Disaster risk perception and local resilience near the “Duboko” landfill: Challenges of governance, management, trust, and environmental communication in Serbia

1.1 Literature review

Perceived severity of environmental threats – rather than mere awareness of their existence – has emerged as a significant factor driving proactive public behaviour, including waste minimisation and proper waste segregation [64]. Therefore, strategies for risk communication concerning landfill management should consider critical demographic factors, especially age, to encourage meaningful public involvement in mitigation initiatives [64]. Among demographic factors, age and gender are particularly influential in shaping how individuals perceive environmental risks and threats associated with landfills.

Gender plays a critical yet complex role in risk perception. Numerous studies have shown that women, particularly white women, tend to perceive greater environmental threats from landfills and hazardous waste sites compared to men [65,66,67,68]. In contrast, white men consistently report the lowest perceived levels of risk, while non-white women and men display heightened and more similar levels of concern [65,66,67,68]. This phenomenon is often known as the “white male effect,” in which white men tend to minimise environmental risks more than other demographic groups [32,34]. Importantly, the influence of gender on risk perception does not operate in isolation but intersects with other factors such as race, class, and social roles. For example, women of colour have been recognised as active agents in resisting environmental injustices, emphasising the need to examine risk perception through an intersectional lens [67,69]. Furthermore, gendered perceptions are shaped by power relations, access to knowledge, and socially defined community roles, which in turn affect how risks are internalised and acted upon [70,71]. In some contexts, particularly within non-white or economically marginalised populations, gender-based differences are less pronounced, with both men and women expressing similar levels of environmental concern [65,67]. Environmental justice offers a vital perspective for understanding how marginalised communities perceive environmental and disaster risks. It highlights both the distributional aspect, which concerns who carries ecological burdens, and the procedural aspect, referring to participation in decision-making processes [15,16,72–79]. In post-socialist settings, these injustices tend to be exacerbated by transitional governance, limited public involvement, and institutional weaknesses [16,77].

In communities directly affected by landfills, research has even found cases where gender differences in environmental concern or activism are negligible, suggesting that local dynamics may override general trends [8]. Among migrant populations, ecological risks have been shown to affect physical and mental health differently by gender: women more frequently report impacts on physical health, while men are more likely to experience mental health effects related to environmental stressors [80].

Age is similarly crucial in shaping risk perceptions and behavioural responses. Younger individuals, especially those raising children, consistently express greater concern regarding landfill risks and are more inclined to take action to mitigate them [8,81]. Conversely, older adults, especially those aged 50 and above, often exhibit a lower willingness to support or fund initiatives aimed at mitigating long-term environmental health risks. This trend suggests a diminished sense of urgency or personal susceptibility [81,82]. Some studies suggest that age may have a more substantial influence on environmental risk perception than other sociocultural variables such as profession or information sources, potentially due to generational differences in ecological awareness and values [81].

In addition to demographic characteristics, public trust has been widely recognised as a pivotal factor in shaping community responses to landfill siting and waste management initiatives [83–87]. Community attitudes toward environmental projects are often grounded not only in scientific assessments of risk but also in the perceived credibility, transparency, and integrity of the institutions and decision-makers involved. A general scepticism toward technological solutions and expert knowledge remains a significant challenge in managing environmental risks, including those related to landfills [85,88]. Trust levels are further shaped by personal experience, economic dependency, and willingness to accept certain risks in exchange for financial benefits. For instance, communities that derive economic advantage from landfill operations may report higher trust in managing institutions and greater tolerance toward perceived risks [89]. Conversely, communities with a history of exclusion or negative interactions with waste management authorities tend to display deep scepticism and opposition to new projects [84]. In such cases, resistance is not necessarily irrational but often rooted in a lack of shared values or transparency. Indeed, scepticism can catalyse more inclusive and negotiated project outcomes [84].

