Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

People’s attitudes towards the agrifood system influence the value of ecosystem services of mountain agroecosystems

  • Enrique Muñoz-Ulecia ,

    Contributed equally to this work with: Enrique Muñoz-Ulecia, Daniel Martín-Collado

    Roles Conceptualization, Formal analysis, Investigation, Visualization, Writing – original draft

    Affiliation Unidad de Producción y Sanidad Animal. Centro de Investigación y Tecnología Agroalimentaria de Aragón (CITA), Instituto Agroalimentario de Aragón (IA2) (CITA-Univ. de Zaragoza), Zaragoza, Spain

  • Alberto Bernués ,

    Roles Conceptualization, Methodology, Supervision, Writing – review & editing

    ‡ AB, DO, MR, MS and ES also contributed equally to this work.

    Affiliation Unidad de Producción y Sanidad Animal. Centro de Investigación y Tecnología Agroalimentaria de Aragón (CITA), Instituto Agroalimentario de Aragón (IA2) (CITA-Univ. de Zaragoza), Zaragoza, Spain

  • Daniel Ondé ,

    Roles Formal analysis, Methodology, Software, Writing – review & editing

    ‡ AB, DO, MR, MS and ES also contributed equally to this work.

    Affiliation Department of Psychobiology & Behavioral Sciences Methods, Complutense University of Madrid, Pozuelo de Alarcón, Madrid, Spain

  • Maurizio Ramanzin ,

    Roles Conceptualization, Methodology, Supervision, Writing – review & editing

    ‡ AB, DO, MR, MS and ES also contributed equally to this work.

    Affiliation Department of Agronomy, Food, Natural resources, Animals and Environment (DAFNAE), University of Padova, Legnaro, Padova, Italy

  • Mario Soliño ,

    Roles Formal analysis, Methodology, Software, Writing – review & editing

    ‡ AB, DO, MR, MS and ES also contributed equally to this work.

    Affiliations Institute of Marine Research, CSIC, Vigo, Spain, Complutense Institute for International Studies (ICEI), Finca Mas Ferré, Edif. A. Campus de Somosaguas, Madrid, Spain

  • Enrico Sturaro ,

    Roles Conceptualization, Methodology, Supervision, Writing – review & editing

    ‡ AB, DO, MR, MS and ES also contributed equally to this work.

    Affiliation Department of Agronomy, Food, Natural resources, Animals and Environment (DAFNAE), University of Padova, Legnaro, Padova, Italy

  • Daniel Martín-Collado

    Contributed equally to this work with: Enrique Muñoz-Ulecia, Daniel Martín-Collado

    Roles Conceptualization, Data curation, Methodology, Supervision, Writing – review & editing

    Affiliation Unidad de Producción y Sanidad Animal. Centro de Investigación y Tecnología Agroalimentaria de Aragón (CITA), Instituto Agroalimentario de Aragón (IA2) (CITA-Univ. de Zaragoza), Zaragoza, Spain


Studies covering the social valuation of ecosystem services (ES) are increasingly incorporating people’s attitudes, which allows social heterogeneity to be identified. This is especially relevant in mountain areas, where diverse complex interactions occur among the environment, the socioeconomic system, and a wide variety of farming practices. In this context, we aimed to: (i) identify the attitudinal dimensions that build people views about the agrifood system; and (ii) analyse how these attitudinal dimensions influence the value given to ES delivered by mountain agroecosystems of two European countries. We conducted a survey with a sample of 1008 individuals evenly distributed in the Italian Alps and Spanish Mediterranean mountain areas to collect information on people’s attitudes toward: (i) the economy and the environment; (ii) rural development and agricultural intensification; (iii) food quality, production, and consumption; and (iv) agricultural and environmental policies. The survey included a choice experiment to assess the value that individuals attach to the most relevant ES provided by mountain agroecosystems in these areas (i.e., landscape, biodiversity, quality local products, wildfires prevention and water quality). The results showed four common attitudinal dimensions, namely Economy over environment, Mass-Market distribution reliability, Agricultural productivism, and Environmentalism and rural lifestyle. These attitudinal dimensions resulted in six groups of respondents. Most groups positively valued an increase in the delivery of all the analysed ES, which suggests that agricultural policies which aim to promote ES are likely to receive social support in the study areas. However, the differing attitudinal dimensions underlying people’s preferences may result in disagreements about the steps to be taken to achieve the desired increase in ES delivery.

1. Introduction

People value things differently, with a very high diversity of individual perceptions of the utility provided by goods and services [1, 2]. This diversity is particularly high in how people perceive the utility derived from ecosystems services (ES) [3]. In this regard, considering people heterogeneity in ES values has been recommended to support inclusive decision making [3], which is frequently addressed by including explanatory variables, such as people’s attitudinal profiles and socio-demographic characteristics [46].

In the human-environment interactions field where ES valuation studies are found, research into the relationship between people’s attitudinal profiles and their valuation of ES is increasing. Researchers have focused mostly on studying the influence of attitudes towards the environment on people’s willingness to pay (WTP) or preferences for different ES provision levels [79]. These studies have generally found widespread environmental concern (in European Union countries and the United States), and a positive relation between pro-environmental attitudes and high ES values. With agroecosystems, where diverse interactions occur among the environment, socioeconomic systems and farming practices, people’s attitudes towards topics, such as lifestyle, agrifood chains, or what taxes are intended for, among others, may also influence their ES valuation [10, 11]. Therefore, exploring attitudes beyond the degree of environmental concern, including aspects like the agrifood system, the economy-environment relation, the quality and marketing of food products, maintaining the rural lifestyle or different agricultural practices (e.g., organic vs. conventional), may provide a more comprehensive view of society’s perception of ES provided by agroecosystems [12].

Mountain agroecosystems are widely recognised as being highly multifunctional, and supply a variety of cultural, regulating, supporting, and provisioning ES to local inhabitants, visitors, and society in general [12, 13]. However, these agroecosystems are immersed in a continuous process of farming abandonment and intensification in more favourable areas, which impacts ES delivery and raises concerns about their proper management over time [14, 15]. So, identifying social groups according to their attitudes towards the agrifood system and their valuation of the ES delivered by agroecosystems may shed light to design sustainable policies to enhance rural development and food security, while being aware of potential conflicts between different social groups and stakeholders [1618]. In this regard, [19] explored these aspects and found that people’s valuation of the ES provided by mountain agroecosystems varied between social groups with pro-environmental and agricultural productivism attitudes.

