Introduction

Member states in the United Nations approved the Sustainable Development Goals (SDGs) at the historic 2015 Sustainable Development Summit. The SDGs aim to create a global transition toward sustainability. The purpose of SDG 12 is to promote the global adoption of sustainable consumption and production patterns. Current global consumption and production patterns rely heavily on natural resource consumption, which has deleterious effects on the planet (Johnstone and Lindh 2022). During the 20th century, wars and rapid postwar economic growth caused environmental degradation, jeopardizing the ecological systems on which humans rely. Since 2019, the global spread of the coronavirus disease 2019 (COVID-19) has gradually increased public awareness on the importance of environmental protection issues (Severo et al. 2021). Traditional farming methods are highly dependent on the use of pesticides and other chemicals, resulting in soil erosion and degradation, biodiversity loss, agriculture-related environmental degradation, eutrophication of water bodies, and aggravation of the greenhouse effect, among other negative environment-related consequences (Azadi et al. 2011). Certain food safety issues are exacerbated by the traditional food production methods. Several studies investigating the influence of conservative or organic farming on farmers indicate that negative news about food safety increases public mistrust of food safety and even undermine farmers’ interests (Hayat et al. 2022). Although the public is concerned about food safety, the World Health Organization estimates that 600 million people worldwide are ill each year owing to contaminated food, resulting in 420,000 deaths (Fahey et al. 2022).

Organic agriculture is an ecological mode of production that promotes environmental protection in most countries and regions worldwide (Ramakrishnan et al. 2021; Sarkar et al. 2021). Organic food production currently occurs in over 180 countries worldwide, and the organic agricultural industry is expanding (Willer et al. 2020). Organic food production adheres to specific principles, including the avoidance of genetic engineering in the production of organisms and their products; avoidance of chemical synthesis of pesticides, fertilizers, growth regulators, and feed additives, including the material; observation of natural rules and ecological principles; coordination of planting and breeding balance; and maintenance of the stability of the production system of an agricultural method (Hamzaoui-Essoussi and Zahaf 2012). Organic industry encompasses the development of organic concepts, fundamental organic research, research and promotion of organic production technology, organic production, organic certification, organic market, organic consumption, and organic supervision (Grassauer et al. 2022). Organic farmland constitutes 71.5 million hectares worldwide, and China ranks third with 3.135 million hectares. The global organic market generates nearly 97 billion euros in retail sales. China’s total retail sales amount to 8.1 billion euros, which is approximately 8% of total sales. This has made China the world’s fourth largest market for organic food. In 1999, only 11 million hectares of organic farmland were present worldwide. Currently, 71.5 million hectares of organic farmland are present worldwide. According to the 2023 Organic Product Certification and Industrial Development Report, issued by the National Administration of Market Supervision and Management, China’s organic product sales in 2022 constituted 877.6 billion yuan, making China the world’s fourth largest organic agricultural producer (Kong 2023).

Conventional agriculture uses pesticides and fertilizers during production that can pollute the air, water, soil, wildlife, and living environments (Maeder et al. 2002). The World Health Organization estimates that 3.5–5 million people worldwide suffer from acute pesticide poisoning each year, with approximately 300,000 deaths (Faber 2020). According to the French Institute for Cancer Research, the agricultural population is 2.6 times more likely to develop brain tumors than the average population; these adverse health conditions are caused by pesticides, insecticides, and growth hormones, which are widely used in agriculture (Gatto et al. 2021). Consequently, an increasing number of consumers are opting for organic food choices to avoid the damage caused by substances such as pesticides and fertilizers to their health (Cakirli and Theuvsen 2020). Furthermore, the Organic Center of America demonstrated 30% less energy consumption during the production of one kilogram of organic corn compared with the production of nonorganic corn; 20% less energy consumption during organic soybean production than conventional methods, in which organic grass-fed cattle require 50% less energy than the conventional method for grain-fed cattle. Organic farming is a strong carbon sequester in which the soil locks approximately 30% more carbon than conventional farming, effectuating a 15% or greater reduction in carbon dioxide emissions from these lands. Additionally, reduction in carbon dioxide emissions reduces the rate of global warming (Clark 2020). The high selling price of organic products motivates farmers to choose organic production and consumers to produce organically themselves. Most consumers of organic food believe that participating in the production of organic food can lead to healthier lifestyles and eating habits (Hansmann et al. 2020). Additionally, to effectively respond to food safety issues and ecological crises, governments at all levels in China provide support to organic farms, including finding land, funding farm infrastructure, and production training (Kong 2023). Nevertheless, as the urbanization rate in China continues to increase, the availability of rural land suitable for organic cultivation is becoming increasingly constrained. Consequently, the potential for organic food production in urban areas must be explored to offer novel insights and opportunities for advancing organic food production and understanding the attitudes and perceptions of urban residents toward organic production.

