Preference for Components of a Horticultural Therapy Program for Alleviating Job Stress Among Secondary School Teachers
Article information
Abstract
Background and objective
This study investigated the program components preferred by secondary school teachers to design effective horticultural therapy programs to relieve their job stress.
Methods
An online survey was conducted from April 8 to 28, 2025, targeting 307 middle and high school teachers. The questionnaire included 8 items on demographic characteristics, 4 items on emotional labor and job stress levels, and 10 items on preferred components of horticultural therapy programs.
Results
Job stress levels were measured on a scale from 0 (low) to 10 (high), with the majority of the respondents (57.8%) scoring 7 or higher. The most individual strategies for relieving job stress were exercise (31.9%) and travel (25.9%). Regarding preferred program components, most teachers favored programs consisting of a total number of 10 or fewer sessions (85%), held once per week (74.3%), with each session lasting 60 minutes or less (70.7%), and involving 10 or fewer participants per session (78.2%). Preferred times for program participation were after work (31.8%) and during lunch breaks (32.6%). Indoor settings were preferred by 42% of respondents as the location of program activity, and group therapy was the most preferred format (33.3%). Lastly, in terms of preferred activities, respondents most commonly selected gardening (34%), followed by craft (33.9%) and cooking activities (25.3%).
Conclusion
These findings are expected to serve as a foundational reference for the design, institutional adoption, and further research of horticultural therapy programs aimed at relieving job stress among secondary school teachers.
Introduction
Modern society is undergoing rapid transformation, characterized by an emphasis on new forms of knowledge and the pervasive use of information technology. These shifts are contributing to various mental and psychological stresses experienced by individuals as members of both society and organizations (Jung, 2006). While a moderate level of job-related stress is generally recognized as beneficial, helping employees achieve their performance goals (Kim and Uhm, 2014), excessive or prolonged stress can have detrimental effects. These include a decline in work quality, reduced creativity, and diminished engagement (Lee and Kim, 2010). The European Commission defines job stress as “a pattern of emotional, cognitive, behavioral, and physiological reactions to adverse and noxious aspects of work content, work organization, and the work environment; a state characterized by high levels of arousal and distress, and sometimes feelings of not coping” (Lee, 2011).
In this context, the educational environment is also becoming increasingly complex and rapidly evolving, leading teachers to experience various forms of job stress (Baek, 2002). Secondary school teachers, in particular, are subject to continuous psychological and emotional strain due to factors such as expectations for academic performance, student guidance responsibilities, interactions with students’ parents, and the burden of administrative tasks (Kim et al., 2016). This stress not only diminishes teachers’ job satisfaction but can also negatively affect the quality of education and teacher-student relationships (Ok, 2009). Consequently, it is necessary to explore effective strategies for alleviating and managing teachers’ job stress.
Horticultural therapy, a therapeutic approach utilizing plants, is known to have positive effects on emotional stability, stress relief, and improved quality of life (Kim, 2023). Numerous studies have demonstrated the psychological healing effects of horticultural activities (Lee and Choi, 2012), suggesting their potential to alleviate emotional burnout and job stress, particularly among teachers.
However, research on the effects of horticultural therapy for secondary school teachers is relatively limited, and studies on how teachers perceive horticultural therapy are also lacking. To develop effective horticultural therapy programs aimed at reducing job stress among teachers, a participant-centered and customized design is essential. This necessitates a preliminary survey to incorporate teachers’ perceptions and preferences.
Therefore, we surveyed secondary school teachers about their levels of emotional labor and job stress, their current stress management strategies, and their experiences with and willingness to participate in horticultural therapy programs. Based on these data, we aimed to identify optimal elements for designing and implementing a horticultural therapy program specifically tailored to alleviate teachers’ occupational stress.
Research Methods
The survey for this study was conducted with the cooperation of teachers from private middle and high schools located in Daegu Metropolitan City and Gyeongsangbuk-do Province. The participating schools included Hyosung Girls’ High School, Hyosung Middle School, Daegeon High School, Daegeon Middle School, Muhak High School, Muhak Middle School, Geunhwa Girls’ High School, Geunhwa Girls’ Middle School, Seongeui High School, and Seongeui Middle School. The survey was administered online from April 8 to April 28, 2025, and a total of 307 responses were collected. All responses were compiled, organized, and used for analysis.
