A Study on the Heterogeneity of Consumer Psychological Mechanisms of Dual Decision-Making Agents in Forest Educational Tourism: The Moderating Effect of Family Decision-Making Empowerment

Abstract

The consumption decision-making in educational tourism exhibits dual-agent characteristics, requiring alignment of consumption intentions between both agents to generate actual purchasing behavior. However, research on this characteristic is still relatively scarce. Understanding the psychological mechanisms and heterogeneity of consumption decision-making among students and parents in forest educational tourism is crucial for implementing precise consumer incentive strategies in related tourist attractions. This study constructs a theoretical model of the consumer psychological mechanism of dual decision-making agents in forest educational tourism, incorporating perceived value and perceived risk based on the Theory of Planned Behavior. A structural equation model is employed to validate the explanatory power and heterogeneity of this theoretical model, as well as to explore the moderating effect of family decision-making empowerment. The results indicate that the formation of the consumer psychological mechanism of dual decision-making agents in forest educational tourism is heterogeneous: the negative impact of perceived risk on perceived behavioral control and the positive impact of perceived behavioral control on consumption intention are only valid in the student group, not in the parent group; perceived behavioral control serves as a mediator only in the relationship between perceived value, perceived risk, and consumption intention for the student group, without any mediating effect for the parent group; family decision-making empowerment moderates certain paths in the consumer psychological influence mechanism of forest educational tourism decision-making agents. This study expands the Theory of Planned Behavior, enriching the research perspective on factors influencing consumption psychology, exploring the heterogeneity of dual decision-making agents in educational tourism, and examining the impact of family decision-making empowerment on consumer psychology. The findings provide relevant tourism enterprises and forest attractions with a deeper understanding of the consumption psychology of dual decision-making agents in forest educational tourism, offering a scientific basis for tourism enterprises and forest attractions to optimize marketing strategies, while also enhancing the consumption experience on the demand side.

1. Introduction

Learning is closely related to tourism [1]. Even when education is not explicitly recognized as a primary objective or motivation for travel, learning occurs as an incidental outcome of tourists’ continuous exposure to novel and diverse environments, coupled with the imperative to navigate unfamiliar and challenging situations encountered during their journeys [2,3]. Consequently, tourism is widely recognized as a critical context for experiential and informal learning [4,5]. Forest resources provide a high-quality environment and rich educational experiences for this type of learning-oriented tourism. In a global context, the market share of learning-oriented tourism products is rapidly expanding due to government support and increasing demand. Benefiting from the unique benefits of forests and the country’s rich forest resources, this growth is particularly evident in forest educational tourism, which has clearly become an important component of educational tourism.

As an innovative approach that integrates tourism and education, forest educational tourism is gaining increasing attention worldwide. In Europe and North America, the forest school model has been widely adopted and has become a key component of educational reform [6,7]. As a specific form of educational tourism, the forest school model not only enhances students’ awareness of the environment but also contributes to their mental health [8,9]. In recent years, the Chinese government has gradually implemented a series of policies to encourage the undertaking of educational tourism for primary and secondary school students with the objective of promoting an innovative educational method that combines knowledge and action [10]. In response to these governmental initiatives, the China National Committee for Forest Awareness has proposed the establishment of a comprehensive study and education activity system aimed at engaging “300 million young people entering the forest” by 2025 [11], which provides new ideas for business development in scenic spots within forests. The research progress and existing gaps in the factors influencing the consumption psychology of educational tourism, the decision-making agents in consumption, and family decision-making empowerment are described in detail as follows.

First, existing research on the factors influencing the consumption psychology of educational tourism has largely focused on perspectives such as the motivations for educational tours, perceived value, self-efficacy, and opinions of relevant groups. For example, King et al. analyzed the motivations for educational tourism from a stakeholder perspective [12]; Li et al. explained the benefits of family-based educational tourism for children from the perspective of perceived value [13]; Dou et al. explored the emotional learning outcomes gained by young people from short-term overseas educational travel, noting that self-efficacy is one of the key variables influencing consumer psychology [14]; and Chen proposed that the opinions of relevant groups significantly affect tourists’ choice of educational tourism destinations [10]. However, existing studies have not systematically reviewed the theoretical frameworks of previous research nor proposed an integrated and effective explanatory framework. While some scholars have utilized the Theory of Planned Behavior to validate the significant positive effects of behavioral attitudes, subjective norms, and the educational context on the willingness to engage in traditional cultural study tours [15], these studies have overlooked two critical factors—perceived value and perceived risk—that influence consumer psychology. Furthermore, most research has measured constructs without considering specific consumption scenarios or the impact of external contextual factors [12,13,14], which clearly undermines the rigor and scientific validity of the survey measurement process. Therefore, this paper extends the Theory of Planned Behavior, incorporating the variables of perceived value and perceived risk within the context of forest educational tourism to enhance the explanatory power of the model and provide a deeper analysis of the consumption psychology of forest educational tourism.

Second, consumer decision-making in educational tourism has a dual-agent characteristic, where students are the experiential consumers and parents are the purchasers [16,17]. For consumption behavior to occur, the consumption intentions of both agents must align. Furthermore, due to the differing roles of students and parents in the decision-making process, as well as variations in values shaped by age and upbringing, there is heterogeneity in how students and parents perceive the value of educational tourism and their respective psychological mechanisms. It is essential to consider both students and parents as decision-making agents during the consumption decision-making process. A dual-agent analytical framework can reveal the interactive roles and relationships between the two in the consumer decision-making process. However, the existing research on educational tourism has predominantly explored consumer behavior from the singular perspective of either parents or students [18], and there is a lack of empirical studies that investigate the heterogeneity of consumer psychological mechanisms from the dual perspectives of the experiencer and the purchaser. In light of this context, this study explores the heterogeneity and boundary conditions of the consumer psychological mechanisms of dual decision-making agents in forest educational tourism. Our aim is to clarify the differences in consumption focus and decision-making behavior characteristics between the two agents, thereby addressing the gap in existing research that has predominantly focused on a single decision-making perspective.

Third, due to the separation of roles between purchasing and consumption experiences in the decision-making process of educational tourism, examining the factor of family decision-making empowerment is highly beneficial for analyzing the formation process of consumer psychological mechanisms. While existing research has explored the relative influences of parents and children in family vacations and group travel decisions [19], there has been limited investigation into how the level of family decision-making empowerment moderates the formation process of family members’ travel consumer psychological mechanisms. Therefore, it is essential to validate the moderating effect of family decision-making empowerment in this study.

This study makes several potential contributions and innovations. First, it introduces the variables of perceived value and perceived risk within the context of forest educational tourism, extending the Theory of Planned Behavior. At the same time, the questionnaire used in this study includes specific information on forest attraction environments, pricing, duration, and courses, allowing for a more realistic simulation of consumer decision-making and enhancing the validity and reliability of the results. Second, from the dual perspective of experiencers and buyers, this study examines the heterogeneity of the consumption psychology of forest educational tourism, revealing differences in decision-making roles. Third, the study introduces family decision-making empowerment as a moderating factor, exploring its impact on consumer psychology in the family decision-making process. The findings provide a theoretical basis for tourism businesses and forest attractions to develop more targeted marketing strategies for forest educational tourism programs.

