Enhancing urban agriculture through participants’ satisfaction: The case of Seoul, Korea

Land Use Policy 69 (2017) 123–133

Contents lists available at ScienceDirect

Land Use Policy journal homepage: www.elsevier.com/locate/landusepol

Enhancing urban agriculture through participants’ satisfaction: The case of Seoul, Korea

MARK

⁎

Joo-seok Oh, Sei-yong Kim

Department of Architecture, Korea University, 352, Bldg. of Engineering, Anam Campus, Anam-dong 5-ga, Seongbuk-gu, Seoul, Korea

A R T I C L E I N F O

A B S T R A C T

Keywords: Urban agriculture Policy evaluation South Korea Citizen satisfaction Regression analysis

South Korea has been lending a variety of official support to promote the urban agriculture (UA), especially after enacting Act on Development and Support of Urban Agriculture (Act of UA) in 2011. Under this policy, which aims at spreading the UA effectively in dense cities, in particular, the government categorizes spaces used for the UA into five types in accordance with their characteristics and has extended customized support. But it has never been explored if the Korean version of the UA has satisfied the participants and there exist any differences in the satisfaction level among participants of each spatial type. With a goal to explore the policy effectiveness, therefore, this research carried out interviews and surveys involving 808 citizens who have taken part in UA activities in 68 farmlands in South Korea's capital city of Seoul and then analyzed them with the ANOVA and the hierarchical regression analysis methods The results revealed statistically significant differences among the different UA types in terms of the respondents’ overall satisfaction, the current status of the UA operation, and policy measures that affect the satisfaction level as well as their effects. Overall, however, the participation of female citizens, the supply of physical facilities by public entities, and the increase in the time respondents were involved in UA and their cultivating areas turned out to positively affect their satisfaction level.

1. Introduction Urban Agriculture (UA), which had mainly been witnessed in underdeveloped nations, has spread fast throughout the world after the economic crisis (Bryld, 2003), with it gaining ground in advanced countries such as the United States as more have concerned about the food security and tried to find ways for a sustainable land use. (Bohn and Viljoen, 2011; Lovell, 2010; Viljoen and Bohn, 2009) Over the course, UA got to be understood as part of the landscaping to boost the quality of the urban life as well as agricultural activities to vitalize local communities. Such extended meanings have led many governmental entities to set up diverse laws and action plans to officially introduce and extend UA. Since then, a slew of researches, including the one by Lovell (2010), have proven that UA has positive effects on the selfsupply of crops within a city, the productive landscape and the creation of sustainable urban environments. Also, a variety of attempts have been made recently to overcome limitations the current way of the UA operation has by, for example, integrating it with the urban planning and the land use. Such attempts have served as a catalyst for policymakers in many countries and cities to either draw or revise relevant laws to make the UA official (Deelstra and Girardet, 2000), which have

⁎

nor only boosted the size of its cultivation areas but sparked the interests of urban residents directly and indirectly. (Nugent, 2000). In this sense, the multi-stake holder action plan for the urban management is needed with long-term perspectives to maintain the policy in a stable and sustainable fashion (Mougeot, 2000). The evaluation of the public participation, which is one of the key pillars of UA, is also required with a short-term view for its immediate implementation. (Moskow, 1999; Guitart et al., 2012), and major researchers, including Travaline and Hunold (2010), Moskow (1999), and Guitart et al. (2012), have adopted the participation-observation methods as practical research tools to draw the general picture of the UA in different regions. Asia, which is expected to house some 66 percent of the world's total population by 2050 (United Nation, 2014), has never been an exception: the nations there have recently begun to perceive the importance of the sustainable urban development and the value of UA for urban planning and land use (Hara et al., 2005; Roberts and Kanaley, 2006), and they have actively pushed for relevant policy measures particularly over the past 20 years. South Korea set up the Act on Development and Support of Urban Agriculture (Act of UA) in 2011. Albeit being belated compared to other

Corresponding author. E-mail addresses: [email protected] (J.-s. Oh), [email protected] (S.-y. Kim).

http://dx.doi.org/10.1016/j.landusepol.2017.04.053 Received 17 September 2015; Received in revised form 30 April 2017; Accepted 30 April 2017 0264-8377/ © 2017 Elsevier Ltd. All rights reserved.

Land Use Policy 69 (2017) 123–133

J.-s. Oh, S.-y. Kim

the year (MAFRA, 2013). What boost the public interests in the UA were (1) the increased demand for safe agricultural products and (2) increased income and leisure time (Park and Yang, 2014). In line with the growing popularity of community gardens and home gardens in advance nations, more local people have also got to pay attention to the UA and a growing number of environment groups and private organizations led the moves and supported them, which finally resulted in the government's enactment of the Act of the UA which has 24 articles. In accordance with the provision of Article 1, the goal of the regulation is to “develop a nature-friendly urban environment and contribute to harmonious development of cities and rural communities by raising urban residents‘ understanding of agriculture.” What it means can be understood in two aspects: firstly, the subject of this crop-raising project is neither specific organizations nor practitioners but general urban residents (Oh and Kim, 2014), and secondly, the main goal of the UA in Korea is to contribute to the sustainable landscape within a city, rather than promoting the production of crops, as stated in researches by Kim et al. (2016), Le Roux et al. (2014) and Lovell (2010), among others. It also runs counter to some existing views, such as Zezza and Tasciotti (2010), which define the UA as “the production of crop and livestock goods within cities and towns.” The Act of UA states five spatial types – UA in residential buildings, UA in neighborhood, UA in city centers, UA in a farm or park, and UA for school education (National Law Information Center of Korea, 2016) – to make the UA more efficient in extremely dense metropolises. But, as mentioned above, the statute serves as a guideline, rather than a mandatory legislation, and recommends each local government to choose a suitable type and apply details in enacting an ordinance of their own.

