1
Analysing the effects of declining local food
stores on sense of place of the elderly: the case of
Reduzum
Bachelor’s Thesis June 17th, 2019 Anne Altenburg 11326042 Human GeographyUniversiteit van Amsterdam Supervisor: Rowan Arundel
2
Abstract
There has been a significant change in the retail landscape of food stores in the Netherlands. Especially a lot of local food stores have closed in the rural areas of the province of Friesland. This study first investigates the change in accessibility to food stores for bicycle and walking between 2002 and 2019 for the province of Friesland by using GIS methods. A network analysis was conducted for 15 minutes walking and cycling. The results reflect the trend of food stores closing in the smaller rural villages while the amount has increased in the larger villages and cities. Most loss of access to food stores occurred in the north-easterly and western regions of the province, which are also shrinking regions or anticipated shrinking regions.
A food store can also have a social function which can be especially important to the (less mobile) elderly. To research the changes in shopping routines related to grocery shopping and the change in sense of place of the elderly inhabitants as a result of the closure of the local supermarket, the study zooms in on Reduzum. This has become a village with no direct access to a food store within 15 minutes walking or cycling. A total of thirteen elderly inhabitants were interviewed on change in time space routines, what role the supermarket had for them and the village and elements of sense of place. Results showed that the supermarket had an important social role as a meeting place which was thought to be especially important to the elderly inhabitants. It was also a place where one was updated on the news in the village which contributed to a sense of community. The closure of the store was experienced as very regrettable. The respondents had to change their time space routines in the sense that they had to travel further and that they planned further ahead for their groceries. For the respondents with a car the transition was more easily made than for the respondent without car who now received help from family and neighbours in accessing food stores. Online grocery shopping was also used. The feelings of attachment regarding the closed supermarket were based on the loss of convenience and more importantly on the social function the supermarket had for them and the community as a whole. The closure did not have a large influence on the feelings of attachment related to the village as a whole since these were mostly based on the already existing sense of community.
3
Contents
1. Introduction………...4-5 2. Theory………...6-9 2.1 Sense of place……….7 2.1.1 Place attachment……….7-8 2.1.2 Place dependence………...8-9 3. Methodology……… 10-16 3.1 Case study description………103.2 Quantitative research methods………...11-14 3.2.1 Data……….12-13 3.2.2 GIS methods………13-14 3.3 Qualitative research methods……….14-16 3.3.1 Sampling methods………...15-16 4. Results………...17-29 4.1 GIS analysis………....17-24 4.2 Zooming in on Reduzum………25-29 4.2.1 Changing routines………26
4.2.2 The (social) role of the supermarket………27
4.2.3 Additional consequences……….28
4.2.4 Place Dependence………....28
4.2.5 Ageing in place………29
5. Conclusion……….30
6. Discussion & acknowledgements………..31
7. References………..32-35
List of figures
Figure 1: Concept map ………...9Figure 2: Map of Reduzum and surroundings………...10
Figure 3: Map showing the change in food stores in Friesland between 2002 and 2019…...17
Figure 4: Zoom-in of the previous map showing the change in food stores between 2002 and 2019 in the area of Reduzum………...18
Figure 5: Map showing the change in service areas for bicycle in the area of Reduzum comparing 2010 and 2015………19
Figure 6, 7, 8: Maps showing the loss and gain of service areas of food stores for bicycle comparing different years………20
Figure 9, 10, 11: Maps showing the loss and gain of service areas of food stores for bicycle comparing different years………..21
Figure 12: Map comparing population density with the loss of service areas of food stores for bicycle between 2002 and 2019………...22
Figure 13: Map comparing percentage of inhabitants aged 65 and older per district and loss of service areas of food stores for bicycle comparing 2002 and 2019………..23
Figure 14: Map comparing shrinking regions and anticipated shrinking regions with loss of service areas of food stores for bicycle comparing 2002 and 2019………..24
4
1. Introduction
In the age of globalization, urbanization and economies of scale, there has been a significant change in retail facilities (Paddison & Calderwood, 2017). Economies of scale have been a contributor to the disappearance of many local small-scale food stores which have been replaced by large chain stores (Clarke & Banga, 2010; Paddison & Calderwood, 2017). Other contributors are population decline and aging populations which are increasingly putting pressure on local retail facilities in some regions of the Netherlands (Fries Sociaal Planbureau, 2017). Population decline and an aging population mainly occur in rural municipalities and are associated with decline of public support for different types of facilities and services and a decreasing liveability of the area (Haartsen & Venhorst, 2010; Ritsema van Eck et al., 2013). Decreasing numbers of retail facilities have therefore especially been a concern for rural areas in the Netherlands (Paddison & Calderwood, 2007; Christiaanse & Haartsen, 2017). This thesis will only focus on food stores since they provide local residents with daily necessities and are thus traditionally part of their daily routine more than other stores. Additionally, they can be places of social interaction and provide a sense of community. Food stores here are mainly understood as supermarkets and stores registered under food commodities in the Chamber of Commerce.
This thesis particularly focuses on the north-easterly province of Friesland in the Netherlands. Friesland is one of the more rural provinces in the Netherlands wherein many of its regions face population decline driven by an aging population and young people moving to more urbanized places (Fries Sociaal Planbureau, 2017). Currently one in five people in Friesland is over the age of 65 (which is above the national average of 18%) and this number will increase to one in four for the year 2030 (ibid.). Also in this province, local food stores have been disappearing over the years. Between the years 2013 and 2018, Friesland has seen the highest decline in supermarkets of all provinces (Distrifood Dynamics, 2019). Furthermore, the average distance to a supermarket is also highest in Friesland at 1.4 kilometres (CBS, 2019). The travel distances to supermarkets can be over twice as high for people living in rural areas compared to urban areas and this can be even more in shrinking regions (CBS, 2019; KpVV, 2015; Steenbekkers & Vermeij, 2013). Even though mobility has increased over the years and the overall accessibility in the Netherlands is high due to a well-developed road network, the closing of these (local) supermarkets can still be a problem for people without motorized transport like the elderly (Christiaanse & Haartsen, 2017; Fries Sociaal Planbureau, 2017). Supermarkets can also be important for them because of secondary purposes like a post office or a pharmacy (Amcoff et al., 2011). Given this, the research area will be the province of Friesland as the pressure on local food supply is highest here resulting in the most decline and decreasing accessibility to the food stores for the elderly.