2. Theoretical Foundation and Research Hypotheses

2.1. Educational Tourism

Educational tourism is typically defined as travel that involves learning as a primary or secondary purpose, involving travel from one’s residence to unique environments for educational purposes [20]. The core characteristic of educational tourism is that the motivation for learning can be a primary or secondary factor driving the travel, and the learning can be formal (through experts or guides) or informal (self-driven) [21]. Currently, research on educational tourism primarily focuses on three aspects. The first is the study of the effects of the educational tourism industry, which indicates that educational tourism not only provides educational value but also has a significant positive impact on regional economic growth [22]. The second is the study of tourist satisfaction with educational tourism experiences, exploring how memorable travel experiences serve as mediators that affect travel motivation and overall tourist satisfaction [23]. The third is the study of the psychological factors that influence consumption, where perceived value and perceived risk are key factors affecting tourists’ consumption psychology; these factors significantly influence tourists’ decision-making processes, thereby impacting their consumption behavior [18]. Despite the progress made in educational tourism research, there are still limitations. Most existing studies are review articles, with relatively few empirical studies and an incomplete theoretical framework. Additionally, current research primarily focuses on student populations, or the role of parents as purchasers in educational tourism [19], lacking a dual-perspective comparison. Therefore, this study is based on the Theory of Planned Behavior and employs structural equation modeling to validate the consumption psychological mechanisms of dual decision-making agents in forest educational tourism, aiming to address these research gaps and provide a new theoretical framework for the study of forest educational tourism.

2.2. Theory of Planned Behavior

The Theory of Planned Behavior (TPB) is an extension of the Theory of Reasoned Action (TRA) [24]. This theory posits that an individual’s behavioral intention directly influences actual behavior, while behavioral intention is affected by attitude toward the behavior, subjective norms, and perceived behavioral control [25]. Specifically, attitude refers to the individual’s willingness to engage in a particular behavior, encompassing both negative and positive attitudes; subjective norms pertain to the perceived social pressures from others or organizations regarding whether one should perform a specific behavior; and perceived behavioral control refers to the perceived ease or difficulty of performing the behavior [26].

The Theory of Planned Behavior (TPB), as one of the key theories in consumer behavior research, has been widely applied in tourism studies, especially with regard to analyzing tourists’ decision-making processes. For example, Lin et al. explored the factors influencing tourists’ intention to revisit forest tourism destinations based on TPB and Protection Motivation Theory (PMT) [27]; Chen and Tung extended the TPB model to analyze the decision-making factors in consumers’ choice of green hotels [28]; and Nguyen et al. applied TPB to study the determinants of tourists’ travel intentions during the COVID-19 pandemic, highlighting the importance of government trust in tourism decision-making [29]. In addition, TPB has also been applied in the field of animal tourism, studying tourists’ ethical responsibility when participating in animal-related activities and their purchase intentions [30,31]. These studies demonstrate that the application of TPB in tourism not only reveals tourists’ behavioral intentions but also provides an in-depth understanding of the psychological mechanisms behind their decision-making processes.

Although TPB has been widely applied in various tourism topics, research in the field of educational tourism remains relatively limited. For example, Brune used the TPB model to investigate whether agricultural study tours could promote tourists’ purchase intentions towards local agricultural products [32], but similar applications have not yet been explored in the context of forest educational tourism. Overall, the Theory of Planned Behavior is typically used to analyze tourists’ decision-making processes before visiting tourism destinations [33], and by exploring intrinsic psychological factors, it provides theoretical support for marketing strategies and planning for tourist destinations. Few studies have integrated the key variables of perceived value and perceived risk, which significantly influence consumer decisions, into the TPB model. This study attempts to address this gap by constructing a comprehensive model of the consumption psychology mechanism in forest educational tourism based on TPB, with the aim of developing a framework for analyzing the consumption psychology mechanism of dual decision-making agents in forest educational tourism.

2.3. Perceived Value

Perceived value in forest educational tourism refers to the benefits that dual decision-making agents perceive from purchasing forest educational tourism products relative to the energy, expenses, and time invested in the consumption process [34,35]. In the marketing field, there is a perspective that “customers do not need products; they need the value that products bring”. Therefore, consumers’ evaluation of a product’s value is a prerequisite for determining whether they develop consumption intention [36]. Consumers’ perception of a product’s value contributes to the formation of a positive behavioral attitude [37]. Additionally, there is a relationship between consumers’ perceived value and the social pressures related to purchasing the product, as well as the perceived resources they possess (such as information, expenses, and time) or anticipated obstacles [38]. Overall, the greater the perceived benefits purchasing forest educational tourism products, the more positive the decision-makers’ attitudes toward these products will be, leading to a higher susceptibility to social pressures, a greater sense of adequacy in their perceived resources, and a stronger ability to overcome challenges associated with educational tourism, resulting in more intense consumption intention. Based on this, Hypotheses 1, 2, 3, and 4 are proposed:

H1. 

The perceived value of forest educational tourism by dual decision-making agents significantly positively influences subjective norms.

H2. 

The perceived value of forest educational tourism by dual decision-making agents significantly positively influences behavioral attitude.

H3. 

The perceived value of forest educational tourism by dual decision-making agents significantly positively influences perceived behavioral control.

H4. 

The perceived value of forest educational tourism by dual decision-making agents significantly positively influences consumption intention.

2.4. Perceived Risk

Perceived risk in forest educational tourism refers to the uncertainty that dual decision-making agents may have regarding the accuracy of their expected outcomes before purchasing forest educational tourism products, as well as the potential physical, financial, and psychological risks that could lead to unpleasant experiences during the forest educational tourism experience [39]. Perceived risk theory is frequently used to study consumer purchasing behavior, and existing research has validated that perceived risk has a negative impact on purchase intention [40]. Consumers’ perception of potential risks associated with purchasing behavior can lead to the formation of negative behavioral attitude [41]. Additionally, there is a relationship between consumers’ perceived risk and the social pressures they perceive, as well as the resources they possess or anticipated obstacles [42]. Overall, the greater the perceived risks associated with purchasing forest educational tourism products by dual decision-making agents, the more negative their attitude toward purchasing these products will be, leading to reduced susceptibility to social pressures, a reduced sense of adequacy with regard to their perceived resources, and a weaker ability to overcome challenges in educational tourism, resulting in lower consumption intention. Based on this, hypotheses 5, 6, 7, and 8 are proposed:

H5. 

The perceived risk of forest educational tourism by dual decision-making agents significantly negatively influences subjective norms.

H6. 

The perceived risk of forest educational tourism by dual decision-making agents significantly negatively influences behavioral attitude.

H7. 

The perceived risk of forest educational tourism by dual decision-making agents significantly negatively influences perceived behavioral control.

H8. 

The perceived risk of forest educational tourism by dual decision-making agents significantly negatively influences consumption intention.