Prior research on organic food production emphasizes large-scale production and marketing of organic food (Azadi et al. 2011; Cakirli et al. 2020). Extant literature on organic production on small farms primarily focuses on rural areas (Czyżewski and Kryszak 2023; Sargani et al. 2023); studies on the growth of organic food production in urban areas are limited. Organic food production has attracted the attention of the scientific community as a pro-environmental strategy for mitigating climate change (Zhang et al. 2020). We reviewed previous studies related to pro-environmental behavior and observed that most theories used in these studies were based on the theory of planned behavior (TPB) (Yadav and Pathak 2017; Gansser and Reich 2023), with scholars indicating the shortcomings of these theories and advocating the adoption of alternative theories (Sniehotta et al. 2014; Cakirli and Theuvsen 2020). Zhang et al. (2020) compare the TPB and value-belief-norm (VBN) in the adoption of agricultural production methods to mitigate climate change. They conclude that the VBN theory has greater predictive power for altruistic behaviors, such as climate change mitigation. Stern et al. (1999) introduce the value theory and New Ecological Paradigm research into the norm activation model, forming the VBN theory. The model framework of the VBN theory mainly includes the dimensions of value, belief, and personal norm and is further subdivided into value, new environmental paradigm, awareness of consequences, ascription of responsibility, and personal norm. These five variables function together to form an inseparable causal chain, with each variable in the chain related to the next variable, which may be directly related to variables further downstream in the chain (Stern 2000). Prior research has extensively utilized the VBN model to explore and identify the major factors of pro-environmental actions (Rezaei-Moghaddam et al. 2020; Fornara et al. 2016; Steg et al. 2005). Hence, we employ the VBN theory to examine the phenomenon of noncommercial, urban organic production. Moreover, some researchers have emphasized the need to expand the VBN model in accordance with the research context. Trust, as a psychological factor, significantly increases pro-environmental behavioral intentions (Carfora et al. 2021; Nuttavuthisit and Thøgersen 2015). Several studies have applied trust as an extension of the VBN model to environmentally conscious contexts. Examples include consumers’ decisions to visit green hotels (Choi et al. 2015), the intention to separate municipal solid waste (Li et al. 2018), and the consumption of organic food (Yang et al. 2023). However, this expansion has not occurred in organic food production. Therefore, this study uses trust in organic food production as an extension of the VBN model to improve its predictive ability.

The global food safety crisis is gaining prominence, and consumer demand for organic products is expanding. Organic agriculture has an enormous potential for market expansion. Acceptance of organic food production not only increases farmer revenue but also protects the environment by eliminating the use of poisonous chemicals and fertilizers (Ullah et al. 2015). As a major exporter of organic products worldwide, China has broad prospects for organic food production. In this context, exploring the factors influencing organic food production improves the development of urban organic food production in China, meets people’s demands for healthy food, and enhances the efficiency of environmental governance. This study determines the specific factors related to the intention toward organic food production and growth based on the VBN theory.

Literature review

Theoretical foundation

According to Schwartz’s (1977) norm activation framework, individuals’ perceptions of conceivably detrimental consequences and their attributions (whether to adopt pro-environmental behaviors) motivate personal norms that determine whether they engage in pro-environmental behaviors. De Groot and Steg (2009) demonstrate significant correlations among awareness of consequences, ascription of responsibility, personal planning, and the propensity to protect society. Their findings provide empirical support for explaining the normative activation model. Following this, Stern et al. (1999) incorporate variables such as values and ecological worldviews, expand the norm activation model, and establish a VBN model. The VBN model focuses on the harmonious coexistence between man and nature, emphasizing that individuals appropriately perform their personal moral responsibilities toward the natural environment and consider the detrimental consequences and personal responsibilities to enhance their awareness of the risk perception of environmental protection behavior (Zainol et al. 2023). Psychological interventions strengthen positive environmental protection behaviors (Peck et al. 2020). The VBN model offers an essential theoretical basis for studying psychological variables in behavior. Based on the VBN model, the causal model transcends from personal values (i.e., biospheric, altruistic, and egoistic values) to the ecological worldview, awareness of consequences, ascription of responsibility, personal norms, and trust, and finally transitions into pro-environmental behavioral intentions and the actual behavior of organic food production (Gkargkavouzi et al. 2019; Karpudewan 2019; Al Mamun et al. 2022).

Hypotheses development

Values and belief

Numerous studies report that individuals placing higher value on issues other than their own self-interest, such as pro-social, altruistic, or biospheric values, are more likely to participate in pro-environmental behaviors (Osburg et al. 2019). However, values do not have a significant direct influence on behavior in general, and the link between general values and behavior is mediated by other variables such as behavior-specific beliefs or personal norms (Steg et al. 2005). Following the confirmation of the relationship between value-based goal differences and behavioral intention differences in value-based research on environmental protection behavior, scholars believe that different value orientations directly affect individuals’ beliefs in protecting the environment. This result aligns with value orientation (Ghazali et al. 2019; Hiratsuka et al. 2018). Personal value is divided into three aspects in the VBN theory: biospheric, altruistic, and egoistic values (Stern 2000). Numerous studies have demonstrated the positive influence of biospheric and altruistic values on forming ecological values (Osburg et al. 2019; Zhang et al. 2020). Nevertheless, divergent perspectives exist regarding the potential positive and negative ramifications of egoistic values on pro-environmental beliefs. Most studies confirm the positive effects of egoistic values on the ecological worldview (Zhang et al. 2020; Gupta and Sharma 2019; Lee et al. 2013). However, according to De Groot and Steg (2009a), individuals with strong self-interest values weigh the costs and benefits of environmental protection behaviors. When the perceived benefits of environmental protection are high, egoistic values foster positive pro-environmental beliefs. In the context of this study, the small-scale organic food production by urban botanists not only satisfies the inheritance of traditional Chinese farming culture but also produces healthy and secure food for them and their families using environment-friendly production methods at a low cost. Considering these factors, we propose the following hypotheses:

H1a: Biospheric values positively affect the ecological worldview.

H1b: Altruistic values positively affect the ecological worldview.

H1c: Egoistic values positively affect the ecological worldview.

