Dark Sky Heritage

(1)

Dark Sky Heritage

Perceptions of local dairy farmers on a ‘Dark Sky Park’ for the Wadden Sea region

Sander Schuil (S2908174)

Bachelor Project

Human Geography & Urban and Regional Planning Faculty of Spatial Sciences, University of Groningen Supervisor: ir. B.M. Boumans, MSc

Words: 6600 June 10, 2019

(2)

Abstract

‘Dark Sky Parks’, the latest labelling extensions for heritage sites, are thought to create new

opportunities for regions to stand out. Recent research suggests however, that heritage-building is not rarely met with local resistance. Especially ‘traditional’ economic sectors might not benefit from a Dark Sky Park inscription. Therefore, a case study in the Wadden Sea region was conducted to explore local perceptions on the current intention to develop a Dark Sky Park for said region. Local dairy farmers were surveyed for this purpose. The results indicate that dairy farmers are not necessarily against a Dark Sky Park. However, A Paired Sample T-Test confirmed that dairy farmers are more reluctant in their stance towards an official Dark Sky Park denomination than only an intention to limit light pollution in the area. Two Multiple Linear Regressions were used to test the variables that might explain these views. It was found that dairy farmers who highly value the starry sky in their

municipality, are more inclined to think that limiting the light pollution in the region is a good idea.

However, motivations by anthropocentric concern and the valuation of jobs over the protection of the environment leads to a more negative stance towards limiting light pollution. In turn, support for an official ‘Dark Sky Park’ grows when farmers believe it will deliver tourism opportunities for the region. Lower support, however, was explained by anthropocentric concern and a higher confidence in the future of one’s dairy company. Further research might investigate perceptions across the entire local population and integrate qualitative research methods. Some results also indicate the region might lack in participative governance, whether this is true and possibly influences local attitudes towards heritage policies, might be explored by further research.

(3)

1. Introduction

1.1 Background

Societal developments as globalization, time-space compression and the consequential marginalization of peripheral areas, increasingly give rise to the need for products as well as spaces and places to construct an image or brand in order for them to remain competitive in a post-modern world (Blichfeldt & Halkier, 2014; Nogué et al., 2017). This development is especially relevant for rural areas, which are increasingly transforming from purely agricultural production areas, to economies that include consumption-based land uses (Olson & Munroe, 2012). One of the ways for areas to stand out is the appointment of areas with special heritage statutes, fed by cultural or natural features that are deemed important to preserve.

A good example of such an area is the Wadden Sea region, which got nominated as UNESCO World Heritage site in 2009 (Krauss, 2005). The labelling of these areas and the protective measures that inevitably come with them, however, are not rarely met with local resistance (Krauss, 2005;

Liburd & Becken, 2017). The prospective status of the Wadden Sea area as World Heritage Site was no exception. Although most stakeholders agreed that the Wadden Sea region needed protection, local community representatives were predominantly against the World Heritage denomination itself (Krauss, 2005). In short, the symbioses of heritage, tourism and ‘’an often-destructive role of politics in crafting brands’’ results in a strained and complex relationship between preservation and

promotion, a difficult challenge in contemporary heritage management and planning (Ryan &

Silvanto, 2010).

Since around 2000 and especially in the last decade, a new form of heritage labelling has emerged, focused on protecting the visibility of the starry sky by developing and naming ‘Dark Sky Parks’

(Charlier & Bourgeois, 2013). In many urbanized areas, this starry night sky is considered a lost heritage (Ibid.), but nevertheless a resource that can potentially be recovered—unlike other forms of lost heritage (Duriscoe, 2001 cited in Collison & Poe, 2013). Dark Sky initiatives, rarely created on their own, are often built on previously attained nominations and natural or cultural parks (Charlier &

Bourgeois, 2013). In line with this observation, over 40 organizations in the Netherlands signed an agreement (2016), in which they promise to put effort in increasing the darkness of the night sky (and thus limit light pollution) in the Wadden Sea region (Waddenzee.nl, 2018). Their goal is to expand the World Heritage denomination with a ‘Dark Sky Park’ for the entire Wadden Sea region, thereby reinforcing and improving on the natural and cultural heritage and identity of the region (Ibid.). But these—mostly governmental—organizations are of course not the sole stakeholders in the area. The mainland region has a long tradition of intensive agricultural use, for example (Sijtsma et al., 2012).

Today, lots of livestock farms in the region use light for the effectuation of their businesses. Next to streetlights, industrial areas, greenhouse complexes and sports fields, their livestock barns are one of the only other possible producers of sizable light scattering in the rural region (see also cover photo).

Indeed, some of their properties distinctly light up on night scans of the Dutch Northern-Friesland region, for example (Waddenzee.nl, 2016). By limiting these light-emissions, outsiders might be inclined to imagine benefits for (nature) preservation and the tourism sector, where locals that depend on the more traditional economic sectors may fear consequences for their livelihood (Walker &

Fortmann, 2003; Jones & Shaw, 2012; Rodríguez‐Darias et al., 2016). However, for a successful and sustainable destination branding of heritage sites, consensus building in stakeholder groups is of paramount importance (Krauss, 2005; Ryan & Silvanto, 2010; Senil, 2011; Jones & Shaw, 2012). It determines the amenability regarding possible legislative measures (Rosario, 2007), which can only be objectively considered if all perspectives of stakeholders—and the factors, interests and values that create them—are known (Krauss, 2005; Bekken & Liburd, 2017).

This research therefore aims to uncover the perspectives that the ‘traditional’ economical sector has concerning the intent to expand the World Heritage denomination with a ‘Dark Sky Park’ and identify possible causal factors. Therefore, the following main-research question was drafted:

What are the views of local dairy farmers regarding the intent to work towards a ‘Dark Sky Park’ for the Wadden Sea region?

(5)

The following sub-questions will be addressed:

- Are local dairy farmers more reluctant towards an expansion of the World Heritage

denomination with that of an official ‘Dark Sky Park’ status than only an intent to reduce light pollution in the area?

- What are the underlying factors and motives that form the perceptions of local dairy farmers regarding a ‘Dark Sky Park’ and reducing the light pollution in the area?

1.2 Structure

The next chapter provides an overview of existing literature about heritage-tourism issues and relates this to new ‘Dark Sky’ heritage-processes. It also covers the factors that might influence the

perceptions of local farmers regarding these concepts. The choice for an extensive research design in combination with a questionnaire survey is explained in the methodology section, together with an overview of ethical considerations. The methodology section is followed by the results section, which presents the descriptive results from questionnaire data as well as the results of statistical analyses.