nations, Korea made the policy comprehensive by systematically arranging UA which had mainly been operated by individuals and local communities in a small scale after factoring in diverse factors such as land types and their way of being managed and supported (Kim et al., 2016). In particular, the Act of UA stipulates a total of five spatial categories of farmlands in consideration of UA activities in major metropolises such as the capital city of Seoul, whose population density is 17,000/km2, and set up legal foundations to support them. As a result, the area reserved for UA in South Korea has grown over sixfold in five years, (Seoul Metropolitan Government, 2015a), which proves that the policy has had some positive effects for the densely populated city. Despite such a history of the policy-based UA development, however, no in-depth researches and evaluation on South Korea's UA policy have been made so far, and little has been known in the international arena. Many non-Korean studies have stressed the need to look into different types of UA-related measures in different regions (Bakker et al., 2000; Brown and Jameton, 2000; Cole et al., 2008; DrakakisSmith et al., 1995), and local experts have also pointed to the significance of the policy improvement through a consistent monitoring of participants' satisfaction level (Kim et al., 2016). In addition, despite the need for risks and benefits to be addressed for an active policy making (Mougeot, 2000), there do exist limitations in active discussions to improve policy measures or to make them be legally binded, as UA activities are often deemed recreational rather than something that plays a major role in land use. In order to overcome the aforementioned limitations, South Korea let its Act of UA be based upon the National Land Planning and Utilization Act, which makes it compulsory in terms of creating farmlands. But it has not been evaluated either at home or abroad about how the policy tool has affected the UA development in Korea. Accordingly, this paper put the focus on the quantitative comparison and the analysis of the user satisfaction level by mobilizing a total of 808 UA participants in Seoul, which is one of the most brisk cities in terms of the municipal government's support for UA and its growth accordingly, via face-to-face interviews, surveys and their statistical analysis. Also, research models were drawn up with several independent variables, such as the socio-demographic characteristics of different groups of the respondents, their major reasons for joining UA, their satisfaction with official support, in order to learn which elements mainly affected the participants' satisfaction level. Through such an analysis of major policy measures and their effectiveness, this paper tried to explore ways to fine-tune them to promote the sustainability of UA.

2.2. Morphological characteristics of UA in Seoul As seen in Fig. 1, the capital city of Seoul, which is the most denselypopulated city in Korea, puts forth a total of four types of the UA by modifying the five categories stated in the Act of UA: (1) “Housing” uses interior/exterior spaces of private and multi-unit houses, including residential buildings with neighborhood type UA (2) “In-city” employs the rooftops of public facilities, (3) “Farm-park” takes advantage of city land that has been abandoned, and (4) “Education” uses schools or other locations where experiential and educational UA can occur (Seoul Metropolitan Government, 2014). Since setting up the ordinance in 2011, the municipal government has come up with relevant supportive measures, such as lending support for creating vegetable gardens on rooftops of 267 buildings and providing 43,785 box-typed vegetable pots for 15,866 places (Seoul Metropolitan Government, 2015a). Such efforts led the surge in UA in the city: in 2011, a total of 29-ha land was reserved for UA, but the area mostly quadrupled to 118 ha as of the end of 2015, which constitutes 0.2 percent of the city's total area. Noteworthy is that around 35 percent of this land is in some leftover spaces within the downtown or schools in the city while the rest is located in its suburban regions, which means that both the peri-UA and the intra-UA have grown at the same time. As shown in Table 1, the Seoul city government has also extended

2. Background 2.1. History and status of implementing UA in Korea and Seoul Since 1990s, voices have been grown in South Korea for the necessity to introduce the UA after local researchers began to look into a variety of agricultural activities in urban areas. As the comprehensive opinion poll in 2011 showed that many citizens were in favor of the UA, the government was encouraged to set up the Act of UA in November of

Fig. 1. Four main types of UA by the Seoul Metropolitan Government. (a) Housing UA on an apartment rooftop using box gardens; (b): In-city UA on a public rooftop; (b) Farm-park UA on large plots that private entities offer to applicants; and (d): Educational UA providing training and experiences to citizens.

124

Land Use Policy 69 (2017) 123–133

J.-s. Oh, S.-y. Kim

Table 1 Physical/social functions and policies by UA type. House PR Programs Education Provision

PR of public function Experiencing program Agro-technology education Farming material distribution Farming area distribution a

In-city

Farm-park

Educational

○ ○ ○ ○ ○ box garden

public share ○ lease

public share ○ lease

○ share/lease

Article 1, 2 and 10 of the Act of UA, which state major goals of the statute and the public support, and the respondents' satisfaction level was measured for an quantitative analysis to draw implications for the development and promotion of UA activities in Seoul.

comprehensive supports to participants in order to promote UA not only in physical terms but social and cultural aspects as stated in the Act of UA. Here, cultivating lands are classified as either box gardens or vegetable gardens, and public entities have taken a leading role in running UA by employing supportive centers and staff in order to offer desirable training programs and to promote UA activities further (Seoul Metropolitan Government, 2015a).

3.1.1. Act of urban agriculture in Korea and Seoul This paper firstly chose major articles of the Act of UA, as shown in Fig. 3, to select independent variables in order to determine the evaluation items. Out of 24 articles, Article 1 stipulates the purpose of its enactment, Article 2 puts forth major perspectives and objectives that the act aims at ultimately, and Article 8 defines the nature of South Korea's UA by presenting five spatial types of UA. Article 10 describes the establishment and the management of public support centers, including the distribution of seeds, box gardens, cultivation tools, and cultivating lands, as well as programs for hands-on experiences and training and relative personal relationship to enhance social and cultural effects of UA. Factoring in aforementioned details, this paper looked into four out of the five UA types that the Seoul metropolitan government adopts, and the policy objectives and the current status of the public support as independent variables. But such details that cannot be examined in quantitative/qualitative aspects as well as some appendix measures on the formation of relevant committees and reward/punishment to name a few, were not dealt with here.