Beliefs

Ecological worldview and awareness of consequence

The association between the ecological worldview and behavior is weak, possibly because behavior-specific beliefs and personal norms may buffer the relationship between the ecological worldview and behavior (Steg et al. 2005; Bamberg 2003). Awareness of consequences refers to an individual’s awareness regarding the adverse consequences that occur in others owing to the non-implementation of the target behavior. Based on the VBN theory, the ecological worldview is believed to encourage pro-environmental behavior by increasing awareness of the negative consequences of environmental behavior. Prior research has consistently underscored the significant influence of the ecological worldview in fostering pro-environmental behaviors by enhancing awareness of the negative consequences of environmental issues (Gupta and Sharma 2019; Chen 2014). Accordingly, we propose the following hypothesis:

H2: Ecological worldview positively affects awareness of consequences.

Awareness of consequences and ascription of responsibility

Ascription of responsibility refers to an individual’s sense of responsibility for the adverse consequences of failure to implement the target behavior (Ghazali et al. 2019; Kiatkawsin and Han 2017). Most people who realize that their actions may have negative impacts associate themselves with possible negative consequences and show a stronger sense of responsibility. Numerous studies evince a positive correlation between awareness and responsibility (Golob et al. 2018; Rui et al. 2021; Obuobi et al. 2022). For example, Rui et al. (2021) proves that the more aware people are about the environmental problems created by automobile use, the stronger their accountability. Hence, we propose the following hypothesis:

H3: Awareness of consequences positively affects the ascription of responsibility.

Norms

Personal norms

Schwartz (1977) believes that social norms are at the level of a social structure, whereas personal norms are internalized attitudes. Although individuals may internalize social norms, in some cases, their behaviors do not conform to these norms. Unless the personal norm is appropriate and relevant in a situation, the individual norm will neither activate nor lead to the corresponding behavior. Hence, within the framework of the VBN theory, Stern (2000) employs the concept of belief as the antecedent variable for norms to foster the development of personal norms pertaining to the environment. Numerous empirical studies demonstrate that, with a strong belief in the ecological worldview, awareness of consequences, and ascription of responsibility values, personal norms are capable of guiding corresponding behaviors (Park and Ha 2014; Steg et al. 2005; Golob et al. 2018; Park et al. 2022). Thus, we propose the following hypotheses:

H4a: Ecological worldview positively affects personal norms.

H4b: Awareness of consequences positively affects personal norms.

H4c: Ascription of responsibility positively affects personal norms.

Trust in organic food production

Recent research indicates that the VBN theory may be slightly changed to improve prediction power and model parsimony (Hiratsuka et al. 2018). This study also examines the predictive impact of trust in explaining intentions toward organic food. Trust is defined as a party’s belief or perception (Stephen 2001). The level of trust varies according to the level of belief, which is an aspect of norms. Additionally, it is challenging for consumers to achieve trust without belief, and trust has been introduced as an expanding factor of norms in the VBN theory in previous studies (Li et al. 2018; Yang et al. 2023). This study perceives the ecological worldview, awareness of consequences, and ascription of responsibility as botanists’ beliefs toward organic food production; their favorable attitude toward organic food production may also influence trust. Therefore, we propose the following hypotheses:

H5a: Ecological worldview positively affects trust in organic food production.

H5b: Awareness of consequences positively affects trust in organic food production.

H5c: Ascription of responsibility positively affects trust in organic food production.

Intention toward organic food production

Based on previous studies on the VBN theory, personal norms are hypothesized to impact pro-environmental behavior constructs, such as the adoption of organic food production, green consumers, green passengers, and recyclers (Rezaei-Moghaddam et al. 2020; Kiatkawsin and Han 2017; Steg et al. 2005). An individual feels stronger obligations and responsibilities for a matter that involves more expectations for related things, which creates a stronger intent to engage in behaviors that are beneficial to the matter. According to Cakirli and Theuvsen (2020), the intention to produce organic food is formed through personal norms and perceived effects of a specific invention. Rezaei-Moghaddam et al. (2020) also find that personal norms have a strong impact on farmers’ intentions to continue pro-environmental behaviors such as producing clean and environment-friendly technology. Therefore, personal standards can be used as predictors of intentions. Hence, we propose the following hypothesis:

H6a: Personal norms positively affect intention toward organic food production.

Trust is critical in the decision-making process associated with the adoption of pro-environmental behavior, given that only a few consumers are aware of the history of pro-environmental activity, whereas most are unable to verify it (Carfora et al. 2021; Nuttavuthisit and Thøgersen 2015). Residents’ trust is particularly relevant in this study because a general lack of knowledge about organic food production is exposed to the manufacturing and preparation of this food category. Some studies have proposed that individuals’ and organizations’ intentions and pro-environmental behaviors are influenced by their personal views, norms, trust, and the strength of existing connections (McTiernan et al. 2021; Kornilaki et al. 2019; Yang et al. 2023). Consequently, individuals should instill trust in the authenticity of the chosen organic food production methods. Hence, we propose the following hypothesis:

H6b: Trust in organic food production positively affects the intention toward organic food production.

Growing organic food

The immediate antecedent of behavior is perceived as behavioral intention, which has been confirmed in many studies and theories (Wang and Mangmeechai 2021; Ajzen 2002). Ajzen (2002) asserts that intention to perform is an immediate antecedent of actual behavior. People are more likely to perform desired actions than actions forced on them. Thus, this study perceives the intention toward organic food production as a key factor motivating the growth of organic food. Hence, we propose the following hypothesis:

H7: Intention toward organic food production positive affects organic food production.

