The assumed motives of opponents to infrastructure facilities: the planners’ perspective or fair distribution of risk? Maarten Wolsink Jeroen Devilee
AESOP conference Napoli, Italy 11-14 July 2007
This paper is a first draft of the following publication (please cite as follows): M Wolsink, J Devilee (2009) The motives for accepting or rejecting waste infrastructure facilities. Shifting the focus from the planners' perspective to fairness and community commitment. Journal of Environmental Planning and Management 52 (2), 217-236
Department of Geography and Planning, University of Amsterdam, Nieuwe Prinsengracht 130, NL-1018 VZ Amsterdam, The Netherlands. phone +31 20 525 6229 E-mail:
[email protected] Jeroen Devilee is currently at: Social and Cultural Planning Agency Parnassusplein 5 2511 VX The Hague, The Netherlands Keywords: Risk perceptions; Planners’ perspective; Backyard; Waste management; Infrastructure; Facility siting
1
The assumed motives of opponents to infrastructure facilities: the planners’ perspective or fair distribution of risk? Maarten Wolsink Jeroen Devilee Summary In environmental planning, decision-making on land use and public policy on space making increasingly causes conflicts. Risk management and perceptions have become crucial and the focus in siting theory has now shifted to the balance of local concerns of host communities with the nationwide concerns of the9333 users of the facilities. In environmental psychology, geography and risk research clear advancements are made by empirically investigating these conflicts. In planning and policy, the planners' perspective remains dominant. This rather one-dimensional approach of the resistance to facility siting is less founded in empirical evidence and it includes easy assumptions about the motives of opponents: residents merely 'protecting their 'turf'', exclusively focusing on their backyard. Unfounded assumptions also concern the locals' perceptions of the facility itself or the utility of what the investors supposed to be a public good. An empirical study is presented on risk perceptions, with a large-scale survey in six decision processes on different waste facilities. A scale is developed to measure the planners' perspective of motives for opposition. The analysis of the "backyard motives" data shows that the crucial factors in perceived risk perceptions are not any personality trait (e.g. egotism, ignorance, irrationality). Crucial factors appear to be perceived environmental injustice, procedural fairness, and personal commitment to community values. The inability of current planning to deal with these phenomena is a clear example of the changed context of decision-making in the risk society. Introduction In environmental planning, decision-making on land use is increasingly controversial. Public policy on space making is related to the social attribution of risks and therefore it often leads to conflicts. Public facility location theory originally focused on the question where to locate public facilities, given efficiency and equity concerns (DeVerteuil, 2000), but the focus has shifted to how public facility patterns can be configured in order to balance the local concerns of host communities with the wider concerns of the investors and users of these public facilities. To counteract frequently emerging social resistance toward siting facilities several strategies and policy lines have been propagated, but most of the strategies and planning policies put forward are not without their pitfalls, mainly because there is often no real understanding of social acceptance and resistance to new facilities (e.g. Wüstenhagen et al. 2007). The analysis, the proposed strategies, and the implemented policies are often based on unilateral approaches. This paper starts by presenting a brief discussion on the nature of the problem: Is it an issue of land use in spatial or economic terms, or is it largely a problem of policy and governance? All planning approaches seem to have one thing in common: They make very strong assumptions about the behaviour, motives, and preferences of the people involved in location conflicts, by reducing the motives of opponents to simple protection of ‘the backyard’. This simplistic categorization by using labels such as “nimby” (“not-in-myback-yard”) of public opposition is now widely discredited (Snary, 2004; Wolsink 2006; Schively 2007) but references to this typecasting suggesting that it is a prominent explanation of opposition to siting facilities keep being published frequently, usually without any adequate investigation of the motives of the people involved (recent examples: Deng, 2003; 2
Kaldellis, 2005; Lehr and Inhaber, 2003; Caplan et al. 2007; Gamboa and Munda 2007). Although different approaches to get to the core of the problem have been employed, the original approach to the siting problem is an economist analysis (O’Hare 1977; the catchy 1 NIMBY-label had not yet been invented ). In this approach, the issue is the divergence between cost and benefits at the various levels. The benefits for people confronted with a facility are usually small, as these are defined as a public good, whereas the costs have to be carried by the people in the community hosting the facility. However, utility is rather unidimensional, whereas local and individual cost are complex. They generally become manifest as uncertainties and perceived annoyance and risk. Simple cost-benefit approaches result in compensation strategies (Groothuis et al., 1998; Caplan et al. 