Renewable and Sustainable Energy Reviews 81 (2018) 399–412

Contents lists available at ScienceDirect

Renewable and Sustainable Energy Reviews

journal homepage: www.elsevier.com/locate/rser

Residential electricity consumption behavior: Influencing factors, related MARK

theories and intervention strategies
⁎
Zhifeng Guoa, Kaile Zhoua,b, , Chi Zhanga, Xinhui Lua, Wen Chena, Shanlin Yanga,b
a
School of Management, Hefei University of Technology, Hefei 230009, China
b
Key Laboratory of Process Optimization and Intelligent Decision-making, Ministry of Education, Hefei University of Technology, Hefei 230009, China

A R T I C L E I N F O A BS T RAC T

Keywords: The proportion of residential electricity consumption in the total energy consumption has increased rapidly in
Residential electricity consumption the past decades all over the world. It is becoming increasingly important to promote household energy
Influencing factors conservation for the sustainable development of a country in the case of resource constraints. This paper
Behavior interventions reviews and evaluates the existing research works which are related to the residential electricity consumption
Theories
behavior. Particular attention is given to the following aspects. (1) Factors influencing residential electricity
consumption in social psychology. (2) Theories of social psychology in understanding residential electricity
consumption behavior. (3) Different interventions aiming at encouraging households to reduce electricity
consumption. Finally, we discuss the challenges and opportunities of research on residential electricity
consumption behavior in the big data era.

1. Introduction consumption bills and related electricity consumption behavior.

At present, many scholars have done a lot of researches on
Environmental pollution and scarcity of resources have become residents’ energy-saving behavior [3]. From the perspective of content,
major factors that affect the sustainable development of global the researches mainly focus on the following three aspects. The first
economy since 21st century [1]. Irrational use of energy has led to aspect is the analysis of the main factors that have effects on residential
environmental pollution and unsustainable development for a long electricity consumption behaviors, including household characteristics,
time. Electricity has played an essential role in our society [2]. socio-economic factors, social-psychology factors and related environ-
However, the main way of generating is thermal power, which mental behavior theory [17]. The second aspect is the statistical
accompanied with air pollution. In this sense, reducing electricity analysis of electricity consumption's data. The regular pattern of
consumption has been important implications for sustainable devel- electricity consumption was found [6]. For example, residents with
opment of society. The power demand of residents is a part of the social the similar load profile can be clustered into one class with the help of
power demand. The proportion of residential electricity consumption cluster analysis. The third aspect is the implementation of intervention
has growing with increasing household appliances and population. [11]. The researchers made an intervention on residents based on
Residents only know the monthly consumption of electricity before the psychological factors and mode of action. So we can reduce electricity
popularity of smart meters. They have few information about the daily consumption by means of energy-saving intervention strategy.
consumption of electricity, and people always receive the past elec- With the development of big data and cloud computing, it is more
tricity consumption bills after several months. The comparison of convenient to study the behavior hiding behind the electricity con-
electricity consumption was lacked. Meanwhile, residents do not care sumption [34]. By means of the Internet technology, researchers can be
about consumption of electricity. With the popularity of first generation more convenient to investigate the behavior of residents that related to
of smart meters, people could know their electricity consumption every electricity consumption. During big data era, it is possible to store and
day and even every 15 min [5]. People can pay off their electricity analysis massive data. Power Company can give feedbacks about
consumption bills on the Internet. The pattern of electricity consump- electricity consumption to customers more frequently through big data
tion is more like the use of mobile phone. During this period, people analysis [25]. We can draw more reliable conclusions with the support
have begun to pay more attention to electricity consumption gradually, of data mining, and further make more effective interventions.
so information makes the residents really give attention to their The purpose of this paper is to review and summarize literatures

⁎
Corresponding author at: School of Management, Hefei University of Technology, Hefei 230009, China.
E-mail addresses: zhoukaile@hfut.edu.cn, kailezhou@gmail.com (K. Zhou).

http://dx.doi.org/10.1016/j.rser.2017.07.046
Received 6 January 2017; Received in revised form 23 March 2017; Accepted 10 July 2017
1364-0321/ © 2017 Elsevier Ltd. All rights reserved.
Z. Guo et al. Renewable and Sustainable Energy Reviews 81 (2018) 399–412

related to residential electricity consumption. Firstly, the paper ana- more children (children 0–9 years old) had a negative impact on
lyzes the vital factors influencing residential electricity consumption. the electricity consumption, and having children has reduced the
Secondly, we review the social psychological factors, their mechanism average electricity consumption. Gram-Hanssen found that the
of effecting, and related behavior theory. Finally we summarize the five effect was significant in the Danish region, but was not in
most common intervention strategies and research direction of resi- Belgium. Wallisn et al. [87] analyzed the influence of the number
dential electricity consumption in big data era. of adolescents on electricity consumption, they found that
adolescents’ frequent purchasing of IT appliances led to higher
2. Social psychological factors influencing residential electricity consumption. Cramer et al [24]. found that children
electricity consumption behavior under 3 years old had no significant effect on household elec-
tricity consumption in the American family. However, for the
Home is the basic unit of electricity consumption, therefore the family has children over 3 years old, there was a significant
reduction of electricity consumption per household will reduce the impact on consumption of electricity. However, in Ireland, Leahy
electricity consumption of the whole society [27]. So we need to find and Lyons [7] showed that families with children had no
the factors which have significant impacts on household consumption significant difference in electricity consumption. Nakamura [28]
of electricity in order to reducing electricity consumption. Different showed that the mother having child was easier to participate in
families have different structure, cultural background and ideological some schools and community organizations, and they tend to
concept. Under the influence of kinds of factors and their interactions, know the information of energy conservation.
each family has a different load profile. Meanwhile, different load (3) Age composition of family members. According to Yohanis et al.
profile also reflect different family types and consumption behaviors. [8], the age of family member had influence on household
This section summarizes the characteristics of the household from electricity consumption. And electricity consumption is relatively
many literatures [29,30,32,33] including: (1) Number of family mem- high, when the age of the family member is 50–65. Electricity
bers. (2) Children. (3) Age composition of family members. (4) Level of Consumption is relatively lower, when the age of family member is
education. (5) Social status of family. (6) Family economic situation. less than 50 years old or over 65 years old. Leahy and Lyons [7]
(7) The type of a house. The effects of different family characteristic pointed out that electricity consumption of household where age of
factors on household electricity are described in the following. the family member is between 45 and 64 was significantly higher
than that of 35–44 years old in Ireland. Household electricity
(1) Number of family members. The relationship between the num- consumption decreased significantly when age of the family
ber of households and electricity consumption has been studied member is more than 64 years old. Mcloughlin et al. [10] found
by many scholars. The majority of literatures showed that the that household electricity consumption of family where age of the
number of households has a positive impact on electricity member is 18–35 was less than 36–55 or 56. Researchers believed
consumption. With the increasing number of households, house- that this is middle-aged family has more children and rooms. So
hold electricity consumption will also increase. Leahy and Lyons the consumption of electricity is more. Kavousian et al. [26] found
[7] studied the electricity consumption of single and double that in the United States, these families whose age of the family
people in Ireland. By comparison, they found that a single member is more than 55 or 19–35 consume less electricity.
apartment have less than 19% of the electricity consumption Filippini et al. [27] found in the India area, the family whose age
per week. Yohanis et al. [8] studied the relationship between the of responsible members is less than 45 have less consumption than
number of households and electricity consumption in an apart- the family whose age of responsible members is older.
ments in Northern Ireland. The results showed that the apart- (4) Level of education. The educational level of the family member has
ment lived with four people or more people are used to consume influence on electricity consumption. The conclusion is also
highest average annual electricity consumption. And there was no uncertain. Bartiaux and Gram-Hanssen et al. [9] found that
obvious difference between houses lived with two people and household electricity consumption decreased significantly as the
three people in average annual electricity consumption. Bartiaux level of education increased. The family members with higher
and Gram-Hanssen [9] investigated the relationship among the degree of education consumed less electricity than the family
number of families, housing type and electricity consumption. members with low education level. However, According to
The results indicated that the correlation between the number of Cramer et al. [24], the educational level of family members had
households and electricity consumption is the most significant. In no significant impact on electricity consumption both in the United
the three types of housing (independent, semi-independent, States and Holland.
apartment). The number of households has always been signifi- (5) Social status of the family. Social status has different influence on
cantly associated with electricity consumption. electricity consumption according to current research's conclu-
(2) Children. Children is also an important factor affecting electricity sions. Mcloughlin et al. [10] found that socio-economic status of a
consumption [35]. Many scholars have made researches on the family had a significant impact on household electricity consump-
relationship between the composition of the family members and tion. And there was a significant positive correlation between
electricity consumption. There are two opposite results. A part of socioeconomic status and household consumption. The higher
the studies found that the composition of family members had a social status of household consumption was accompanied with
significant impact on consumption of electricity. For example, more electricity consumption. However, according to Leahy and
Mcloughlin et al. [10] studied on the relationship between Lyons [7], family members' economic status had no significant
household electricity consumption and whether the family have impact on household electricity demand.
children or not. They found that the family had children consume (6) Family economic situation. Economic situation mainly involves
more electricity than the family had no children. Brounen et al. two aspects [54]. On the one hand, it is the family income; On the
[12] found that the family with a child consume more 1/5 other hand, it is the family disposable income. These two aspects
electricity than the family without children. With the growth of reflect the economic situation of a family. A large number of
the child, household electricity consumption will also increase. In literatures showed that household electricity consumption in-
Brounen's opinion, children tended to play computer, watch TV, creased with income. Yohanis et al. [8] found that in Northern
play games or other electrical device, these activities would lead Ireland, the households whose annual incomes of more than
to consume electricity. However, Bartiaux and Gram-Hanssen [9] 30,000 Irish pounds consumed more electricity than the low
had opposite conclusion. They found that the family with two or income families which has annual income of 10,000 Irish pounds.

