From Perception to Adoption: The Established Psychological Social Distance Measure as a Criterion for Citizens’ Willingness to Accept Sustainable Engineering Solutions

Abstract

Urbanization increases pluvial flood risk by expanding impermeable surfaces, which is a trend likely to intensify with climate change. Permeable pavement (PePav) made from industrial byproducts, in accordance with circular economy principles, may improve soil permeability. Public acceptance remains a critical barrier to its implementation. Existing measures of willingness to accept (WtA) new technologies are inconsistent, limiting interdisciplinary collaboration. Therefore, a concise WtA scale was adapted from the Bogardus Social Distance Scale to assess acceptance of PePav at varying levels of proximity in residential contexts, from public flood-prone roads to private yards. The scale was evaluated across three studies: Study 1 (N = 195) and Study 2 (N = 187) utilized mixed student samples, while Study 3 (N = 625) involved a non-student sample. The 5-item solution, identified through factor analysis in Study 1, consistently demonstrated a unidimensional and cumulative structure and satisfactory reliability, even after the proposed PePav ingredient modification in subsequent studies. The scale correlated with recycling experience and professional background, indicating convergent validity, but not with flooding or informal construction experience, across all samples. Study 3 provided evidence of external validity by incorporating empirically well-established Theory of Planned Behavior (TPB) constructs and showing that WtA predicted PePav use beyond TPB variables and demographics. The scale also showed measurement invariance across sample type (student vs. general population) and different levels of construction experience. The constructed WtA scale is suitable for efficiently assessing professional and public acceptance of circular building materials and may have broad cross-disciplinary relevance. This enables timely, targeted interventions and informed policy decisions to advance sustainable technologies in the built environment, with substantial implications for education, professional policy, and sustainable engineering. Nevertheless, further validation is required.

1. Introduction

According to the latest report by the Intergovernmental Panel on Climate Change, human behavior and activities are the primary drivers of global warming, which has been widely recognized as a critical threat to the lives of people [1]. Urbanization, a hallmark of modern civilization, poses significant environmental challenges. The use of non-permeable materials contributes to pluvial flooding, particularly during heavy rainfall and in areas with inadequate drainage systems. This issue affects millions globally and is expected to worsen due to climate change and urban expansion. One viable option is to restore soil permeability by implementing permeable pavement (PePav), which may include industrial waste materials to align with sustainable and circular economy principles. However, the successful deployment of such technologies depends partially on public acceptance [2]. Understanding citizens’ perceptions and concerns is therefore essential.

Previous research on the willingness to accept (WtA) innovative, sustainable construction solutions revealed notable conceptual and methodological challenges. At the conceptual level, definitions vary considerably across disciplines, and there is often ambiguity in distinguishing willingness from related constructs such as intention or behavior. Methodologically, the WtA construct has been measured with varying levels of complexity, starting from single-item measures of this hypothetical construct [3,4,5]. While practical, this technique hinders the construction of more sophisticated conceptual models and the use of advanced statistical analysis. On the other side, researchers developed robust, multidimensional scales [6,7] capturing actual behavior [8,9]. Furthermore, discipline-specific measuring methods frequently lack scalability and cross-disciplinary applicability. All of this poses a challenge to the transdisciplinary approach and generalization of research results, emphasizing the importance of conceptual clarity and methodological consistency when assessing public acceptability of sustainable, innovative engineering solutions. To address these challenges, the current study outlines three investigations that use the Social Distance Scale (SDS) [10] to assess citizens’ willingness to pay for sustainable innovative construction solutions, with a particular emphasis on PePav.

The Bogardus SDS, originally developed for assessing cultural and heritage-based intergroup relations, has been adapted as a tool for evaluating public WtA sustainable engineering solutions from conceptual, psychometric, and practical perspectives. Conceptually, the SDS assesses social acceptance by presenting scenarios that represent varying degrees of closeness to social groups, ranging from visitors to spouses. Respondents evaluate the acceptability of situations, such as having an individual from a particular group as a visitor, citizen, colleague, or neighbor, thereby illustrating different levels of social proximity. The scale has two important properties. It is unidimensional because it tracks a single underlying construct (i.e., social distance) along a linear continuum of increasing intensity. The scale also has cumulative properties since it is a Guttman-type scale. As a result, the items are arranged in a hierarchical order; an affirmative response to a more intimate level of contact logically implies acceptance of all previous, less intimate levels. This unidimensional and cumulative scaling approach enables the assessment of WtA sustainable engineering solutions by suggesting that support for a less proximate option (e.g., application in the participants’ backyards) generally extends to more distant contexts (e.g., streets and roads). This facilitates a nuanced understanding of public preferences for sustainable solutions across various settings. Psychometrically, the SDS demonstrates strong internal consistency, reliability, and validity [11,12,13], thereby maintaining contextual relevance across diverse sustainability domains. Practically, the SDS is concise, easily administered, and requires only basic reading skills, thereby minimizing socio-demographic bias. The scoring process is straightforward and does not require specialized training for examiners. The resulting scale uses a Likert-type format and a unidimensional, cumulative structure to assess psychological proximity to the proposed innovation, both in personal and urban contexts.

The primary objective of this research was to evaluate the psychometric properties of the WtA scale, its predictive validity with respect to behavioral intentions, and its cross-validity against the Theory of Planned Behavior (TPB). Additionally, the scale’s sensitivity to contextual factors, such as prior relevant experiences (e.g., with recycling, environmental activism, construction, or floods) and professional expertise or background, was examined to assess its potential for nuanced population segmentation. Finally, the scale’s practical applicability for research, policymaking, and public acceptance initiatives is discussed, particularly in domains where sustainable engineering innovations are crucial. The research framework integrates contemporary scientific perspectives, including biopsychosocial and systemic approaches, evidence-based practice, and transdisciplinary thinking.

1.1. Climate Change, Urbanization and Pluvial Flooding: Background

Climate change is increasing the frequency of high-intensity, short-duration rainfall, while rising urbanization raises extra demands for boosting the capacity of urban drainage systems [11]. Furthermore, the expansion of impermeable surfaces (roofs, roads, pavements) greatly reduces permeable surfaces in urban areas, forcing the bulk of rainwater to produce surface runoff, which is subsequently collected and carried through sewage infrastructure [14,15,16,17]. During the urbanization process, vegetation that previously played a role in retaining and recirculating water through evapotranspiration is removed, while soil compaction during construction further limits infiltration. As a consequence, the rate of surface runoff increases, the natural hydrological regime is altered, and the risk of flooding and erosion in downstream channels increases. Reduced infiltration can also lead to a decline in base flows in watercourses, diminished recharge of underground aquifers, and the degradation of natural habitats in aquatic ecosystems and along riverbanks [15,18]. Impervious surfaces reduce the potential for infiltration, so stormwater no longer penetrates the soil as it did before urbanization. This causes an increase in surface runoff, characterized by larger water volumes, higher peak flows, and shorter duration. The excess water is directed toward the sewer infrastructure, whose capacity is most often insufficient to receive such large amounts of precipitation in a short period of time. The consequence is the occurrence of pluvial flooding, which may include temporary accumulation of water on roadways and sidewalks, flooding of low-lying terrain and underground structures such as subways, underground roads and passages, basements, etc. [15,19,20,21].

The impacts of pluvial flooding are diverse, extending across real estate, social, and environmental dimensions. Accumulated water on roadways interferes with or completely disrupts traffic flow, while flooded basements and residential and commercial buildings cause significant economic losses [19,22]. Furthermore, floods can damage public health by spreading pollution and increasing the risk of infectious diseases [23,24]. At the same time, disruptions in the operation of public utilities, such as water supply and electricity distribution, further complicate daily life and require urgent intervention. An additional problem is the transport of pollutants by surface runoff into water bodies, which may lead to, depending on local conditions, degradation of water quality and adverse impacts on biodiversity [16,21,25].

1.2. The Urban Drainage Approaches and Solutions

The traditional approach to urban drainage is still largely focused on gray infrastructure, which includes conventional structures for stormwater collection and transportation, with the occasional application of stormwater retention/detention facilities [26,27]. However, due to changing climatic conditions and accelerated urbanization, the existing sewer infrastructure is often overwhelmed by stormwater inflow, resulting in pluvial flooding occurring almost every year. The construction of urban stormwater systems, which typically falls behind the rate of urbanization, aggravates the situation even more. The limitations of existing stormwater sewer systems necessitate the implementation of new and sustainable urban drainage solutions [28,29]. Progress in this area is expected through the implementation of modern stormwater management systems, which aim not only to reduce peak flows but also the overall volumes of stormwater runoff through retention and infiltration while simultaneously improving wastewater quality [11,26]. A possible solution for reducing the inflow of stormwater into sewer collectors is to reduce storm runoff through an increase in infiltration into the ground through the application of nature-based solutions (NBS) [14,30,31]. These solutions can include the application of pervious materials such as pervious concrete and asphalt or prefabricated elements made from ordinary concrete, building ceramics, or natural stone, but placed in a way to form openings between adjacent elements. Recommended areas of application for such pavements are pedestrian sidewalks, bicycle trails, and parking lots [32,33].

Pervious concrete is designed with a high percentage of open pores due to a lack of fine aggregate fraction (0/2 mm), which makes it highly permeable for water (filtration coefficient ranges between 0.2 × 10⁻² m/s and 0.54 × 10⁻² m/s, although it is possible to reach much higher values) [34]. There are two possibilities for its installation: pervious concrete placed on site and prefabricated pervious concrete pavers. One of the main advantages of prefabricated pervious concrete pavers is their sustainability potential, which can be perceived through several points. Primarily, by reducing or completely excluding the fine aggregate fraction, it preserves natural river sand, whose extraction affects a wide range of ecosystem services (shape of the riverbed, biological environment, chemical environment, and human environment) [35]. Secondly, it is possible to design pervious concrete in a way to incorporate different waste and recycled materials as partial replacements for cement and coarse natural aggregate. There are already studies related to the incorporation of fly ash, ground granulated blast furnace slag, and solidified wastewater sludge as partial replacements of cement, and recycled concrete aggregate and crushed brick aggregate as partial replacements of natural aggregates in these composites [36,37,38]. Finally, once their service life is reached, individual pavers could be crushed and reused as aggregate for the production of new concrete [39]. The drawbacks of previous concrete pavers are mainly related to the requirement for their regular maintenance, such as vacuuming and cleaning to avoid closure of the voids and reduction in permeability [34]. An essential aspect for researchers that is frequently disregarded is citizens’ willingness to adopt these sustainable materials in their daily lives, as the end consumers of various solutions in urban environments.

1.3. Importance of Human Factor in Application of Sustainable Solutions

The concept of sustainability dates back over 310 years, when Hans Carl von Carlowitz introduced the term Nachhaltigkeit to describe the sustainable management of wood resources [40]. The modern understanding of “sustainable development” began to take shape in the early 1970s, in response to growing ecological concerns and calls for a stable society ensuring long-term ecological and economic balance [41]. By the 1980s, it had evolved into a recognized paradigm [41,42]. The Brundtland Report [43] established the foundation of sustainable development by linking environmental protection with social and economic progress [42]. Its influence was far-reaching, spurring the rise in sustainability science [44]. This discipline is based on multidisciplinary techniques and was created via collaboration across the scientific and social sciences. Addressing global change requires moving beyond disciplinary boundaries and embracing co-designed, co-produced research. Since the 1970s, psychology has made significant contributions to environmental research [45]. This contribution reflects the recognition that environmental issues are not only technical but also rooted in human perception, decision-making, and behavior. The institutionalization of sustainability science, supported by the creation of the International Science Council, reinforced the integration of natural and social sciences [44]. Within this framework, the human factor plays a dual role, serving both as a driver of environmental change and as a key agent in finding solutions. Analysis of the human factor in sustainable engineering includes activities that directly contribute to environmental protection and the advancement of ecological practices. In the context of sustainable environmental engineering, this entails efforts to mitigate barriers and enhance public acceptance of innovative sustainable solutions. A holistic strategy can promote more congruence between human actions and desired environmental consequences, resulting in a more sustainable future.