3. Material and methodologies 3.1. Research setting and study area The public evaluation of a government policy would serve as a crucial way to assess it in a practical term and provide policy implications for its sustainability (Dunn, 2003; Heckman, 1991; Heckman and Vytlacil, 2007). In this sense, this paper carried out exploratory surveys and their statistical analysis involving active participants at 68 locations in 12 autonomous districts in Seoul out of 25 regions, as shown in Fig. 2, where lands for UA were built thanks to the public support. Major considerations for this analysis are as follows: First, this research carried out both face-to-face interviews of participants involved in different types of UA in Seoul and surveys on their satisfaction level with major policy measures in order to detect any differences among different groups and to paint more comprehensive picture by making up for this quantitative study. This paper also aims to learn how key variables, such as gender, age, and the level of income, would quantitatively affect the respondents' satisfaction. Secondly, this paper selected major policy measures which can be quantitatively evaluated by participants. As shown in Fig. 3, a total of 12 items out of

3.1.2. Operational conditions, vision of policies Previous researches, such as Travaline and Hunold (2010), referred not only the characteristics of respondents as well as local conditions and the examination on the operation status as crucial microscopic

Fig. 2. Location map of study area and survey areas (Map is based on Seoul GIS).

125

Land Use Policy 69 (2017) 123–133

J.-s. Oh, S.-y. Kim

Fig. 3. Systematic diagram of the “Act on Development and Support of Urban Agriculture of Korea”.

Table 2 Variables and methodologies used in the quantitative analysis (n = 808). Category

Questionnaires/Variables

Measurement/ Unit

Variable Code

Gender

Dummy

Age Occupation Income

$ per month

Gender (women) – – –

Methodology

Socio-demographic information

Satisfaction on main purpose of UA Article 2

Satisfaction on official provision Article 10

Satisfaction on provision & Operation status Article 1

Aspect of Hobby and Leisure Teaching-learning activity Crop cultivation Experience on rural and agriculture

Likert (1–5)

Hobby Learning Cultivation Experience

ANOVAC, HRA(I.V)d

Advertisement of public functions of UA Experiencing programs on UA Education on agricultural technology Distribution of farming materials

Likert (1–5)

PR Programs Agro-Edu Provision

ANOVA, HRA(I.V)

Amount of crop Nature-friendly urban environment Cultivating areas Participation time

Likert (1–5)

Amount Env Area Hr.

ANOVA HRA(I.V)

Satisfaction

HRA(D.V)

Sqm Hour(s) per week

Overall satisfaction rate on UA & UArelated activities Overall satisfaction on UA

Likert (1–5)

a. Hierarchical Regression Analysis, b. Independent Variable, c. One-way Analysis of Variance d. Dependent Variable.

paper also tried to learn if and how much UA activities help participants recognize the eco-friendly urban environment, which is one of key ultimate goals the Act of UA aims at, and took it as a group of independent variables.

elements to assess UA in a macroscopic term. In this sense, this paper examined the size of the land cultivated by participants in each category, and the time an individual spent for UA activities per week, while checking the socio-demographic information as seen in Table 2. This

126

Land Use Policy 69 (2017) 123–133

J.-s. Oh, S.-y. Kim

each group of the respondents shows the different level of satisfaction with 1) UA in general, 2) main perspectives and purposes stated in Article 2, and 3) the public support shown in Article 10. ANOVA is deemed as a useful tool to verify the average gap among more than two groups (Boone and Boone, 2012; Clason and Dormody, 1994; Norman 2010). Second, the Hierarchical Regression Analysis (HRA) was conducted to see if and how much major indices have meaningful effects on the overall satisfaction with UA. Here, three models of indices were put in turn: socio-demographic characteristics (3.1.2), respondents' satisfaction with major policy measures (3.1.1), and the current status of the operation. As for arranging the indices, this paper follows guidelines presented by several previous researches: the Baptiste et al. (2015) employed demographic variables as regression analysis in consideration of the public involvement in studying policy evaluation; and Maas and Hox (2004) and Willms (1999) called for applying the hierarchical structure, stressing the need to take into account individual characteristics at the first stage. Accordingly, this paper put the socio-demographic characteristics at the first step of the HRA, as shown in Fig. 4, and the group of variables on the satisfaction with major policies were inserted at Model 2 and 3, so as to explore the effectiveness of the UA policies. Meanwhile, Likert-type scale which this paper adopted is deemed as a technically ordinal scale, and scholars have different opinions on its usage for policy evaluations. But Johnson and Creech (1983) and Zumbo and Zimmerman (1993) noted that point 5 or more can be accepted as relatively continuous variables, and this paper took it as a reference point in taking the surveys and conducting quantitative research.

3.2. Data collection In order to measure major indices mentioned in 3.1, three rounds of survey were conducted involving a total of 808 UA participants residing in Seoul in April and June 2015 and April 2016. The quantitative data collected is shown in Table 2. To get more trustworthy results, the surveys were carried out faceto-face rather than through the web-based method Dillman and Bowker (2001), with 13 polltakers all taking at least a bachelor's degree in architecture, urban planning or landscaping. A total of 68 places were under the subject of the surveys: 24 vegetable gardens in apartment and private houses (n = 201), 11 UA farms (n = 202) in eight autonomous districts, 10 vegetable gardens (n = 203) within the campuses of middle/high schools and colleges in seven districts, and 23 rooftop gardens (n = 202) in nine districts. All respondents had been receiving technical and administrative supports from the government, as they had took part in UA by applying for the official program offered either by the municipal government or each district office. Of them, 42 middleand high-schoolers had not managed gardens on their own, as their UA activities were part of their education courses. To measure the satisfaction, the five-point Likert Scale was adopted, with one meaning very low and five very high, just as policy, sociology, and psychology studies widely use (Ostroff 1992; Göb et al., 2007). Questions on the general status quo, including the demographic characteristics, the participation time and the size of vegetable gardens, were drawn up with the normal data. Before taking the poll, researchers gave enough explanations on questions, and the respondents’ comments were tape-recorded in order to obtain more precise results for the quantitative analysis and to minutely detect any differences in their opinions.