400
Z. Guo et al. Renewable and Sustainable Energy Reviews 81 (2018) 399–412

One possible reason was that higher income families tended to 3.1. Social cognitive theory
have a large area of housing and household appliances. This led to
more electricity consumption. However, efficiency of home appli- The famous American social psychologist Bandura first put forward
ances and electronic devices is also vital to energy saving. Wealthy social cognitive theory [66]. The theory claims that the motivation
families have ability to buy smart appliances and install related of the individual is the result of the interaction among the individual,
home power management systems, some poor families can only environment and behavior (see left panel Fig. 1). The individual refers
afford energy saving bulb. This situation will eventually change to people's belief, goal, attitude, intention, emotion, and so on, which
with the development of economic. More and more people will reflects the individual's cognitive ability. Environment refers to the
choose energy-saving equipment. Ozkan [81] used smart home resources, the consequence of action and physical conditions, and it
power management systems to reduce electricity cost and improv- reflects the objective conditions of individual. The interaction between
ing energy efficiency while maintaining their comfort. The result the individual and behavior shows that the individual's beliefs, goals,
indicated that the system provided improvements in terms of the attitude, intention and emotion will affect the one's behavior, and vice
energy consumption reduction of about 5–16% compared to versa. The interaction between behavior and environmental factors
conventional electronic devices usage. shows that the environment is the result of the behavior, however the
(7) The type of a house. House is the basic unit of people's life, behavior can also change the objective environment to make the
different lifestyles are determined by the type of a house. In this subject more suitable. The interaction between personal factors and
sense, different house will display various load profile. On one environmental factors shows that the main beliefs, goals, attitudes and
hand, different type of a house display different consumption emotions are determined by environmental factors, but the effect of the
behavior, on the other hand, individuals have different attitude environment on the subject is not absolute.
towards energy in different place. For example, there are the According to social cognitive theory, self-efficacy and outcome
distinction behavior between office and home settings. Littleford expectations are the most important concepts. Self-efficacy refers to
et al. [82] investigated the relationships between energy consump- extent of confidence in one's own action. As a result, the higher the self-
tion behaviors performed in office and home settings. The result efficacy is, the more likely he is to perform the action. Outcome
indicates that the home setting is an important feature of the expectation is the result of one's likelihood of completing a task.
energy use behavior. Dixon et al. [83] discovered the sense of Positive outcomes are expected to promote individual behavior, while
community is key factor in workplace energy conservation. negative outcomes are expected to hinder individual behavior.
Therefore, type of a house is a vital factor that has an effect on As far as energy, a more pluralistic viewpoint is depicted in Fig. 1.
electricity consumption behavior. There are two panels in the figure. The left panel is macro factors that
originate from social cognitive theory, and the right panel is micro
factors that are summarized form many related papers. To some extent,
3. Theories of social psychology in understanding residential the right panel is concrete form of left panel. In reality, there are three
electricity consumption behavior aspects that are related with individual's behavior. Namely, culture,
society and economic. Individual's decision environment is consist of
From a psychological point of view, the basis of a person's behavior these three factors. Culture could have the most lasting impact on
is driven by psychological factors [68]. Therefore, there is a complex people, education is a way to spread culture. Economic is important
psychological process behind the energy consumption behavior, which factors that are have positive effect on energy-saving behavior. Many
involves a number of psychological factors. In recent years, many scholars have many factors has been researched the relationship
scholars have studied the behavior of energy saving from the perspec- between energy-saving behaviors and these factors. For example,
tive of psychology, and analyzed how these factors have influence on income has positive effect on electricity consumption by many scholars
behavior. This section summarizes various factors that affect the [84] as mentioned earlier. Kyriakopoulos et al. [97] studied the issue of
behavior and the mode. The main psychological factors are attitude, technological advancements and the social acceptability of renewable
belief, values, culture, habits, preferences, subjective norms, environ- energy technologies and point importance of renewable energy tech-
mental awareness, and self-efficacy, etc. [69]. These psychological nologies’ diffusion in the everyday life of people. Education on energy
factors affect the behavior indirectly. Person's behavior is influenced issues is vital to students. They tended to form correct energy attitudes,
by psychological factors through a variety of ways. Some scholars have lifestyle and behavior that will be good for whole society with the help
studied the effect of psychological factors from the viewpoint of of environmental education, and education has the most lasting impact
environmental behavior, and put forward the new social psychology on people. Ntona et al. [85] researched on student's viewpoints and
theory based on classical theory. attitudes about energy and its usage related to the environment, Their
The behavioral theories are mainly social cognitive theory, social findings indicated that the education need a vital change towards an
norms theory, and theory of reasoned action, theory of planned environmentally sustainable orientation. In addition to the school, the
behavior, goal-oriented behavior model, value-belief–norm theory, family is an important place to cultivate habits of saving energy. Social
norm activation theory, self-regulated behavior change theory and factors, such as subsidies have influence on electricity consumption
ABC theory [70–75]. Different behavioral theories are based on behavior. Nicolini and Tavoni [98] verified if policy support for
different psychological factors and the general pattern of effecting. renewable electricity have been effective in promoting renewables in
Each theory has its own advantages and disadvantages. And as far as the five largest European countries during 2000–2010. The result
the theory is concerned, it has been developing and improving. indicated that these policies have been effective in promoting renew-
Therefore, there exists many connections among different theories, able energy, both in the short and in the long term. In general,
For example, from the theory of reasoned action to the theory of economic, social, cultural factors form a vital environment that has
planned behavior, and from the theory of planned behavior to the important influence on decision process. Culture has a lasting and
model of goal-oriented behavior. Another example is that value-belief- stable influence on behavior. Social and economic factor have transient
norm theory is the perfect of norm activation theory. Therefore, none of effect, for example, if welfare can not last or people's financial status
the theories can explain all the environmental behavior, and each gets bad, people will choose cheap products without considering
theory has its own limitations. Therefore, in the study of electricity environmental factors.
consumption behavior, vital psychological factors should be deter- In aspect of environmental behavior theory, many scholars re-
mined by a variety of experiments. Table 1 presents a brief description garded social cognition theory as the research framework and analyzed
of each behavior theory. the effect of various psychological factors. Thøgersen and Grønhøj [67]

401
Z. Guo et al. Renewable and Sustainable Energy Reviews 81 (2018) 399–412

Table 1
A brief description of behavioral theory.

Theory Author Description Refs.

Social cognitive theory Bandura Subjective initiative is the result of the interaction between the individual factors, the environment and behavior. [66]
Social norms theory Elster Social norms that created and observed by people; individual is expected to share in the group [80]
Theory of reasoned action Fishbein, Ajzen Person's behavior is determined by the behavior intention, behavior intention is affected by the attitude and [53,54]
subjective norm
Theory of planned behavior Ajzen Actual behavior is determined by behavior intention, and the behavior intention is affected by the attitude, [52]
subjective norms and perceived behavioral control effect
Model of goal-directed Perugini Attitude, subjective norm and perceived behavioral control did not directly affect the behavior intention, but the [59]
behavior affect behavior intention through desire which were an intermediate variables
Norm-activation-model Schwartz People's environmental behavior is the result of norm, belief and value [50]
Value-belief-norm theory Stern Values, beliefs, and personal norms affect a person's behavior gradually, from values to action in turn [51]
Self-regulated behavior change Bamberg The theory suggests that a person's behavior can be changed under certain conditions, even if the behavior has [55]
theory been a habit
ABC theory Guagnano Individual behavior is the result of mutual effect between attitude and the external context [57]

Macro factors Micro factors

Environmental
factors Economic factors
resources welfare
others financial status
action results employment
physical
condition

Social factors
cohesion
Behaviors
inclusion
acceptability

Individual factors
belief Cultural factors
goal family
attitude environment
intention humanistic
emotion background
education profile

Fig. 1. Interactions among environment, behaviors, and person [66].