Public concern about landfill issues does not automatically arise; instead, it is often triggered when waste management becomes highly visible or starts to affect daily life. The salience of landfill-related risks and the extent to which they are perceived as urgent or threatening depend on local socio-political context and public awareness [86]. For the successful implementation of waste management policies – particularly those involving public-private partnerships – citizen confidence in government oversight, the competence of developers, and the reliability of waste treatment technologies is crucial [87]. State ownership and clearly defined long-term liability mechanisms can also help mitigate public concerns. Broader trust in policy frameworks is influenced by the presence of economic incentives (e.g. landfill taxes), institutional stability, and effective coordination among government bodies [90]. Even in technologically advanced systems, public distrust may persist due to lingering concerns about environmental and health safety, revealing a disconnect between scientific claims and community sentiment [85].

Education is another key determinant of risk perception and environmental behaviour. Individuals with higher educational attainment are more likely to recognise the ecological and health impacts of landfill sites and to support responsible waste management policies [91–93]. University students, particularly those pursuing studies in technical or environmental fields, display progressively deeper understanding and more critical perspectives on landfill siting and operations as they advance academically [91]. Education has also been linked to greater support for stakeholder participation in environmental decision-making, as well as stronger endorsement of collective accountability for waste-related issues [91,92]. More educated individuals are also more aware of the broader social and ecological consequences of poor waste management, such as traffic congestion, aesthetic degradation, and contamination from leachate [93].

Although education significantly improves knowledge and attitudes, its effect on actual behaviour is less straightforward. Educational campaigns and awareness initiatives can increase consciousness about waste reduction, but they do not always result in behavioural change. Additional measures – such as incentives, institutional support, and culturally sensitive outreach – are often required to translate knowledge into action [55]. Nonetheless, along with age and gender, education remains one of the most consistent predictors of how individuals and communities perceive and respond to landfill-related risks [54].

In contrast, there is limited evidence linking employment status directly to awareness or perception of landfill-related risks. Existing studies primarily explore the impact of employment on waste generation and exposure rather than on risk awareness or environmental attitudes [94–98]. This highlights a potential area for future research.

Household-level factors also influence risk perception and behaviour. Social context, leadership roles, and family dynamics contribute significantly to how environmental information is processed and acted upon [99–102]. Households in which leaders model environmentally responsible behaviour and encourage waste reduction tend to exhibit greater awareness and higher levels of pro-environmental behaviour. One study found that family leadership accounted for nearly 28.4% of the variance in household awareness and engagement with waste management issues, while motivation and education explained the rest [99]. Despite general awareness of the health hazards associated with landfills, this awareness does not always lead to active support for waste management policies [100,102]. However, households that display higher awareness – particularly those with children or younger members – are more likely to adopt positive behaviours and demonstrate stronger engagement with local environmental initiatives [8,100,102].

In sum, landfill risk perception is shaped by a dynamic interplay of demographic characteristics, trust in institutions, educational background, and social context. These factors not only influence how communities perceive environmental risks but also determine the extent to which they are willing to take action in response to them. Understanding these interrelationships is crucial for developing inclusive, responsive, and practical strategies in environmental and disaster risk management.

2.1 Hypotheses

This study combines theories of risk perception, frameworks on institutional trust, and socio-ecological models of community resilience to better understand what shapes how residents living near the “Duboko” regional landfill perceive environmental risks and their confidence in public institutions. The research hypotheses are designed to reflect the potential influence of demographic, socio-economic, geographic, and informational factors.

Hypothesis 1. Demographic characteristics have a notable effect on how people perceive risk and trust institutions. It is anticipated that factors such as gender and age are closely tied to how individuals experience and respond to environmental risks. Specifically, it is expected that women and older adults (those 45 and above) will report feeling more concerned about these risks and will have lower levels of trust in public authorities and emergency services.

Hypothesis 2. Socio-economic background plays a key role in shaping views on disaster readiness, safety, and resilience. Participants with higher levels of education (including university and postgraduate degrees), secure employment, higher household incomes, and smaller household sizes are expected to express greater trust in institutional responses and exhibit more emotional stability. On the other hand, those with only elementary education, those who are unemployed, or those earning less than 215 EUR per month are expected to report higher levels of anxiety, feel less in control, and place less trust in authorities.

Hypothesis 3. Where someone lives and how close they are to the landfill are likely to influence their sense of risk and vulnerability. Living closer to the landfill – especially within 5 km – and residing in rural areas is expected to be associated with a greater concern about pollution, public health, and overall environmental degradation. These individuals are also likely to report lower trust in institutions compared to people living farther away or in more urban settings.