2.5. Subjective Norms, Behavioral Attitude, Perceived Behavioral Control, and Consumption Intention

This study introduces the Theory of Planned Behavior into the field of forest educational tourism consumption. In the TPB model, behavioral attitude refers to the evaluations of preference and attractiveness that the dual decision-making agents have regarding forest educational tourism courses and projects. Subjective norms refer to the social pressures perceived by the dual decision-making agents concerning the purchase of forest educational tourism products. Perceived behavioral control pertains to the resources (such as information, funding, and time) or anticipated obstacles that the dual decision-making agents have regarding the consumption behavior of forest educational tourism products.

Existing research has widely confirmed that the three variables of subjective norms, behavioral attitude, and perceived behavioral control in the Theory of Planned Behavior are significantly positively related to consumption intention, and that there are also mutual influences among these three variables [43]. Han et al. used the TPB model to predict individual intentions in bicycle tourism [44], with the results indicating that behavioral attitude, subjective norms, and perceived behavioral control have a significant positive impact on tourists’ intention to participate in bicycle tourism. Qiu et al. confirmed that behavioral attitude, subjective norms, and perceived behavioral control directly influence the environmentally friendly behavioral intentions of both Generation Z and older generations [45]. Based on the above research, this study posits that in the field of forest educational tourism consumption, the three core variables of behavioral attitude subjective norms and perceived behavioral control of the dual decision-making agents positively influence purchasing intentions.

Generally speaking, the more positive the attitude of dual decision-making agents toward forest educational tourism consumption, the greater the pressure from social networks, and the more adequate the resources they possess, the stronger their intention to purchase forest educational tourism products will be. Based on this, the following hypotheses 9, 10, and 11 are proposed:

H9. 

Subjective norms significantly positively influence the consumption intention of dual decision-making agents in forest educational tourism.

H10. 

Behavioral attitude significantly positively influences the consumption intention of dual decision-making agents in forest educational tourism.

H11. 

Perceived behavioral control significantly positively influences the consumption intention of dual decision-making agents in forest educational tourism.

2.6. The Moderating Effect of Family Decision-Making Empowerment

Family decision-making empowerment refers to the size and proportion of decision-making authority that family members have in internal family decisions. A higher level of family decision-making empowerment indicates a more equal status among family members, meaning that the individual perspectives of family members are more likely to be reflected in the family’s consumption decisions [46]. The level of family decision-making empowerment can influence individuals’ subjective norms, behavioral attitudes, and perceived behavioral control. Specifically, family members with higher decision-making authority may develop a more positive attitude toward forest educational tourism and feel supported and encouraged by other family members, which can effectively enhance their perceived behavioral control [47]. Conversely, the opposite may occur. The degree of power family members have in the decision-making process directly affects their sense of control over the execution of behaviors, thereby influencing behavioral intentions and actual consumption behaviors [48]. This positions family decision-making empowerment as an important moderating variable that impacts consumer psychology and behavior within the TPB framework. Differences in perceived behavioral control among groups with varying levels of family decision-making empowerment can also lead to distinct consumer psychology and behavioral expressions during decision-making. Existing research has introduced social power theory to study the influence of children’s social power or parental social power on family purchasing decisions, validating the impact of children on family decision-making and indicating the necessity of considering both parental and child power in the family consumption decision-making process [49]. In the context of forest educational tourism consumption, the dual decision-making agent characteristic is present; thus, it is essential to fully understand the role of decision-making agents’ perceived level of family decision-making empowerment in the formation of their consumption intention. Therefore, this study introduces family decision-making empowerment as a moderating variable to explore its level in the consumer psychological influence mechanism of forest educational tourism decision-making agents. Therefore, the following hypothesis 12 is proposed:

H12. 

The consumer psychological mechanism for forest educational tourism exhibits significant differences among groups with different levels of family decision-making empowerment.

In summary, this study constructs a theoretical model based on the Theory of Planned Behavior, incorporating six variables: perceived value, perceived risk, subjective norms, behavioral attitude, perceived behavioral control, and consumption intention, to explore the heterogeneity of the consumption-related psychological mechanism of dual decision-making agents in forest educational tourism while examining the moderating effect of family decision-making empowerment, as shown in Figure 1.

3. Research Methods

3.1. Questionnaire Survey and Measurement

Our questionnaire consists of two main sections: one section measures the demographic variables of the dual decision-making agents in forest educational tourism, while the other section assesses six latent variables: behavioral attitude, subjective norms, perceived behavioral control, perceived value, perceived risk, and consumption intention. The questionnaire also provided specific information related to forest educational tourism that could influence consumption intention, including details about the forest destinations (located 130–160 km from the residential area), forest environment images, duration (at least one night’s stay), accommodation and meal prices, and the cost of forest educational courses. The measurement of behavioral attitude primarily references the scale designed by Zhou et al. (2018) [50], which examines the interest level of students and parents in forest educational tourism courses and programs. Subjective norms are measured based on the scale designed by Yadav et al. (2016) [51], while perceived behavioral control is measured using the scale developed by Cao (2020) [52]. Perceived value is assessed according to the scale by Sánchez et al. (2006) [53], while perceived risk focuses on the potential economic, physical, and psychological harm students may encounter during forest educational tourism. Consumption intention is measured using the scale designed by Zhang et al. (2021) [54]. All scales employ a five-point Likert scale, with response options ranging from “1 = strongly disagree” to “5 = strongly agree”. Family decision-making empowerment is measured using a single item: If you and your child (parent) discuss whether to participate in such a forest educational tourism activity, who ultimately makes the decision?

To facilitate understanding and ensure precise measurement, this study designed two sets of questionnaires for students and their parents (see

Table 1. Confirmatory factor analysis for the student group.

Variables	Measurement Items	Student Group
		Factor Loadings	CR	AVE	Cronbach’s α
Perceived value (PV)	The time invested in participating in forest educational tourism is worthwhile	0.801	0.878	0.645	0.874
	Forest educational tourism allows me to learn and make progress	0.875
	Collaborative experiences with others in forest educational tourism help me better adapt to society	0.845
	Forest educational tourism is more beneficial than extracurricular classes for the same cost	0.676
Perceived risk (PR)	There may be falls or injuries during the forest educational tourism process	0.859	0.890	0.669	0.889
	There may be damage to or loss of belongings during the forest educational tourism process	0.852
	There may be mosquito bites during the forest educational tourism process	0.800
	There may be discomfort related to food and accommodation during the forest educational tourism process	0.757
Subjective norms (SN)	I also want to participate in the forest educational tourism since my classmates are joining	0.862	0.817	0.535	0.808
	I also want to participate in the forest educational tourism since my friends are joining	0.838
	I also want to participate in the forest educational tourism since my relatives’ children are joining	0.631
	I will participate in the forest educational tourism as the school is encouraging it	0.546
Behavioral attitude (BA)	My level of interest in knowledge about natural organisms in the forest environment	0.559	0.827	0.550	0.822
	My level of interest in learning survival skills in forest and other outdoor environments	0.742
	My level of interest in activities that enhance personal courage	0.891
	My level of interest in activities that cultivate teamwork and cooperation	0.737
Perceived behavioral control (PBC)	My family’s financial capability can support my participation in educational tourism	0.633	0.844	0.521	0.843
	I can obtain enough information to participate in forest educational tourism	0.775
	I have the time to participate in forest educational tourism	0.730
	I have the physical strength to participate in forest educational tourism	0.781
	I have the ability to cope with unexpected situations that may arise during forest educational tourism	0.680
Consumption intention (CI)	As long as it benefits my growth, I will participate in forest educational tourism	0.519	0.879	0.604	0.881
	As long as I am interested, I will participate in forest educational tourism	0.653
	I am willing to recommend forest educational tourism to my classmates	0.960
	I am willing to recommend forest educational tourism to my friends	0.934
	I am willing to recommend forest educational tourism to my relatives	0.727

and

Table 2. Confirmatory factor analysis for the parent group.