This data is held against the theoretical framework and its derived conceptual model here as well. The conclusion section harbours a summary of the most important findings, contains a reflection on the quality of the study and provides recommendations for further research.

2. Theoretical framework

Notable is the fact that in a lot of the publications surrounding the Dark Sky Park initiative for the Wadden Sea region, the creation of an unique tourist environment is explicitly mentioned as an advantage (Clafis, 2018) Municipalities note that more can be done about the ‘experience of the area’

and on ‘the area of culture’ (Ibid.). Thus, from local governments’ perspectives, a lot of attention seems to be given to the benefits that a ‘Dark Sky’ brand will deliver to the development of the area, especially to the tourism sector.

Tourism itself is often presented as facilitating and increasing social interactions. It would represent the post-modernistic landscape, a world parting from ideas of domination and moving towards mutual understanding and a reduction of conflict (Poria & Ashworth, 2009). According to these authors however, heritage tourism may actually inhibit such understanding and even increase conflict. The authors use the term heritagization to describe the process where heritage tourism is used to attain certain social goals. A distinction can be made between the process of heritagization, that results in the eventual presentation of heritage sites for heritage tourism, and cultural tourism, which focuses on the conservation or preservation of heritage. While the latter thus focuses on sustaining and protecting in the literal sense, the former is primarily built on certain ideological frameworks and can eventually increase segregation (Ibid.). This is because in heritagization, all kinds of (in)tangible qualities are attributed to heritage sites (Porter & Salazar, 2005). These qualities inherently reinforce certain value-based systems, thereby unavoidably undermining others (Ibid.). As a consequence, the endeavor of designating areas with a heritage status can be used to legitimize certain ideologies, strengthen ‘us versus them’ thinking and subsequently cause conflicts (Ibid.). In public debates around the nomination process for the Wadden Sea region to become UNESCO World Heritage for example, adherents to the view that the region is an important ‘pure nature’ landscape wanted to restrict human interference in the ecosystem. But the local population framed the region as cultural heritage.

Therefore, they used constructs like ‘the Frisian identity’ as a means to emphasize their interests (Krauss, 2005). Even when stakeholders agree on certain heritage values, the actual management of a protected site can still favor certain ideological frameworks (Jones & Shaw, 2012; Rodríguez‐Darias et al., 2016) and thereby limit developments in the area for adherents of other ideologies (Jones & Shaw, 2012).

The starry sky is increasingly seen as a true environmental amenity in Europe, whose protection and enhancement imparts new opportunities for local, sustainable tourism development (Charlier &

Bourgeois, 2013). Organizations like UNESCO actively promote the reducing of light pollution impacts on biodiversity, the revealing of nocturnal landscapes and the cultural heritage of the starry sky (Ibid.). The official recognition of the starry sky as distinct invaluable ‘object’ to protect and label

(6)

was difficult to develop however, as it does not entirely fit in existing heritage-making practices. The starry sky itself, for example is not threatened, it is only its visibility that may be limited. Besides, it is difficult to argue that the starry sky is cultural heritage itself, but it can be associated with

astronomical heritage and religion. Therefore, the compromise hitherto is to view nocturnal darkness as a supportive addition to existing heritage sites, reflecting the connection between natural and cultural heritages (Ibid.). The criteria for starry skies are subsequently built around their aesthetical connection with nocturnal landscapes and can be considered to give an indication of the naturalness of the spaces in which they are visible (Ibid.). Dark Sky Parks can therefore be seen as an evolution of the protectionist responses to natural areas distant from the city, that have been in place since the 19^th century (Piolle, 1993). The parks can serve as areas of darkness and silence in which city dwellers come to ‘slow down’, an extension to the classic city-dwellers’ longing for nature (Gwiazdzinki, 2002)

.

For these reasons, the starry sky as a heritage phenomenon, is closely related to the landscape approach in heritage (Charlier & Bourgeois, 2013).

Indeed, urban and rural spaces are increasingly understood as being inter-related and co-dependent in function (Olson & Munroe, 2012). With an apparent role for city-dwellers in rural areas’

heritagization processes, it is therefore not difficult to imagine frictions. Taking this into account, Woods (2009) warns about possible uneven outcomes in landscape conflicts because of the strong economic and demographic role urban areas have in this relationship. ‘Urban’ aesthetical views about landscape can be imposed by environmentalist-dominated governments on the regional level, resulting in conflicts between traditional economies and ‘new’ consumption-based views (Walker & Fortmann, 2003). This brings forward the question of who landscapes ‘belong’ to (Ibid.) and returns us to the notion that heritage creation can be used as a tool to exclude, much like gentrification processes in cities (Senil, 2011).

In the literature about locals’ perceptions on matters of environmental protection and heritage-sites, general views differ extensively (Rodríguez‐Darias et al., 2016; Krauss, 2005; Rasoolimanesh et al., 2015). Respective attitudes are therefore heavily depended on context (Pienaar et al., 2013).

Subsequently, a multitude of factors like age, gender, level of education, income and occupation have been shown to influence local perceptions (Rodríguez‐Darias et al., 2016; Pienaar et al., 2013).

Rodríguez‐Darias et al. (2016) found that the degree of involvement in an area was the most important variable that explained positive or negative attitudes towards protective sites. A higher involvement in planning-processes has also been shown to increase support for World Heritage Sites (Rasoolimanesh et al., 2015). Whereas in other studies, a higher level of education was positively related towards heritage-building practises (Jones & Dunlap, 1992;You et al., 2014). Locals’ values and believes, in the form of environmental and cultural attitudes also influence their perceptions (Rasoolimanesh et al., 2015). And if benefits for tourism within a resident’s community are envisioned, support can increase as well (Ibid.). Additionally, Zuo et. al. (2017), have linked a higher faith in governments with greater support towards concerning processes. To visualize the processes outlined above, a conceptual model (figure 1) was created.

Figure 1: Conceptual model simplifying the processes around heritage discussions

(7)

2.1 Hypotheses

The null-hypotheses that are tested in this study are:

H01: Dairy farmers are equally positive about limiting light pollution in the Wadden Sea Region as about expanding the World Heritage site with a ‘Dark Sky Park’.

H02: The factors derived from the literature do not influence perceptions regarding limiting light pollution in the Wadden Sea Region.

H03: The factors derived from the literature do not influence perceptions regarding the expansion of the World Heritage site with a ‘Dark Sky Park’.

The alternative hypotheses, or the expected results derived from the theory are:

H1: Farmers are more positive about limiting light pollution in the Wadden Sea Region than about expanding the World Heritage site with an official ‘Dark Sky Park’.

H2: The factors derived from the literature influence perceptions regarding limiting light pollution in the Wadden Sea Region.