Hypothesis 4. Housing conditions and property ownership are linked to feelings of safety and psychological reactions. People who live in homes they own or in property owned by their family feel more personally at risk from environmental harm. They are also expected to show more concern about losing property value or facing long-term consequences than those living in rented spaces.

Hypothesis 5. Access to information and civic involvement are essential predictors of trust and preparedness. Those who are well-informed about landfill-related issues and actively involved in environmental or community efforts are expected to feel better prepared, more optimistic, and more positive about the transparency of institutions. In contrast, individuals with limited access to reliable information or who are less civically engaged may feel more distrustful, less empowered, and more emotionally affected by the potential risks tied to the landfill.

This study draws on insights from risk perception theory, institutional trust theory, and existing literature on the social determinants of safety to propose several hypotheses examining how young individuals perceive security through different lenses. These hypotheses, grouped by independent variables, can be classified into three primary categories: demographic attributes, socio-economic status, and levels of institutional trust and exposure.

2.2 Sample characteristics

Using a multi-stage random sampling method, a total of 1,180 respondents were surveyed from nine local communities in Serbia: Čačak, Užice, Požega, Lučani, Kosjerić, Čajetina, Bajina Bašta, Ivanjica, and Arilje (Table 1). The gender distribution is balanced, with 49.2% of respondents identifying as male and 50.5% as female. The largest age group consists of individuals aged 30–44 (32.9%), while those over 60 years make up 17.8% of the sample. In terms of education, most respondents have completed secondary education (55%), while 26.5% hold higher education degrees, and only 2.8% possess a master’s or doctoral degree. Regarding their place of residence, the majority of respondents (60.6%) live in urban areas, while 17% reside in suburban or semi-rural areas, and 22.4% live in rural settlements. Regarding the distance from the regional landfill “Duboko,” most respondents (59.07%) live more than 10 km away, while only 8.73% live within 1 km of it. In terms of employment, 64.5% of respondents are employed (including those with occasional or seasonal jobs), 16% are unemployed, 15.7% are retired, and 3.8% are students. Most participants reside in family-owned properties (40.3%), while 32% own their property independently, and 27.7% live in rented housing. The household structure indicates that the majority of respondents live in households with three members (34.7%) or four or more members (26.2%). The most represented monthly income bracket is between €515 and €770 (35.8%), followed by 28.4% earning between €215 and €515. A below-average income (under €215) is reported by 9.2% of respondents, while 12% have a monthly income exceeding €1,280.

2.3 Questionnaire design

To better understand how people living near the regional “Duboko” landfill perceive environmental and disaster-related risks, a structured field survey was carefully planned and conducted. The project was titled “Disaster Risk Perception and Community Resilience around the ‘Duboko’ Landfill: Public Trust, Vulnerability, and Environmental Communication in Serbia.” Its central focus was to explore how the public perceives environmental hazards, how much they trust institutions, and how aware and psychologically affected they are by the landfill’s presence, all within the broader framework of waste management and community resilience.

The data collection was conducted entirely through face-to-face interviews, utilising a paper-based questionnaire in selected nearby communities. Before participating, each respondent was informed about the study’s purpose and reassured that their answers would be kept confidential and participation was voluntary.

The questionnaire consisted of seven sections, totalling 52 closed-ended questions and one open-ended item. These were organised as follows:

1. Demographic and socioeconomic characteristics (11 items) – This section collected essential background information, including gender, age, education, employment status, household income, household size, property ownership, and distance from the landfill. These variables formed the basis for interpreting how different groups perceive risks.

2. Risk perception and environmental impact (6 items) – This section measured concerns about the landfill’s effects on air, water, and soil, as well as its perceived impact on health, property values, and community well-being.

3. Psychological responses and perceived control (4 items) – This section of the survey examined emotional reactions, such as stress and optimism, as well as perceptions of personal agency and the effectiveness of individual actions.

4. Environmental values and engagement (4 items) – Focused on personal attitudes toward environmental protection, this section also examined how often people participate in eco-friendly activities and how much they value nature conservation.

5. Perception of disaster and potential threats (4 items) – This part assessed how worried people are about possible environmental or technological accidents – like fires or financial consequences – and how much importance they place on being informed about such risks.

