top of page

Posts   Themes    Archive

Urban diversity and how to measure it: an operational definition of classes and scales

Paper presented at the 18th International Seminar on Urban Form, Montreal, 2011.

Sara Sardari Sayyar, Lars Marcus



Keywords: diversity, measurement tools, urban form, scale, categorization


Abstract

 

Diversity as an essential factor for liveability, economic growth, and attractiveness in cities was stressed already half a century ago by Jane Jacobs (1961, 1969). Its importance has only grown and today, diversity is stated as fundamental for current creative societies and their need for knowledge spillovers, and referred to by economists as “Jacobian externalities”. More specifically, in urban morphology and design, we can find various trends that have tried to achieve such aims under the banner of for example mixed-use. The problem is, firstly, that the definitions are not strict enough to categorize what we mean with diversity. Secondly, there are not proper analytical tools to measure diversity so that we can compare various areas. For measurements there are various challenges ahead. For instance, cities may seem diverse on one scale but widely homogeneous on another. Consequently, application of analysis on various scales and precise categorizations are essential for the development of new knowledge on urban diversity.

This paper addresses these needs, firstly, by a conceptual discussion on diversity in contradiction to specialization and homogeneity in cities. Secondly, by setting up a framework for the measurement of diversity, both as an economic phenomena and how it can be supported by urban form. The overall framework for this project is a thorough discussion and testing of a redefinition of Jacobs’ four criteria for spatial diversity in cities, introduced in the Death and Life of Great American Cities. Preliminary studies suggest strong correlation between all these criteria, in their developed and analytically formalized definitions for ground floor activities in the inner city of Stockholm.


Introduction

 

Cities and regions are today generally considered to be the greatest complex systems we have, with an unusually large set of interrelated variables. Although the study of cities, also as complex systems, using different methodological approaches has a long tradition in urban geography and planning (Wilson, 2000), there are still many unknown factors and we are far from a comprehensive theory (Batty, 2007). There is today research going on in a great variety of fields that explore urban complexity from their particular point of view using a wide range of methods and tools (Wilson, 2000; Batty, 2005). In the end, such research will enhance our knowledge regarding patterns and relations on different scales of cities as well as contribute to a deeper understanding of different critical variables. Such knowledge once obtained will be of the greatest use in future development and refinement of cities.

One of the most fundamental variables in cities, not least when considered as complex systems, is their degree of diversity. Diversity was pointed out already half a century ago by Jane Jacobs (1961) as an essential factor for liveability, economic growth, and the general attractiveness of cities. Its importance in urban debate has since steadily grown and diversity is put forward as maybe the most decisive prerequisite for urban and regional growth in the knowledge economy as well as in the information society in general, due, not least, to their enhanced need for ‘knowledge spillovers’ (Glaeser et al., 1992). In addition, the importance of diversity in attractiveness for retail facilities and in marketing studies is paramount (Teller & Reutterer, 2008).

The notion of ‘knowledge spillovers’, critical for all of urban economy, originally refers to a phenomenon in cities as production centres, where economists define cities as “the spatial concentration of economic actors” (Glaeser et al., 2001). For any such production centre the ‘knowledge spillover’ within or between firms is of fundamental importance for economic growth. The term ‘spillover’ here comes close to the term ‘externalities’ and is sometimes used interchangeably (Ref ?). For our contemporary information societies, the notion has come to be of even greater importance, since the level of productivity here often is evaluated by the ‘innovation rate’, which is highly influenced by the amount of ‘knowledge spillovers’ among the agglomerated industries in a city or region.

The phenomenon of ‘knowledge spillovers’ is today studied in urban economics using two partly contradictory ideas or theories (Beaudry & Schiffauerova, 2009). On the one hand the idea of ‘Marshallian spillovers’[1] or the ‘MAR model’, which puts emphasis on ‘knowledge spillovers’ between highly specialised firms within an industry, and on the other hand ‘Jacobean spillovers’ or ‘dynamic externalities’[2], which stresses the importance of ‘knowledge spillovers’ between industries. While both theories point out the critical importance of ‘knowledge spillover’ for innovation and economic growth, the Marshallian model stresses the importance of such spillover between similar firms, that is, its importance for ‘localisation economies’, while the Jacobean model stresses the importance of ‘knowledge spillovers’ between firms of great diversity, that is, its importance for ‘urbanisation economies’. Diversity, thus, has a more decisive role in the latter theory.