2007), or in more or less sophisticated auctions in which communities may “volunteer” (Inhaber, 1991; Quah and Tan, 1998). The policy option of compensation does not necessarily have to be financial. Van Baren (2001) reports about a connection of a community with a highway, and such provisions can be offered to make a community ‘volunteer’ for a facility. A different approach to the problem is based on the concept of ‘transaction costs’ and financial risk. Interests of residents are defined mainly as their investments in their house and land use (settlement, ownership, greenery, renovations and improvements) and their primary motive is to ‘protect their turf’ (Dear 1992). Resistance is a rational response to uninsured risk and Fischel (2001) even suggests the creation an insurance market for it. The investments in house ownership, which are fairly inflexible, are threatened by decisions that create uncertainty about the future, Residents hold property rights in goods that derive their value from the zoning of their surroundings, something the owners cannot control or only at a great cost, because it is a public decision-making process. Since housing, in particularly ownership is inflexible, bounded to a location, and within a spatially fixed context, the new developments create uncertainty the only possible response is to resist the developments that create the risk (Deng 2003). Although the economist approach continues to label the problem in terms of space, property, and land use (“my backyard”), the cost-benefit approach does not consider the geographical and timing aspects of the problem. Strategies using incentives to change the cost-benefit balance may easily fail (Kearney and Smith, 1995). Essentially, the economic analysis reduces resistance in location conflicts to a situational determined response to the threat of ‘vested interests’. This negative reaction may be expressed in a protectionist attitude toward perceived risky technologies (Marks and von Winterfeld, 1984) or toward people who are perceived as dangerous (Takahashi, 1997; Wolsink 2006). Nevertheless, in economic terms the core of the problem is a response that is based on the rationality of the individual utility character: Opposition to facility siting is a reaction driven by selfish motives (Dear, 1992; Kraft and Cleary, 1991). This is a very strong assumption, and highly questionable because seldom there is any evidence of selfishness provided and it disqualifies all opposition. Aim of this study The strong selfishness-assumption has already been criticized more than 15 years ago (Freudenberg and Steinsapir 1991; Freudenburg and Pastor 1992) but in planning the easy label is still used without any reference to its dubious status. The dubious status of the nimby ‘concept’ and the presumptuous persistence of its application in the planners’ perspective have now been illustrated in many ways and this discussion will not be repeated here (McAvoy 1999; Birmingham 2000; Snary 2004; Wolsink 2006; Schively 2007). This paper aims at empirically investigating the motives for resistance of residents in such a way, that some basic assumptions in the planners’ perspective can be tested and nuanced. The criticism on those assumptions is twofold:: 3
Criticism focused on assumptions about local residents. An important argument for shifting decision-making power to national and regional authorities is that citizens (and municipalities) are assumed to act in the same selfish manner in the sense of economic rationality. This has often been empirically falsified and the significance is revealed of investing in good relations with the public by having an open minded for other motivations, for example on human service facilities (Piat, 2000; Zippay 2007) and renewable energy (Wolsink, 2007a&b). Almost all reports on facility siting mention the large differences in the reactions of local residents, based on a wide variety of motives. Furthermore, unfounded assumptions may concern the locals’ perceptions of the facility itself or of the public good the facility is to serve. In the planners’ perspective there is no doubt of the utility of what is supposed to be a public good, and the technocratic assumptions of investors and authorities often also concern the ignorance and irrationality of the public (Brown, 1987; Inhaber, 1991; Maillebouis, 2003). Criticism concerning the structure and administrative handling of decision-making processes. In the current systems of physical planning, other stakeholders are seldom given a proper role in the decision-making process. The planners’ perspective still has a technocratic inclination and it does not legitimately address local groups and local residents’ risk perceptions (McAvoy, 1999). In decisions that attribute and distribute risk among local communities, the lack of trust in responsible agents and authorities is always at stake (Slovic et al., 1991). Risk research and psychologal risk perception research has often revealed the evidence for this in facility siting (Greenberg and Williams 1999). Opposition may emerge from such factors as perceived inequity and unfairness (Adger, 2002; Capek, 1993; Lake, 1996; Lober, 1995) whereas the most fundamental critique on the attribution of backyard motives to people questioning the risks that are imposed on them according to the perspective of planning authorities also concerns the issues of injustice, equity, and fairness (Wolsink 2006, 2007a; Hermansson 2007). It is also important to notice that the fairness issue