Figure 1 illustrates the research framework for this study based on the above hypotheses.

Fig. 1
figure 1

Research Framework.

Full size image

Research methodology

To validate the previously proposed hypotheses, we conducted cross-sectional data collection in China between January and February 2022. Considering that this period was during the Chinese New Year and that some regions were still implementing coronavirus disease 2019 (COVID-19) epidemic-related lockdown policies, we adopted an online survey to collect data. The target population was urban residents with the ability to grow crops independently in their gardens or balconies. Owing to the large population of urban residents in China, obtaining a complete list of residents for the entire country to create a sampling framework was not feasible. Therefore, we employed the judgmental sampling method, a nonprobability sampling approach, to selectively invite respondents who met specific criteria to participate in the survey, based on their expertise and experience (Sekaran and Bougie 2016). This sampling method enhanced the representativeness and effectiveness of our sample and ensured that it aligned with the study’s objective, that is, investigating the key factors that influence the intention to grow organic food. For calculating the minimum sample size, this study used G* Power software and employed an F test with an effect size of 0.15, α level of 0.05, power of 0.80, and number of predictors 9. The minimum calculated sample size was 114 (Faul et al. 2009). Moreover, sample heterogeneity is a factor that must be carefully considered in sampling design. Urban farmers may differ in many aspects, such as their planting scale and methods and socioeconomic characteristics, which may influence their views and practices regarding planting behavior. To mitigate and avoid heterogeneity, we enlarged the dataset at the data collection stage by referring to the minimum sample size calculated by G*Power. A large sample size improves the likelihood of reflecting the overall characteristic distribution (IntHout et al. 2015). Furthermore, large sample size has higher flexibility for censoring and filtering based on sample characteristics. Therefore, we implemented a judgmental screening question before commencing the survey: “Do you have the capability to grow crops in your balcony or garden?” After completing data collection, the responses from participants who answered “No” to this question were excluded from the final sample, resulting in 671 valid questionnaires that aligned with the research background of the initially collected 847 questionnaires. Additionally, in the background survey, we included a question related to organic food: “Do you have experience with organic food?” The responses with the option “haven’t” were excluded. Finally, 621 valid questionnaires that met the research context and exceeded the minimum required sample size were obtained.

The original measurement items used in this study were developed in English; therefore, before data collection, professional translators were hired to translate the questionnaire and align it with the Chinese research context. To ensure the reliability and validity of the survey questionnaire, we implemented both pre- and pilot-testing procedures. Specifically, during the early stages of questionnaire preparation, a pre-test was conducted in which an experienced professor and several doctoral candidates reviewed the initial English questionnaire and provided feedback to address some of its details. Moreover, to prevent bias resulting from reading difficulties in the translated questionnaire, two professional language experts were hired to perform English-to-Chinese translation. After completing these steps, a small-scale pilot test was conducted involving 20 urban residents with planting experience. Respondents were divided into various age groups. Responses from the pilot test were excluded in the final sample. We recorded the average time taken by the pilot-test respondents to complete the online questionnaire (https://www.wjx.cn/vm/QSXiHma.aspx#), ensuring that the online survey link and QR code presented no technical barriers to opening, inputting, or submitting responses.

Survey instrument

The respondents were required to sign an informed consent form before completing the questionnaire. During this process, all participants were informed about the following: (a) primary purpose of the study, (b) participant’s right to withdraw or refuse to participate in the study, (c) ultimate flow of data, (d) confidentiality of the study, (e) there was no correct answer to any question in the study, and participants were required to answer genuinely, and (f) a brief explanation of the keyword, Urban Farming, was given. Subsequently, the questionnaire was divided into two parts: background and latent variable scale questions. The background questions included age, sex, education level, average monthly income, place of residence, and the sample screening questions mentioned above. A Likert scale was used in the scale questions primarily to measure the respondents’ values (biospheric, altruistic, and egoistic values), beliefs (ecological worldview, awareness of consequences, and ascription of responsibility), norms (personal norms, and trust in organic food production), and the intention and production of organic food.

The measurement items used in this study were integrated and adapted from existing research; while a 7-point Likert scale (1 = very unimportant, 7 = very important) was used to measure biospheric, altruistic, and egoistic values, and another 7-point Likert scale (1 = strongly disagree, 7 = strongly agree) was used to measure the ecological worldview, awareness of consequences, ascription of responsibility, personal norms, trust in organic food production, and intention toward organic food production and growth. The measurement items used in this study were modified according to the participants. This study unified measurement projects from different studies. In previous measurement projects, researchers used different expressions and some items had fewer than five measurements. The direct combination of these nonuniform measurements can lead to a certain degree of misunderstanding among respondents and may eventually result in common method bias (CMB) arising from data collection. In the process of adapting the questionnaire, the expression of similar measurement items was first deleted; for example, in the awareness of consequences, this study used only the dimensions of “global warming,” “environmental quality,” “life quality,” avoiding boredom and similarity of the questionnaires, and reducing the reading fatigue of the respondents. Second, the study unified the keywords in the questionnaire, such as trust in organic food production, intention toward organic food production, and organic food production, to avoid confusion in reading. Both the measurement items and sources of these components are presented in Supporting Material: S1. Survey Instrument. The specific measurement items for this study were integrated and adapted from specific articles: biospheric, altruistic, and egoistic values (Han et al. 2016); ecological worldview (López-Mosquera and Sánchez 2012); awareness of consequences (Choi et al. 2015; López-Mosquera and Sánchez 2012); ascription of responsibility (Ünal et al. 2019; López-Mosquera and Sánchez 2012); personal norms (Ünal et al. 2019; Choi et al. 2015); trust in organic food production (Chen 2010); intention toward organic food production (Chen and Deng 2016; Maichum et al. 2016); and organic food production (López-Mosquera and Sánchez 2012; Walton and Austin 2011).