H3: The factors derived from the literature influence perceptions regarding the expansion of the World Heritage site with a ‘Dark Sky Park’.

3. Methodology

3.1 Research design

This thesis uses an extensive research design. It is the goal of such a design to find generalizable patterns in large-n research data, which could indicate causality (Clifford et al., 2016). The goal is to find the factors that might generally influence the opinions of local dairy farmers regarding the research themes, a quantitative research approach fits this approach. Although an intensive research method e.g. interviews could reveal deeper, fundamental details in case studies (Ibid.), this particular research on the other hand, tries to first identify possible factors by controlling for a larger-n and other factors (e.g. socio-demographic factors) that might have cross-linear relationships and correlations in the target population. Because of time constraints, a follow-up intensive research approach or mixed- methods approach was not feasible, but this would nevertheless be an interesting way to possibly reveal deeper, subjective understandings of respondent motives.

Since there is a wish for a large ‘Dark Sky Park’ in the Northern Netherlands, this research focused on said area by using a case study research strategy. Case studies are an apt method to learn about peoples meaning-making in a complicated socio-cultural context (Taylor, 2016). Analytical

generalization will be used as a way to compare statistically generalized findings for the population of this specific case study to findings and theories from existing literature, in order to see if the latter can be supported (Yin, 2014 cited by Taylor, 2016, p. 587). The population selected for this research consists of all dairy cattle companies in the Dutch municipalities that border the Wadden Sea World Heritage property (see also figure 3). This specific group of farmers was selected since they have a bigger impact on light-emission reduction as compared to farmers more inland. It is also expected they have a more outspoken opinion on the matter, because their close proximity to the Wadden Sea. This strategy might have improved the response rate as well. The selection makes it difficult however, to make assumptions about the opinions of other dairy-cattle farmers in the (Northern) Netherlands, their different geography might result in different perceptions.

Because of limited resources, a random sample of 500 companies was generated in Microsoft Excel and send a letter (appendix 1) with a link to an online questionnaire. While this sampling strategy ensures an equal chance of selection, a downside of the strategy is that the geographical distribution of the population may result in under-representation of certain geographical areas (McLafferty, 2016). If opinions in excluded geographical areas differ significantly from the acquired sample, this means the sampling technique was biased and results might not be entirely valid. (Ibid.).

(8)

3.2 Questionnaire survey

Primary data was collected through questionnaire survey research. Surveys are an efficient way to collect data about the characteristics, behaviours and attitudes inside populations (McLafferty, 2016).

Questionnaire surveys are especially useful for finding out people’s opinions on environmental, social and political problems (Ibid.). Since the theoretical framework of this research warrants an interest in the respondents’ opinions on matters related to all these three themes, it is a particularly valuable research instrument. The online questionnaire (appendix 2) was divided into six subsections, primarily to pre-sort questions on theme and also to increase clarity for respondents. The first section consists of questions that are primarily enabling the classification of respondents into groups, including socio- economic variables that might influence respondents’ attitudes to the problems relevant for this research (Rodríguez‐Darias et al., 2016). The second section contains the important dependent variables that measure the opinions of respondents regarding a reduction of light-emissions (Limit_LE_Wad) and an official expansion of the World Heritage site with a Dark Sky Park

(ExpandDS_Park). In this section, respondents’ attitudes towards tourism, the starry sky and heritage frameworks are also measured to be able to identify and statistically control for them in the analyses.

The other three sections contain questions with the same goal, about regulatory pressure, light-

emissions in relation to dairy farms, and environmental attitudes respectively. The last section contains questions about environmental attitudes that were adapted from the Environmental Attitudes Inventory and have been thoroughly tested for validity (Milfont & Duckitt, 2010).

Most questions were composed as fixed-response questions. While this offers a limited set of responses and might prevent respondents to express their ‘true’ views, it does make the process of answering the questions and the use of statistical techniques easier (McLafferty, 2016). A semi- continuous Likert scale was used in all questions that ask for a respondents’ views. Because it is important to keep questionnaires simple (Ibid.), respondents were required to rate statements on a scale of 1 (totally disagree) to 10 (fully agree). The resulting variable can be interpreted as interval and therefore means can be used in statistical analyses (Ibid.).

In the results section of this thesis, descriptive statistics present important aggregate results derived from the response data. Statistical tests were then performed in IBM SPSS that were interpreted to determine whether validity of the stated hypotheses can be assumed for the broader population.

Whether statistically significant differences can be assumed between the opinions about limiting light pollution in the Wadden Sea region and the expansion of the UNESCO World Heritage the Wadden Sea with a Dark Sky Park denomination, was tested through a Paired Sample T-Test. Potentially explanatory variables for these two variables that arose from the literature were statistically tested and cross-controlled for in two Multiple Linear Regressions. Whether all tested variables qualify for valid usability in these tests was first determined by a mix of collinear statistics, central limit theorem assumptions, and analyses of residuals—where relevant.

3.3 Ethics

For every question in the questionnaire, possible ethical problems have been carefully considered.

Since asking about incomes can be an undesirable exercise for respondents, for example, it was decided to not include such a question. The residence of respondents was determined on the four numbers in their ZIP-code, which strikes a good balance between anonymity and spatial

analysability in GIS. Answers in the questionnaire cannot be traced to individual respondents and respondents were explicitly informed about this. Respondents were also guaranteed that their data will be treated confidentially and will not be shared outside of the university. To rule out any potential harm in the population, respondents or other parties will not be supplied with results.

Respondents were not given any pressure to complete the survey in a specific time frame and were given full freedom to decide if they wanted to participate. They were also given information to help them with this decision (e.g. an estimation of the time needed to complete the survey).

Questions and conclusions were always drafted with a cultural awareness towards possible beliefs and ethical views in the agricultural community in mind. Additionally, I am not affiliated with any parties that might be stakeholders in the subject of this thesis. While I thus emphasize my neutrality, the online nature of the questionnaire also guaranteed my inability to directly influence any

respondents.

(9)

4. Results

4.1 Respondent demographics

The respondents (n=56) submitted their completed surveys between April 30 and May 12, 2019. The valid response rate was 11,2%. Among the respondents were 18 females (32,1%) and 38 males (67,9%). Their age varied between 22 and 80 with a mean age of 51. Figure 2 shows the full age distribution.

Indicated in figure 3 are the ZIP-code (postcode 4) areas in which the respondents reside. Notable is that municipalities with a relatively low number of dairy farmers are also under-represented in number of respondents. Under-representation is especially evident for the islands, with only one respondent from Texel. It is suspected that the non-existence of respondents from the Frisian municipality

‘Waadhoeke’ has to do with an error in the origin address file that was used to make a sample.