In this context, the purpose of this study was to analyse how people attitudes towards the agrifood system may be modulating their values of the ES provided by European mountain agroecosystems. Specifically, our first aim was to identify the attitudinal dimensions that build people’s views about: (i) the economy and the environment, (ii) rural development and agricultural intensification, (iii) food quality, production, and consumption, and (iv) agricultural and environmental policies. Our second aim was to analyse how these attitudinal dimensions influence the values attached to the ES provided by mountain agroecosystems in two European countries.] followed a standard methodological approach that consisted in including a selection of attitudinal statements in choice models that maximise heterogeneity [4, 19, 20]. However, this method shows marked differences in people’s attitudes due to the selection of those statements that receive the most extreme responses [21]. Moreover, the selection of opinion variables without considering their psychometric properties, might generate endogeneity problems in econometric models [10, 22]. In this study, we follow a more solid alternative to include people’s attitudes in choice experiments, which consists in considering the latent attitudinal dimensions that underlie and shape people’s responses [21, 23].

2. Materials and methods

2.1. Study areas

Our study builds on the research performed by previous studies [19, 24, 25]. The study areas are located in High Nature Value (HNV) mountain regions of Spain and Italy. In Spain, the ‘Sierra y Cañones de Guara’ Natural Park in the Spanish Pre-Pyrenees is a protected area of 807 km2 in northeast Spain. It is a Mediterranean mountainous area characterised by extensive, low-input low-output livestock farming systems that form a highly heterogeneous agricultural landscape [26]. These agroecosystems are recognised for their provision of numerous ES, such as biodiversity conservation, wildfires control, and the supply of quality products linked to the territory [25]. In Italy, the Autonomous Province of Trento in the Italian Alps covers an elevated mountainous area of 6200 km2 in northern Italy. This region works mainly in traditional dairy cattle agroecosystems which, similarly to the Spanish case study, provide diverse ES, such as a wide variety of local cheeses, highly valued landscapes and biodiversity conservation [2729]. However, the combined processes of agriculture intensification and abandonment that have taken place in recent decades in both areas have led to a disruption of the ES that these agroecosystems provide society with [30, 31].

2.2. Survey design

The Ethics Committee of the Agrifood Research and Technology Centre of Aragón, Spain, approved the research protocol and questionnaire content (no. CEISH_2021_3). Data anonymity was granted to the participants in the survey, who expressed their oral consent to provide the information contained in the questionnaire. Focus groups were performed in each study area to collect local inhabitants’ opinions about the relation between livestock systems and the environment to identify the ES provided by the livestock agroecosystems under study. Based on these focus groups, a questionnaire was designed and a pilot face-to-face interview was held with 70 respondents to check the questionnaire’s coherence. The final questionnaire included three sections: i) attitudinal statements, ii) ES valuation, and iii) respondents’ socio-economic data (e.g., age, gender, family size, level of education, level of income, relationship and involvement in farming and environmental associations). Interviews were held with 1008 people. In each study area, 102 adult residents in the mountain areas under study (i.e., local population) were interviewed face-to-face in all the rural areas, and 402 adults were interviewed by a professional online panel to represent the general population in the regions where the study areas are located (i.e., Aragon in Spain, Trento Province in Italy). Further details about the questionnaire design and sampling processes are found in [19, 24, 25].

2.2.1. Attitudinal statements.

We defined 20 Likert statements on aspects that might play a direct or indirect role for ES provision, which have been previously used by [19]. Statements were grouped into four main topics: (i) economy and environment, (ii) rural development and agricultural intensification, (iii) food quality, production, and consumption, and (iv) agricultural and environmental policies (Table 1). Statements were randomly (using the Microsoft Excel randomization tool) presented to respondents, who were asked to state their level of agreement with each statement on a five-point Likert scale ranging from 1—“Totally disagree” to 5—“Totally agree”, with an intermediate option for “Neutral”.

Table 1. Statements used in the questionnaire and descriptive statistics.

2.2.2. Ecosystem services valuation.

A discrete choice experiment was used to value the most relevant ES provided by the mountain agroecosystems under study, which were identified in the initial pilot interview. The use of choice experiments instead of other valuation methods is based on its higher robustness to analyse several attribute levels [32]. The experiment represented different agriculture policy scenarios, whose implementation would lead to distinct ES provision levels. Both attributes (cultural, supporting, regulating and provisioning ES, and cost) and levels (improvement, maintenance, and decrease) were defined to ensure that they were intuitive for the general public, scientifically accepted, and also appropriate for identifying changes in ES provision [25, 33, 34]. ES were site-specific, which allows people to value their local and known environment. The cultural, supporting and provisioning ES were the same in both areas, namely landscape, biodiversity, and quality local products, respectively. Regulating ES referred to forest wildfires prevention in Spain and water quality in Italy. It was explicitly clear that the cost of each alternative corresponded to the amount of money that each family member older than 18 would have to pay as annual tax. Further details about the choice experiment design are found in [19, 24, 25].

During the choice experiments, individuals were asked to choose their preferred alternatives from 30 choice sets divided into six blocks [5]. Each choice set consisted of three alternatives described by attributes (ES and costs of agri-environmental schemes), which took multiple levels according to the scenarios created following an efficient fractional factorial design and orthogonal and balanced combinations for the final design. Choices required the respondents making a trade-off between ES levels and tax costs and, thereby, stating their preferences for the ES levels (Fig 1) with the option to opt-out to avoid forced choices.

Fig 1. Example of the choice set presented to the respondents in the Spanish ‘Sierra y Cañones de Guara’ Natural Park.

From [25]. Policy A, B, and Current policy, refers to ES improvement, decrease, and maintenance, respectively.

