A Study on Priority of Classification System for Agro-healing Using the Analytic Hierarchy Process

Hierarchical Criteria Composition and Evaluation Scale

In this study, the criteria for the classification system were established based on the opinions of experts in related fields. These criteria help to determine the importance of, and prioritize the establishment of, the classification system for Korean agro-healing, referring to the Act on Research, Development and promotion of healing agriculture (2021), Study on methodology and evaluation system of care farming (RDA, 2017), and Korean Standard Industrial Classification (KSIC) (Statistics Korea, 2020). It also involves a value chain system that includes a series of activity processes related to the creation of added value (Porter, 1985). The main criteria consist of three elements including agro-healing input industry, agro-healing service industry, and agro-healing-related (derived) industry; the details of each element are shown in Table 1. Further, there are also 11 sub-criteria, as shown in Table 2. When looking at the sub-criteria for each main criterion, the “agro-healing input industry” is composed of the agro-healing construction industry, agro-healing input manufacturing, crop cultivation and breeding, and the manufacturing of agri-food and related processed goods. “Agro-healing service industry” consists of treatment and rehabilitation, social rehabilitation, and community service. Finally, “agro-healing-related/derived industry” is comprised of training and education institutions for agro-healing experts, research and development (R&D) and consulting services related to agro-healing, industrial and expert organizations related to agro-healing, and other agro-healing support services.

Table 1

Descriptions of the main criteria

Table 2

Descriptions of the sub-criteria

The evaluation scale used for pairwise comparisons is a 9-point scale (Table 3) that includes the reciprocal value proposed by Saaty (1988), which is based on the idea that an average human cannot compare more than 7 objects at the same time (7 ± 2), as per Miller's (1956) “Stimulus-Response” theory. This scale requires n(n−1)/2 comparisons if the immediate lower level consists of n elements (Chung and Yoon, 2008). Pairwise comparisons are performed by assigning scores to the semantic expression (response options) in a survey questionnaire, and the respondent assigns scores to items on the scale that are regarded as relatively more important among two comparators (Lim and Lee, 2019).

Table 3

Pairwise comparison 9-point scale for Analytic Hierarchy Process preferences

Validation Criteria

Validation of the AHP analysis can be confirmed through a consistency ratio (CR), which is a value obtained by dividing the consistency index (CI) by the average of the random index (RI; Vaidya and Kumar, 2004). In general, the smaller the CR value, the greater the consistency of the judgment (decision making); with respect to the criterion, if the value is less than 0.2 (20%), then it may be that the respondent makes judgments with some consistency (Lee, 2007; Park, 2012). If the CR value is 0.2 or greater, it is regarded as inconsistent (Vaidya and Kumar, 2004) and re-survey is recommended.

Three or four experts participated in the existing AHP analysis. However, a limitation was raised as far as the overall decision-making process may be distorted by motivational bias due to the small number of participants. Thus, KDI (2008) established guidelines for deriving the results of the AHP analysis by increasing the number of experts to 7 or 8. As such, the AHP analysis is primarily used in decision-making processes that require expertise. Unlike general survey methods, this approach helps to obtain significant decision-making results with a small sample. Therefore, it is important to consider which expert group in the relevant field is selected as a participant, and how faithfully and consistently the participants responded (Saaty, 2008; Kim et al, 2007).

Subjects and Data Collection

From October 21 to October 30, 2021, a survey was administered to 79 students and faculty at 6 educational institutions that foster agro-healing specialists. These respondents comprised a group of experts with at least 5 years of experience in agro-healing and related industries, and who were currently operating healing farms (rural or urban) or working at agro-healing education institutions including hospitals, clinics, and private medical institutions; private institutions such as welfare centers; horticultural therapy associations; and universities.

As a result of measuring the individual CR for 79 response sheets, it was found that the CR of 61 response sheets exceeded 0.2, rendering them unsuitable for analysis. According to Hong (2011), the AHP technique is performed using pairwise comparisons of survey responses, which are respondents' subjective judgments. When comparing the relative weights between evaluated factors, if a pair of items is not evaluated as A > C and is instead evaluated as C > A when A > B and B < C in the degree of relative weights, the response lacks consistency, which indicates a lack of reliability. Therefore, these comparisons are related to the quality of the evaluation. In this study, when considering these validation criteria, 18 responses with CR < 0.2 (Saaty, 1990) were selected for the final analysis, as they were reflective of the fact that the evaluation was performed with an acceptable level of consistency.

Analysis Method

An analysis of demographic data was performed using SPSS 25.0 (IBM, USA) for the final selected responses. Then, the AHP analysis to establish the classification system for agro-healing was performed using DRESS 1.3 (CHOISH, KOREA), an AHP analysis software package that is capable of analyzing the importance (weights) and priority level of the hierarchical structure of this study (Fig. 1). Since AHP analysis is a technique that can be used to select an optimal alternative by classifying multiple attributes hierarchically and performing pairwise comparisons on the importance of each attribute, it is used as a decision-making tool that enables rational choices of respondents by presenting complex problems in a simple and systematic manner (Lee, 2011; Saaty, 1990).

Fig. 1

Hierarchical structure of the classification system.