6. Information, media, and communication (4 items + 1 open-ended question) – This section reviewed how informed citizens feel, their satisfaction with media coverage, trust in official information, and how they prefer institutions to engage with them.

7. Institutional trust and transparency (19 items) – The most detailed section examined the extent to which people trust various authorities – local, national, environmental, and scientific – as well as their views on transparency, institutional effectiveness, and whether improved communication could alleviate public concerns.

A five-point Likert scale was used throughout the questionnaire, allowing respondents to indicate the strength of their opinions, whether they were assessing trust, concern, or agreement, depending on the topic. The questionnaire underwent several revisions, guided by expert feedback and small-scale pilot tests, to ensure that the questions were clear and thematically coherent. Reliability was confirmed through Cronbach’s alpha scores, which ranged from 0.80 to 0.88, indicating solid internal consistency. Specialists also verified the content and construct validity, as well as the existing models related to disaster risk and environmental communication. Overall, this tool provides a solid foundation for analysing public attitudes and shaping more effective risk communication in communities that live in the shadow of large-scale waste facilities, such as the “Duboko” landfill.

2.4 Analyses

To gain a more precise and more nuanced understanding of how nearby communities perceive risks associated with the “Duboko” regional landfill – including their level of trust in institutions, emotional responses, and willingness to get involved – researchers employed a layered and comprehensive statistical approach. All analyses were conducted using IBM SPSS Statistics (Version 26), with a significance level set at the standard two-tailed p-value threshold of less than 0.05.

The first phase of analysis focused on descriptive statistics, which helped sketch out broad patterns in public opinion. This included insights into how people view environmental damage, health hazards, institutional transparency, and the landfill’s overall effect on daily life. To make sense of the data, researchers calculated averages, standard deviations, and frequency distributions, providing a solid overview of the sample.

Next, the study moved into the more comparative territory, using statistical tools to explore whether personal and socioeconomic background factors – such as gender, education, where people live, and their assessment of their financial status – had any noticeable impact on risk perception or trust in institutions. Independent-sample t-tests were used to detect any gender-based differences. In contrast, one-way ANOVA was used to assess variations across groups defined by education, urban vs rural location, and self-assessed income level. In situations where equal variance assumptions did not hold up, alternative methods like Welch’s ANOVA and the Brown–Forsythe test ensured that the results stayed valid. Whenever significant results emerged, post hoc tests such as Tukey’s HSD were applied to figure out exactly which groups differed.

To better understand how various factors interacted, the research team used Pearson’s correlation coefficients. These calculations revealed the strength and direction of connections between variables, such as institutional trust, financial background, and public awareness, forming a crucial foundation for the predictive models that followed.

To pinpoint which factors most strongly influenced how people perceived risk and trusted institutions, a set of multiple linear regression models was built. These followed a step-by-step or hierarchical structure, with demographic factors coming first, followed by economic variables, and finally, trust and informational exposure. This layering enabled researchers to determine the extent to which each block of variables contributed to the final explanation. Key diagnostics, such as checking for multicollinearity through VIF and tolerance values, ensured that the models were statistically sound.

To make the data more manageable and easier to interpret, six composite variables were created (Table 2). Each one corresponded to a central theme covered by the survey

• Perceived risk perception and environmental impact of the landfill,

• Psychological responses and sense of control,

• Environmental values and civic involvement,

• Perceptions of disaster risk and potential threats,

• Public awareness and communication,

• Trust in institutions and transparency.

Each of these variables was calculated as an average of several closely related survey items, all rated on a five-point Likert scale – from strong disagreement or low perception to strong agreement or high perception. This approach provided continuous variables that could be used in more in-depth statistical analysis, facilitating easier comparisons across different domains. To ensure the reliability of these groupings, Cronbach’s alpha was used to test internal consistency, with all six scales demonstrating solid reliability (α values ranging from 0.80 to 0.88). All statistical procedures were conducted in accordance with established ethical standards. The study adhered to the principles outlined in the Declaration of Helsinki and received ethical approval from the Institutional Committee of the Scientific–Professional Society for Disaster Risk Management under protocol 003/2025.

All participants were informed about the purpose and anonymity of the survey and provided their consent before participation. Participation was entirely voluntary, and respondents were assured of the confidentiality of their responses.