Common method bias

To examine and avoid common method bias (CMB) potentially induced during data collection, we implemented both ex ante preventive measures and post hoc screening remedies. Ex ante preventive measures primarily included informed consent procedures before data collection, random distribution of measurement modules, keyword parsing, and clear instructions for item presentations. Prior research acknowledges that, while CMB can be controlled to some extent, it is impossible to completely prevent it (Podsakoff et al. 2003). To assess the extent of bias in the final samples, Harman’s single-factor test was used to examine the maximum variance accounted for by a single factor. The results revealed that the maximum variance accounted for by a single factor was 35.86%, which remained below the stringent threshold of 40% required to address CMB. Furthermore, we used Kock’s (2015) full collinearity tests to show the variance inflation factor values for biospheric values (2.230), altruistic values (2.211), egoistic values (1.198), ecological worldview (1.806), awareness of consequences (3.393), ascription of responsibility (1.977), personal norms (2.107), trust in organic food production methods (3.823), intention toward organic food production (2.693), and growing organic food (4.057). Overall, the results suggest that CMB does not exist in this study.

Data analysis method

After confirming that the CMB did not have any impact on the research results, we used the Web Power online statistical tool to assess the multivariate normality of the data (https://webpower.psychstat.org/w./tools/index). The results indicated a p-value of less than 0.05, suggesting the presence of non-normality in the research data, making it suitable for subsequent partial least squares structural equation modeling (PLS-SEM) analysis. PLS-SEM is a causal prediction method comprising two parts: a structural model (or internal model) and measurement model (or external model). It also has practical research applications. We used PLS-SEM to measure the external and internal structures of this study owing to several reasons: (a) PLS-SEM was used to predict the target structure, (b) the structural model had a formatively measured construct; (c) this study had multiple independent, dependent, and mediating variables. Moreover, the research model was complex, with multiple coplanarities and indicators. This aspect made it more suitable for the research method of PLS-SEM; and (d) PLS-SEM had strong relative robustness for data with non-normal distribution (Sarstedt et al. 2017).

Findings

Table 1 presents a descriptive analysis of respondents’ characteristics. The results of the age group show a high proportion of young people participating in this survey, with 76.9% under the age of 40 years and a majority of 18–30-year-olds, with a significantly higher enthusiasm for the topic in the younger group than in the middle-aged and older groups. Regarding income, the peak is 5001–7500 CNY, and decreases evenly on both sides. The educational distribution of the respondents shows that most of the respondents have a bachelor’s degree or equivalent, which is different from that shown in our previous survey on agriculture. In summary, the results show that, unlike previous research, youthfulness, high education, and high income appear among those concerned with urban organic farming.

Table 1 Demographic Characteristics.
Full size table

Measurement model

This study evaluated the model before measuring the external structural model and examined its convergent validity, discriminant validity, and internal consistency. The results (presented in Supporting Material: S2. Loading and Cross Loading) illustrate that all loadings show values greater than 0.7. However, all loadings have a value greater than that of the adjacent cross-loading (Hair et al. 2017). Table 2 demonstrates that the average variance extraction (AVE) values are higher than 0.50, indicating that the model achieved overall convergent validity. Meanwhile, the composite reliability values are between 0.877 and 0.945, which are significantly higher than the minimum level of 0.70 required to ensure internal consistency. The values of the heterotrait–monotrait (HTMT) ratio shown in Table 3 are below the threshold of 0.90 (Henseler et al. 2015), indicating a significant difference between all HTMT values. Meanwhile, the results of the Fornell–Larcker criterion indicate that the square root of the AVE is larger than the correlations of all other constructs. Therefore, discriminant validity is established for all the constructs.

Table 2 Reliability and Validity.
Full size table
Table 3 Discriminant Validity.
Full size table