The theory of Jacobean spillovers is of a later date and is founded on Jane Jacobs’ reflections and ideas in urban economics, especially as expressed in her later books (Jacobs, 1970, 1985), where she argues that the sheer number and variety in the division of labour creates a higher inherent capacity in urban economies to develop a wider range of goods and services (Jacobs, 1970). In spite of the contradictions between these two fundamental theories; one can today find analytical studies that favour both of them (Beaudry & Schiffauerova, 2009; Audretsch & Feldman, 1998). However, diversity as a critical factor for innovation and economic growth, as argued by Jacobs, is a relatively new approach where the number of studies is rapidly growing.

Furthermore, and in relation to the increasing need for firms in the knowledge economy to consider preferences in the labour market, access to diverse and high quality amenities in cites is becoming increasingly important in urban economics (Glaeser, 2011). In a similar vein, the accessibility to diversity in people is also discussed as an essential factor, with important effects on the economic growth in cities. The idea here is that people attract people and especially attractive is diversity among people, not least for high skilled and innovative people (Florida, 2009). This mirrors also the increasing interest in urban economics for cities as consumption centres (Glaeser et al., 2001), that is, as points of attraction for consumers who look for diverse and exclusive services, as opposed to the traditional interest in cities as production centres. Finally, in retail marketing, diversity among commercial activities is recognised as an immensely important factor in generating profit and therefore a variable of increasing interest in gravity-based location studies (T. Drezner & Z. Drezner, 2008). Commercial diversity is, furthermore, specifically pointed out by Jacobs as an indicator of more general diversity, since it normally is dependent on diversity in other fields or scales (Jacobs, 1961, p. 148).  

The essential question from the point of view of urban planning and design is then: what are the central factors that support and enhance diversity? While it is likely that such factors especially will be found in different socio-economic institutions, like bank-systems, rent policies and innovation centres, Jacobs (1961) introduced the urban fabric in itself as a strategic factor that through its ability to create diversity, co-presence and variations of accessibility, can create efficient economic pools of uses. In this context she listed four specific criteria for the generation of what in this paper more specifically is referred to as urban diversity: primary functions, short blocks, old buildings and density (Jacobs, 1961). In recent decades we have also seen many followers of her ideas, especially in urban design, where there have been many attempts to develop diversity under the banner of for example mixed-use, not least in New Urbanism (e.g. Duany 2000).

The problem is, firstly, that the definitions of diversity lack rigor so that it is difficult to be sure what is actually meant with diversity in different cases. Secondly, we lack analytical tools that can measure diversity with any precision so that we can compare and evaluate different urban areas. Together, this means that arguments for or against concepts like mixed-use are often conducted without proper basis in empirical data. To develop useful measurements, we need to be precise on, first, what category of urban phenomena we are investigating, that is, answering the question: diversity of what? Second, we need to be clear on what scale we measure diversity, since cities may be diverse on one scale but prove to be most homogeneous on another, that is, answering the question: diversity on what scale? Therefore, distinct definition of scale and category is essential for the development of precise analysis and measurement of urban diversity.

         This paper addresses these needs by setting up a framework for discussion of the methodological difficulties in measuring diversity. Since the ultimate end is to develop a deeper understanding of how urban design can support, develop and sustain urban diversity, there are two sides to this. On the one hand, we need to develop means whereby we can measure diversity in an economic context, and, on the other hand, we need to develop variables of urban form that can be shown to influence urban diversity. Together this will be given the form of a thorough investigation of Jacobs’ four criteria for diversity introduced in the Death and Life of Great American Cities in 1961. In the end, the project aims to contribute to a more informed discussion on methods and ideas on how urban form structured and shaped in urban planning and design, acts as a means to develop spatial structures in our cities adequate for our current and more complex knowledge economy.