4. Results 4.1. Overview of socio-demographic variables and UA participation

3.3. Methodology Of the 808 respondents, 44.39 percent, or 356 people, were male citizens, and 55.61 percent, or 444, were women. Some 10.1 percent (81 persons) were aged below 20, with those in their 20 s and 30 s accounting for 35.91 percent (288), those in their 40 s and 50 s taking up 30.3 percent (243), and citizens in their 60 s and older 23.29 percent (190). Students accounted for 25.81 percent (207) of the total respondents, followed by office workers 23.32 percent (187), housewives and those with no occupation 22.57 percent (181), the self-employed

Policy evaluation and the quantitative research on a group influenced by the policy measures often take different methodologies according to their purpose, and no internationally approved single method exists (Steudler et al., 2004). In order to perform the analysis referred in the Data Collection paragraph, this paper adopted the following two methodologies: First, the one-way analysis of variance (ANOVA) was used to see if

Fig. 4. Conceptual representation of research models, variables, and procedures.

127

Land Use Policy 69 (2017) 123–133

J.-s. Oh, S.-y. Kim

Fig. 5. Results of ANOVA test of mean values on satisfaction, working hours, and area. a. Comparison of average UA satisfaction level of each type b. Comparison of participation time per person in each type c. Comparison of average cultivating size per person in each type

UA, farm-park and in-city type participants turned out to feel more satisfactory: In Particular, farm-park participants, who had comparatively enjoyed spacious places, gave higher marks in terms of hobby and cultivation of crops, while in-city type people showed relatively higher satisfaction level in education term. As shown in Table 4, in-city type participants were found to show statistically higher levels of satisfaction than those in other categories in terms of public support stated in Article 10, except for the experiential programs aspect. Noteworthy, however, is that the house-type UA participants were found to be less satisfactory in this term, excluding the supply of materials, which indicates that they tend to shun any training programs or they have had limited chances to take part in them. Following comments gave such stories, which were echoed by many according to the interviews. (Respondent#453) “I am sowing seeds at home as a hobby and that's all. I don't want to bother to be trained or take part in relevant events for this job. I've got three box gardens from the district office, and I was able to find how to plant and grow them by a simple online search.” (Respondent#599) “I failed even to have my plants bud last year, maybe because I am a total stranger in this field. And I did not know if such public training programs exist. It would be good if I got through such systematic training courses, but it seems bothersome to find the one and to attend it as I am busy.” Drawing attention is that the respondents in the education-type UA turned out to be less satisfied with public education which runs counter to what this type of UA is meant to be, as shown in Table 3. Such a low level of their contentment with UA-related educational programs indicates that related policy measures appear to have weighted too much on in-city and farm-park type participants.

and those with special occupations 15.12 percent (126) and labor workers 12.59 percent (101). Noteworthy is that the research on the current status ahead of taking the surveys pointed to the significance of efforts to promote UA further in Korea, which is in its burgeoning stage. After all, 79.3 percent of the 599 respondents, excluding students and other citizens that had taken part in the UA activities as short-term education programs, said they have got to join the project after seeing advertisements by the Seoul metropolitan government or their district offices, while the rest said the media coverage on the issue and the recommendations by their acquaintances led them to take part in UA. Among the 203 respondents who were in the education-type UA, a whopping 89.6 percent said the promotion at school or via other channels were the No. 1 motivation of their participation. It also deserves attention that respondents in different groups show differences in their UA cultivation areas and working hours. Participants in the farm-park type turned out to have managed the largest size of gardens of 30.42 m2 on average, and spent the longest hours of 11.66 h per week. But they visited the site 1.2 times per week, far fewer than their counterpart in the in-city category who logged 3.3 times per week, which appears to be attributable to the matter of access. In fact, 49 percent of the farm-park participants said they used their own vehicles or the public transportation to get to their gardens mostly located in the outskirt of the city, while 81 percent of the in-city type participants said they can reach their gardens on foot.

4.2. Major policy evaluation results of the act of UA (a) in Fig. 5 shows differences in the overall satisfaction of the 808 respondents involved in the four types of UA activities. While the satisfaction level logged the statistical differences (F = 12.885, p < 0.001) by type, the post examination (scheffe) found that the satisfaction among farm-park (3.83) and in-city (3.71) type participants were higher than that among those in the house (3.34) and education (3.28)-type UA. Similar results were drawn out in the respondents' evaluation on the major objectives of UA and their satisfaction with the government's major supportive measures, as shown in Table 3. Upon the four objectives stated in Article 2 – hobby and Leisure, education, experiencing a rural life, and cultivation of crops through

4.3. HRA results and model comparison The statistics showed that each group showed differences in their participants’ satisfaction with the public support and the effects of the objectives UA aims at, as well as in the current status of operations. Based upon the results, the HRA method was used to learn which elements affect the satisfaction level. Such a statistical structure was found to be proper as the figures increase according to the application of each 128

Land Use Policy 69 (2017) 123–133

J.-s. Oh, S.-y. Kim

Table 3 Results of ANOVA test of Article 2 (purpose and scope) of Act of UA (n = 808). Category

Types of UA

Mean

Standard deviation

95% confidence level mean Lower

Upper

f-statistic

Comparison

c > a, b, d

Hobby

House In-city Farm-park Education

2.90 3.06 3.77 2.90

1.324 1.110 1.246 1.397

2.71 2.91 3.60 2.71

3.08 3.21 3.95 3.09

21.761***

Learning

House In-city Farm-park Education

2.43 2.72 2.52 2.58

0.957 1.104 1.078 1.059

2.30 2.56 2.37 2.43

2.57 2.87 2.67 2.72

2.537

Experience

House In-city Farm-park Education

2.38 2.95 2.67 2.80

0.773 1.281 0.937 1.158

2.28 2.77 2.54 2.64

2.49 3.12 2.79 2.96

10.371***

b, d > a, c

Cultivation

House In-city Farm-park Education

3.18 2.86 3.60 2.24

1.205 1.109 1.040 1.028

3.02 2.70 3.46 2.09

3.35 3.01 3.74 2.38

56.124***

c > a > b > d

stages, still has a long way to go before being self-sustainable. Lastly, how the participants' satisfaction with the status of the physical operation affects dependent variables were dealt with in the Model III, with the results also revealing gaps among different groups. Regardless of their types, however, it was found that the longer the participants were involved in the UA activities and relevant training sessions, their satisfaction level rose. Secondly, except house-type participants, the larger spaces the respondents cultivated and yielded a larger amount of crops, they were more satisfied with UA (Fig. 6).