3.1 Descriptive analysis

The descriptive statistics are presented across six thematic sections, aligned with the structure of the questionnaire: (1) demographic and socioeconomic characteristics; (2) risk perception and environmental impact; (3) psychological responses and perceived control; (4) environmental values and engagement; (5) perception of disaster and potential threats; and (6) information, media, and communication. Together, these categories provide a detailed overview of participants’ profiles, their concerns related to landfills, emotional and behavioural reactions, environmental attitudes, perceived risks, and the effectiveness of institutional communication.

Notably, the highest average score emerged for the statement, “Local communities near the landfill are particularly vulnerable to risks associated with it” (Mean (M) = 3.24, Standard deviation (SD) = 1.25), highlighting a strong perception among respondents that those living closest to the “Duboko” landfill face elevated environmental risks. Similarly, high scores were observed in responses to items regarding long-term health hazards (M = 3.16, SD = 1.13) and concerns about soil pollution (M = 3.10, SD = 1.17), indicating widespread public unease about the landfill’s potential long-term effects on both ecological and human health.

Moderate levels of concern were also reflected in scores related to river contamination (M = 3.07, SD = 1.16) and air quality decline (M = 3.03, SD = 1.23), suggesting a broader awareness of environmental impacts stemming from the landfill.

By contrast, the lowest average score appeared for the statement, “Potential landfill issues could affect the value of your property” (M = 2.74, SD = 1.76), indicating that while environmental and health issues weigh heavily in public opinion, economic concerns – especially regarding personal property – are viewed as less significant. The overall scale average (M = 3.06, SD = 1.30) reflects a generally elevated sense of risk and awareness about the landfill’s environmental repercussions (Table 3, Figure 1).

Figure 1

Descriptive statistics of youth perceptions of personal safety.

The highest average score in this group was linked to the statement, “You can influence solving the ‘Duboko’ landfill problem” (M = 3.07, SD = 2.24), indicating a moderate sense of personal agency among respondents. However, the relatively large standard deviation points to considerable variation in how individuals perceive their influence over environmental outcomes.

At the opposite end, the lowest agreement was recorded for the item, “How stressed are you about the current situation with the ‘Duboko’ landfill?” (M = 2.55, SD = 1.29), suggesting that while the issue is recognised, the emotional stress it triggers remains relatively low. Respondents also provided moderate ratings regarding their perception of how effective they feel their voice is in addressing the problem (M = 2.74, SD = 1.22) and their ability to remain optimistic about potential solutions (M = 2.78, SD = 1.07).

Taken together, the average score for this set of psychological indicators (M = 2.78, SD = 1.53) portrays a cautious equilibrium between perceived control, emotional involvement, and hopefulness. However, the broad range of responses reveals varied psychological engagement across the population (Table 4, Figure 2).

Figure 2

Descriptive statistics on psychological responses and perceived control.

The strongest agreement in this group was recorded for the statement, “Preserving the local ecosystem is essential for the quality of life in your community” (M = 3.16, SD = 1.10), closely followed by concern for environmental protection (M = 3.24, SD = 1.17) and support for local waste reduction efforts (M = 3.15, SD = 1.14). These results reflect a clear orientation toward valuing ecological preservation and sustainable community initiatives.

In contrast, the statement regarding “Participation in actions related to environmental protection” received the lowest mean score (M = 2.64, SD = 1.49), indicating that although pro-environmental attitudes are prevalent, actual engagement in related activities is less consistent. The higher standard deviation further suggests considerable differences in civic involvement across respondents.

Overall, the composite score for this set (M = 3.05, SD = 1.23) indicates broadly positive environmental values, although it also highlights a moderate gap between concern and direct action (Table 5, Figure 3).

Figure 3

Descriptive statistics on environmental values and engagement.

The highest average score within this group was associated with the statement, “Regular workshops on environmental protection would improve your understanding of landfill issues” (M = 4.20, SD = 0.93), indicating a strong perceived need for educational initiatives that enhance public knowledge and engagement. Nearly as high was the belief that “A fire at the landfill threatens your quality of life” (M = 3.88, SD = 1.11), indicating widespread concern over potential environmental emergencies.