The contents of the survey are summarized in Table 1. First, demographic and occupational characteristics were assessed through eight items, including gender, age, marital status, workplace, homeroom teacher status, job titles, employment status, and years of work experience. Second, based on the questionnaire developed by Back (2024), emotional labor and job stress were each measured using two items rated on a 10-point Likert scale. In addition, one item assessed individual stress management strategies, while three items explored participants’ perceptions of horticultural therapy programs, as well as their prior experience with and willingness to participate in such programs. Third, preferences for the components of horticultural therapy programs were assessed using eight items, based on the research findings of Ji (2009) and Kim (2006): implementation time, type of program delivery, total number of sessions, frequency per week, number of participants, location for activities, duration of each session/intervention, and type of activity.
The collected data were coded using Microsoft Excel and analyzed using SPSS (version 19.0; IBM, USA, 2012). The reliability of the items measuring emotional labor and job stress was confirmed with a Cronbach’s alpha of .84. Descriptive statistics, including frequencies and percentages, were used to determine individual preferences based on the collected responses. A chi-square goodness-of-fit test was conducted to determine the statistical significance of the observed distributions. Additionally, a chi-square test of independence was used to examine group differences across specific attributes.
Results and Discussion
Demographic characteristics of the respondents
A total of 307 respondents participated in the survey for this study. Among them, 139 were male (45.3%) and 168 were female (54.7%), indicating a slightly higher proportion of female respondents. The largest age group was individuals in their 40s (121 respondents, 39.4%), followed by those in their 30s (75 respondents, 24.4%). This distribution reflects the predominance of mid-career teachers in educational settings (MOE and KEDI, 2024). Regarding marital status, the majority of respondents were married (207 individuals, 67.4%), followed by unmarried individuals (97 respondents, 31.6%), and other (e.g., divorced, widowed; 3 respondents, 1.0%) (Table 2).
Occupational Characteristics of the Respondents
An analysis of the occupational characteristics of the respondents is shown in Table 3, which includes school level, homeroom teacher status, job titles, employment status, and years of work experience. Of the total, 131 respondents (42.7%) were employed in middle schools, while 176 (57.3%) worked in high schools, indicating a slightly higher proportion of high school teachers. In terms of homeroom teacher status, 140 respondents (46.0%) were homeroom teachers, whereas 167 (54.0%) were non-homeroom teachers. Regarding job titles, 144 respondents (46.9%) were classified as general teachers (including subject teachers, nutrition teachers, health teachers, and counselors), 82 (26.7%) as senior teachers (e.g., grade-level coordinators, academic affairs directors, and research directors), 16 (5.2%) as administrators (vice principals and principals), and 65 (21.2%) as other types of teachers (e.g., fixed-term/temporary or subject-specific teacher). This data indicates that general teachers constitute a significant proportion of the staff within the schools surveyed. With respect to employment status, 178 respondents (58.0%) were full-time teachers, 115 (37.5%) were fixed-term/temporary teachers, 9 (2.9%) were part-time instructors, and 5 (2.0%) were classified as other (e.g., dispatched teachers or those employed directly by the Office of Education). Full-time teachers made up the largest portion of the sample. In terms of work experience, 44 respondents (14.3%) had less than 3 years of experience, 81 (27.2%) had 3 to 10 years, 95 (30.9%) had 10 to 20 years, and 87 ( 27.6%) had 20 years or more of experience.
Levels of emotional labor and job stress among teachers
In contemporary society, teachers are encountering increasing challenges in effectively fulfilling their professional roles. Violations of teachers’ rights—rights that are essential to maintaining their professional status and carrying out their educational responsibilities—are rapidly increasing (Lim and Do, 2014). Within this context, teachers are experiencing significant levels of stress, which threatens their mental health and psychological well-being. This stress can ultimately have a detrimental impact on their teaching performance (Lee and Lee, 2009). Given these circumstances, researchers emphasize the need for even more studies aimed at enhancing teachers’ psychological well-being (Lim and Do, 2014). Therefore, it is essential to make concerted efforts to reduce individual teacher stress and to identify effective strategies to prevent and manage job stress, thereby fostering a happier and more supportive school environment (Jeong, 2022).