goes far beyond the application of formal mandatory participation, that is restricted in most planning systems to formal consultation and appeal following the real planning process. In general, it is not even about the techniques of how to apply participation, but on how to achieve a process that will be perceived as fair (Smith and MacDonough 2001). Premise of the study The premise of the current research was that individual reactions to facility siting exhibit a large variation in both motives and attitudes. The ‘concept’ of NIMBY motives would only have some explanatory meaning in case “the underlying activity is widely supported while the facility to do it is opposed locally” (Lober 1995; Sjöberg and Drottz-Sjöberg 2001). Obviously, attitudes may sometimes differ significantly from behaviour. As the NIMBY logic implicitly interprets this difference by attributing very specific motives to it, we decided to develop an indicator for these motives. Hence, the aim of this research was to translate the planners’ perspective assumptions about the existence of backyard motives into testable hypothesis. The basic hypothesis is: H1: It should be possible to create a consistent scale (and reliable in methodological terms) to measure the rejection of a facility based on motives that refer to one’s backyard. In line with the referred critical elaborations of the NIMBY as a valid concept, two more hypothesis are formulated. Although there may seem to be explanatory power in the developed ‘backyard-scale’ its significance remains limited compared to other factors. These primarily concern attitudes to and perceptions of risk, and about the way such risk are 4
distributed in society. All literature on facility siting reveals the crucial role of risk perceptions (Boholm, 2004) whereas in the ‘risk society’ the distribution of such ‘bads’ as risks is the essence of political decision making (Beck 1992, Ch.1). If the claim that backyard motives a crucial in opposition to new infrastructure facilities, (because the planners’ perspective is that existence of such facilities as such is supported) these motives should provide extra explanatory power to intentions to resist beside other atitudinal factors. H2a: The scale measuring the tendency to reject facilities based on ‘backyard’ motives will be related to negative attributes, in particular perceived risks, but beyond that it adds extra explanatory power to behavioural intentions. However, as most critics of the use of the NIMBY-label say, the distribution of risk within society is what really counts, and in the minds of public that is confronted with a proposed decision this mainly becomes manifest in the way such distribution is perceived as ‘fair’. In our research we also investigated the processes in all cases, so we can formulate following hypothesis: H2b: The extra explanatory power is conditional: it will be dependent on conditions related to the decision-making process. The fairness has two component: distributional fairness (related to equity) and procedural fairness. These two can be separated analytically, but in the practice of public perception fairness they are strongly connected. Furthermore, the term ‘procedural’ suggest that the fairness concerns formal institutions such as mandatory consultation procedures, but the perceived fairness rather concerns the actual process in which decisions are taken than the legal and technical rules that form the procedure (Smith and McDonough 2001). This second sub-hypothesis concerns the limited significance of backyard-motives. We expect the impact of backyard motives to be dependent on such factors as the type of facility, the character of the decision-making process, and the phase in that process. Method The focus of this research was on the assumptions about local residents’ attitudes and behaviour, in particular the development of a scale that measures references to the backyard for resitiance towards a proposed facility. This research focused on waste facilities, because in the Netherlands so called speed-up legislation has been implemented and changes to the planning systems are proposed to overcome resistance to infrastructure facilities (Wolsink, 2003). This speed-up legislation was implemented with a strong reference to backyards motivations as well as waste infrastructure (deJong and Wolsink, 1997). As only waste management facilities were included in the study, the variation in type of facilities remained limited to different types of waste management and to cases where planned facilities were built versus cases where the proposals had been cancelled. This also implied that meant the construction of a reliable scale to measure an individual’s inclination to motivate his/her resistance was framed in terms of waste management, but apart from that our questionnaire was designed according to social psychological attitude-behaviour modelling that is widely considered valid (Ajzen, 1991, 2001). As noted, the term NIMBY contains normative or implicit interpretations about motives for opposition, and because of that “theoretical” distinction it must also be empirically separated from behavioural and attitudinal opposition. Definitions of NIMBY usually remain implicit and neither much literature concerning experience with indicators of thebackyard motivation exist (Wolsink 2006). Hence, first the concept of the situational and motivational context of the phenomenon was described to use in a pilot study, aiming at the development of a set of reliable items. In the pilot study, 137 respondents with a balanced variation of age, gender, and level of education, answered 3*17 5