Respondents also expressed considerable apprehension about potential ecological incidents (M = 3.81, SD = 1.21) and the economic consequences of landfill-related problems, including their impact on agriculture and tourism (M = 3.69, SD = 1.21). These findings suggest that the public recognises not only immediate environmental risks but also broader socioeconomic impacts associated with the landfill.

The overall mean score (M = 3.89, SD = 1.12) for this set points to a heightened awareness of both current and future threats from the “Duboko” landfill, reinforcing the public’s call for better preparedness, transparency, and environmental education (Table 6, Figure 4).

Figure 4

Descriptive statistics on the perception of disaster and potential threats.

The results revealed moderate levels of institutional confidence, with the highest mean score attributed to the statement “Satisfaction with the way the media reports on the landfill issues” (M = 2.82, SD = 1.22), indicating a slightly more favourable perception of media coverage compared to other information sources. Perceptions of being well-informed about the landfill (M = 2.75, SD = 1.21) and trust in the accuracy of information provided by local officials (M = 2.71, SD = 1.21) remained consistently moderate.

Although none of the items reflect strong institutional trust or confidence, the overall composite score (M = 2.76, SD = 1.21) suggests a cautious, possibly sceptical public stance toward the adequacy and credibility of information dissemination regarding the “Duboko” landfill issue. The uniformity of standard deviations across items further points to a stable but moderately low level of institutional trust within the sample (Table 7, Figure 5).

Figure 5

Descriptive statistics on perceptions of information, media, and communication.

Among the surveyed items, the highest average score was noted for the statement “Greater transparency of institutions would increase your trust” (M = 3.27, SD = 1.11), indicating a broad belief that openness plays a crucial role in strengthening public trust. Respondents also showed relatively strong agreement with the statement “Better public information would reduce citizens’ concerns” (M = 3.24, SD = 1.18), indicating that improved communication is perceived as a crucial way to alleviate public concerns about the landfill situation.

Moderate levels of trust were recorded for scientific institutions (M = 3.01, SD = 1.25), national environmental inspectors (M = 2.97, SD = 1.49), and government institutions (M = 2.97, SD = 1.26), suggesting a steady but cautious confidence in these bodies. Trust in local authorities was somewhat lower (M = 2.78, SD = 1.22), possibly reflecting reservations about their ability to manage the issue effectively.

Ratings for institutional clarity (M = 2.98, SD = 1.21) and transparency (M = 2.95, SD = 1.34) mirrored the overall trust pattern, further underscoring a moderate stance.

The overall composite score (M = 3.02, SD = 1.26) reflects a view that, while generally optimistic, it remains critical, highlighting the potential for institutions to build greater trust through more transparent communication and openness (Table 8, Figure 6).

Figure 6

Descriptive statistics on perceptions of institutional confidence and trust in security policy.

3.2 Correlations and effects of socio-demographic factors on disaster risk perception and community resilience

3.3 Predictors of perceived disaster risk and community resilience around the “Duboko” landfill: Regression model analysis

To examine how various socio-demographic, economic, and spatial factors influence residents’ perception of disaster risk and environmental impact, a multivariate linear regression analysis was conducted across six thematic dimensions: (1) risk perception and environmental impact, (2) psychological responses and perceived control, (3) environmental values and civic engagement, (4) perception of disaster and potential threats, (5) disaster preparedness and crisis response, and (6) information, media, and communication. Each model employed a consistent set of independent variables, including gender, education level, settlement type, proximity to the landfill, employment status, property ownership, and number of household members.

The results for each thematic domain are presented in Table 13. Before conducting the regression analyses, the assumptions of normality, linearity, multicollinearity, and homoscedasticity were tested and confirmed, ensuring the robustness and reliability of the models. This methodological rigour strengthens the validity of the findings and supports the credibility of the conclusions drawn from the statistical approach applied in this study.

The multivariate regression analysis on environmental risk perception revealed a statistically significant model (R ² = 0.056, Adj. R ² = 0.051, F(7, 1.180) = 10.15, p < 0.001), indicating that the included predictors collectively explain approximately 5.1% of the variance in perceived environmental risk.