Remarkably, a few respondents from Groningen seem to reside outside of the sampled municipalities.

The reason for this is unknown, but they were left inside the analyses to obtain a higher n. All in all, the geographical representativeness of the data is mediocre, it was already established that this is partially caused by the used sampling technique (section 3.2).

Figure 3: Map with locations of respondents’ residence, sampling area, the World Heritage property and current Dark Sky Parks in the area

Figure 2: Age distribution

(10)

4.2 Questionnaire results

Table 1: Descriptive statistics

The descriptive statistics of respondent answers on the most important survey questions are listed in table 1. Firstly, it is interesting to learn that only 34% of respondents knew there are intentions to limit the light pollution in the Wadden Sea region and its vicinity (variable 1). Variable 2 in the table shows that on a range from 1 (primarily natural heritage) to 10 (primarily cultural heritage), respondents answer the question ‘How would you describe the Wadden Sea region and its surrounding region?

with a mean score of 4.84. Since this is higher than 4.5, respondents seem on average inclined to think about the Wadden Sea region as being a region of cultural heritage, which is in line with what Krauss (2005) observed in the local population of the region. We will later see that this result is not

significantly contributing to ‘Dark Sky’ heritage opinions however, whereas Krauss (2005) saw this region-framing as primary factor in the World Heritage discussion. All other questions (statements) in table 1 were measured on a scale from 1 to 10, with 1 meaning total disagreement and 10 meaning total agreement. When asked if respondents think that ‘being able to see the starry sky is an important quality of the municipality where their company is located’, respondents strongly agree with 6.98 points on average. Respondents also agree that it is a good intention to limit the light-emissions in their municipality, with a mean of 6.75. Survey question 10 which asked the same question, but for the Wadden Sea area as a whole, gives similar though slightly higher results. The variance in answers among respondents regarding question 14 (to what extent they agree with it being a good idea to officially expand the ‘UNESCO World Heritage the Wadden Sea’ with a ‘Dark Sky Park’

denomination), is relatively high. Opinions of respondents regarding this matter seemingly deviate more widely than in other survey questions. Despite this, on average respondents slightly agreed with the statement, but are nonetheless more reluctant about an official ‘Dark Sky Park’ denomination than about just limiting the light-emissions. An explanation for this might be found in the following commentary:

‘’Another stamp from UNESCO seems unnecessary to me. This is often followed by idiotic regulations, devised by someone behind his desk who has never been here.’’

Continuing, we find that the respondents strongly agree (6.5 points) with it being nonsensical to obligate farmers to limit their light pollution. Furthermore, it stands out that respondents in question 13 only slightly agree (5.46 points) that a better visibility of stars will benefit tourism in their

municipality. A similar result was obtained for question 14, respondents slightly agree that it is a good idea to attract more tourists to their municipality.

Because the whole intention to limit light pollution in the Wadden Sea region is presented as having started as a bottom-up process (Waddenzee.nl, 2018), it is interesting to learn that respondents

(11)

do not feel included in the decision-making process around the Dark Sky initiative (question 18, 3.54 points). This might have to do with the early state of the process but might also expose a governance problem. What may point towards a governance problem, however, is the fact that on average,

respondents moderately disagree with the statement that they are treated as an equal party in plans that concern their region (question 17, 3,8 points). A respondent about this:

‘’These self-proclaimed nature experts from the big city determine how or what nature should look like in the Netherlands at the expense of people who have worked with nature their whole lives’’.

Another respondent:

‘’Our municipality currently uses sneaky schemes concerning tourism. They take land from you to do something random for ‘tourism’. I’m quite open towards tourists […], but the problem is they [tourists] look negatively to farmers, while the nature problem lies largely with the consumer, including the tourist.’’

These and other similar comments indicate there are a lot of emotions connected towards some of the themes in the questionnaire. They are good examples of the urban-rural tensions which are covered in the theory. In the following subsections it will be explored whether these opinions statistically

influence the views about light-emission reduction and/or a ‘Dark Sky Park’.

4.3 Paired Samples T-Test

To answer the question: ‘Are local dairy farmers more reluctant towards an expansion of the ‘World Heritage the Wadden Sea’ denomination with that of an official ‘Dark Sky Park’ status

(ExpandDS_Park), than only an intent to reduce light pollution in the area (Limit_LE_Wad)?’, a Paired Sample T-Test was performed. The variable ExpandDS_Park was measured by asking respondents whether they agree with it being a good intention to officially expand the ‘UNESCO World Heritage the Wadden Sea with a Dark Sky Park. They were required to do so on a scale from 1 to 10, 1 meaning total disagreement and 10 meaning total agreement. The same scale was used on variable Limit_LE_Wad, for which respondents were asked to report to what extent they think it is a good intention to limit the light-emissions in the Wadden Sea region as much as possible. The full output of the Paired Sample T-Test can be found in appendix 3, the results in table 2.

Table 2: Results Paired Samples T-Test

The fourth column shows a P value of .000 which is smaller than the used significance level of 0.05, and thus predicts a significant difference in the population. The null-hypothesis H01, is therefore rejected. The alternative hypothesis H1: Farmers are more positive about limiting light pollution in the Wadden Sea Region than about expanding the World Heritage site with an official ‘Dark Sky Park’, is accepted. It can furthermore be assumed with 95% confidence, that the mean value of ExpandDS_Park in the population lies between 1.029 and 1.971 points lower than the mean value of Limit_LE_Wad.

4.4 Multiple Linear Regression: Limit Light Pollution

To explore what factors might predict the views regarding Limit_LE_Wad and ExpandDS_Park in the population, potentially relevant variables—according to the literature—were entered in two Multiple Linear Regression analyses. Any significant outcomes can be interpreted as having been controlled or

‘purified’ for all other variables in the corresponding analysis. The control-variables that were used include socio-economic variables like gender, age and level of education, but also environmental

(12)

attitudes, heritage framing of the region, views about tourism, views on participation in the region, and more. Appendix 4 houses a complete overview of the variables and their meanings.

Most of the variables, however, did not have a significant linear relation with Limit_LE_Wad in the population. They are left out in table 3, which only shows significantly contributing variables. From the model summary in the appendix we learn a linear relation is assumed between Limit_LE_Wad and the predicting model as the model as a whole is significant with a P value of 0.00. Therefore, the Null- Hypothesis H02 is rejected and the alternative hypothesis H2 is accepted: The factors derived from the literature, influence perceptions regarding limiting light pollution in the Wadden Sea Region.

Moreover, a good fit of the model is assumed with an adjusted R Square value of .734. Corrected for the number of variables, the model predicts roughly 73% of the variance in Limit_LE_Wad.

