th
Proceedings of the 20 International Conference on Automation & Computing, Cranfield University, Bedfordshire, UK, 12-13 September 2014
Behaviour interventions for water end use:An integrated model Yang Fu, Wenyan Wu Faculty of Art and Creative Technologies, Staffordshire University Stafford, Staffordshire, UK
[email protected],
[email protected] Abstract—Water authorities are facing the challenges to guaranty the need of consumers and water saving due to the climate change. In order to solve the problems, engineers need to understand how consumers use their water and their water use behaviours. This research is to re-analyse the effective technologies for researching the water use behaviours (behavioural models) in the water end use field, and proposes an integrated model for better understanding water end use behavioural intervention. It is recommended that the self-regulation attitude and the process of behavioral change play important roles and influent greatly on the behavioural intervention of water end use. An expanded theory of planned behavior model is used as the basis in the model and self-regulation theory also is adapted to modify the interventions. The integrated model proposed in this paper is to better understand the details of water end use behaviours, and intervention and its impact of energy saving. Keywords-water end use; behaviour intervetnion; water saving; modelling
I.
INTRODUCTION
Water conservation and prediction at the household level is an increasingly important issue to urban utilities [1] which take a considerable amount (representing over 65%) of the total water consumption (including industrial consumption etc.) [2]. Population growth, changes in land use, and climate change are putting pressure on existing water resources worldwide and it is not certain that supplies are adequate to meet the increasing demand for water [3]. The main issue in many regions about water is to diminish the water consumption and provide higher standard purified water [4]. The traditional water demand management methods to change consumers’ behaviours for saving water purpose include upgrading the network and modifying the water pricing feature (management strategies) etc. [5][6]. The method to save water from the engineering opinion is to adjust and optimize the water use pattern and upgrade the pipeline network. It cannot be denied that water saving by the engineering methods does make a difference. Former research have already been done in water use areas not only for the household level but also for the community and urban water supply system level. For the management strategies, governments and water providers are planning, or already have, introduced a lot of projects [7], such as non-pricing policies, incentives of water saving device and water use restrictions.
Another well-known approach for quantifying, conserving and predicting water demands of individual end uses is using the modelling method to understand about the individual’s attitude and behaviour for using water in the house [8]. This approach could help to have a better understanding to the behaviour of consumers and it is the way to evaluate and simulate the water policy, social factors and economic factor in the water industries. However, inevitably, the behaviour model, or even the basis of water end use model itself have disadvantages and limitations. Therefore, the proposal of integrated models of water conservation is sorely needed [9]. The aim of this paper is to propose a theoretial integrated end use model through the examination of current water use behavioural models and previours social and economical water use studies. And it could be different from the other similar models as this model take the behavioural interventions for consideration. The behavioural change model part could simulate the process of behavioural change independently by following the time series. The paper begins with a review of influcent factors for water end use and some sigificances casued by fully understanding how people think about water and their behaviour changed. Following with a section that investigates the findings of water end use behavioural studies that leads to the proposal of a new integrated end use behavioural model. The significant contributions of this integrated model are: •
characterisation the subjective and objective parameters for the water end use simulation study
•
development of the integrated model for better understanding the behavioural intervention
•
the implementation of an expanded TPB (Theroy of Planned Behaviour)
•
Application of self-regulation theory for the behaviour change intervention modelling II.
ISSUES ABOUT WATER USE BEHAVIOUR
Considering the condition for climate change and the increase of human population, the resources of drinking water is considered as a precious part of the nature which need be conserved and effectivly used by all of us. Reducing demand by improving the efficiency of water use necessitatesis essential and understanding that how water is used and in what ways water savings can be realized [4].
For the water use field, researches are mainly focused on the development of the BCM (behaviour change model) and application of the BCM levels. The application for BCM level the mature theory of BCM and apply it for water use research. There are a lot of research to change people’s water use behaviour or specific water-relevant behaviours [10] in this level. Specifically, the behaviour science can be used to understand users’ attitude and reactions towards different intervention of the water related business. For example, advertising campaigns conducted by Australia States and Territories in 1996 were cantered on a “water is money” concept and followed a price increase for water billing. Moreover, research in the water use field is to reduce the water consumption for household, community and individual. A study run by Kurz and Donaghue used combination information leaflets, labels and feedback to change water use behaviour and they have achieved about 23% reduction in 2005 [6]. The development of BCM level means studies make it into BCM more specific framework for water use. The behaviour change method used for the water research is based on the original and the modern psychological research and use it to develop a new framework for water use research. Like the UKWIR project CU02 in 2012, the research team applied the MINDSPACE behaviour change framework to encourage greater use of “self service” facilities and build a new framework for water use behaviour change. The application of BCM in the water use field is using the social and economic model to simulate the human behaviour influence to the water consumption. Inevitablly, This method has its weakness. The explanation for their disadvantages is that these method normally just select the behaviour change parameters, such as attitude and intention, as input factors for caculating by the mostly used modelling techniques. Even though a large number of researchers have presented the models of water end use stimulation and prediction [4][11][12], they are still not sufficient to address the issues about how our human behaviours changed in the water end use process and how the changed behaviours influence the water consumption. III.