The concept of WtA in the field of sustainable engineering is extensively cited and investigated as a prerequisite for social acceptance and an indicator of the potential successful adoption of such technologies. There are inconclusive research findings on the (un)successful adoption rates of new technology. According to Gourville [46], the failure rate of new technology adoption in the market is significantly high, with estimates ranging from 40% to 60%. In contrast, some authors contend that the failure rate for new products is not as pronounced, typically reported to be 40% or lower [47]. Despite presented disagreement, there is no doubt that the adoption of sustainable technology solutions is linked to user acceptance [48,49,50,51]. Public acceptance can mitigate the risks of cost overruns, project failures, or cancellations [52,53].

Research results reveal that acceptance of technology is a complex phenomenon, involving willingness and intention, which are closely linked but frequently used interchangeably [54]. In the current literature on sustainability, the specific lexical and scientific status of the willingness and related terms may be observed. Willingness denotes an individual’s openness and readiness to take action [55]. The will is generally understood as the human capacity for autonomous decision-making, allowing individuals to make choices independently of external influences, which encompasses both the ability to decide and the perseverance to pursue a set purpose [56]. Contemporary definitions emphasize the voluntary nature of willingness, describing it as the ability to act without reluctance or hesitation [57]. In psychological contexts, willingness is viewed as a vital factor in behavioral change, reflecting an individual’s motivation and readiness to take action or pursue personal growth. Additionally, it is acknowledged as a precursor that prompts behavioral actions in favorable situations [58]. Intention is a structured and goal-oriented concept. The APA Dictionary of Psychology defines it as a conscious decision made prior to engaging in a specific behavior, often linked to clear objectives and directives in experimental research [59,60]. Willingness and intention are usually empirically introduced in order to assess the adoption of the new technologies. Adoption pertains to the acceptance or initiation of new practices. This process involves both the commencement of utilizing something and the granting of official approval [61,62].

Besley et al. [63] highlighted the conceptual similarities between willingness and intention constructs, noting that both indicate a form of ‘readiness’ to engage in behavior and are often empirically correlated. However, Gibbons et al. [64] emphasize a significant distinction: while willingness indicates readiness to act, intention involves a more structured and planned approach to behavior execution. Accordingly, in psychology studies on sustainability, willingness is considered a motivational precursor to behavior, especially important in contexts of change and growth, which reflects readiness rather than commitment. Intention plays a central role in planned behavior theories, representing a commitment to perform a specific action. Pomery et al. [65] note that willingness can be conceptualized as an individual’s openness to opportunities, specifically their readiness to engage in a particular behavior when favorable circumstances arise. In contrast to intention, which necessitates careful deliberation regarding both the behavior and its potential consequences, willingness often requires minimal precontemplation [66,67]. Behavioral intention is typically characterized as a structured decision-making process that encompasses the assessment of behavioral alternatives, the evaluation of potential outcomes, and a conscious choice to either engage in or refrain from the behavior. According to social behavior models, intention is viewed as the product of a deliberate, goal-directed cognitive process [68]. A body of empirical evidence suggests that willingness and intention can function as independent predictors of behavior [65]. A meta-analysis conducted by Webb and Sheeran [69] found that interventions designed to change intentions resulted in medium to large effects, whereas changes in intention accounted for only small to medium effects on behavior. This analysis further revealed that the effects of interventions on behavior were not entirely mediated by intention. Those findings highlighted the necessity for further investigation into not-so-strictly reasoned behavioral processes, such as willingness.

Several studies examined the intricate interplay between willingness and intention. Intention has demonstrated greater reliability in predicting planned behaviors, while willingness showed a strong association with spontaneous behavioral engagement when opportunities present themselves [70]. Concerning specific behaviors, the predictive validity of willingness has been shown to extend beyond that of intention among distinct age groups [64]. Research comparing behavioral intention (BI), behavioral expectation (BE), and behavioral willingness (BW) has revealed variability in predictive strength influenced by factors such as respondents’ age, prior experiences, and the behavioral context [65]. Furthermore, attitudes have been identified as consistent predictors of willingness across various public engagement behaviors, as emotional processing plays a more significant role in reactive and unplanned behaviors [71]. Notably, self-efficacy has been consistently identified as a strong predictor of intention [71,72,73], while its influence on willingness has been found to be minimal [63]. This may explain why attitude, instead, was found to be the more consistent predictor of willingness across different modes of public engagement. It is important to underscore that, unlike intention, willingness is contingent upon situational factors [64]. Thus, when assessing willingness, it is imperative to clearly define scenarios, allowing respondents to provide behavioral responses based on contextual conditions rather than personal attributes.

1.4. Willingness in the Relevant Sustainable Environmental Solutions Research

Over the past fifteen years, several research studies addressed WtA as a central construct, searching for the most suitable conceptual and operational definitions of the term. Those studies fall under four thematic areas: Construction & Technology [74,75,76,77,78,79,80]; Environment & Sustainability [81,82,83,84,85,86,87,88,89]; Economics [90,91,92,93,94,95,96]; and Education [97,98,99]. This research reveals considerable divergence in the conceptualization of willingness and intention, reflecting disciplinary priorities and methodological traditions. The primary determinants of willingness to adopt technologies or sustainable practices across various sectors include technological characteristics and perceived benefits, economic factors, organizational and regulatory environments, information and knowledge, general versus local acceptability, and psychological factors such as attitudes, norms, and perceived control. Technologies perceived as useful and simple to implement are more likely to be adopted. In contrast, technological complexity can be an obstacle [75,76,79,99]. Technology costs, financial incentives, and expected economic benefits influence willingness to adopt [77,90,94]. In organizational contexts, regulatory frameworks, market demand, professional authorities, and organizational support also play important roles [80,81,84,87,89,92,95,98,99]. More information and knowledge increase the willingness to implement innovations, while a lack of information is a significant barrier [82,83,86]. Prior experience, particularly with comparable technologies, facilitates a more accurate evaluation of their respective benefits and limitations [85,97]. Citizens often express higher acceptance of technologies than of immediate local implementation (NIMBY effect) [77,88,91].

Psychological and social factors are recognized as key predictors of readiness and intention to adopt new practices. Within this research cluster, broader value and normative frameworks, including environmental awareness and pro-environmental practices such as recycling, significantly influenced attitudes toward technologies intended to address environmental challenges [82,83,84,89,93]. Stakeholder attitudes also directly predict an organization’s readiness to implement practices, such as minimizing construction waste [80,84,92,99]. Cultivating social responsibility and professional values can increase willingness to address social and environmental challenges [96]. Research employing models such as the TPB indicates that beliefs regarding the likely consequences of the behavior, normative expectations of others, and the presence of factors that may facilitate or impede behavioral performance can significantly influence WtA and actual behavior [97,99]. The subsequent section provides a detailed overview of the TPB framework and examines how its key components shape WtA’s innovative solutions.

Regarding scale format, the selection of items and scales was aligned with the prevailing research principles within each thematic area. Studies in Construction & Technology, Environment & Sustainability, and Education most frequently employed Likert-type scales, whereas Economics predominantly utilized binary or price-based measurements. Parasuraman proposed an even more elaborate framework, particularly relevant for research focusing on WtA, through the development of the Technology Readiness Index, a multi-item assessment tool designed to evaluate individuals’ WtA new technologies [74]. Accordingly, it appeared plausible to develop a conceptually grounded and psychometrically robust measure of WtA.

According to core psychometric principles [100], such a measure should have a unidimensional structure to ensure internal coherence and meet established reliability and validity standards. Importantly, it should be contextually appropriate, capable of assessing WtA sustainable engineering solutions while maintaining conceptual independence, allowing its application across various sustainability research domains. Additionally, the instrument should be brief and easily administrable, requiring only basic reading skills, minimizing socio-demographic bias, and enabling sensitive data collection. Scoring procedures should be straightforward, requiring no specialized training or advanced qualifications from examiners. For these reasons, the Bogardus SDS was chosen as a base for the WtA measurement.

1.5. The Origins and Application of the Bogardus SDS

The concept of social distance was introduced by sociologist Robert Park in the 1930s [101] in an effort to represent through quantifiable indicators the degrees of intimacy and comprehension that characterize individual and interpersonal connections in general. Social psychologist Emery Bogardus, driven by a desire to understand the reality and dynamics of social groups, further developed Park’s ideas, which led to the creation of one of the most cited and empirically utilized psychometric tools, the Bogardus SDS [10]. According to Mather et al. [102], this scale has been used for decades to enhance insights in the social sciences and beyond. The Bogardus scale assesses the degree of acceptance in social interactions. Respondents are presented with various scenarios (e.g., having someone from a particular group as a visitor to their country, a citizen, a colleague, or a neighbor), each representing a different level of closeness to the given social group. For example, cohabitation implies a different level of closeness than friendship or marriage. Respondents indicate the acceptability of each scenario for the groups. It is based on the Guttman scale, where the selected possibility implies answers to all others. Therefore, it assumes that all the ‘less distant’ options are sine qua non prerequisites for a ‘more distant’ option (i.e., if someone chooses that they would marry an out-group member, then it is assumed that they would also accept this person as a neighbor). Therefore, the scale can assess acceptance in specific contexts or serve as a general indicator.

The Bogardus SDS was originally developed to research cultural and heritage-based intergroup relations, but it is now extensively used to assess attitudes toward various social groups of interest within a given community. The Bogardus SDS has certain unique properties that make it a useful tool for understanding people’s WtA in order to develop new, sustainable engineering solutions. First, its unidimensionality [103] indicates that the scale items measure a single theoretical concept, in this case, the WtA sustainable technological solutions. Second, the scale is cumulative. It can be assumed that accepting less distant options would encompass the WtA or more distant solutions. Specifically, a respondent expressing a certain degree of WtA will also endorse items that express lesser degrees of WtA. For instance, a respondent’s WtA a proposed sustainable engineering solution in their yard will also likely accept the same solution in all urban areas. As a result, this scale assists in assessing preferences for proposed solutions intended for residents’ living spaces and surrounding areas, thereby pinpointing where individuals are WtA or distancing themselves from them. In practice, if there is a requirement to adopt specific solutions, this scale may help define the extent to which these solutions are acceptable to citizens and in what proximity. For example, a solution could be deemed suitable in critical flooding areas but not across an entire country or in every citizen’s yard. Furthermore, if someone asserts that a solution is appropriate in their yard, it is reasonable to assume that it is also acceptable in other flooded or urban areas.