model just as shown in Table 5. Finally, the house-type UA logged 72 percent, in-city type 74.5 percent, farm-park type 95.5 percent and education type 72.3 percent (adjusted R). In the Model 1 which includes the socio-economic variables of the respondents, it deserves attention to the correlations between the UA activities and female citizens. Such a linkage has mentioned mainly in researches focusing on the food security in developing nations and the economic aspect of UA, but this paper found a specific relation between the women and UA activities in Seoul where they usually take place as a kind of hobby, which calls for a need for more in-depth researches further. The Model II was designed to learn the satisfaction with the policy support which positively affects the UA participants, as well as the impact the satisfaction with the major goals of UA has on dependent variables. The results varied upon each type, indicating that the different groups of participants complemented each other so as to achieve the objectives of UA – hobby, learning, having experiences and cultivating crops – in Seoul as a whole. Noteworthy is that the state support for cultivating lands and seeding helps boost the satisfaction level of all participants, which indicates that UA in Seoul, which is in its early

4.4. Analysis of results by UA type 4.4.1. HRA results of house UA The HRA on a total of 201 respondents in the house-type UA showed that the participation by female citizens in Model I and the objectives of UA as a hobby and experiences turned out to have positive effects on the participants' satisfaction level. In particular, the physical support was found to be the most crucial element with the greatest implication on the satisfaction level, which indicates that no public support would mean a failure to boost participants' satisfaction. But education/training

Table 4 Results of ANOVA test of Article 10 (Establishment of support centers) of Act of UA (n = 808). Category

Typology

Mean

Std. dev.

95% confidence level mean Lower

Upper

f-statistic

Comparison

PR

House In-city Farm-park Education

1.58 3.00 2.28 2.39

0.674 1.234 0.804 1.098

1.49 2.82 2.16 2.24

1.68 3.17 2.39 2.54

70.377***

b > d > c > a

Program

House In-city Farm-park Education

1.68 2.72 3.24 2.71

0.640 1.103 1.069 1.145

1.59 2.57 3.09 2.55

1.77 2.87 3.38 2.86

84.394***

b > c, d > a

Agro-edu

House In-city Farm-park Education

1.85 3.20 3.23 2.24

0.801 1.164 1.211 0.989

1.73 3.03 3.06 2.10

1.96 3.36 3.40 2.37

87.894***

c > b, d > a

Provision

House In-city Farm-park Education

2.47 3.04 2.36 3.05

0.980 1.242 0.835 1.167

2.34 2.86 2.24 2.89

2.61 3.21 2.48 3.22

23.749***

b, c > d > a

129

Land Use Policy 69 (2017) 123–133

J.-s. Oh, S.-y. Kim

Table 5 Standardized coefficients (beta) and p-values of HRAa (Hierarchical Regression Analysis) results of all UA typologies and variables (n = 808). House (n = 201) Beta

In-city (n = 200) Beta

Farm-park (n = 203) Beta

Education (n = 204) Beta

Model I (Socio-demographic variables) Gender (women) 0.164*** Middle school −0.028 High school −0.037 Undergraduate 0.078 Graduate 0.099 < 20 years old −0.021 20–40 years old −0.012 40–60 years old −0.008 > 60 years old −0.011 Student 0.047 Blue collar 0.010 White collar 0.050 Self-employed 0.029 Unemployed 0.080 < 2 thousand dollars −0.096* 2–4 thousand dollars −0.057 > 4 thousand dollars 0.037

0.101* −0.033 0.002 0.089 0.041 −0.089 0.022 0.081 −0.044 0.040 0.003 0.011 −0.005 −0.015 0.051 −0.085 0.026

0.019 0.004 −0.100 −0.023 −0.049 −0.020 −0.063 0.010 −0.072** −0.096** −0.155*** −0.113* −0.105* −0.008 −0.017 0.012 0.003

0.107** −0.040 −0.043 −0.067 −0.088 0.029 −0.007 0.054 0.012 −0.070 −0.030 −0.056 −0.012 0.000 −0.002 −0.040 −0.035

Model II (Satisfactory evaluation of Act on UA) Hobby 0.114** Learning −0.107** Experience 0.086 Cultivation 0.060 PR. 0.016 Program −0.006 Agro-edu 0.018 Provision 0.324***

0.203*** 0.769 0.030 .109** −0.045 −0.693 0.122** 0.174***

−0.023 0.114*** 0.109*** 0.196*** 0.001 0.231*** 0.096** 0.286***

0.104 0.140** 0.297*** 0.069 0.078 −0.113 0.067 0.125*

Model III (Operation condition and satisfaction) Env 0.039 Amount 0.058 b Space 0.042 c Hr 0.256***

0.212*** 0.139** 0.126** 0.266***

−0.074** 0.130*** 0.137*** 0.165***

0.183** 0.061 0.082* 0.191***

Model Summary (based on Model III) F-statistic 18.749*** R/R2/Adj. R2 0.449/.761/.720 Durbin-Watson 1.888

21.043*** 0.162/0.557/0.745 2.212

61.669*** 0.368/0.839/0.955 1.749

19.187*** 0.205/0.662/0.723 1.942

* = p < 0.05. ** = p < 0.01. *** = p < 0.001. a All inputted data are based on results of Model III. b Provided area by (local) government or public institution (sqm).

programs provided by the Seoul government were found to have failed to do good to participants, given the aforementioned interviews and the analysis which showed that UA with educational objectives would rather spoil the participants' appetite. On the contrary, UA as a hobby was proven to be an element that enhance their satisfaction.

activities. Factoring in the fact that General buildings are private properties, and South Korea's law on construction does not have rules on the landscape of rooftops for every size of buildings, it is recommended to find ways to legally link the Act of UA with relevant legislations.