2.3. Statistical analysis

We firstly performed an Exploratory Factor Analysis (EFA) to reveal the respondents’ attitudinal dimensions (i.e., factors) that underly their attitudes towards: the economy and the environment; rural development and agricultural intensification; food quality, production and consumption; and agricultural and environmental policy. Having determined and described the factors, we included them in a latent-class choice model to explore their influence on people’s ES valuation. Finally, we tested any differences between the latent classes (social groups hereafter) identified in the choice model regarding socioeconomic variables using the Kruskal-Wallis and Chi2 tests for numeric and categorical variables, respectively.

2.3.1. Attitudinal dimension analysis.

We used the polychoric correlation matrix as the input matrix, which is appropriate when analysing ordinal variables (i.e., attitudinal statements; see [35]). We employed the Psych package of R [36] and the estimator Unweighted Least Squares (ULS). The resulting factors identified groups of related scores to the attitudinal statements, which allowed us to look more closely at the respondents’ inner attitudinal structure by determining attitudinal dimensions. The Parallel Analysis (PA) was followed to identify the optimum latent structure of data and, thus, the adequate number of factors. To test the model’s fit, we checked different indicators (e.g., Root Mean Square Error of Approximation (RMSEA), Non-Normed Fit Index (NNFI), Root Mean Square Residuals (RMSR), and residuals).

2.3.2. Choice modelling.

We used a latent-class choice model [37] with random parameters to estimate each individual’s utility that derived from a particular ecosystem service with the Latent GOLD® software (v. 5.1). This model type allows different social groups to be isolated using the attitudinal dimensions identified in the EFA, which enables the unobserved preference heterogeneity among the respondents from each social group to be considered. For each social group, we examined the relation between individuals’ preferred policy choices (dependent variable) and the levels of the attributes in the alternatives they chose, i.e., landscape maintenance, biodiversity preservation, maintaining water quality/preserving forest fires, provision of high-quality local food products and annual cost (independent variables). The effect of the attributes on choice probability was evidenced by parameter estimates. The sign of a parameter value showed the extent to which the presence of an attribute in each scenario influenced the probability of choosing it. A further description of the latent-class choice model specifications is found in the S1 Appendix.

3. Results

3.1. Psychographic analysis—Respondents’ attitudinal dimensions

The EFA gave similar results in both study areas, with the best solution suggested by the PA. It consisted in four factors with adequate fit indices values (Spain: RMSEA = 0.043, NNFI = 0.903, RMSR = 0.031 (< 0.44 Kelley criterion), 4.6% of the residuals > 0.05; Italy: RMSEA = 0.053, NNFI = 0.899, RMSR = 0.032 (< 0.44 Kelley criterion), 5.1% of the residuals > 0.05).

An initial analysis, which considered the pooled sample indicated that statement AP17 (Table 1) was problematic because it showed salient factor loadings on more than one factor. This was why, we removed statement AP17 and re-analysed the dataset. The best solution of the EFA, suggested by the PA, was four factors. The fit indices values were RMSEA = 0.049, NNFI = 0.900, and RMSR = 0.029 (lower value than Kelley criterion 0.032). Only 4.8% of the residuals values were > 0.05. Table 2 shows the configuration matrix of the factor loading on the 19 statements final scale.

Table 2. Configuration matrix of the 19 statements factor loadings for attitudinal factors (oblique rotation).

These four factors represented the attitudinal dimensions that we hypothesised would influence or help to explain the respondents’ ES valuation, and can be described as follows:

  • Factor 1 –Economy over environment: it represented attitudes towards the generation of economic wealth over nature, acceptance of the livestock impact on the environment, and lack of interest of environmental policies.
  • Factor 2 –Mass-Market distribution reliability: it represented attitudes towards new technologies for product transformation and large supermarkets as better guarantors of product quality and safety than traditional stores.
  • Factor 3 –Agricultural productivism: it grouped attitudes towards maximising economic benefits from natural resources, intensive agriculture advantages and productivity-oriented premiums.
  • Factor 4 –Environmentalism and rural lifestyle: it was formed by half the presented statements, which gathered all the attitudes related to a socio-environmentally concerned way of life and towards rural life.

Some factors were correlated to a limited extent: Economy over environment correlated positively with Mass-Market distribution reliability (0.37) and negatively with Environmentalism and rural lifestyle (-0.32). Mass-Market distribution reliability correlated positively with Agriculture productivism (0.25) and negatively with Environmentalism and rural lifestyle (-0.16). All the other correlations between factors came close to zero.

3.2. Choice experiment—Respondents’ ecosystem services valuation

The model that showed the best fit while allowing a comprehensive explanation resulted in six social groups and had a McFadden pseudo R2 of 0.57. Figs 2 and 3 summarise the latent-class model results. Fig 2 describes the latent groups according to the attitudinal dimensions, while Fig 3 shows the value given to ES provided by HNV agroecosystems across social groups. Full model outputs are provided in the S1 Appendix (A1 and A2 Tables in S1 Appendix).

Fig 2. Different social groups formed in the latent-class model from the attitudinal dimensions.

Scores indicate the contribution of each attitudinal factor to the different latent social groups. Black stars refer to trends (90%, 1 star) and significant attitudes defining social groups (95% and 99% for 2 and 3 stars, respectively). For more detailed statistical results, see A1 Table in S1 Appendix.

Fig 3. Respondents’ ES valuation for the different latent social groups in the different scenarios.

Fig 3a–3c, represent how different social groups valued ES attributes in various scenarios. A more detailed description of scenarios is provided in [24, 25]. Black stars refer to trends (90%, 1 star) and significant scores (95% and 99% for 2 and 3 stars, respectively). Deviation bars denote standard error. For more detailed statistical results, see A2 Table in S1 Appendix.