One key issue concerning the psychological processes that underpin the Bogardus SDS in the context of measuring WtA sustainable engineering solutions requires further investigation. Despite its long-standing tradition and numerous empirical studies, the debate continues: is social distance simply a type of attitude? Some researchers argue that it is, implying that social distance contains all components of attitude, cognitive, conative, and emotional, and has the ability to predict conduct [102,104]. Conversely, other authors argue that it is inappropriate to equate social distance with attitudes. They suggest that social distance is most closely linked to the conative component, which may or may not align with the affective and cognitive dimensions of attitude [105,106]. In this context, social distance can lead individuals to demonstrate discriminatory behavior towards the subject at hand [107]. This aligns conceptually with the proposal to apply this scale within an interdisciplinary approach to sustainable engineering. Addressing this issue is of practical necessity. The relationship between opinion, affection, and intention is complex [108]. Moreover, an extremely negative attitude often goes beyond a simple refusal to adopt innovative solutions. A greater social distance or reduced willingness to embrace certain solutions does not invariably indicate a negative opinion about them; in some cases, it might simply reflect a lack of familiarity with the evaluated solution. These dilemmas underscore the need for testing the WtA in the context of conceptually and empirically robust theoretical frameworks [69], such as the widely applied TPB in studies related to sustainable engineering.

1.6. The TPB: Origins, Overview, and Applications in Environmental and Sustainability Research

The TPB, developed by Icek Ajzen in 1991 [109], is one of the most widely used and empirically validated theories in social psychology. The TPB is a robust and flexible theoretical framework for understanding and predicting human behavior across various domains. Its emphasis on cognitive decision-making provides valuable insights into the mechanisms underlying behavioral intentions and actions. It offers a solid foundation for theoretical exploration and practical application in efforts to influence and change behavior for individual and societal benefit.

The TPB (Figure 1) proposes that human conduct is shaped by three basic sorts of considerations [110]: (1) Behavioral beliefs—these are beliefs regarding the likely outcomes of a specific behavior; (2) Normative beliefs—these pertain to beliefs about the expectations of others concerning that behavior; and (3) Control beliefs—these involve beliefs about the factors that may either facilitate or obstruct the execution of the behavior. Therefore, attitudes toward the behavior (ATB), subjective norms (SN) and perception of behavioral control (PBC) lead to the formation of a behavioral intention (BI).

BI is conceptualized as an indicator of the extent to which individuals are predisposed to enact a specific behavior and is considered the most immediate predictor of that behavior [109]. The formation of BI is influenced by three primary determinants: ATB, SN, and PBC. BI is jointly determined by an individual’s attitude, which captures the degree of favorability or unfavorability toward the behavior in question; SN encompasses the perceived social pressures to engage in the behavior; and PBC denotes the perceived ease or difficulty associated with performing the behavior, akin to the construct of self-efficacy [111,112]. ATB was identified as one of the critical predictors in contemporary research concerning the application of the TPB within the context of environmental science [111]. Numerous studies substantiate that a more favorable attitude correlates with an increased likelihood of behavioral enactment [2,113]. For example, the extent to which citizens perceive proposed sustainable solutions as harmless, affordable, or reliable directly influences their WtA for such solutions [114]. SNs are articulated as the social pressures that individuals perceive from significant others during the decision-making process regarding behavior [115]. Individuals who obtain support for specific behaviors from key figures in their lives, such as close friends or colleagues who encourage them to adopt sustainable solutions, exhibit a greater likelihood of engaging in those behaviors [116,117,118]. PBC significantly enhances the predictability of the model. It refers to an individual’s perception of how easy or difficult it is to engage in a particular behavior [109]. The easier the behavior is to perform, the greater the perceived control one has over it [119]. In essence, whether a person believes they can adopt sustainable solutions or feels constrained in their ability to do so is a critical factor in their involvement in sustainable behavior [120,121,122]. When individuals have a sufficient degree of real control over their actions, they are expected to act in accordance with their intentions when opportunities arise [110]. The influence of ATB and SN on BI is moderated by PBC [110]. Generally, a more favorable ATB and SN, combined with higher PBC, lead to a stronger intention to engage in the desired behavior. Consequently, when individuals possess adequate actual control over their behavior, they are likely to fulfill their intentions when the opportunity arises. In this context, intention is viewed as the immediate precursor to behavior. If PBC accurately reflects actual control, it can effectively serve as a predictor of the behavior in question [123].

A bibliometric analysis conducted by Si et al. [111] indicates that the TPB was first applied in environmental science research in 1995. Between 2008 and 2018, the average annual growth rate of publications related to TPB in the Environmental Science category defined by Scopus reached an impressive 62.19%. The authors projected that the total number of publications for 2019 could exceed 180. This expanding body of research not only demonstrates the TPB’s broad applicability in predicting various individual behaviors but also highlights the increasing interdisciplinary integration within environmental science, particularly with fields such as social sciences, energy, and engineering.

The TPB is widely applied in sustainability research to predict intentions and behaviors in areas including environmental protection, waste management, green consumption, resource conservation, and sustainable transportation [124,125]. Researchers frequently incorporate control variables such as culture, education, age, gender, and income to enable comparisons across demographic groups, and studies often focus on specific populations such as households, students, young consumers, and farmers [111]. In practice, the TPB serves as a foundational framework for designing interventions that promote sustainable behavior by targeting, according to the theory, the most relevant constructs: ATB, PBC, and SN. Among the three components, ATB and PBC are generally regarded as the most influential in shaping environmentally significant behaviors, while SN plays a context-dependent role, particularly in specific demographic segments [111,112]. Overall, TPB not only provides a robust tool for empirical analysis of sustainability-related behaviors but also facilitates interdisciplinary integration across environmental, social, and economic dimensions, supporting both theoretical understanding and practical applications.

1.7. The TBP and WtA

Numerous studies have applied the TPB to examine determinants of WtA across various environmental and sustainability contexts. Recent research has consistently shown that willingness is a crucial predictor of pro-environmental intention, operating alongside and often beyond traditional TPB components. Studies have examined diverse contexts such as electronic waste recycling [126,127], green product purchasing [128], and financial contributions to pollution reduction [129], demonstrating that willingness shapes intention across behaviors. Extensions of the TPB have incorporated additional predictors such as normative factors and perceived control in energy conservation [117,130], morality and self-identification in water conservation [131], and comparisons with alternative models such as the Norm Activation Model in commuting choices [132]. Importantly, these studies highlight that willingness continues to explain pro-environmental intentions even when additional TPB-related factors are considered. Similarly, applications emphasizing environmental knowledge and concern, including water conservation [133], air quality improvement [129], and green vehicle purchasing [134], demonstrate that willingness consistently predicts sustainable choices over and above the standard TPB framework. Demographic influences have also been observed in WtA sustainable hotels [135], electric vehicles [136], zero-waste tourism [129], and green roofs [137]. Collectively, these findings underscore the importance of WtA in understanding and promoting sustainable behavior. Accordingly, empirical findings within the TPB framework consistently show that WtA sustainable, innovative solutions are strongly associated with the BI to adopt them, albeit with variations in the strength of association and contributing factors across sustainability domains. Given that the TPB specifies distinct predictors of intention, it was deemed meaningful to examine the external validity of WtA alongside other TPB constructs. Evaluating WtA within the TPB domain is particularly relevant in light of the previously acknowledged conceptual uncertainty regarding whether the social distance construct, underlying the proposed WtA scale, merely captures attitudinal dispositions or constitutes a conceptually and psychometrically distinct variable. Distinct in the sense that it appears to be most closely related to the conative component of behavior, thus aligning this measure with the notion of readiness to act without reluctance or hesitation, a precursor to a structured, goal-directed decision often linked to clear objectives, i.e., BI.

Building on the presented empirical findings, this paper conceptualizes readiness for adoption using an integrated framework. The framework encompasses psychological determinants from TPB, including ATB, SN, and PBC. It also incorporates knowledge and awareness of technology, distinguishing between general and local acceptability. WtA is positioned as an intermediate construct that links cognitive-psychological and contextual factors with technology support, including willingness to pay.

2. Materials and Methods

The main goal of the research is to develop a measurement tool for assessing citizens’ WtA regarding PePav. The assessment performed concerning PePav is based on (1) the Ø-Waste-Water project proposed solution (PePav_Ø-W-W) and (2) neutral (PePav_N). According to the Ø-W-W project proposal, PePav_Ø-W-W is designed to incorporate selected municipal and industrial wastes in line with defined design specifications. These may include blast furnace slag, fly ash, solidified wastewater treatment sludge, and cathode ray tube glass [38]. In contrast, PePav_N is conceived to utilize neutral, harmless, value-adding material derived from hazardous industrial waste via the patented MID-MIX^® process [138]. Neutral may contain processed oils, contaminated soils, petrochemical and pharmaceutical residues, industrial and municipal sludge, paints, tar, emulsions, and other hazardous waste, including asbestos [138]. A pilot study [114] focused on the construction of the WtA scale, evaluation of its initial psychometric properties, along with associations with sociodemographic variables and relevant personal experience. Across three consecutive studies presented in this paper, the scale underwent additional improvement and consecutive psychometric validation, and the WtA concept was cross-validated by examining its associations with TPB constructs.

2.1. Three-Study Design for WtA Measure Validation

Study 1: The SDS properties WtA measure was examined in a sample of students from the University of Belgrade. An exploratory factor analysis was applied to assess unidimensionality. Beyond established unidimensionality, it is also important to explore the scale’s cumulativity, a defining characteristic of SDS, in which endorsing a more intimate level of contact implies acceptance of all preceding, less intensive levels. This was assessed via Mokken Scale Analysis using the automated item selection procedure (AISP) and by calculating three different Loveringer H indices [139]. First, AISP was used to determine whether our set of items forms a unidimensional, cumulative scale, using a range of recommended thresholds (0.00 to 0.55). The inter-item coefficient (Hij) assesses whether a specific pair of items maintains a consistent hierarchical order by measuring the extent to which participants endorse a more intimate social distance item while rejecting a less intensive one. The item-specific coefficient (Hi) captures the extent to which a single item conforms to the cumulative structure in relation to the full item set. Finally, the total scalability coefficient (H) quantifies the overall degree of cumulativity of the entire scale. These indices were interpreted according to [139], where <0.30 indicates unscalable items, 0.30 to 0.40 weak, 0.40 to 0.50 moderate, and >0.50 strong scalability.

Recent work highlighted the critical need to understand how young adults, including university students, form and develop acceptance of sustainable infrastructure [140]. Consequently, Studies 1 and 2 focused on student samples for two specific reasons. First, students represent future citizens living in urban environments, including professionals and policymakers who will make decisions on how to manage sustainable infrastructure [141]. Second, understanding how this subgroup conceives of psychological readiness could be an important first step toward developing effective policies for technological adoption [142]. For these reasons, it was considered that students may act and think as major stakeholders who might either drive or derail sustainable agendas forward. The measure was applied to two target objects: (1) PePav_Ø-W-W, and (2) PePav_N. Since the sample included students from two faculties, the Faculty of Philosophy (P) and the Faculty of Civil Engineering (CE), differences in PePav acceptance among groups with varying levels of engineering and construction knowledge, as well as exposure to construction practice and waste management, were also investigated. Within the CE, students were drawn from different departments, including Hydraulic and Environmental Engineering (HEE) as well as General Civil Engineering (GCE), which allowed for a more nuanced analysis of how specific educational backgrounds and experience within the construction field might influence responses.