constructed the framework to analysis electricity saving behaviors in establishment of structural equation model. The results suggest that:
households based on social cognition theory (see Fig. 2). In addition to (1) we should change the socio-structural environment to be more
the psychological factors, the social structural factor was also put into convenient for energy saving behaviors. (2) Feedback about their
model as an influencing factor. Self-efficacy and outcome expectation household's electricity consumption is a vital factor that has influence
are the main factors influencing the behavior intention. The socio- on their own behavior.
structural factor and perception of others’ behavior are known as
structural factors. The proposing assumptions are as follows. (1) 3.2. Social norms theory
Individual's behavior is affected by outcome expectations, self-efficacy,
socio-structural factor, perception of others’ behavior and goals. And Norms first comes from the Latin word Norma, which means a
outcome expectations self-efficacy, socio-structural factor and percep- ruler. However, the standard concepts came into the social psychology,
tion of others’ behavior cannot only have a direct impact on the which owing to the famous social psychologist Sherif's experiment in
behavior, but also indirectly influence behavior through goal. (2) The 1936 [61]. From then on, the norm theory has become an important
outcome expectation, self-efficacy, perception of others’ behavior, and theoretical in many social sciences. Norms in the field of sociology are
socio-structural factor have mutual influence among themselves. (3) also known as social norms that is proposed formally by Elster [80]. To
Outcome caused by individual's behavior can give feedback on the a certain extent, social norms and norms have the same connotation.
outcome expectations and self-efficacy socio-structural factor. Social norms have different definitions with the development of society.
Researchers verified hypothesis through a questionnaire and the In the earlier years, according to Sherif [61], it was defined as customs,

402
Z. Guo et al. Renewable and Sustainable Energy Reviews 81 (2018) 399–412

Attitude
Outcome
expectations

Intention Behavior
Goals
Subjective
norm
Self-efficacy
Fig. 3. Theory of reasoned action [53,54].

Environment
Behavior al knowledge

Sociostructural
factor Environment
al awareness

Environment
Outcome al attitude
Perception of
Intention Behavior
others behavior Economic
of green of green
background
purchase purchase
Fig. 2. Social cognitive model based on behavior and learning [66].

Technologica
traditions, standards, rules, values, fashion and so on. Recently, a
l development
concept that is more generalized is what the individual is expected to
share in the group. Social norms are divided into descriptive norms and
the injunctive norms in general. The descriptive norm is to point out
what most people are doing, and the injunctive norms is to point out Peer pressure
the behavior that others disapproval of [62].
The impact of social norms on individual behavior is mainly divided
into three stages according to Kelman [63]. The first stage is com-
pliance. During this period, the individual try to change their behavior Energy label
in order to get a reward or avoid punishment on the surface. The
second stage is identification. After a long-term impact, the individual
begins to consciously abide by the expectations of others. The third Fig. 4. Research model [23].
stage is internalization, and in this stage, individual really accept the
external values from the heart, convert the social norms into personal
norms, and regard it as long-term guidance. Social norms on human external pressure to perform or not to perform a certain behavior. And
behavior are mainly reflected in pro-social behavior, especially for the the pressure is caused by others. The contribution of the theory of
pro-environmental behavior [65]. Person's behavior can be influenced reasoned action is to consider the social factors such as the subjective
by social norms through different ways. There are different models norm, it means behavior is not only affected by their own attitude but
according to the different perspective, such as the norm activation also by the outside world.
theory, value-belief-norm theory, and the theory of planned behavior. Many scholars have constructed energy consumption behavior
In the field of consumption of electricity, reduction of electricity model based on theory of reasoned action. Zainudin [23] used
consumption has taken good effect by social norms. The most success- the TRA (theory of reasoned action) to construct the framework,
ful is the case of OPOWER Company [64]. OPOWER send Home which designed to display process of public green purchase intention
Energy Report letters to residents, in order to make intervention. It (see Fig. 4.). In this model, the researchers supposed that public's
also provides social norms information, which compares the household willingness to buy environmental product is influenced by environ-
electricity consumption and their neighbors’ consumption in report. mental knowledge, environmental awareness, environmental attitude,
The results showed that the average treatment effects is 2%. Therefore, environmental attitude, peer pressure and the energy label. The
social norms is of great significance for reducing electricity consump- researchers hypothesized that: (1) Environmental knowledge, environ-
tion. mental awareness, environmental attitude and behavior intention
indirectly influenced people's behavior by means of peer pressure. (2)
Environmental knowledge, environmental awareness, environmental
3.3. Theory of reasoned action attitude, and peer pressure had positive impact on behavior intention.
These hypotheses were verified by constructing a structural equation
Theory of reasoned action (see Fig. 3) was proposed by Fishbein model. The final analysis results showed that environmental knowl-
and Ajzen in 1975 [53,54]. The theory proposes that a person's edge, environmental awareness, environmental attitude and peer
behavior is determined by the behavior intention, and the greater the pressure played a positive role in promoting green energy purchase
behavior intention is, the more likelihood he will perform the behavior. intention, and environmental label on green had a negative impact on
Behavior intention is affected by the attitude and subjective norm, energy purchase. However, economic background and technological
attitude is extent of supporting behavior. Subjective norm refers to the development also have significant influence on green purchase inten-

403
Z. Guo et al. Renewable and Sustainable Energy Reviews 81 (2018) 399–412

tion (see Fig. 4). In this sense, environmental knowledge, environ-

Behavioral
mental awareness, environmental attitude, environmental attitude, attitude
peer pressure, energy label, economic background and technological
development form a more practical framework (see Fig. 4), and many
scholars have done many researches under this framework. Pothitou
et al. [89] gave attention to the impact of knowledge about environ- Subjective Behavior
Behavior
mental, attitudes, habits and energy issues on potential pro-environ- norms intention
mental behavior in households. The results showed that significant
correlations which indicate that residents with positive environmental
values and greater environmental knowledge were more likely lead to Perceived
behavioral Direct
energy-saving behavior. Testa et al. [91] researched the determining impact
control
factors behind individuals' decisions to purchase energy saving pro-
ducts. They found that personal norms and trust in information Fig. 5. The theory of planned behavior [52].
provided by private companies and family and friends have influence
on adoption of energy-saving purchases. Suki [92] investigated the
influence of consumption values (i.e. functional value, social value, processes hidden behind the residential electricity behavior according
emotional value, conditional value, and epistemic value) in environ- to the theory of planned behavior. Scott et al. [16] used the theory of
mental concern on purchasing of green products. They found that planned behavior to do a survey of energy usage through a series of
social value has most important impact on consumers' environmental questions that are related with beliefs in British Yorkshire and Humber
concern as expressed by their purchase of green products. Ramayah area. They measured the relationship between people's attitudes and
et al. [93] analyzed how individual values and attitudes in a developing the rate of adopting energy-saving device. The results showed that: (1)
country influence purchase intention of a green product from the People who use energy saving devices easier were worried about
perspective of the TRA. Results indicated that individual consequences abnormal changes in the global temperature. (2) The extent of
relating to amount of effort and convenience of consumers was familiarity with energy-saving device had impact on the rate of using
negatively related to intention to purchase of green product. Olson- energy saving devices. Residents were more willing to use familiar
Hazboun et al. [94] analyzed the relationship between environmental energy-saving devices. (3) As far as different energy-saving devices, the
beliefs, climate change opinions and support for renewable energy respondents showed significantly different in the behavioral attitude,
through questionnaire of five communities in the Rocky Mountain subjective norm, and perceived behavioral control. Based on the theory
region of the U.S. The results showed that public support for renewable of planned behavior, Abrahamse and Steg [17] put forward the
energy was less related to environmental than to some other factors, following hypothesis. (1) There was a strong correlation between
including economic benefits. Hast et al. [95] analyzed consumers’ household energy consumption and social demographic characteristics
attitudes towards green energy in China and their willingness to (such as income, family population). (2) Energy consumption was
purchase of green electricity or renewable energy systems. The results mainly determined by psychological factors. The results of regression
indicated that willingness to pay for a green electricity product is analysis showed that the use of household energy was determined by
influenced by income, building type, how promising the renewable demographic factors. Yazdanpanah et al. [18] studied on the relation-
energy potential is seen. As mentioned earlier, financial status are high ship between social psychological factors and renewable energy.
related with electricity consumption on matter national level or Researchers concluded that the social psychological factors (such as
individual level. Technological development is another factor that behavior, perception, and subjective norm) had significant impacts on
promote green purchase. Kyriakopoulos and Arabatzis [96] pointed the adoption of the Energy Sources Renewable (RES) project. The
that implementation of electrical energy storage systems should researchers collected information from 260 students by questionnaires.
consider the diffusion of innovative technologies and comparison The relationship between attitude toward RES, behavior attitude,
between innovative technologies and traditional technology. perceived behavior, subjective norm and behavior intention was
analyzed by means of structural equation model. The final results
3.4. Theory of planned behavior showed that the code of ethics and behavior attitude and perceived
behavioral control significantly affected the residents’ willingness to
In 1991, Ajzen [52] proposed the theory of planned behavior (TPB), use renewable energy, but subjective norm and self-recognition did not
which was an extension of the theory of reasoned action (TRA). significantly affect attitude toward renewable energy.
According to the theory, there exists a complicated psychological Botetzagias et al. [19] studied the relationship between different
process behind the individual behavior. The behavior is the results of factors and energy-saving behavior by telephone interview. He re-
a series of mental processes [60]. The actual behavior is decided by garded demographic characteristics, psychological factors and moral
behavior intention, and the behavior intention is affected by the factors as explanatory variables, energy saving behavior is viewed as
attitude, subjective norms and perceived behavioral control (see the explanatory variable for regression analysis. The researchers put
Fig. 5). Behavioral attitude is extent of a person's support or not forward the hypothesis that the power saving behavior depends on the
support a behavior. Subjective norm refers to the social pressure that demographic characteristics, psychological factors and moral factors.
people perceives when to decide whether or not to perform a particular The results showed that age, gender and perceived behavioral control
behavior. Perceived behavioral control refers to the ability to perform a had significant impact on power reduction. Paul et al. [31] constructed
behavior. Accurate perception behavior control reflects the actual a new research framework (Fig. 6) by add concern environmental
control conditions, it can be used as an alternative measure of the variables to the theory of planned behavior. He explored the factors
actual control conditions, so it can also be used as a direct impact on that have influence on the willingness of Indian customers to buy green
the behavior (as shown in Fig. 5). The accuracy of the prediction products by confirmatory factor analysis through a questionnaire. The
depends on the perceived behavioral control. Individual and social results showed that customer attitudes and perceived behavioral
cultural factors such as personality, intelligence, experience, age, control can predict behavioral intention. However, social norms did
gender and cultural background affect the behavior intention and not predict the intention. The results also showed that environmental
behavior indirectly through behavior control behavior attitude, sub- concern and attitude, subjective norm, perceived behavioral control
jective norm and perceived behavioral control. and purchase intention were positively correlated. Therefore, environ-
Some scholars have done a lot of researches on psychological mental concern can indirectly affect purchase intention through