  • Group 1 –Anti-economic prioritisation (14.6% of the responses) grouped attitudes strongly opposed Economy over environment and opposed to Mass-Market distribution reliability. This social group showed support to the current level and were against the deterioration of supporting and regulating ES, while also positively valuing an improvement in provisioning ES. They did not show a clear position about cultural ES.
  • Group 2 –Non-Mass-market environmentalism (21.9%) grouped attitudes in favour of Environmentalism and rural lifestyle and opposed to Mass-Market distribution reliability. They were dissatisfied with the current level of supporting ES and showed support for improving cultural and supporting ES. They did not show a clear position about regulating and provisioning ES.
  • Group 3 –Mass-market environmentalism (7.0%) grouped attitudes strongly in favour of Mass-Market distribution reliability and slightly in favour of Environmentalism and rural lifestyle. This social group was the only one to give positive values at the same time to the deterioration and improvement of one ES (namely, cultural ES), while negatively valuing status quo. They strongly supported improvement in supporting and regulating ES, but were against their deterioration, while positively valuing the current provisioning ES level.
  • Group 4 –Miscellaneous (16.7%) presented no dominant attitudinal dimension. This social group was strongly against the deterioration of cultural and regulating ES, and against the deterioration of provisioning ES. They tended to support improvement in cultural, supporting and regulating ES.
  • Group 5 –Non-Mass-market productivism (25.9%) grouped responses in favour of Agricultural productivism and opposed to Mass-Market distribution reliability. This social group supported improvement in, but against, the deterioration of supporting, regulating, and provisioning ES. For cultural ES, the current level was the most valued.
  • Group 6 –Anti-environmentalism (13.9%) grouped attitudes strongly in favour of Economic over environment and Mass-Market distribution reliability, but strongly opposed to Environmentalism and rural lifestyle. These respondents showed support for improvement in, and were against, the deterioration of regulating and provisioning ES. They were dissatisfied with the current level of cultural ES and favoured their improvement. They negatively valued deterioration of supporting ES in favour of the current level.

Note that Non-Mass-market environmentalism and Mass-market environmentalism (covering one third of the sample) assigned significant positive values to cost estimates (p<0.05 and p<0.01, respectively). This means that these participants did not make the expected trade-off between ES attributes and their associated costs (A2 Table in S1 Appendix). Non-Mass-market productivism and Anti-environmentalism (around 40% of the sample) attempted to maximise costs (p<0.01 and p<0.05, respectively), while the other social groups (Anti-economic prioritisation and Miscellaneous) did not offer any significant outputs for this issue.

The analysis of the socio-economic data collected in the third part of the questionnaire (A3 Table in S1 Appendix) showed that Non-Mass-market productivism and Anti-environmentalism had the lowest income of all the social groups. Significant differences were also found in the respondents’ relationship with the farming sector, being Environmentalism, both Non-Mass-market and Mass-market, who had the lowest proportion. However, no differences were found for the respondents’ age, gender, level of education, relationship with the environment, farming activity and involvement in environmental associations.

4. Discussion

The consideration of people’s attitudes in sociocultural valuation studies strongly influence on ES perception [3]. Most studies have focused on environmental attitudes [e.g., 38, 39], leaving attitudes towards other issues largely unexplored. This research gap is particularly for agroecosystems where attitudes towards the agrifood system and the socioeconomic system at large may have great influence on people’s ES valuation [19]. Our study contributes to bridge this gap by exploring a set of attitudes that goes beyond the human-environment relationship using sound psychometric and choice modelling statistical tools. We specifically explored attitudes towards the economy-environment relation, agriculture, food production and consumption, and rural development.

Our results showed that, despite human-environment attitudes and the valuation of ES delivered by agroecosystems usually being context-dependent [4042], some common attitudinal dimensions shape people’s ES valuation in the Italian Alps and the Spanish Mediterranean mountains. We identified four attitudinal dimensions (namely Economy over environment, Mass-Market distribution reliability, Agricultural productivism, and Environmentalism and rural lifestyle) that strongly influenced the respondents’ valuation of the ES provided by mountain agroecosystems.

4.1. Attitudinal dimensions

The four attitudinal dimensions revealed different perceptions of the agrifood system, with interrelated aspects related associated with the balance among economy and environment, food production and consumption, agriculture, and rural development.

4.1.1. Economy over environment.

This attitudinal dimension shares an epistemological basis with the prevailing socioeconomic paradigm [43], characterised by considering economic growth to be a feasible and advisable social objective [44]. This linkage is sustained by the belief in continuous economic growth as the primary goal, which overrides environment preservation [45] and sustainability [46, 47].

4.1.2. Mass-market distribution reliability.

Food processing, quality standards and distribution channels have largely changed in the last few decades [48, 49] by increasing market concentration, with a significant increase in supermarkets at the expense of traditional shops [50, 51]. This attitudinal dimension refers to people’s perception of reliability on mass-market distribution channels as a guarantee of food product quality and safety. Our results showed that this attitudinal dimension has not always confronted pro-environmental positions, as pointed out in other studies [52].

4.1.3. Agricultural productivism.

This attitudinal dimension is directly related to the well-known paradigm of productivism, which can be associated with the “Green Revolution”. Agricultural productivism has been considered the paradigmatic strategy to feed the world for a long time [53]. Nowadays, most international agriculture institutions’ strategies recognise agriculture as a multifunctional activity [54, 55]. Our results showed that the productivism paradigm is still established on a large share of the studied population, despite the volume of food production in European countries amply meeting requirements, and problems associated with mountain areas’ agriculture being widely recognised as others (abandonment, landscape degradation, etc.) than maximising production [40].

4.1.4. Environmentalism and rural lifestyle.

This attitudinal dimension places attitudes related to both environmental concerns and support of rural lifestyle together. To some extent, it relates the idea of returning to the countryside being the solution to the environmental problems that derive from the current urban lifestyle, which has been profoundly studied by Dunlap and colleagues [46, 56].

4.2. Social groups and ecosystem services values

The four identified attitudinal dimensions generated six social groups with a diversity of perspectives towards agriculture, rural lifestyle, product quality and safety, and economy-environment interaction, which resulted in differing ES valuations. However, we observed that, save a few exceptions, all the social groups supported greater ES delivery. This result is consistent with several studies that have demonstrated a general societal willingness to improve ES delivery (even at the expense of paying more taxes) [24, 25, 57] and widespread serious environmental concerns [5860]. We also found that the respondents’ socioeconomic characteristics (e.g., age, gender and level of education) were not significantly different in the identified social groups, which other studies have also encountered [6]. Below we discuss the different ES valuation across the identified social groups.