items for three different types of waste management facilities: a landfill, a composting plant, and an incinerator. For the main research project, a large-scale field survey, the respondents were persons living in the vicinity of a planned waste facility. As the research was aimed at establishing the differences between different types of decision-making (Van Baren, 2001; Wolsink, 2004), and particularly at the impact on attitudes and behaviour of waste management types and planning phase, the single case study that is common in the literature on facility siting had to be avoided. Six cases were selected from a pool of 103 waste projects in the Netherlands that were large enough to require a mandatory environmental impact report. Selection was based on criteria as size, type of facility, and duration of the decision-making process. Three processes that lasted several years but did not come to fruition (the planned facilities were cancelled) were selected, and three cases resulted in the building of a facility, also after a process of several years. The six cases concerned a variety of types of waste disposal (landfilling, incineration) and waste treatment (recovery, composting), as well as some combinations of waste management. In one case (Table 1), the planned composting facility but not the incinerator was built, and in another case the composting plant was an extension of an existing composting plant and landfill. Table 1. The six selected cases Case
C: activity cancelled B: facility built
Subsample N
Zwolle
C landfill, recovery, composting
298
Nieuwegein
C landfill
281
Leiden
C processing and fermentation plant
259
Landgraaf
B landfill
251
Duiven
B composting plant (C incinerator)
259
Wijster
B extension composting plant
254
(Control) Gouda No waste management facility
253
(Total 1855)
For all cases, a random sample was taken near the proposed facility (maximum radius 5 kilometres). Zip codes were used to compare the samples with population figures resulting from geographic information system (GIS) data. The comparison revealed no significant sample bias impact on our study (Devilee, 2002). Data resulting from the control case were used in the scale analysis. Gouda is an average-sized city situated in an environment that is neither very urban nor very rural, and during the period under study, there was no planning process concerning any waste facility to be sited in the city. First the consistency of the developed backyard-motivation scale is tested, and then its application is presented. We examine the relevancy of the nimby concept by taking into account other important concepts, such as perceived risk and perceived negative consequences. This examination is based on structural equation analyses (LISREL; Jöreskog and Sörbom, 1996), in which we use an attitude-behaviour model developed to investigate the explanations for opposition to and support for waste facility siting. This model is only applied here, as it is described elsewhere (Devilee, 2002). As formulated in the hypothesis the model primarily focuses on intentions, although we included participation in oppositional activities. Because of the invariance among a large part of the population, in most empirical research active political participation is usually less correlated to all research variables than variables indicating psychological involvement (Cohen et al., 2001). 6
The full model includes a large number of manifest variables and the deletion of too many respondents because of occasional missing data had to be avoided. A method was applied to include the maximum number of respondents who delivered substantial information in the analysis, without smoothing down the structural differences between the sub-samples. The treatment of missing data was the method of corrected item means (CIM), in which a missing value on an item is imputed by the sub-sample mean corrected for the other scores of the respondent (Huisman, 2000) thus leaving structural differences between sub-samples intact. The reliability of the scales (including the backyard-scale that is introduced in this paper) was tested before the imputation, to prevent the reliability to reflect the imputation process (Huisman, 2000). First result: backyard motives measured Basic principles for item construction The NIMBY phenomenon needs to be distinguished from attitudes toward a facility and from oppositional behaviour. It’s basic position regarding behaviour, intention and attitudes is shown in Fig. 1. The concept is the inclination toward oppositional behaviour based on backyard motives, whereas the attitude itself is not causing that behaviour. This inclination originates from specific motivational concerns that reflect individual costs versus public goods. Nimby inclination based on “backyard motives”
2 3 Negative attributes of attitudes towards the facility
1
Intention to oppose
4
Oppositional behaviour