From table 3 we conclude that variable QualityStars_M has a significant linear relation in the population. With every point that QualityStars_M increases, Limit_LE_Wad increases with an average of 0.623. In other words, farmers that find the visibility of stars an important quality of their

municipality are more positive towards the intention of reducing light-emissions in the Wadden Sea region. Relative to the total variability in Limit_LE_Wad, the strength of this correlation is moderate (0.349).

We can also assume that AbusingEnviron also has a significant, but negative linear relation in the population. When controlling for all other variables, the more farmers in the population think that humans are abusing the environment, the less positive they are towards reducing light-emissions in the Wadden Sea region. For every mean point that AbusingEnviron increases in the population,

Limit_LE_Wad decreases with a mean of 0.319. Although indicating significance, a semi-partial correlation of -0.191 shows that strength of the correlation is weak. A simple explanation of this odd finding cannot be given. The variable might have an accidental linear relation with Limit_LE_Wad, although the literature does indicate it can be a factor (Rasoolimanesh et al., 2015; Milfont & Duckitt, 2010). Another explanation might be that a more ‘helpless’ stance towards environmental issues results in a more indifferent stance towards limiting light-emissions.

Another environmental attitude (RiverSports) also has a significant, but weak correlation with Limit_LE_Wad. For every mean point increase of RiverSports, Limit_LE_Wad will on average have a 0.262 points lower mean score. In other words, local dairy farmers who are motivated by

anthropocentric concerns in their environmental conservation efforts, will be less positive towards reducing light-emissions in the Wadden Sea region.

The partialled-out correlation between Jobs_VS_Environm and Limit_LE_Wad is even weaker, but also indicates a significant linear relation in the population. In the population, we can expect a mean decrease of 0.281 points in Limit_LE_Wad for every mean point Jobs_VS_Environm increases. So, dairy farmers who value the protection of people’s jobs relatively higher than the protection of the environment, are less positive towards reducing light-emissions in the Wadden Sea region as well.

Table 3: Results Multiple Linear Regression for Limit_LE_WAD

4.5 Multiple Linear Regression: Dark Sky Park

Table 4 below contains the significant findings from the Regression analyses with ExpandDS_Park as dependent variable. Most variables had no significant linear relationship with ExpandDS_Park in the population. The complete overview of the Regression output can be found in appendix 5. The model summary shows a P value of 0.000, with which we can assume a linear relation between

ExpandDS_Park and the predicting model. This means we reject null-hypothesis H03 and accept the alternative hypothesis H3: The factors derived from the literature influence perceptions regarding the expansion of the World Heritage site with a ‘Dark Sky Park’. An adjusted R Square value of .754

(13)

means that the model is a good fit for ExpandDS_Park. Corrected for the number of variables, the model is able to predict roughly 75% of the variance in ExpandDS_Park.

Tables 4 shows AddedValTS_M has a significant, moderately positive correlation with

ExpandDS_Park. It is predicted that a point increase in AddedValTS_M, yields a 0.643 points higher mean score for ExpandDS_Park. This means that local dairy farmers who—to a relatively high extent—think that a better visibility of the starry sky has added value for the tourist-sector in their municipality, will be more positive towards officially expanding the World Heritage status with a

‘Dark Sky Park’ denomination. This result is supported by the findings of Rasoolimanesh et al.

(2015), who found that a higher support for tourism development increased the support for a World Heritage Site.

RiverSports returns as significant contributor in this model, with a weak negative semi-partial correlation. Local dairy farmers who are motivated by anthropocentric concerns in their environmental conservation efforts, will be less positive towards officially expanding the World Heritage

denomination with a ‘Dark Sky Park’ status.

A remarkable finding is that dairy farmers in the population who have a relatively large amount of trust in the future of their company (Trust_Future_C), will be less positive towards officially

expanding the World Heritage denomination with a ‘Dark Sky Park’ status. This correlation is weak, nevertheless significant. This might again be explained by the feeling that such a denomination will come with regulations that limit dairy companies (see also Jones & Shaw, 2012), as was already touched upon in section 4.2. Farmers with low(er) trust in their company’s future, might be more indifferent about such regulations.

Table 4: Results Multiple Linear Regression for ExpandDS_Park

5. Conclusion

This research was able to provide an overview of the views that dairy farmers in the Dutch Wadden Sea region have regarding a ‘Dark Sky Park’. Since dairy farmers are one of the groups that possibly stand to lose the most from policy implications, it is interesting to learn that they do not have a negative stance towards the Dark Sky initiative per se. However, the findings confirm that dairy farmers in the region are more reluctant towards an official expansion of the World Heritage Site with a Dark Sky Park (ExpandDS_Park) than they are towards limiting light pollution (Limit_LE_Wad).

The most important variable in the model that explained the variance in Limit_LE_Wad was the quality attributed to a starry sky. Support for light-emission reduction has a moderately strong, positive relationship with this value attached to the starry sky. Other significantly contributing variables that explain the variance in Limit_LE_Wad also reside in environmental attitudes.

Specifically, a motivation by anthropocentric concern and the valuation of jobs over the protection of the environment lead to a lower valuation of this variable, although these relationships with

Limit_LE_Wad are weak. Farmers with higher anthropocentric concern also seem to be more negative about a Dark Sky Park, which is a weak correlation, but nevertheless supported by Rasoolimanesh et al. (2015), who found that environmental attitudes can indeed influence heritage perceptions.

Additionally, a higher trust in the future of one’s company was found to be weakly related to a lower support for a Dark Sky Park. This might be explained by concerns that policies related to a Dark Sky Park will have implications for business continuity. This is supported by research of Jones & Shaw (2012) who found the same concerns around a World Heritage inscription in Australia. Lastly, this research was able to confirm findings of Rasoolimanesh et al. (2015), who argued that envisioned benefits for tourism inside communities can increase support for Heritage Sites. Exactly this was found: a believe that a better visibility of the stars will produce opportunities for tourism, moderately strongly relates to an increased support for an official Dark Sky Park denomination.

(14)

However, not all possible explaining factors from the literature were confirmed. Unlike in studies conducted by Rodríguez‐Darias et al. (2016) and Pienaar et al. (2013), this research did not confirm the significance of socio-economic and demographic factors on perceptions, for example—although it is suspected that the inclusion of farmers-only might also have caused this insignificance. However, the general inconclusiveness of results in existing literature seems to point towards the importance of context. Generalizations about heritage-making processes should therefore be made very carefully (See also Rasoolimanesh et al., 2015). Indeed, it should be expected that the intensive agricultural tradition in the Netherlands is an exceptional case and not necessarily representative for other areas.

