survey Data Pre and Post intervention energy
perceptions, behaviors, and
literacy
Pre and Post intervention energy
perceptions, behaviors, and literacy
Tab le 2. Behavioral data sample summary.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

4
SCIENTIFIC Data | (2019) 6:291 | https://doi.org/10.1038/s41597-019-0275-3
www.nature.com/scientificdata
www.nature.com/scientificdata/
generated in multiple ways, (4) social and technological innovations aect the amount of energy used by society,
and (5) energy use can be calculated and monitored.
e research team considered several factors when developing the daily energy budget for the EFDs. First,
the RESNET accredited home energy rating (HERS) energy simulations that were developed during design were
reviewed. Second, ve years of historical energy use data for the selected apartments were analyzed to develop a
measured average kWh/day per apartment. Finally, the team established the energy reduction goals (budgeted
kWh/day) to be programmed into the NEXI energy feedback display by multiplying the simulated consumption
by the EFD manufacturer’s suggested multiplier for each unit ranging from 1.7 to 1.8. Table 3 provides an over-
view of the estimated, measured, and budgeted energy goals.
 Weather data for the project location is publicly available through the National
Oceanic Atmospheric Administration (NOAA)23 and Weather Underground24. Historic weather data allows for
a researcher to assess the impact of year to year weather variance and climate to weather variances on the site. A
baseline of 65 °F/18 °C is used to calculated Heating Degree Day (HDD) and Cooling Degree Day (CDD) with
outdoor temperatures measured at the NOAA Richmond International Airport Weather Station (RIAWS). e
station is located at 37.51151°, −77.32344° and is approximately 5 miles (8 kilometers) from the case study site.
We utilized Weather Underground to retrieve ten years of monthly HDD and CDD from RAIWS.
Next, the team referenced the Typical Meteorological Year (TMY3) data for Richmond, VA, USA to develop
a climate benchmark. TMY3 data represents a 30-year average benchmark and was retrieved from the ASHRAE
Handbook of Fundamentals25. TMY3 data is commonly employed in energy simulation models. e authors
leveraged the TMY3 data to unpack 1) simulated performance versus observed performance, 2) yearly weather
variance compared to TMY3, and 3) HDD/CDD during the survey period versus the TMY3.
Fig. 1 Educational and feedback instruments leveraged in the case study. (a) six residents received Targeted
Education Intervention (TEI) and an Energy Feedback Display (EFD). e EFD was programmed with a daily
energy budget that reected energy behavior with a dynamic color display; (b) EFD interface reset at midnight
showing minimal instantaneous use on le with green color, and full daily budget remaining on right with green
color; (c) high real-time consumption on le with fuchsia color, and 20% of daily budget consumed on right
with yellow color; (d) low real-time consumption on le with green color, and 70% of daily budget consumed
on right with red color.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

5
SCIENTIFIC Data | (2019) 6:291 | https://doi.org/10.1038/s41597-019-0275-3
www.nature.com/scientificdata
www.nature.com/scientificdata/

e EECe dataset can be found on the Open Science Framework10. Energy data from each unit was collected
utilizing three methods: 1) utility bills in.pdf format, 2) WegoWise utility tracking service, and 3) NEXI energy
monitor data loggers. e WegoWise and NEXI data are present in this data set. e utility bills were not stored
or shared publicly to protect participant identities. Utility bills were used to spot check the other two records. e
WegoWise data is provided in a Microso Excel table at a monthly interval (kWh/month/unit). e WegoWise
data represents 38 apartments, with 59 contiguous months of data from July 2013 to March 2018.Tables 4 and 5
describe the dataset lesand the monitored variables.
e NEXI data is provided as raw (uncalibrated) CSV les and calibrated CSV les. Each CSV has 11 columns.
Column A is the timestamp. Columns B-F represent the measured electrical current. ese numbers need to be
processed through the calibration formula to get amperage values. Column G represents the voltage readings and
must be processed through a conversion formula for voltage. A guide for the calibration process is included in the
dataset for those who would like develop scripts for managing uncalibrated data from a NEXI device. is will be
particularly useful in the future for cloud computing processes using the NEXI’s WIFI capabilities which allow for
network connected NEXI’s to upload data directly to a server.
Columns H through L are a representation of phase angle. is was an experimental feature, it may provide
some information, but the feature remains in alpha and hasn’t been tested for reliability to the same degree as the
Voltage and Current readings. e NEXI calculates time stamps the number of seconds since July 7, 2017. e
data provided starts at July 7, 2017 and ends at March 22, 2018.
In the data validation folder of the dataset, timeseries analysis are provided exemplifying one of our processes
to ensure the data was valid before advanced analysis and validation processes were executed. It is important to
note that there were no single missing seconds, and under a day’s worth of time gaps in the data signifying the
NEXI’s were properly recording data consistently as long as they were powered. e identied time gaps most
likely reect power outages as the gaps were registered on multiple independent NEXI devices at the same time.
Other data validation processes have been exemplied showing the ability for the data to be disaggregated to
discover the use of the dryer at multiple settings.