This study measured and analyzed the subjective levels of emotional labor and job stress among teachers, utilizing a 10-point Likert scale. The measurement approach was informed by prior research on teachers’ emotional labor, job stress, and burnout conducted by Jang (2022), Lee (2019), and Lee et al. (2017). The results are presented in Table 4. An analysis of levels of emotional labor revealed that the largest proportion of respondents (68 individuals, 22.1%) selected a score of 8. The distribution then followed in descending order: 7 (53 respondents; 17.3%), 10 (42; 13.7%), 5 (41; 13.4%), 9 (36; 11.7%), and 6 (33; 10.7%). The least frequent responses were observed for scores of 3 ( 13; 4.2%), 2 ( 8; 2.6%), 4 (7; 2.3%), 1 (5; 1.6%), and 0 (1; 0.3%) (χ2 = 179.257, p = .000***). These findings align with those of Jang (2022) and Yi (2022), who reported that teachers experience high levels of emotional labor. Regarding levels of job stress, the most common response was a score of 7, reported by 65 respondents (21.2%). This was followed by scores of 8 (45 individuals; 14.7%), 5 (44; 14.3%), 10 (41; 13.4%), 6 (41; 13.4%), and 9 (25; 8.5%). Lower frequencies were observed for scores of 3 ( 16; 5.2%), 2 ( 11; 3.6%), 1 ( 10; 3.3%), 4 (7; 2.3%), and 0 (1; 0.3%) (χ2 = 149.876, p = .000***). These results are consistent with the findings of Kim (2017) and Park (2005), who similarly identified high levels of job stress among teachers. As such, given that teaching is widely regarded as an emotionally demanding profession, research on emotional labor and job stress among teachers has been on the rise. This aligns with the findings of Yang (2022), who reported that secondary school teachers perceive themselves as emotional laborers and experience significant emotional labor and stress in their interactions with students, parents, colleagues, school administrators, as well as in the course of their educational duties. The fact that all survey respondents in this study were teachers at private middle and high schools may have contributed to the elevated levels of perceived job stress. This finding is consistent with the analysis by Park (2003), which indicated that private schools tend to restrict teachers’ participation in school decision-making processes compared to public schools, thereby offering less autonomy and discretion to teachers.
Individual stress management strategies
The individual stress management strategies of survey respondents were analyzed based on the research findings of Eom (2003), Kim (2010), and Song et al. (2017). The categorization of responses was structured in consultation with the research team, and multiple responses were permitted. The results are presented in Table 5. The most commonly reported strategy was exercise, selected by 212 respondents (31.9%), followed by travel (173, 25.9%), reading (113, 17%), cleaning (55, 8.3%), other activities (45 respondents, 6.8%), gardening (40 respondents, 6%), and no activity (27 respondents, 4.1%) (χ2 = 335.221, p = .000***). Notably, only 40 respondents (6%) reported engaging in gardening activities. This relatively low figure may be attributed to limited awareness or lack of direct experience regarding the stress-relief benefits of gardening, especially when compared to more common leisure activities. These findings suggest a need for greater promotion of horticultural therapy (Kim, 2008) and increased opportunities for individuals to engage in gardening as a form of stress management.
Stress management strategies and engagement in horticultural therapy programs: Participation frequency and willingness
The survey results regarding strategies for alleviating job stress and the willingness to participate in horticultural therapy programs are presented in Table 6. In terms of coping strategies or sources of support for relieving job stress, 266 participants (86.6%) reported they were relying on self-management. This was followed by 32 participants (10.4%) who stated they had no coping mechanisms, and 9 participants (3%) who reported seeking professional help (χ2 = 395.225, p = .000***). This finding contradicts the results of Shin (2011), which suggest that systematic management and intervention are more effective than self-reliance in mitigating job stress. However, it aligns with Kim (2013), who found that teachers experiencing higher levels of job stress tend to respond passively—often avoiding the issue or lacking effective coping mechanisms. Regarding the frequency of participation in horticultural therapy programs, 237 respondents (77.2%) indicated that they had never participated, 58 (18.9%) reported participating once or twice, 8 (2.6%) had participated five or more times, and 4 (1.3%) had participated three to four times (χ2 = 469.717, p = .000***). When asked about their willingness to participate in a horticultural therapy program if one were developed and offered, 221 respondents (72%) stated that they would definitely participate. In contrast, 51 respondents (16.6%) indicated they would not participate, 23 (7.5%) preferred to engage in other activities, and 12 (3.9%) responded that they would prefer to do nothing (χ2 = 372.023, p = .000***).