Further research might therefore be advised to explore qualitative approaches. This way, deeper meanings might be uncovered for the relations found in the literature and this specific case study of the Dutch Wadden Sea region. These deeper meanings may be less dependent on context. Moreover, for a consideration of all possible views, follow-up studies should not be limited to dairy farmers alone. Another important weakness of this study is the random sampling technique which might have caused biased results. This is inherent to the geographically uneven distribution of the population and led to an under-representation of e.g. the Wadden Sea islands. Together with an issue in the used address file, this resulted in a weak geographical representativeness of the data.

Although this study did not statistically confirm a relation between a lack in participative-

governance and opinions about the Dark Sky initiative, on average respondents did not feel included in the decision-making process and did not feel treated as an equal party in plans for their region. They also repeatedly expressed this, possibly best summarized by the following commentary: ‘’Please talk with farmers, not about them’’. These feelings are potentially problematic, as the political process of building consensus in stakeholder groups has been described as being of key importance to a

successful and sustainable destination branding of heritage sites (Krauss, 2005; Ryan & Silvanto, 2010; Senil, 2011; Jones & Shaw, 2012). They are also interesting since the Dark Sky initiative is presented as a bottom-up process (Waddenzee, 2018). Which governance approaches are best for current heritage problems and to what extent they are actually implemented might therefore also be interesting for further research.

(15)

6. References

Blichfeldt, B.S. & Halkier, H. (2014). Mussels, tourism and community development: A case study of place branding through food festivals in rural North Jutland, Denmark. European Planning Studies, 22(8), 1587-1603.

Charlier, B. & Bourgeois, N. (2013). Half the park is after dark. L’Espace géographique, 42(3), 200- 212.

Clafis (2018). Overzicht maatregelen lichtreductie Dark Sky Werelderfgoed Waddengebied.

Clifford, N., Cope, M., Gillespie, T., French, S. & Valentine, G. (2016) Getting started in geographical research: How this book can help. In N. Clifford, M. Cope, T. Gillespie & S. French (Ed.), Key

Methods in Geography (pp. 4-16) London: SAGE Publications Ltd.

Collison, F. M. & Poe, K. (2013). “Astronomical tourism”: The astronomy and dark sky program at Bryce Canyon National Park. Tourism Management Perspectives, 7, 1-15.

Gwiazdzinski, L. (2002). The territories of the shadow: think the city, think the night. In Delegation to Regional Planning and Regional Action (DATAR), 5, 21-25.

Jones, R. E. & Dunlap, R. E. (1992). The social bases of environmental concern: Have they changed over time? Rural Sociology, 57(1), 28-47.

Jones, R. & Shaw, B. (2012). Thinking locally, acting globally? stakeholder conflicts over UNESCO world heritage inscription in Western Australia. Journal of Heritage Tourism, 7(1), 83-96.

Krauss, W. (2005). The natural and cultural landscape heritage of Northern Friesland, International Journal of Heritage Studies, 11(1), 39-52.

Liburd, J. J. & Becken, S. (2017). Values in nature conservation, tourism and UNESCO world heritage site stewardship. Journal of Sustainable Tourism, 25(12), 1719-1735.

Milfont, T. L. & Duckitt, J. (2010). The environmental attitudes inventory: A valid and reliable measure to assess the structure of environmental attitudes. Journal of Environmental Psychology, 30(1), 80-94.

Nogué, J., San Eugenio Vela, J. de, & Govers, R. (2017). Visual landscape as a key element of place branding. Journal of Place Management and Development, 10(1), 23-44.

McLafferty S.L. (2016) Conducting questionnaire surveys. In N. Clifford, M. Cope, T. Gillespie & S.

French (Ed.), Key Methods in Geography (pp. 129-140) London: SAGE Publications Ltd.

Olson, J. L. & Munroe, D. K. (2012). Natural amenities and rural development in new urban-rural spaces. Regional Science Policy & Practice, 4(4), 355-371.

Pienaar, E. F., Lew, D. K. & Wallmo, K. (2013). Are environmental attitudes influenced by survey context? an investigation of the context dependency of the new ecological paradigm (NEP) scale.

Social Science Research, 42(6), 1542-1554.

Piolle, X. (1993). The mountain elsewhere ‘’privileged city dwellers’’. Records of the Alpine Geography Journal, 11, 107-111.

Poria, Y. & Ashworth, G. (2009). Heritage Tourism—Current resource for conflict. Annals of Tourism Research, 36(3), 522-525.

(16)

Porter, B. W. & Salazar, N. B. (2005). Heritage tourism, conflict, and the public interest: An introduction. International Journal of Heritage Studies, 11(5), 361-370.

Pritchard, S. B. (2017). The trouble with darkness: NASA’s suomi satellite images of earth at night.

Environmental History, 22(2), 312-330.

Rasoolimanesh, S. M., Jaafar, M., Kock, N. & Ramayah, T. (2015). A revised framework of social exchange theory to investigate the factors influencing residents' perceptions. Tourism Management Perspectives, 16, 335-345.

Rodríguez‐Darias, A., Santana‐Talavera, A. & Díaz‐Rodríguez, P. (2016). Landscape perceptions and social evaluation of Heritage‐Building processes. Environmental Policy and Governance, 26(5), 394- 408.

Rosario, R. (2007). Places worth keeping. In Jones, R. & Shaw, B. (2012). Thinking locally, acting globally? stakeholder conflicts over UNESCO world heritage inscription in Western Australia.

Journal of Heritage Tourism, 7(1), 83-96.

Ryan, J. & Silvanto, S. (2010). World heritage sites: The purposes and politics of destination branding.

Journal of Travel & Tourism Marketing, 27(5), 533-545.

Senil, N. (2011). Reorganising space and time: A comparative analysis of the heritage development of the Chauvet cave and the Millau viaduct. Journal of Alpine Research, 2(99), 8.

Sijtsma, F. J., Daams, M. N., Farjon, J. M. J. & Buijs, A. E. (2012). Deep feelings around a shallow coast. A spatial analysis of tourism jobs and the attractivity of nature in the dutch wadden area. Ocean

& Coastal Management, 68, 138-148.

Taylor, L. Case Study Methodology. In N. Clifford, M. Cope, T. Gillespie & S. French (Ed.), Key Methods in Geography (pp. 581-594) London: SAGE Publications Ltd.

Waddenzee.nl (2018). Beleid. Retrieved on 1 March 2019 from https://www.waddenzee.nl/themas/dark-sky/.

Waddenzee.nl (2016). Annual Report 2016 Dark Sky Park De Boschplaat.

Walker, P. & Fortmann, L. (2003). Whose landscape? A political ecology of the 'exurban' sierra.