4.4.2. HRA results of in-city UA According to the HRA on 200 people involved in the in-city type one, UA for a hobby, learning, and the cultivation of crops positively affected the participants' satisfaction among effects of the UA the public entities aim to achieve. It seems to be attributable to the fact that more than 88 percent of the respondents took part in the UA activities at gardens located on the rooftops of buildings of public institutions and regularly took public training programs. The analysis also showed that the time of their involvement and their contentment with the public training courses led to the improvement in the overall satisfaction with UA, indicating that it would be needed to proactively devise measures to improve such programs in both quantitative and qualitative terms to further satisfy the participants. The crop yield and the cultivation area showed a statistically meaningful value, but the spatial scope in this type is limited as of now to rooftops of public institutions, Thus, it would be necessary to amend relevant regulations to extend the scope to rooftops of private buildings and other structures as a venue for UA

4.4.3. HRA results of farm-park UA The HRA on the 202 participants in the farm-park type UA showed that the correlation between the independent and dependent variables grew from 36.8 percent in Model I, to 83.9 percent in Model II and to 95.5 percent in Model III, which proves that this statistical structure is the most suitable for studying this type of UA. The analysis also found that the cultivation area and the time of the participants' involvement positively affected their satisfaction level. But the distinguished point here is the effect of demographic factors: the participation by the elderly aged over 60, low-income citizens, students and general workers turned out to adversely affect the overall satisfaction. As mentioned in 4.1, this is attributable to the poor access to their farmlands which led the participants to visit there UA sites less frequently. Thus, it would be recommended to take a community-based approach, rather than stick to the current way, to encourage people to join UA activities in farmlands located within or near their town. Factoring in the fact that the participants in this type showed the highest 130

Land Use Policy 69 (2017) 123–133

J.-s. Oh, S.-y. Kim

Fig. 6. Standardized coefficients values of regression model by UA type (based on Modell III). * = p < 0.05, ** = p < 0.01, *** = p < 0.001

participants managed the largest size of land, or 30.42 m2, on average, while the house-type people operated 13.31m2 and the education-type participants a mere 5.10 m2. Such a linkage between the size of farmlands and the satisfaction level indicates that it would be desirable for Seoul to extend spaces reserved for UA down the road for its success. Second, the statistics to verify the effectiveness of UA policy showed that the hierarchical regression analysis (HRA) model, which reflects the socio-economic features, the policy effectiveness, and the current status of the management, is proper, as the figures increase according to the application of each model. In terms of the adjusted R2 value, the house-type UA logged 72%, in-city type 74.5%, farm-park type 95.5% and education type 72.3%. Thirdly, public promotions and relevant events turned out to have little effect on participants' satisfaction, which runs counter to general notions and expectations that they would spark public interests and ultimately attracts more people. Education/training programs on agriculture were also found to have limited effects on the satisfaction level, as it stopped short of heightening the satisfaction level of the farm-park and in-city type participants who managed large spaces, not all users. Fourth, factors that contributed to raising the satisfaction level differ in accordance with the type of UA. The HRA results showed that the agricultural activities as a hobby helped boost the satisfaction level among house-type and in-city type participants; experiences in agriculture made farm-park and education-type participants gratified, the harvest of crops caused in-city and farm-park type respondents to feel happy with UA. In particular, the effect of education/training on UA were found to have resulted in heightening the satisfaction level among all participants, except those involved in the house-type UA, which points to a need to set up a roadmap to provide users with more systematic programs. Lastly, what have caused the participants to be satisfied with UA include 1) physical supports by public entities, 2) their longer hours involved, 3) a larger size of land and 4) the participation by female

satisfaction level, and the distribution of variables that positively affected the overall UA satisfaction of the respondents were found to even throughout the goals that UA aims at, it would be necessary for public entities to strive for further expansion of this type of UA. 4.4.4. HRA results of education UA The survey on the 203 participants in the education-type UA activities showed that the time of participation, area, and public support turned out to affect their overall satisfaction, though the respondents often had limited experiences due to time constraints. Also, their satisfaction with UA activities and lessons positively affected the dependent variables, which indicates that the education type UA met its original goals. But it was found that the governments-led active promotion efforts and relevant training programs failed to help boost the respondents' satisfaction. In this sense, the public entities are required to improve relevant policy measures to enhance the effects of their PR and education programs. It would also be needed to come up with ways to encourage people to join UA activities in a longer period of time, rather than experiencing it as a one-off event, which is expected to further boost the effectiveness of the education-type UA. 5. Summary and discussion This paper dealt with overall circumstances and the state policy that contributed to the implementation of UA in South Korea, a theme that has rarely been covered in international researches. In particular, this article found the effectiveness UA has on the participants differs in accordance with its type. The results of the statistics are as follows: First, the ANOVA test involving 808 UA participants showed that the satisfaction level among the respondents of each type revealed statistical differences. The satisfaction among farm-park and in-city type participants stood at 3.83 and 3.71, respectively, which are higher than that among those in house (3.34) and education (3.28) types. (F = 16.097, p < 0.001). Noteworthy is that the farm-park type 131