4.2.1. Cultural ecosystem services.

These non-marketable and non-extractive public services have been increasingly studied in mountain areas in recent years [61, 62]. As cultural ES valuations are inherently subjective and individual-dependent [63], their appropriate management is a source of conflict between different stakeholders and social groups [6466]. Indeed we found that people from contrasting social groups (Non-Mass-Market Environmentalism, Mass-Market Environmentalism and Anti-Environmentalism) positively valued an increase in cultural ES delivery but would likely support alternative ways of doing so. For example, we observed that, although the people in Mass-Market environmentalism social group did not appreciate the current landscape, they supported both a richer mosaic resulting from more agriculture and a re-naturalised landscape resulting from further agriculture abandonment.

4.2.2. Supporting ecosystem services.

Supporting ES are often identified with biodiversity preservation, which is considered a key driver for the delivery of many other ES [67]. Our results showed that around 70% of the responses supported increased biodiversity, and the Mass-Market environmentalism was the social group with the highest scores. Interestingly, the people in the Non-Mass-market productivism social group supported increased biodiversity despite agriculture intensification and industrialisation having been identified as drivers of ES disruption and biodiversity loss [68, 69].

4.2.3. Regulating ecosystem services.

These ES are related to ecosystem processes and functions, whose valuation is difficult because the benefits they provide society with are not direct and are, thus, difficult to identify [70]. However, regulating ES are increasingly recognised by society because their role is essential for preventing hazards and maintaining ecosystems’ health [71]. This may explain the general positive attitude towards improving regulating ES in our study. Indeed, most social groups gave high scores against decreasing the regulation of ES delivery.

4.2.4. Provisioning ecosystem services.

These marketable and tangible services are private goods that are well recognised by society for being the basis of market trade [72]. Their maximisation usually requires a trade-off with the provision of other services because farmers have to decide between agricultural productivity and pro-environmental practices [73, 74]. We observed that only those groups with a positive or undefined attitude towards Environmentalism and rural lifestyle supported an increase in the provision of local quality products.

4.3. Considerations for stated preferences studies

Notwithstanding choice experiments are commonly used for monetary ES valuations [75], we identified that one-third of the respondents, specifically those with pro-Environment and rural lifestyle (i.e., Non-Mass-market environmentalism and Mass-market environmentalism attitudes), did not show the expected trade-off between levels of attributes and their associated costs. The inclusion of people’s attitudes in the choice model revealed (an otherwise hidden fact) that a significant part of the participants did not react to the costs attribute of the selected alternative, which does not allow the WTP to be calculated for these groups of respondents. For environmentally concerned people, the prices presented in the experiment were not as important as other characteristics when determining the utility of ES. This finding suggests that the WTP calculation might be over-estimated by strong pro-environmental attitudes [20]. However, we should acknowledge that the cost associated with different choice alternatives could be too low for some respondents despite the pilot survey experience. This involves a handicap to implement or select choice alternatives that mean the same to everyone. These results confirm the existing literature, and highlight that methods very much shape valuation outcomes [76]. Previous studies that have applied a mixed logit modelling approach have pointed out the presence of high heterogeneity on several dimensions [24, 77]. Our result encourages the inclusion of people’s attitudes in choice experiments to account for such heterogeneity, and the use of complementary non-monetary approaches to assess the value of ES [75, 78].

4.4. Limitations

We should highlight some limitations in our study that join to those previously acknowledged in [38, 39]. We focused on the most relevant ES identified during participatory workshops held with local stakeholders. Consequently, our research did not cover all the ES provided by the studied agroecosystems. We should note that wildfire prevention and improvement of water quality have been highly recognised and valued by society [24, 77]. This raises questions as to whether less socially recognised ES would lead to different results.

Regarding the link between attitudes and behaviour, despite it is commonly assumed that people behave according to their values, attitudes have been found to have a varying impact on real behaviour, which results from the complex interaction between different external (e.g., economic and cultural) and internal factors (e.g., motivation, awareness, values, locus of control, priorities) [7981]. Moreover, some degree of hypothetical bias underlying choice design and attribute non-attendance (as recognised above for the cost attribute) might also be present in the study [82, 83], so future research should look closely at these relevant issues for economic valuation applications.

4.5. Insights for policy planning

Social and scientific claims for designing transformative policies to face the sustainability challenge are increasing [84, 85]. However, large-scale transitions require coordinated plans and society’s commitment at different levels and scales [8688]. As society integrates heterogeneous social groups with contrasting values and goals, the identification of social heterogeneity can help to design strategies that modify those values and engage hesitant societal sectors in more sustainable behaviours [89, 90].

Our study shows some common attitudinal dimensions related to the agrifood system across study regions that, if confirmed in other European mountain areas, may help to understand the social acceptance of alternatives in agricultural and nature conservation policies. This study also reveals that society might be highly segmented in different social groups regarding these attitudinal dimensions, which can lead to a double reading: on the one hand, despite this segmentation, we found widespread support to improve the delivery of some ES across all social groups (e.g., the general preference for higher biodiversity levels across social groups may facilitate the implementation and success of biodiversity policies, such as the European Biodiversity Strategy 2030) [91]; on the other hand, since a generalised consensus on the desired goal (improve ES delivery) does not always imply agreements about the ways and means to achieve it, conflicts are likely to emerge. In the end, these conflicts may determine the success or failure of policies, as highlighted by other authors [89, 92].

5. Conclusions

This study contributes to understand people’s attitudes towards central topics in the agrifood systems-environment debate, including the livestock environmental impact, the quality and marketing of food products, and rural development. This research work establishes four attitudinal dimensions: Economy over environment, Mass-Market distribution reliability, Agricultural productivism, and Environmentalism and rural lifestyle. These attitudinal dimensions influence the values that people attach to the key ES provided by mountain agroecosystems, which allowed us to identify diverse social groups. The preference for increasing ES delivery in most groups highlights the social demand for policies that aim to increase the delivery of ES in mountains. However, the differing attitudinal dimensions that underly people’s preferences may result in disagreement and conflict about the specific policy measures to be implemented.