Cultural Geographies, 10(4), 469-491.

Woods, M. (2009). Rural geography: Blurring boundaries and making connections. Progress in Human Geography, 33(6), 849-858.

You, W., He, D., Hong, W., Liu, C., Wu, L., Ji, Z., & Xiao, S. (2014). Local people's perceptions of participating in conservation in a heritage site: A case study of the wuyishan scenery district cultural and natural heritage site in South-eastern China. Natural Resources Forum, 38(4), 296-307.

Zuo, B., Gursoy, D. & Wall, G. (2017). Residents’ support for red tourism in China: The moderating effect of central government. Annals of Tourism Research, 64, 51-63.

Cover photo: taken by author (October 2017). On the foreground: cow barns. In the middle: soccer fields. The skyline of Groningen is visible in the background. The glow from the city is reflected from the clouds. Photo was taken at a 25 km distance from Groningen.

(17)

7. Appendix

(18)

7.1 Letter to respondents (Dutch)

(19)

7.2 Questionnaire Survey (Dutch)

(20)

(21)

(22)

(23)

(24)

7.3 Output Paired Samples T-Test

Results for: Limit_LE_Wad and ExpandDS_Park

Paired Samples Statistics

Mean N Std. Deviation Std. Error Mean

Pair 1 Limit_LE_Wad 6.88 56 2.690 .360

ExpandDS_Park 5.38 56 3.084 .412

Paired Samples Correlations

N Correlation Sig.

Pair 1 Limit_LE_Wad &

ExpandDS_Park

56 .823 .000

Paired Samples Test Paired Differences

t df

Sig. (2- tailed) Mean

Std.

Deviation

Std.

Error Mean

95% Confidence Interval of the

Difference Lower Upper Pair

1

Limit_LE_Wad - ExpandDS_Park

1.500 1.758 .235 1.029 1.971 6.385 55 .000

(25)

7.4 Overview of variables in statistical analyses

Variable Name Survey question number

Variable label Measurement level

Man 2 Gender Dummy. Women as

reference category

Age 3 Age Years

DidYouKnowAbout Dummy 1 Did the respondent know about the intention to work towards a reduction of the light-emissions in the Wadden Sea area and environment?

Dummy. Yes/No

Cult_vs_Nat 8 How would you describe the

Wadden Sea area and its surrounding region?

Likert scale 1-10. 1=Mainly natural heritage 10=Mainly cultural heritage

QualityStars_M 9 Being able to see the starry sky is an important quality of the municipality in which my company is situated.

Likert scale 1-10. 1=Totally disagree 10=Totally agree

MoreTourist_M 14 It is a good idea to attract more tourists to my municipality.

Likert scale 1-10. 1=Totally disagree 10=Totally agree AddedValTS_M 13 It has added value for tourism in my

municipality if the starry sky would be better visible.

Likert scale 1-10. 1=Totally disagree 10=Totally agree Trust_Future_C 15 I have trust in the future of my dairy

company.

Likert scale 1-10. 1=Totally disagree 10=Totally agree Easy_Rules 16 I think it is easy to implement rules

for the agricultural sector in my company.

Likert scale 1-10. 1=Totally disagree 10=Totally agree IncludedDecisions_LE 18 I feel included in the decision

processes concerning the reducing of light-emissions in my region.

Likert scale 1-10. 1=Totally disagree 10=Totally agree Equal_Party_Region 17 I feel treated as an equal party in

plans for my region.

Likert scale 1-10. 1=Totally disagree 10=Totally agree Possi_Reduce_LE 19 It is possible to further reduce light-

emissions in my company.

Likert scale 1-10. 1=Totally disagree 10=Totally agree Measure_Willing_cost 22 I am prepared to take measures to

reduce light-emissions in my company if that would cost money.

Likert scale 1-10. 1=Totally disagree 10=Totally agree Nonsense_Force 20 It is nonsensical to obligate farmers

to reduce their light-emissions.

Likert scale 1-10. 1=Totally disagree 10=Totally agree EnvironOrganis

(Environmental Attitudes Inventory)

23 I would like to join an environmental organisation.

Likert scale 1-10. 1=Totally disagree 10=Totally agree RiverSports (Environmental

Attitudes Inventory)

24 One of the most important reasons to keep rivers and lakes clean is for the ability for people to enjoy water sports.

Likert scale 1-10. 1=Totally disagree 10=Totally agree

ScienceSolves

25 Science will be able to solve our environmental problems.

Likert scale 1-10. 1=Totally disagree 10=Totally agree AbusingEnviron

26 Humans are severely abusing the environment.

Likert scale 1-10. 1=Totally disagree 10=Totally agree SaveResources

27 I try to save resources whenever I can.

Likert scale 1-10. 1=Totally disagree 10=Totally agree Jobs_VS_Environm

28 The protection of people’s jobs is more important than the protection of the environment.

Likert scale 1-10. 1=Totally disagree 10=Totally agree GroningenDummy 5 Residence in province of Groningen Reference category = ELSE

HBOWO Dummy 4 Higher level of education Dummy with VMBO, HAVO,

VWO and MBO as reference category

(26)

7.4 Output Multiple Linear Regression

Results for: Limit_LE_Wad

Descriptive Statistics

Mean Std. Deviation N

Limit_LE_Wad 6.84 2.700 55

Man .67 .474 55

Age 50.93 12.525 55

DidYouKnowAbout Dummy .35 .480 55

Cult_vs_Nat 4.85 2.321 55

QualityStars_M 6.96 2.775 55

MoreTourist_M 5.64 2.648 55

AddedValTS_M 5.44 2.892 55

Trust_Future_C 7.87 1.816 55

Easy_Rules 4.91 2.413 55

IncludedDecisions_LE 3.47 2.356 55

Equal_Party_Region 3.80 2.189 55

Possi_Reduce_LE 3.40 2.385 55

Measure_Willing_cost 4.25 2.562 55

Nonsense_Force 6.55 2.924 55

EnvironOrganis 2.69 2.243 55

RiverSports 3.60 2.346 55

ScienceSolves 4.07 2.176 55

AbusingEnviron 5.80 2.620 55

SaveResources 8.25 1.713 55

Jobs_VS_Environm 5.05 2.121 55

GroningenDummy .31 .466 55

HBOWO Dummy .40 .494 55

Model Summary^b

Model R

R Square

Adjusted R Square

Std. Error of the Estimate

Change Statistics R Square

Change

F

Change df1 df2

Sig. F Change

1 .918^a .843 .734 1.391 .843 7.790 22 32 .000

a. Predictors: (Constant), HBOWO Dummy, Man, AbusingEnviron, Measure_Willing_cost, DidYouKnowAbout Dummy, MoreTourist_M, GroningenDummy, Jobs_VS_Environm, Possi_Reduce_LE, ScienceSolves, Cult_vs_Nat, RiverSports, SaveResources, Age, Nonsense_Force, EnvironOrganis, Trust_Future_C, Equal_Party_Region, Easy_Rules, QualityStars_M, IncludedDecisions_LE, AddedValTS_M

b. Dependent Variable: Limit_LE_Wad

(27)

Coefficients^a

a. Dependent Variable: Limit_LE_Wad Model

Unstandardized Coefficients

Standardized Coefficients

t Sig.