Study 2: The internal structure of the final WtA measure, as identified in Study 1, was subjected to replication using exploratory factor analysis and the Mokken scale analysis. In Study 1, the description of the neutral element may have implicitly suggested that it was composed of asbestos, a substance potentially perceived as strongly aversive by participants. Therefore, an additional aim of Study 2 was to replicate the findings of Study 1 using an alternative, less objectionable definition of the neutral element. Differences between students with and without formal education in construction were also analyzed.

Study 3: To validate the WtA measure in a non-student context, the study was expanded to include a larger and more diversified sample from the general Serbian population. The sample included individuals coming from either a construction or a non-construction background to understand whether different levels of professional knowledge and experience affect participants’ scores on the WtA scale.

Three primary objectives were pursued. First, the adequacy of a single-factor model of PePav_Ø-W-W acceptance was assessed through confirmatory factor analysis (CFA), with the aim of determining whether the model structure identified in prior studies remained consistent within a general population sample. As in Studies 1–2, a Mokken scale analysis was applied to examine the measure’s cumulativity. Study 3 also tested for measurement invariance across sample type (student vs. general population) and construction experience (yes vs. no) using the CFA approach. Second, variations in participants’ experiences related to recycling, environmental protection, construction, and flood exposure were examined as potential influences on attitudes toward sustainable practices. Third, evidence for the external validity of the WtA measure was provided through an analysis of its associations with theoretically relevant constructs. In particular, relationships between WtA, intention to adopt PePav_Ø-W-W, and the three core components of the TPB [109,115] were explored. Furthermore, distinctions between individuals with and without a professional background in construction were investigated. While urban planners and policymakers are undoubtedly core stakeholders in the implementation of sustainable engineering solutions, the primary objective of this research is to provide a validated tool for measuring citizen WtA. This focus is justified, as public opposition and limited social license are often primary barriers to sustainable infrastructure, leading to delays or cancellations [143,144]. Then, quantifying the psychological distance of the general public is a necessary precursor to planning or policy decisions [145]. The constructed scale might provide planners with a metric to gauge readiness before physical construction begins [146].

2.2. Participants, Materials and Procedures, Data Analysis and Ethical Considerations

This manuscript provides information on all data exclusions and analyses. To improve clarity, sample characteristics, materials, procedures, data analysis methods, and results are presented separately for each study. Data analysis was conducted within the statistical programs: (a) R [147] using the basic stats, psych [148], rstatix [149], emmeans [150], parameters [151], lm.beta [152], BetterReg [153], dynamic [154], and lavaan [155] packages; and (b) FACTOR [156]. The local ethics committee approved all three studies, Helsinki Declaration guidelines were followed, and all respondents gave consent.

3. Results

3.1. Study 1

Study 1 explored the internal structure of the WtA measure on a sample of University of Belgrade students towards two objects: (1) PePav_Ø-W-W and (2) PePav_N. As outlined in the brief study description, the research sample included the Faculty of Philosophy students’ sample (S_P) and the Faculty of Civil Engineering students’ sample (S_CE), as well as two departments within the S_CE: Hydraulic and Environmental Engineering (S_HEE) and General Civil Engineering (S_GCE). Participants received course credits for their contribution. Accordingly, Study 1 also examined differences in PePav acceptance among student groups varying in engineering and construction knowledge and in their exposure to construction practice and waste management.

As detailed in

Table 1. Characteristics of the student samples from Study 1.

Variable		N (%)
Gender	Male	71 (36)
	Female	124 (64)
Study program	SP	88 (45)
	SGCE	61 (31)
	SHEE	46 (24)
Recycling experience	Yes	131 (67)
	No	64 (33)
Environmental protection activities and experience	Yes	133 (68)
	No	62 (32)
Construction experience	Yes	128 (66)
	No	67 (34)
Flood experience	Yes	55 (28)
	No	140 (72)

, a total of 214 individuals accessed the study. After excluding cases with missing data, the final sample included 195 valid responses. The average participant age was 23 years (M = 23.27, SD = 2.43), with 71 males and 124 females.

The sample size was sufficient for exploratory factor analysis implementation based on the general rules of thumb [157] and the number of participants to items ratio [158,159] and observed correlation coefficient magnitudes (rs > 0.57). A sensitivity power analysis revealed that a 3 (between) × 2 (within) ANOVA could detect small-to-medium main effects and interactions (f < 0.22) considering the sample size, α = 0.05, and power = 0.80. A 2 (between) × 2 (within) ANOVA detects small-to-medium main effects and interactions (f < 0.20) considering the sample size, α = 0.05, and power = 0.80. Participants accessed an online questionnaire implemented on Google Forms (https://docs.google.com/forms/d/e/1FAIpQLSea8m_fNbjpC549xnJn_VMkWk63BYz6HjrLbx0FgP7V0peJjg/viewform?usp=header) (accessed on 22 April 2026). After giving consent, participants answered the WtA PePav_Ø-W-W and PePav_N measures, a set of demographic questions (gender, age, study program, recycling experience, participation in environmental protection activities, experiences in construction work, and flood experience), and other measures going beyond the scope of the present work are described in detail in Svetozarević and colleagues [114,160].

The study lasted approximately 60 min. The Bogardus SDS [10] was modified to measure how much participants would accept PePav_Ø-W-W. First, the SDS’s original version presents a case of a single, forced-choice item where participants choose an answer reflecting the closest degree of intimacy towards a social group. Since this answering method might not always correspond to participants’ mental models [161], following Weinfurt and Moghaddam [11], the Likert approach was used to treat each of the seven options as a separate item. Before answering the questions, participants read that PePav_Ø-W-W may include blast furnace slag from steel and copper production, fly ash, solidified wastewater treatment sludge, and cathode ray tube glass. Thereafter, participants rated their willingness to cover spaces described in each item with PePav_Ø-W-W on an 8-point scale (1 = I would not cover the space with PePav under any circumstances, 8 = I would cover the space with PePav_Ø-W-W without a doubt). Second, the included WtA items were selected to satisfy two criteria: a) they should reflect options to cover spaces with PePav, and b) they are personally relevant, provoking varying psychological distances to PePav in the participants. The same approach was applied when measuring WtA towards PePav_N. Prior to responding, participants read that neutral is a powdery substance made from the technological processing of different types of waste, including asbestos. Additionally, participants found out that neutral can be used to construct PePav. After reading about neutral, participants answered the seven WtA items towards PePav_N using the 8-point scale described above.

Table 2. Descriptive statistics, item loadings, and item-specific coefficients of scalability for the final five WtA PePav_Ø-W-W measure items for Study 1.

I Would Use PePavØ-W-W to Cover:	M	SD	Min.	Max.	Sk (SE)	Ku (SE)	Item Loadings *	Hi
Item 1: all streets and roads	6.10	2.07	1	8	−1.05 (0.13)	0.05 (0.38)	0.85	0.71
Item 2: streets and roads in flooded areas	5.50	2.14	1	8	−0.50 (0.11)	−0.90 (0.17)	0.93	0.77
Item 3: my street	5.87	2.12	1	8	−0.75 (0.13)	−0.59 (0.24)	0.95	0.78
Item 4: my college/work environment	5.88	2.10	1	8	−0.74 (0.12)	−0.61 (0.24)	0.94	0.76
Item 5: my yard	5.77	2.17	1	8	−0.68 (0.12)	−0.82 (0.23)	0.73	0.62

* Reported item loadings reflect the loadings of each item on the WtA PePav measure factor. Item loadings were computed via an exploratory factor analysis.

and Figure 2 present the item descriptive statistics and item loadings. The range of observed item values indicated that the participants used the full range of possible responses. However, mean item scores gravitated mostly towards higher values in all items (ranging from 5.54 to 6.21). The skewness/standard error ratio exceeded all items’ critical value (z = |2.58|), indicating skewed data. Likewise, the kurtosis/standard error ratio exceeded the critical values for items 2, 3, and 5. In Table 2, M is the mean value; SD is the standard deviation; Min. is the minimum observed item value; Max. is the maximum observed item value; Sk is the item’s skewness value; Ku is the item’s kurtosis value; SE is the standard error of skewness and kurtosis, respectively; and Hi is the item-specific coefficient of scalability.

In summary, participants’ responses to the five items were not normally distributed, and their answers reflected a general acceptance of covering the spaces described in the items with PePav_Ø-W-W. The psychometric properties of the final five-item WtA measure were evaluated to establish evidence for its unidimensionality and reliability [162], and cumulative structure [163,164]. First, to assess internal structure, a robust exploratory factor analysis (EFA) was performed on a polychoric correlation matrix to account for the non-normal distribution of the categorical item data. Following the iterative removal of two redundant items identified through the Expected Residual Correlation Change index (EREC [165,166,167]; Appendix A), both standard [168] and optimal [165] implementations of parallel analysis supported a single-factor solution. This unidimensional model explained 82% of the total variance, with all items exhibiting strong factor loadings and excellent internal consistency (α = 0.92). This reliability coefficient remained stable despite the reduction in the item pool, confirming the efficiency of the shortened scale.

The measure exhibited cumulative scale structure across all recommended AISP thresholds (0.00 to 0.55). As shown in Table 2, all item-specific coefficients (H_i) were well above this threshold, and the overall total-scalability coefficient (H = 0.73) further quantified a high degree of cumulativity for the entire scale. Additionally, inter-item coefficients (H_ij) ranged from 0.55 to 0.88, confirming that participant response patterns consistently adhered to the intended hierarchical order. Collectively, these findings demonstrate that the WtA measure is a reliable, unidimensional instrument with a robust cumulative structure. The group comparisons on the WtA measures are related to whether the participants’ answers differed on the WtA PePav_Ø-W-W and the PePav_N measures, depending on their study program and experiences with recycling, environmental protection, construction, and floods.

A 3 (study program: S_P, S_GCE, S_HEE; between) × 2 (WtA measure: PePav_Ø-W-W vs. PePav_N, within-participants independent variable) ANOVA was conducted to study whether participants’ answers on the general WtA PePav_Ø-W-W and the WtA PePav_N measures depended on their study program. Also, a 2 (experiences with recycling, environmental protection, construction, and floods all had two levels: yes vs. no) × 2 (WtA measure: PePav_Ø-W-W vs. PePav_N, within-participants independent variable) mixed ANOVA was repeated four times to examine whether participants’ reported experiences affect their answers on two different WtA measures. Significant main effects were followed up with pairwise comparisons using the Bonferroni-Holm procedure (only in the case of the study program with three levels). Significant interaction effects were planned to be probed by studying group differences on (a) WtA for PePav_Ø-W-W and (b) WtA for PePav_N.

Table 3. The WtA PePav and neutral measures’ mean scores and standard deviations, depending on their study program.

Variable	Study Group	WtA PePavØ-W-W		WtA PePavN
		M	SD	M	SD
Study program	SP	6.24	1.84	6.04	2.02
	SGCE	5.52	1.82	5.20	2.10
	SHEE	5.42	1.82	5.26	1.95

and

Table 4. The WtA PePav and neutral measures’ mean scores and standard deviations, depending on experiences with recycling, environmental protection, construction, and floods.

Variable	Previous Experience	WtA PePavØ-W-W		WtA PePavN
		M	SD	M	SD
Recycling experience	Yes	5.89	1.82	5.66	2.06
	No	5.69	1.94	5.46	2.08
Environmental protection activities and experience	Yes	5.93	1.78	5.78	2.04
	No	5.59	2.01	5.19	2.08
Construction experience	Yes	5.87	1.85	5.80	1.97
	No	5.74	1.88	5.20	2.19
Flood experience	Yes	5.90	1.73	5.66	1.96
	No	5.79	1.91	5.57	2.11

present the descriptive statistics of these two measures across participants’ study programs and experiences.