Besides their practical function for groceries, local retail facilities like food stores can also function as social hubs where informal social interaction takes place which contributes to a sense of community (Jorgensen, 2010; Paddison & Calderwood, 2007; Christiaanse & Haartsen, 2017; Clarke & Banga, 2010). This social function can to be particularly important for less mobile people like the elderly and closing of retail facilities can therefore result in a feeling of loss and less social contact (Gardner, 2011; Stanley et al., 2011). Additionally, local food stores can contribute to a sense of place where people attach feelings of connectedness to their environment (Christiaanse & Haartsen, 2017). Sense of place refers to the meanings people have attached to a place (Cresswell, 2009). In
5 this thesis, the focus will only be on the emotional aspect (place attachment) and the behavioural aspect (place dependence) of the concept sense of place. These concepts which will be further introduced in the theory section.
Christiaanse and Haartsen (2017) conducted research on this social aspect in their paper which looked at the reactions to the closure of a local supermarket in the province of Groningen and what meaning that supermarket had for the respondents. They suggested that further research should be conducted on the accessibility of facilities for the elderly. This thesis will focus both on accessibility and the influence of closure on the elderly and their grocery shopping routines and sense of place.
This research will consist of a mixed methods approach in which first change in accessibility to food stores will be analysed by using Geographical Information Systems (GIS). The second part will focus on the social impact of the closure on the elderly living in one of the identified areas with low access, namely the village Reduzum, by conducting interviews. As a result, this thesis has two main questions since there are two different levels of analysis which cannot be combined into one sentence. Prior to answering these, an overview will be given on the related theory and the methods used. Finally, the results will be presented and a conclusion will be given to answer the following research questions :
How is change in accessibility to food stores spatially distributed in Friesland?
How has the closure of the supermarket in Reduzum influenced grocery-shopping routines and the sense of place of the elderly inhabitants?
6
2. Theory
The closing of (rural) food stores has proven to be especially problematic for the less mobile elderly in rural areas who do not own a car (Amcoff et al., 2011). People with a car are able to maintain social networks farther from home, but less mobile elderly are more dependent on local social networks and meeting points (Gray et al., 2006). Their so-called activity spaces are thus more constrained than those of car drivers. An activity space is the space in which most activities of an individual take place (Holloway & Hubbard, 2001). In general, most activities are carried out in a local structure of movements with some activities further away. Elderly leave the house less often when they get older and the share of outdoor activities related to getting groceries and personal care increases (Harms, 2008). The movements further away (to e.g. family members) are more difficult for the elderly without motorized transport and they might therefore rely on public transport or other people. Mobility is therefore not only important for accessibility, but also for social relations.
Research conducted by Mollenkopf et al. (2004), which also included the Netherlands, showed that amongst the elderly, walking was the most popular mode of transport, followed by car and bicycle and lastly public transport. However, car use amongst the elderly has increased over the years as they are able to drive independently longer at a higher age (Harms, 2008). Furthermore, there is a significant difference in mobility between men and women where men are more mobile than women (ibid.). Also amongst the elderly, more men drive a car than women. In addition, the percentage of elderly driving as passenger also increases when age gets higher.
Another trend is that of the e-bike. Study shows that mostly elderly people use the e-bike in the Netherlands (Ministerie van Infrastructuur en Milieu, 2015). The same study also showed that elderly are cycling more and for longer distances.
Ageing in place
Disappearing retail facilities can be a reason for elderly to move to larger villages or cities where there are enough facilities (Steenbekkers & Vermeij, 2013; Fries Sociaal Planbureau, 2017). This undermines the ability for people to age in place, although it is often what they prefer (Morley, 2012). Here the definition of aging in place of Morley (2012) will be used and it refers to “the ability to live in one’s own home and community safely and independently as one ages” (p. 489). When a facility closes, alternative support may be needed from neighbours and family members to enable the elderly to age in place (ibid.). This is also true for food stores. Neighbours and family members may need to drive them to the store of need to deliver the groceries at home (Fries Sociaal Planbureau, 2017).
The social function of food stores
As mentioned in the introduction, local retail facilities like food stores can function as social hubs where informal social interaction takes place (Jorgensen, 2010; Paddison & Calderwood, 2007; Christiaanse & Haartsen, 2017; Clarke & Banga, 2010). This social function can to be particularly important for less mobile people like the elderly and is also important for enhancing the sense of community and social interaction (Gardner, 2011; Paddison & Calderwood, 2007; Stanley et al., 2011). Moreover, results of interviews of Amcoff et al. (2011) indicated that the social function of a
7 food store was often more important than its function for food supply. They further note that especially elderly go to a local store to meet others and therefore closure can lead to a lost opportunity to interact with people outside a small group of neighbours.
2.1 Sense of place
Local stores, like food stores, can contribute to a sense of place. Sense of place has received a lot of attention in literature, but there is still a lot of ambiguity regarding the concept. Here, sense of place is defined as emotive bonds and meanings people attach to a place and it consists of the dimensions place attachment (emotional connection to place), place identity (beliefs about the relationship between self and place) and place dependence (degree of a particular place to satisfy the needs of a person compared to other places) (Jorgensen, 2010; Christiaanse & Haartsen, 2017; Boerebach, 2012; Foote & Azaryahu, 2009). These attachments are not always strong or positive. Here, only place attachment and place dependence will be further elaborated upon as food stores are less likely to be connected to ones’ identity.
The emotions associated with place are not always consciously experienced but changes in place dynamics can trigger them to emerge (Foote & Azaryahu, 2009). Furthermore, sense of place can also be shared by a group of people like a community (ibid.). The concept of sense of place is mostly used here as an overarching term in which other concepts are combined.
2.1.1 Place attachment
In this research, the definition of place attachment of Brown et al. (2012) will be used: “Place attachments are the positive bonds people form with places, arising from affective, behavioural, and cognitive ties between individuals or groups and their socio-physical settings.”. It can grow when people use and invest time and meaning in places. Places are understood as “a meaningful site that combines location, locale, and sense of place.” (Cresswell, 2009). The attachment to place can also be influenced by personal characteristics like age and gender (Hidalgo & Hernández, 2001). In order to explain place attachment in a more founded way, the following three (interrelated) concepts will be discussed: social interaction, symbolic meaning and sense of community.