B Std. Error Beta

Semi- Partial Correlation

VIF (Collinearity

statistics)

1 (Constant) 8.399 2.337 3.594 .001

Man .467 .530 .082 .882 .384 .062 1.757

Age .003 .025 .015 .130 .897 .009 2.811

DidYouKnowAbout Dummy

-.816 .520 -.145 -1.568 .127 -.110 1.740

Cult_vs_Nat .065 .099 .056 .658 .515 .046 1.478

QualityStars_M .623 .125 .640 4.977 .000 .349 3.368

MoreTourist_M .112 .098 .109 1.139 .263 .080 1.881

AddedValTS_M .220 .146 .236 1.509 .141 .106 4.970

Trust_Future_C -.363 .180 -.244 -2.014 .052 -.141 2.995

Easy_Rules -.110 .127 -.098 -.864 .394 -.061 2.628

IncludedDecisions_LE -.188 .147 -.164 -1.282 .209 -.090 3.350

Equal_Party_Region .152 .155 .124 .984 .332 .069 3.208

Possi_Reduce_LE .150 .100 .133 1.499 .144 .105 1.597

Measure_Willing_cost -.053 .090 -.050 -.581 .565 -.041 1.500

Nonsense_Force -.070 .097 -.076 -.725 .474 -.051 2.245

EnvironOrganis .113 .117 .094 .962 .343 .067 1.933

RiverSports -.262 .099 -.228 -2.662 .012 -.187 1.492

ScienceSolves .029 .120 .023 .240 .812 .017 1.891

AbusingEnviron -.319 .117 -.310 -2.721 .010 -.191 2.639

SaveResources -.093 .153 -.059 -.605 .549 -.042 1.924

Jobs_VS_Environm -.281 .134 -.221 -2.096 .044 -.147 2.257

GroningenDummy .399 .529 .069 .753 .457 .053 1.702

HBOWO Dummy -.389 .482 -.071 -.807 .425 -.057 1.582

(28)

7.5 Output Multiple Linear Regression

Results for: ExpandDS_Park

Descriptive Statistics

Mean Std. Deviation N

ExpandDS_Park 5.31 3.072 55

Man .67 .474 55

Age 50.93 12.525 55

DidYouKnowAbout Dummy .35 .480 55

Cult_vs_Nat 4.85 2.321 55

QualityStars_M 6.96 2.775 55

MoreTourist_M 5.64 2.648 55

AddedValTS_M 5.44 2.892 55

Trust_Future_C 7.87 1.816 55

Easy_Rules 4.91 2.413 55

IncludedDecisions_LE 3.47 2.356 55

Equal_Party_Region 3.80 2.189 55

Possi_Reduce_LE 3.40 2.385 55

Measure_Willing_cost 4.25 2.562 55

Nonsense_Force 6.55 2.924 55

EnvironOrganis 2.69 2.243 55

RiverSports 3.60 2.346 55

ScienceSolves 4.07 2.176 55

AbusingEnviron 5.80 2.620 55

SaveResources 8.25 1.713 55

Jobs_VS_Environm 5.05 2.121 55

GroningenDummy .31 .466 55

HBOWO Dummy .40 .494 55

Model Summary^b

Model R

R Square

Adjusted R Square

Std. Error of the Estimate

Change Statistics R Square

Change

F

Change df1 df2

Sig. F Change

1 .924^a .854 .754 1.523 .854 8.535 22 32 .000

a. Predictors: (Constant), HBOWO Dummy, Man, AbusingEnviron, Measure_Willing_cost, DidYouKnowAbout Dummy, MoreTourist_M, GroningenDummy, Jobs_VS_Environm, Possi_Reduce_LE, ScienceSolves, Cult_vs_Nat, RiverSports, SaveResources, Age, Nonsense_Force, EnvironOrganis, Trust_Future_C, Equal_Party_Region, Easy_Rules, QualityStars_M, IncludedDecisions_LE, AddedValTS_M

b. Dependent Variable: ExpandDS_Park

(29)

Coefficients^a

Model

Unstandardized Coefficients

Standardized Coefficients

t Sig.

B Std. Error Beta

Semi- Partial Correlation

VIF (Collinearity

statistics)

1 (Constant) 4.353 2.559 1.701 .099

Man -.451 .580 -.069 -.777 .443 -.052 1.757

Age .025 .028 .103 .910 .370 .061 2.811

DidYouKnowAbout Dummy

-.868 .570 -.136 -1.524 .137 -.103 1.740

Cult_vs_Nat .031 .109 .023 .282 .780 .019 1.478

QualityStars_M .156 .137 .141 1.140 .263 .077 3.368

MoreTourist_M .055 .107 .047 .511 .613 .034 1.881

AddedValTS_M .643 .160 .605 4.026 .000 .272 4.970

Trust_Future_C -.478 .197 -.283 -2.423 .021 -.163 2.995

Easy_Rules .103 .139 .081 .738 .466 .050 2.628

IncludedDecisions_LE .144 .161 .111 .896 .377 .060 3.350 Equal_Party_Region -.018 .170 -.013 -.106 .916 -.007 3.208

Possi_Reduce_LE .081 .110 .063 .735 .468 .050 1.597

Measure_Willing_cost -.019 .099 -.016 -.190 .851 -.013 1.500

Nonsense_Force .044 .106 .042 .411 .683 .028 2.245

EnvironOrganis -.038 .128 -.027 -.293 .771 -.020 1.933

RiverSports -.306 .108 -.234 -2.835 .008 -.191 1.492

ScienceSolves -.028 .131 -.020 -.212 .833 -.014 1.891

AbusingEnviron -.057 .128 -.048 -.442 .661 -.030 2.639

SaveResources -.016 .168 -.009 -.095 .925 -.006 1.924

Jobs_VS_Environm -.115 .147 -.079 -.784 .439 -.053 2.257

GroningenDummy -.331 .580 -.050 -.571 .572 -.039 1.702

HBOWO Dummy .103 .527 .017 .195 .847 .013 1.582

a. Dependent Variable: ExpandDS_Park

Dark Sky Heritage