Component preference analysis for the development of a horticultural therapy program
To investigate preferences regarding the components of a horticultural therapy program, survey items were developed in consultation with the researchers of this study. These items covered various aspects of program design, including the program implementation time, type of delivery, total number of sessions, weekly frequency, number of participants, location of activities, and duration of each session/intervention. The survey design was informed by the findings of Ji (2009) and Kim (2006). Participants were classified into two groups based on their job stress levels, using the job stress measurement tool provided by the Korea Occupational Safety and Health Agency (KOSHA). Individuals scoring 7 or higher—representing the top 25% —were categorized as the high stress risk group (hereinafter referred to as high-risk group), while those scoring below 7 were classified as the low stress risk group (hereinafter referred to as low-risk group). A frequency analysis was conducted for each group, and a chi-square test of independence was performed to determine whether there were significant differences between the two groups. The results are presented in Table 7.
Regarding the preferred time for program implementation, lunchtime was the most favored option (100 respondents, 32.6%), followed by after work (98 respondents, 31.8%), during exam periods (61 respondents, 19.9%), during vacation periods (38 respondents, 12.4%), and on weekends or holidays (10 respondents, 3.3%) (χ2 = 98.033, p = .000***). Among the high-risk group, preferences were as follows: after work (62 respondents, 35.0%), lunchtime (58 respondents, 32.8%), exam period (31 respondents, 17.5%), vacation period (19 respondents, 10.7%), and weekends or holidays (7 respondents, 4.0%). In the low-risk group, lunchtime was the most preferred (42 respondents, 32.3%), followed by after work (36 respondents, 27.7%), exam period (30 respondents, 23.1%), vacation period (19 respondents, 14.6%), and weekends or holidays (3 respondents, 2.3%). However, the difference in preference between the high- and low-risk groups was not statistically significant (χ2 = 3.972, p = .410NS).
Regarding the preferred type of program delivery, the largest proportion of respondents (145 participants, 47.2%) reported no specific preference for the format. Group therapy was the next most preferred option, selected by 102 respondents (33.3%), followed by individual therapy, chosen by 60 respondents (19.5%) (χ2 = 35.303, p = .000***). In the high-risk group, a similar trend was observed: the majority (85 participants, 48.0%) indicated no specific preference regarding the delivery format. Group therapy was selected by 53 participants (29.9%), while individual therapy was preferred by 39 participants (22.0%). A comparable pattern emerged in the low-stress group, where the most frequent response was again indifference to the format (60 participants, 46.2%). This was followed by a preference for group therapy (49 participants, 37.7%) and individual therapy (21 participants, 16.2%). However, no statistically significant differences were found between the high- and low-risk groups in terms of format preference (χ2 = 2.736, p = .255NS). Overall, the results indicate a clear preference for group therapy over individual therapy. This may reflect participants’ desire for a delivery method that allows them to engage collectively despite their busy schedules, as well as their interest in achieving psychological relief and stress reduction through positive interactions with colleagues (Kim, 2016).
Regarding the total number of program sessions, the majority of respondents (261 individuals, 85.0%) preferred 10 sessions or fewer. This was followed by preferences for 11–20 sessions (35 individuals, 11.4%), 31 or more sessions (8 individuals, 2.6%), and 21–30 sessions (3 individuals, 1.0%) (χ2 = 597.482, p = .000***). In the high-risk group, 148 participants (83.6%) preferred 10 sessions or fewer, followed by 20 (11.3%) who preferred 11–20 sessions, 8 (4.5%) who preferred 31 or more sessions, and 1 (0.6%) who preferred 21–30 sessions. In the low-risk group, 113 participants (86.9%) preferred 10 sessions or fewer, followed by 15 (11.5%) who preferred 11–20 sessions, 2 (1.5%) who preferred 21–30 sessions, and none (0.0%) who preferred 31 or more sessions. However, the difference between the two groups was not statistically significant (χ2 = 6.703, p = .082NS).