Variables	Measurement Items	Parent Group
		Factor Loadings	CR	AVE	Cronbach’s α
Perceived value (PV)	I believe that the time invested by children in participating in forest educational tourism is worthwhile	0.879	0.923	0.750	0.921
	Forest educational tourism aligns with my expectations for my child’s development	0.924
	I believe that collaborative experiences with others in forest educational tourism help children better adapt to society	0.882
	I believe that forest educational tourism is more beneficial than extracurricular classes for the same cost	0.771
Perceived risk (PR)	There may be falls or injuries during the forest educational tourism process	0.839	0.860	0.607	0.859
	There may be damage to or loss of belongings during the forest educational tourism process	0.706
	There may be mosquito bites during the forest educational tourism process	0.814
	There may be discomfort related to food and accommodation during the forest educational tourism process	0.751
Subjective norms (SN)	I also want my child to participate in the forest educational tourism since my relatives’ children are joining	0.847	0.909	0.716	0.904
	I also want my child to participate in the forest educational tourism since my friends’ children are joining	0.891
	I also want my child to participate in the forest educational tourism since my colleagues’ children are joining	0.929
	I will let my child participate in the forest educational tourism as there is information in the WeChat group	0.700
Behavioral attitude (BA)	I believe my child’s level of interest in knowledge about natural organisms in the forest environment (Parent)	0.781	0.912	0.721	0.911
	I believe my child’s level of interest in learning survival skills in forest and other outdoor environments	0.856
	I believe my child has a level of interest in activities that enhance personal courage	0.890
	I believe my child has a level of interest in activities that cultivate teamwork and cooperation	0.865
Perceived behavioral control (PBC)	My family’s financial capability can support my child’s participation in educational tourism	0.620	0.797	0.441	0.793
	I can obtain enough information to help my child participate in forest educational tourism	0.664
	My child has the time to participate in forest educational tourism	0.602
	My child has the physical strength to participate in forest educational tourism	0.762
	I believe my child has the ability to cope with unexpected situations that may arise during forest educational tourism	0.662
Consumption intention (CI)	As long as it benefits my child’s growth, I will take my child to participate in forest educational tourism	0.522	0.923	0.676	0.922
	As long as my child is interested, I will take my child to participate in forest educational tourism	0.560
	I am willing to recommend forest educational tourism to other parents	0.913
	I am willing to recommend forest educational tourism to my colleagues	0.945
	I am willing to recommend forest educational tourism to my relatives	0.924
	I am willing to recommend forest educational tourism to my friends	0.944

). It is important to note that the questionnaire items were measured within the specific context of forest educational tourism consumption, providing information on the costs associated with transportation, accommodation, and course fees incurred by the decision-making agents. This approach offers a more realistic reflection of the true consumption intention of the decision-making agents compared to general inquiries about consumption intention, thereby ensuring the authenticity of the measurements related to the consumer psychological mechanism in forest educational tourism.

3.2. Data Collection

The formal survey was conducted from December 2020 to January 2021. Due to pandemic control measures, this study utilized the Questionnaire Star platform to distribute questionnaires to ordinary middle schools in four cities: Shenyang, Anshan, Harbin, and Dalian. A total of 398 student questionnaires and 930 corresponding parent questionnaires were collected. Considering the difficulty in ensuring the quality of self-administered questionnaires, strict screening of the student and parent questionnaires was conducted using response time, designated items, and reverse-coded items. Ultimately, 312 valid student questionnaires and 300 valid parent questionnaires were obtained, with response rates of 78.4% and 32.3%, respectively. We designed two sets of questionnaires targeting two groups: students and their parents. The pairing of questionnaires was achieved by having students fill in their parents’ nickname on the student questionnaire. During the initial phase of data collection, the questionnaires were distributed online in Harbin’s regular middle school classes, with multiple class teachers distributing them to student and parent groups. However, the response rate for valid parent questionnaires was low (over half were invalid based on directed and reverse questions), which prevented sample pairing. Subsequently, separate questionnaires were distributed to the two groups (students and parents) in other cities’ regular middle schools, resulting in two independent samples rather than paired samples. Since the two sets of questionnaires were independent, the test items were appropriately adjusted according to the respondents’ level of understanding. For instance, the item, “Study tours help me better adapt to society” in the student questionnaire was phrased as “I think study tours help children better adapt to society” in the parent questionnaire. Before the formal distribution of the questionnaires, several experts compared and reviewed the items in both questionnaires to ensure the core content of the measurement indicators was consistent, thereby ensuring the content validity of both sets of questionnaires.

The determination of the sample size took into account expected effect size, confidence level, margin of error, and anticipated response rate, ensuring that the study has sufficient statistical power. By covering the cities of Shenyang, Anshan, Harbin, and Dalian, the study ensured the geographic diversity of the sample, enhancing the generalizability of the findings. Furthermore, the collected 312 student questionnaires and 300 parent questionnaires not only met the requirements for structural equation modeling and other complex data analyses, but also provided an ample data set to support subsequent statistical analyses, thereby ensuring the scientific rigor and reliability of the research results. In summary, the selected sample size meets the research requirements in terms of representativeness and analytical validity. In this study, to ensure that middle school students and their guardians were fully informed and voluntarily participated, we provided guardians with detailed information outlining the research objectives and participants’ rights. The questionnaire included a participation confirmation option at the beginning, allowing students to actively express their willingness to participate and informing them that participation was voluntary and could be withdrawn at any time. We also promised to strictly maintain the confidentiality of participants’ personal information, using it solely for research purposes. These measures ensured that the study complied with ethical review requirements. Subsequently, the study employed SPSS 26 software to conduct common method bias tests and descriptive statistical analyses of the questionnaire data. SAS 9.4 software was used to perform multilevel statistical analyses on the filtered data to examine whether there were significant differences in the responses among the four regions of Shenyang, Anshan, Harbin, and Dalian. Using “students’/parents’ intention for forest educational tourism consumption” as the outcome variable, a one-way random effects ANOVA was conducted, revealing intra-class correlation coefficients of 0.039 and 0.015, with no significant between-group heterogeneity. This indicates that there is no significant relationship between students’/parents’ intentions toward forest educational tourism consumption and the different survey regions; therefore, the regional differences can be disregarded in terms of their impact on the consumption intention of dual decision-making agents. Consequently, the data from the four regions can be utilized for subsequent analyses.