Social interaction
Social interaction is an important factor related to place attachment as it can contribute to the process of positive bonding and feeling more connected to a place. Meaning is often given to places through interaction (Milligan, 2011). The place shapes, constrains and influences social interaction. Frequent interaction in certain places, like local food stores, often results in place attachment (ibid.). Local food stores can be one of the few public spaces in small villages where people can spontaneously meet and interact with each other (Bovenhoff & Meier, 2015). This function was found to be especially relevant to the elderly. The closure of a food store can disrupt feelings of attachment and can result in feelings of loss (Milligan, 2011; Christiaanse & Haartsen, 2017).
Symbolic meaning
Village stores can have a symbolic meaning to the community (Amcoff et all., 2011). This symbolic meaning is strongly linked with the other concepts related to place attachment as it is also a result of those processes. When a rural facility closes, this can be upsetting because the facility can have an
8 attached symbolic meaning to various people, even though they did not make use of it (Kyle et al., 2004). To illustrate, a local food store in the Friesian town Sloten was going to close due to bankruptcy. Inhabitant were requested to shop there more often in order to save the store, but it did not work. However, when it was announced that the store was going to close, inhabitants were shocked and upset even though they did not buy their groceries there ("Plaatselijk Belang: "Supermarkt in Sloten hoeft niet dicht"", 2014; "Supermarkt Sloten sluit in januari haar deuren", 2014). These reactions can be connected to the general belief that a village should have a supermarket and that it contributed to a certain status of the village (Christiaanse & Haartsen, 2017). Hence, the symbolic meaning of a supermarket in a village does not only have to be based on its function for grocery shopping, but can also be based on the symbolic meaning for the village and, to a smaller degree, as a social meeting place (ibid.). According to Farmer et al. (2012), the symbolic meaning can be based on the cultural, economic or social value the feature has for the community. Sense of community
As illustrated by the symbolic meaning, a local food store can not only have a personal meaning, but also a meaning for the community as a whole. The definition that is frequently used to describe sense of community is that of McMillan & Chavis (1986) who define sense of community as "a feeling that members have of belonging, a feeling that members matter to one another and to the group, and a shared faith that members' needs will be met through their commitment to be together.". Public places that foster informal social interaction can contribute to the sense of belonging to a community (Christiaanse & Haartsen, 2017).
The presence of stores can also encourage people to take a walk to them which enlarges the chances of them encountering people with whom they can interact (Lund, 2002; Kim & Kaplan, 2004; Bovenhoff & Meier, 2015). This is beneficial for the sense of community (Lund, 2002). Local food stores are also a place where one can be updated on news in the village. For example, one of the reviews of a local food store in the village IJsbrechtum said that the person appreciated the store because the owner always updated everyone on the latest news. When people are up to date on what is going on in the community, they can feel more included which can contribute to a sense of community (Bovenhoff & Meier, 2015).
2.1.2 Place dependence
Place dependence and place attachment share ground on some levels as place attachment also includes the behavioural aspect. However, the difference is that place dependence looks at the degree to which a place can satisfy the needs relative to other places (Brown et al., 2012). It accompanies two elements: the quality of the place regarding availability of social and physical resources and how it compares to other places (Pretty et al., 2003). People do not always consciously compare places and their ability to satisfy their needs, but certain events can increase awareness (ibid.). A food store closing could be one of those events.
Time space routines
Hägerstrand is the leading researcher on time-space routines. Time-space routines refer to the notion that “individuals routinely and repeatedly draw upon resources of time and space in the conduct of
9 their everyday lives.” (Holloway & Hubbard, 2001). These routines can also be mapped out and show the spatial extent of movements of an individual over a certain period of time. This concept is not directly part of place dependence but it can help to describe how shopping routines regarding getting groceries have changed as a result of the closure of the food store. This in turn can influence the place dependence as inhabitants needed to make use of another food store. This store can be different in quality of social and physical resources.
10
3. Methodology
3.1 Case study description
The case study area for the qualitative research is the village Reduzum (Roordahuizum in dutch). It is located in the municipality Leeuwarden and counted 1100 inhabitants in 2018 (“Informatie woonplaats Reduzum”, n.d.). The village entered a period of population decline and decay in the 80’s of the 20th century. Inhabitants then started initiatives which focussed on making the village more attractive. For example, without help of the municipality, they managed to build and entirely new neighbourhood and expand the school (“het success van Reduzum”, n.d.). This indicates that there is a strong active community which is committed to maintain the liveability of their village.
Since then, the number of inhabitants has increased. Now, there are a lot of sports activities and many other activities in which inhabitants participate. Although there is an active community, the amount of facilities has declined over the years. The only remaining food provisioning services left in Reduzum are the butcher the mobile greengrocer which delivers fresh vegetables and fruit once a week. There used to be a lot more food stores and the village even counted three bakeries at one point. Between 2010 and 2015, the local supermarket closed its doors. One of the inhabitants saved the supermarket, but it was not feasible and the store permanently closed within two years. The nearest supermarkets are in Mantgum and Grou and can be accessed by approximately 20 minutes cycling or a ten minute drive. These supermarkets are larger chain supermarkets and were already used by inhabitants before the closure because of the larger product range.
11 Research strategies
This research consists of multiple levels of analysis where the quantitative analysis focusses on the entire province of Friesland and the qualitative analysis focusses on the village Reduzum. By using a mixed methods approach, it is possible to offset weaknesses and combine strengths of both quantitative and qualitative research methods (Bryman, 2016). Firstly, the ecological validity is better ensured by including qualitative research which means that the research results are more likely to reflect the real world (Bryman, 2016). The qualitative research provides more depth and exploratory research on what the findings of the GIS actually mean for the elderly inhabitants. By only analysing accessibility changes due to decline in food stores, the social impact and personal experiences of the inhabitants would not be taken into account. Moreover, better policy recommendations can be made by incorporating qualitative research.
On the other hand, qualitative research is often said to be subjective and interpretative and by including the quantitative GIS analysis, objective data is incorporated which allows for a more comprehensive description of the area of inquiry and how it relates to the broader spatial context (Bryman, 2006).
The analysis is based on literature and in this sense deductive, but it also accompanies an inductive element as it is not clear how decline in food stores changed the accessibility to them in Friesland and how that impacted the local elderly inhabitants. The research questions are:
How is change in accessibility to food stores spatially distributed in Friesland?
How has the closure of the supermarket in Reduzum influenced grocery-shopping routines and the sense of place of the elderly inhabitants?
Sub-questions have been formulated bellow for both the quantitative and qualitative section.