Figure 1. The conceptual position of the NIMBY inclination Measuring the inclination by merely using indicators of behaviour or attitudes alone (Groothuis and Miller, 1994) is meaningless (Sjöber end Drottz-Sjöberg 2001). NIMBY would then simply be synonymous with an unfavourable evaluation without any indication of the motives (relation 1, Fig. 1). Nor can the inclination be measured indirectly by, for example, comparing the levels of rejection among persons living near the proposed facility with a control group or with a group living at a larger distance from the facility (Lober and Green, 1994; Marks and von Winterfeldt, 1984). Such might only show the aggregate effect at the level of sub-samples, whereas an indicator on the individual level is needed. Furthermore, the measurement must concern all individuals; those who have the inclination toward rejection based on backyard motives, but also those who are not motivated to oppose for those reasons. NIMBY disputes usually focus on opponents and the conventional description ignores the supporters (Smith and Marquez, 2000). Hence, the our indicator is a bi-polar scale for the inclination to oppose a facility for specific reasons, concerning the difference between the local and/or individual cost one the one hand, and the general, public benefits on the other. The items reflect the individualist type rationality that causes the lack of support for public goods in so-called social dilemmas. The original analysis of the phenomenon was already defined in terms of social dilemma theory, but that was limited to only the most rigid variant of the prisoners’ dilemma (O’Hare, 1977). In a more sophisticated characterization, the social structure of facility siting is a “volunteer’s timing dilemma” (Devilee, 2002; Weesie, 1993). 7
Scale items had to be variants of the recognition that only the population living near a certain site will be confronted with the cost of the facility, whereas others (the initiators, the investors or the society as a whole) enjoy the benefits, similarly to a scale previously applied in studying public acceptance of wind energy infrastructure (Wolsink 2000; Wolsink 1988). The items also had to indicate the inclination to shift the burden of a waste facility to others (“anywhere, but not here”). On the other side of the scale we must find those individuals that would accept the burden as soon as they recognise that the facility creates a public good. Such an inclination public has clear moral implications, just like the use of the term NIMBY usually contains a strong moral component. Parker et al. (1995) and Bratt (1999) have shown that the inclusion of personal norms is a useful extension of attitude-behaviour models in cases of behaviour that has moral implications. Two types of personal norms are relevant in the social dilemma that characterizes participation in decision-making on facility siting. These are distinguished by Kerr (1995; Kerr et al. 1997), namely personal norms about equity and personal norms about commitment. The equity norms refer to feeling of fairness of processes and legitimacy of decision-making and eventually to the distribution of costs and benefits in society (Adger, 2002; Beck 1992). Within the social dilemma of facility siting, oppositional behaviour is collective action that may fail if the number of active participants remains limited. This situation is also a typical social dilemma (Olson, 1965) and the commitment to others involved in those collective actions is relevant here. The intention to oppose may be a result of unfavourable attitudes that, as such, do not tell us anything about the motives (Fig.1, Relation 1). Motives concerning the “backyard” may also influence the intention to oppose, and this relation (Relation 3) reflects the impact of the NIMBY inclination. However, this does not mean that there is no relation between attitudes and the NIMBY inclination. In hypothesis H2a we formulated that we expect a clear relation, simply because there must be a recognition that the facility is negatively impacting the residents near the site of the facility. As the social dilemma is about the balance between costs and benefits, the costs must be recognized in the first place. Hence, this recognition (Relation 2) only reflects the cognitive components (beliefs) of the negative attributes within the attitudes, but not the positive ones. Crucial to the planners’ perspective on backyard motives is, however, that Relation 3 is significant and substantial: The “backyard” motivated inclination should add to the explanation of intentions and ultimately influence oppositional behaviour. Pilot study Principal component analysis was used to confirm the consistency of the full set of items examined in the pilot study. As all items had to be indicators of one latent variable high loads of the items on the first component were examined. These appeared to reflect the concept of an inclination toward NIMBY behaviour in a reasonably consistent way. The factor loadings on this confirmatory component analysis are shown in Table 2. The strongest indicators in the pilot seemed to be the hard-boiled, individualist utilitymaximizing items (“I don’t think it’s very necessary to bear a part of the collective burden …”) reflecting the supposed economic rationality (“It’s quite stupid to accept ...” and “it’s completely logical ….”), versus the willingness to accept individual cost in support of the public good (“If good arguments can be found ….”). Items that tend to express a preference to balance the costs and benefits with a reference to actors (“Whoever wants to make the profits also has to bear the burden”) or of equity principles (“I don’t think it’s fair to saddle another municipality…”) do not load on the first component. Some items with elements of commitment to equity and fairness (“The costs resulting from waste facilities should be borne by all of us”) moderately load on the first component. These were taken into consideration in the subsequent phases of the research.