404
Z. Guo et al. Renewable and Sustainable Energy Reviews 81 (2018) 399–412

Attitude Attitude
Autonomous
motivation

Environmental Subjective Purchase Actual Positive

concern norm intention behavior anticipatory effect

Negative
Perceived Desire Intention Behavior
anticipatory effect
personal
control

Perceived
Direct impact behavioral control

Fig. 6. Research model [31].

Past behavior
attitude, subjective norm and perceived behavioral control, and it can
also directly affect the purchase intention. So when a person has a Subject norm
positive attitude towards green products and a high environmental
concern, he is more likely to buy green products.
Fig. 8. Research model [58].

3.5. Model of goal-directed behavior autonomous motivation factor and past behavior to model of goal-
directed behavior. He intended to find most significant explanatory
Perugini and Bagozzi [59] proposed a model of goal-directed variable through the survey analysis. The results showed that auton-
behavior based on theory of reasoned action and theory of planned omous motivation factor has most signification ability to predict
behavior (see Fig. 7). The theory pointed out that attitude, subjective behavior than other variables, such as behavior intention, subjective
norm and perceived behavioral control did not directly affect the norm, perceived behavioral control, and past behavior.
behavior intention, but the affect behavior intention by desire was an
intermediate variable. And he put forward the concept of anticipatory
effect. Anticipatory effect a dynamic reflection of reality. Anticipatory 3.6. Norm-activation-model
effect is divided into positive anticipatory effect (such as the desire for
success) and negative anticipatory effect (such as fear of failure). In the The norm-activation-model was proposed by psychologist Schwartz
model of goal-directed behavior (see Fig. 7), attitude, subjective norm, in 1977 [50]. The model explains the intrinsic mechanism of people's
positive anticipatory effect, negative anticipatory effect, perceived pro-social behavior in the perspective of psychology. Pro-social beha-
behavioral control have effect on desire that have effect on behavior vior refers to all kinds of behavior that be beneficial to others or society,
intention. Behavior intention ultimately affects the behavior. Model of such as: environmental protection, energy-saving behavior, water
goal-directed behavior was an improvement of the theory of reasoned conservation and other environmental behaviors. Schwartz [50] de-
action and theory of planned behavior. They pointed out that psycho- scribed the process of people's environmental behavior (see Fig. 9) by
logical factors have important effect on behavior such as desire. Some introducing the conception of social norms, person norms, awareness
researchers have applied the model of goal-directed behavior to energy. of consequence, and ascription of responsibility. Social norms refer to
For example, Webb et al. [58] established a new theoretical frame- values, attitudes and behaviors that most people identify with in
work based on Model of goal-directed behavior (Fig. 8). He added society. Person norms refer to the moral obligation of a person.
Person norms are the specific embodiment of social norms and
mainstream values of the society. And what he behavior criterion
Attitude reflected to the individual is the personal norms. Awareness of the
consequences is that a person's attention to the consequences that
caused by person's choice. Ascription of responsibility is refers to the
negative consequences of cognition that caused by ignoring the action.
Subject According to norms of social activation theory, social norms have effect
on the individual through personal norms, but the person norms do not
lead to environmental behaviors definitely. To some extent, it depends
on the person's situation, which means that environmental behavior is
Positive also influenced by awareness of the consequences and ascription of
Desire Intention Behavior
anticipatory effect responsibility. In other words, Person norm will cause the environ-
mental behavior when awareness of the consequences and ascription of
responsibility are relatively high.
Negative Zhang et al. [44] studied on the factors that have effect on electricity
anticipatory effect saving behaviors in the company using the norm-activation-model. The
researchers pointed out that electricity saving behavior in company is
different from electricity saving behavior in family. For example,
Perceived workers who consume electricity in a company do not need to worry
behavioral control about spending money. In this case, saving electricity behavior is
driven by moral norms. Researchers added organization electricity
Fig. 7. Model of goal-directed behavior [59]. saving climate into the original model (see Fig. 10). They measured

405
Z. Guo et al. Renewable and Sustainable Energy Reviews 81 (2018) 399–412

Awareness of Ascription of
consequence responsibility

Social norms Personal norms Environmental behavior

Fig. 9. Normative-activation-model [50].

explanatory variable through the questionnaire and analysis data by ment through questionnaire and analysis the relationship among them
partial least squares method. Results showed that when the employees with the help of structural equation model. The results showed that the
have higher person norms, they are more likely to save electricity. relationship between the sense of responsibility and the responsibility
However, higher organization electricity saving climate has weakened attribution is not significant. Han [41] combined value-belief-norm
the influence of personal norms that has positive effect on energy theory with the theory of planned behavior to form a new theoretical
consumption behavior. When employees feel a higher ascription of framework (see Fig. 12). The lower part is the value belief norm theory
responsibility, awareness of consequence and organization electricity and the upper part is the theory of planned behavior. The results
saving climate, the staff is more likely to form strong personal norms. showed that the integrated framework performs better than the single
Personal norms have positive impact on energy saving behavior. Han theoretical framework.
et al. [45] combined the norm-activation-model with model of goal-
directed behavior in order to establish a new theoretical framework. 3.8. Self-regulated behavior change theory
The research results showed that the integrated framework has more
explanatory power than the norm activation theory or model of goal- Bamberg [55] put forward self-regulated behavioral change theory
directed behavior. (SSBC) based on the theory of norm activation theory and the theory of
planned behavior in 2013. The theory suggests that a person's behavior
3.7. Value-belief-norm theory could be changed under certain conditions, even if the behavior has
been a habit (see Fig. 13). Self-regulated behavior change theory is
Stern et al. [51] put forward the value-belief-norm theory that mainly applied to the analysis process of complex behavior. The self-
combined with the value theory, norm-activation-theory and the new regulated behavior change theory is divided into four interdependent
environmental paradigm in 1999. The new environmental paradigm is stages, that is, pre-decision, pre-action, action, and post-action. Four
proposed by Dunlap and Liere Van [56]. It is an effective tool to states proceed in turn. There are also goal intention, behavioral
measure the environmental awareness, and NEP is short for new intentions, implementation intention and new behavior in the modal.
ecological paradigm (see Fig. 11). The theory pointed out that values, The goal intention refers the desire to achieve some kind of goal. The
beliefs, and personal norms affect people's behavior gradually. Value- goal intention is through a series of psychological processes as well as
belief-norm theory consists of five variables, namely, values, NEP, the outside interference realization. Aiming at producing a goal, a
awareness of consequence, attribution of responsibility, environmental person need to feel the negative impact of their current behavior first of
ethics (see Fig. 11). Values is mainly consist of the ecological values, all, it will have a sense of responsibility. This sense of responsibility will
altruistic values, and self-interested values. NEP represents the general produce negative emotions, and ultimately in the role of social norms
belief of people. Value-belief-norm has been expressed the transition and negative emotions, resulting in the intention of the target. When a
from values to environmental behavior by a series of psychological person in the role of multiple factors to produce a target intention, it
changes. In value-belief-norm theory, a person's environmental beha- will be the transition to the second stage, which is, the stage of behavior
vior will occur only when he thinks it is his obligation to protect the and the stage of behavioral intention. Behavior intention refers to a
environment. People will have NEP in their minds under the influence person's specific behavior to achieve the goal. The main factors that
of a variety of values. And awareness of consequence, attribution of affect the realization of intention are behavior plan, cognitive plan and
responsibility and environmental ethics proceed from the initial values the ability to keep the plan. When the goal of the implementation of the
to energy saving behavior in turn. intention to reach, after entering the stage of behavior. Behavior is
In recent years, many scholars have studied the psychological mainly used in the third stage of the behavior, so as to achieve the real
factors that affect residents’ behavior by means of the value-belief- change behavior.
norm framework. Fornara et al. [20] used the value-belief-norm theory In recent years, many scholars have begun to utilize self-regulated
to study the relationship among the individual values, NEP, the behavior theory to study residents’ energy consumption behavior.
consequences of consciousness, responsibility attribution, environmen- Based on the framework, Nachreiner et al. [21] investigated the
tal ethics and the willingness to buy green equipment. They measures relationship between smart meter information feedback and energy
people’ values, NEP, awareness of the consequences, attribution of saving behavior. In the first stage, researchers provided residents with
responsibility, environmental ethics, and the purchase of green equip- information of electricity consumption. In this case, electricity con-

Organizational
electricity
saving climate

Awareness of Ascription of Person Electricity saving

consequence responsibility norm behavior

Fig. 10. Research model [44].