In this section, we have presented the results of analyzes on the dataset to ensure that they follow the expected pat-
terns. As shown in Fig.2, the energy data that was captured at the appliance (sub-circuit) level for the heat pump,
dryer, and water heater, manifests rationale variability and reects the expected patterns in the units and validates
the quality of the data. Figure2a,b shows the variation of energy consumption for the heat pump. Figure2a shows
the variations of daily energy consumption of heat pumps across the period of data collection. e heat pump
represents both cooling and heating load variations. As expected, the elevated energy consumption around July
and August (representing the cooling load) decreases due to seasonal transition between August and November,
where the heating load causes a sharp increase in energy consumption. Furthermore, these graphs show the var-
iability across dierent units reecting the diversity in occupant behavior when it comes to operating the loads.
Figure2b shows the average daily variation of the heat pump energy averaged across all units for all days. is
graph illustrates the expected variation of energy consumption which reects the activities of occupants at home.
As noted, the occupants of the units are senior citizens and therefore, during the day energy consumption remains
relatively high.
On the other hand, the seasonality pattern is not observed in the energy consumption of dryers as shown in
Fig.2c. In other words, the observations show the distributed use of the dryers and the variations across dierent
units. e interactional behavior impact could be better observed in Fig.2d, in which the dierences in dryer
use including the time of use and the demand from each unit has been highlighted. e observed variations of
the appliance use patterns support the validity of energy use with respect to occupant behavioral patterns across
dierent units.
Description Unit
AUnit
BUnit
CUnit
DUnit E Unit
F
Simulated kWh/day 3.94 3.94 3.94 3.78 3.62 3.62
Measured kWh/day 7.99 8.96 9.39 11.03 12.07 9.40
Budgeted kWh/day 6.78 6.78 6.78 6.78 6.20 6.20
Tab le 3. Energy Feedback Display budgeting.
Sample Number of Samples Temporal range Description Data
Tabular Data 38 units, 20
Building occupants July 2013–March
2018
Occupant Demographics, Monthly-time series
power data, Pre and Post educational intervention
energy literacy, behaviors, and environmental beliefs
EECe Dataset
5_15_19.xlsx
NEXI data 6 units July 7, 2017–
March-22-2018 Second data at the circuit level for the entire unit,
dryer, hot water heater, and HVAC (mini-split) unit A-F_EST.csv
Tab le 4. Dataset Description.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

6
SCIENTIFIC Data | (2019) 6:291 | https://doi.org/10.1038/s41597-019-0275-3
www.nature.com/scientificdata
www.nature.com/scientificdata/
Furthermore, Fig.2e,f represent the daily energy consumption and demand of water heaters at dierent
time scales. e former presents the daily variations for the water heater energy consumption across the entire
period of data collection. Although most of the units show similar load behavior with relatively constant energy
Fig. 2 Energy consumption patterns at appliance level. e variations reect the changes in occupant behavior
in these units that were monitored.
Data
Category Subcategory Monitored Variables
Energy Hot Water Heater Electricity
Energy Dryer Electricity
Energy Heating/cooling (HC) Electricity
Inhabitants Attitudes (AD) Ecological beliefs
Inhabitants Attributes (AT) Consumption Behaviors
Tab le 5. Dataset Mapping to Mahdavi and Taheri’s (2017) ontology19.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

7
SCIENTIFIC Data | (2019) 6:291 | https://doi.org/10.1038/s41597-019-0275-3
www.nature.com/scientificdata
www.nature.com/scientificdata/
consumption, the occupant behavior impact could be also observed as it has been reected on the variations of
consumption across dierent days. For unit E the change in consumption behavior could be observed across
dierent months. Although these units are single occupancy units, some units have been occupied by up to two
occupants at dierent times of the data collection period. Figure2f shows the average daily variation of energy
demand for dierent units. Again, this graph shows the dierences in behavior of the occupants in the targeted
units and shows reects the impact of behavioral dierences in interacting with loads.
Figure3 illustrates the relationship between energy use for the heat pump circuit during a cooling day. As
expected, the heat pumps begin to draw energy as the temperature begins to rise in the morning and loads drop
with temperature decline in the evening. e dierence in behaviour can also be observed in the energy use.
Figure4 illustrates the dierence in monthly energy use for the 38 unit sample (from which the nal 20 unit
survey sample, and 6 unit NEXI sample were selected) over a year. e TEI + EFD units, TEI units, and control
units, were compared to analyze the impact of the TEI on energy consumption over a year. e TEI + EFD units
average monthly energy consumption was much lower than the TEI units and control group which was expected
due to the TEI + EFD units having solar generation osetting their use.