A study program main effect was found, with F(2, 384) = 8.07, p < 0.001, and η²_p = 0.04. S_p demonstrated higher WtA than S_HEE, t(378) = 3.24, p = 0.002, d = 0.42, and S_GCE, t(387) = 3.43, p = 0.002, d = 0.40. There was no difference between S_GCE and S_HEE (t(378) = 0.09, p = 0.931, d = 0.01). People who took part in environmental protection activities had higher WtA than people who did not (F(1, 386) = 4.82, p = 0.029, η²_p = 0.01). Participants did not differ based on whether they had experienced recycling (F(1, 386) = 0.90, p = 0.343, η²_p < 0.01), construction (F(1, 386) = 3.11, p = 0.079, η²_p = 0.01), or floods (F(1, 386) = 0.20, p = 0.654, η²_p < 0.01). Regardless of the PePav type (project proposed solutions/neutral) utilized to evaluate WtA, the main effect of the WtA measure was not found. When comparing PePav_Ø-W-W to PePav_N, participants consistently displayed similar WtA regardless of study programs (F(1, 386) = 1.38, p = 0.241, η²_p < 0.01), recycling experience (F(1, 386) = 1.33, p = 0.249, η²_p < 0.01), environmental protection (F(1, 386) = 1.35, p = 0.247, η²_p < 0.01), construction experience (F(1, 386) = 1.34, p = 0.247, η²_p < 0.01), or experience related to flooding (F(1, 386) = 1.33, p = 0.250, η²_p < 1.01). Additionally, the interaction effect was therefore not significant for study programs (F(1, 386) = 0.05, p = 0.952, η²_p < 0.01), experience with recycling (F(1, 386) < 0.01, p = 0.995, η²_p < 0.01), environmental protection (F(1, 386) = 0.32, p = 0.572, η²_p < 0.01), experience with construction (F(1, 386) = 1.29, p = 0.258, η²_p < 0.01), or experience related to flooding (F(1, 386) < 0.01, p = 0.983, η²_p < 0.01).

3.2. Study 2

Study 2 aimed to replicate the internal structure of the final WtA measure from Study 1 using an enlarged S_GCE. Additionally, the Study 1 description of neutral implied that it is made from asbestos, which some participants might find repulsive. Therefore, another aim of Study 2 was to replicate Study 1 results using a different neutral definition. Neutral was presented to the respondents in this study as a powdered material obtained by additional processing of industrial waste and classified as non-hazardous waste according to the regulations of the Republic of Serbia. As in Study 1, the student participation pool was used to recruit participants. The sample consisted of 187 participants, and no participants had missing values.

Table 5. Characteristics of the student samples from Study 2.

Variable		N (%)
Gender	Male	50 (26)
	Female	137(73)
Study program	SP	84 (45)
	SGCE	103 (55)
Recycling experience	Yes	147 (79)
	No	40 (21)
Environmental protection activities and experience	Yes	142 (76)
	No	45 (24)
Construction experience	Yes	44 (24)
	No	143 (76)
Flood experience	Yes	59 (32)
	No	128 (68)

presents other sample characteristics relevant to this study.

The Study 2 sample was somewhat younger on average than the participants in Study 2 (M = 20.79, SD = 6.13). The samples from both studies were similar in gender composition, recycling, environmental protection, and flood experiences. However, fewer participants from Study 2 reported having construction experiences than in Study 1. One possible explanation is that Study 2 asked participants whether they actively participated in construction or remodeling, while participants in Study 1 answered whether they actively participated in construction work and remodeled their work or living space as a separate question. Because the Study 1 question is a more occurrent behavior, one could expect that more individuals would answer that they had this experience. Also, due to changes in the CE curriculum structure, the potential size of the S_HEE sample at the time of conducting Study 2 did not meet the strict requirements for the intended complex statistical analyses. The obtained sample size was sufficient for exploratory factor analysis implementation based on the general rules of thumb [157], the number of participants/items ratio [158,159], and observed correlation coefficients (rs > 0.57). The results of a sensitivity power analysis suggested that a 2 (between) × 2 (within) ANOVA would detect small-to-medium main effects and interactions (f < 0.21) considering the sample size, α = 0.05, and power = 0.80. The study was administered on an online platform (https://www.1ka.si/) (accessed on 7 November 2023), and its procedure did not differ from that described in Study 1. After giving consent, participants filled out the two WtA measures and answered demographic questions.

Aside from differences in collecting data about participants’ construction experiences, Study 2 also administered a different neutral description. Given that people generally hold strong negative attitudes toward asbestos [169,170], any mention of asbestos was removed from the description to reduce the likelihood of shifting participants’ responses toward less positive answers. With this neutral description it was possible to obtain a more reliable answer to the key question: is the Ø-W-W-proposed PePav solution perceived as a meaningful advancement compared to a hypothetical, technologically derived sustainable alternative? Hence, this procedure aimed to test whether variations in material descriptions could influence participants’ WtA responses.

The psychometric properties of the WtA PePav measure in Study 2 were evaluated to determine if the unidimensionality, reliability, and cumulative structure observed in Study 1 were replicated. To assess the internal structure, a robust exploratory factor analysis (EFA) was performed on a polychoric correlation matrix. Both the classical and optimal implementations of parallel analysis suggested a one-factor solution, which explained 75% of the data variance. All items exhibited acceptable factor loadings (

Table 6. Descriptive statistics, item loadings, and item-specific coefficients of scalability for the final five WtA PePav measure items for Study 2.

I Would Use PePavØ-W-W to Cover:	M	SD	Min.	Max.	Sk (SE)	Ku (SE)	Item Loadings *	Hi
Item 1: all streets and roads	6.33	1.71	1	8	−0.71 (0.14)	−0.54 (0.34)	0.74	0.70
Item 2: streets and roads in flooded areas	5.59	1.74	1	8	−0.31 (0.12)	−0.66 (0.19)	0.82	0.79
Item 3: my street	5.97	1.74	1	8	−0.49 (0.14)	−0.63 (0.27)	0.94	0.90
Item 4: my college/work environment	6.29	1.55	1	8	−0.38 (0.12)	−1.14 (0.14)	0.86	0.82
Item 5: my yard	5.77	1.89	1	8	−0.49 (0.12)	−0.74 (0.20)	0.79	0.76

* Reported item loadings reflect the loadings of each item on the WtA PePav measure factor. Item loadings were computed via an exploratory factor analysis.

and Figure 3), and no redundant items were identified, successfully replicating the unidimensional structure found in the previous study. This single-factor model demonstrated excellent internal consistency with a Cronbach’s alpha of 0.90.

Beyond the verification of dimensionality and reliability, the cumulative nature of the scale was again tested via Mokken Scale Analysis. The AISP consistently suggested a one-dimensional, cumulative scale under thresholds ranging from 0.00 to 0.55. Evidence for the hierarchical structure was strong, with inter-item H_ij coefficients ranging from 0.57 to 0.80. Furthermore, both the item-specific H_i coefficients (Table 6 and Figure 3) and the overall H coefficient (0.66) exceeded the 0.50 threshold for strong scalability. These findings confirm that the hierarchical response pattern remains robust in Study 2, indicating that the WtA PePav measure functions as a reliable and valid cumulative instrument.

Descriptive statistics for the measure across different study programs and experience levels are provided in

Table 7. WtA PePav and neutral measures’ mean scores and standard deviations depending on study program and experiences with recycling, environmental protection, construction, and floods.

Variable		WtA PePavØ-W-W		WtA PePavN
		M	SD	M	SD
Study program	SP	6.25	1.39	6.07	1.39
	SGCE	5.78	1.47	5.48	1.50
Recycling experience	Yes	6.10	1.41	5.86	1.44
	No	5.60	1.56	5.32	1.56
Environmental protection activities and experience	Yes	6.04	1.41	5.75	1.44
	No	5.84	1.57	5.73	1.62
Construction experience	Yes	5.95	1.51	5.81	1.62
	No	6.00	1.44	5.72	1.44
Flood experience	Yes	6.12	1.35	5.74	1.62
	No	5.93	1.50	5.75	1.42

.

Study 2 used the same analytical approach to make group comparisons as in Study 1. Psychology students showed greater levels of WtA than their engineering colleagues, F(2, 370) = 12.68, p < 0.001, η²_p = 0.03. Additionally, students who reported previous engagement in recycling activities presented higher WtA scores than those without such involvement, F(1, 370) = 8.09, p = 0.005, η²_p = 0.02. No significant differences were found with respect to prior participation in environmental protection or construction activities, F(1, 370) = 0.36, p = 0.550, η²_p < 0.01; F(1, 370) = 0.01, p = 0.916, η²_p < 0.01, respectively, or flood-related experience, F(1, 370) = 0.30, p = 0.583, η²_p < 0.01.

The WtA measure’s main effect was not observed, regardless of the PePav type (project proposed solutions/neutral) used to assess WtA. Participants consistently showed similar WtA when comparing PePav_Ø-W-W to PePav_N, irrespective of study programs (F(1, 370) = 2.71, p = 0.100, η²_p = 0.01), recycling experience (F(1, 370) = 2.68, p = 0.103, η²_p = 0.01), environmental protection (F(1, 370) = 2.63, p = 0.106, η²_p = 0.01), construction experience (F(1, 370) = 2.63, p = 0.106, η²_p = 0.01), or experience with flooding (F(1, 370) = 2.63, p = 0.106, η²_p = 0.01). Likewise, the interaction effect was not significant for study programs (F(1, 370) = 0.15, p = 0.694, η²_p < 0.01), recycling experience (F(1, 370) = 0.01, p = 0.920, η²_p < 0.01), environmental protection (F(1, 370) = 0.27, p = 0.606, η²_p < 0.01), construction experience (F(1, 370) = 0.16, p = 0.689, η²_p < 0.01), or flooding experience (F(1, 370) = 0.39, p = 0.530, η²_p < 0.01). The findings suggest that the absence of HEE student participation appears not to have affected the results, with Study 2 largely replicating those of Study 1.

3.3. Study 3

In Study 3, the measure was validated beyond a student sample, using a larger sample from the general population living in Serbia. First, Study 3 aimed to examine whether the proposed single-factor model of PePav acceptance also fits the data collected from a general population using a CFA approach. Second, the focus was on the differences between recycling, environmental protection, construction, and flood experiences of the Study 3 sample. Third, Study 3 provided evidence of the measure’s external validity, the extent to which the WtA measure correlates with other related constructs and predicts relevant outcomes. Specifically, Study 3 examined the relationship between WtA, intentions to use PePav, and three intention components from the TPB [109,115].

All participants over 18 years old living in Serbia and fluent in Serbian were eligible for participation. Data was collected by sharing the online questionnaire link (https://www.1ka.si/) (accessed on 22 November 2023) on social media and online forums and using snowball sampling. A total of 625 participants gave their consent and successfully finished the study (

Table 8. Characteristics of the samples from Study 3.