406
Z. Guo et al. Renewable and Sustainable Energy Reviews 81 (2018) 399–412

Value Individual
Belief Behavior
norm
Values of the
ecological

Altruistic Awareness of Ascription of Environmental Environmental

NEP
values consequence responsibility ethics behavior

Self -interested
values

Fig. 11. Value-belief-norm [51].

Attitude
toward the
behavior

Perceived
Behavioral
behavioral
intention
control

Subjective
norm

Biospheric Ecological Awareness of Ascribed Sense of obligation to take

value worldview consequences responsibility pro-environmental action

Fig. 12. Research model [41].

sumption information makes residents know their electricity consump- saving skills, and people begin to make a plan for saving energy. In the
tion, people begin to set an energy-saving target. In the second stage, fourth stage, people will be change energy-saving behavior into habits.
the researcher provided more detailed information on the electricity Based on the SSBC framework, Mack and Tampe-Mai [22] developed
consumption, including each home appliance. In this case, residents an information system for making interventions on electricity con-
learned consumption of each home appliance. They know which home sumption behaviors, the system mainly included: (1) electricity con-
appliance consumes most, and people begin to form behavior intention. sumption and related data visualization (2) feedback chart (3) con-
In the third stage, the researchers provided people with many energy- sumption analysis module (4) what is the tile (5) module the source

Predecision Preaction Action Postaction

Emotions anticipated
with goal progress

Salient social Goal Behavioural Implementation

Personal norms New behavior
norms intention intention intention

Negative
go
emotion Attitude
toward and Action
perceived planning
Perceived behavioural cognitive
Perceived Recovery
goal control over planning
responsibility self-effancy
feasibility alternative maintenanc
behavioural e self-
change efficacy
Perceived strategies
negative
consequences
of own
behaviour

Fig. 13. SSBC model [55].

407
Z. Guo et al. Renewable and Sustainable Energy Reviews 81 (2018) 399–412

Positive providing information and result feedback are most used. Some
researchers try make a comprehensive intervention strategy by means
of the combination of five basic interventions. The results showed that
the effect of the comprehensive intervention strategy is better than the
single intervention strategy. The following briefly describe different
Behavior intervention strategies and the corresponding literature.
occurrence
External
(1) Commitment. Commitment is one of the most common interven-
conditions
Adverse Favorable tion strategies. Commitment is achieved by oral contract or written
contract in which people promise to change their behavior for the
Behavior does not reduction of electricity consumption. There are various forms of
occur setting goal for people. At the beginning, the effect of intervention
strategies were obvious, but the effect of intervention begin to get
Negative
worse after a period of time.
Katzev and Johnson [13] sent a paper contract to different
Attitude
families, and it required people to reduce energy consumption by
Fig. 14. ABC model [57]. 10%. After a period of time, household that accept contract
consumed less electricity compared with these who refused to
and climate information (6) comparison module (7) target setting accept the contact. The researchers further did another interven-
module (8) energy saving module (9) energy-saving skills module. The tion experiment. The experiment compared commitment with
system is used to support the behavior of different stages in the SSBC other intervention strategies such as rewards and comprehensive
model. In the energy and climate information module, people can intervention strategy. The results showed that comprehensive
understand the current global climate and energy status of the world, intervention strategy has best effect on reduction of energy
as well as the important significance of emission reduction. Under the consumption. Commitment is used as an intervention strategy
intervention of all kinds of information, the residents have the goal frequently in earlier year, in recent years, commitment is less used,
intention and tend to show more energy-saving behaviors. at the end of experiment, the residents’ behavior is as usual as
before. In this case, Commitment cannot lead to sustained energy-
3.9. ABC theory saving behavior.
(2) Goal setting. Goal setting is to set a goal for the family, such as
Guagnano et al. [57] proposed ABC theory in the study of garbage reduction of 5% or 15% electricity consumption. The target value
collection issues. The theory considers that individual behavior (B) is can either be set by the family itself, or it be set by the researchers.
the result of mutual effect between attitude (A) and the external context Target value cannot only be determined in the form of absolute
(C). The theory deems that a person's behavior is not only affected by value, but also determined in the form of proportion. Goal setting
the individual's attitude, but also by the external environment. ABC is used by most researchers. The goal that set too high or too low
theory emphasizes the impact of external factors (see Fig. 14). The often led to worse effect.
horizontal axis represents the external conditions. The vertical axis Becker [14] set two targets for different households, one is
represents the attitude. When external conditions and attitudes are relatively easy to reach (saving 2%) and another is difficult to reach
greater than zero, behavior will occur. Behavior will not occur when the (saving 20%). At the same time, the experimental group received
external conditions and attitudes are less than zero. When the external feedback every week. In the experimental period, researchers
conditions are in a favorable position, action will occur. provided with information about energy-saving skills. The results
In the ABC theory, Guagnano emphasized the influence of external showed that these household that received a higher goal of saving
environment on the behavior of a person. He pointed out that when a 20% electricity saved most (15.1%). The experimental indicated
person's attitude tends to zero, the external environment have a that higher goal of reducing electricity consumption is achieved by
determined effect on person's behavior. At this point, if the external providing with energy-saving information. Harding and Hsiaw
environment is very favorable for behavior, it will greatly promote the [46] executed a plan that aims to encourage people to set a goal by
occurrence of behavior; on the contrary, if the external environment is themselves in Northern Illinois. The results showed that the
unfavorable for behavior, it will greatly prevent the occurrence of a reasonable goal of saving electrical consumption is achieved, and
person's behavior. In the same manner, when a person implements a household saved average of nearly 11%, which is significantly
behavior that costs much, the occurrence of behavior will depends on higher than those whose targets are too low or too high.
the attitude. ABC theory provides a new way of thinking for the Abrahamse et al. [47] performed an energy-saving intervention
research of energy saving behavior. in Groningen. The households were divided into two groups. The
first group received tail information, tail feedback, and goal setting
4. Intervention strategies (5%). The second group is regarded as the control group without
any intervention. During 5 months, researchers performed three
The number of household electrical equipment has been increasing aforementioned interventions, and the final results showed that
with the development of social economy, the same as the proportion of comprehensive intervention strategy reduces the energy consump-
residential electricity consumption. Therefore, advocating energy-sav- tion by 5.1%.
ing behavior has been an important issue for society. For a long time, (3) Providing information. Providing information is another common
household electricity behavior did not get enough attention. With the intervention strategy. Information is about environmental pollu-
number of household equipment increases, electricity consumption tion, the significance of saving electricity and the tips of saving
accounted for the proportion of whole electricity consumption has energy. Providing information cannot only improve residents’
increased too. In this sense. A number of studies have shown that awareness of energy conservation, but also increase the knowledge
intervention strategies can be used to reduce energy consumption. In of energy conservation. There are also a variety of ways to provide
this section, common five kinds of intervention strategies is reviewed, information. Common way of providing information is mass
including commitment, goal setting, providing information, reward, communication. The aim of providing information is to change
result feedback. Among many intervention strategies, goal setting, people's consumption behavior from the heart, so providing