Land Use Policy 69 (2017) 123–133

J.-s. Oh, S.-y. Kim

Bryld, E., 2003. Potentials, problems, and policy implications for urban agriculture in developing countries. Agric. Hum. Values 20 (1), 79–86. Clason, D.L., Dormody, T.J., 1994. Analyzing data measured by individual Likert-type items. J. Agric. Educ. 35 (4), 31–35. http://dx.doi.org/10.5032/jae.1994.04031. Cole, D.C., Lee-Smith, D., Nasinyama, G.W. (Eds.), 2008. Healthy City Harvests: Generating Evidence to Guide Policy on Urban Agriculture. CIP/Urban Harvest and Makerere University Press, Lima, Peru. Deelstra, T., Girardet, H., 2000. Urban agriculture and sustainable cities. In: Bakker, N., Dubbeling, M., Gündel, S., Sabel-Koshella, U., de Zeeuw, H. (Eds.), Growing Cities, Growing Food. Urban Agriculture on the Policy Agenda. Zentralstelle für Ernährung und Landwirtschaft (ZEL), Feldafing, Germany, pp. 43–66. Dillman, D.A., Bowker, D.K., 2001. The Web questionnaire challenge to survey methodologists. In: Reips, U.-D., Bosnjak, M. (Eds.), Dimensions of Internet Science. Pabst Science, Lengerich, pp. 159–178. Drakakis-Smith, D., Bowyer-Bower, T., Tevera, D., 1995. Urban poverty and urban agriculture: an overview of the linkages in Harare. Habitat Int. 19 (2), 183–193. http:// dx.doi.org/10.1016/0197-3975(94)00065-a. Dunn, W.N., 2003. Public Policy Analysis: An Introduction. Prentice Hall, New Jersey, NJ. Ellis, F., Sumberg, J., 1998. Food production, urban areas and policy responses. World Dev. 26 (2), 213–225. Göb, R., McCollin, C., Ramalhoto, M.F., 2007. Ordinal methodology in the analysis of Likert scales. Qual. Quantity 41 (5), 601–626. http://dx.doi.org/10.1007/s11135007-9089-z. Guitart, D., Pickering, C., Byrne, J., 2012. Past results and future directions in urban community gardens research. Urban For. Urban Green. 11 (4), 364–373. Hara, Y., Takeuchi, K., Okubo, S., 2005. Urbanization linked with past agricultural landuse patterns in the urban fringe of a deltaic Asian mega-city: a case study in Bangkok. Landscape Urban Plann. 73 (1), 16–28. http://dx.doi.org/10.1016/j.landurbplan. 2004.07.002. Heckman, J.J., Vytlacil, E.J., 2007. Econometric evaluation of social programs, part I: causal models, structural models and econometric policy evaluation. Handb. Economet. 6, 4779–4874. http://dx.doi.org/10.1016/s1573-4412(07)06070-9. Heckman, J.J., 1991. Randomization and Social Policy Evaluation. NBER Techincal Working paper, pp. 107. Johnson, D.R., Creech, J.C., 1983. Ordinal measures in multiple indicator models: a simulation study of categorization error. Am. Sociol. Rev. 48 (3), 398–407. http://dx. doi.org/10.2307/2095231. Kim, Soeun, Oh, Jooseok, Kim, Seiyong, 2016. Research on residents’ preference for outdoor spaces to vitalize community gardens within joint housing complexes – with the focus on four joint housing complexes in Seoul. J. Urban Des. Inst. Korea Urban Des. 17 (3), 57–70. Le Roux, D.S., Ikin, K., Lindenmayer, D.B., Blanchard, W., Manning, A.D., Gibbons, P., 2014. Reduced availability of habitat structures in urban landscapes: implications for policy and practice. Landscape Urban Plann. 125, 57–64. http://dx.doi.org/10.1016/ j.landurbplan.2014.01.015. Lovell, S.T., 2010. Multifunctional urban agriculture for sustainable land use planning in the United States. Sustainability 2 (8), 2499–2522. Maas, C.J., Hox, J.J., 2004. Robustness issues in multilevel regression analysis. Stat. Neerlandica 58 (2), 127–137. http://dx.doi.org/10.1046/j. 0039-0402.2003. 00252.x. McClintock, N., 2010. Why farm the city? Theorizing urban agriculture through a lens of metabolic rift. Camb. J. Reg. Econ. Soc. rsq005. Ministry of Agriculture Food and Rural Affairs of Korea, 2013. The First Step Toward Fostering Urban Agriculture. Press release of MAFRA (June 2) Retrieved from http:// www.mafra.go.kr/list.jsp? & newsid=155444697§ion_id=b_sec_1 & pageNo =2 & year=2015 & listcnt=10 & board_kind=C & board_skin_id= C3 & depth=1 & division=B & group_id=3 & menu_id =1125 & reference=2 & parent_code=3 & popup_yn=N & tab_yn=N. Moskow, A., 1999. The Contribution of Urban Agriculture to Gardeners, Their Households and Surrounding Communities: The Case of Havana, Cuba. For Hunger-proof Cities Sustainable Urban Food Systems. International Development Research Centre, Ottawa, pp. 77–83. Mougeot, L.J., 2000. Urban Agriculture: Definition, Presence, Potentials and Risks. Growing Cities, Growing Food: Urban Agriculture on the Policy Agenda. pp. 1–42. National Law Information Center, 2016. Act on Development and Support of Urban Agriculture. Retrieved from. http://www.law.go.kr/eng/engLsSc.do?menuId=1& query=urban+agriculture&x=0&y=0#liBgcolor0. Norman, G., 2010. Likert scales, levels of measurement and the laws of statistics. Adv. Health Sci. Educ. 15 (5), 625–632. http://dx.doi.org/10.1007/s10459-010-9222-y. Nugent, R., 2000. The impact of urban agriculture on the household and local economies. In: Bakker, N., Dubbeling, M., Gündel, S., Sabel-Koshella, U., de Zeeuw, H. (Eds.), Growing Cities, Growing Food. Urban Agriculture on the Policy Agenda. Zentralstelle für Ernährung und Landwirtschaft (ZEL), Feldafing, Germany, pp. 67–95. Oh, J.S., Kim, S.Y., 2014. Study on the introduction and planning of urban agriculture within multiple housing complex – focused on the drawing planning index and ways to its application. J. Urban Des. Inst. Korea 15 (5), 47–66. Ostroff, C., 1992. The relationship between satisfaction, attitudes, and performance: an organizational level analysis. J. Appl. Psychol. 77 (6), 963–974. http://dx.doi.org/ 10.1037/0021-9010.77.6.963. Park, H.S., Yang, S.H., 2014. Review and Implication of Urban Agriculture of Seoul. Seoul Institute Retrieved from. https://opengov.seoul.go.kr/research /455563. Parra-López, C., Groot, J.C., Carmona-Torres, C., Rossing, W.A., 2009. An integrated approach for ex-ante evaluation of public policies for sustainable agriculture at landscape level. Land Use Policy 26 (4), 1020–1030. Roberts, B., Kanaley, T. (Eds.), 2006. Urbanization and Sustainability in Asia: Case Studies of Good Practice. Asian Development Bank, Manila, Philippines.