Variable		N (%)
Gender	Male	259 (41)
	Female	366 (59)
Construction background	Yes	99 (16)
	No	526 (84)
Recycling experience	Yes	445 (71)
	No	180 (29)
Environmental protection activities and experience	Yes	473 (76)
	No	152 (24)
Construction experience	Yes	260 (42)
	No	365 (58)
Flood experience	Yes	181 (29)
	No	444 (71)

). No participants were excluded, in line with the predefined exclusion criteria (Study 1). On average, participants were 31 years old (M = 31.28, SD = 14.28). Two hundred fifty-nine participants identified as males, while females composed the rest of the sample. 99 participants came from a construction background (architects, construction engineers, project managers, and construction workers).

Following the Hair et al. [157] guidelines, the sample size was above the suggested N for CFA models with five or fewer constructs, more than three items per construct, and high item communalities (>0.60), indicating that the Study 3 sample was adequate to test the single-factor model. Sensitivity power analysis suggested that a 2 (between) × 2 (within) ANOVA could detect relatively small main effects and interactions (f < 0.11) considering the sample size, α = 0.05, and power = 0.80. The study was administered on an online platform (https://www.1ka.si/) (accessed on 22 November 2023). The questionnaire lasted approximately 15 min. After giving consent, participants were asked to complete the WtA measure, attitudes, SN, PC, and intentions measures. Then, participants answered demographic questions (gender, age, recycling experience, participation in environmental protection activities, experiences in construction work, and flood experience).

The WtA measures were the same as in Study 2. The remaining measures are: 1) PePav attitudes (α = 0.85). Following research measuring attitudes in the ecological sustainability sphere [113,171,172,173] and the TBP tradition [115,174,175], it was decided to measure PePav attitudes via a semantic differential measure. Participants read a list of eight different adjective pairs (e.g., risky-safe) presented as two endpoints on a continuum from left to right. Participants were asked to rate PePav on a 1–5 scale (1 = the adjective on the left completely describes PePav, 5= the adjective on the right completely describes PePav). 2) PePav subjective norms (α = 0.80). The SN measure typically used in TPB research [109,174,176] was adapted to the PePav context. Participants answered how much other people (family, friends, colleagues, present/future boss) would approve of using PePav using a 1–5 scale (1 = would not approve at all, 5 = would fully approve). PePav perceived control (α = 0.81). Once again, the PC measure from the TPB tradition [109,174,176] was adapted to PePav. Participants read five items (e.g., It is easy for me to apply PePav.) and answered using a 1–5 scale (1 = completely disagree, 5 = completely agree). Intentions to use PePav (α = 0.88). Participants read three items (e.g., I want to use PePav.) and answered using a 1–5 scale (1 = completely disagree, 5 = completely agree).

Table 9. Descriptive statistics, item loadings, and item-specific coefficients of scalability for the final five WtA PePav measure items for Study 3.

I Would Use PePavØ-W-W to Cover:	M	SD	Min.	Max.	Sk (SE)	Ku (SE)	Item Loadings *	Hi
Item 1: all streets and roads	6.12	1.84	1	8	−0.78 (0.07)	−0.20 (0.17)	0.79	0.64
Item 2: streets and roads in flooded areas	5.54	1.90	1	8	−0.29 (0.06)	−0.77 (0.09)	0.87	0.71
Item 3: my street	5.86	1.86	1	8	−0.53 (0.06)	−0.63 (0.12)	0.91	0.73
Item 4: my college/work environment	6.02	1.72	1	8	−0.59 (0.07)	−0.26 (0.16)	0.87	0.70
Item 5: my yard	5.67	1.92	1	8	−0.55 (0.06)	−0.40 (0.11)	0.82	0.66

* Reported item loadings reflect the loadings of each item on the WtA PePav measure factor. Item loadings were computed via an exploratory factor analysis.

summarizes the mean scores and standard deviations of the five PePav items (as illustrated in Figure 4). The item means and standard deviations did not differ much from the estimated descriptive statistics in Studies 1–2. Once again, the participants used the full range of available responses.

Participants’ answers gravitated toward larger responses, with mean item values ranging from 5.54 to 6.12. This was also corroborated by a significant Skewness/standard error ratio (z > |2.58|) for all items. Furthermore, a significant Kurtosis/standard error ratio (z > |2.58|) was observed for items 2, 3, and 5.

The psychometric evaluation of the WtA PePav measure in Study 3 aimed to replicate previous findings within a general population sample using a CFA approach. Following the unidimensional theoretical assumption of the SDS [10], the fit of a single-factor measurement model was tested. To account for the ordinality and non-normality of the five WtA PePav items, the weighted least squares mean and variance-adjusted estimator (WLSMV) was used, as it does not assume normally distributed exogenous variables [157]. Model fit was evaluated through the Dynamic Fit Index (DFI) framework [177], which constructs fit cut-offs [178,179] tailored to the specific model context (e.g., sample size, number of items, and factor reliability). As presented in

Table 10. Model and dynamic fit indices generated on Levels 1 and 2 of misspecification (Study 3).

Fit Index	CFI	RMSEA	SRMR
Model	0.992	0.118	0.022
Level 1 (small model misspecification)	0.991	0.084	0.015
Level 2 (medium model misspecification)	0.981	0.127	0.022

, the CFI was higher than the Level 1 cut-off, indicating a close fit, while the SRMR met the Level 2 cut-off, indicating a fair fit [166]. Although the RMSEA fell between Levels 1 and 2, this was interpreted as a fair fit within the DFI framework, which provides a direct explanation for why the value exceeds traditional rules without invalidating the model; specifically, RMSEA frequently overestimates misspecification in parsimonious models with small degrees of freedom [180,181]. Collectively, the excellent CFI and acceptable SRMR, along with the DFI-contextualized RMSEA, substantiate the validity of the single-factor structure.

Evidence for the scale’s reliability and homogeneity was likewise robust. The composite reliability (CR = 0.81) and average variance extracted indices (AVE = 0.73) were above the recommended cut-offs of 0.70 and 0.50, respectively [157], suggesting that all items represent the same latent factor. In line with these findings, the item-total correlations ranged between 0.71 and 0.91, suggesting that all items contributed significantly to scale homogeneity [182]. Finally, the cumulative properties of the scale were successfully replicated in the general population sample. Mokken Scale Analysis confirmed the hierarchical structure, with inter-item H_ij coefficients ranging from 0.57 to 0.78. Both the item-specific H_i coefficients (Table 9) and the overall H coefficient (0.69) were strong, exceeding the 0.50 threshold for high scalability. These results confirm that the hierarchical response pattern remains stable across samples, supporting the interpretability of the WtA PePav measure as a valid cumulative instrument.

A series of nested multi-group confirmatory factor analysis (CFA) models for categorical data was used to evaluate the measurement invariance of the WtA PePav instrument. Invariance testing was conducted across two grouping variables: Sample Type (General Population versus Student) and Construction Experience (Yes versus No). Following the methodology outlined in [157], measurement invariance was examined at three hierarchical levels. Initially, configural invariance was established to verify that the unidimensional factor structure is conceptualized consistently across groups. Subsequently, metric invariance was tested by constraining factor loadings to equality, thereby confirming that the relationship between individual items and the latent construct is equivalent across groups. Finally, scalar invariance was assessed to ensure equivalence of item thresholds, indicating that individuals with the same latent trait level have an equal likelihood of endorsing a given response category irrespective of group membership. Model fit was primarily evaluated using the Comparative Fit Index (CFI) and the Standardized Root Mean Square Residual (SRMR). Consistent with [157], invariance was supported if changes in CFI were less than 0.010.

The results for Sample Type are presented in

Table 11. Measurement Invariance Fit Indices for Sample Type and Construction Experience.

Grouping Variable	Model	CFI	SRMR	RMSEA
Sample Type	Configural	0.987	0.018	0.104
	Metric	0.986	0.021	0.091
	Scalar	0.992	0.019	0.110
Construction Experience	Configural	0.985	0.019	0.112
	Metric	0.981	0.026	0.105
	Scalar	0.990	0.020	0.118

. The configural model demonstrated excellent fit, confirming a common factor structure. Metric invariance was supported, as changes in the Comparative Fit Index (ΔCFI = −0.001) and the Standardized Root Mean Square Residual (ΔSRMR = 0.002) were negligible. Scalar invariance was also established. The fit indices improved in the most parsimonious model, with changes remaining below the acceptable thresholds (ΔCFI = 0.006, ΔSRMR = −0.002). Results for Construction Experiences are also detailed in Table 11. Configural invariance was successfully established. The model achieved metric invariance (ΔCFI = −0.004, ΔSRMR = 0.006), indicating equivalent factor loadings across groups. Scalar invariance was further supported, with the scalar model exhibiting a better fit than the metric model (ΔCFI = 0.009, ΔSRMR = −0.006). Across both grouping variables, the WtA PePav measure scale demonstrated strong (scalar) invariance, indicating that the measurement instrument is unbiased across these groups and allowing for meaningful comparisons of latent means.

In Table 11, CFI refers to the Comparative Fit Index (higher values indicate better fit), RMSEA refers to the Root Mean Square Error of Approximation (lower values indicate better fit), and SRMR refers to the Standardized Root Mean Square Residual (lower values indicate better fit). To conclude, the data supports an acceptable fit of the one-factor PePav acceptance model.

3.4. Differences in Recycling, Environmental Protection, Construction, and Flood Experiences

Assuming the same approach as in Studies 1 and 2, differences in the WtA towards PePav and neutral based on their recycling, environmental protection, construction, and flood experiences were examined. A main effect of the WtA measure was observed for every type of examined relevant experience. Participants consistently showed higher WtA when considering PePav_Ø-W-W compared to PePav_N, regardless of whether they had a construction vocational background or general construction experience, as well as experience with environmental protection activities, recycling, or flooding (F(1, 1246) = 5.85, p = 0.016, η²_p = 0.01). Participants with a construction background (architects, construction engineers, project managers, and workers) reported lower levels of WtA than other participants (F(2, 1246) = 17.71, p < 0.001, η²_p = 0.01). Practical experience with construction work, irrespective of construction background, did not significantly affect WtA (F(1, 1246) = 3.11, p = 0.078, η²_p < 0.01). Participants with experience in environmental protection activities had higher WtA scores than those without (F(1, 1246) = 15.91, p < 0.01, η²_p < 0.01), as did those with experience in recycling (F(1, 1246) = 18.00, p < 0.001, η²_p = 0.01). In contrast, participants’ WtA concerning the use of PePav did not differ based on flood-related experience (F(1, 1246) = 2.54, p = 0.111, η²_p < 0.01).

Table 12. The WtA PePav and neutral measures’ mean scores and standard deviations, depending on their study program and experiences with recycling, environmental protection, construction, and floods.

Variable		WtA PePavØ-W-W		WtA PePavN
		M	SD	M	SD
Construction background	Yes	5.36	1.76	5.21	1.92
	No	5.93	1.53	5.70	1.61
Construction experience	Yes	5.73	1.70	5.54	1.79
	No	5.92	1.49	5.68	1.59
Environmental protection activities and experience	Yes	5.96	1.54	5.71	1.67
	No	5.48	1.65	5.34	1.65
Recycling experience	Yes	5.96	1.53	5.74	1.68
	No	5.54	1.67	5.31	1.64
Flood experience	Yes	5.92	1.49	5.68	1.59
	No	5.95	1.64	5.74	1.74

contains the descriptive statistics of the PePav measures computed in Study 3.