         In this paper, we focus on the two critical issues of classification of economic activity and differentiation of urban scales, starting with a theoretical discussion and short review of current methodological development in both, followed by a preliminary empirical test in a group of neighbourhoods in Stockholm. It should be emphasised that we do recognise the complexity behind the generation of urban diversity such as the distribution of economic activity, as well as the wide range of variables that influence it, including other than urban form. However, it should be kept in mind that the current study is conducted from an urban design perspective where the examination of the observed patterns from an urban form perspective is of specific interest to be able to inform and improve such practice.

 

2. Classification

 

2.1 Background

Fundamental for any type of analysis of urban space is to develop or choose a satisfying system of classification (Harvey 1969: Wilson 2000). As in most empirical studies, the field of study in spatial analysis presents a rich set of individualities that needs to be sorted one way or another to be accessible for relevant study. Such a classification depends on the aims of the enquiry, where the very same individuals can be sorted very differently depending on these aims of. Hence, the development of an adequate classification system asks for the greatest rigour and precision. At the same time, this has proven to be one of the most difficult tasks in spatial analysis.

In our particular case, the need of a clear classification arises from the aim to analyse and measure diversity in urban space, more specifically, diversity in economic activity in urban space, which first of all needs to be kept separate from the amount or the density of economic activity in urban space. The principle behind the choice of classes, the number of classes, as well as their attributes, will all have decisive effects on the final diversity values. For example, it we measure the diversity in an area concerning primary functions, for instance divided into residential and working population, this will be very different from its diversity concerning economic sectors, such as official, commercial and industrial uses. Moving further down the hierarchy measuring the diversity in for example the commercial sector and more specifically retailing, which can be divided based on the type of offered goods, such as, clothes, shoes and furniture, this will yield different values yet. This means that an area can have a high diversity and a low diversity at the same time, all depending on the classification is used. In the following a survey of various systems of classifications and some of their theoretical foundations is conducted as support and background to finding an appropriate system for the kind of study we want to perform.

 

2.2 Various systems of classification

Classification is regarded as “the basic procedure by which we impose some sort of order and coherence upon the vast inflow of information from the real world. ... [It is] maybe regarded as a means for structuring reality to test hypothesis” (Harvey, 1969, p. 326). Today we often have access to a vast amount of data, available from various sources, which makes extensive analysis on various scales more feasible than ever. However, it also produces difficulties in sorting and using data for proper and useful results. Consequently, sorting information into classes and sets will be necessary to ease the manipulation of data in any investigation process. However, different classes could be proposed for the same data depending on the purpose of the investigation. Therefore, the efficiency and adequacy of a classification system designed for a specific study can not be evaluated independently of the purpose of the study. The importance of a primary question or a presupposed hypothesis is often stated, in order to have a proper interrelationship between theory and classification (Harvey, 1969). Additionally, as Wilson emphasises one should be aware that, “there is no absolutely right way to do categorization. But any account of the population in categories provides a statement of initial condition” (Wilson, 2000, p. 8), and the condition will be present at various steps of the study. [What does this mean?]

Two main procedures of classification in scientific research can be identified, ‘logical division’ or ‘deductive classification’ and ‘grouping’ or ‘inductive classification’ (Harvey, 1969, p. 334). The first process is described as division of the data using a universal set which is often applied at different levels as a series of steps, where at each step a set of properties is used to differentiate between classes. This type of classification is highly determined by criteria chosen beforehand, which assumes that there is good prior knowledge about the studied phenomena. Accordingly, this procedure is also named as classification from above. On the other hand, ‘grouping’ is a type of classification from bellow. Here, the process is conducted as an inductive procedure by which the phenomenon being examined is searched for patterns, regularities, or significant interrelationship concerning various factors, so that the data is grouped based on such common properties. Grouping is by Harvey said to have the possibility to yield a much more realistic classification, since it is a classification applied without presumptions or prior chosen theory. Using this method a theory of interrelationship in the data can be said to be constructed, the classification system in itself becomes part of the end-product in that the classifications system in itself constitutes knowledge about the studied phenomenon (Harvey, 1969).