408
Z. Guo et al. Renewable and Sustainable Energy Reviews 81 (2018) 399–412

information often have the effect of changing people's behavior. intelligent device, namely intelligent device is two difficult for people to
Hutton and Mcneill [15] assessed the effect of the Low Cost/No use. Another reason is that people are not interested in consumption
Cost energy conservation program of the US Department of information, so intelligent hardware device cannot always be useful.
Energy, The project tried to achieve the goal of saving energy by Hargreaves et al. [37] found that after a period of time, the residents
provide people with energy-saving manual and a nozzle control tend to adjust the status of intelligent device into background mode,
device. The results showed that these families that received the rather than having the positive interaction with it. Burchell et al. [42]
manual could master skills to reduce energy consumption. studied the effect of feedback that based on intelligent community.
Hargreaves et al. [37] tried to make an intervention by persuading Different from the traditional intelligent hardware feedback mode,
people to install the smart meter. Residents can obtain the real- people gained their own electricity consumption information through
time condition of electricity consumption with the help of smart the weekly e-mail feedback that sent by researchers. E-mail consists of
meter. However, after a year, the researchers found that most not only people's own electricity consumption information, but also the
families let the smart meters to run in the background state. The average electricity consumption of the whole intelligent community
use of smart meters does increase the people's knowledge of and some useful energy saving tips. The results showed that people can
electricity consumption, but after a long time, the smart meter participate in this form of feedback in long term. And they can draw
does not reduce the electricity consumption due to various reasons. women's attention, ultimately people tend to change the behavior.
Komatsu and Nishio [38] investigated the effect of providing Matsui et al. [43] installed electricity consumption monitoring system
information in Japan, they provided household with their neigh- in people's house. It provided people with electricity consumption per
bor's electricity consumption information. The results showed that 15 min through the web page. In addition to the people's recent
a household's consumption of electricity is associated with their electricity consumption data and some tips, people's knowledge
neighbor's electricity information. improved obviously after a year's intervention, and electricity con-
(4) Rewards. Rewards is an intervention that encourages people to sumption is decreased. Karjalainen [40] investigated people's prefer-
reduce the electricity consumption by giving them a certain ence for forms of feedback. He displayed eight different chart to people
amount of incentives. Awards mainly consists of economic reward through the questionnaire analysis. The results showed that people is
and social reward. Economic reward is to give the family a certain most interested in three kinds of information, namely, electricity
economic incentives, and the number of economic incentives can expenses, proportion of household appliances accounted and other
be fixed or changed with the quantity of saving. Social reward people's electricity consumption information.
mainly refers to encouragement through public propaganda.
Economic rewards can receive obvious effect of intervention at 5. Research direction of residential electricity
once, and social incentives are better than the economic incentives
in the long term. In recent years, the Internet has become an indispensable part of
Handgraaf et al. [48] conducted a research on the effect of people's daily life. The rapid development of cloud computing technol-
economic rewards and social rewards. He selected company's ogy and big data provide new opportunities for the study of residential
employees as target sample in Holland. Social rewards are positive behavior. However, many new challenges follow too. During the
descriptive reviews. And each of rewards have two forms, namely Internet era, the residents are not only the electricity consumption
give people rewards public or secretly. Researchers provided groups, but also belong to net citizen. More and more people are willing
different rewards according to employees’ weekly electricity con- to share some of their daily activities on the Internet, which are true
sumption during the period of 13 weeks. The results showed that reflection of the way of life, values, habits, educational level, and so on.
social incentives have better effect than the economic incentives. Residents’ social network activities and online shopping information
And public reward is better than the secret reward. Bertoldi et al. can be used to analyze the energy consumption behavior of residents.
[49] pointed out that in earlier years, economic incentives are Compared with the traditional way of questionnaire, the residents’
mainly used to support the promotion of efficient technology behavior in the social network more truly reflects the pattern of the
equipment for reducing the CO2 emissions, but not to change the residents’ behavior. So analysis of residents’ behavior by Internet is an
people's behavior. important direction to electrical consumption behavior. The Internet
(5) Feedback. Feedback provides the family with their electricity environment has also bring some new challenges in the field of
consumption and energy saving tips. Form of feedback is very rich residential electricity. For example, first of all, how to get the informa-
according to frequency. There are continuous feedback, daily tion of daily activities of residents and analyze them is a big challenge.
feedback, weekly feedback and monthly feedback, etc. According The second is how to discover the electricity consumption behavior
to the feedback content, there are daily electricity consumption, according to people's online information. Residents left a lot of
the monthly electricity consumption and annual electricity con- information on the Internet, but this information is not a direct
sumption. Some feedback includes synthetic index, such as daily mapping of consumption behavior. So it is necessary to conduct a
growth rate, month growth rate, and the rankings in the region. multidimensional analysis. In this case, the pattern behavior of
From the way of feedback, it is not only text messages but also web consumption behavior can be discover. Therefore, how to explore the
feedback through the Internet. Feedback's content can also be a pattern of behavior becomes a new challenge for this field. Based on the
series of number, and it can also be combined with multimedia summary of the previous literatures, this section puts forward two
pictures. Different form of feedback will have different effects. aspects of the analysis of the behavior in the Internet. The first is how
to use the Internet and data mining technology to dig the typical load
Carroll et al. [4] studied effect of provide knowledge of energy profile of electricity consumption and related pattern of behavior. The
efficiency tips with people through feedback, and the results showed second is to develop personalized intervention strategies for each
that the household who received a feedback once a month reduced the family based on the pattern.
power consumption by 2.9%. Nilsson et al. [36] provided feedback
through installing intelligent hardware device in people's house. (1) Big data analysis. With the help of big data and cloud computing
Feedback information includes daily electricity consumption, week technology, it is possible to store and analyze massive electricity
electricity consumption and electricity consumption trend. The results data, and to explore the pattern of electricity consumption. From
showed that intelligent hardware device did not have significant impact the point of time, the daily pattern of electricity consumption can
on electricity consumption. The researchers pointed out that an be explored. The pattern of the week, month, quarter and year, and
important reason is that people are lack of ability of understanding peak and valley will be found easily. From the perspective of

409
Z. Guo et al. Renewable and Sustainable Energy Reviews 81 (2018) 399–412

region, it can be used to explore the pattern of electricity researchers have made intervention experiment depend on single
consumption in different regions. It is also easy to discover these behavioral theory. In recent years, some scholars have begun to
households which share with similar load profile by means of data combine different behavioral theories into a new theoretical
mining. Electricity demand predication is another import issue framework and got good results. Therefore, it is a new trend to
that could be done batter by the big data technology. For example, combine different behavioral theories to make the intervention
Rathod and Garg [88] discovered electricity consumption pattern experiment, because each behavioral theory has some limitations,
at regional level in a city and extract knowledge concerning to the and the combination of different theories can overcome the
electricity consumption by data mining. Pre-processing of data, limitations. Any behavioral theory is not suitable in all cases, so
application of DM algorithms and discovered knowledge are the combination of the theory with regional cultural factors to form
involved. Zhou et al. [76] discovered consumption patterns of a new framework is the challenge of intervention strategy research.
residential users by fuzzy c-means (FCM) clustering method in The family is the basic unit of electricity consumption. Every family
Jiangsu Province in China. He found eight kind of typical has a different load profile, which reflects the behavior of family
electricity consumption profiles of residential users, and different members. Different families have different values and ways of life.
profile indicated people's different lifestyles. Hussain et al. [77] During the big data era, we can use cloud computing technology to
forecasted total electricity consumption by means of Holt-Winter mine the pattern of behavior that hidden behind load profile.
and Autoregressive Integrated Moving Average (ARIMA) models Furthermore, we can make different intervention strategies ac-
from 1980 to 2011 in Pakistan. The result was useful for Power cording to the different family. In this case, each family will receive
Company. Ramos et al. [78] discovered typical load profile by most suitable intervention strategies. In previous studies, scholars
combining classification and clustering. First, they clustered daily have provided the same intervention strategies for all families in a
load profiles into several typical profiles. Second, they used labeled region. In the big data era, the pattern of action could be
sample to train a multi-class model. Finally, they did the out-of- discovered in all aspects, so as to make a personalized intervention
sample prediction. This was a meaningful research, because strategy for each family according to the pattern of human
clustering will be difficult during big data era. Furthermore, behavior.
discovering people's behavior pattern hide behind typical load
profile is important. In order to achieve this goal. However, mining 6. Conclusions
users’ behavior is more difficult, because the people's behavior is
not a complete mapping form electricity consumption data. In This paper first the review vital factors that influencing people
earlier studies, researchers often investigated the behavior through energy consumption in the aspect of social psychology. We also review
the questionnaire. However, the questionnaire has problem of related behavioral theories that have effect on energy consumption
simple and low credibility. In the Internet environment, we can behavior, including social cognitive theory, social norms theory, theory
grasp the behavior of the residents from different perspectives. of reasoned action, theory of planned behavior, goal-oriented behavior
Social network is an important way to analyze the behavior of the model, value-belief–norm theory, norm activation theory, self-regulate
residents. Through the collection and analysis of data from social behavior change theory and ABC theory. Then, a review of the five
network, we can have a batter knowledge of the people's behavior, common strategies of intervention in residential electricity consump-
so as to get the pattern of the residents’ electricity consumption tion is conducted, namely, commitment, goal setting, providing in-
behavior. In this case, we should combine energy with big data into formation, reward, and feedback. Finally, in big data era, this paper
energy big data. Energy issue should be handled by many related points out the challenges and opportunities in field of residential
disciplinary. Zhou and Yang [79] proposed a framework of the electricity consumption.
interdisciplinary research of energy, social and information The factors reviewed in this paper are mainly related with social
science. Under this framework, a new perspective is that people psychological, there are many other factors that we do not consider.
energy consumption behavior can be analyzed in time dimension, Intervention strategies need to be determined in specific environment,
user dimension and spatial dimension. During the big data era, we cannot give a comprehensive review in aspect of the combination
there is a stronger connection between energy consumption and energy, information and behavior. This is the direction of our future
behavior. It is beneficial for us to discover typical load profile and research.
behavior pattern. Zhou et al. [86] pointed out that Energy Internet
will be a new development form of energy system. It realized the Acknowledgements
integration of energy flow, information flow and business flow. In
this case, electricity consumption was not merely an energy issues. This work is supported by the National Natural Science Foundation
Recently, deep learning, a very powerful machine learning algo- of China (Nos. 71501056, 71690235), China Postdoctoral Science
rithm, is applied in electricity load forecasting. For example, Foundation (No. 2017M612072), the Fundamental Research Funds
Dedinec et al. [90] used a deep belief networks make short-term for the Central Universities (No. JZ2016HGTB0728), Anhui Provincial
electricity load forecasting based on the Macedonian hourly Natural Science Foundation Program (No. 1608085QG165), Anhui
electricity consumption data during the period 2008–2014. The Provincial Philosophy and Social Science Planning Project (No.
results showed that the mean absolute percentage error was AHSKQ2015D42), and the Foundation for Innovative Research
reduced by up to 8.6% when using deep learning compared with Groups of the National Natural Science Foundation of China (No.
traditional method. Once obtaining typical load profile, on the one 71521001).
hand, the power company can convey it to the residents through
the visual feedback technology and stimulate the attention of References
residents for consumption of electricity. On the other hand, it
can be better to develop intervention strategies according to the [1] Zhou K, Yang S, Shen C, et al. Energy conservation and emission reduction of
pattern of consumption. China's electric power industry. Renew Sustain Energy Rev 2015;45:10–9.
[2] Zhou K, Yang S. Demand side management in China: the context of China's power
(2) Intervention framework and individual intervention strategies. The industry reform. Renew Sustain Energy Rev 2015;47:954–65.
theoretical basis of the experimental intervention is mainly based [3] Lopes MAR, Antunes CH, Martins N. Energy behaviours as promoters of energy
on various behavioral theory, such as social cognitive theory, social efficiency: a 21st century review. Renew Sustain Energy Rev 2012;16(6):4095–104.
[4] Carroll J, Lyons S, Denny E. Reducing household electricity demand through smart
norm theory, theory of reasoned action, theory of planned beha- metering: the role of improved information about energy saving. Energy Econ
vior, and norm activation theory. In the earlier years, the