Structural model

Bootstrapping analysis was conducted using SmartPLS 3, and the coefficients of determination (R2) and effect size (f2) were obtained using subsampling bootstrap (5000). Predictive relevance (Q2) values were calculated to evaluate the predictive relevance of the PLS path model. The R2 value for ecological worldview in this model is 0.326, indicating that the exogenous construct explains 32.6% of the variance in ecological worldview. The R2 for awareness of consequences and ascription of responsibility is 0.423 and 0.445, respectively, indicating that the exogenous construct in this model explains 42.3% and 44.5% of the variance in awareness of consequences and ascription of responsibility, respectively. The coefficient of determination for personal norms is 0.451, and the exogenous construct explains 45.1% of the variance in personal norms. Moreover, the exogenous construct in this model demonstrates a 40.9% trust in organic food production. The R2 for intention toward organic food production is 0.478, representing 47.8% of the intention toward organic food. The R2 value for growing organic food is 0.610, indicating that the exogenous construct represents 61% of the variance of growing organic food. Hair et al. (2017) propose that the importance of exogenous constructs in explaining endogenous constructs can be measured by assessing the effect size (f2). Given the thresholds for explaining effect sizes, an amount close to or less than 0.02 is small, close to 0.15 is moderate, and higher than 0.35 is substantial (Avkiran and Ringle 2018). In this study, the effect size of egoistic values on ecological worldview is 0.010, while that of ecological worldview on personal norms is 0.000. The ecological worldview on trust in organic food production and ascription of responsibility on trust in organic food production is 0.000, and personal norms on intention toward organic food production is 0.005. Overall, the minimum effect size is less than 0.02. The biospheric values of the ecological worldview shows an effect size of 0.111, whereas the altruistic values of the ecological worldview exhibits an effect size of 0.031, and the ascription of responsibility on personal norms shows an effect size of 0.036. These effect sizes are higher than 0.02 and lower than 0.15, indicating smaller effect sizes. Awareness of consequences and personal norms (f2 = 0.225) and awareness of consequences and trust in organic food production (f2 = 0.162), which have effect sizes higher than 0.15 but lower than 0.35, exhibit moderate effect sizes. Ecological worldview to awareness of consequences (f2 = 0.734), awareness of consequences to ascription of responsibility (f2 = 0.802), trust in organic food production to intention toward organic food production (f2 = 0.486), and intention toward organic food production to growing organic food (f2 = 1.561) demonstrate effect sizes higher than 0.35 and larger effect sizes. Finally, the values of predictive relevance (Q2) for blindfolding are greater than zero, indicating that the model has predictive relevance.

Path analysis

The structural model used in this study was based on assumptions made in previous studies. Bootstrapping analysis was performed using SmartPLS to evaluate the effects of previously proposed latent variables. Table 4 presents the results of the path analysis: biospheric (β = 0.378, p < 0.001, CI = (0.279, 0.465)), altruistic ((β = 0.205, x < 0.001, CI = (0.114, 0.299)), and egoistic values (β = 0.087, p = 0.016, CI = (0.020, 0.156)), which have positive and significant effects on ecological worldview. Subsequently, H1a, H1b, and H1c are supported. Ecological worldview (β = 0.650, p < 0.001, CI = (0.587, 0.711)) exhibits a positive and significant effect on awareness of consequences, thus supporting H2. Awareness of consequences (β = 0.667, p < 0.001, CI = (0.607, 0.728)) also shows a positive and significant effect on ascription of responsibility and supports H3. Awareness of consequences (β = 0.543, p < 0.001, CI = (0.451, 0.623)) and ascription of responsibility (β = 0.190, p < 0.001, CI = (0.099, 0.274)) have a significant positive effect on personal norms, while ecological worldview (β = −0.020, p = 0.311, CI = (−0.087, 0.048)) does not show a positive effect on personal norms. These results support H4b and H4c and reject H4a. Similarly, awareness of consequences (β = 0.478, p < 0.001, CI = (0.392, 0.574)) and ascription of responsibility (β = 0.221, p < 0.001, CI = (0.120, 0.311)) have a significant positive effect on trust in organic food production methods, whereas ecological worldview (β = −0.011, p = 0.392, CI = (−0.088, 0.054)) does not show a positive effect on trust in organic food production methods. Therefore, these results support H5b and H5c and reject H5a.

Table 4 Hypotheses Testing.
Full size table

Moreover, personal norms (β = 0.067, p = 0.077, CI = (−0.010, 0.141)) have nonsignificant effect on intention toward organic food production, thus rejecting H6a. Trust in organic food production methods (β = 0.647, p < 0.001, CI = (0.584–0.713)) have a significant positive effect on intention toward organic food production, thus supporting H6b. Finally, the intention toward organic food production (β = 0.781, p < 0.001, CI = (0.738, 0.823)) exhibits a positive and significant impact on growing organic food. This result supports H7. Figure 2 shows an important path diagram for all the above effects. Simultaneously, the path coefficient results indicate that zero is not present in the maximum and minimum confidence intervals in H1a, H1b, H1c, H2, H3, H4b, H4c, H5b, H5c, H6b, and H7. The t-values for these assumptions are greater than 3.090, supporting the hypotheses.

Fig. 2
figure 2

Measurement model.

Full size image

This study simultaneously examined the specific partial mediation effects. Table 5 shows that the intention toward organic food production does not significantly mediate the relationship between personal norms and organic food production. Moreover, the direct effects indicate that the ecological worldview has a nonsignificant influence on trust in organic food production and personal norms. However, the mediation effects shows that the ecological worldview plays a significant mediating role between biospheric values, altruistic values, egoistic values, and trust in organic food production. Specifically, the ecological worldview plays an essential role in mediating the impact of biospheric, altruistic, and egoistic values on trust in organic food production and personal norms. Despite the lack of a direct and significant impact of the ecological worldview on trust in organic food production and personal norms, the ecological worldview has emerged as a crucial factor in elucidating the intricate relationships between biospheric, altruistic, and egoistic values and trust in organic food production and personal norms.

Table 5 Mediating Effect.
Full size table

Discussion

First, value-related variables (biospheric, altruistic, and egoistic values) have significant positive effects on the ecological worldview. This indicates that biospheric, altruistic, and egoistic values inherent in individuals have significant effects on the construction of an ecological worldview. This is consistent with the process through which individuals develop ecological worldview. Previous pro-environmental studies have confirmed the positive effects of biospheric and altruistic values on the ecological worldview (Osburg et al. 2019; Zhang et al. 2020). Most previous studies on pro-environmental behavior show that egoistic values have a negative or nonsignificant effect on the ecological worldview (Gupta and Sharma 2019; Lee et al. 2013). In this study, egoistic values are positive and significant for the ecological worldview, confirming the conjecture of the study. This indicates that the formation of an individual’s ecological worldview is stimulated when individuals with strong egocentric values observe that the benefits of environmental protection behavior exceed their inputs. Compared with previous research, this study emphasizes noncommercial fields and social issues, which explains why egocentric values have a positive effect on ecological worldview. Furthermore, growing environmental consciousness within Chinese society and the increasing emphasis on healthy lifestyles have contributed to the emergence of additional intrinsic values of organic food production, such as improving social influence and inheriting traditional culture. These added values hold greater significance for individuals in noncommercial contexts who prioritize spiritual self-interest values.