3.2 Quantitative research methods
For the quantitative analysis GIS tools are used in ArcMap 10.5.1 to analyse change in accessibility to food stores in Friesland. The research units in this section are the food stores. The corresponding sub-questions of this section are :
- How did the spatial distribution of local access to food stores change in Friesland between 2002 and 2019 for walking and cycling?
- What are the population characteristics of the areas that experienced a decrease in access? GIS methods are commonly used to measure access to facilities, like food stores (Charreire et al., 2010; Van Hoesen et al., 2016). There are many ways to approach an accessibility analysis and most studies made use of a buffer surrounding the food store (Charreire et al., 2010). This is a helpful way of identifying areas which lie outside a pre-determined absolute distance from the store, but does not account for the infrastructure people actually need to use to get there (Larsen & Gilliland, 2008; McEntee & Agyeman, 2010). The method that is used here to is to create service areas based on a network analysis which makes use of the street network from June 2018 provided by Waterschappen. This year was deliberately chosen because errors would be less likely compared to the newest data since it was already frequently corrected. The network analysis is conducted for walking and bicycle since these are the two most important modes of transport for elderly who do not have access to motorized transport.
12 In order to research accessibility to food stores, maximum travel times need to be chosen. A commonly used acceptable distance from food stores is 1000 meters which corresponds to approximately 15 minutes of walking (Apparicio et al., 2007). Distances over 1000 meters are not perceived as walking distance by the Dutch Central Bureau of Statistics (CBS). The other time-frame for bicycle is also set at 15 minutes to be able to show the difference in distance that can be crossed for the modes of transport. The bicycle has not been included often in literature as a mode of transport in the analysis. However, as mentioned in the theory, the bicycle is still a frequently used mode of transport amongst the elderly, especially in the Netherlands (Mollenkopf et al.,2004; Harms, 2008).
A limitation of this GIS analysis method is that it only looks at availability of the food store and accessibility to it and does not incorporate characteristics regarding quality and costs of food and specific supply of different services, as well as the characteristics of the store itself (e.g. personnel and athmosphere) (Penchansky & Thomas, 1981; McEntee & Agyeman, 2010). However, most of these characteristics can become apparent by the qualitative research section which is an important reason for including this type of research.
3.2.1 Data
The quantitative analyses is based on a dataset provided by Distrifood Dynamics. This dataset includes all major supermarkets and small-scale independent food stores registered under supermarket/food commodities in the Chamber of Commerce (Kamer van Koophandel). The dataset includes the years 2002, 2006, 2010, 2015 and 2019. These stores are geocoded by using a six digit postcode found on the ArcGIS online database. The camping stores are deleted as people outside the camping are not allowed to make use of it. In addition, also the driving stores are deleted because their movement is not reflected in the single address and including these would lead to incorrect representation.
A dataset containing the road network provided by the Waterschappen 2018 was used for the network analysis. A shortcoming is that the stores had to be snapped to the nearest road in order to conduct a network analysis. As a result the locations are slightly different than the addresses. The stores also did not snap to the exact same place in all years, this however only led to a difference of a couple of meters in most cases (<5 meters).
In addition, postcodes have likely changed over the years. Forexample, some municipalities have merged into new municipalities (Súdwest-Fryslân and de Fryske Marren). These changes are not accounted for. Another shortcoming is the fact that not all addresses of the stores are correct. In the earlier years, stores had to send their address themselves and these were not checked, also not on spelling. This, at least, applies for the year 2002. Distrifood Dynamics later started automatically checking the locations via google maps.
To compare the results with population characteristics, data from the CBS and Topografische Dienst Kadaster was used from 2018. It was not possible to calculate differences in population characteristics through the years because the borders of districts and municipalities were different. Therefore the most recent data of 2018 was chosen to be incorporated.
13 3.2.2 GIS methods
After the geocoding of all food stores, the food stores were snapped to the road network in order to be able to conduct a network analysis. First, data of the road network of geofabrik was used. However, this network turned out to have too much errors which turned out to be too difficult to fix. Eventually, the road network data of the Waterschappen of June 2018 was used. This dataset is less detailed than the one of geofabrik but also less complicated and without errors. Two things were needed to build the network dataset: the length of the road segment and the amount of minutes it took to cross it. The length in meters was calculated with calculate geometry. The following formula was used to calculate the minutes for walking: (length of road segment/1.4)/60. The average speed of walking is 5 kilometre per hour which translates into 1.4 meter per second. It has to be divided by 60 to get minutes. The formula for minutes of cycling was (length of road segment/5)/60. The average speed for cycling is 18 kilometre per hour which translates into five meter per second.
The service areas for 15 minutes walking and cycling were created for all years with the network analyst toolbar in ArcMap. The following years were chosen to visualize change in service areas over the years: 2002, 2010 and 2019. These years were selected because comparing all years would not add value and would result in too many maps and this comparison already gives a good overview of change over the years. Furthermore, these years almost have the same time period between them. The years 2002 and 2010, 2002 and 2019 and 2010 and 2019 have been compared on gain and loss in service areas. In order to do this, loss, gain and overlap had to be identified with the help of the tool ‘union’. To identify loss a selection was made in the attribute table for which the FID of, for example, 2002 was >=0 and the FID of 2019 -1. For gain it would be the other way around and for overlap both FID were >=0. The resulting files for overlap were too large for ArcMap to draw, especially in combination with the shapefile for loss and gain. The dissolve tool to reduce the amount of polygons to one also did not work. As a result, only loss and gain in service areas are visualized. No calculations have been made with this data because these numbers would be incorrect due to the shortcomings of the data mentioned by the data section.
Table 1.
Quantitative concept operationalization
Concept Dimension Indicator Variable
Decline in food stores
Amount of food stores
Decline in access to food stores by foot
Decline in access to food stores by bicycle
Amount of food stores present in the Distrifood Dataset
Not being within a 15 minutes travel distance by foot to a food store using the road network
Not being within a 15 minutes travel distance by bicycle to a food store using
14 the road network
Elderly
Food stores
Population Age
People with age 70 and older All major supermarkets and small-scale independent food stores registered under supermarket or food commodities (levensmiddelen) in the Chamber of Commerce
3.3 Qualitative research methods
In the second part, the spatial is narrowed to the village Reduzum. This village has been selected because of the closure of a local supermarket between 2010 and 2015. As a result, there is no supermarket accessible within a 15 minute walking or cycling distance. In addition, this was one of the more recent cases of closure which makes it more likely for elderly to recall memories and experiences related to the closure. Lastly, this village was also selected due to convenience reasons because it was one of only areas which experienced relatively recent closure and was accessible by bicycle from my parents’ house in Friesland.