Among the predictors, property ownership emerged as the strongest statistically significant predictor (β = 0.172, t = 5.81, p < 0.001), suggesting that individuals who own additional property perceive environmental risks as more severe. The level of education (β = −0.121, t = −3.92, p < 0.001) and settlement type (β = −0.119, t = −4.12, p < 0.001) also significantly influenced risk perception. Specifically, higher education and living in urban areas were associated with lower perceived environmental risk.

Other significant but weaker predictors included gender (β = −0.062, t = −2.13, p = 0.034) and number of household members (β = −0.058, t = −2.03, p = 0.042), both indicating a slight decrease in risk perception. Variables such as employment status and proximity to potential environmental hazards did not reach statistical significance (p > 0.05). Overall, although the model reached statistical significance, the low R ² indicates that a significant portion of the variability in environmental risk perception remains unexplained by the included socio-demographic factors.

To explore how socio-demographic and spatial characteristics influence emotional responses and perceived psychological control related to the “Duboko” landfill, a multivariate linear regression analysis was conducted. The model, which included seven predictors – gender, education, settlement type, proximity to the landfill, employment status, property ownership, and household size – was statistically significant (R ² = 0.030, Adj. R ² = 0.024, F(7, 1.180) = 5.27, p < 0.001). Although the model explains only 2.4% of the variance in psychological responses, it nevertheless reveals several meaningful relationships.

Among the examined predictors, education, settlement type, and property ownership were found to be statistically significant. Higher education levels were associated with lower levels of psychological distress (β = −0.122, p < 0.001), indicating a greater sense of emotional stability and perceived control among more educated individuals. Similarly, settlement type showed that residents in rural areas experienced more intense psychological responses (β = −0.111, p < 0.001), possibly due to a heightened sense of exposure or limited access to institutional support. In contrast, individuals who own property exhibited stronger emotional reactions (β = 0.073, p = 0.016), likely reflecting a greater sense of personal vulnerability regarding potential environmental consequences.

Other factors, such as gender, employment status, proximity to the landfill, and household size, did not have statistically significant effects in this model. Despite the modest explanatory power, the results highlight the significance of structural vulnerability – specifically, education level, place of residence, and property status – in shaping psychological responses to environmental threats. These findings can inform the design of targeted communication and intervention strategies aimed at strengthening emotional resilience in affected communities.

To assess the socio-demographic and spatial factors influencing pro-environmental values and civic participation concerning the Duboko landfill, a multivariate linear regression analysis was conducted. The model considered predictors such as gender, education, type of settlement, proximity to the landfill, employment status, home ownership, and household size, yielding statistically significant findings (R ² = 0.037, Adj. R ² = 0.032, F(7, 1.180) = 6.60, p < 0.001).

Although the model explains a modest 3.2% of the variance in environmental values and engagement, the results underscore significant trends related to structural vulnerability and social positioning. Several predictors emerged as statistically significant. Gender demonstrated a link to environmental engagement, with women showing slightly stronger values (β = −0.079, p = 0.008), suggesting a greater sense of environmental responsibility among female participants. Education also displayed a notable negative impact (β = −0.096, p = 0.002), indicating that individuals with lower educational attainment report heightened environmental concerns and involvement, likely reflecting their direct experiences with ecological degradation. Settlement type was another key variable (β = −0.109, p < 0.001), showing that rural residents are more likely to value the environment and engage more actively than urban residents.

Moreover, property ownership was positively predicted by environmental engagement (β = 0.121, p < 0.001), suggesting that property owners are more attentive to environmental sustainability, possibly due to a perceived long-term interest in local ecological conditions. Other factors, including proximity to the landfill, employment status, and household size, were not statistically significant (p > 0.05). In summary, the findings suggest that gender, education, place of residence, and property ownership have a substantial influence on environmental values and engagement among the surveyed population.

Perceptions of disaster risk among residents living near the “Duboko” landfill were further explored through a multivariate linear regression model, which incorporated a range of socio-demographic and spatial predictors. The model reached statistical significance (R ² = 0.030, Adj. R ² = 0.024, F(7, 1.180) = 5.29, p < 0.001), indicating that the selected variables collectively account for just over 2% of the variance in perceived disaster threats. While modest, this level of explanation still offers valuable insights into the structural influences that shape public concern.