INFULENCE FACTORS FOR WATER END USE
A range of infuence factors for water use behaviors have been identified. All the factors, which influent the consumer’s behaviours could be defined as direct drives or indirect drivers [4][13][14]. Bradley Jorgensen et al. concluded his research result about the drivers for consumer’s water use in his research (see Table 1) [4]. All the analysis about the drivers for water use have different discretionary characteristic espacially for outdoor use. The reason for the consumption diversity of outdoor use is due to the habit of people using their outdoor facilitits and the condition (size and kinds) of them. Indoor water use situation could be more stable compared with outdoor use. The influence factors for indoor water use include house condition, regulatory enviorment and personal characteristics. Moreover, by considering the water use behaviour itself, the behaviours are more likely to be influeced by
consumers’ mind. Specifically, behavioural belief, normative belief and control belief (Theory of Planned Behaviour) are the main factors, which could alter what consumers are thinking about their water use behaviours. Those factor are more relied on the education standard and the responsibilty they could feel about their water use condition etc. And those factors are directly or indirectly influenced by water management stragegies and policy from Water unitiles and authorities . In sumary, the impacts of these influecing factors on the water end use consumption, the processing influence end use, how important it could be and how they interact with each other, are still in question. Researches are intended modelling all the factors which change water end use. TABLE I.
DRIVERS FOR WATER END USE[4]
Direct-Drivers
Indirect-Drivers
Climate/seasonal variability (e.g.,tariff structure and pricing etc.) Regulations and ordinances (e.g., water restrictions etc.) Property characteristics (e.g., lot size, pool, bore etc.) Household characteristics (e.g., household composition etc.) Person characteristics(eg. Intention, knowledge about water) Incentives/disincentives (e.g., water restrictions, local government planning regulations etc.)
Person characteristics (eg. subjective norm, behaviour Control ) Institutional trust (i.e., trust in the water provider) Inter-personal trust (i.e., trust in other consumers) Fairness Environmental values conservation attitudes
&
Socio-economic factors (e.g., income, household composition, age, gender, education, etc.) Intergenerational equity
IV.
WATER END USE MODELLING
A. Technologies Water end-use modelling is an approach for quantifying and predicting water demands of individual end uses using the relationships built on the data monitored at individual household scale [15]. The main purpose for residential end-use modelling is to quantify end-uses by different water use devices. Changes in appliances and water use behaviour lead to different water demand patterns. An end-use model, therefore, acts as a predictive model and can be utilized in the design stage of water distribution network and water suppply networks where no household water meters are installed [16]. Generally, this modelling method is based on the time at the volume of end-use events start and end. The basic structure of end-use modelling method is shown in Figure 1. The water end-use modelling requires the application of analytical techniques, (likes the stochastic modelling, multi-variable regression and Bayesian networks) and household water use database for supporting that modelling work [17]. A lot of researches have presented that models for water end use simulation and prediction are capable [4][11][12].
demand management programs (rebates or water smart meters etc.) can be effective for water conservation.