Regarding the preferred number of sessions per week, the majority of respondents (228 individuals, 74.3%) preferred once a week. This was followed by twice a week (48 individuals, 15.6%), other frequencies (26 individuals, 8.5%), and three or more times per week (5 individuals, 1.6%) (χ2 = 409.469, p = .000***). In the high-risk group, 136 participants (76.8%) preferred once a week, followed by 26 (14.7%) who preferred twice a week, 12 (6.8%) who selected other frequencies, and 3 (1.7%) who preferred three or more times per week. In the low-risk group, 92 participants (70.8%) preferred once a week, followed by 22 (16.9%) who preferred twice a week, 14 (10.8%) who selected other frequencies, and 2 (1.5%) who preferred three or more times per week. However, there was no statistically significant difference between the two groups (χ2 = 2.031, p = .566NS).
Regarding the preferred number of participants per program session, the majority of respondents (240 individuals, 78.2%) preferred 10 or fewer participants. This was followed by 51 respondents (16.6%) who preferred 11 to 20 participants, 12 respondents (3.9%) who preferred 21 to 30 participants, and 4 respondents (1.3%) who preferred 31 or more participants (χ2 = 479.463, p = .000***). In the high-risk group, 143 participants (80.8%) preferred 10 or fewer participants, followed by 25 (14.1%) who preferred 11 to 20, 5 (2.8%) who preferred 21 to 30, and 4 (2.3%) who preferred 31 or more participants. In the low-risk group, 97 participants (74.6%) preferred 10 or fewer participants, followed by 26 (20.0%) who preferred 11 to 20, 7 (5.4%) who preferred 21 to 30, and none (0.0%) who preferred 31 or more participants. However, the difference between the two groups was not statistically significant (χ2 = 6.118, p = .106NS).
Regarding the preferred location for program activities, the highest proportion of respondents preferred indoor settings (129 individuals, 42.0%), followed by indoor-outdoor hybrid settings (119 individuals, 38.8%) and outdoor settings (59 individuals, 19.2%) (χ2 = 28.013, p = .000***). In the high-risk group, 71 participants (40.1%) preferred indoor settings, followed by an equal number (71 participants, 40.1%) who preferred indoor-outdoor hybrid settings, and 35 participants (19.8%) who preferred outdoor settings. In the low-risk group, 58 participants (44.6%) preferred indoor settings, followed by 48 participants (36.9%) who preferred indoor-outdoor hybrids, and 24 participants (18.5%) who preferred outdoor settings. However, the difference between the two groups was not statistically significant (χ2 = .626, p = .731NS). Notably, 119 respondents (38.8%) expressed a preference for indoor-outdoor hybrid activities, suggesting that while there is a strong overall preference for indoor settings, mixed-format programs are also well-received. These findings highlight the importance of flexibility in the spatial design and operation of horticultural therapy programs.
Regarding the preferred duration of each session or intervention, the majority of respondents (217 individuals, 70.7%) indicated a preference for 60 minutes or less, followed by 79 individuals (25.7%) who preferred 60 to 90 minutes, 10 individuals (3.3%) who preferred 90 to 120 minutes, and 1 individual (0.3%) who preferred 120 minutes or more (χ2 = 389.169, p = .000***). Among the high-risk group, the most preferred session duration was 60 minutes or less, selected by 131 individuals (74.0%), followed by 60 to 90 minutes (38 individuals, 21.5%), 90 to 120 minutes (7 individuals, 4.0%), and 120 minutes or more (1 individual, 0.3%). Similarly, in the low-risk group, the most preferred duration was also 60 minutes or less, chosen by 86 individuals (66.2%), followed by 60 to 90 minutes (41 individuals, 31.5%), 90 to 120 minutes (3 individuals, 2.3%), and none who preferred 120 minutes or more. However, no statistically significant difference was observed between the two groups (χ2 = 4.967, p = .174NS).