The findings indicate that the type of settlement played a particularly significant role. Individuals residing in rural areas expressed greater concern about disaster risks compared to their urban counterparts (β = −0.109), highlighting the heightened vulnerability and environmental awareness among those in less urbanised settings. Similarly, property ownership emerged as a positive predictor (β = 0.121), suggesting that those who own residential property – either independently or through family – tend to perceive environmental and disaster-related risks as more serious, likely due to their long-term investment and potential exposure to loss.

Although weaker in effect, the number of household members was negatively associated with disaster perception (β = −0.055, p = 0.048), indicating a trend where individuals in smaller households may feel more exposed or less supported in the face of environmental threats. Other variables, including gender, education, employment status, and proximity to the landfill, did not demonstrate statistically significant effects in this model.

The analysis explored patterns in information-seeking behaviour, media usage, and communication about environmental risks, utilising a multivariate linear regression model. This model achieved statistical significance (R ² = 0.031, Adj. R ² = 0.026, F(7, 1.180) = 5.50, p < 0.001), indicating that the socio-demographic and spatial variables included account for approximately 2.6% of the variance in media engagement concerning environmental issues.

Among the predictors, gender emerged as a significant factor (β = −0.103, p = 0.001), indicating that women tend to exhibit slightly higher levels of engagement with information and media than men. Education level also significantly influenced the results (β = 0.00128, p < 0.001), with respondents who had less education showing greater involvement in media and communication about environmental matters. This inverse correlation may indicate a reliance on external information sources when formal education fails to provide sufficient ecological literacy.

Additionally, employment status displayed a minor yet significant adverse impact (β = 0.066, p = 0.033), suggesting that individuals who are unemployed or informally employed are more likely to engage with media to remain informed about environmental issues. Conversely, property ownership was correlated positively with higher levels of media and communication engagement (β = 0.087, p = 0.004), likely reflecting a stronger inclination to stay informed due to increased personal interest in local environmental conditions.

Other predictors, including settlement type, landfill proximity, and household size, did not exhibit statistically significant impacts within this model. Overall, although the explained variance remains limited, the results suggest that demographic factors, such as gender, education level, employment status, and property ownership, play a role in shaping how people engage with environmental information through media and communication platforms. These findings can inform the development of more effective public outreach and information dissemination strategies tailored to specific demographic segments.

A multiple linear regression model was used to assess trust in institutions and perceptions of transparency, incorporating seven socio-demographic and spatial predictors. The model demonstrated statistical significance (R ² = 0.074, Adj. R ² = 0.069, F(7, 1.191) = 13.62, p < 0.001), suggesting that the variables account for roughly 6.9% of the variance in institutional trust and perceived transparency, marking it as the model with the highest explanatory power among those analysed in this study.

Several predictors appeared particularly significant. Settlement type harmed trust (β = 0.142, p < 0.001), with individuals in rural areas reporting notably lower levels of institutional trust. A similar trend was observed with education (β = 0.129, p < 0.001), suggesting that individuals with lower educational backgrounds may have slightly more faith in institutions, possibly due to less critical engagement with these processes or differing perspectives on trust. Furthermore, gender (β = 0.086, p = 0.003) showed that women are more likely to indicate reduced levels of institutional trust and transparency. In contrast, property ownership (β = 0.157, p < 0.001) correlated positively with increased trust, possibly reflecting a sense of greater influence or alignment with formal systems among property owners.

Interestingly, the number of people in a household revealed a notable negative correlation (β = 0.132, p < 0.001), implying that those residing in smaller households tend to have higher institutional trust. Variables such as distance to the landfill and employment status exhibited no significant effects in this model. Overall, these findings demonstrate that perceptions of institutional trust and transparency are influenced not just by educational and spatial elements but also by gender roles, ownership patterns, and household composition, emphasising the complex nature of public trust in governance systems.

4.1 Recommendations

Drawing from the recognised systemic shortcomings in waste management and risk governance in Serbia, Table 14 below outlines structured recommendations designed to tackle significant challenges. Each recommendation corresponds to a specific issue, specifies the stakeholders responsible for its implementation, and assesses the feasibility of implementing it. These initiatives aim to facilitate a shift toward a more resilient, inclusive, and EU-compliant waste management system that robustly protects environmental and public health.