The research units of this section will be the elderly inhabitants of Reduzum. The sub- questions related to the qualitative part of the research are:
- How have time-space routines of the elderly respondents changed, specifically relating to everyday shopping needs?
- What role did the supermarket fulfil for both the respondents and the village?
- What are the place attachments related to the supermarket and what are they based on?
These questions will be answered by conducting semi-structured interviews. This type of interview can make use of an interview guide which can contain some important themes and optional questions (Bryman, 2016). This gives the interview some structure while still leaving room for follow-up questions and improvisation. Themes that were discussed are changing grocery shopping routines, the role of the supermarket, the social function of the supermarket, contribution to the sense of community and experiences related to the closure. Other themes have also passed in conversation.
15
3.3.1 Sampling methods
In total 13 people have been interviewed of which: one group interview of ten people (with ages starting somewhere in the 70’s till 90), an interview over the phone with a woman of 77, an interview with a woman of 70 without a car and an interview with the local butcher. There is a strong overrepresentation of women in the sample and most respondents or their partner still drove a car. All respondents have lived in Reduzum for a long time or all their lives. Most interviews have been held in their native tongue, namely Frisian.
For sampling, non-probability sampling methods were used. This method does not involve random selection and examples of non-probability sampling are convenience sampling and snowball sampling (Bryman, 2016). These two sampling methods are both used in this research. Convenience sampling refers to respondents who are simply available to the researcher. A critique is that respondents are likely not representative of the group as a whole (Bryman, 2016). Therefore, findings can often not be generated.
However, convenience sampling is still often used in social research because probability sampling often involves a lot of preparation and time which is often not feasible within the time frame and resources of the research (ibid.). Convenience sampling could also still provide a base for further research and links with existing findings could be made.
The snowball sampling method refers to new respondents being found via already participating respondents (Bryman, 2016). Snowball sampling is to some extent a form of convenience sampling, but is different because the researcher has first made contact with people who are relevant to the research and thus based on probability sampling. An advantage of this is that people are more likely to be open and trustful towards the researcher.
In this research, snowball sampling was not used in this exact form. The respondents of the group interview were found via other people who were not participating themselves. I made use of the network connections of my father to get into touch with inhabitants and from there on I was forwarded to two different people before the last person invited me over for coffee with the group of ten elderly people. It is therefore a mix of convenience and snowball sampling.
Two other respondents have been found as a result of advertising. I was allowed to post a request for respondents on the Facebook page and website of Reduzum. In this request I shortly described my research and the target group for interviews, namely 70+ and preferably without a car. This is a form of purposive sampling in which research units are selected according to the research questions (Bryman, 2016). Two people responded to this, one interview was over the telephone and the other one was at their home. The telephone conversation was however not really in-depth and the respondent still drove a car.
As a result of the interview on the telephone, the butcher was also selected as an interesting respondent as he also sells other basic products and bread from a bakery. The respondent mentioned that there is no supermarket in Reduzum anymore, but they still have the butcher who also sells a lot of other products. It is a sort of small supermarket and also offers to deliver groceries to people’s homes. The butcher was therefore also interviewed to research, amongst other things, the role of the
16 butcher in the village, what people made use of it, why they chose to also sell other products and whether they already did this before the supermarket closed.
Limitations
The group interview was useful in getting a good indication of the case study area and it also provided the opportunity for discussion. However, it constrains the opportunity to go into depth on personal experiences. Furthermore, the respondents of the group interview all drove a car, or their partner did. However, the license of some people would be withdrawn within the next years. Based on the theory and research questions, these people would be less relevant to interview, but it does provide for an interesting comparison between elderly with motorized transport and elderly without. In addition, they had already thought about how they would acquire their groceries when they could not drive anymore. This is also interesting to predict future trends in changing routines.
Additionally, the respondents said that there were hardly any elderly in the village without a car and the people who did not have a car had all coped with the closure in a similar way. Therefore, although people without a car are underrepresented in the sample, it is still likely that the results, at least partly, apply to their situation as well.
17
4. Results
4.1 GIS analysis
The first part of the GIS analysis focusses on the sub-question: “How did the spatial distribution of local access to food stores change in Friesland between 2002 and 2019 for walking and cycling?” The amount of food stores and their locations have significantly changed as can be seen in Figure 3. The larger villages and cities can be easily identified according to the number of dots clustered together. A trend that can be observed based on this map is that many food stores in the smaller villages have disappeared while the amount has increased in the larger villages and cities. The major cities refer to Leeuwarden, Sneek, Drachten and Heerenveen. It should be noted that some overlapping dots can be the same store which has been snapped to a slightly different location on the road network. These then falsely suggest closure.
18 Figure 4. Zoom-in of the previous map showing the change in food stores between 2002 and 2019 in the area of Reduzum. The same legend applies.
The zoomed-in version shows that multiple food stores surrounding Reduzum have closed over the years, including the one in Reduzum itself. As a result, it has become a village which lies outside the service areas of the nearest food store in just five years’ time (Figure 5 on the next page). Based on these maps, Reduzum was selected as an area of interest.
19
Figure 5. Map showing the change in service areas for bicycle in the area of Reduzum comparing 2010 and 2015.
The following maps show the change in service areas for bicycle and walking in terms of loss and gain for the years 2002, 2010 and 2019. The maps have here been cropped to the same page to make it easier to compare them and observe differences. A file with all full-size maps can be requested by mail. This thus applies to all maps presented in this thesis. When comparing the maps 2002-2010 & 2010-2019 most loss and gain has occurred between 2002 and 2010. In general, there has been more loss of service areas than gain. This means that increasingly less areas can access a food store within 15 minutes cycling or walking.
20
Figures 6, 7 and 8. Maps showing the loss and gain of service areas of food stores for bicycle comparing different years.
21
Figures 9, 10 and 11. Maps showing the loss and gain of service areas of food stores for bicycle comparing different years.
According to these maps, gain has occurred relatively random although it seems that some gain was mostly in the area of the major cities. Most loss has occurred in the north-easertly and western regions of the province when looked at the spatial distribution of loss. If this correlates with some population characteristics will be looked at in the next part.