3.3. Structural Equation Modeling

Structural equation modeling (SEM) is a multivariate statistical analysis method that can simultaneously estimate the direct and indirect relationships among observed variables while allowing for the consideration of latent variables behind the observed variables [47]. This study applies structural equation modeling to analyze the data obtained from the questionnaire survey, aiming to examine the model of dual decision-making agents in forest educational tourism. The sample size in this study meets the requirements for data analysis using AMOS. The Chi-Square to Degrees of Freedom Ratio (χ²/df), Comparative Fit Index (CFI), Parsimony Normed Fit Index (PNFI), and Root Mean Square Error of Approximation (RMSEA) are commonly used indices for evaluating model fit in structural equation modeling. The χ²/df ratio assesses the trade-off between model fit and complexity, with lower values indicating better fit. The CFI compares the fit of the proposed model to a baseline model (with no relationships among variables), with values closer to 1 indicating superior fit. The PNFI takes model parsimony into account, with values above 0.50 generally indicating a good fit, suggesting that the model effectively balances complexity and explanatory power. The RMSEA evaluates the discrepancy per degree of freedom, with values below 0.08 indicating acceptable fit. Together, these indices offer a comprehensive assessment of the model’s validity and fit, ensuring it appropriately reflects the underlying data relationships.

4. Results Analysis

4.1. Demographic Characteristics of Respondents and Information on Educational Tourism Consumption

Males accounted for 45.8% of the student sample, while females accounted for 54.2%. In terms of grade level, students in the first and second years of junior middle school represented a significant majority (87.9%). Regarding past experiences with forest educational tourism, 32.4% of students had no participation history, 29.2% had participated 1–2 times, 31.1% had participated 3–5 times, and 7.3% had participated 6 times or more. Considering that demographic characteristics such as gender and past experiences with forest educational tourism may influence consumption psychology, this study describes their distribution to aid in interpreting the findings related to the consumption psychological mechanisms of the student group. The grade distribution of the student sample aligns with the target market for educational tourism, indicating that the respondents in this study are representative. The conclusions drawn can provide valuable insights for the sustainable development of the forest educational tourism industry.

In the parent sample, in terms of education level, 40.7% of respondents had completed junior high school or below, 47.7% had completed high school or vocational college, and 11.6% had obtained a bachelor’s degree or higher. Regarding family annual per capita disposable income, 36.3% of respondents reported earning less than 20,000 RMB, 38% earn between 20,000 RMB and 50,000 RMB, 22.7% earn between 50,000 RMB and 100,000 RMB, and 3% earn above 100,000 RMB. In terms of past purchasing frequency of forest educational tourism products, 39.7% of parents have no purchasing experience, 33% have purchased such products 1–2 times, 20.7% have purchased such products 3–5 times, and 6.6% have purchased such products 6 times or more. Considering that demographic characteristics such as education level, household disposable income, and past experiences in purchasing forest educational tourism products may influence consumption psychology, this study describes their distribution to aid in interpreting the findings related to the consumption psychological mechanisms of the parent group.

4.2. Reliability and Validity Analysis of the Scales

This study uses SPSS 26 and AMOS 28 software to conduct reliability and validity tests for each group of latent variables in the student and parent questionnaires. As shown in Table 1, the standardized factor loadings of the variables in the student questionnaire range from 0.519 to 0.960, all exceeding the threshold of 0.400. The composite reliability (CR) of the latent variables ranges from 0.817 to 0.890, with all variables surpassing the acceptable level of 0.700. According to the criteria set by Hair et al. (2019) [55], the composite reliability of the latent variables is satisfactory. The average variance extracted (AVE) is an indicator that measures the explanatory power of a latent variable over its observed variables, with a value greater than 0.360 considered acceptable and a value above 0.500 considered ideal. The average variance extracted (AVE) values of the latent variables range from 0.521 to 0.669, all meeting the threshold of 0.500 for discriminant validity, thus indicating satisfactory convergent validity of the latent variables and an overall high model quality.

As shown in Table 2, in the parent questionnaire, the standardized factor loadings of the variables range from 0.522 to 0.945, all exceeding 0.400. The composite reliability of the latent variables ranges from 0.797 to 0.923, all exceeding the threshold of 0.700. According to the criteria of Hair et al. (2019) [55], the composite reliability of the latent variables is considered satisfactory. With the exception of the perceived behavioral control latent variable, the average variance extracted (AVE) values of the latent variables in the parent questionnaire range from 0.607 to 0.750, all meeting the discriminant validity threshold of 0.500. This study does not delete the measurement items for the perceived behavioral control latent variable, due to the following reasons. First, this study aims to conduct a comparative analysis between the student and parent groups, so it is necessary to maintain consistency in the measurement of latent variables. Second, deleting the measurement indicators related to time feasibility and economic conditions, which are highly regarded by parents in forest educational tourism decision-making, would result in a measurement deficiency in the latent variable’s content. Third, Fornell notes that AVE is a conservative indicator; even if AVE is below 0.5 but the composite reliability (CR) exceeds 0.7, the convergent validity can still be considered acceptable [56]. The AVE value for the perceived behavioral control latent variable is 0.441, which does not meet the ideal threshold of greater than 0.500 but remains within the acceptable range of above 0.360. Since the composite reliability of perceived behavioral control is greater than 0.7, it is deemed acceptable [57]. Overall, the convergent validity of the latent variables is satisfactory, and the model’s quality is high.

4.3. Path Analysis and Mediation Effect Test

4.3.1. Path Analysis

After conducting reliability and validity analyses on the scales, this study utilizes AMOS 28 software and employs the maximum likelihood estimation method to assess the fit of the hypothesized models. The results show that the fit indices for the student model are χ²/df = 2.748, CFI = 0.897, PNFI = 0.742, RMSEA = 0.075, while the fit indices for the parent model are χ²/df = 2.563, CFI = 0.923, PNFI = 0.773, RMSEA = 0.072. These indicate that both the student model and the parent model meet the standard values for fit indices, demonstrating good model compatibility.

Based on

Table 3. Passage inspection.