  1. 1. Georgescu-Roegen N. Choice, Expectations and Measurability. Q J Econ 1954;68:503–34.
  2. 2. McFadden D. Conditional logit analysis of qualitative choice behavior. In: Zarembka P, editor. Front. Econom., New York: Academic Press; 1974, p. 105–42.
  3. 3. Tauro A, Gómez-Baggethun E, García-Frapolli E, Chavero EL, Balvanera P. Unraveling heterogeneity in the importance of ecosystem services: Individual views of smallholders. Ecol Soc 2018;23.
  4. 4. Choi AS, Fielding KS. Environmental attitudes as WTP predictors: A case study involving endangered species. Ecol Econ 2013;89:24–32.
  5. 5. Hensher DA, Rose JM, Greene WH. Applied Choice Analysis. Second. Cambridge: Cambridge University Press; 2015.
  6. 6. Lau JD, Hicks CC, Gurney GG, Cinner JE. Disaggregating ecosystem service values and priorities by wealth, age, and education. Ecosyst Serv 2018;29:91–8.
  7. 7. Eusse-villa LF, Franceschinis C, Thiene M, Meyerhoff J, Mcbratney A, Field D. Attitudes and Preferences towards Soil-Based Ecosystem Services: How Do They Vary across Space? Sustainability 2021;13:19.
  8. 8. Racevskis L, Lupi F. Valuing Forest Ecosystem Services: A Comparison of the Effects of Attitudes and Demographic Characteristics on Willingness to Pay. Am. Agric. Econ. Assoc. Annu. Meet., 2006, p. 1–9.
  9. 9. Wardropper CB, Mase AS, Qiu J, Kohl P, Booth EG, Rissman AR. Ecological worldview, agricultural or natural resource-based activities, and geography affect perceived importance of ecosystem services. Landsc Urban Plan 2020;197:103768.
  10. 10. Hess S, Beharry-Borg N. Accounting for Latent Attitudes in Willingness-to-Pay Studies: The Case of Coastal Water Quality Improvements in Tobago. Environ Resour Econ 2012;52:109–31.
  11. 11. Vredin Johansson M, Heldt T, Johansson P. The effects of attitudes and personality traits on mode choice. Transp Res Part A Policy Pract 2006;40:507–25.
  12. 12. Martín-López B, Leister I, Cruz PL, Palomo I, Gr A, Harrison PA, et al. Nature’ s contributions to people in mountains: A review. PLoS One 2019;14:1–24. Funding.
  13. 13. European Environment Agency. Europe’s ecological backbone: recognising the true value of our mountains. 2010.
  14. 14. Boix-Fayos C, Boerboom LGJ, Janssen R, Martínez-Mena M, Almagro M, Pérez-Cutillas P, et al. Mountain ecosystem services affected by land use changes and hydrological control works in Mediterranean catchments. Ecosyst Serv 2020;44:101136.
  15. 15. Durán M, Canals RM, Sáez JL, Ferrer V, Lera-López F. Disruption of traditional land use regimes causes an economic loss of provisioning services in high-mountain grasslands. Ecosyst Serv 2020;46.
  16. 16. Castro AJ, Martín-López B, García-LLorente M, Aguilera PA, López E, Cabello J. Social preferences regarding the delivery of ecosystem services in a semiarid Mediterranean region. J Arid Environ 2011;75:1201–8.
  17. 17. Knickel K, Redman M, Darnhofer I, Ashkenazy A, Calvão Chebach T, Šūmane S, et al. Between aspirations and reality: Making farming, food systems and rural areas more resilient, sustainable and equitable. J Rural Stud 2017;59:197–210.
  18. 18. Hoffmann I, From T, Boerma D. Ecosystem services provided by livestock species and breeds, with special consideration to the contributions of small-scale livestock keepers and pastoralists. Rome: 2014.
  19. 19. Rodríguez-Ortega T, Bernués A, Alfnes F. Psychographic profile affects willingness to pay for ecosystem services provided by Mediterranean high nature value farmland. Ecol Econ 2016;128:232–45.
  20. 20. Aldrich GA, Grimsrud KM, Thacher JA, Kotchen MJ. Relating environmental attitudes and contingent values: How robust are methods for identifying preference heterogeneity? Environ Resour Econ 2007;37:757–75.
  21. 21. Taye FA, Vedel SE, Jacobsen JB. Accounting for environmental attitude to explain variations in willingness to pay for forest ecosystem services using the new environmental paradigm. J Environ Econ Policy 2018;7:420–40.
  22. 22. Hoyos D, Mariel P, Hess S. Incorporating environmental attitudes in discrete choice models: An exploration of the utility of the awareness of consequences scale. Sci Total Environ 2015;505:1100–11. pmid:25461111
  23. 23. Johnston RJ, Boyle KJ, Vic Adamowicz W, Bennett J, Brouwer R, Ann Cameron T, et al. Contemporary guidance for stated preference studies. J Assoc Environ Resour Econ 2017;4:319–405.
  24. 24. Faccioni G, Sturaro E, Ramanzin M, Bernués A. Socio-economic valuation of abandonment and intensification of Alpine agroecosystems and associated ecosystem services. Land Use Policy 2019;81:453–62.
  25. 25. Bernués A, Rodríguez-Ortega T, Ripoll-Bosch R, Alfnes F. Socio-cultural and economic valuation of ecosystem services provided by Mediterranean mountain agroecosystems. PLoS One 2014;9. pmid:25036276
  26. 26. Riedel JL. Bases para la gestión sostenible del Parque de la Sierra y Cañones de Guara: interacciones entre la ganadería y la dinámica de la vegetación. University of Zaragoza, 2007.
  27. 27. Bittante G, Cecchinato A, Cologna N, Penasa M, Tiezzi F, De Marchi M. Factors affecting the incidence of first-quality wheels of Trentingrana cheese. J Dairy Sci 2011;94:3700–7. pmid:21700060
  28. 28. Zendri F, Sturaro E, Ramanzin M. Highland summer pastures play a fundamental role for dairy systems in an Italian alpine region. Agric Conspec Sci 2013;78:295–9.
  29. 29. Cocca G, Sturaro E, Gallo L, Ramanzin M. Is the abandonment of traditional livestock farming systems the main driver of mountain landscape change in Alpine areas? Land Use Policy 2012;29:878–86.