22 To answer the second subquestion “What are the population characteristics of the areas that experienced a decrease in access?” the population characteristics from the CBS aggregated in districts (wijken) were included. The first map shows the population density with the loss of service areas for bicycle comparing 2002 and 2019 as overlay. This layer had been chosen because service areas for bicycle are most visible and the comparison between 2002 and 2019 is more comprehensive. It can be observed that loss did not occur in the most densely populated areas. Overall, Friesland is not very densely populated. Loss appears to have mostly occurred in the less densely populated areas and also sometimes in the areas surrounding more densely populated areas. Also, there is less loss surrounding the city areas.
Figure 12. Map comparing population density with the loss of service areas of food stores for bicycle between 2002 and 2019.
23 For the next map, the percentage of people aged 65 and older have been displayed together with the loss of service areas because the share of elderly is often an indicator of shrink (Haartsen & Venhorst, 2010). There does not seem to be a clear overlap between the share of elderly and locations with loss of service areas. In general, however, the share of elderly per district is relatively high with 10-20% and 20-30% of the inhabitants being 65 or older being the most frequent. Reduzum also falls within an area with one in five to one in roughly three people being 65 and older.
Figure 13. Map comparing percentage of inhabitants aged 65 and older per district and loss of service areas of food stores for bicycle comparing 2002 and 2019.
24 Lastly, the loss of service areas for bicycle comparing 2002 and 2019 were compared to the shrinking regions and anticipated shrinking regions identified by the government (Erfgoedmonitor, n.d.). The map shows that most loss has occurred within shrinking regions or anticipated shrinking regions. According to a factsheet on the shrinking regions and anticipated shrinking regions in Friesland and Groningen of 2016 (bron) the people experienced that the liveability had decreased in those areas. Closing facilities can indeed lead to a declined liveability (Haartsen & Venhorst, 2010; Ritsema van Eck et al., 2013). These results could be interesting for further research on the closing of food stores in shrinking regions and liveability.
Figure 14. Map comparing shrinking regions and anticipated shrinking regions with loss of service areas of food stores for bicycle comparing 2002 and 2019.
25
4.2 Zooming in on Reduzum
The butcher was interviewed as an additional person of interest. They provide a diverse range of additional (mostly shelf life) products. They also provide delivery service of products from the store, but this service is hardly used anymore. Furthermore, the customers were of all ages and there was not a particular group overrepresented.
When the supermarket closed, they adapted their range of products and included bread and some other products. The butcher perceived its role in the village are purely food provisioning and not as a social hub. This used to be different when his mother ran the store because then people talked more and people exchanged more stories and news. After his mother left the store, some regular customers stopped coming. He mentioned that the owner of a store is important to customers for deciding whether or not they are going to shop somewhere.
Nowadays, most customers come for the by-products instead of the meat. This also became apparent by the other interviews. The people who came there only went there occasionally and for small purchases. The sales have dropped and the butcher did not see a bright future of the store. They cannot compete with the large chain supermarkets where most inhabitants shop. This is a trend he also observed in other villages. If the store closes, this would mean that the last food store in Reduzum would disappear.
It also became apparent in the interviews that a lot of inhabitants also still make use of the mobile greengrocer who delivers fresh fruit and vegetables one a week. This is also convenient in the sense that these products can be heavy and the respondents did not have to buy these in the supermarket and carry them on the bicycle. However, they also expect this service to disappear as soon as he turns 65 because he does not have a successor. This would then be one of the many mobile stores that have already disappeared in the village.
Most respondents frequently used the supermarket in Reduzum. Since it was not a big store, they still sometimes needed to go to the larger supermarkets to get specific products. However, you could also ask the owner to order something and then he would get it.
In general, people knew the owner well. That was also one of the reasons why they liked going there. The respondents also all stressed the importance of the owner of a store in their decisions on whether or not they went there. They were also willing to travel further if they did not like the owner of the store more nearby. According to them this also plays a large role in the survival of a store. For example, when the store was taken over, people had already changed their routines to doing groceries in the neighbouring villages. “The new owner was a very sweet woman and she did her best, but she also interacted with the people differently” *.
*All quotes have been translated from either Dutch to English or Frisian to English. Consequently, some translation errors might have occurred. Original recordings can be provided on request.
26
4.2.1 Changing routines
After the store closed, an initiative was set up to arrange a bus to drive people to the supermarket who could not drive themselves. This initiative failed because of a lack of enthusiasm. The respondents of the group interview thought that this was because most neighbours and family members already fulfilled this function and it also constrained you in the sense that you had to comply to predetermined time frames. However, the respondent without a car still thought it would be nice to have a service like this.
Most respondents now went to the Jumbo in Grou for their groceries and also Mantgum was mentioned. Interestingly, according to google maps this is approximately 20 minutes cycling, but when I asked the respondents how long it was, they all said half an hour. Apparently, there is a difference between the objective and subjective distance. However, they did not seem to think that the distance was too long.
The respondents also told that there were few people in the village who did not own a car. These people got help from their children and neighbours who either brought them groceries or drove them to the food stores. The butcher also mentioned that an elderly inhabitant bought an e-bike to become more mobile. The respondent without a car also had to arrange different ways to get to the store. Once a week, her daughter picks her up to do groceries together and the neighbours also frequently offer to drive her. In cases where cycling is not an option and there is no opportunity of driving with somebody to a food store, public transport is still an important mode of transport. She also makes use of the delivery service of the Jumbo which she very much enjoyed as “it gives you a bit of independence”.
The respondents of the group interview also expected their children to help them when they could not drive anymore and were also planning to use the delivery service of the Jumbo. Even though they would like for a supermarket to return to Reduzum, they also acknowledged that it is not feasible anymore in modern times to own a small food store in a village. Therefore, online shopping was thought to be the future trend for elderly without motorized transport. These delivery vehicles were already spotted driving through the village on an almost daily basis. However, they also saw the down sides of this:
“Sometimes you just want to have the feeling of being in a supermarket, to see the products”. “It is also the social contact that you miss when ordering via the app”.
The closure of the store also caused a change in time-space routines. The respondents all adapted their grocery-shopping routines in the sense that they planned further ahead. One respondent, for example, mentioned that she went once every three to four weeks to Grou with the car to buy most groceries and did the smaller purchases by bicycle. When the supermarket in Reduzum was still open, they went to the supermarket more frequently and could also stop by when they forgot something. Some mentioned that they sometimes go to the butcher when they forgot something, but not often.