410
Z. Guo et al. Renewable and Sustainable Energy Reviews 81 (2018) 399–412

2014;45(C):234–43. 2014;122:17–23.
[5] Schultz PW, Estrada M, Schmitt J, et al. Using in-home displays to provide smart [37] Hargreaves T, Nye M, Burgess J. Keeping energy visible? Exploring how house-
meter feedback about household electricity consumption: a randomized control holders interact with feedback from smart energy monitors in the longer term.
trial comparing kilowatts, cost, and social norms. Energy 2015;90:351–8. Energy Policy 2013;52(3):126–34.
[6] Batih H, Sorapipatana C. Characteristics of urban households‫ ׳‬electrical energy [38] Komatsu H, Nishio KI. An experimental study on motivational change for
consumption in Indonesia and its saving potentials. Renew Sustain Energy Rev electricity conservation by normative messages. Appl Energy 2015;158:35–43.
2016;57:1160–73. [40] Karjalainen S. Consumer preferences for feedback on household consumption of
[7] Leahy E, Lyons S. Energy use and appliance ownership in Ireland. Energy Policy electricity. Energy Build 2011;43(2–3):458–67.
2010;38(8):4265–79. [41] Han H. Travelers' pro-environmental behavior in a green lodging context:
[8] Yohanis YG, Mondol JD, Wright A, et al. Real-life energy use in the UK: how converging value-belief-norm theory and the theory of planned behavior. Tour
occupancy and dwelling characteristics affect domestic electricity use. Energy Build Manag 2015;47:164–77.
2008;40(6):1053–9. [42] Burchell K, Rettie R, Roberts TC. Householder engagement with energy con-
[9] Bartiaux F, Gram-Hanssen K. Socio-political factors influencing household con- sumption feedback: the role of community action and communications. Energy
sumption of electricity: a comparison between Denmark and Belgium. In: Eceee Policy 2016;88:178–86.
summer study proceedings energy savings what works & who delivers; 2005. [43] Matsui K, Ochiai H, Yamagata Y. Feedback on electricity usage for home energy
[10] Mcloughlin F, Duffy A, Conlon M. Characterising domestic consumption of management: a social experiment in a local village of cold region. Appl Energy
electricity patterns by dwelling and occupant socio-economic variables: an Irish 2014;120(120):159–68.
case study. Energy Build 2012;48(19):240–8. [44] Zhang Y, Wang Z, Zhou G. Antecedents of employee electricity saving behavior in
[11] Wilson C, Marselle MR. Insights from psychology about the design and imple- organizations: an empirical study based on norm activation model. Energy Policy
mentation of energy interventions using the behaviour change wheel. Energy Res 2013;62(7):1120–7.
Soc Sci 2016;19:177–91. [45] Han HS, Myong J, Jinsoo H. Cruise travelers' environmentally responsible
[12] Brounen D, Kok N, Quigley JM. Residential energy use and conservation: decision-making: an integrative framework of goal-directed behavior and norm
economics and demographics. Eur Econ Rev 2012;56(5):931–45. activation process. Int J Hosp Manag 2016;53(3):94–105.
[13] Katzev RD, Johnson TR. A social-psychological analysis of residential consumption [46] Harding M, Hsiaw A. Goal setting and energy conservation. J Econ Behav Organ
of electricity: the impact of minimal justification techniques. J Econ Psychol 2014;107:209–27.
1983;3(3–4):267–84. [47] Abrahamse W, Steg L, Vlek C, et al. The effect of tailored information, goal setting,
[14] Becker LJ. Joint effect of feedback and goal setting on performance: a field study of and tailored feedback on household energy use, energy-related behaviors, and
residential energy conservation. J Appl Psychol 1978;63(4):428–33. behavioral antecedents. J Environ Psychol 2007;27(4):265–76.
[15] Hutton RB, Mcneill DL. The value of incentives in stimulating energy conservation. [48] Handgraaf MJJ, Jeude MAVLD, Appelt KC. Public praise vs. private pay: effects of
J Consum Res 1981;8(3):291–8. rewards on energy conservation in the workplace. Ecography 2013;86(2):86–92.
[16] Scott FL, Jones CR, Webb TL. What do people living in deprived communities in [49] Bertoldi P, Rezessy S, Oikonomou V. Rewarding energy savings rather than energy
the UK think about household energy efficiency interventions?. Energy Policy efficiency: exploring the concept of a feed-in tariff for energy savings. Energy Policy
2014;66(1):335–49. 2013;56(2):526–35.
[17] Abrahamse W, Steg L. How do socio-demographic and psychological factors relate [50] Schwartz SH. Normative influences on Altruism 1. Adv Exp Soc Psychol
to households' direct and indirect energy use and savings?. J Econ Psychol 1977;10:221–79.
2009;30(5):711–20. [51] Stern PC, Dietz T, Abel T, et al. A value-belief-norm theory of support for social
[18] Yazdanpanah M, Komendantova N, Ardestani RS. Governance of energy transition movements: the case of environmentalism. Human Ecol Rev 1999;6(2):81–97.
in Iran: Investigating public acceptance and willingness to use renewable energy [52] Ajzen I. From intentions to actions: a theory of planned behavior. In: Action
sources through socio-psychological model. Renew Sustain Energy Rev control. Springer Berlin Heidelberg; 1985. p. 11–39.
2015;45(45):565–73. [53] Ajzen I, Fishbein M. Understanding attitudes and predicting social behavior.
[19] Botetzagias I, Malesios C, Poulou D. Electricity curtailment behaviors in Greek Prentice-Hall; 1980.
households: different behaviors, different predictors. Energy Policy [54] Ajzen I. From intentions to actions: a theory of planned behavior. In: Action
2014;69(6):415–24. control. Springer Berlin Heidelberg; 1985. p. 11–39.
[20] Fornara F, Pattitoni P, Mura M, et al. Predicting intention to improve household [55] Bamberg S. Changing environmentally harmful behaviors: a stage model of self-
energy efficiency: the role of value-belief-norm theory, normative and informa- regulated behavioral change. J Environ Psychol 2013;34(4):151–9.
tional influence, and specific attitude. J Environ Psychol 2015;45:1–10. [56] Dunlap RE, Van Liere KD. The new environmental paradigm. J Environ Educ
[21] Nachreiner M, Mack B, Matthies E, et al. An analysis of smart metering information 1978;9:10–9.
systems: a psychological model of self-regulated behavioural change. Energy Res [57] Guagnano GA, Stern PC, Dietz T. Influences on attitude-behavior relationships.
Soc Sci 2015. Environ Behav 1995;27:699–718.
[22] Mack B, Tampe-Mai K. An action theory-based electricity saving web portal for [58] Webb D, Soutar G, Saldaris P, et al. Self-determination theory and consumer
households with an interface to smart meters. Util Policy 2016. behavioural change: evidence from household energy-saving behaviour. J Environ
[23] Zainudin N, Siwar C, Choy EA, et al. Evaluating the role of energy efficiency label on Psychol 2013;35(35):59–66.
consumers’ purchasing behaviour. Apcbee Procedia 2014;10:236–330. [59] Perugini M, Bagozzi RP. The role of desires and anticipated emotions in goal-
[24] Cramer JC, Miller N, Craig P, et al. Social and engineering determinants and their directed behaviours: broadening and deepening the theory of planned behaviour.
equity implications in residential electricity use. Energy 1985;10(12):1283–91. Br J Social Psychol 2001;40(1):79–98.
[25] Gans W, Alberini A, Longo A. Smart meter devices and the effect of feedback on [60] Conner M, Armitage CJ. Extending the theory of planned behavior: a review and
residential electricity consumption: evidence from a natural experiment in avenues for further research. J Appl Social Psychol 2006;28(15):1429–64.
Northern Ireland. Energy Econ 2011;36(3):729–43. [61] Sherif M. The psychology of social norms. New York: Harper; 1966.
[26] Kavousian A, Rajagopal R, Fischer M. Determinants of residential consumption of [62] Cialdini RB, Reno RR, Kallgren CA. A focus theory of normative conduct: recycling
electricity: using smart meter data to examine the effect of climate, building the concept of norms to reduce littering in public places; 2010. p. 1015–26.
characteristics, appliance stock, and occupants' behavior. Energy 2013;55:184–94. [63] Kelman HC. Compliance, identification, and internalization: three processes of
[27] Filippini M, Pachauri S. Elasticities of electricity demand in urban Indian house- attitude change. J Confl Resolut 1958;2(1):51–60.
holds. Energy Policy 2004;32(3):429–36. [64] Allcott H. Social norms and energy conservation. J Public Econ
[28] Nakamura H. Effects of social participation and the emergence of voluntary social 2009;95(914):1082–95.
interactions on household power-saving practices in post-disaster Kanagawa, [65] Keizer K, Schultz PW. Social norms and pro-environmental behavior[J]. Microb
Japan. Energy Policy 2013;54(2):397–403. Ecol 2012;53(3):367–8.
[29] Anderson B, Lin S, Newing A, et al. Electricity consumption and household [66] Bandura A. Social foundations of thought and action: a social cognitive theory.
characteristics: implications for census-taking in a smart metered future. Comput Pearson Schweiz Ag 1986.
Environ Urban Syst 2016. [67] Thøgersen J, Grønhøj A. Electricity saving in households—a social cognitive
[30] Yau OHM, Mccoll-Kennedy JR. House structure, household social characteristics, approach. Energy Policy 2010;38(12):7732–43.
applicance ownership and electricity consumption: Elderly Australian consumers. [68] Yang S, Zhang Y, Zhao D, et al. Who exhibits more energy-saving behavior in direct
In: Proceedings of the 2nd international conference on comparative management; and indirect ways in china? The role of psychological factors and socio-demo-
1989. p. 176–81. graphics. Energy Policy 2016;93:196–205.
[31] Paul J, Modi A, Patel J. Predicting green product consumption using theory of [69] Katzev RD, Johnson TR. A social-psychological analysis of residential electricity
planned behavior and reasoned action. J Retail Consum Serv 2015;29:123–34. consumption: the impact of minimal justification techniques. J Econ Psychol
[32] Susanti L, Fithri P, Bestarina K. Demographic characteristics in correlation with 1983;3(3–4):267–84.
household electricity use. Ind Eng Manag Sci Appl 2015 2015:959–68. [70] Madden TJ, Ellen PS, Ajzen I. A Comparison of the theory of planned behavior and
[33] Newsham G, Birt BJ, Rowlands IH, et al. the effect of household of household the theory of reasoned action. Personal Soc Psychol Bull 1992;18(1):3–9.
characteristics on total and peak electricity use in summer. Energy Stud Rev 2011. [71] Rossi Ashley N, Armstrong James B. Theory of reasoned action vs. theory of
[34] Anderson B, Lin S, Newing A, et al. Electricity consumption and household planned behavior: testing the suitability and sufficiency of a popular behavior
characteristics: implications for census-taking in a smart metered future. Comput model using hunting intentions. Hum Dimens Wildl 1999;4(3):40–56.
Environ Urban Syst 2016. [72] Martin Jack. Self-regulated learning, social cognitive theory, and agency. Educ
[35] Susanti L, Fithri P, Bestarina K. Demographic characteristics in correlation with Psychol 2004;39(2):135–45.
household electricity use. Ind Eng Manag Sci Appl 2015;2015:959–68. [73] Bamberg S. Changing environmentally harmful behaviors: a stage model of self-
[36] Nilsson A, Bergstad CJ, Thuvander L, et al. Effects of continuous feedback on regulated behavioral change. J Environ Psychol 2013;34(4):151–9.
households' consumption of electricity: potentials and barriers. Appl Energy [74] Vitali M, Pernici B, O’Reilly UM. Learning a goal-oriented model for energy efficient