Figure 1. General diagram of the model for household water end-use consumption
A stochastic end-use model based on end-use category frequency of use, demographics, event occurrence likelihood and flow duration to simulate water demand patterns have been developed by Blokker, Vreebury and Dijk [16]. Blokker used a Poisson rectangular pulse model derived from metering studies and surveys. This final model has achieved up to 93% accuracy compared to the truly metering data. Bayesian conditional framework and ANNs(Artificial netural networks) have also been used as very good methods for the water end use modelling. Hsiao & Mountain used the Bayesian framework develops a forecasting model for heating water and the residential water demand uses [18]. The dummy variables and the transforming variables in this model are combined with aggregated loads, appliance ownership and demographic information. This study was applied to 396 households and with acceptable relative errors ranging from 0.081 to 0.298. Christopher Bennett and Rodney A. Stewart[17] used the data from SEQ (South-east Queensland) in Australia (over 250 households’ water end-use data) to build an ANNs model. The ANNs provide the technique for aligning the databases to extract the key determinants for different end-use category. Finally, their model had error proportion values of 0.33, 0.37, 0.60, 0.57, 0.57, 0.21 and 0.41 for toilet, tap, shower, clothes washer, dishwasher, bath and total internal demand, respectively. B. Social and Economic Model Social models for water end use behaviour are used for predict household water consumption by considering the social and human behavioural factors. Syme et al. investigated the house owners’ attitudes against their water consumption result during a whole year in Perth, Australia [19]. His model showed that effective demand management strategies could change the users’ behaviours in combination with water price adjusts. The other example about the social model is to use the alternative water sources and technologies to investigate the receptivity of community by Clark Brown [20]. As a result, all the water saving ability and reuse power relies on the condition of behaviour change capacity. For the economic models, it indicates that price, price structuring and use restrictions have a direct influence on household water use [4]. Some researchers suggest that household characteristics plays a major role in the determination of the water use, and behaviours needs to be informed by local information about how to use and save water in the house. Moreover, such as Kenney et al.’s research[6] pointed out that residential demand depends on a lot of factors like price. Therefore, the non-price
However, despite of the accuracy of all the different kinds of models, currentwater use models take no differences between the objective and subjective parameters for simulation. All the parameters are selected to input the model and calculated by the model itself. Taking the social model of household outdoor water consumption as example [1], it analysed and input some social factors concerned attitude and behaviour detials for calculation, but it did not give any information about behavioural intervention. Some economic models like hoffmanne et al.’s[21] model and Kenney et al.’s[6] model, are considered economic and social factors, even involved in behavioural parameters (subjective parameters), however, most of those models only regard behaviour formation factors as a single or multi parameters in the models but not designed and considered the behaviour intervention. We believe that the development of behaviour model as an initial calculation model should be taken into account to play a role for understanding the behaviour process. This proposed concept is based on the complication of human behaviour and lead to a better understanding for consumers’ behaviour precisly about how social and economic factors influence on water use and by what methods we can adapt to change consumers’ behaviours for water conservation purpose. According to the review, like example models above, many studies have discovered different factors acting on water end use. However, no study was able to attribute all the variation in water use to the factors they examined [4]. V.
DEVELOPMENT OF THE INTEGRATED MODEL
A. Architecture and framework The primary aim of this research is to improve the understanding of behavioural water end-use change interventions by proposing an integrated water end use model. This model will consider how important system variables respond to changes in input variables overtime. Research already pointed out that integrated model can be used to answer questions regarding what variables to be measured, its measure frequency, speed for intervetion and the functional form of the outcome responses as a result of decisions regarding the timing, spacing and dosage levels for intervention components [22]. In order to achieve this goal, this paper develops an integrated model for end-use simulation and psychological considerations. For the end-use simulation and modelling component, we rely on the artificial neural networks (ANNs) which have been regards as the most suitable technique to exploit the water use data and develop water end-use model [17]. If the model can be given sufficient sample, it could be trained with less errors and higher accuracy. For the psychological component, this research will present a behavioural model as a part of the integrated model for studying attitudes, intentions, subjective norms and other modelling variables. These variables may be impacted by an intervention, which can lead to the conservation of the end-use and more reasonable water use habits over time. In this part, the widely accepted
theory of planned Behaviour (TPB[23]) is used as a basic modelling technology for the psychological model and will be expanded as ETPB model. ETPB is also connected with self-regulation theory to control the feedback or inverse response in the behavioural change processing. Figure 2 shows the general conceptual diagram for the intelligent integrated philosophy model for household water end use.