27
4.2.2 The (social) role of the supermarket
The respondents all found the closure of the store very regrettable. One respondent referred to it as “It was just as if a small part of the village had been cut off”. The supermarket did not only have a food provisioning role, but it also fulfilled a social role on both an individual level and for the village as a whole since it was a meeting point for inhabitants.
“We knew the owners well and each time I went there we had a great laugh. That is something special. You also met people there, for example, whom you had not seen in a while and then you would arrange to meet for coffee someday soon. That is something very important I think, also for elderly. Even if it is just a short conversation”.
The social function was also thought to be especially important to the elderly as illustrated by the following quotes.
“The younger inhabitants will not go to the supermarket for social interaction, they already have enough of that. They often just pick up their groceries on their way home from work”.
“I can imagine when you are older and you do not drive a car and there is no food store anymore, that you really miss that store. It is also a social happening.”
“It was especially bad for the elderly. For example, there was a woman who dropped of her mother in law, who was in a wheelchair, in front of the store. The neighbour was then waiting for her in the store and then they chit chatted for a bit and she did that with everyone. It was actually a social activity for her, to be among people again”.
These quotes all relate more to the social interaction element related to place attachment. It is in line with the theoretical framework where research indicated that the social function of a food store is especially important for the elderly (Gardner, 2011; Paddison & Calderwood, 2007; Stanley et al., 2011; Amcoff et al., 2011). This was also related to the smaller activity spaces of the elderly (Holloway & Hubbard, 2001). Furthermore, the social function of the store seemed to be almost equally important, if not more, as the food provisioning function. However, it could be that answers have been somewhat steered as a result of the research description or the questions.
The supermarket also had a function of social control or community care. For example, when a regular customer did not visit in a while, the owners would worry and ask if everything was alright. One respondent added that “It was nice to have a trusted environment”. This was also thought to contribute to the sense of community.
Something that was also thought to contribute to the sense of community was the role of the supermarket in spreading news. One respondent said:
“You went to store and, for example, heard that X was in the hospital and then you could immediately buy a card to send them. You were updated on all the news in the village”. This was affirmed by the other respondents. This role was something they valued and missed. The
connection with some other inhabitants was reduced, especially with the younger inhabitants.
28
4.2.3 Additional consequences of closure
The closing of the store also had influence on the amount of people walking on the streets. The respondents mentioned that after the store closed, less people were seen walking on the streets. One respondent also said that the village became less lively and liveable. This is also reflected in the theory where it was discussed that the presence of stores can encourage people to take a walk to them which in turn could be beneficial for social interaction and the sense of community (Lund, 2002; Kim & Kaplan, 2004). It also relates to a certain symbolic value for the village in the sense that the village now felt ‘emptier’. For example, one respondent mentioned that when the store closed, some younger inhabitants experienced it as a loss even though they hardly made use of it. This is in line with the story of the supermarket in Sloten and the research of Christiaanse & Haartsen (2017). Another consequence was that the flow shop next to the supermarket also had to close as a result of the closure of the supermarket. The closure had caused such a decrease in customers that it was not feasible anymore. One respondent mentioned that she always liked to walk through the flower shop on her way to the supermarket. When the supermarket closed, this route could no longer be made in combination. This illustrates that the presence of stores influences each other.
4.2.4 Place dependence
The comparison between supermarkets and their capability to satisfy the needs of respondents will be made between the supermarket in Reduzum and the Jumbo in Grou since most respondents no go to the Jumbo in Grou. These stores are compared on their ability to satisfy the needs of the respondents regarding availability of physical and social resources (Pretty et al., 2003).
The Jumbo is better in satisfying their needs regarding the availability of products. Since the store is bigger, it has a larger product range. However, when looked at the social resources, the supermarket in Reduzum was evaluated as better. The respondents also encountered people in the Jumbo in Grou with whom they had a chat, but this was different compared to the supermarket in Reduzum. There the social contact was more meaningful and they talked longer. Furthermore, as previously mentioned, the supermarket in Reduzum was also the place where one was updated on all the news in the village. However, an advantage of the Jumbo was that they encountered more people from other villages and cities. This can be beneficial in expanding their social network.
Overall, the respondents with car had adapted quite easily to the store closing because they saw it coming and the other stores where easily accessible by car. They also already went to the other stores sometimes. This switch was harder for the respondent without a car who depended more on the local supermarket. Mobility is therefore an important element.
29
4.2.5 Ageing in place
“The elderly move because there are almost no facilities left”.
This was something that came up in the group discussion. Some people mentioned that friends of them had moved to surrounding villages and sometimes cities because of a need of certain facilities. Conversely, it also became apparent by the interviews that the decreasing amount of facilities was not a reason for them to move. These seem to be somewhat contradictory statements. The underlying reasons are, however, different. The first statement is more connected to a need of certain facilities which the village cannot offer, like health care. The second statement related more to the sense of community. All respondents mentioned that it was not a reason for them to move because the village has such a nice and active community where people looked out for one another. This is also referred to as ‘mienskip’ in Frisian. They have lived here for so long and built a good social network in the village. The duration of residence and age are probably important factors related to place attachment. It was also mentioned in the theory that attachment to place can be influenced by personal characteristics (Hidalgo & Hernández, 2001)
.
There are also a lot of sports clubs and other activity groups which also foster the social interaction. They do miss the closed facilities like the supermarket and the ATM, but it is not a motivation for them to move.The delivery service was also experienced as beneficial for being able to age in place. They proposed that a class or workshop could be useful in learning how to order groceries online for the elderly who are not very digital.
In sum, the closure of the store has been experienced as a loss for both food provisioning and social functions. The closure of the store also led to a change in time-space routines where people now planned further ahead and made more large purchases. The respondents with a car adapted more easily to the closure than the one who did not have a car. These people received help from neighbours and family in accessing food stored and also online shopping was used.
Furthermore, the respondents had certain place attachments to the store which were mostly based on the social role of the store and what role it fulfilled in the community (social interaction, meeting place, news updates, trusted environment where you know the people and owners, personal attention). Also, in this village the supermarket was a meeting place, which was thought to be especially important to the elderly (Bovenhoff & Meier, 2015). The closure led to the feeling that the village had become less lively and liveable since it was one of the many facilities to also disappear. However, although the supermarket was missed, closure did not lead to a significant change in place attachment regarding the village as a whole. The perceived ability to age in place was not altered as the sense of community was the most important factor in their motivation to stay living in Reduzum.