e EECe dataset is focused around a set of time-series datasets which can be analyzed with a variety of so-
ware packages. We encourage the use of Python, which was used in our example les via the Jupyter Notebook
Platform to create very accessible HTML les. e use of Plotly is also recommended to create interactive visu-
alizations of the time-series datasets. e eece dataset also provided access to detailed information about the
occupants which to our knowledge has not been combined with timeseries data for seniors in aordable energy
Fig. 3 Relationship between heat pump energy consumption and temperature. Sample NEXI data for the 6
units in the EECe dataset for a day.
Fig. 4 Monthly average energy use (kWh). Representing complete 38 unit sample with interviewed units, TEI
units, and control units. Lines represent a yearly average for each group.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

8
SCIENTIFIC Data | (2019) 6:291 | https://doi.org/10.1038/s41597-019-0275-3
www.nature.com/scientificdata
www.nature.com/scientificdata/
ecient housing. e contextual data (perceptions, energy literacy, environmental beliefs), are being prepared for
another publication and we encourage users of this dataset to contact the corresponding authors for assistance in
leveraging this data. e EECe dataset will continue to be updated as the longitudinal multiphase study which
allowed for the collection of the EECe dataset is continuing. e authors encourage dialogue from the commu-
nity to allow for the most useful data to be contributed to the greater scientic community.

An example of the Python code used to analyze the raw energy monitor data is publicly available on the Open
Science Framework data repository10. The data can also be analyzed using software which handles tabular
timeseries data such as R or MATLAB. e code used to aggregate and plot the data for this paper are publicly
available on the data repository on OSF.
Received: 30 May 2019; Accepted: 19 September 2019;
Published: xx xx xxxx