There are various systems of classification available today, such as SNI (Svensk Näringsgrensindelning: Swedish standard industrial classification) in Sweden, which covers all kinds of organisations, including economical organisations, such as firms. This is similar to other exploratory classifications, like the ones presented by Wilson (2000) that defines organisations in hierarchical systems from coarser to finer levels of economic sectors. Such hierarchical systems of classifications are mainly based on the type of economic activity the organisations is involved in.

Concerning commercial diversity, a review of the literature related to retail studies, revealed that there is no standard definition or classification system for various types of retail facilities, such as type of shop or type of location. Most studies use loosely defined classes, such as, shopping malls, convenience shops and shopping streets as types of agglomerated retail, where especially their attractiveness for consumers is in focus, and, moreover, they are more or less never put together in a distinct system. Normally, such facilities are classified according to their size, their number of shops and the variety in the goods they offer. In extension to our particular interest in urban morphology and urban design, it is important to note that we find few attempts of classification of retail markets according to spatial criteria such as accessibility or density (Gibbs, 2010; Harry & Associates, 2010). On account of this lack of any standard systems of classification for retail activities in general and even less so when it comes to their spatial criteria, of interest for our current investigation, the procedure of grouping seems most apposite for the development of such a classification system.             

 

3. Scale and related issues

 

The choice of scale in studies of urban diversity will influence both the observed patterns of diversity and its interpretation. For instance, an area detected as relatively homogeneous at one scale may appear to be rather heterogeneous ay another scale. So, the issue of deciding the proper and useful scale in diversity studies is of fundamental importance. The problem of scale is confronted also in other disciplines, such as landscape ecology and is of course a central issue in all branches of geography. In the case of landscape ecology it is due to the fact that one is dealing with ecological phenomena on different scales (Gibson et al., 1999; Meentemeyer, 1989; M. G. Turner et al., 1989). Also in urban studies there is a wide range of references to scale, such as local/global, large-scale/small-scale, neighbourhood, district, city and region. However, these terms are more often than not most vaguely defined. Furthermore, the theoretical understanding of the inherent difficulties in moving between scales is often lacking. An overview of definitions of the scale as well as other related concepts will therefore support the current study.

 

3.1 Scale and other relevant terms

The term scale is highly used in many fields and is, furthermore, often interpreted differently in different disciplines. Besides, there are many related concepts like; level, resolution, extent, and hierarchy, used as replacements or synonyms. In scientific terms, scale is defined as spatial, temporal, quantitative or analytical dimensions and is used to measure and study objects and processes (Gibson et al., 1999). It could be mentioned here that temporal dimensions, in comparison to other scales, is generally less investigated in urban studies due to difficulties to examine urban patterns over long time spans. Level, as a related term to scale, addresses the issue of scale in relation to locations along a scale. For instance, on a spatial scale, micro-, meso- and macro-levels refer to small, medium and large-sized phenomena. Extent is defined as magnitude of a dimension used in measurements. In relation to space, extent could differ from one sqm, to thousands of sqm. And finally, resolution refers to the precision used in the analysis and measurements (ibid).

The interrelationship between scale and hierarchy as a concept is also recognized in various analysis, where Wilson (2000) explains that “scale is a form of hierarchy and clarity of vision in this respect is critical”. In the definition of hierarchy lies also the idea that there is a conceptually or casually linked system for grouping phenomena along an analytical scale (Gibson et al., 1999). Hierarchical studies, therefore typically covers various interrelated levels. Implementation of hierarchical analysis on various levels from local to regional level, data hierarchy from coarse to fine resolution, and hierarchy of extents from small to large as the main methods in the study will ease dealing with complexity of investigations. Moreover, the outcomes of hierarchical analysis in exploring and understanding the urban diversity and its context will contribute to decision making processes from large scale to district level.