Effect Path	Student Model				Parent Model
	Path Coefficient	Standard Error	p	Conclusion	Path Coefficient	Standard Error	p	Conclusion
H1: PV → SN	0.469	0.069	0.001	Support	0.510	0.057	0.001	Support
H2: PV → BA	0.540	0.059	0.001	Support	0.423	0.058	0.001	Support
H3: PV → PBC	0.680	0.055	0.001	Support	0.784	0.050	0.002	Support
H4: PV → CI	0.388	0.109	0.001	Support	0.320	0.144	0.018	Support
H5: PR → SN	0.040	0.066	0.528	Reject	−0.013	0.063	0.799	Reject
H6: PR → BA	−0.041	0.063	0.487	Reject	0.091	0.057	0.109	Reject
H7: PR → PBC	−0.108	0.056	0.040	Support	0.009	0.069	0.941	Reject
H8: PR → CI	−0.002	0.056	0.936	Reject	−0.073	0.055	0.237	Reject
H9: SN → CI	0.201	0.071	0.002	Support	0.229	0.076	0.002	Support
H10: BA → CI	−0.059	0.069	0.354	Reject	0.011	0.059	0.824	Reject
H11: PBC → CI	0.187	0.097	0.061	Support	0.218	0.157	0.179	Reject

Note: PV: perceived value, PR: perceived risk, SN: subjective norms, BA: behavioral attitude, PBC: perceived behavioral control, CI: consumption intention.

and Figure 2 and Figure 3, in the student model, hypotheses H1–H4 are all supported at the 1% significance level, indicating that the perceived value of forest educational tourism significantly positively influences students’ subjective norms (β = 0.469 ***), behavior attitude (β = 0.540 ***), perceived behavior control (β = 0.680 ***), and consumption intention (β = 0.388 ***). Secondly, among hypotheses H5–H8, only H7 is supported, as the perceived risk of forest educational tourism has a significant negative impact on perceived behavior control (β = −0.108 **). The other hypotheses (H5, H6, and H8) are not supported by the sample data. Furthermore, hypotheses H9 and H11 are validated at the 1% and 10% significance levels, respectively, indicating that subjective norms (β = 0.201 ***) and perceived behavior control (β = 0.187 *) significantly positively influence students’ consumption intentions with regard to forest educational tourism, while hypothesis H10 is not supported by the sample data.

In the parent model, hypotheses H1–H4 are supported at the 1% and 5% significance levels, showing that the perceived value of forest educational tourism significantly positively influences parents’ subjective norms (β = 0.510 ***), behavior attitude (β = 0.423 ***), perceived behavior control (β = 0.784 ***), and consumption intention (β = 0.320 **). Additionally, hypotheses H5–H8 are not supported by the sample data. Furthermore, hypothesis H9 is validated at the 1% significance level, indicating that subjective norms (β = 0.229 ***) significantly positively influence parents’ consumption intentions with regard to forest educational tourism, while hypotheses H10 and H11 are not supported by the sample data.

4.3.2. Mediation Effect Test

This study utilized AMOS 28 software and employed the bias-corrected percentile Bootstrap method to test the mediating effects of subjective norms, behavioral attitude, and perceived behavioral control on the relationships between perceived value and perceived risk and consumption intention in the context of dual decision-making agents in forest educational tourism. The test results are presented in

Table 4. Standardized Bootstrap mediating effect test.

Effect Type	Path	Student Model					Parent Model
		Effect Value	Standard Error	Bias-Corrected 95% CI			Effect Value	Standard Error	Bias-Corrected 95% CI
				Lower	Upper	p			Lower	Upper	p
Total Effect	PV → CI	0.577	0.058	0.463	0.685	0.001	0.613	0.054	0.502	0.715	0.001
Direct Effect	PV → CI	0.388	0.109	0.181	0.625	0.001	0.320	0.144	0.052	0.620	0.018
Indirect Effect	PV → CI	0.189	0.082	0.014	0.344	0.033	0.293	0.124	0.044	0.534	0.020
Specific Indirect Effect	PV → SN → CI	0.094	0.030	0.020	0.137	0.002	0.117	0.027	0.027	0.137	0.002
	PV → BA → CI	−0.032	0.026	−0.079	0.029	0.326	0.005	0.015	−0.029	0.035	0.805
	PV → PBC → CI	0.127	0.049	0.003	0.201	0.045	0.171	0.079	−0.048	0.273	0.163
Total Effect	PR → CI	−0.011	0.054	−0.117	0.095	0.841	−0.073	0.063	−0.192	0.056	0.267
Direct Effect	PR → CI	−0.002	0.056	−0.117	0.104	0.936	−0.073	0.055	−0.169	0.046	0.237
Indirect Effect	PR → CI	−0.010	0.022	−0.062	0.027	0.550	0.000	0.027	−0.052	0.057	0.957
Specific Indirect Effect	PR → SN → CI	0.008	0.008	−0.008	0.025	0.462	−0.003	0.011	−0.028	0.016	0.743
	PR → BA → CI	0.002	0.003	−0.002	0.013	0.321	0.001	0.004	−0.006	0.011	0.649
	PR → PBC → CI	−0.020	0.009	−0.038	0.000	0.048	0.002	0.013	−0.022	0.034	0.820

.

In the student model, the perceived value of forest educational tourism by the student group has a significantly positive total effect, direct effect, and total indirect effect on consumption intention, with effect values of 0.577 ***, 0.388 ***, and 0.189 **, respectively. Among the three specific indirect effects, the mediating effect of behavior attitude on the relationship between perceived value and consumption intention does not reach significance, while subjective norms and perceived behavior control play important partial mediating roles in the relationship between perceived value and consumption intention for the student group. From the perspective of effect values, the mediating effect of perceived behavior control (0.127 **) on the relationship between perceived value and consumption intention is greater than that of subjective norms (0.094 ***). The perceived risk of forest educational tourism by the student group has not reached significance in terms of total effect, direct effect, and total indirect effect on consumption intention. Among the three specific indirect effects, the mediating effects of subjective norms and behavior attitude on the relationship between perceived risk and consumption intention do not reach significance, while perceived behavior control plays an important complete mediating role in the relationship between perceived risk and consumption intention for the student group, with an effect value of −0.020 **.

In the parent model, the perceived value of forest educational tourism by the parent group has a significantly positive total effect, direct effect, and total indirect effect on consumption intention, with effect values of 0.613 ***, 0.320 **, and 0.293 **, respectively. Among the three specific indirect effects, the mediating effects of behavior attitude and perceived behavior control on the relationship between perceived value and consumption intention do not reach significance, while subjective norms play an important partial mediating role in the relationship between perceived value and consumption intention for the parent group, with an effect value of 0.117 ***. The perceived risk of forest educational tourism by the parent group has not reached significance in terms of total effect, direct effect, and total indirect effect on consumption intention. In the three specific indirect effects, the mediating effects of subjective norms, behavior attitude, and perceived behavior control between perceived risk and consumption intention for the parent group also do not achieve significance.

4.4. Multi-Group Structural Equation Model

The student and parent samples are divided into equal and unequal groups based on the level of family decision-making empowerment, and multi-group structural equation modeling analysis is conducted using AMOS 28 software.

Based on the student sample, the fit indices for the multi-group model are as follows: the χ²/df value ranges from 2.002 to 2.067, the CFI value ranges from 0.865 to 0.882, the PNFI value ranges from 0.695 to 0.751, and the RMSEA value ranges from 0.057 to 0.059. These parameters indicate that the baseline model fits well with the two groups of data, and the model structure is consistent between the groups. Through the test of structural consistency, multi-group analysis is applicable.

Table 5. Comparison of estimation results of multi-group structure model based on family decision-making empowerment.