8
Table 2. Items and factor loadings in one principal component analysis (N = 134) Items It’s quite stupid to accept a waste facility in your own neighbourhood. Because I don’t think it’s very necessary to bear a part of the collective burden, I don’t accept a waste facility in my neighborhood. It’s completely logical for me that a waste facility should be sited in someone else’s neighborhood. If good arguments can be found to site a waste facility in my neighborhood instead of in someone else’s, I will accept such a facility. I don’t accept a waste facility in my neighborhood, because I think that somebody else would not accept it in his or her own neighborhood. Because a waste facility has to be built somewhere, I don’t object in advance to it being sited in my neighborhood. Accepting a waste facility in your own neighborhood means that you don’t represent your own interests strongly enough. I don’t feel like shouldering the burden of a problem that is also caused by others, by accepting a waste facility in my neighborhood. As far as I’m concerned, a waste facility should be sited in somebody else’s neighborhood. Life is competitive: if a waste facility is sited in someone else’s neighborhood it is not sited in mine The costs resulting from waste facilities should be borne by all of us. It’s only common sense not to object in advance to a waste facility being built in your own neighborhood. With respect to the location of a waste facility, I certainly want to contribute in one way or another to solving a problem that is also caused by others. Whoever wants to make the profits also has to bear the associated burden: A waste facility may cost me something. As long as a waste facility is not built in my neighborhood, I don’t object to it. I’m willing in some way to pay extra in order to contribute to the costs of building a waste facility in another municipality. As a matter of fact, I don’t think it’s fair to saddle another municipality with a waste facility. Item scales range from 1 (fully disagree) to 5 (fully agree)
Factor loading .85 .78 .74 -.74 .72 -.68 .67 .66 .63 .60 -.54 -.50 -.43 -.37 .33 -.14 .06
Table 3. Items selected for field survey ( = .84; N = 134) Item
Mean
if item deleted
The costs resulting from waste facilities should be borne by all of us. (-)
1.85
.83
As far as I am concerned, a waste facility should be sited in someone else’s neighbourhood
2.23
.84
I don’t accept a waste facility in my neighbourhood because I think that somebody else would not accept it in his or her own neighbourhood.
2.37
.82
It’s quite stupid to accept a waste facility in your own neighbourhood.
2.43
.80
If good arguments can be found for siting a waste facility in my neighbourhood instead of in someone else’s, I’ll accept it. (-)
2.44
.81
Because a waste facility has to be built somewhere, I don’t object in advance to it being sited in my neighbourhood. (-)
2.50
.82
2.95
.82
I don’t feel like shouldering the burden of a problem that is also caused by others, by accepting a waste facility in my neighbourhood. Items scales range from 1 (fully disagree) to 5 (fully agree) (-) Item recoded: 1 (fully agree) to 5 (fully disagree)
9
A further selection focused on items with an absolute factor loading > .50, which also fulfilled three criteria: a) diversity in content, b) diversity in the answer (positive or negative), and c) the results of a reliability-test. This resulted in scale of seven items that was very reliable in the pilot study ( = .84) and only included items that improved the reliability of the scale. All items, item means, and indicators of their contribution to the scale reliability are given in Table 3. As the first, the fifth, and the sixth item in the scale are formulated positively (acceptance, even in the own backyard), the answers were recoded for the reliability analysis. Comparing the scale means for the different types of facilities and different levels of education tested the robustness-test of the scale: the scale did not lose its reliability for different levels of education. An analysis of variance revealed that we did not have to differentiate the scale with respect to the type of facility. The pilot study results formed the basis for the final item construction. In addition to performing the component analysis and the reliability tests, we the respondents were interviewed about the interpretation of the formulation of the items. Unintended meanings of particular formulations were found this way and some items were reformulated based on this information. Reliability of the scale in the multiple cases survey The aim of developing the scale was to include it in the explanatory model of attitudes and behaviour in decision-making on waste management facilities. In the multiple cases survey, six scale items in Table 3 were included in the final questionnaire. The one that received the most extreme score was not included, in order to prevent the variable of becoming too skewed. The reliability test of the NIMBY items in the survey showed that a reliable scale could indeed be constructed. Six items in the scale were included based on the pilot study results, but in the other two items in Table 4 the normative element in the NIMBY inclination is obviously revealed. Table 4: Final backyard-motivation scale (nimby-inclination; N = 1525; Item
= .86) Mean
if item deleted
As far as I am concerned, a waste facility should be sited in someone else’s neighbourhood.
1.95
.86
It is in conflict with my ideas about equity. (Eq)
2.11
.84
If good arguments can be found for siting a waste facility in my neighborhood instead of in someone else’s, I’ll accept it (–)
2.26
.85
Because a waste facility ultimately has to be built somewhere, I do not object in advance to it being sited in my neighbourhood. (–)
2.31
.85
It is quite stupid to accept a waste facility in your own neighbourhood.