Second, within the framework of VBN theory, numerous studies emphasize that the impact of value on pro-environmental behavior is generally moderate. Consequently, the theory emphasizes the establishment of a connection between beliefs and stimulating norms using terminal variables (Steg et al. 2005; Bamberg 2003). This study also demonstrates that the ecological worldview has significant positive effects on the awareness of consequences, awareness of consequences on ascription of responsibility, and ascription of responsibility on norm-related variables (personal norms and trust in organic food production methods). Thus, a chain of relationships is established. These findings indicate that the establishment of a positive ecological worldview can inform residents regarding the positive or negative impacts of related behaviors on the environment, helping them develop a sense of responsibility and increasing their willingness to adopt pro-environmental behaviors subsequently. This result is consistent with those of previous studies (Gupta and Sharma 2019; Chen 2014). Furthermore, based on the previous review, the ecological worldview and awareness of consequences have a direct effect on the norm in some studies (Liobikienė and Poškus 2019). Therefore, this study also investigates the direct relationship between ecological worldviews and awareness of the consequences of personal norms to enhance the rigorousness of the research findings. The results demonstrate that the ecological worldview has no significant effect on the two aspects of the norm (personal norms and trust in organic food production methods), whereas awareness of consequences has a positive and significant effect on personal norms and trust in organic food production methods. These findings indicate that belief-related factors further down the relationship chain are more likely to have a direct effect on norms. Specifically, in the process of continuously strengthening beliefs, awareness of negative outcomes caused by the environment has been able to cross the formation of individuals’ internal responsibility consciousness and thus directly influence norms. As a relatively abstract and generalized concept, it is difficult for an ecological worldview to directly influence pro-environmental norms in a particular context. Its primary purpose is to act as an intermediary connecting individual values with societal norms.

Third, norms, such as personal norms and trust, have been proposed by many scholars as important factors in pro-environmental behavioral intentions (McTiernan et al. 2021; Cakirli and Theuvsen 2020; Rezaei-Moghaddam et al. 2020; Kornilaki et al. 2019). This study partially proves this result through the VBN theory, in which trust in organic food production is a key factor that directly affects pro-environmental behavioral intentions. Trust is expressed as the degree of personal trust in organic production methods to achieve environmental protection, which leads botanists to instill higher trust in the environmental protection brought about by organic food production. Consequently, they are more willing to consume organic foods. However, personal norms do not significantly affect organic food production. The most likely explanation for this result is that small-scale organic food cultivation under the existing conditions is largely promoted by trust in organic production methods, which is also confirmed by the path coefficients (0.647) and R2 (0.478). Thus, we conclude that the uncertainty or distrust of production methods currently hinders urban botanists’ willingness to cultivate organic food. Once the trust issue is resolved, and the added value of organic production is considered, people are more willing to cultivate organic food without the constraints of personal norms. This indicates the value of incorporating trust into the VBN model in the context of this study.

Finally, intention is the most important factor affecting an individual’s actual behavior, as demonstrated in previous studies (Wang and Mangmeechai 2021; Ajzen 2002). A positive and significant relationship was observed between these two variables leading to the following conclusion regarding organic food production: the stronger the urban botanists’ intention to produce organic food, the more likely they were to do so. Therefore, when encouraging more people to actively produce organic food, an individual’s positive intention to produce organic food must be maintained.

Conclusion

Environmental and food safety concerns have increased social and academic interest in organic food production. Organic food production is promoted in rural areas worldwide as an environment-friendly method of sustainable agriculture. However, its application in urban areas is less prevalent. Therefore, this study focuses on urban botanists and builds a theoretical framework based on the VBN theory with the constructs of value (biospheric, altruistic, and egoistic values), belief (ecological worldview, awareness of consequences, and ascription of responsibility), norms (personal norms and trust in organic food production), intention toward organic food production, and growing organic food. All three variables (biospheric, altruistic, and egoistic values) positively impact the development of ecological worldview. Additionally, under the concept of belief, a positive relationship exists between the ecological worldview and awareness of consequences and between awareness of consequences and ascription of responsibility. This study confirms the positive impact of the ascription of responsibility on norms (personal norms and trust in organic food production) and the influence of awareness of consequences on norms (personal norms and trust in organic food production). To expand on the findings of this study, we present the concept of trust in organic food production as a broader construct within the norms of the VBN theory and provide empirical evidence supporting its positive influence on intention toward organic food production. However, this study does not confirm the positive impact of personal norms on intention toward organic food production, which further underscores the importance of trust in organic food production in facilitating the promotion of intention toward organic food production. Based on the aforementioned findings, this study contributes to the advancement of the VBN theory. Simultaneously, a unique perspective on growing urban noncommercial small-scale organic food is used to propose a new development direction for organic food production. Furthermore, it offers pertinent organizations a valuable reference point for devising future strategies to foster the sustainable development of urban organic agriculture.