1. McCoy, A. P. et al. e Impact of Energy Ecient Design and Construction on LIHTC Housing in Virginia. 102 (A eport by the
Virginia Center for Housing esearch (VCH) at Virginia Tech for Housing Virginia 2015).
2. McCoy, A., Zhao, D., Paige, F. & Tech, V. Sustaining Energy Eciency: Longitudinal Evidence of Virginia’s Low-Income Housing Tax
Credit Prope rties. 33 (A eport by the Virginia Center for Housing esearch (VCH) at Virginia Tech for Housing Virginia 2017).
3. U.S. Energy Information Administration. Electric Power Monthly with data for February 2019. 266 (EIA, 2019).
4. Brehmer, B. Dynamic decision maing: Human control of complex systems. Acta Psychol. (Amst.) 81, 211–241 (1992).
5. Sterman, J. Business Dynamics: Systems inking and Modeling for a Complex World. (Irwin/McGraw-Hill, 2000).
6. Ehrhardt-Martinez, ., Donnelly, . A. & Laitner, J. A. Advanced Metering Initiatives and esidential Feedbac Programs: A Meta-
eview for Household Electricity-Saving Opportunities. 2010. In eport Number E105 for theAmerican Council for an Energy-
Ecient Economy140 (2010).
7. Fischer, C. Feedbac on household electricity consumption: a tool for saving energy? Energy Ec 1, 79–104 (2008).
8. Ouyang, J. & Hoao, . Energy-saving potential by improving occupants’ behavior in urban residential sector in Hangzhou City,
China. Energy Build 41, 711–720 (2009).
9. Stinson, J., Willis, A., Williamson, J. B., Currie, J. & Smith, . S. Visualising Energy Use for Smart Homes and Informed Users. Energy
Procedia 78, 579–584 (2015).
10. Paige, F. & Agee, P. EECe, an Energy Use and Occupant Behavior Dataset for Net Zero Energy Aordable Senior esidential
Buildings. Open Science Framework, https://doi.org/10.17605/OSF.IO/2AX9D (2019).
11. ECMF Team. EarthCra MultifamilyTechnical Guidelines (2016).
12. VHDA. VHDA Low Income Housing Tax Credit Manual (2019).
13. Nair, G., Gustavsson, L. & Mahapatra, . Factors inuencing energy eciency investments in existing Swedish residential buildings.
Energy Policy 38, 2956–2963 (2010).
14. Poortinga, W., Steg, L. & Vle, C. Values, Environmental Concern, and Environmental Behavior: A Study into Household Energy
Use. Environ. Behav. 36, 70–93 (2004).
15. Delmas, M. A., Fischlein, M. & Asensio, O. I. Information strategies and energy conservation behavior: A meta-analysis of
experimental studies from 1975 to 2012. Energy Policy 61, 729–739 (2013).
16. Zografais, N., Menegai, A. N. & Tsagarais, . P. Eective education for energy eciency. Energy Policy 36, 3226–3232 (2008).
17. McMain, A. H., Malone, E. L. & Lundgren, . E. Motivating esidents to Conserve Energy without Financial Incentives. Environ.
Behav. 34, 848–863 (2002).
18. DOE. Energy Literacy : Essential Principles and Fundamental Concepts for Energy Education (2012).
19. Mahdavi, A. & Taheri, M. An ontology for building monitoring. J. Build. Perform. Simul 10, 499–508 (2017).
20. Jazizadeh, F., Afzalan, M., Beceri-Gerber, B. & Soibelman, L. EMBED: A Dataset for Energy Monitoring through Building Electricity
Disaggregation. In Proceedings of the Ninth International Conference on Future Energy Systems - e-Energy 2018, https://doi.
org/10.1145/3208903.3208939, 230–235 (ACM Press, 2018).
21. Jan DeWaters; Energy Literacy Survey : A Broad Assessment of Energy-related nowledge Attitudes and Behaviors (2009).
22. Dunlap, . E. e New Environmental Paradigm Scale: From Marginality to Worldwide Use. J. Environ. Educ 40, 3–18 (2008).
23. Local Climatological Data Station Details: ICHMOND INTENATIONAL AIPOT, VA US, WBAN:13740 | Climate Data
Online (CDO) | National Climatic Data Center (NCDC). Available at, https://www.ncdc.noaa.gov/cdo-web/datasets/LCD/stations/
WBAN:13740/detail (Accessed: 15th May 2019).
24. ichmond, V. A. History. Weather Underground Available at: /history/daily/us/va/richmond/IC/date/2019-5-15. (Accessed: 15th
May 2019).
25. ASHAE, A. S. of H., efrigerating and Air-Conditioning Engineers. 2013 ASHRAE Handbook: Fundamentals. (American Society
of Heating, efrigerating and Air-Conditioning Engineers Inc, 2013).

e authors would like to thank the Virginia Center for Housing Research (VCHR), which initiated this study
in partnership with Housing Virginia, Virginia Housing Development Authority, and Dominion Energy.
Andrew McCoy, Director of VCHR, played an instrumental role in developing the idea for the original study
which allowed for the creation of this dataset. e authors would also like to thank and acknowledge Santhosh
irukkumaran, Emma Coleman, Greg Pfotenhauer, Dong Zhao, Byung-Jun Kim and Lata Kodali who assisted
in installation of the energy monitoring devices and preliminary analysis of this dataset. Lastly, we would like to
acknowledge all of the study participants in this study who cannot be named to protect their personal privacy.
is data would not exist without their wiliness to share their experience and knowledge with the research team
and the world.

F. Paige, as the project PI, Paige coordinated with all involved parties for the study, provided guidance on
preparing the dataset, provided supplementary occupant data via the survey instrument such as demographics,
environmental perceptions, energy literacy, prepared sections of the manuscript and reviewed all elements of
the manuscript. P. Agee, collected and organized all of the energy data for this study, prepared sections of this
Content courtesy of Springer Nature, terms of use apply. Rights reserved

9
SCIENTIFIC Data | (2019) 6:291 | https://doi.org/10.1038/s41597-019-0275-3
www.nature.com/scientificdata
www.nature.com/scientificdata/
manuscript, provided review for co-authored elements of this manuscript. F. Jazizadeh, reviewed the complete
manuscript, provided guidance on data representation for an open access audience, developed the technical
validation section of this manuscript.

e authors declare no competing interests.

Correspondence and requests for materials should be addressed to F.P. or F.J.
Reprints and permissions information is available at www.nature.com/reprints.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional aliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre-
ative Commons license, and indicate if changes were made. e images or other third party material in this
article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons license and your intended use is not per-
mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the
copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
e Creative Commons Public Domain Dedication waiver http://creativecommons.org/publicdomain/zero/1.0/
applies to the metadata les associated with this article.
© e Author(s) 2019
Content courtesy of Springer Nature, terms of use apply. Rights reserved

1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com