2.35
.83
I do not consider it fair. (Eq)
2.44
.83
I do not feel like shouldering the burden of a problem that is also caused by others, by accepting a waste facility in my neighbourhood.
2.68
.84
I do not accept a waste facility in my neighbourhood because I think that somebody 2.89 else would not accept it in his or her own neighbourhood. Items scales range from 1 (fully disagree) to 5 (fully agree) (–) Item recoded: 1 (fully agree) to 5 (fully disagree) (Eq) Item originally formulated as indicator for commitment to equity
.85
As explained in the “Basic principles” section, specific personal norms may be significant in cases concerning the shift of a burden to others. Following Kerr (1995), items were included in the questionnaire to measure the personal norms about commitment (to others) and the 10
personal norms about equity. The two items that were included in the scale (Table 4) were added to the questionnaire in an effort to measure the personal norm about equity in social dilemma situations: “It (= siting the facility here) is in conflict with my ideas about equity” and “I don’t consider it fair”. These items could not be distinguished from the items resulting from the pilot measuring the preferences in the social dilemma. Obviously, the norms about equity are at stake when local residents develop an inclination to resist unwanted activities in their neighbourhood. The relation between facility siting issues and environmental injustice is apparently recognized by the residents (Capek, 1993). The personal norm of equity prescribes that the input and the output in the social dilemma be balanced. The two items match very well with the normative aspect as measured with the NIMBY items. The reliability was improved by including these items in the scale. The resulting scale supports hypothesis H1, as it can be considered very reliable (α = .86). The item means (recoded for three positive items) indicate the preferences in the social dilemma. It appears that, on average, the respondents slightly disagree with most of the statements, although this should be interpreted carefully. The items indicate an inclination that might be considered socially less acceptable and there may be a tendency not to be reluctant with supporting the statements. Within the context of this study, it is not the central tendency (the average support) what counts, but the differences between individuals (the variance). All items have a clear variance and the scale has clear potential explanatory power as well be shown in the next sections. The statements with the highest means were “Because others would not accept it either” and “I don’t feel like shouldering the burden of a problem that is also caused by others.” The item “Site it in someone else’s neighbourhood” was the one the respondents disagreed with the strongest. The overall picture is that the item-means indicate that there is still a substantial minority that prefers not to volunteer in the social dilemma, and a smaller minority that also wants to shift the burden to others. Further results: backyard motives related to attitudes To investigate the backgrounds of intentional and behavioural opposition, we developed a model based on Ajzen’s (1991) theory of planned behaviour. The model was tested and includes, in addition to the NIMBY inclination, the following concepts (Devilee 2002): Perceived benefits: 7-item scale ( = .79); (such as source of employment, efficient use of raw materials etc.) Perceived costs: 9-item scale ( = .83); (ugly, noise annoyance, nuisance from birds etc.) Risk perception: 7-item scale ( = .81); (frightened for accidents, installation ‘not dangerous’, installation is ‘time bomb’ etc.) Personal norm about commitment: 2 items ( = .65); (“they expect me to disapprove..”; “I promised others to reject”) Intention to accept: single item 5-point scale from strong opponent to strong proponent; Opposition to the waste facility: 9 dichotomous items combined in one strong cumulative scale (Loevinger’s H = .50; from “signed petition” to “contacted a lawyer”). In the theoretical model perceived health consequences had been included. However, this 3item scale ( = .83) was discarded from the tested model, because its relations with the other variables in the model were not significant (p > .01). Perceptions of health consequences were clearly related to the intention to accept the facility, but there was also a rather strong correlation with perceived risk. Due to collinearity this scale did not add significant impact to the explanation by perceived risk. As usual in research in which actual political participation (beyond voting) is measured, the oppositional behaviour is a very skewed variable, since 86% in the total sample had taken no action at all. The observation that variables indicating psychological involvement, such as the 11
intention to oppose, are much better related to the other research variables than active political participation (Cohen et al., 2001) is clearly valid in this case. The full model explains 66% of the variance of the “NIMBY inclination,” and 45% of the “intention to accept.” The joint effect (.42 in the full model) of the direct and indirect paths of perceived risk is much stronger than the effect of the NIMBY variable, and this indicates that “perception of risk” is the crucial factor with respect to the intention to accept a waste facility. This is in line with prospect theory that tells us that avoidance of risk generally has heavier weight in an assessment of new situations than do possible benefits (Tversky and Kahneman, 1992). The variable “oppositional behaviour” is explained only to a limited degree (8%), mainly due to its extreme skewness. The final model as reported by Devilee (2002) is used to examine the differences between two sub-samples in the following sections. Model for residents in cases where facilities were not built Attitudes and perceptions on facilities easily shift once facilities are actually built, so their interdependence may also be changed (Wolsink 2007b). Figure 2 presents the structural model for residents in cases where planned facilities were not built. The model of the subsample of the first three cases (Table 1) has a good fit (Jöreskog and Sörbom, 1996). The model explains 63% of the variance of the NIMBY inclination, 48% of the intention to accept, and 5% of the opposition. The dotted line between “perceived costs” and “intention to accept” indicates an insignificant relation (p > .01). What is very salient in the model is the relationship of perceived risk with the NIMBY inclination (γ = .48) and with the intention to accept (γ = -.21).