Also in relation to the definition of scale, issues still arise regarding human perception of space. For instance human geographers use concepts like: absolute, relative and conceptual scale (Gibson et al., 1999), which are defined by the scale of the relationship between the dimension and the object. Absolute scales refer to those that exist independent of the objects or processes being studied. For instance, scales applied in remote sensing or grid systems normally both use absolute scale. On the other hand, relative scales are defined by, rather than define, the objects and processes under study. Finally the conceptual scale brings up terms like global and local with their conceptual level of “comprehensiveness” or “contextuality” (Meyer et al., 1992; in Gibson et al., 1999). The concepts ‘absolute’ and ‘relative’ is also found in relation to ideas about space. This is a classical debate where, according to Harvey, ‘absolute space’ is understood as a “container of all material objects” (Harvey, 1969, p. 195), and takes an Euclidian point of view on space (Meentemeyer, 1989). ‘Relative space’, on the other hand, understands space as “a positional quality of the world of material objects or events” (Harvey, 1969).

 

Discussion

 

To summarise, for proper analysis of urban diversity it is necessary to have a multi level approach, covering micro-, meso- and macro-studies, in order to get a necessary broader perspective. This approach presents the possibility to move between the scales and look for interrelations and common patterns. Furthermore, if we accept the presumption in urban economics that diversity leads to regional urban growth (Glaeser et al., 2001, 1992) the issue of on what scale we talk about diversity in the city becomes critical and most likely there is strong interaction here between scales. Not least, this opens for influencing such diversity using urban design, which primarily concerns the local scale but can maybe be proven to, indirectly, also influence other scales. On the other hand, economical growth generated by urban design on the local level can be considered as a proxy for meso- and macro-changes. Urban studies, like investigations in urban economics, mostly focus on regional or micro-levels, with less studies concerning the levels in between. In this regard, hierarchical analysis covering specific neighbourhood, group of neighbourhoods, districts and finally the regional level is suggested to be important in future studies. The use of relative scale related to the importance of individuals’ perception of urban diversity is suggested in this project, using configurational analysis as developed theoretically and methodologically in space syntax research, more particularly using available software tools (depthmap, confeego and place syntax tool). Regarding the choice of proper classification for the study of urban diversity, we will, on the one hand use a classification system that comes closest to our needs (Horton xx) and, on the other hand, conduct a grouping procedure using the data for the areas under study. This method presents the possibility to explore the limitations of both approaches that hopefully will lead to a better understanding of the subject.

 

4. Experimental study

 

The following study intends to test the results of the discussion above concerning classes and scales in a group of neighbourhoods in Stockholm. The study applies the cognitive scales proposed above as well as the grouping of commercial activities and the diversity among them The commercial centres of each neighbourhood are classified according to their level of integration in the urban system through configurational analysis on three levels: global integration (radius 30) representing accessibility on the urban scale, and local integrations (radius 3) representing accessibility on the neighbourhood scale and radius 6, representing accessibility on the district scale. These analyses are applied using soft ware techniques developed in space syntax research (Hillier & Hanson, 1984; Hillier 1996). The study tests the hypothesis that the level of integration of the centres at various scales can capture the area of accessible demand, that is, the accessible population in this case including both residents and working population. Put the other way around it can be said to capture how far shops in each centre can reach at different distances. Such spatial factors as accessibility measured in this way are clearly influenced by urban form and how it spatially structures the neighbourhoods, and it seems likely that they are essential to the way commercial centres perform and are sustain over time.

The general hypothesis is that neighbourhoods with a low level of accessibility at the urban and district level will tend to contain mostly shops offering convenience goods (group C), which primarily cater for local demands. Neighbourhoods with higher integration at the urban and district level will, besides convenience shopping, begin to offer more comparability shopping (group B), which need a larger pool of demand. Finally, neighbourhoods with a high accessibility at the urban and district level will, while keeping both convenience shopping and comparability shopping, add a new group of shops offering speciality shopping (group A), including more exclusive and specialised retail, food markets and restaurants, which need much larger pools of demand.