citizens. Such results echoed some previous researches including Deelstra and Girardet (2000), and highlighted the importance of women' participation to make UA more sustainable though what it means is not women's dominances in this field but some positive effects UA-related measures in Seoul have on female citizens. The following is a guideline for fine-tuning relevant policy measures based upon the aforementioned analysis and observations. First, it would be desirable for the municipal government to secure larger plots for UA activities by revising regulations on urban planning. Of around 60,000 box-type gardens that the Seoul municipal government had offered to citizens between 2011 and 2013, only 59 percent has been in usage to date amid the lack of public interests (Seoul Metropolitan Government, 2015b). Given that those involved in the incity and farm-park types, who managed a relatively larger size of land, were found to have shown a higher level of satisfaction with UA, securing more land for UA to help more people manage larger spaces would be one of crucial ways to promote the project further down the road. It is encouraging that the state added the urban agriculture park to one of city park categories in the National Land Planning and Utilization Act, to which the Act of UA belongs. In order to spur such movements further, more attention should be paid in amending rules on urban planning. Second, it is advisable to adjust the range of public supports for UA activities in a way to scale it back. True, the statistics showed that the state support for facilities and crops helped raise the satisfaction level of participants. But it would nothing but a metabolic rift if the public holds a key of the whole cycle of the UA management, just as McClintock (2010) pointed out. It is in this context that it would be necessary to come up with measures to have UA stand on its own feet by, for example, diversifying relevant education programs which were found to have some positive effects, so as to overcome limitations the government-led UA system has (Ellis and Sumberg, 1998). A slew of researches so far have emphasized the significance of policy measures in introducing and expanding UA. As there exists a clear limit in securing large-scale farmlands within a dense city, it is important to devise elaborate public plans to find places available for UA by making, for example, the most of idle plots in urban areas. (Aubry et al., 2012; Parra-López et al., 2009). In line with such previous researches, this paper looked into the current status of UA in South Korea and its capital city of Seoul, and the statistical results here also highlighted a key role of the state policy in the growth of UA. It is in this context that the proactive introduction of policy measures and the revision of existing ones in a way to encourage multi-functional land use would be imperative in improving users' satisfaction that ultimately leads to cities' sustainable development. Acknowledgment This study is funded by the NRF (National Research Foundation of Korea) (Project No: NRF-2012R1A2A2A01046490). References Aubry, C., Ramamonjisoa, J., Dabat, M.H., Rakotoarisoa, J., Rakotondraibe, J., Rabeharisoa, L., 2012. Urban agriculture and land use in cities: an approach with the multi-functionality and sustainability concepts in the case of Antananarivo (Madagascar). Land Use Policy 29 (2), 429–439. Bakker, N., Dubbeling, M., Gündel, S., Sabel Koschella, U., Zeeuw, H.D., 2000. Growing Cities, Growing Food: Urban Agriculture on the Policy Agenda. A Reader on Urban Agriculture. DSE. Baptiste, A.K., Foley, C., Smardon, R., 2015. Understanding urban neighborhood differences in willingness to implement green infrastructure measures: a case study of Syracuse, NY. Landscape Urban Plann. 136, 1–12. http://dx.doi.org/10.1016/j. landurbplan.2014.11.012. Bohn, K., Viljoen, A., 2011. The edible city: envisioning the continuous productive urban landscape (CPUL). Field 4 (1), 149–161. Boone, H.N., Boone, D.A., 2012. Analyzing likert data. J. Ext. 50 (2), 1–5. Brown, K.H., Jameton, A.L., 2000. Public health implications of urban agriculture. J. Public Health Policy 21 (1), 20–39. http://dx.doi.org/10.2307/3343472.

132

Land Use Policy 69 (2017) 123–133

J.-s. Oh, S.-y. Kim

United Nation. (2014, July 10). World’s population increasingly urban with more than half living in urban areas. United Nation. Retrieved from http://www.un.org/en/ development/desa/news/population/world-urbanization-prospects-2014.html. Viljoen, A., Bohn, K., 2009. Continuous productive urban landscape (CPUL): Essential infrastructure and edible ornament. Open House Int. 34 (2), 50–60. Willms, J.D., 1999. Basic concepts in hierarchical linear modeling with applications for policy analysis. Handb. Educ. Policy 473–493. http://dx.doi.org/10.1016/b978012174698-8/50046-1. Zezza, A., Tasciotti, L., 2010. Urban agriculture, poverty, and food security: empirical evidence from a sample of developing countries. Food Policy 35 (4), 265–273. http:// dx.doi.org/10.1016/j.foodpol.2010.04.007. Zumbo, B.D., Zimmerman, D.W., 1993. Is the selection of statistical methods governed by level of measurement? Can. Psychol./Psychol. Can. 34 (4), 390–400. http://dx.doi. org/10.1037/h0078865.

Seoul Metropolitan Government, 2014. Management Plan for Roof Top and Box Garden. Department of public economic affairs Retrieved from. http://agro.seoul.go.kr/ archives/4364. Seoul Metropolitan Government, 2015a. Masterplan of Seoul Urban Agriculture 2.0. Seoul Agriculture Center Retrieved from. http://opengov.seoul.go.kr/sanction/ 4829075. Seoul Metropolitan Government, 2015b. Introduction of Urban Agriculture with Seoul Citizen. Seoul Agriculture Center. Retrieved from. http://opengov.seoul.go.kr/ sanction/1494957. Steudler, D., Rajabifard, A., Williamson, I.P., 2004. Evaluation of land administration systems. Land Use Policy 21 (4), 371–380. http://dx.doi.org/10.1016/j.landusepol. 2003.05.001. Travaline, K., Hunold, C., 2010. Urban agriculture and ecological citizenship in Philadelphia. Local Environ. 15 (6), 581–590.

133