considering researches already been done for characterize the factors of water end-use. We used the TPB to define the main parameters of subjective and the objective parameters are from the water use researchers’ results [24] [25]. TABLE II. SUBKECTICE WATER END USE Subjective Parameters
Figure 2. General diagram of the Integrated model for household water end-use consumption change
Influence factors of water end use in this model will be separated into two kinds by reviewing the former researches about water use behavioural change. The criterion of selection for those two kinds of parameters is based on TPB and the social and economic model reviewed above. This paper suggests using ETPB model as a independent model and integrated with ANNs end use model for simulation, As we believe that behavioural change model should be existed as a part in the end use model for better understanding the behavioural interventions but not just input behavioural, social and economic factors as parameters. By considering all the improvements for end use modelling above, this model could not only have a better understanding of behavioural intervention, but also can be used to fix the weak points of former end use modelling, especially for guiding proper behaviours for saving water. B. Subjective and Objective Parameters As discussed above, all the parameters are classified into two types to distinguish their differences. This work could help us gain a better understanding about the parameters which could influence on the human mind and which could not. The subjective parameters are those parameters that could be influenced by human mind, such as attitude, attention to the different household water use behaviour which could only exist in the TPB/ psychological model. The objective parameters are the those that already exist and could not be changed easily or at all, such as education level, house size and whether the located in rural area or not. Parameters are chosen by
AND
OBJECTIVE PARAMETERS FOR
Objective Parameters
Attitude
Household characteristics
Subjective norm
Climate/seasonal variability
Moral norm
Person characteristics
Beliefs about behaviours
Regulations and ordinances
Intention
Incentives/disincentives
Past behaviours
Socio-economic factors
Descriptive norms
Property Characteristics
C. Expannded Theory of Planned Behvio;ur Ajzen and Fishbein’s (1980) theory of planed action is probably the best-know attitude-behaviour model and incorporates external factors (normative social influences) on behavioural intention. In this theory, the underlying assumption is that humans are rational and make systematic use of available information [26].This theory builds on expectancy value theory to incorporate normative social influences on behavioural intention. It gives the concept for linkages between beliefs, attitudes, perceived social norms and behaviours.
Figure 3. General diagram of the expanded theory of planned behaviour model
S. J. Kantole et al. [27] found that the TPB model is useful for explaining intentions to conserve water and the most significant influence on intentions to save water could be explained by subjective normative feeling and the ages. A later study by kantole et al.[28] about the ability of persuasive communications with different levels of intensity to water saving influence proved that it did not support the dominant mediation role. Po et al.[29] and Syme and Nancarrow [19] used TPB model to explain the extent to which intended behaviour could predict the user responses for the water supply system.
A review of literature on household water conservation points out the importance of the additional predictors of household water use [30]. In addition to the input parameters of TPB, we also suggest that the measures of descriptive norms, moral norms and former behaviour are also need take roles in the processing of behaviour interventions. Descriptive norms is the parameters define normal behaviour which means what people ought to [31] and the moral norm is to measure the duty to conserve [32]. Note that the TPB has its own weakness as it did not describe clearly the interaction of subjective norms, it could be necessary to add more influence factors into calculation. Figure 3 shows the diagram for the expanded theory of planned behaviour model. D. Self-Regulation Carver and Scheier (1998)’s theory [33] in psychology has been great influenced on this research. he proposed that human behaviour is goal-directed and regulated by feedback control process. It reflects the human preceived power and capacity and proposes how those parameters leading to further human behaviours. In this paper, this theory is adopted and implemented as a controller in the model of ETPB that adjusts the precevied behavioural control. Control engineering knowledge is combined with modelling stage to simulate the human behaviour intervention processing [34]. E. Survey and data collection To support this model, the data about water use and survey is needed. As mentioned above, the drives of water use and saving can be seen as the support to choose our input for modelling especially for TPB model. However, this is still not sufficient to meet our simulation requirment. Cara Beal et al., 2011 [35] concluded with a discussion of the general characteristics including the social perspectives of water consumption, sociodemographic trends, gender and education; household water appliances. All the factors could affect the household water use from 1985 residents in Australia. The analysis conclusion showed which factors could influence on the water use much more. We will use it to identify the different parameters influences to the model. The third method to support the parameters selection is the TPB model. In Icek Ajzen’s research to the TPB model, the sample TPB standard questionnaire will help this model to know which factors could influence and they will be selected as subjective parameters [23]. If this model is used as a dynamic model, all the parameters in the model should be long-time observed. F. Modelling Technical Basis The Expanded TPB’s standard mathematical representation relies on path analysis of Structural Equation Modelling (SEM). There are many examples of developed social and economic model by using this method. For instants, Geoffrey J. Syme et al.(2004)’s model [1] for predicting and understanding home garden water use, it used SEM analysed 11 parameters which could influent the garden water use and proved that its high accuracy. Daniel E. Rivera in 2012 [36] reviewed system identification and control engineering could offer
optimization of behavioural interventions. In his research, he gave the examples of computer programmed TPB model by using SEM for water use study[37]. By transferring the equations of ETPB to higher-order derivatives, accompanied with the help of controller of self-regulation, it is possible to better understand the behaviour interventions by using modelling method. JEmeterio et al. in 2011 applied this mathematical and modelling method to build a dynamic model for describing behavioural interventions for weight loss and body composition change which has been proved a success with high accurarcy. All the subjective parameters calculated in the SEM and dynamic model then input with objective parameters and water use data into ANNs end use model. By trained properly, ANNs end use model could give prediction results of water end use by concerning the behavioural change intervention processing details. VI.
CONCLUSION
This paper has reviewed a number of water end use models and the behavioural change technologies which applied in water use. It is found that current research have not suficiently address on what happened in the processing of water end use behavioural intervention. The most models for water end use have considered the social and economic factors by using some behavioural change theory. Also, little researches were able to define all the factors which could influence on water end use behaviour. We suggest that the influencing factors of the water use behaviour should be classified into subjective and objective parameters in order to have a clear understand of its intervention. The research in this paper proposed an integrated model by using the ETPB model. This model could be used as new behavioural model for simulation of water end use consumption even behaviours, and this integrated model could also be combined with energy model for energy saving purpose. Extensions of this model include implement work with practical application and modification of model. A longterm goal is to develop a dynamic water end use model which is based on this theoretical model and could be used for guiding consumer to lead to proper water conservation behaviours and saving energy, but not just water. ACKNOWLEDGMENT The authors acknowledge the financial support of the European the Seventh Framework Program (FP7) WatERP (318603) and FP7 Marie Curie ActionsSmartWater (PIRSES-GA-2012-318985). REFERENCES Geoffrey J. Syme, Quanxi Shao (2004) Predicting and understanding home garden water use. Presence: Land and Urban Planning, 68(2004): 121-128. [2] Tingyi Lu (2007) Research of domestic water consumption: a field study in Harbing, China. Master of Science thesis,Loughborough University, September, 2007. [3] Bates, B., Kundzewics, Z.W., Wu, S. and Palutikof, J.(2008). Climate Change and Water; IPCC Technical Report VI, IPCC Secretariat, Geneva. [1]
[4]
[5]
[6]
[7]
[8] [9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
Bradley Jorgensen, Michelle Graymore, Kevin O’Toole.(2009). Household water use behaviour: An integrated model, Journal of Environmental Management 91(2009) 227-236. Chen, Y., Zhang, D., Sun, Y., Liu, X., Wang, N., Savenije, H.H.G., 2005. Water demand management: a case study of the Heibe River Basin in China. Physics and Chemistry of the Earth, Parts A/B/C 30. 408-419. Kenney, D., Goemans, C., Klein, R., Lowery, J., Reidy, K., 2008. Residential water demand management: lessions from Aurora, Colorado. Journal of American Water Resources Association 44, 192-207. Marshen, J. and Pickering, P. 2006. Securing Australia’s urban water supplies: opportunities and impediments. A disscussion paper prepared for the Department of the Prime Minister and Cabinet, Marsden Jacob Associates, Camberwell. Peter W.Meyer et al. (1999) Residential End Uses of Water, American Water Works Association, 1999. Land & Water Australia, 2006. Water Perpectives: Outcomes of an Expert Work-shop Scoping Social and Institutional Research Questions in Support of the Implemention of the National Water Initiative. Land & Water Australia, Canberra, ACT. Ezra M. Markowitz and Bob Doppelt (2009). Report: Reducing Greenhouse Gas Emissions Through Behavioural Change: An Assessment of Past Research On Energy Use, Transportation and Water Consumption, January 2009. Luaky-Froukh, M. (2001). Decision Support System for Domestic Water Demand Forecasting and Management. Water Resources Management, 15, 363-382. Polebitski, A. and Palmer, R. (2010). ”Seasonal Residential Water Demand Forecasting for Census Tracts.” J. Water Resource Plan and Manage. 136(1), 27–36. Corral-Verdugo, V., Frias-Armenta, M., Perez-Urias, F., Orduna-Cabrera, V. and Espionza-Gallego, N., 2002. Residential water consumption, motivation for conserving water and the continuing Tragedy of the Commons. Environmental Management 30, 527-535. Gregor, G., Di Leo, M., 2003. Repeated behavior and enviornmental psychology: the role of personal involvement and habit formation in explaining water consumption. Applied Social Psychology 33(6), 1261-1296. K. Rathnayaka, H. Malano. (2011). Review of residential urban water end-use modelling: 19th International Congress on Modelling and Simulation, Perth, Australia, 12-16 December 2011. M.Blokker, E., Vreeburg, J., & Dijk, J. (2010). Simulating residential water demand with a stochastic end-use model. Journal of Water Resources Planning and Management, 136(19), 19–26. Christopher Bennett, Rodney A.Stewart, Cara D. Beal (2013). ANN-based residential water end-use demand forecasting model. Expret Systems with Applications 40(2013) 1014-1023. Hsiao, C., Mountain, D., & Illman, K. (1995). A Beyesian integration of end-use metering and conditional-demand analysis. Journal of Business & Economic Statistics, 13(3), 315-326. Syme GJ & Nancarrow BE (1992). Predicting public involvement in urban water management and planning. Environment and Behavior 24 (6), 738–758. Clark, J.M., Brown, R.R, 2006. Understanding the factors that influence domestic water consumption within Melbourne. Australian Journal of Water Resources 10, 251-258. Hoffmann, M., Worthington, A., Higgs, H., 2006. Urban water demand with fixed volumetric changing in a large
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33] [34]
[35]
[36]
[37]
municipality: the case of Brisbane, Australia. The Australian Journal of Agricultureal and Resource Economics 50, 347-359. J.-Emeterio Navarro-Barrientos, Daniel E. Rivera (2011)A dynamical model for describing behavioural interventions for weight loss and body composition change, in: Mathematical and Computer Modelling of Dynamical Systems, 17: 2, 183-203, 12 January 2011. Icek Ajzen (1991). The Theory of Planned Behaviour. Organization Behaviour and Human Decision Processes. 50, 179-211. Campbell, H.E., Johnson, R.M., Larson, E.H., 2004. Prices, devices, eople, or rules: the relative effectiveness of policy instruments in water conservation. Review of Policy Research 21, 637-662. Nancarrow, B.E., Po, M. and Porter, N.B. 2004. Why Doesn’t it work? The Incorporation of Community Culture in the Evalucation of Water Sensitive Urban Designs. In cities as Catchments – WSUD 2004; 2004 International Conference on Water Sensitive Urban Design, Adelaide, Australia. Andrew Darnton, Practical Guide: An overview of behaviour change models and their uses, Government Social Research Unit, UK, 2008. S. J. Kantola, Syme, G. J., and A. R. Nesdale (1993). The Effects of Appreised Severity and Efficacy in Promoting Water Conservation: An Iformational Analysis. Journal of Applied Social Psychology, 12, 164-182. Doi: 10.1111/j.1559-1816.1983.tb02328.x S. J. Kantola, S. J., Syme, G. J., & Campbell, N. A.(1982). The Role of Individual Differences and External Variables in a Test of the Sufficiency of Fishbein's Model to Explain Behavioral Intentions to Conserve Water. Journal of Applied Psychology, 12, 70-83. Doi: 10.1111/j.15591816.1982.tb00850.x Po, M., Nancarrow, B., Levistion, Z., Porter, N., Syme, G., & Kaercher, J. (2005). Predicting Community Behaviour in Relation to Wastewater Reuse: What Drives Decisions to Accept or Reject? Melourne: CSIRO. Russell, S. and Fielding, K. (2010). Water demand management research: A psychological prespective. Water Resources Research, 46, W0253502, doi:10.1029/2009WR 008408. Cialdini, R.B., Reno, R.R. and Kallgren, C.A. (1990). A focus theory of normative conduct: Recycling the concept of norms to reduce littering in public places. Journal of Personality and Social Psychology, 58(6), 1015-1026. Manstead, A.S.R. and Parker, D. (1995). Evaluating and extending the theory of planned behaviour. In W.Stroebe and M. Hewstone (Eds), European Review of Social Psychology Bulletin, 34, 913-923. C.S. Carver and M. F. Scheier. On the Self-Regulation of Behavior. Cambridge University Press, 1998. D.E. Rivera, M.D. Pew, and L.M. Collins. Using engineering control principles to inform the design of adaptive interventions a conceptual introduction. Drug and Alchol Dependence, 88(2):S31-S40, 2007. Beal. C, Stewart, RA., Fielding, K. (2011) A novel mixed method smart metering approach to reconciling differences between preceived and actual residential end use water consumption. Journal of Cleaner Production, 2011, doi:10.1016/j/jclepro.2011.09.007. D.E. Rivera. Optimized behavioural interventions: what does system identification and control engineering have to offer?. 16th IFAC Symposium on System Identification (SYSID 2012), 2012. T. Raykov and G.A. Marcoulides, A First Course in Structural Equation Modeling, 2nd ed. Erlbaum, Mahwah, NJ,2006.