  30. 30. Bernués A, Riedel JL, Asensio MA, Blanco M, Sanz A, Revilla R, et al. An integrated approach to studying the role of grazing livestock systems in the conservation of rangelands in a protected natural park (Sierra de Guara, Spain). Livest Prod Sci 2005;96:75–85.
  31. 31. Sturaro E, Marchiori E, Cocca G, Penasa M, Ramanzin M, Bittante G. Dairy systems in mountainous areas: Farm animal biodiversity, milk production and destination, and land use. Livest Sci 2013;158:157–68.
  32. 32. Oviedo JL, Caparrós A. Comparing Contingent valuation and ChoiCe modeling using field and eye—traCking lab data. 2014.
  33. 33. Layke C. A new eruption attachment for impacted teeth. Washington DC: 2009.
  34. 34. Faccioni G. Ecosystem Servies and Sustainability Evaluation of Alpine Dairy Cattle Systems. Università degli Studi di Padova, 2019.
  35. 35. Forero CG, Maydeu-Olivares A, Gallardo-Pujol D. Factor Analysis with Ordinal Indicators: A Monte Carlo Study Comparing DWLS and ULS Estimation. Struct Equ Model A Multidiscip J 2009;16:625–41.
  36. 36. Revelle W. Procedures for Psychological, Psychometric, and Personality Research 2021.
  37. 37. Magidson J, Vermunt JK. Latent class models. Thousands Oakes; 2004.
  38. 38. Spash CL, Urama K, Burton R, Kenyon W, Shannon P, Hill G. Motives behind willingness to pay for improving biodiversity in a water ecosystem: Economics, ethics and social psychology. Ecol Econ 2009;68:955–64.
  39. 39. Bartczak A. The role of social and environmental attitudes in non-market valuation. An application to the Białowiez˙a Forest. For Policy Econ 2015;50:357–65.
  40. 40. Bernués A. Animals on the land: Ecosystem services and disservices of grazing livestock systems. Meat Cris Dev More Sustain Ethical Prod Consum Second Ed 2017:67–92.
  41. 41. Costanza R, de Groot R, Braat L, Kubiszewski I, Fioramonti L, Sutton P, et al. Twenty years of ecosystem services: How far have we come and how far do we still need to go? Ecosyst Serv 2017;28:1–16.
  42. 42. Bernués A, Tello-García E, Rodríguez-Ortega T, Ripoll-Bosch R, Casasús I. Agricultural practices, ecosystem services and sustainability in High Nature Value farmland: Unraveling the perceptions of farmers and nonfarmers. Land Use Policy 2016;59:130–42.
  43. 43. Pirages DC, Ehrlich PR. Ark II: Social responses to environmental imperatives. San Francisco: W.H. Freeman; 1974.
  44. 44. Ward JD, Sutton PC, Werner AD, Costanza R, Mohr SH, Simmons CT. Is decoupling GDP growth from environmental impact possible? PLoS One 2016;11:1–14. pmid:27741300
  45. 45. Spash CL. Routledge Handbook of Ecological Economics: Nature and Society. Routledge Int Handbooks 2017:445.
  46. 46. Dunlap R. The new environmental paradigm scale: From marginality to worldwide use. J Environ Educ 2008;40:3–18.
  47. 47. Dunlap RE, Van Liere KD. The “new environmental paradigm.” J Environ Educ 1978;9:10–9.
  48. 48. Samuel MV, Shah M, Sahay BS. An insight into agri-food supply chains: A review. Int J Value Chain Manag 2012;6:115–43.
  49. 49. Saitone TL, Sexton RJ. Agri-food supply chain: evolution and performance with conflicting consumer and societal demands. Eur Rev Agric Econ 2017;44:634–57.
  50. 50. Sexton RJ, Xia T. Increasing Concentration in the Agricultural Supply Chain: Implications for Market Power and Sector Performance. Annu Rev Resour Econ 2018;10:229–51.
  51. 51. Lee J, Gereffi G, Beauvais J. Global value chains and agrifood standards: Challenges and possibilities for smallholders in developing countries. Proc Natl Acad Sci U S A 2012;109:12326–31. pmid:21149723
  52. 52. Codron JM, Siriex L, Reardon T. Social and environmental attributes of food products in an emerging mass market: Challenges of signaling and consumer perception, with European illustrations. Agric Human Values 2006;23:283–97.
  53. 53. Horlings LG, Marsden TK. Towards the real green revolution? Exploring the conceptual dimensions of a new ecological modernisation of agriculture that could “feed the world.” Glob Environ Chang 2011;21:441–52.
  54. 54. FAO. The State of Food and Agriculture 2021. Making agrifood systems more resilient to shocks and stresses. Rome: 2021.
  55. 55. UN. United Nations Food Systems Summit 2021. Science and Innovations for Food Systems Transformation and Summit Actions. 2021.
  56. 56. Dunlap RE, Van Liere KD, Mertig AG, Jones RE. Measuring endorsement of the new ecological paradigm: A revised NEP scale. J Soc Issues 2000;56:425–42.
  57. 57. Huber R, Finger R. A Meta-analysis of the Willingness to Pay for Cultural Services from Grasslands in Europe. J Agric Econ 2019:1–27. pmid:32406423
  58. 58. de Groot M, Drenthen M, de Groot WT. Public visions of the human/nature relationship and their implications for environmental ethics. Environ Ethics 2011;33:25–44.
  59. 59. Grendstad G, Wollebaek D. Greener Still? An Empirical Examination of Eckersley’s Ecocentric Approach. Environ Values 1998;30:653–75.
  60. 60. Thomson SCG, Barton MA. Ecocentric and Anthropocentric attitudes toward the environment. J Environ Psychol 1994;14:149–57.
  61. 61. Schirpke U, Scolozzi R, Dean G, Haller A, Jäger H, Kister J, et al. Cultural ecosystem services in mountain regions: Conceptualising conflicts among users and limitations of use. Ecosyst Serv 2020;46.
  62. 62. Kosanic A, Petzold J. A systematic review of cultural ecosystem services and human wellbeing. Ecosyst Serv 2020;45:101168.