The distinction made by Holton (1985) between various types of goods corresponds well to the classification of shops above. He introduces three different types of goods: convenience goods, shopping goods and specialty goods. The first group describes goods that need to be frequently and immediately accessible, where time is more important than comparing the price and quality. Shopping goods are goods where the consumer enjoys shopping and want to compare quality and price. Finally, specialty goods consists of goods of a more exclusive and special kind where such things as brands are important but can still concern a wide range of types of goods such as watches and clothes but also groceries. However, Horton’s classification describes types of goods without the spatial context where they are offered. More specifically, if for example shopping goods like clothes, are offered in a shop that is located in a highly local centre, segregated from surrounding neighbourhoods, and with a limited access to consumers, that is, a small demand area, the shop, performs more like a convenience shop and to a degree transforms the shopping good into a convenient good. The point we want to make is that there is reason to believe that one type of shop will perform differently depending on its location in relation to both other shops and local demand and thereby also change the character of the goods it offers.


Figure1: the location of eight neighbourhoods in southern Stockholm


Accessibility, in addition to density and diversity of shops in commercial centres, are all recognised as essential variables for the attractiveness of centres in retail marketing studies (Teller & Reutterer, 2008; T. Drezner & Z. Drezner, 2008). Accessibility is then almost exclusively analysed as accessibility by cars as the major means of transportation, while diversity is measured by mix of tenants and type of offered goods. The current study aims to give different definition of these by using accessibility analysis for pedestrian movement as developed in space syntax research, that is, moving to analysis at a cognitive level of urban space. This will also be extended by using the place syntax tool[3], for analysis of density and diversity.

The eight selected areas: Farsta, Skärholmen, Hammarbyhöjden, Bagarmossen, Gubbängen, Hökarängen, Midsommarkransen, and Västertorp (Figure 1), are located in the southern part of Stockholm. They are all planned and developed around 1930-1970, and have all access to the underground system. Each area has access to a planned neighbourhood centre, inspired by neighbourhood unit planning. Part of the planning was that two of the areas, Farsta and Skärholmen, were to function as larger centres serving more than their local neighbourhood, but a group of neighbourhoods. Skärholmen is in current planning considered to be a strategic point for business growth in Stockholm municipality that can contribute to the regional growth (Annual Report 2010 - Stockholm Business Region). Among the other centres of the chosen areas, Bagarmossen, Gubbängen and Hökarängen are generally characterised as highly local centres, while others, like Midsommarkransen, is considered to be a little more active. The residential population of the areas vary between 4500-10800, and the number of registered retail activities for each area, excluding trade in motor vehicles and ambulatory activities in stalls, are between 27 to 119, where Farsta and Skärholmen have the highest numbers (Table 1).

Neighbourhood

Retail shops

Residents

Working population

Total population

Farsta

119

10887

4172

15095

Skärholmen

99

7710

3733

11443

Hammarbyhöjden

32

5450

969

6419

Bagarmossen

27

10201

992

11193

Gubbängen

17

4571

1556

6127

Hökarängen

32

7989

997

8986

Midsommarkransen

47

8078

3071

11149

Västertorp

27

5524

745

6269

Table1: Number of retail shops, residents and working population in each area.


The axial map[4], which is used for configurational analysis, is comprised of 66.000 lines and covers all of the municipality of Stockholm as well as some other municipalities in the vicinity. Data used for the accessibility analysis includes census data for all residential and working population from early 2000. Data regarding various economical activities include all economic activity in 2006 sorted according to branch code. Both data sets have extremely detailed resolution on address level, yet at certain steps in the study the data is aggregated to facilitate comparison between neighbourhoods.


Empirical results


Integration analysis at the global level (radius 30) shows that suburban areas to the south of the inner city are better integrated than suburbs to the south-west (Figure2).  From the map we can see how Skärholmen and Farsta both are far less integrated with their surrounding neighbourhoods at this level of analysis than the inner city. Midsommarkransen and Västertorp, being closer to the inner city measured metrically, are still less integrated than Gubbängen.  Integration analysis at radius 9, shows that Skärholmen, Midsommarkransen, Gubbängen and Hammarbyhöjden are fairly integrated with their adjacent neighbourhoods (Figure 3).