As in Studies 1 and 2, the interaction effect was not significant for construction professional, vocational, and educational background (F(1, 1246) = 0.13, p = 0.719, η²_p < 0.01), recycling experience (F(1, 1246) < 0.01, p = 0.959, η²_p < 0.01), environmental protection (F(1, 1246) = 0.22, p = 0.641, η²_p < 0.01), construction experience (F(1, 1246) = 0.06, p = 0.881, η²_p < 0.01), or flood-related experience (F(1, 1246) = 0.01, p = 0.930, η²_p < 0.01).

Table 13. Correlations between WtA PePav_Ø-W-W, attitudes, subjective norms, perceived control, and intentions to use PePav.

Variable	1	2	3	4
WtA PePav (1)
PePavØ-W-W attitudes (2)	0.58 ***
PePavØ-W-W subjective norms (3)	0.55 ***	0.59 ***
PePavØ-W-W perceived control (4)	0.11	0.12 *	0.18 **
Intentions to use PePav (5)	0.43 ***	0.47 ***	0.49 ***	0.38 ***

* = p < 0.05. ** = p < 0.01. *** = p < 0.001.

and Figure 5 present the correlations between WtA PePav and the TPB constructs. As expected, WtA PePav correlated positively and moderately with PePav attitudes and SN, supporting external convergent validity, that is, the extent to which the measure correlates with theoretically related constructs. Conversely, the small and non-significant correlation between WtA PePav and PBC may indicate external divergent validity, reflecting the measure’s limited association with less related constructs.

A three-step hierarchical regression analysis was implemented to provide evidence of incremental criterion validity (the extent to which a measure can explain variance in a meaningful outcome accounting for other predictors) for the WtA PePav measure (

Table 14. Results of a 3-step hierarchical multiple regression analysis predicting intentions to use PePav.

Predictors	Model 1				Model 2				Model 3
	b	β	t(617)	r2sp	b	β	t(509)	r2sp	b	β	t(506)	r2sp
Intercept	2.81		16.70 ***		−0.56		−2.42 *		−0.58		−2.53 *
Construction B	0.08	0.03	0.66	<0.01	−0.01	0.00	−0.08	<0.01	−0.05	−0.02	−0.54	<0.01
EXPrecycl	−0.22	−0.10	−2.32 *	0.01	−0.06	−0.03	−0.80	0.01	−0.06	−0.03	−0.81	<0.01
EPA	−0.25	−0.11	−2.57 *	0.01	−0.18	−0.08	−2.25 *	<0.01	−0.16	−0.07	−2.02 *	<0.01
EXPcon	−0.11	−0.05	−1.04	<0.01	−0.03	−0.02	−0.38	<0.01	−0.04	−0.02	−0.47	<0.01
EXPflood	0.08	0.04	0.94	<0.01	0.09	0.04	1.30	<0.01	0.10	0.04	1.41	<0.01
Gender	0.16	0.08	1.92	0.01	0.10	0.05	1.53	<0.01	0.10	0.05	1.48	<0.01
Age	0.01	0.17	3.62 ***	0.02	0.01	0.11	2.98 **	0.01	0.01	0.11	3.09 **	0.01
PePavØ-W-W At					0.25	0.26	6.72 ***	0.04	0.19	0.20	4.84 ***	0.02
PePavØ-W-W SN					0.33	0.26	6.60 ***	0.04	0.27	0.22	5.23 ***	0.03
PePavØ-W-W PC					0.30	0.28	8.60 ***	0.07	0.30	0.28	8.62 ***	0.07
WtA PePavØ-W-W									0.09	0.15	3.68 ***	0.01
	R2 = 0.07 *** R2adj = 0.06				R2 = 0.40 *** R2adj = 0.39 ΔR2 = 0.32 ***				R2 = 0.41 *** R2adj = 0.40 ΔR2 = 0.01 ***

* = p < 0.05. ** = p < 0.01. *** = p < 0.001. PePav_Ø-W-W and At–PePav_Ø-W-W are attitudes. PePav_Ø-W-W and SN–PePav_Ø-W-W are subjective norms. PePav_Ø-W-W and PC–PePav_Ø-W-W are perceived control. WtA PePav_Ø-W-W is Willingness to Accept PePav_Ø-W-W. b is the unstandardized regression coefficient. β is the standardized regression coefficient. t is the value of the statistical test testing the significance of the regression coefficient. r²_sp is the squared semipartial correlation coefficient. R² is the coefficient of multiple determination. R²_adj is the corrected coefficient of multiple determination. ΔR² is R² change in the larger model.

).

Specifically, this approach tested whether WtA regarding PePav predicted intentions to use PePav beyond demographic variables and TPB intention components. Model 1, which included a set of demographic control variables (age, gender, construction background, recycling, environmental protection, construction, and flood experiences) as predictors of intentions to use PePav_Ø-W-W, was significant, F(3, 617) = 6.84, p < 0.001. Including the three TPB intention components additionally explained 32% of the variance in the outcome, F(3, 618) = 109.64, p < 0.001. As expected, all three components positively predicted intentions to use PePav_Ø-W-W. Finally, Model 3, including WtA PePav, further explained another 1% of the variance, F(1, 617) = 13.56, p < 0.001. Therefore, WtA PePav_Ø-W-W positively predicted intentions to use PePav_Ø-W-W (albeit to a lesser degree) over and above demographic variables and the TPB intention components.

4. Discussion

A discussion based on the most significant results is given below. The current set of research examined the internal structure, convergent, divergent, and incremental criterion validity of a novel measure of WtA PePav.

4.1. Structure and Reliability of the WtA Measure

In both student (Studies 1 and 2) and general population samples (Study 3), the five-item WtA scale solution demonstrated a clear one-factor structure, aligning with theoretical expectations based on the SDS Model [10]. Exploratory and confirmatory factor analyses consistently supported this unidimensional structure, and the internal consistency of the scale was excellent (α = 0.90–0.92). Additionally, the scale showed excellent cumulativity properties across all three studies. The fact that the hierarchical ordering of items remained invariant suggests that the underlying logic of social proximity is a stable construct within this cultural context, regardless of an individual’s specific experience in the construction sector or their educational background. The results are consistent with research based on the Bogardus scale applied in various fields, such as studies on the pandemic [107] and attitudes toward individuals with mental illnesses [183,184] and different ethnic groups and immigrants [161,185], as well as other social groups, such as the homeless [186].

4.2. Replication and Impact of Neutral Description

A potential strength of this research lies in the replication of the measure’s psychometric properties using different descriptions of the material neutral, which participants were informed could be used to construct PePav. In Study 2, the explicit reference to asbestos was removed to assess whether wording affected responses. Despite this change, the scale retained its structure and reliability, and participants’ responses remained consistent with those observed in Study 1. This suggests that participants’ WtA toward PePav is relatively stable and not easily altered by minor variations in material descriptions, reinforcing the robustness of the measure. In environmental and sustainability research, it is common to use the same questionnaire while varying the tested solution or intervention. This enables comparative analysis of different sustainable innovations while maintaining the reliability and consistency of measurement. Developing universal, technologically agnostic instruments [187,188,189] and integrated assessment models [189,190,191] applicable across various contexts has been considered a means of facilitating a more systematic evaluation of sustainability-oriented competencies. For example, such an approach has been applied in studies on the evaluation of blue–green versus grey flood prevention measures [30]. It has also been utilized in assessments of awareness and barriers to sustainable practices [192], where a mixed-methods design complemented self-report questionnaires that are susceptible to self-report bias.

4.3. Group Differences in WtA

Participants from a general population sample with construction-related backgrounds reported lower WtA scores. Previous research highlights the multidimensional and non-linear nature of this relationship [193]. Experiences with failed implementations and limited effectiveness of similar solutions can diminish openness to new interventions [194,195,196,197,198]. Professionals who perceive innovative solutions as having limited effectiveness relative to costs, technical requirements, or local conditions are less likely to support them [199,200,201]. Risk perception also influences professionals’ willingness to support preventive measures. Those who assess flood risk as low or believe more efficient and less demanding alternatives exist often demonstrate reduced WtA regarding proposed interventions [198,202,203,204]. The institutional context of decision-making can indirectly shape professional attitudes, as experts evaluate solutions within the regulatory framework, available resources, planning procedures, and the distribution of responsibilities among actors. These factors can affect perceptions of limitations, potential risk, or practical feasibility [205,206,207]. Collectively, these findings underscore the complexity of factors influencing experts’ attitudes toward self-sustaining innovative solutions. Professional experience, risk perception, and institutional context may contribute to various mechanisms of reduced willingness, indicating a need for further research into the multidimensional determinants of professionals’ attitudes. Similarly, Studies 1 and 2 identified distinct differences among study programs with respect to construction background. Specifically, S_P reported greater WtA than their colleagues from S_GCE and S_HEE. Hence, these two studies also suggest that expertise has an important role in how individuals perceive PePav: construction-related expertise might increase skepticism towards PePav. What is certain, though, is that clear differences regarding attitudes toward PePav already emerge in the aforementioned pilot study [114] at the level of two student groups with different educational backgrounds (psychology/civil engineering), which carries important implications. The future civil engineers expressed more doubts regarding whether the proposed PePav solution was safe, useful, practical, harmless, and truly needed. Actually, even students from two different civil engineering study groups (general civil engineering vs. hydraulic and environmental engineering) reported significantly different restraints in this regard. Nonetheless, students of civil engineering still perceived the proposed PePav solution as equally resistant as did the psychology students. The absence of a significant association between construction or remodeling experience and WtA may be attributed to the ways in which professionals develop attitudes toward innovative, sustainable solutions. These attitudes are shaped through prolonged specialized education and structured practice [97,139]. The findings suggest that unstructured, informal experience, without didactic principles or professional engagement, does not influence these attitudes.

Group comparisons across the three studies indicated that prior experiences with recycling and environmental protection were generally positively correlated with higher willingness to adopt sustainable, innovative construction solutions. This outcome is consistent with the broader body of literature on pro-environmental behavior, which suggests that prior engagement is often associated with increased openness to sustainable innovations. Specifically, individuals who regularly engage in recycling tend to exhibit heightened awareness of environmental issues and demonstrate a greater propensity to embrace new technologies and practices that promote sustainability [85,98,208]. The observed lack of a significant association between flood-related experience and willingness to adopt PePav solutions contrasts with prior research indicating that disaster experience typically increases risk perception and motivation for protective behavior [209,210,211]. This result can be understood through the lens of the “risk perception paradox,” which posits that heightened risk awareness does not always lead to protective action or the adoption of innovative NbS [212]. The relationship between disaster experience, risk perception, and behavioral responses may be mediated by several factors. Experiencing floods does not necessarily ensure that individuals understand the protective functions of NbS or ecosystem-based solutions [213]. Many citizens associate flood protection primarily with conventional engineering infrastructure rather than with blue–green interventions such as PePav [214]. Limited public familiarity with ecosystem-based flood mitigation strategies may therefore weaken the relationship between past flood experience and behavioral intentions [213,215]. Beyond individual experience, institutional trust plays a critical mediating role in shaping preparedness and protective behavior. Trust in authorities shapes how individuals interpret risk information, assess the credibility of proposed measures, and define their own responsibility for action. Consequently, institutional trust may prove a stronger predictor of protective behavior than prior disaster experience alone [212,216]. Additionally, the characteristics of local flood events, which may be episodic or moderate in intensity, can shape how risk is perceived and attributed [217]. Floods in Serbia exhibit significant heterogeneity due to the unique combination of topographical features, ranging from the Pannonian plain to mountain ranges, as well as varying climatic influences and anthropogenic factors. Consequently, Serbia faces diverse flood risks, including slow river floods in the north and severe torrential flows in the central and southern regions [218]. For these reasons, the questionnaire explicitly defined pluvial floods to establish a clear frame of reference for respondents when assessing their readiness to adopt PePav in flood prevention.Another finding of the preliminary study [114] indicated that, particularly among psychology students, having a close relative affected by flooding was associated with a greater WtA for innovative solutions than being directly affected oneself. This may reflect the humanistic values and orientation toward caring of these students. Thus, subjective factors may contribute to decision-making in this group. However, these findings should be interpreted with caution due to the pilot study’s limited sample size, design, and the indirect nature of this observation.