Path	Student Model			Parent Model
	Equality	Inequality	Critical Ratio	Equality	Inequality	Critical Ratio
H1: PV → SN	0.332 ***	0.604 ***	1.914	0.476 ***	0.541 ***	−0.156
H2: PV → BA	0.389 ***	0.713 ***	1.174	0.398 ***	0.453 ***	−0.635
H3: PV → PBC	0.701 ***	0.651 ***	0.500	0.832 ***	0.756 ***	−1.594
H4: PV → CI	0.368 ***	0.308	0.024	0.096	0.421 **	1.361
H5: PR → SN	0.112	−0.038	−1.297	0.056	−0.061	−0.999
H6: PR → BA	−0.043	−0.041	0.084	0.098	0.096	−0.338
H7: PR → PBC	−0.019	−0.214 **	−2.031	0.002	0.016	0.115
H8: PR → CI	−0.069	0.077	1.364	−0.144 *	−0.043	1.243
H9: SN → CI	0.145 *	0.312 ***	1.542	0.344 ***	0.143	−1.603
H10: BA → CI	−0.076	−0.009	0.297	0.057	−0.048	−0.952
H11: PBC → CI	0.280 **	0.110	−1.014	0.370 *	0.218	−0.591

Note: * indicates p < 0.1, ** indicates p < 0.05, *** indicates p < 0.01, and no asterisk indicates no significant statistical significance.

displays the path coefficients for different levels of family decision-making empowerment. The findings show several similarities: perceived value significantly positively influences subjective norms, behavioral attitudes, and perceived behavioral control across both family decision-making empowerment groups. In contrast, perceived risk does not significantly affect subjective norms, behavioral attitudes, or consumption intentions in either group. Furthermore, subjective norms exhibit a significant positive effect on consumption intentions across both groups, while behavioral attitudes do not significantly influence consumption intentions in either group. These conclusions align with the results of the overall model analysis. The differences in the multi-group model are as follows: perceived value and perceived behavioral control have a significant positive impact on consumption intentions in the equal group, consistent with the overall model results. However, this effect is not significant in the unequal group. Conversely, perceived risk has a significant negative impact on perceived behavioral control in the unequal group, which is also consistent with the overall model results, while no significant effect is observed in the equal group.

Based on the parent sample, the fit indices for the multi-group model are as follows: the χ²/df value ranges from 1.933 to 2.002, the CFI value ranges from 0.899 to 0.905, the PNFI value ranges from 0.728 to 0.794, and the RMSEA value ranges from 0.056 to 0.058. These parameters indicate that the baseline model fits well with the two groups of data, and the model structure is consistent between the groups. Through the test of structural consistency, multi-group analysis is applicable. Table 5 displays the path coefficients for different levels of family decision-making empowerment. The findings show several similarities: perceived value significantly positively influences subjective norms, behavioral attitudes, and perceived behavioral control across both family decision-making empowerment groups. In contrast, perceived risk does not significantly affect subjective norms, behavioral attitudes, or perceived behavioral control in either group. Additionally, behavioral attitudes do not significantly influence consumption intentions in either group, which aligns with the results of the overall model analysis. The differences are as follows: perceived value has a significant positive impact on consumption intentions in the unequal group, consistent with the overall model results, but this effect is not significant in the equal group. Conversely, subjective norms have a significant positive effect on consumption intentions in the equal group, which is consistent with the overall model results, while no significant effect is observed in the unequal group. Notably, perceived risk significantly negatively affects consumption intentions in the equal group, and perceived behavioral control significantly positively influences consumption intentions in the equal group, representing new findings that differ from the overall model analysis results.

5. Discussion

(1)

The consumption psychological mechanisms of dual decision-making agents in forest educational tourism exhibit differences. The perceived value of forest educational tourism products positively influences subjective norms, behavioral attitude, perceived behavioral control, and consumption intention among dual decision-making agents, a finding that aligns with the conclusions of Diddi et al. (2019) [36]. Additionally, perceived risk only affects the student group’s cognition regarding their resources and ability to participate in forest educational tourism activities, but does not significantly influence the final consumption intention of dual decision-making agents or their attitudes towards forest educational tourism products. While this conclusion challenges the prevailing notion that perceived risk has a negative effect in many studies, it somewhat aligns with the findings of Wen et al. (2022) [17]. This may be attributed to students’ and parents’ motivation to develop resilience and the ability to withstand challenges through participation in forest educational tourism. Furthermore, the participation of relatives, friends, and others in forest educational tourism consumption enhances the consumption intention of both the student group and the parent group. Additionally, the perception of resources and abilities that students believe they possess for participating in forest educational tourism consumption also reinforces their consumption intention, which aligns with the conclusions of Hultman et al. (2015) [58]. In contrast, the impact of parents’ perception of their resources and abilities on consumption intention is not significant. This may be attributed to the parental role, as parents, being the purchasers of forest educational tourism, do not partake in the experiential process of the activities. In most cases, parents recognize the benefits and cost-effectiveness of forest educational tourism activities and are more likely to overcome potential difficulties in purchasing these products to promote their children’s physical and mental development [59]. Additionally, the interest level of both students and parents in the courses and programs of forest educational tourism does not significantly affect consumption intention. This may be because the behavioral attitude latent variable only reflects individuals’ subjective emotions towards specific behaviors and fails to adequately convey the impact of individuals’ perceptions of the usefulness and ease of specific behaviors on their behavioral intentions [60]. Therefore, behavioral attitude does not accurately predict the consumption intention of students and parents regarding the purchase of forest educational tourism products.

(2)

Our study reveals the mediating roles of subjective norms and perceived behavioral control, enriching the internal mechanism of the consumption psychology formation of dual decision-making agents in forest educational tourism. This mechanism specifically manifests by influencing how the perceived value of participating in forest educational tourism prompts dual decision-making agents to be more influenced by the participation of relatives, friends, and others, thereby strengthening their willingness to participate in and recommend forest educational tourism products. Additionally, the student group’s perception of their resources and capabilities for engaging in forest educational tourism activities reinforces the impact of perceived value on their willingness to participate in and recommend these activities, while also somewhat weakening the influence of perceived risk on their willingness to engage in and recommend forest educational tourism activities.

(3)

The level of family decision-making empowerment affects the consumption psychology of dual decision-making agents in forest educational tourism. Through the analysis of the sample data, among the student group perceiving unequal family decision-making empowerment, a relatively high proportion of students are empowered to make consumption decisions. As a result, when making travel consumption decisions, they may focus more on the potential frustrations and uncertainties they might face while participating in forest educational tourism, as well as consider their own resources and capabilities related to participation, along with the benefits the activities may provide. In contrast, these situational factors have less significant influence on students in the group perceiving equal family decision-making empowerment. In the parent group perceiving unequal family decision-making empowerment, a relatively high proportion of parents empower students to make consumption decisions. Therefore, when making consumption decisions, parents, who are funding the endeavor, primarily consider the benefits and cost-effectiveness of the expenditures related to forest educational tourism. On the other hand, in the parent group perceiving equal family decision-making empowerment, parents have relatively greater authority and thus tend to consider factors from the perspective of consumption experience, including social circle consumption behaviors, activity risks, and students’ coping abilities.