Theoretical implications

This study combined the VBN theory with organic food production in the context of noncommercial cities in China. However, this aspect has rarely been discussed in previous studies. The above discussion indicates that the values, beliefs, and norms in the VBN theory are relevant to the process of organic food production. This significant and positive influence proves the validity of the VBN theory in the context of green sustainable development. Furthermore, while numerous studies demonstrate the broad applicability of the VBN theory to environment-related social issues (Ghazali et al. 2019), few have integrated the VBN framework with organic food production. As organic food producers in a noncommercial city, urban botanists’ values directly affect their attitudes toward the relationship between themselves and the ecological environment, subsequently affecting the construction of norms and, ultimately, the individual’s pro-environment intentions and behaviors. This dynamic process explains the psychological decision-making of Chinese agricultural producers, particularly regarding the adoption of organic production methods. It also presents a significant reference for scholars to study the pro-environmental behaviors associated with organic food production in the future. In addition, trust was introduced into the VBN theoretical framework as a norm-related variable, and was found to be a good predictor of organic food production intentions. Therefore, an extension of the VBN theoretical framework may enable readers and researchers to improve their understanding of urban botanists’ decisions to produce organic food, particularly when new eco-friendly products and technologies emerge.

Practical implications

From the perspective of value, this study shows that the formation of an individual’s ecological worldview is affected by the joint effects of biospheric, altruistic, and egoistic values. According to the VBN theory, values have a significant impact on individuals’ beliefs about engaging in pro-environmental behaviors. Values act as guiding principles in people’s lives and the formation of individual values is influenced by their experience, education, environment, and people around them. Therefore, when encouraging individuals to form new environmental behaviors, they should focus on improving their absorption of environmental protection knowledge at the educational level. For example, communities regularly conduct promotional lectures on organic food production.

From the perspective of beliefs, the establishment of positive ecological world values, increased awareness of the benefits of adopting organic food production for urban botanists, and a sense of responsibility for environmental protection contributed to the establishment of personal norms and enhanced respondents’ trust in organic food production. An ecological worldview can be interpreted as an individual’s general belief about the connection between humans and nature. Therefore, it is an effective method to actively establish a connection between humans and nature, develop people’s realization of the importance of ecological balance for human survival, break traditional human-centered thinking, and allow people to actively take responsibility for environmental protection.

From the perspective of norms, this study includes both personal norms and trust in organic food production. However, only the positive effect of trust in organic food on intention to produce organic food was confirmed. Regarding trust in organic food, compared with the pollution of soil and water resources caused by traditional agriculture, organic food production can reduce environmental harm to a certain extent and promote sustainable development. The government and relevant departments should increase the publicity and explanation of organic food production technology to build a clear understanding among the public about the organic food production process, considering that individuals are more likely to trust the publicity of nonprofit organizations, such as the government, than other nonofficial organizations.

To maintain the organic food production intentions of urban botanists, the government should formulate targeted long-term support and incentive policies to replace short-term incentives. Similarly, relevant organizations should provide technical guidance for organic production technologies or products. For example, the government can cooperate with universities to publish relevant guidelines for urban private gardens. Universities and related researchers should also be encouraged to develop seeds suitable for urban private gardens. Subsequently, it reduces the resistance among people to adopt organic food production methods and makes them realize that organic food production is a long-term sustainable development measure supported by the government.

Limitations and future directions

Based on the limitations of this study, we suggest directions for future research. First, it was conducted in the context of the COVID-19 pandemic. The strict policies to combat COVID-19 in China have caused high uncertainty regarding travel and livelihood security, particularly regarding the purchase of fresh vegetables. Despite government efforts to deploy resources to reduce the negative effects of the lockdown, small outbreaks continue to create uncertainty regarding the security of fresh vegetables. Considering the instability of the social environment during an epidemic, the results of this study may not fully explain the different behaviors that may exist during and after the epidemic. Accordingly, future studies should incorporate more social factors to better understand the current intentions and other behavioral factors toward organic food and organic food production. Second, as a key variable of the VBN theory, personal norms do not have a significant effect on pro-environmental behavior in this study. However, this may result from multiple factors, such as the current developmental status of organic food production in China, public’s understanding of organic food production, and role of personal norms. Therefore, future research should continue to model personal norms as part of the norms while introducing additional norms related to the research background to expand the VBN model and increase its explanatory power. Moreover, this study was conducted in China using only small-scale data, and a random sample was obtained from the Internet. Owing to the diversity in China’s demographics, different social backgrounds may impact the demand for organic food. Therefore, the results of this study do not represent all individuals in China. However, it is important to emphasize that this study focuses on interpretation rather than generalization of the sample. Future research should be conducted in different social and regional contexts to provide further support for the results and literature.

Previous research has focused mainly on exploring the influence of external factors on personal environmental behavior. This study offers a good direction for future research, considering the impact of deep intrinsic psychological factors on individual behavior. Future research should consider conducting more comprehensive work, combining quantitative and qualitative research to comprehensively explore the main factors that influence individual environmental conduct, such as fuzzy set qualitative comparative analysis (), or introducing analytical methods such as necessary conditions analysis, based on the results, to provide more accurate evidence for future policy guidance, such as promoting public pro-environmental behaviors. Additionally, the reasons urban botanists choose organic food production have not been comprehensively discussed. Whether the prices of organic products or other external incentives influenced the decisions of urban botanists is excluded in the discussion. Future research should consider including these factors and in-depth discussion on the extensions and policy guidelines related to urban agriculture based on this study.