The highest preference was for the total number of program sessions to be ten or fewer, conducted once a week for 60 minutes or less. This aligns with the findings of Choi (2022). A review of previous studies on the effects of alternative therapies based on the number of sessions per week shows that, in a meta-analysis of music therapy, Yim (2006) found that the average effect size generally increased as the number of weekly sessions increased. Therefore, it is necessary to design program elements that both align with teachers’ preferences and reflect current research trends, in order to maximize program effectiveness. No statistically significant differences were found in preferences for the components of horticultural therapy programs according to stress level (high-risk vs. low-risk groups). This indicates that preferences for program elements are consistent regardless of stress level. This finding is consistent with Jeong (2022), who also reported no significant differences in preferences for forest therapy program components based on participants’ stress levels. Therefore, when designing horticultural therapy programs aimed at reducing stress among secondary school teachers, it is implied that preferences for program elements may be affected more by teachers’ coping strategies for stress than by the level of stress itself.
Finally, respondents were surveyed regarding their preferred horticultural therapy program activities. This survey item was developed in consultation with the research team, drawing upon the findings of Ji (2009), which examined annual trends in horticultural therapy programs. Multiple responses were allowed. The results indicated that the most preferred activity was gardening (222 respondents, 34%), followed closely by craft activities (221 respondents, 33.9%), cooking (164 respondents, 25.3%), learning activities (34 respondents, 5.21%), and other activities (11 respondents, 1.68%) (χ2 = 316.543, p = .000***). Among participants in the high-risk group, the most preferred activity was crafting (133 respondents, 35.5%), followed by gardening (115 respondents, 30.7%), cooking (101 respondents, 26.9%), learning activities (19 respondents, 5.1%), and other activities (7 respondents, 1.9%). In contrast, participants in the low-risk group showed the highest preference for gardening (107 respondents, 38.6%), followed by craft activities (88 respondents, 31.8%), cooking (63 respondents, 22.7%), learning activities (15 respondents, 5.4%), and other activities (4 respondents, 1.4%). However, no statistically significant differences were found between the two groups (χ2 = 4.926, p = .295NS). The activity categories were developed based on Kim (2023)’s classification of gardening activities and were included in the survey accordingly.
As such, horticultural therapy has been shown to be effective in relieving stress and enhancing quality of life by using plants to promote interpersonal connections and stimulate the five senses—sight, hearing, touch, taste, and smell (Jung, 2006). Furthermore, Hong (2006) found that horticultural therapy activities offer not only opportunities for plant cultivation but also provide social, psychological, and emotional healing benefits.
Conclusion
This study surveyed secondary school teachers to assess their levels of job stress and emotional labor, as well as their individual stress management strategies. The objective was to identify the most effective components for designing and implementing horticultural therapy programs aimed at reducing stress. The findings are intended to serve as foundational data for the development of tailored horticultural therapy programs to help alleviate teacher job stress. The following conclusions and recommendations are based on a comprehensive analysis of the study results.
First, a significant proportion of respondents scored 7 or higher on the emotional labor and job stress scales, indicating that secondary school teachers are experiencing persistent psychological and emotional strain. This result is consistent with the findings of Song and Lee (1997), who reported that higher stress levels are associated with increased psychological and physical burden. Therapeutic interventions for emotional recovery are needed.
Second, teachers reported using personal stress relief strategies such as exercise, travel, and reading, and there was also a clear demand for professional counseling and alternative therapeutic approaches. Notably, the majority of participants expressed a positive attitude toward the introduction of horticultural therapy programs, supporting both the feasibility and validity of developing such programs for teachers.
Third, an analysis of program component preferences indicated that teachers generally favored group-based activities over individual ones. This aligns with the findings of Min (2023), which demonstrated that interactive activities with colleagues can have a positive effect on reducing occupational stress. Additionally, the location, frequency, and duration of such programs are likely to be effective if designed and implemented in accordance with most horticultural therapy programs.
Based on the above conclusions, we propose the following recommendations for future research. To mitigate the job stress faced by secondary school teachers, it is necessary to seek group-oriented and systematic approaches alongside existing short-term, individual-centered interventions. This aligns with the findings of Shin (2011). The positive responses to the survey items regarding the implementation of horticultural therapy programs suggest that such programs can be effectively utilized as a means of alleviating job stress. Moreover, in line with Yang (2025), it is essential to develop practical program components based on institutional support, taking into account the time constraints and high workload demands characteristic of secondary education settings. Therefore, future studies should include empirical testing of the effectiveness of horticultural therapy through pre- and post-intervention assessments. It is also recommended that program components be tailored to reflect diverse variables such as teachers’ working conditions, age, gender, and coping strategies.