Figure 2: Spatial integration analysis, Global integration (radius 30), local integration within longer walking distances(radius 9)


Figure 3: Spatial integration on highly local level (radius 3)


At this level of analysis, based on the prior assumptions for grouping, Hammarbyhöjden and Gubbängen are the areas with highest integration with the urban core (group A), while Skärholmen despite its planning approach is well integrated at walking distances, and is not connected at global level, hence it is part of group (B) as well as Midsommarkransen. Farsta is fragmented at both local levels in addition to global level and is included among other remaining neighbourhoods within group (C) with low integration within walking distances.  These areas are assumed to perform at local level. It should be emphasized here that as stated all areas have access to subway system, so they are accessed by visitors arriving by subway, current study considers the funndamental infrastructure formed by urban form.

 

Accessibility analysis (Figure 4 and 5) from retail shops to residents as customers on local level (within 500 m/3 axial lines) shows that shops in Midsommarkransen have on average the highest access (within 260-326), while Farsta (98-2300) and Hökarängen (275-1800) have the lowest. The importance of the working population as additional customers is evaluated by accessible total population, where account is taken for both residents and working people. Analysis of access to total population reveals that shops in Skärholmen (within 80-5200) and Midsommarkransen (within 900-4600) have access to more people, where Farsta stays at next level (between 100-4000), even though Farsta has the highest total population. Hökarängen (280-2500), Gubbängen (650-2900)and Västertorp (250-2250) have the lowest ranges.    Analysis within longer walking distances (1500m/9axial line) illustrates that Hammarbyhöjden and Midsommarkransen have the highest access to both residential population and total population (residents:10-20000, total population: 20-35000)  which is almost double the highest ranges for Farsta,  and Skärholmen (total accessible population at most 20000). Västertorp and Hökarängen have the lowest access to both groups of population (highest access to total population less than 15000).

 

Figure 4: Comparison between population groups and accessibility of each group to shops within walking distances in neighbourhoods (aggregated level).


Figure 5: accessible total population to shops (left), and other retail shops(right) within 500m/3axial lines on address point level


To summarise these two types of analysis, one can say that Midsommarkransen and Hammarbyhöjden are well integrated within an area larger than their local neighbourhood and have therefore access to the highest population in the group, while Skärholmen is well integrated at local level but less integrated with neighbouring areas. On the other hand, Farsta, while being well-known to be visited by large quantities of customers is not that well integrated with its neighbouring areas, which clearly shows it dependency on the underground system. Even though Gubbängen is integrated at the global level it has low access to local population. Other neighbourhoods have both low integration and low accessible population.

Analysis of accessibility from the shops in each area to other retail shops within 500 metres, shows that Farsta and Skärholmen by far have the most shops, most likely due to the fact that they are planned as large commercial centres. At the distance of 1500 metres, we get the similar number of accessible shops in Midsommarkransen and Hammarbyhöjden as in Skärholmen and Farsta. Comparing also the access to restaurants from shops, we find similar numbers. It is interesting to note the growth in number of accessible restaurants for Midsommarkransen and Hammarbyhöjden at 1500 metres, a growth much higher than in the two commercial centres Skärholmen and Farsta (Figure 6).

Comparison between the types of shops in each neighbourhood reveals significant differences among the areas (Figure 2). However, both convenience shops and other retail shops are found in all areas, but most of the shops fall into group (C) as highly local shops. Midsommarkransen and Hammarbyhöjden have a few shops in group (B) facilitating comparative shopping. Farsta and Skärholmen are exception because of their size and function as regional centres. All other neighbourhoods are dominated by convenience shops (Figure 7).