Recent publications highlight ongoing methodological challenges in evaluating the integration of sustainability within disciplines such as engineering and construction [219,220,221,222,223], largely attributable to the lack of standardized and validated assessment instruments [224,225,226]. This deficiency impedes the comparability of findings and delays the systematic assessment of sustainability-related competencies across professional sectors. The WtA scale emerges as a promising tool, demonstrating sensitivity to expert knowledge and experience, robust psychometric properties, and cost-effectiveness, while remaining accessible for research involving industry professionals. The objective of this early phase of WtA scale testing was to adapt the SDS, a reliable and well-established instrument, in accordance with transdisciplinary research principles. This study aimed to assess its psychometric properties, its sensitivity to factors like sociodemographic background and relevant experience, and its validity compared to TPB. Results from all three studies support the importance of professional knowledge and experience. This highlights the recognized need to involve key stakeholder groups—including decision-makers, experts, private sector representatives, civil society organizations, and end-users—in research aimed at ensuring the sustainability of technological innovations [75,80,84,92,96,99,113,173,196,197,198,199,200,201,202,203,204,205,206,207]. The generalizability of these results is determined by the fact that the data were collected only from a Serbian population. Previous studies have shown that cultural differences can significantly affect outcomes in sustainable engineering [11,97,112,116,146,198].

4.4. Relationship with TPB Constructs and Behavioral Intentions Measurement

Study 3 contributed to the validation of the scale by contextualizing it within the TPB [109]. The findings revealed a moderate correlation between WtA and PePav attitudes and SN, indicating strong convergent validity. This correlation implies that the assessment tool correlates meaningfully with other measures aimed at evaluating similar constructs, thus reinforcing the premise that diverse methodologies for assessing a particular concept yield coherent results, supporting the idea that they are truly measuring the same construct [227]. The observed weaker association with PBC is a relatively common finding in sustainability research [228,229,230]. It may suggest that participants perceive inherent limitations regarding their ability to engage with PePav, indicating that attitudes and norms may exert a more substantial influence on acceptance than perceived feasibility. This aligns with established findings in the literature under the TPB framework, which consistently highlights the predominant role of attitudes and norms [231,232,233].

Differential roles of TPB variables in sustainability research have been documented, and Yuriev et al. [125] articulated significant concerns regarding the adequacy and operational efficacy of the TPB for predicting green behaviors. The complexity of the phenomena being investigated suggests that additional factors, which influence pro-environmental behaviors, may not be encompassed within the TPB framework [125,234,235]. This study also corroborates these findings by showing that WtA might be an additional construct explaining sustainability intentions. In line with theoretical reasoning [54,64,65,66,67,99], Study 3 showed that WtA and intention are closely related but distinct, with WtA potentially predicting the formation of intention. Importantly, WtA contributed uniquely to the prediction of BI to use PePav, beyond traditional TPB components and demographic factors. Although the incremental variance explained was modest (ΔR² = 0.01), this finding supports the scale’s predictive validity and suggests that WtA captures a distinct motivational aspect relevant to sustainable material adoption. Regarding the predictive utility of the adapted SDS, it is important to note that the incremental variance of the WtA measure observed in Study 3, while statistically significant, represents a small effect size according to traditional benchmarks [236]. However, this result should be evaluated within the context of incremental validity [237]. Because TPB already accounts for a substantial proportion of the variance in BI, any additional contribution by a new construct is necessarily constrained. Indeed, several studies reported small incremental effects of additional predictors on intention measures when controlling for TBP components [117,130,133]. Relatedly, within psychological science, it is increasingly recognized that small effects can have cumulative practical importance, especially when they pertain to large-scale civic behaviors or public policy [238]. For urban planners, even a marginal increase in the understanding of citizens’ WtA provides a more nuanced toolkit for risk assessment. The WtA scale could capture a unique dimension of psychological readiness rather than serve as a dominant predictor of intention. This assertion is supported by an acknowledgment of the intricate relationships among attitudes, such as WtA, intention, and behavior. These relationships are influenced by a myriad of contextual and individual factors, as delineated within the extensive TPB framework literature [97,99,112,124,125].

5. Limitations of the Study

While the sample included students and members of the general population from both construction and non-construction fields, with varying levels of knowledge and experience, it did not include key stakeholders. These stakeholders, such as urban planners, community representatives, and policymakers, may have offered perspectives, concerns, and motivations that were not captured. This limitation is particularly pertinent for informing educational, communication, or policy strategies aimed at implementing sustainable solutions.

Although the WtA scale demonstrated excellent psychometric properties, its application was limited to a specific domain, namely construction-related contexts with a focus on PePav. This domain-specific use may restrict the generalizability of the findings to other fields or types of interventions, and further research is needed to examine the scale’s applicability in broader or different contexts.

The study used self-reported measures of experience, attitudes, and WtA, which are susceptible to social desirability bias and recollection mistakes. It is unclear whether respondents who expressed a desire to support or acquire PePav will do so in real-world scenarios. This constraint is especially important when interpreting subjective experiences with flooding and evaluating novel remedies.

The lack of an expected link between flood-related experience and WtA may reflect participants’ difficulty translating difficult personal situations with floods into clear expectations about how specific solutions, such as PePav panels, would work. This shows a possible disconnect between real-world experience and the perceived importance of technical mitigation techniques.

Finally, the scale was validated within Serbia’s cultural limits. As a result, the findings must be evaluated within the socioeconomic context of a developing Southeast European economy.

6. Conclusions

This research confirms the value of an interdisciplinary framework for studying willingness to accept (WtA) sustainable building materials, bridging psychological, construction, architectural, and environmental perspectives. The adapted social distance-based WtA measure demonstrated strong reliability, validity, and incremental predictive power beyond established Theory of Planned Behavior constructs, capturing a distinct motivational dimension related to acceptance.

Empirical evidence from three studies supports the measure’s stable unidimensional structure and applicability across diverse populations and survey methods. Its brevity and clarity facilitate practical use in assessing public and professional attitudes toward circular construction materials. Notably, acceptance varied by background: environmental engagement correlated with higher WtA, while construction professionals showed more caution, underscoring the importance of targeted stakeholder engagement.

The WtA measure remained robust under varying informational contexts, indicating its utility in public opinion research despite message framing differences. Positioned within the TPB framework, WtA emerges as a complementary construct that precedes and enhances preliminary prediction of intentions to adopt sustainable innovations. WtA can be considered a preliminary tool for early-stage research and assessment, pending further validation in broader stakeholder groups and against real-world behavior. Overall, the WtA scale provides a theoretically sound, empirically validated, and practical tool to assess acceptance and readiness for circular building materials, informing strategies to promote sustainable innovation adoption in the built environment.

7. Recommendations

The WtA scale, evaluated through three studies detailed in this research, represents an initial instrument intended for exploratory research and assessment. Therefore, the findings should be interpreted with caution, as further validation is required across diverse subpopulations and in relation to observed real-world behaviors. Moreover, additional validation across regions varying in socio-economic features and environmental conditions could provide further insights into the scale’s robustness and the confines of the conditions of the proposed measure.

Given the WtA scale’s concise format (1–2 min), clear wording, standardized administration procedure, strong psychometric performance, and consistent response patterns, including sensitivity to respondents’ experience, proximity to application, and the specific components of innovative solutions, the scale seems to be suitable for research applications across a wide range of subpopulations. Research may benefit from testing its use among national demographics; targeted subgroups (e.g., adolescents, seniors); professional sectors (e.g., architects, construction engineers, technologists); stakeholder groups (e.g., workforce, managers, university faculty, researchers, policymakers, end-users); and market segments (e.g., purchasers of innovative sustainable construction solutions).

Since the WtA scale enables the examination of respondents’ subjective perceptions and self-reported attitudes, it could be integrated with supplementary methodologies to address methodological challenges associated with questionnaire surveys. For instance, it may be combined with data from multiple sources, objective behavioral indicators, validity measures, or statistical modeling techniques as appropriate.

Regarding relevant stakeholders and the professional context, the use of the WtA scale may facilitate a systematic assessment of professionals’ readiness to adopt self-sustaining interventions. This enables the identification of diverse attitudes and preferences within professional communities. For instance, this scale might serve as a probabilistic foundation for structured discussions within these communities (e.g., toolbox talks). It may also help identify individuals likely to support innovative solutions for broader dissemination of ideas, such as through door-to-door engagement. The scale underpins research enabling comparability of results and the systematic evaluation of sustainability-oriented competencies across diverse professional contexts. Additionally, it seems worthwhile to explore the broader applicability of the WtA scale across various categories of innovative and sustainable solutions. The findings of this study may have practical implications for decision-makers and regulators responsible for planning and approving urban infrastructure projects. The WtA scale developed in this research offers a straightforward tool for assessing public acceptance of sustainable engineering solutions, such as PePav, at the initial stages of project planning. This approach may enable early identification of potential social barriers to implementation. These barriers can then be addressed through targeted stakeholder communication or public participation processes. Consequently, the information derived from the scale could assist planners and policymakers in evaluating the social feasibility of introducing innovative stormwater management solutions.

The finding that both theoretical knowledge and practical experience in civil engineering may contribute to lower WtA in relation to PePav highlights the need to enhance awareness of PePav’s broader benefits among current and future professionals. Strengthening the integration of sustainable stormwater management concepts within university curricula and professional training for engineers may represent one potential strategy. Furthermore, as recent research indicates, greater emphasis should be placed on communicating the long-term environmental and social advantages of PePav, which may not be immediately apparent from construction or cost perspectives. Early involvement of experts in planning and implementation processes can help identify barriers and tailor communication strategies to address specific needs and challenges. Employing targeted communication strategies and participatory processes allows professionals to articulate concerns and contribute to solution optimization, thereby reducing resistance and enhancing acceptance of innovations.

Because the WtA scale has shown moderate but stronger prediction of intended behavior when cross-validated with constructs from TPB, future research could explore its role as a moderating variable within multivariate frameworks, including testing in connection with observed real-world behaviors. This approach could help clarify interactions among factors influencing the sustainable innovation implementation, while also offering tentative insights into its ecological validity and applicability in real-life contexts.

As mentioned in the Section 5, the cross-cultural generalizability of the WtA scale remains an open question. Future research should prioritize multi-region validation to assess the scale’s robustness across diverse international contexts. Furthermore, future investigations should explore how the measure performs across varying socio-economic strata and environmental conditions. Testing the scale in regions with different levels of “green” policy maturity or varying degrees of climate change vulnerability would provide deeper insights into the boundary conditions of the proposed measure.