411
Z. Guo et al. Renewable and Sustainable Energy Reviews 81 (2018) 399–412

adaptive applications in data centers. Inf Sci 2015;319(C):152–70. [87] Wallisn H, Nachreiner M, Matthies E. Adolescents and electricity consumption;
[75] Liu PL, Lai HC, Yu YC. et al. Understanding household electricity saving in taiwan Investigating sociodemographic, economic, and behavioural influences on electri-
through the theory of planned behavior: Focusing on differences between user city consumption in households. Energy Policy, Energy Policy 2016;94:224–34.
types. In: Proceedings of IEEE international conference on industrial engineering [88] Rathod RR, Garg RD. Regional electricity consumption analysis for consumers
and engineering management. IEEE; 2015. p. 554–8. using data mining techniques and consumer meter reading data. Int J Electr Power
[76] Zhou K, Yang C, Shen J. Discovering residential electricity consumption patterns Energy Syst 2016;78:368–74.
through smart-meter data mining: a case study from China. Uti Policy [89] Pothitou M, Hanna RF, Chalvatzis KJ. Environmental knowledge, pro-environ-
2017;44:73–84. mental behaviour and energy savings in households: an empirical study. Appl
[77] Hussain A, Rahman M, Memon JA. Forecasting electricity consumption in Energy 2016:184.
Pakistan: the way forward. Energy Policy 2016;90:73–80. [90] Dedinec A, Filiposka S, Dedinec A, et al. Deep belief network based electricity load
[78] Ramos S, Duarte JM, Duarte FJ, et al. A data-mining-based methodology to forecasting: an analysis of Macedonian case. Energy 2016;115:1688–700.
support MV electricity customers’ characterization. Energy Build 2015;91:16–25. [91] Testa F, Cosic A, Iraldo F. Determining factors of curtailment and purchasing
[79] Zhou K, Yang S. Understanding household energy consumption behavior: the energy related behaviours. J Clean Prod 2015:112.
contribution of energy big data analytics. Renew Sustain Energy Rev [92] Suki NM. Consumer environmental concern and green product purchase in
2016;56:810–9. Malaysia: structural effects of consumption values. J Clean Prod 2015;132:204–14.
[80] Elster J. Social norms and economic theory. J Econ Perspect 1989;3(4):99–117. [93] Ramayah T, Lee JWC, Mohamad O. Green product purchase intention: some
[81] Özkan HA. Appliance based control for Home Power Management Systems. Energy insights from a developing country. Resour Conserv Recycl 2010;54(12):1419–27.
2016;114:693–707. [94] Olson-Hazboun SK, Krannich RS, Robertson PG. Public views on renewable energy
[82] Littleford C, Ryley TJ, Firth SK. Context, control and the spillover of energy use in the rocky mountain region of the United States: distinct attitudes, exposure, and
behaviours between office and home settings. J Environ Psychol 2014;40:157–66. other key predictors of wind energy. Energy Res Soc Sci 2016;21:167–79.
[83] Dixon GN, Deline MB, Mccomas K, et al. Saving energy at the workplace: the [95] Hast A, Alimohammadisagvand B, Syri S. Consumer attitudes towards renewable
salience of behavioral antecedents and sense of community. Energy Res Soc Sci energy in China—the case of Shanghai. Sustain Cities Soc 2015;17:69–79.
2015;6:121–7. [96] Kyriakopoulos GL, Arabatzis G. Electrical energy storage systems in electricity
[84] Wallis H, Nachreiner M, Matthies E. Adolescents and electricity consumption; generation: energy policies, innovative technologies, and regulatory regimes.
Investigating sociodemographic, economic, and behavioural influences on electri- Renew Sustain Energy Rev 2016;56:1044–67.
city consumption in households[J]. Energy Policy 2016;94:224–34. [97] Kyriakopoulos GL, Arabatzis G, Chalikias M. Renewables exploitation for energy
[85] Ntona E, Arabatzis G, Kyriakopoulos GL. Energy saving: views and attitudes of production and biomass use for electricity generation. A multi-parametric litera-
students in secondary education. Renew Sustain Energy Rev 2015;46:1–15. ture-based review. AIMS Energy 2016;4(5):762–803.
[86] Zhou K, Yang S, Shao Z. Energy Internet: the business perspective. Appl Energy [98] Nicolini M, Tavoni M. Are renewable energy subsidies effective? Evidence from
2016;178:212–22. Europe. Renew Sustain Energy Rev 2017:412–23.

412