  63. 63. Bryce R, Irvine KN, Church A, Fish R, Ranger S, Kenter JO. Subjective well-being indicators for large-scale assessment of cultural ecosystem services. Ecosyst Serv 2016;21:258–69.
  64. 64. Guo Z, Zhang L, Li Y. Increased dependence of humans on ecosystem services and biodiversity. PLoS One 2010;5. pmid:20957042
  65. 65. Wilkes-Allemann J, Pütz M, Hirschi C, Fischer C. Conflict situations and response strategies in urban forests in Switzerland. Scand J For Res 2015;30:204–16.
  66. 66. Smith HF, Sullivan CA. Ecosystem services within agricultural landscapes-Farmers’ perceptions. Ecol Econ 2014;98:72–80.
  67. 67. Balvanera P, Quijas S, Martín-López B, Barrios E, Dee L, Isbell F, et al. The links between biodiversity and ecosystem services. In: Potschin M, Haines-Young R, Fish R, Turner RK, editors. Routledge Handb. Ecosyst. Serv., Routledge; 2016, p. 45–61.
  68. 68. Bowler DE, Heldbjerg H, Fox AD, de Jong M, Böhning-Gaese K. Long-term declines of European insectivorous bird populations and potential causes. Conserv Biol 2019;33:1120–30. pmid:30912605
  69. 69. IPBES. Global assessment report on biodiversity and ecosystem services of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services. Bonn: 2019.
  70. 70. Kadykalo AN, Kelly LA, Berberi A, Reid JL, Findlay CS. Research effort devoted to regulating and supporting ecosystem services by environmental scientists and economists. PLoS One 2021;16:e0252463. pmid:34048482
  71. 71. Fürst C, Frank S, Inkoom JN. Managing Regulating Services for Sustainability. In: Potschin M, Haines-Young R, Fish R, Turner RK, editors. Routledge Handb. Ecosyst. Serv., Routledge; 2016, p. 630.
  72. 72. Galford G, Ricketts T. Ecosystem Services: Provisioning. In: Potschin M, Haines-Young R, Fish R, Turner RK, editors. Routledge Handb. Ecosyst. Serv., Routledge; 2016, p. 630. pmid:27028754
  73. 73. Banerjee S, Cason TN, de Vries FP, Hanley N. Transaction costs, communication and spatial coordination in Payment for Ecosystem Services Schemes. J Environ Econ Manage 2017;83:68–89.
  74. 74. Midler E, Pascual U, Drucker AG, Narloch U, Soto JL. Unraveling the effects of payments for ecosystem services on motivations for collective action. Ecol Econ 2015;120:394–405.
  75. 75. Gómez-Baggethun E, Barton DN, Berry P, Dunford R, Harrison PA. Concepts and Methods in Ecosystem Services Valuation. In: Potschin M, Haines-Young R, Fish R, Turner RK, editors. Routledge Handb. Ecosyst. Serv., Routledge; 2016, p. 630.
  76. 76. Hansjürgens B, Schröter-Schlaack C, Berghöfer A, Lienhoop N. Justifying social values of nature: Economic reasoning beyond self-interested preferences. Ecosyst Serv 2017;23:9–17.
  77. 77. Bernués A, Alfnes F, Clemetsen M, Eik LO, Faccioni G, Ramanzin M, et al. Exploring social preferences for ecosystem services of multifunctional agriculture across policy scenarios. Ecosyst Serv 2019;39:101002.
  78. 78. Potschin M, Haines-Young R. Defining and Measuring Ecosystem Services. In: Potschin M, Haines-Young R, Fish R, Turner RK, editors. Routledge Handb. Ecosyst. Serv., Routledge; 2016, p. 630. pmid:27045459
  79. 79. Schultz PW. Conservation Means Behavior. Conserv Biol 2011;25:1080–3. pmid:22070255
  80. 80. Kollmuss A, Agyeman J. Mind the Gap: Why do people act environmentally and what are the barriers to pro-environmental behavior? Environ Educ Res 2002;8:239–60.
  81. 81. Gifford R, Nilsson A. Personal and social factors that influence pro-environmental concern and behaviour: A review. Int J Psychol 2014;49:141–57. pmid:24821503
  82. 82. Schmidt J, Bijmolt THA. Accurately measuring willingness to pay for consumer goods: a meta-analysis of the hypothetical bias. J Acad Mark Sci 2020;48:499–518.
  83. 83. Lew DK, Whitehead JC. Attribute non-attendance as an information processing strategy in stated preference choice experiments: Origins, current practices, and future directions. Mar Resour Econ 2020;35:285–317.
  84. 84. Díaz S, Settele J, Brondízio ES, Ngo HT, Agard J, Arneth A, et al. Pervasive human-driven decline of life on Earth points to the need for transformative change. Science (80-) 2019;366. pmid:31831642
  85. 85. IPCC. Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge: 2021.
  86. 86. Folke C, Österblom H, Jouffray JB, Lambin EF, Adger WN, Scheffer M, et al. Transnational corporations and the challenge of biosphere stewardship. Nat Ecol Evol 2019;3:1396–403. pmid:31527729
  87. 87. Brondizio ES, Ostrom E, Young OR. Connectivity and the governance of multilevel social-ecological systems: The role of social capital. Annu Rev Environ Resour 2009;34:253–78.
  88. 88. Ostrom E. Governing the Commons. Gov Commons 1990.
  89. 89. Iglesias A, Garrote L, Bardají I, Santillán D, Esteve P. Looking into individual choices and local realities to define adaptation options to drought and climate change. J Environ Manage 2021;293:112861. pmid:34087643
  90. 90. Dessart FJ, Barreiro-Hurlé J, Van Bavel R. Behavioural factors affecting the adoption of sustainable farming practices: A policy-oriented review. Eur Rev Agric Econ 2019;46:417–71.
  91. 91. European Commission. EU Biodiversity Strategy for 2030. Bringing nature back into our lives. 2021. pmid:34013622
  92. 92. Wojtynia N, Van Dijk J, Derks M, Groot PWG, Marko K. A new green revolution or agribusiness as usual? Uncovering alignment issues and potential transition complications in agri-food system transitions. Agron Sustain Dev 2021;41:20.