Spheres of Influence:

Social and Economic Integration Indicators / Alternative Methods for Deliniating CMA's

 

 

Self Directed Graduate Studies Geography Project

Prepared for Tim Davies, Adjunct Professor of Geography, Director, Geography Division, Statistics Canada

by Warren Munroe

Major contributors: Henry Puderer & Peter Murphy, Statistics Canada

July, 2000


Abstract

This report describes how alternative indicators of social and economic integration were developed and then tested through a telephone survey of residents living outside of Kingston Ontario. A random sample of telephone numbers led to interviews of twenty five people from each of three areas, approximately 20, 40, and 60 kilometers from Kingston. This procedure was then repeated for a second tangent, with sampling done at the same distances. Five indicators were tested: information, communications, goods, services, and social activity. High and low values were arbitrarily ascribed to the last three variables, while communications was tested with 'phone calls to' and 'phone calls from' Kingston. The information variable was tested using 'newspaper readership' and 'radio listening' habits. Questions were posed asking about the most recent activity in order to increase accuracy but did not inquiry about frequency nor duration.

The results of the spheres of influence study suggest that there are differences in the size of area covered by each of the social & economic integration indicators tested. Social needs and minor health care services are seen to be met locally, while economic needs are not. The 'place of work' indicator presently used to delinate central metropolitan areas, which closely matches social activity and minor health care, appears to be a conservative estimate of integration. Both the purchase of goods above and below $1000 from Kingston reachs much further than place of work. It appears that local rural retailers are not able to compete in the supply of goods. Many respondants mentioned that the last place that they shopped was at a particular 'box store' in Kingston. Since these stores are often set up on the edge of large centers, the 'goods' variable may not have tested integration of the core and periphery, as much as it may have tested the integration of the urban fringe and periphery.

Though the location of the purchases has not been formally tested, several phenomina may be interesting to consider. Telecommunications, including the internet, provide new opportunities for improved access to the marketplace. At the same time, deregulation of the wholesale retail structure (not unrelated to advancements in technologies) has allowed 'box stores' to florish. Internet shopping, while convenient for the consumer, moves the time consuming and labor intensive activity of sorting individual orders to the wholesaler/retailer, as well as the delivery costs. Instead, 'box stores' have taken the advantage of the deregulation of the retail / wholesale market by providing goods to consumers on pallets, therefore reducing sorting and shelving costs. Consumers may be traveling further to these stores which are set up on the urban fringe, in order to select goods for themselves. Not only is this consumer activity moving out of the core to the urban fringe, but 'box stores' may be changing the quality of life for urban and rural residents, while increasing spheres of influence.


TABLE OF CONTENTS

1. Introduction *

2. Overview / Literature Review *

3. Methodology *

3.1 Orientation *

3.2 Spatial Elements *

3.2 a) Urban-Rural Site Selection *

3.2 b) Areas surveyed *

3.3 Indicators & Survey Questions *

3.3 a) Economic indicators *

3.3 b) Social indicators *

3.4 Measures *

3.5 Distances and Mapping *

3.6 The Survey Process *

3.5 a) Respondents *

3.5 b) Telephone Procedures *

3.5 c) Question ordering *

4. Results and Discussion *

4.1 Response *

4.2 Significance levels *

4.3 Areal Extent of Various Indicators *

4.3 a) Information and Communication *

4.3 c) Goods *

4.3 d) Social Activities *

4.4) Spheres of Influence *

5. Summary *

Bibliography

Appendix A: Responce Tables (due to time constraints these are not yet posted. Email a request) A-1

Appendix B: Survey Questions B-1

 


List of Figures

Figure 1: Example of csd's to be included

Figure 2: Example of a scattergram of extents

Figure 3: Graph of Spheres of Influences for Kingston Ontario

 

List of Maps

Map 1: Kingston Place of Work for Ontario csd's

Map 2: Kingston Area: crosssections and telephone areas

Map 3: Distance to roads

Map 4: Spheres of Influence estimates


1. Introduction

The classification of administrative areas requires the evaluation of particular variables in relation to place. The method of classifying central metropolitan areas involves the use of several indicators. In Canada, density and population are used to indicate areas of high and low inhabitation, while socio-economic integration of the people in an area is measured by 'place of work'. But are there other more appropriate indicators of integration? What of areas of high socio-economic integration and low numbers of people working in the urban core? Measures of integration may involve variables other than location of employment. For example, people in areas surrounding the urban core may utilize it's retail and commercial goods and services, or it's arts and entertainment, but may be employed else where. This then leads us to question, "to what degree is a jurisdictional delineation preconditioned by the method of analysis?" and "what are the chances of making an error when a particular methodology is employed?" The purpose of this study is to test variables other than 'place of work', in order to examine their potential for describing socio-economic integration.

 

2. Overview / Literature Review

Delineations between urban and rural areas are at first relative and subsequently given absolute values. The United Nations concluded in 1955 that "there is no point in the continuum from large agglomerations to small clusters or scattered dwellings where urbanity disappears and rurality begins" (United Nations demographic yearbook, 1955). Yet, according to Cloke, there is "some sort of dichotomous continuum from truly rural to truly urban communities" (Cloke, 1977). Sorokin suggested that the transition between rural and urban settings was one characterized by gradations which reflected variations in "occupation, environment, community size, population density, homogeneity, social differentiation, mobility, and systems of interaction." (Sorokin and Zimmerman, 1922) Around the world, countries use widely different criteria for delineations. These usually include measures of agricultural activity in relation to industry and manufacturing. Few countries make use of measures of density or population concentration. Yet, with a concentration in population, facilitated by a surplus in goods, the processing of raw materials allows for specialization and differentiation in labor activities. An urban area then is a "concentration of differentiated but functionally integrated specialization." (Lampard, 1955) An economic association, one that affords specialization and yet has a need for spatial association would seem to be of particular importance in defining core areas (Childe). On the other hand, core areas require a degree of social association. Peripheral areas are then defined in relation to core areas. In this manner, peripheral areas might be described as areas of relative homogeneity in terms of labor activity, as well as being areas of lower population and density. Yet, in order to remain true to the hypothesis put forward by Cloke regarding the dichotomous nature of the continuum, it is helpful to consider the variations in rural settings and homogeneity in urban settings; for this must be an exercise in perception.

Canada's large area has supported many diverse local activitives and it was not long ago that living on the land required specialized skills not easily matched by urban dwellers. Urban / rural difference may best be seen in terms of variations in control over natural processes. Those living 'on the land' contend more directly with nature's extremes, while cities attempt to remove these variations in preference for 'human standards'. This suggests that urban and rural dwellers have more in common with themselves rather than between each other. While rurality maybe seen to be more physical, and urbanity may be seen to be more human, there is still interaction since there is movement; not only of raw resources and refined goods (quantity and quality?) but also of people.

Urban cores may differ amongst themselves in a number of ways. Some are primarily transportation nodes, others are administrative centres, while others may be have manufacturing as the principle function. In the case of the latter example, globalization and restructuring have resulted in 'out migration'. As early as the 1970’s, the growth of many North American cities began to slow and there was an increased proportion of the population living in non-metropolitan areas. Berry described the outflow from metropolis into hinterland as population deconcentration. Dispersion was evident when looking at reduced population gradients and internal population redistribution in rural areas (Bourne, 1988). In 1989, Champion used the decline in urban population as a measure of deconcentration. He listed seventeen reasons for counter-urbanization, many related to economics. