SEARCH

SEARCH BY CITATION

Keywords:

  • Australia;
  • water conservation;
  • behaviours;
  • sustainable living;
  • drought;
  • prestige

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Approaches to understanding consumption behaviours
  5. The study area
  6. Price of water in the study area
  7. Methodology
  8. Understanding stated and actual water use
  9. The prestige of sustainable living
  10. Conclusion
  11. Acknowledgements
  12. References

The paper outlines water conservation behaviours and assesses the level of congruity between the stated water use of householders against their actual metered consumption. A profile of high water users in three parts of South Australia is offered: two metropolitan areas differing in socio-economic characteristics and a regional town. The research used a postal questionnaire, a follow-up telephone interview and corresponding household water meter readings. Location, household size and annual household income have significant predictive qualities for high per capita water use. The number of times gardens were watered in a week, watering the garden more often than was permitted under the restrictions, and the manner in which conservation behaviours were carried out helped predict high per capita water use. Participants had an accurate idea of the magnitude of their water use and how it compared with that of other households. High water users knew that they were high consumers of water. Implications of the findings for water demand management are briefly outlined.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Approaches to understanding consumption behaviours
  5. The study area
  6. Price of water in the study area
  7. Methodology
  8. Understanding stated and actual water use
  9. The prestige of sustainable living
  10. Conclusion
  11. Acknowledgements
  12. References

Increasing resource consumption (water, energy) is a concern where supply battles to keep pace with demand. Limited storage capacity of reservoirs, anticipation of future use, the development of new infrastructure and alternative sources are key factors in demand–supply management (OECD 2009). Water demand management needs to predict future water demands and on that basis develop alternative water supplies or put in place strategies aimed at ensuring future supplies are adequate to meet the demand, whilst allowing for often significant lag times between the implementation of strategies and the realisation of additional supplies. Concerns over how to meet water demands are not a new problem for water mangers (Balchin 1958; Gordon-Spencer et al. 1973); however, whereas urban water planning previously entailed projecting future water demand, identifying exploitable water resources, and designing the infrastructure from source to the point of delivery, society's values and preferences are changing and thus water management, policies and practices need to evolve accordingly (Boland and Baumann 2009). There has been international movement towards sustainable consumption (Jackson 2004), and greater efficiency in resource use.

Given the limited availability of exploitable water resources, the pros and cons of alternative water supplies (grey and reclaimed water reuse) and decentralised systems are being widely debated. The development of alternative supplies is not limited by technological knowledge but by public acceptance and institution regulation and policy frameworks. There is a strong need for public consultation and public acceptance of alternative water supplies before they can be considered as viable supply options. Further, the risks associated with each component of centralised and decentralised water systems need to be clearly defined and regulated. The latter is often hindered by complexities over jurisdictional boundaries (Leflaive 2008).

The social, environmental and economic costs associated with water supply projects is leading to a stronger focus on demand management than supply augmentation (Baumann 1990). There has been much consideration on the economic drivers of water demand. Despite this, there is debate on the value of market-driven approaches and restrictions as effective means of curtailing water demand (Aisbett and Steinhauser 2011; Alston and Mason 2008; Olmstead and Stavins 2008; Pumphrey et al. 2008). Further, during times of water scarcity multiple strategies are often deployed simultaneously, making it difficult to identify the determinant of lower water consumption. Market-driven approaches raise questions of equity and water as a basic right and for this reason non-price mechanisms are widely favoured in demand management (Olmstead and Stavins 2008). Besides the economic drivers, non-market approaches to water demand management and conservation planning range from the technological solutions (such as eco-settings on washing machines), restrictions on land use development (and concomitant number of water connections on a supply system), restrictions (limit water use), infrastructure management (leak detection and repair), and media campaigns appealing to behavioural change (Inman and Jeffrey 2006).

According to Baumann (1990, 13) ‘not all practices that reduce water use should be considered desirable. The beneficial effects of the reduction in water use (loss) must be considered greater than the adverse effects associated with the commitment of other resources to the conservation effort’. For water management to be regarded as ‘conservation’ there should be no net loss in social welfare (Baumann 1990). Driven by the demand for high water supply standards, and growing environmental concern among consumers and water service providers alike, it is becoming common place for water conservation planning to be seen as an integral part of urban water services provision (Baumann 1990), but a need for regulatory policy to form a substantial part of urban water conservation policies (Pumphrey et al. 2008).

Making sense of high water user behaviour

Given water managers' focus on sustainable consumption over the exploitation of resources, the need for data to inform policy decisions is of paramount importance. A better understanding of the drivers of water consumption and the behaviours and characteristics of high water users will assist water demand management (Gregory and Di Leo 2003; OECD 2008). A profile of high water users cannot easily be given as a list of attributes as high users possess a variable and complex conglomeration of differing social, demographic and economic characteristics. In addition, the relationship between a demographic attribute and water use is not simple. Beal et al. (2008) argue that older people used more water but speculated that the higher water use was a function of age-related activities (due to the amount of time spent at home and flushing the toilet) rather than age as such. Likewise, the age–water use relationship described by Yabiku et al. (2008) was attributed to the desire among younger families wanting lawns as play areas for children over desert gardens. An analysis reveals that higher water use occurs more as a result of water demanding behaviours (Domene et al. 2005; Yabiku et al. 2008) than demographic characteristics alone. There is a need to avoid a simplistic analysis between these factors and actual consumption. Likewise the availability or installation of water-saving technologies does not necessarily result in lower consumption where technologies are not taken up or where behavioural changes offset the benefits (Campbell et al. 2004; Olmstead and Stavins 2008).

The study examined the relationship between stated and actual residential water use during a protracted drought in Australia. Specifically, the objectives were to compile a profile of the characteristics and behaviours of high water users in three study areas in South Australia and conduct an analysis of the level of congruity between respondents' self-proclaimed water conservation behaviours and their actual water use. In the next section of the paper the differing methodological approaches to understanding consumption behaviours provide some insight into the choice of methodologies in water use studies. An outline of the characteristics and price of water in the study areas and details of the data collection are given to provide context for the results that follow. The results section is structured as follows: first, the actual water use of residents is presented; this is followed by two sections that highlight the characteristics and behaviours typical of householders with high water use; based on this analysis, the predictors of high water use are then outlined. The closing discussion points to the prestige associated with sustainable living.

Approaches to understanding consumption behaviours

  1. Top of page
  2. Abstract
  3. Introduction
  4. Approaches to understanding consumption behaviours
  5. The study area
  6. Price of water in the study area
  7. Methodology
  8. Understanding stated and actual water use
  9. The prestige of sustainable living
  10. Conclusion
  11. Acknowledgements
  12. References

There are different approaches to understanding consumption behaviours. The attitude–behaviour model used in the discipline of environmental psychology builds on the Theory of Planned Behaviour (after Ajzen 1991) in which the intention to perform certain behaviours is predicted from consumers' attitudes, subjective norms and perceived behavioural control. That is, the people who are most likely to conserve water are those that believe water is worthy of conservation, feel a sense of moral obligation to society to conserve water and are physically capable of carrying out efficient water-using behaviours. Put simply, pro-environmental attitudes are seen as a proxy for behavioural intent. However, one shortcoming of the attitude–behaviour approach is that strong pro-environmental attitudes do not necessarily translate into corresponding behaviours (Berenguer et al. 2005). While the attitude–behaviour model has been further developed to include the effects of stimuli such as environmental awareness, reasoned processes (attitudes), unreasoned processes (habits), personal capabilities (age, education, income, environmental awareness) and contextual factors (technology available, rebates, pricing) to ascertain the drivers of water use (Gregory and Di Leo 2003), it falls short when the level of congruity between attitudes and actual water use is unknown (Russell and Fielding 2010). Behavioural patterns shaping water usage matters (Kose et al. 2004 cited in Mui et al. 2007). However, studies that focus on self-reported behaviours alone without correlating them with actual consumption (cf. De Oliver 1999) tend to lead to speculative outcomes. This paper seeks to address this shortcoming by examining both behaviours and actual water use in Australia.

Another approach to understanding consumption behaviours is a sociological focus based on the structuration theory of Giddens (1984) that argues that social behaviour is a product of individual agency in interaction with the social system. Importantly, for this approach behaviours need to be understood as conscious decisions that are shaped by the particular social system. Spaargaren (2003) develops this into The Social Practices Approach in which the actual social practices of consumers (agency) are analysed within the context of the production system (social system) evident in everyday life. In this model resource consumption (or conservation), as an act of individual agency or behaviour, is facilitated or hindered by systems of provision or production. The key attribute of this approach is related to a perspective that does not see ‘environment’ as a key determinant of practice, but rather practices are framed by social settings and are separate from specific anti-social behaviours such as profligate resource use. Shove (2003) postulates that water consumption is inexorably linked with social practices determined by a desire for comfort, cleanliness and convenience. The findings of other empirical studies (Allon and Sofoulis 2006; Gilg and Barr 2006; Saurií 2003; Syme et al. 2000; Troy and Randolph 2006) are in keeping with social practices theory, as are the findings in this paper. Shove and Walker (2010) extend the social practices debate by exploring the development and disappearance of sustainable practices in the context of socio-technical transitions. Yet a technology–practices divide is evident in the findings of Allon and Sofoulis (2006), who report on residents in Western Sydney who preferred to wash dishes in the sink despite having a dishwasher because of the social interaction afforded by the activity. Given the tendency for technological advances to be seen as the panacea for meeting water demand amid dwindling resources, the work of Shove and Walker (2010) is particularly pertinent. The findings of Allon and Sofoulis (2006), as with Shove and Walker (2010, 476) highlight ‘the importance of attending to all requisite elements of practice and individual agency, to forms of practical know-how, bodily activities, meanings, ideas and understandings, as well as to materials, infrastructures and sociotechnical configurations’.

This paper examines the self-stated behaviours of water consumers and correlates these against actual water use, enabling the authors to compile a profile of the characteristics and behaviours of high water users. In the broader study (than is presented here) the attitudes of water users were also examined to assess the level of congruity between conservation attitudes, self-stated behaviours and actual water use. The results showed that a number of attitudes had significant predictive qualities for high water use. However, they were found to be strongly linked to demographics in that once demographic and socio-economic differences were statistically controlled for they were not significant, hence attitudes as predictive qualities for high water use are not discussed further in this paper. Further, for the sake of clarity and length, this paper examines only the former variables while the attitudinal aspects affecting water use are discussed in Pearce et al. (2012).

The study area

  1. Top of page
  2. Abstract
  3. Introduction
  4. Approaches to understanding consumption behaviours
  5. The study area
  6. Price of water in the study area
  7. Methodology
  8. Understanding stated and actual water use
  9. The prestige of sustainable living
  10. Conclusion
  11. Acknowledgements
  12. References

The study was conducted primarily as a result of the protracted and severe drought and long-term, advanced water restrictions, in what is normally a semi-arid climate. Widespread below average rainfall and inflows in the Murray-Darling River catchment between 2002 and 2009 (the time of this study) meant that the entire river system was under severe stress (Bureau of Meteorology 2010), with conditions worse in South Australia due to its position at the end of the river system. Restrictions commenced in 2003 and as the drought progressed became more restrictive. At the time of the study, between July and October 2009, the metropolitan areas were on restrictions (level 3) that limited the use of water externally (watering gardens, washing boats). The restrictions banned certain water-using activities (such as washing down external paving areas and the use of sprinklers), and limited the duration and timing of water-using activities. The country region, as with most country towns, was subject to less restrictive permanent water conservation measures (PWCM) aimed at long-term ‘sensible water use’ (SA Water 2009). For example, PWCM permitted the use of sprinklers and were more lenient on the duration and timing of water-using activities. Both restrictions and PWCM were mandatory with similar fines for breaches in compliance (SA Water 2007); however, enforcement was infrequent and fines minimal.

The three study areas comprised two suburban areas of metropolitan Adelaide (referred to here as Metro North and Metro East) and a regional town (henceforth referred to as Regional). Regional differs from metropolitan Adelaide in that it has a relatively more plentiful water supply, is reliant on local groundwater rather than the water-deficient Murray River for its water supply, and was subject to PWCM rather than restrictions. Regional is slightly cooler and wetter than Adelaide; the annual rainfall is 775 mm and 529 mm respectively.

The study areas were chosen on the basis of their general socio-demographic profile. In Australia an Index of Relative Socio-economic Disadvantage and Advantage (SEIFA) scores socio-economic advantage and disadvantage, education and resources. It is a measure of an area rather than individuals, although it indicates that populations with similar incomes, education and community resources tend to live in close proximity. Metro North comprises a relatively disadvantaged population with 67% in the top four disadvantaged SEIFA deciles 1–4, Regional householders are at decile 4 and Metro East falls within the most advantaged deciles (9 and 10) in South Australia (Australian Bureau of Statistics 2008).

Price of water in the study area

  1. Top of page
  2. Abstract
  3. Introduction
  4. Approaches to understanding consumption behaviours
  5. The study area
  6. Price of water in the study area
  7. Methodology
  8. Understanding stated and actual water use
  9. The prestige of sustainable living
  10. Conclusion
  11. Acknowledgements
  12. References

In South Australia there is a state-wide water-pricing policy (National Water Commission 2011), which means that residents of the three study areas pay the same amount per volume of water consumed. The operating costs in the regional areas are higher than in Adelaide. Thus, in effect, consumers in Adelaide subsidise prices in the rest of the state. This applies to all areas where the principal state water service provider delivers the service.

Water pricing is stepped with tier one set at AUD$0.97 cents per kl for water use up to 120 kl, tier two at AUD$1.88 per kl for water use in excess of 120 kl and tier three at AUD$2.26 per kl for use above 520 kl in single dwelling residential properties. To offset the staged increases in water pricing over the next four years, service fees were reduced in 2009 and residents are now provided with invoices for their water use on a quarterly rather than biannual basis. A range of payment assistance measures and rebates are also available from the water service provider and aid organisations to assist low-income households who are experiencing utility stress (South Australian Government 2009, 37).

Methodology

  1. Top of page
  2. Abstract
  3. Introduction
  4. Approaches to understanding consumption behaviours
  5. The study area
  6. Price of water in the study area
  7. Methodology
  8. Understanding stated and actual water use
  9. The prestige of sustainable living
  10. Conclusion
  11. Acknowledgements
  12. References

Data collection

Data for this paper were collected via a postal survey and a follow-up telephone interview with the same respondent. Potential candidates for the survey were drawn by the water service provider from their customer database for the study areas and included adults who were responsible for paying the water bill, resided in an individually metered dwelling, had resided at the address for at least a year and none of the household occupants were employed by the water service provider. Every tenth householder eligible for the study was identified for inclusion until a non-stratified sample population of 3000 householders was achieved. Completed numerically identified surveys were returned in a de-identified form to the research team. Survey respondents who agreed to an additional follow-up telephone interview also returned a consent form to the water service provider who then passed the respondents' contact details on to a computer-assisted telephone interviewing (CATI) company tasked with obtaining the additional data from willing survey respondents. While the names and addresses of participants were retained by the service provider, they received no data.

Water meter data were provided to the researchers by the water service provider in a de-identified form for each household that had completed both a postal survey and a telephone interview. Water meter data were obtained as two 3-month blocks from mid 2009 to the end of 2009. The dates covered by each quarter were slightly different between the three study areas, as follows: Metro East from July to September and October to December 2009; Metro North from August to October 2009 and November 2009 to January 2010; and Regional from June to August and September to November 2009.

In total, 539 (18%) usable mail surveys were received: 227 for Metro East, 150 for Metro North and 162 for Regional, although there was some variation across individual questions. In addition, data from 438 completed telephone interviews (CATI) were obtained. Overall, respondents were slightly skewed towards older, single and coupled occupants. The response rate limits the extent to which the results can be generalised and without access to the customer database it was not possible to delve deeper into the representativeness of the sample.

Survey content

The survey comprised 34 questions that asked for information on householders' demographics, attitudes and behaviours relating to water conservation and restrictions. The survey questions mostly used Likert scale responses (seven point) to indicate respondents' agreement with or acceptance of the statements and questions, many of which also comprised a number of sub-questions. Respondents were also given the opportunity to provide comments. The survey content was vetted by personnel at the water service provider and ethics approval to conduct the research was obtained.

Data analysis

Logistic regression analyses were applied to establish a profile of high users compared with households with lower levels of water consumption; further explanatory information on binary logistic regression analysis is outlined in the next section. All analyses were carried out using the statistical software package SPSS version 17. Initially bivariate logistic regression analyses were conducted to examine the relationship between the individual demographic predictors and high annual per capita water use. Only items showing an association at the p = 0.25 level were entered into multiple binary logistic regression analyses (Hosmer and Lemeshow 2000). In the regression analyses the relevant predictor variables were entered in one block rather than stepwise procedures (after Field 2005). Predictor variables that were entered into the model but returned as not significant were in turn tested against models containing only significant predictor variables. This process allowed for the comparison of several models, resulting in a final model for high water use containing only indicators that significantly improved the fit of the model.

As the focus of the analysis was on establishing a behavioural profile of householders with high annual per capita water use, each respondent's household's annual water consumption was divided by the number of individuals cohabiting to yield per capita use. Subsequently, households falling into the top 30% of annual per capita water users were classified as high water users. Selection of a high user group based on a cut-off point of 30% rather than a numerical value for water usage was considered the most suitable option to yield a sufficiently large sub-sample of high water users for which analyses were carried out, and reflect the relative nature of top users to the rest of the sample. It would have been preferable to use a lower percentage of top users (e.g. the top 5% or 10%) or indeed an absolute value of water usage, but this was not possible given the relatively small number of data points available for analysis. Utilisation of the top 30% of annual per capita water users resulted in 142 households from the postal survey (136 CATI interviewees – the CATI respondents thus represent a sub-sample of postal survey respondents) being referred to as high users, while 321 postal survey respondents (293 CATI interviewees) were categorised as not falling into the high water use group. Overall, complete datasets (postal survey, CATI and actual water use) were available for 407 households. As a direct consequence of the low number of returned postal surveys (and subsequently completed CATIs) it was not possible to further divide responses by study area and specify models for high water use in each of the three locations. Instead, area (location) was used as an additional indicator of annual per capita water use. Regression analysis was conducted based on complete observations only; that is, households were included in the analysis if water use as well as socio-demographic and behavioural measures were available for the same household.

Binary logistic regression analysis

Binary logistic regression analysis is a form of regression analysis used to predict a binary (i.e. dichotomous) outcome variable using categorical predictor variables; it yields information about how well predictors explain the variance observed for a specific outcome. Good model fit means that a large proportion of the variance observed between different response categories (e.g. ‘Yes’ or ‘No’) can be explained in reference to the predictor(s) included in the model. In the current paper, a guideline to model fit is given in the form of Nagelkerke R2 values. Preparatory steps conducted before establishing the regression model are useful to test for multi co-linearity between predictor variables in order to avoid redundancy in the model. For example, two variables might be highly correlated, hence explaining a third variable in relation to the two means the overlap between the two predictor variables will explain the same amount of variance in the outcome variable, and therefore weaken the model. It is therefore necessary to carefully examine the variables entered into a regression model, especially if the resulting mode is to yield robust, valid results.

One of the main parameter estimates yielded by logistic regression analysis, which is useful for interpretation of the results, are regression coefficients for individual predictors, and their associated odds ratios, significance levels and confidence intervals. In the current paper, mainly odds ratios are reported; these allow for a simple interpretation of the results in the form of likelihood statements; for example, ‘high users are more likely to live in Metro East than the other two areas’. Odds ratios smaller than 1 indicate that a specific predictor variable level is less likely to fall into a specific category than the designated reference level of the same predictor variable. In contrast, an odds ratio that is larger than 1 indicates an increased likelihood of the particular predictor falling into a specific category relative to the reference level.

Understanding stated and actual water use

  1. Top of page
  2. Abstract
  3. Introduction
  4. Approaches to understanding consumption behaviours
  5. The study area
  6. Price of water in the study area
  7. Methodology
  8. Understanding stated and actual water use
  9. The prestige of sustainable living
  10. Conclusion
  11. Acknowledgements
  12. References

The results are presented in three sections: general water use patterns; demographic and socio-economic characteristics; and consumption and conservation behaviours of householders with high annual per capita water use. Odds ratios (ORs; Exp(B)) are reported instead of constant coefficient (B) values; the former, which is merely an exponentiation of the latter, is easier to interpret than the B value, which is given as log-odds. Moreover, Wald χ2 test statistics are reported, which test the hypothesis that the predictor constant (B) = 0, assuming a 95% confidence level of the results not having occurred by chance (i.e. p-level of 0.05).

General water use patterns

Across households, annual per capita water consumption ranged from 0.4 to 729.0 kl, yielding a sample mean of 95.7 kl (SD = 74.3 kl, n = 463). Mean annual per capita consumption for the top 30% was substantially higher (mean = 177.3 kl or 486 litres/person/day, SD = 85.5, n = 142) than households not falling into the high user category (mean = 59.6 kl or 163 litres/person/day, SD = 22.3, n = 321) (Figure 1). Inferential statistical analyses confirmed these patterns to be statistically significant (U = 0.0, Z = –17.2, p < 0.001): annual per capita water use was significantly higher in the top 30% than for all other consumers. These results not only demonstrate the widespread use of water across households, but also validate the 30% cut-off point creating sub-samples with markedly different consumption patterns.

figure

Figure 1. Bar chart representing the mean annual per capita water use (kilolitres) of high (top 30%) user households compared with all other consumers

Download figure to PowerPoint

Demographic and socio-economic characteristics of householders with high annual per capita water use

Three demographic and socio-economic variables were found to have significant predictive qualities for high annual per capita water use [χ2(13) = 92.44, p < 0.001, Nagelkerke R2 = 0.29], including location [Wald (2) = 15.4, p < 0.001], household size [Wald(4) = 43.9, p < 0.001] and annual household income [Wald(7) = 14.9, p < 0.05]. Parameter estimates for the predictor variables and their respective levels are reported in Table 1. Starting with location, high water users were around 70% less likely to reside in Metro North (OR = 0.32, p < 0.01) and approximately 60% less likely to be a resident in the Regional area (OR = 0.39, p < 0.01) than in Metro East; or put differently, high users were more likely to live in Metro East than the other two areas. Furthermore, high water use was more likely in single person compared to larger households. Relative to respondents living by themselves, high water use was 60% less likely for respondents living with another person (OR = 0.40, p < 0.01), around 90% less likely for households comprising three people (OR = 0.11, p < 0.001), 93% less likely for respondents living in households of four people (OR = 0.07, p < 0.001), and 97% less likely for households comprising five or more individuals (OR = 0.03, p < 0.001). Taken together, the more people who cohabit the less likely the household is to have high annual per capita water use.

Table 1. Fit indices and parameter estimates for householders' demographic and socio-economic characteristics as indicators of high annual per capita water use
Predictor variable (reference level)N (%)aWald (df)pOR95% CI
  1. Note: df = degrees of freedom; OR = odds ratio; CI = confidence interval

  2. a

    Rounding accounts for percentages not adding up to 100.

  3. b

    Annual household income is given in AUD$.

Catchment area (Metro East)174 (42.75)15.42 (2)<0.001
Metro North111 (27.27)11.95 (1)<0.010.320.17–0.61
Regional122 (29.98)9.46 (1)<0.010.390.22–0.71
Household size (single person)63 (15.48)43.90 (4)<0.001
Two persons183 (44.96)7.85 (1)<0.010.400.21–0.76
Three persons67 (16.46)24.23 (1)<0.0010.110.04–0.26
Four persons64 (15.72)25.85 (1)<0.0010.070.03–0.20
Five or more persons30 (7.37)17.05 (1)<0.0010.030.01–0.17
Household incomeb (>$156 000)59 (14.50)14.90 (7)<0.05
Up to $20 80044 (10.81)8.73 (1)<0.010.200.07–0.58
$20 801–31 20038 (9.34)4.64 (1)<0.050.320.11–0.90
$31 201–41 60037 (9.09)0.99 (1)>0.050.590.21–1.66
$41 601–52 00030 (7.37)2.81 (1)>0.050.410.14–1.16
$52 001–78 00080 (19.66)0.48 (1)>0.050.750.32–1.71
$78 001–104 00059 (14.50)5.57 (1)<0.050.320.13–0.83
$104 001–156 00060 (14.74)2.04 (1)>0.050.530.22–1.27

Lower household income levels were associated with a smaller likelihood of consuming large quantities of water. Respondents with an annual income of up to AUD$20,800 were 80% less likely to have high annual water use than respondents with an income over AUD$156,000 (OR = 0.20, p < 0.01). Similar results were observed for households with between AUD$20,801 and 31,200 (OR = 0.32, p < 0.05) and between AUD$78,001 and 104,000 (OR = 0.32, p < 0.05), both of which were around 70% less likely to have high water consumption than households in the highest income bracket. Whilst some 95% confidence intervals were relatively narrow, others showed a larger spread, rendering the parameter obtained open to variation within the given interval (Table 1). A larger sample size is needed to yield more exact parameter estimates.

In summary, the following demographic and socio-economic profile was constructed for households with high annual per capita water use: high water consumption was more likely for households in the Metro East area and significantly less likely for larger households and households with lower levels of annual income. In order to control for the impact of demographic and socio-economic factors on water use, all subsequent analyses included location, household size and income.

Behaviours of householders with high annual per capita water use

Analysed in isolation, several behavioural variables had significant predictive qualities for high water use. However, the majority of these were rendered insignificant once demographic and socio-economic differences were controlled for. The results presented below refer only to the three behavioural predictors that were significant for high water use independently of socio-demographic characteristics, namely: weekly average for watering the garden, watering the garden more often than restrictions allow, and whether conservation happens automatically (Table 2).

Table 2. Odds ratios (ORs), 95% confidence intervals (95% CIs) and significance levels (p) for models of householders' consumption and conservation behaviours returned as significant predictors of high annual per capita water use. Categories in brackets were used as the reference level against which all other predictor levels were compared
 Model 1Model 2Model 3Model 4
OR (95% CI)pOR (95% CI)pOR (95% CI)pOR (95% CI)p
  1. a

    Annual household income is given in AUD$.

Catchment area (East)        
North0.35 (0.18–0.67)<0.010.28 (0.14–0.57)<0.0010.29 (0.14–0.58)<0.010.31 (0.15–0.64)<0.01
Regional0.40 (0.22–0.73)<0.010.43 (0.23–0.81)<0.010.47 (0.25–0.88)<0.050.49 (0.26–0.93)<0.05
Household size (five or more people)        
Single person31.84 (6.25–162.21)<0.00136.41 (7.06–187.93)<0.00157.71 (9.25–360.07)<0.00162.51 (9.63–405.73)<0.001
Two persons11.03 (2.40–50.68)<0.0111.29 (2.43–52.52)<0.0118.49 (3.28–104.29)<0.00115.97 (2.76–92.31)<0.01
Three persons3.27 (0.65–16.52)>0.053.82 (0.75–19.57)>0.056.51 (1.05–40.36)<0.056.94 (1.09–44.33)<0.05
Four persons2.25 (0.43–11.80)>0.052.28 (0.43–12.13)>0.053.65 (0.57–23.30)>0.053.69 (0.56–24.36)>0.05
Annual household income (up to $20 800)a        
$20 801–31 2001.49 (0.48–4.63)>0.052.51 (0.73–8.67)>0.052.84 (0.84–9.60)>0.052.45 (0.70–8.54)>0.05
$31 201–41 6002.89 (0.94–8.80)>0.052.95 (0.89–9.81)>0.052.97 (0.89–9.88)>0.052.49 (0.71–8.54)>0.05
$41 601–52 0002.38 (0.71–7.99)>0.052.25 (0.63–8.05)>0.052.32 (0.64–8.33)>0.052.66 (0.69–10.19)>0.05
$52 001–78 0003.56 (1.35–9.40)<0.054.26 (1.49–12.18)<0.014.34 (1.54–12.22)<0.014.08 (1.40–11.89)<0.05
$78 001–104 0001.52 (0.50–4.67)>0.051.76 (0.52–5.99)>0.051.58 (0.46–5.39)>0.051.44 (0.41–5.11)>0.05
$104 001–156 0002.64 (0.87–8.05)>0.053.04 (0.93–9.95)>0.053.53 (1.08–11.53)<0.053.28 (0.96–11.18)>0.05
>$156 0005.03 (1.67–15.11)<0.016.42 (1.96–21.03)<0.016.77 (2.09–21.94)<0.016.34 (1.89–21.33)<0.01
Weekly garden watering1.24 (1.02–1.52)<0.051.28 (1.02–1.60)<0.05
Water more than restrictions allow (agree)        
Disagree0.43 (0.21–0.87)<0.050.47 (0.22–0.98)<0.05
Neutral0.57 (0.10–3.29)>0.050.58 (0.10–3.44)>0.05
Conserving automatically (neutral)        
Agree0.12 (0.02–0.57)<0.010.12 (0.02–0.60)<0.01
Disagree0.12 (0.02–0.65)<0.050.10 (0.02–0.55)<0.01
Average frequency of garden watering

Under the restrictions garden watering was limited to once per week. When householders were asked how often they watered their gardens, responses ranged from 0 to 10 times per week, yielding a sample mean of 1.59 times (SD = 1.19). As shown in Figure 2, households falling into the top 30% of annual per capita water use reported slightly higher averages (mean = 1.76, SD = 0.09) than households not in the top 30% (mean = 1.51, SD = 0.07); the difference in water use between the two groups was statistically significant (U = 17 633, p < 0.01). After controlling for the influence of demographic factors, the number of times respondents watered their garden (per week) was predictive of per capita water consumption [χ2(14) = 89.45, p < 0.001, Nagelkerke R2 = 0.29]: the more often the garden was watered, the higher the likelihood of high annual per capita water use (OR = 1.24, p < 0.05) (Table 2, model 1). Note that watering the garden more than once per week constitutes a breach of restrictions only if mains water was used. If alternative sources were used it would not constitute a breach, however the results indicate that mains water was being used for the excess watering.

figure

Figure 2. Mean frequency of the number of times in a week that respondents' watered their garden

Download figure to PowerPoint

Compliance with garden watering restrictions

Of householders who responded to the survey item I water the garden more often than restrictions allow’, more than 85% of respondents reported to have complied with restrictions (Figure 3). In contrast, 12.3% agreed with the statement; that is, they admitted to non-compliance with restrictions, and 2.1% were neutral. Whilst the majority of respondents said that they complied with restrictions, 19.1% of the high water users stated that they watered their garden more often than restrictions allowed, but only 8.9% of respondents not in the high water users group agreed with the statement. In a similar vein, 79% of high water users did not water their gardens more often than restrictions allowed, but close to 90% of those not in the top 30% of water users responded the same way. In both groups, roughly 2% of respondents were undecided on the issue. Exceeding restriction levels when watering the garden was indicative of high water use beyond the variation accounted for by socio-demographics [χ2(15) = 97.53, p < 0.001, Nagelkerke R2 = 0.32] (Table 2, model 2). Householders who were adhering to restrictions (as indicated by their disagreement with the statement ‘I water my garden more often than restrictions allow’) were 57% less likely to fall into the high water user group than those who indicated non-compliance with restrictions (OR = 0.43, p < 0.05).

figure

Figure 3. Percentage of responses to the survey item ‘I water my garden more often than restrictions allow’

Download figure to PowerPoint

Conserving water as an ‘automatic’ behaviour

The majority of householders (84%) agreed with the survey item ‘I don't have to think about conserving water in my household because I do it automatically’, 12.6% disagreed and 3.4% were neutral. Of those in the high water users group, 83.7% agreed and 8.9% disagreed with the survey item, whereas 84.0% of those not in the high water users group agreed and 14.3% disagreed with the statement (Figure 4). High water users were neutral on the topic more often than other householders (7.4% and 1.7%, respectively). When used as an indicator of high water use, the automaticity with which conservation takes place held significant predictive qualities [χ2(15) = 100.31, p < 0.001, Nagelkerke R2 = 0.33] (Table 2, model 3). In particular, compared with respondents who were neutral, householders who indicated that conservation happened automatically or did not happen automatically were close to 90% less likely to be high water users (OR = 0.12, p < 0.01, and OR = 0.12, p < 0.05, respectively). These patterns suggest that it may be the awareness of personal behaviours as such that determines water use patterns, more so than whether conservation requires thought or happens automatically. That is, respondents who agreed that water conservation happened automatically might be conserving water; however, those who disagreed with the statement might also be conserving water but in their case they have to think about their behaviours and perhaps remind themselves to turn the tap off while cleaning their teeth (it does not happen automatically). Those respondents who were neutral (that is, did not have sufficient awareness of whether or how their own conservation practices occurred) were more likely to be high water users.

figure

Figure 4. Percentage of responses to the survey item ‘I don't have to think about conserving water in my household because I do it automatically’

Download figure to PowerPoint

Demographic and behavioural predictors of high water use

A final model, for which the three demographic (location, household size, annual household income) and behavioural predictors were entered, returned five of the six predictors as significant for high water usage [χ2(18) = 100.95, p < 0.001, Nagelkerke R2 = 0.34] (Table 2, model 4). Compliance with restrictions was rendered insignificant once the average frequency of garden watering was entered into the model [χ2(2) = 4.08 p > 0.05]. This suggests that both variables explain the same variation, and that demographic indicators, average watering frequency [χ2(1) = 4.44, p < 0.05] and the conservation automaticity [χ2(2) = 7.23, p < 0.05] are sufficient to explain more than a third of the variation observed for high water use.

The prestige of sustainable living

  1. Top of page
  2. Abstract
  3. Introduction
  4. Approaches to understanding consumption behaviours
  5. The study area
  6. Price of water in the study area
  7. Methodology
  8. Understanding stated and actual water use
  9. The prestige of sustainable living
  10. Conclusion
  11. Acknowledgements
  12. References

This study found that a number of demographic and socio-economic factors had significant predictive qualities for high annual per capita water use, namely household size, location and annual household income; these factors will be discussed first, followed by a discussion of the behavioural predictors of high water use.

While households with a greater number of occupants consumed more water per household than households with fewer occupants, when water use was analysed on a per capita basis, single people, and small households used more water. Similarly, a study of Sydney, the Blue Mountains and Illawarra region showed that the highest domestic water use occurred in single-person households, followed by couples then families (Everett et al. 2004). Water-based evaporative coolers are used extensively in Australia for domestic temperature control (Australian Greenhouse Office 2001), which may partly account for the high per capita water use. Evaporative coolers have low energy requirements compared with refrigerative air conditioners, which make them a cost-effective choice, but they are profligate water users (Karpiscak et al. 1998). Garden watering and evaporative coolers consume an amount of water independent of the number of people living on the property, thus when water use is analysed on a per capita basis, households with fewer occupants tend to have higher water use than larger households. In contrast, the relationship between household size and per capita water consumption exhibits an ‘economies of scale’ in larger households. Barrett and Wallace (2009), Beal et al. (2008) and Everett et al. (2004) report similar findings. As the trend in housing in Australia moves towards a greater predominance of single- or two-person households (Australian Bureau of Statistics 2008) this will have implications for water demand if concomitant water-related efficiencies in housing developments do not occur. This highlights the need for water saving technologies and water-sensitive urban design to become common practice or legislated in the building industry – a move which has begun in various states in Australia.

When examining water use behaviours, De Oliver (1999) found strong spatial correspondence in relation to water consumption and characteristics such as income, education, political affiliation and ethnicity. Similarly, our study found that location and income were strong predictors of high water consumption. High water use has been linked to a greater number of water-using amenities, types of appliances (Barrett and Wallace 2009; Everett et al. 2004; Harlan et al. 2009; Troy and Randolph 2006), swimming pools and spas, the size of gardens (Domene et al. 2005), and the sophistication of garden-watering systems (Loh and Coghlan 2003). Occupants of high water-using households comprising large homes and gardens have the greatest capacity to consume water, but likewise have the greatest capacity to conserve water (Campbell et al. 2004), and it may be argued that they are also most able to afford water-saving technologies. A lack of motivation to use less water or water-related lifestyle interests (such as gardening) may, however, prevent this from happening.

Our results indicate that people had an accurate idea of the magnitude of their own water use and how it compared with other households. The fact that respondents in this study were so forthcoming about their excess water-using behaviours is noteworthy as numerous studies (cited in De Oliver 1999) found a tendency for people to provide socially acceptable responses in surveys. Participants in the latter study saw themselves as more frugal in their water use than others.

This study found a link between high water use and excess garden watering. Usually, during non-drought periods, over half of Adelaide's domestic water use occurs externally. Although external water use did decline with the onset of mandatory restrictions, the fact that external water use remained high shows the importance that some residents place on maintaining a healthy garden (Allon and Sofoulis 2006; Domene et al. 2005; Harlan et al. 2009; Saurií 2003). Results from Syme et al. (2004) showed that people with a higher interest in their garden tended to use more water externally. The link between lifestyle and water use is captured in the comments by Allon and Sofoulis (2006, 47) that water use is ‘entangled with users' habitual enjoyment of the services, technologies and experiences that water makes possible’ and by Head and Muir (2007, 902) that ‘aspirations towards water conservation are in tension with the pleasure derived from water’. Given the importance of gardens as a symbol of social distinction, in the latter study, the higher income earners did not consider cost when choosing their style of garden or the amount of water used when irrigating. In this study, Metro East, a stately established suburb of Adelaide, is regarded as a prestigious location. As this study implies, and numerous studies have shown (Martin et al. 2004; Kahana et al. 2003; Spiniti et al. 2004 all cited in Yabiku et al. 2008), the maintenance of property value and the importance of a healthy garden as a symbol of economic status within a neighbourhood is more important than conserving water. Domene et al. (2005, 532) highlight the importance of aesthetically pleasing gardens as a means of prestige and gaining social distinction and thus caution against raising the price of water as a means of water demand management because for this sector of the population it could further entrench the symbolism of lush gardens as a sign of wealth. Thus, for water demand strategies to be effective they will need to consider the importance of a gardening lifestyle among certain sectors of the population; in Australia gardening is a popular pastime. The broader social and financial costs of water restrictions should thus not be underestimated. During the drought in New South Wales some residents who did not want to breach restrictions but had a keen interest in maintaining their garden purchased rainwater tanks for that sole purpose. In doing so, they abided by the restrictions (by not using excess mains water on the garden), but the cost of storing the rainwater far exceeded that of using mains water, yet it was a price deemed worthy by the residents. While this behaviour maintains high water-using habits that are counter to the conservation objective of restrictions, it highlights the amenity or aesthetic value that some people place on healthy gardens regardless of the cost (Moy 2012). Similarly, in Adelaide, when restrictions became increasingly severe, the number of applications to drill domestic bores (an expensive option) increased sixfold (Verity 2011), again highlighting the price that people place on maintaining their gardens. Lower socio-economic groups with an interest in maintaining a healthy garden, however, may not be able to afford the extravagant garden–watering alternatives mentioned above, and for those groups the welfare costs associated with restrictions (Brennan et al. 2007; Roibás et al. 2007; Syme et al. 2004) will be more sorely felt.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Approaches to understanding consumption behaviours
  5. The study area
  6. Price of water in the study area
  7. Methodology
  8. Understanding stated and actual water use
  9. The prestige of sustainable living
  10. Conclusion
  11. Acknowledgements
  12. References

Spaargaren (2003) calls on international researchers to contribute to the sustainable consumption debate, in particular those offering non-European case studies. Our findings show a link between high water use and the ‘social practice’ of excessive garden watering in South Australia and thus are in keeping with the social practices approach (Shove 2003; Spaargaren 2003). In our study high water users, despite being aware of the water scarcity and restrictions, knew that they were high users and yet they did not alter their water-consuming behaviours. The willing breaches of restrictions among high-income households may mean that demand management strategies such as increasing the price of water or conservation awareness campaigns would have little effect in these households. Thus, although targeting high water users may have the potential to reap the greatest water savings, they may be the most challenging group in which to achieve a reduction in water use (because of the importance they place on their water-using practices and their ability to pay for the high water use).

Some lessons from previous case studies may apply to sectors of the population who are not only high water users, but knowingly continue their profligate behaviours. First, an approach similar to that outlined by Walton and Hume (2011) could be applied in that only one behaviour was targeted. Although the four-minute shower was targeted in the latter study, of relevance to this paper, profligate garden watering could be targeted. Second, the most cost-effective water use reductions may be achieved if the behaviour-targeted campaigns are also location specific; that is, high water use locations can be targeted rather than a general, large-scale approach where many households may already be frugal in their water use. Further, for behavioural change campaigns to be effective they need to personalise the impacts of not doing the water conservation behaviour (Gilg and Barr 2006) and offer alternatives that do not lead to any loss in social welfare or status. Opulent and aesthetically pleasing examples of native species landscapes could be offered. Marketing attractive alternative garden styles to high water users may influence behavioural change on two levels: first, through a reduction in the anticipated negative consequences associated with watering the garden less often; and second, through offering consumers an alternative behaviour; that is, not in conflict with their lifestyle desires or the prestige value they place on their garden. High water users need to be able to see the net increase in their social welfare for any reduction in water use to be sustainable. There is growing prestige associated with sustainable living (cf. Oliver and Lee 2010; Palm and Tengvard 2011), thus marketing the prestige value of symbols of sustainable consumption (such as native gardens) should form the core of campaigns targeting profligate resource users.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Approaches to understanding consumption behaviours
  5. The study area
  6. Price of water in the study area
  7. Methodology
  8. Understanding stated and actual water use
  9. The prestige of sustainable living
  10. Conclusion
  11. Acknowledgements
  12. References