 

References

 

Audretsch, D. B. & Feldman, M. P. (1998). Innovation in Cities: Science-Based Diversity, Specialization and Localized Competition. C.E.P.R. Discussion Papers. Retrieved June 19, 2011, from http://ideas.repec.org/p/cpr/ceprdp/1980.html

Batty, M. (2008). Cities as Complex Systems: Scaling, Interactions, Networks, Dynamics and Urban Morphologies. UCL Center for advanced spatial analysis. Retrieved June 13, 2011, from http://discovery.ucl.ac.uk/15183/

Beaudry, C. & Schiffauerova, A. (2009). Who’s Right, Marshall or Jacobs? The Localization Versus Urbanization Debate. Research Policy, 38(2), 318-337. Retrieved June 19, 2011,

Drezner, T. & Drezner, Z. (2008). Validating the Gravity-Based Competitive Location Model Using Inferred Attractiveness. Annals of Operations Research, 111, 227-237.

Florida, R. (2009). Who’s Your City?: How the Creative Economy Is Making Where to Live the Most Important Decision of Your Life. Basic Books.

Gibbs, R. J. (2010). Retail Markets Smart Code Module. Retrieved from http://www.transect.org/modules.html

Gibson, C. C., Ostrom, E. & Ahn, A. (1999). The Concept of Scale and the Human Dimensions of Global Change: A Survey. Ecological Economics, 32(2), 217-239.

Glaeser, E. L., Hedi D. Kallal, Jose A. Scheinkman & Andrei Shleifer. (1992). Growth in Cities. Journal of Political Economy, 100(6), 1126-52. Retrieved June 19, 2011,

Glaeser, E. L. (2011). Triumph of the City: How Our Greatest Invention Makes Us Richer, Smarter, Greener, Healthier, and Happier. Penguin Press HC, The.

Glaeser, E. L., Kolko, J. & Saiz, A. (2001). Consumer City. Journal of Economic Geography, 1, 27 -50. Retrieved June 13, 2011,

Harry, S. & Associates. (2010). Retail: Sustainable CommercrceSmart Code Module.

Harvey, D. (1969). Explanation in Geography. St. Martin’s Press.

Hillier, B. & Hanson, J. (1984). The Social Logic of Space. Cambridge University Press.

Holton, R. H. (1958). The Distinction Between Convenience Goods, Shopping Goods, and Specialty Goods. The Journal of Marketing, 23(1), 53-56. Retrieved July 4, 2011,

Jacobs, J. (1961). The Death and Life of Great American Cities. Modern Library.

Jacobs, J. (1970). The Economy of Cities. Vintage.

Jacobs, J. (1985). Cities and the Wealth of Nations. Vintage.

Meentemeyer, V. M. (1989). Geographical Perspectives of Space, Time, and Scale. Landscape Ecology, 3(3/4), 163-173.

Meyer, W. B., Gregory, D., Turner, B. L. & McDowell, P. F. (1992). The Local-global Continuum., in: Geography’s Inner Worlds, (pp. 255-79). Rutgers Univ. Press.

Teller, C. & Reutterer, T. (2008). The Evolving Concept of Retail Attractiveness: What Makes Retail Agglomerations Attractive When Customers Shop at Them? Journal of Retailing and Consumer Services, 15(3), 127-143. Retrieved June 24, 2011,

Turner, M. G., Dale, V. H. & Gardner, R. H. (1989). Predicting Across Scales: Theory Development and Testing. Landscape Ecology, 3(3/4), 245-252.

Wilson, A. G. (2000). Complex Spatial Systems: The Modelling Foundations of Urban and Regional Analysis. Northcote House Pub Ltd.

 

Notes


[1] Marshallian externalities or spillover, refers to A. Marshall.

[2] This concept was introduced by E. Glaeser in his article “Growth in Cities” (Glaeser et al., 1992) that he refers to as Dynamic Externalities in addition to MAR knowledge spillovers (Marshall-Arrow-Romer). These concepts are based on knowledge spillover as externalities resulted from agglomeration of industries in one city.

[3] For further description of the Place syntax tool, see Ståhle, Marcus and Karlström, 2005.

[4] The axial map is merged from several different axial maps made by researchers from the research group Spatial

Analysis and Design at the School of Architecture, and by the consultant firm Spacescape.

© LARS MARCUS
architect and professor in Urban Design

  • LinkedIn
  • Instagram
bottom of page