Personal norm commitment
.33 .48 Nimby inclination
Perceived risk
.18 Perceived cost
-.21 -.09
Perceived benefits
.21 GFI =
-.36 Intention to oppose
-.23
Oppositional behaviour
.921; AGFI = .907; RMR = .044
Figure 2. Explaining support of and opposition to cancelled projects (N = 793) Perceived risk has a total effect of -.38 by the combination effect of the direct impact risk and the indirect path via NIMBY. Perception of risk is the crucial factor with respect to the intention to accept a waste facility. Apparently the avoidance of risk is an important reason for not being willing to accept a waste facility, but it is only partly linked to “backyard” motives. This supports Hypothesis H2a. There are two paths to the intention to accept a waste facility. The direct paths in Figure 2 from “perceived risk” (γ = -.21) to the intention to accept represent a reason to oppose a waste facility without any reference to the backyard. The 12
perception whether or not the facility will create benefits adds to this intention, while there is no relation at all with the NIMBY inclination (H2a). The direct paths represent to what extent local residents oppose a waste facility for reasons that are directly associated with a waste facility and not with the position taken in the volunteer’s timing dilemma. The path that has been considered as most evident by policy makers is the direct influence of the NIMBY inclination. Following their argument, it is logical that local residents immediately put the problem in a social dilemma context and decide to oppose a waste facility. The indirect path via the NIMBY inclination represents the validity of this line of reasoning. The relationship between the NIMBY inclination and the intention to accept is clearly significant (β = -.36). This shows that at least a substantial part of the population living in a neighbourhood where a planned facility was cancelled was clearly motivated by the perception that they had to carry the burden for others and that this could not be considered fair. This perceptions is predominantly influenced by perceived risk, but also by the personal norm about commitment. Here we see that the NIMBY inclination is strongly based on personal norms. It is influenced by the norm about commitment to others and the scale itself contains items that are indicators of the norms about equity. Model for residents in cases where facilities were built In the model for the three cases of residents in cases where the planned facilities were actually built (Figure 3) the relation between the perceived costs and the NIMBY inclination is not significant (p > .01). This is a clear difference from the model figure presented in Figure 2. The same holds for the relation between the perceived benefits and the intention to accept, which is not significant. The fit of the model is slightly better for this subsample than for the sample in cases of cancelled facilities. Personal norm commitment
.52 .45 Nimby inclination
Perceived risk
-.03 Perceived cost
-.43 -.06
Perceived benefits
.08 GFI =
-.14 Intention to oppose
-.38
Oppositional behaviour
.926; AGFI = .912; RMR = .044
Figure 3: Explaining support and opposition in cases where facilities were built (N = 718) Risk perception is even more prominent in the model. The negative consequences seem to have boiled down to the perceived risk only. The perceived risk concerns safety and accidents, while the perceived costs are negative consequences such as smell, attraction of birds, etc. The perceived risk also contributes to the NIMBY inclination, but now this path is hardly important because of the relation between the NIMBY inclination and the intention to accept, which has become rather weak (β = -.14). 13
In these cases, the residents were able to gain some experience with the waste facility. Moreover, information processing measured by a strong cumulative scale and controlled for the significant effect of type of facility (F (1595,2) = 16.3; p < .01) appeared to be significantly higher in those cases (F (1595,1) = 97.6; p