The authors acknowledge The Australian Research Council Linkage Program and SA Water in funding the research. We also acknowledge the participants in the research and the following individuals: Kelly Westell, Kristen Pellew (SA Water), Carmel McCarthy, and Pawel Skuza (Flinders University). The authors would like to acknowledge the contribution made by the referees of this paper.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Approaches to understanding consumption behaviours
  5. The study area
  6. Price of water in the study area
  7. Methodology
  8. Understanding stated and actual water use
  9. The prestige of sustainable living
  10. Conclusion
  11. Acknowledgements
  12. References
  • Aisbett E and Steinhauser R 2011 Maintaining the common pool: voluntary water conservation in response to increasing scarcity. Crawford School Research Paper 11 Crawford School of Economics and Government, Australian National University, Canberra 18 118 (http://ngsm.anu.edu.au) Accessed August 2012
  • Ajzen I 1991 The theory of planned behaviour Organizational Behavior and Human Decision Processes 50 179211
  • Allon F and Sofoulis Z 2006 Everyday water: cultures in transition Australian Geographer 37 4555
  • Alston M and Mason R 2008 Who turns the taps off? introducing social flow to the Australian water debate Rural Society 18 131140
  • Australian Bureau of Statistics 2008 Year book Australia 2008 1301.0 Canberra (www.abs.gov.au/AUSTATS/) Accessed August 2010
  • Australian Greenhouse Office 2001 Analysis of potential for minimum energy performance standards for evaporative air conditioners National Appliance and Equipment Energy Efficiency Program Report Australian Greenhouse Office (www.energyrating.gov.au/library/pubs/tech-evapac2001.pdf) Accessed 12 October 2006
  • Balchin W G V 1958 A water use survey The Geographical Journal 124 476482
  • Barrett G and Wallace M 2009 Characteristics of Australian urban residential water users: implications for water demand management and whole of the system water accounting framework Water Policy 11 413426
  • Baumann D D 1990 Water conservation issues: introduction to water supply and conservation planning Water Resources Update 83 918
  • Beal C, Stewart R A, Spinks A and Fielding K 2008 Believed versus actual household water use: disaggregating end uses to inform water conservation attitudes in Queensland Australia (www.efficient2011.com/technical/paper/234.pdf) Accessed May 2011
  • Berenguer J, Corraliza J A and Martin R 2005 Rural–urban differences in environmental concern, attitudes and actions Eur J Psychol Assess 21 128138
  • Boland J J and Baumann D 2009 Water resources planning: past, present and future in Russell C S and Baumann D D eds The evolution of water resource planning and decision making Edward Elgar, Cheltenham 114
  • Brennan D, Tapsuwan S and Ingram G 2007 The welfare costs of urban outdoor water restrictions The Australian Journal of Agricultural and Resource Economics 51 243261
  • Bureau of Meteorology 2010 Annual climate summary for South Australia (www.bom.gov.au/climate/current/annual/sa/summary.shtml) Accessed September 2010
  • Campbell H E, Johnson R M and Larson H E 2004 Prices, devices, people, or rules: the relative effectiveness of policy instruments in water conservation Review Policy Research 21 637662
  • De Oliver M 1999 Attitudes and inaction: a case study of the manifest demographics of urban water conservation Environment and Behavior 31 372394
  • Domene E, Saurií D and Parés M 2005 Urbanization and sustainable resource use: the case of garden watering in the metropolitan region of Barcelona Urban Geography 26 520535
  • Everett B, Kemp A and Napper K 2004 Residential water use in Sydney, the Blue Mountains and Ilawarra Independent Pricing and Regulatory Tribunal of New South Wales Research Paper RP26 Sydney (www.ipart.nsw.gov.au/files/RP26.pdf) Accessed May 2012
  • Field A 2005 Discovering statistics using SPSS Sage Publications, London
  • Giddens A 1984 The constitution of society Polity, Cambridge
  • Gilg A and Barr S 2006 Behavioral attitudes towards water saving? evidence from a study of environmental actions Ecological Economics 57 400414
  • Gordon-Spencer P, Cotton C J N, Hansford-Miller F H and Rees J A 1973 The demand for water in South-East England: discussion The Geographical Journal 139 3642
  • Gregory G D and Di Leo M 2003 Repeated behaviour and environmental psychology: the role of personal involvement and habit formation in explaining water consumption Journal of Applied Social Psychology 33 12611266
  • Harlan S L, Yabiku S T, Larsen L and Brazel A J 2009 Household water consumption in an arid city: affluence, affordance, and attitudes society and natural resources International Journal 22 691709
  • Head L and Muir P 2007 Changing cultures of water in eastern Australian backyard gardens Social and Cultural Geography 8 889905
  • Hosmer D W and Lemeshow S 2000 Applied logistic regression John Wiley and Sons, New Jersey
  • Inman D and Jeffrey P 2006 A review of residential demand-side management tool performance and influences on implementation effectiveness Urban Water Journal 3 127143
  • Jackson T 2004 Negotiating sustainable consumption: a review of the consumption debate and its policy implications Energy and Environment 15 10271051
  • Kahana E, Lovegreen L, Kahana B and Kahana M 2003 Person, environment, and person–environment fit as influences on residential satisfaction of elders Environment and Behavior 35 434453
  • Karpiscak M M, Babcock T M, France G W, Zauderer J, Hopf S B and Foster K E 1998 Evaporative cooler water use in Phoenix Journal of the American Water Works Association 90 121130
  • Kose H, Sakosuke S and Iizuka H 2004 Measurement of water, electronic power and gas consumption and questionnaire about act of water usage in the apartment house Proceedings of the 30th CIBW062 International Symposium on Water Supply and Drainage for Buildings 15–17 September Centre Scientifique et Technique du Batiment, Paris
  • Leflaive X 2008 Alternative ways of providing water emerging options and their policy implications OECD, Paris
  • Loh M and Coghlan P 2003 Domestic water use study: in Perth, Western Australia 1998–2001 Water Corporation Perth, Australia
  • Martin C, Warren P and Kinzig A 2004 Neighborhood socioeconomic status is a useful predictor of perennial landscape vegetation in small parks and surrounding residential neighborhoods in Phoenix, Arizona Landscape and Urban Planning 69 355368
  • Moy C 2012 Rainwater tank households: water savers or water users? Geographical Research 50 204216
  • Mui K W, Wong L T and Law L Y 2007 Domestic water consumption benchmark development Hong Kong Building Services Engineering Research and Technology 28 329335
  • National Water Commission 2011 Review of pricing reform in the australian water sector National Water Commission, Canberra
  • OECD 2008 Household behaviour and the environment: reviewing the evidence OECD, Paris
  • OECD 2009 Managing water for all: an OECD perspective on pricing and financing OECD, Paris
  • Oliver J D and Lee S-H 2010 Hybrid car purchase intentions: a cross-cultural analysis Journal of Consumer Marketing 27 96103
  • Olmstead S and Stavins O 2008 Comparing price and non-price approaches to urban water conservation. Fomdazione Eni Enrico Matter 225 (www.feem.it/Feem/Pub/Publications/WPapers/default.htm) Accessed June 2011
  • Palm J and Tengvard M 2011 Motives for and barriers to household adoption of small-scale production of electricity: examples from Sweden Sustainability: Science Practice and Policy 7 615 (www.google.com.au/archives/vol7iss1/1001-006.palm.html) Accessed September 2012
  • Pearce M, Willis E, Mamerow L, Jorgensen B and Martin J 2012 A rural-urban divide? Attitudinal differences towards water restrictions in South Australia Australasian Journal of Regional Studies 18 390417
  • Pumphrey R G, Edwards J A and Becker K G 2008 Urban and rural attitudes towards municipal water controls: a study of a semi-arid region with limited water supplies Ecological Economics 65 112
  • Roibás D, Garcia-Valinas M A and Wall A 2007 Measuring welfare losses from interruption and pricing as responses to water shortages: an application to the case of Seville Environmental and Resource Economics 38 231243
  • Russell S and Fielding K 2010 Water demand management research: a psychological perspective Water Resources Research 46 W05302 doi:10.1029/2009WR008408
  • Saurií D 2003 Lights and shadows of urban water demand management: the case of the metropolitan region of Barcelona European Planning studies 11 229243
  • SA Water 2007 Change to enforcement process for restriction breaches Media release 15 October (www.sawater.com.au/SAWater/WhatsNew/NewsRoom/Media+Releases+Archive.htm) Accessed 11 August 2010
  • SA Water 2009 In detail: permanent water conservation measures (www.sawater.com.au) Accessed August 2010
  • Shove E 2003 Converging conventions of comfort, cleanliness and convenience Journal of Consumer Policy 26 395418
  • Shove E and Walker G 2010 Governing transitions in the sustainability of everyday life Research Policy 39 471476
  • South Australian Government 2009 Transparency statement part A: water and wastewater prices in metropolitan and regional South Australia 2009–10 South Australian Government, Adelaide
  • Spaargaren G 2003 Sustainable consumption: a theoretical and environmental policy perspective Society and Natural Resources 16 687701
  • Spinti J, St Hilaire R and Van Leeuwen D 2004 Balancing landscape preferences and water conservation in a desert community HortTechnology 14 7277
  • Syme G J, Nancarrow B E and Seligman C 2000 The evaluation of information campaigns to promote voluntary household water conservation Evaluation Review 24 539578
  • Syme G J, Shao Q, Po M and Campbell E 2004 Predicting and understanding home garden water use Landscape and Urban Planning 68 121128
  • Troy P and Randolph B 2006 Water consumption and the built environment: a social and behavioral analysis NSW Environmental Trust and CRES paper no 5 Cities Future Research Centre, Sydney
  • Verity A 2011 Attitudes towards prescription of groundwater in the Adelaide metropolitan region Unpublished Honours thesis, School of the Environment, Flinders University
  • Walton A and Hume M 2011 Creating positive habits in water conservation: the case of the Queensland Water Commission and the Target 140 campaign International Journal of Nonprofit and Voluntary Sector Marketing 16 215224
  • Yabiku S T, Casagrande D G and Farley-Metzger E 2008 Preferences for landscape choice in a Southwestern desert city Environmental Behavior 40 382400