6. Conclusions and Limitations

6.1. Conclusions

This study expands the Theory of Planned Behavior by introducing perceived value and perceived risk variables within the context of forest educational tourism consumption, analyzing the formation mechanism and heterogeneity of the consumption psychology of dual decision-making agents from the perspectives of students and parents, and examining the moderating role of family decision-making empowerment. The conclusions are as follows:

(1)

There is heterogeneity in the formation of the consumption psychology mechanism of dual decision-making agents in forest educational tourism; perceived value significantly positively influences the subjective norms, behavioral attitude, perceived behavioral control, and consumption intention of these agents. While perceived risk has a significant negative impact on perceived behavioral control among the student group, its negative influence on the consumption intentions of dual decision-making agents has not been validated. Subjective norms positively influence the consumption intention of the dual decision-making agents, and perceived behavioral control positively impacts the consumption intention of the student group, whereas behavioral attitude does not accurately predict the consumption intention of the dual decision-making agents.

(2)

In the relationship between the perceived value of forest educational tourism of dual decision-making agents and their consumption intention, subjective norms play a partial mediating role, while the mediating effect of behavior attitude is not significant. The partial mediating effect of perceived behavior control is validated only in the student model. In the relationship between the perceived risk of forest educational tourism and consumption intention, only the complete mediating effect of perceived behavior control is validated in the student model, while the other mediating paths are not supported.

(3)

Family decision-making empowerment has a moderating effect on certain paths of influence in both the student and parent models. In the student model, perceived value and perceived behavioral control significantly positively influence consumption intentions in the equal group, but this effect is not significant in the unequal group. Conversely, perceived risk significantly negatively impacts perceived behavioral control in the unequal group, while this effect is not significant in the equal group. In the parent model, perceived value significantly positively influences consumption intentions in the unequal group, but this effect is not significant in the equal group. Additionally, subjective norms significantly positively influence consumption intentions in the equal group, while this effect is not significant in the unequal group. Furthermore, perceived risk has a significant negative impact on consumption intentions in the equal group, and perceived behavioral control significantly positively influences consumption intentions in the equal group, representing new findings that differ from the overall model analysis.

6.2. Theoretical Implications

First, this study incorporates the variables of perceived value and perceived risk into the Theory of Planned Behavior (TPB) model, expanding and refining the traditional TPB model. This comprehensive TPB model helps to deepen our understanding of the formation mechanisms of tourism consumers’ psychological behaviors and effectively enhances the applicability and explanatory power of the TPB model in tourism consumption psychology research.

Second, this study explores the heterogeneity of the consumption-related psychological mechanisms of dual decision-making agents in forest educational tourism for the first time. Given that educational tourism features a separation between purchasing payment and consumption experience, it is essential to thoroughly analyze the similarities and differences in the consumption decision characteristics of dual decision-making agents in educational tourism products. This study responds to that need by focusing on the consumption context of forest educational tourism, thus broadening the research perspectives on educational tourism consumption decisions and their influencing factors.

Third, this study investigates the boundary conditions affecting the consumption psychological mechanisms of dual decision-making agents in forest educational tourism. The specific consumption decision contexts may exhibit different mechanisms of influence on consumption psychology. By considering the separation of roles in purchasing payment and consumption experience within educational tourism consumption decisions, this study introduces the moderating variable of family decision-making empowerment. The findings reveal that there are differences in the consumption psychological mechanisms of dual decision-making agents under varying levels of family decision-making empowerment. This research identifies important variables influencing educational tourism consumption decisions and serves as a significant supplement to related studies on the influencing factors of consumption psychology in educational tourism, while also expanding the research perspectives within the field of dual decision-making agent consumption studies.

6.3. Practical Implications

The consumption decision-making of forest educational tourism has dual agent characteristics. To identify relevant scenic areas and enterprises which will allow us to refine the market positioning of educational tourism products and develop scientific and precise marketing strategies, it is essential to clarify the differences in consumption focus and decision-making behavior characteristics of the dual decision-making agents in forest educational tourism. This study provides the following insights for targeted marketing of forest educational tourism destinations and other stakeholders:

(1)

The first step should be to enhance perceived value. Perceived value has a significant positive impact on the subjective norms, behavioral attitude, perceived behavioral control, and consumption intention of dual decision-making agents. In response, forest educational tourism scenic areas should strive to enhance the perceived value of forest educational tourism products among parents and students. On one hand, scenic areas should focus on promoting the learning value of forest educational tourism products. When marketing through advertisements and social media, they should convey information about how forest educational tourism products benefit children’s learning and development. On the other hand, scenic areas should emphasize the social value of forest educational tourism products, with marketing efforts highlighting how these products help children better adapt to society and how they enhance social interactions.

(2)

The second step should be to manage activity risks. Aside from having a significant negative impact on perceived behavioral control among the student group, perceived risk does not have a significant effect on the subjective norms, behavioral attitude, and consumption intention of dual decision-making agents. This requires scenic areas to adopt a dialectical view of the roles that parents and students play regarding perceived risk in the formation of their consumption psychology concerning forest educational tourism products. On one hand, relevant enterprises should consider the motivations of parents and students who hope to gain experience and education from forest educational tourism activities when designing these products, ensuring that the products are challenging. On the other hand, enterprises should also implement safety controls to address potential physiological risks, such as falls and acclimatization issues, as well as economic risks like loss or damage to belongings, in order to alleviate student’s concerns about participating in forest educational tourism.

(3)

The third step should be to emphasize word-of-mouth communication and marketing information design. Subjective norms positively influence the consumption intention of dual decision-making agents, while perceived behavioral control only positively affects the consumption intention of the student group. In this regard, forest educational tourism scenic areas should focus on peer influence when marketing their products. They can utilize social groups and new media channels to promote word-of-mouth marketing, stimulating the participation intention of decision-making agents. Additionally, scenic areas should design messages during the marketing process that enhance the self-efficacy of the student group, helping them build confidence in successfully completing the forest educational tourism experience.

(4)

The fourth step should be to identify family decision-makers and design persuasive strategies accordingly. The level of family decision-making empowerment influences the consumption psychology of dual decision-makers in forest educational tourism. Therefore, marketers need to consider both parental power and children’s power in the family consumption decision-making process when promoting forest educational tourism products. This approach should broaden the marketing channels for these products, ensuring that relevant information effectively reaches both parents and students. Additionally, marketers should strive to identify the family decision-makers and design targeted persuasive strategies for marketing forest educational tourism products.

6.4. Limitations

This study has certain limitations. First, it focuses on the consumption intentions of dual decision-making agents in forest educational tourism, and due to the level of industry development and research difficulty, it does not explore the actual consumption behaviors of these agents. Future research could utilize experimental methods to measure the true consumption behaviors and their formation mechanisms through actual promotions and the organization of forest educational tourism activities. Second, due to space constraints, this study does not delve into other pathways and boundary conditions of the consumer psychological influence mechanism of dual decision-making agents in forest educational tourism. Future studies could explore the impacts of factors such as motivation, trust, and educational experiences on the formation of consumption psychology among these agents.