30 5. Conclusion
This thesis research has tried to give insight in the changing availability and accessibility of food stores in the province of Friesland. It has then zoomed in to the village Reduzum to research the importance of a local food store and the experiences and consequences related to closure.
First the availability and accessibility of food stores were analysed for the years 2002 till 2019 by using GIS methods which resulted in maps. The maps showed that there had been a significant change in food stores. A trend could be observed of food stores closing in the more rural areas of the province while the amount of food stores increased in the larger villages and cities. The maps comparing the service areas of the food stores for bicycle and walking showed that more loss had occurred than gain. Most loss had occurred in the north-easterly and western regions of the province.
The areas with a loss of service areas were then compared to population characteristics per district. First was the comparison with population density. Loss appeared to have mostly occurred in lower densely populated area. Also, loss also did not occur in the areas surrounding the major cities. The second comparison was between the percentage of people of 65 and older. There was no clear overlap between the layers. Lastly, the regions with population shrink and anticipates shrinking regions were compared to the areas with loss of service areas. It could be observed that most loss had indeed occurred in one of these regions.
Based on the results of the maps, Reduzum was selected as case study for the qualitative part of this research for which interviews had been held. All respondents had made frequent use of the supermarket in Reduzum and they found the closure of the store very regrettable.
The first sub-question related to the change in time space routines regarding grocery shopping. The respondents adapted their routines in the sense that they planned further ahead. They bought more groceries at once and for a longer period of time compared to when the supermarket in Reduzum was still open. The people without a car had to arrange some form of motorised transport to bring them to the nearest food stores. They often received help from family and neighbours who went with them and also online shopping was used.
The larger supermarkets which they now went to like the Jumbo in Grou were better regarding the product range. People also met more people from outside the village there. However, the quality of the conversations was considered to be less compared to the conversations the respondent had in the supermarket in Reduzum. There people were more relaxed and took more time to have a proper conversation.
When the store closed, the respondents not only missed the convenience of having a store in the village, but also the social function. It thus fulfilled both the role of food provisioning and social meeting place. It was a spot where people met each other and had social interaction. It was also the place where people were updated on the news in the village. The social role of the supermarket was perceived as particularly important to the elderly.
The respondents all liked the supermarket and had positive emotions attached to it. These were mainly based on the social role of the supermarket. One element that also turned out to be important for this was the owner of the store. The respondents all liked the owner of the supermarket in Reduzum and this personal contact added to the positive attachment and made it a more trusted environment. The closure of the store did not seem to have a significant impact on the place attachment regarding the village. They did, however, worry more about the liveability of the village. Nonetheless, the decrease of facilities was not a motivation for them to move. The strong sense of community in the village was more influential on the sense of place.
31
6. Discussion & acknowledgements
Discussion
The conclusions that have been made should be read with caution.
As was already mentioned in the data section of the quantitative research methods, changes in road networks and postcodes have not been accounted for. It can therefore not be assumed that these maps reflect the real world situations. Therefore, the network analysis can only be used to give an indication of how the situation changed, which is still a useful base for further research.
Also, multiple different years have been compared. For example, the maps comparing loss and gain in service areas for bicycle and walking for 2002 and 2019 only compare those two years and do not take the other years into account. To still give some sort of insight in the change over years, 2010 was included. In addition, the population characteristics from only 2018 were incorporated, which does not make for a good comparison.
With the qualitative research the sample is not a good representation of the population. There was a large overrepresentation of elderly women and people that still owned a car. In addition, some conclusions are also based on things the respondents assumed or thought. I still included these assumptions as they often include elements of feelings and emotions which are elements of interest in this thesis. Besides, the closure of the store happened quite some years ago so it was sometimes difficult the make the respondents of the group interview recall their emotions they had approximately eight years ago.
Despite these shortcomings, a lot of things that were mentioned in the interviews were also found in literature. These findings combined therefore still have value in terms of quality.
Acknowledgements
This thesis research would not have been possible without the dataset provided by Distrifood Dynamics. I would like to thank them for being so helpful and willing to share it.
Other people I would like to thank are the village committee of Reduzum and other inhabitants that helped me with finding respondents or that were respondents themselves, you have all been very helpful and kind. for letting me post on the Facebook page and the website and helping me with finding respondents.
32
References
Willems, J. (n.d.). Oude supermarkt in Gasselternijveen [photograph title page]. Retrieved from
https://imgd.rgcdn.nl/0f63a4b29d38451a8e093768605dc7a9/opener/De-oude-supermarkt-in-Gasselternijveen-foto-RTV-Drenthe-Jeroen-Willems.jpg
Apparicio, P., Cloutier, M. S., & Shearmur, R. (2007). The case of Montreal's missing food deserts: evaluation of accessibility to food supermarkets. International journal of health geographics, 6(1), 4. A. Steenbekkers, L. Vermeij (2013). De Dorpenmonitor - Ontwikkelingen in de Leefsituatie. Sociaal Cultureel Planbureau. Den Haag.
Brown, B., Altman, I. and Werner, C. (2012). Place Attachment. International Encyclopedia of Housing and Home, pp.183-188.
Christiaanse, S., & Haartsen, T. (2017). The influence of symbolic and emotional meanings of rural facilities on reactions to closure: The case of the village supermarket. Journal of Rural Studies, 54, 326-336.
Cresswell, T. (2009). Place. International Encyclopedia of Human Geography, pp.169-177.
Paddison, A., & Calderwood, E. (2007). Rural retailing: a sector in decline?. International Journal of Retail & Distribution Management, 35(2), 136-155.
Haartsen, T., & Venhorst, V. (2010). Planning for decline: anticipating on population decline in the Netherlands. Tijdschrift Voor Economische En Sociale Geografie, 101(2), 218-227.
J. Ritsema van Eck, F. van Dam, C. de Groot, A. de Jong (2013). Demografische ontwikkelingen 2010-2040. Ruimtelijke effecten en regionale diversiteit. Planbureau voor de Leefomgeving (PBL). Den Haag.
Gardner, P. J. (2011). Natural neighborhood networks—Important social networks in the lives of older adults aging in place. Journal of Aging Studies, 25(3), 263-271.
Morley, J. (2012). Aging in Place. Journal Of The American Medical Directors Association, 13(6), 489-492.
Clarke, I., & Banga, S. (2010). The economic and social role of small stores: a review of UK evidence. The International Review of Retail, Distribution and Consumer Research, 20(2), 187-215.