General Characteristics of Survey Respondents

The demographic characteristics of the respondents are presented in Table 4. There were 8 males (44.4%) and 10 females (55.6%); respondents were most commonly in their 50s (n = 13; 72.7%) or 40s (n = 3; 16.7%), and there was 1 person each (5.6%) in their 30s and in their 60s. With respect to respondents’ careers in agro-healing and related industries, most (n = 11; 61.1%) had 5 to 10 years of experience, while some (n = 4 ; 22.2%) had 15 or more years of experience, and a few (n = 3; 16.7%) had 10 to 15 years of experience. In terms of their current workplace, respondents were able to provide multiple responses; participants most commonly worked in educational institutions such as universities (33.3%), followed by private institutions such as horticultural treatment associations (14.8%) and welfare centers (11.1%); urban agro-healing farms (11.1%); plant production (11.1%); rural agro-healing farms (7.4%); hospitals, clinics, and private medical institutions (3.7%); and as agro-healing activists (3.7%).

Table 4

Characteristic of survey respondents

AHP Analysis

The results from the analysis of the relative weights and priority levels of the main criteria and sub-criteria for the agro-healing classification system are shown in Table 5. With respect to the main criteria, the agro-healing service industry (0.52) showed the highest weight, followed by the agro-healing related (derived) industry (0.32) and agro-healing input industry (0.16). The CR for the main criteria was found to be 0.00, indicating that respondents provided consistent responses, as identified during the pairwise comparison.

Table 5

Pairwise comparison matrix for the criteria using AHP.

The weights and priority levels for the sub-criteria are as follows: In the agro-healing input industry, the CR for the sub-criteria was found to be 0.01, while crop cultivation and breeding (0.42) showed the highest weight, followed by the manufacturing of agri-food and related processed goods (0.32), the agro-healing construction industry (0.13), and agro-healing input manufacturing (0.13). In the agro-healing service industry, the CR for the sub-criteria was 0.00, and the weight was the highest for community service (0.47), followed by social rehabilitation (0.30), and treatment and rehabilitation (0.23). In the agro-healing-related/derived industry, the CR for the sub-criteria was 0.00, indicating response consistency. Other agro-healing support services (0.33) had the highest weight, followed by training and education institutions for agro-healing experts (0.28), R&D and consulting related to agro-healing (0.20), and industrial and expert organizations related to agro-healing (0.18).

Complex weights were calculated based on the analysis results to determine the priority level of all evaluation items for establishing the agro-healing classification system (Table 6). These weights were obtained using a general method that is often used to calculate relative weights, in which the weight of each evaluation item of the main criteria was calculated using the AHP (pairwise comparison), and the value was multiplied by the weight of each evaluation item of the sub-criteria (Lee at al., 2017).

Table 6

The AHP criteria ranking and their complex weights.

Of the 11 sub-criteria, community service (0.2411) had the highest complex weight, followed by social rehabilitation (0.1574), treatment and rehabilitation (0.1197), other agro-healing support services (0.1064), training and education institutions for agro-healing experts (0.0901), crop cultivation and breeding (0.0675), R&D and consulting related to agro-healing (0.0650), industrial and expert organizations related to agro-healing (0.0582), and manufacturing of agri-food and related processed goods (0.0526). The agro-healing construction industry (0.0210) and agro-healing input manufacturing (0.0210) criteria were found to have the lowest complex weights.

When examining the above analysis findings, which were grouped based on sub-criteria per main criterion, all three items (community service, which featured the highest complex weight; social rehabilitation, which ranked 2nd; and treatment and rehabilitation, which ranked 3rd) fell within "agro-healing service industry" as the main criterion. Jeong et al. (2020) reported that anyone can become a target of agro-healing, even if they do not have specific physical or emotional problems, and that there should be widespread awareness that anyone can experience agro-healing activities. This is in line with the fact that community service showed the highest complex weight among the 11 sub-criteria. Community services — providing care services for nursery schools, daycare centers, and the elderly with dementia, as well as education-related services for the general public, such as lifelong education and meditation shelters — represent the most important factor to be considered when creating an agro-healing classification system. In addition, the identification of social rehabilitation, and treatment and rehabilitation as high-priority factors suggests that the agro-healing system and its associated activities should be established and developed with a focus on the importance and role of agro-healing activities (RDA, 2016). This is important, as these activities can focus on healing those who are under work-related stress or poor health, as well as those in need of medical and social treatment (such as those living with mental illness, depression, learning disabilities, substance misuse, and those with social dissatisfaction).

For complex weights of sub-criteria that fall under the main criterion of "agro-healing-related/derived industry," other agro-healing support services ranked 4th, training and education institutions for agro-healing experts ranked 5th, R&D and consulting related to agro-healing ranked 7th, and industrial and expert organizations related to agro-healing ranked 8th; indicating an intermediate level of weights among the three main criteria. This finding is considered to reflect experts' opinions on the need for a classification system for the service industry, which is in charge of public relations to establish and activate agro-healing, as well as agro-healing related/derived industry as support activities that foster the development of agro-healing experts through the provision of training and education institutions. As interest in agro-healing farms and agro-healing increased after the Act on Research, Development and Promotion of Healing Agriculture came into effect, the need for both information on agro-healing activities at agro-healing farms and research on its effects has frequently been raised (Jang et al., 2020). In this context, the segmentation and organic linkage of agro-healing-related support services, expert training and education institutions, and R&D and consulting are required for agro-healing activity support and in-depth research to confirm its effects.

For the sub-criteria of "agro-healing input industry", crop cultivation and breeding ranked 6th, the manufacturing of agri-food and related processed goods ranked 9th, and the agro-healing construction industry and agro-healing input manufacturing tied for 10th, showing the lowest weights. This means that research exploring agro-healing construction industry and agro-healing input manufacturing awareness and systems should be promoted. Crop cultivation and breeding was found to be the most important factor among the four sub-criteria for the agro-healing input industry. This finding may have resulted from the fact that horticultural crops and livestock-raising activities provided by agro-healing farms are directly related to agro-healing services, which is an essential element underlying all agro-healing activities.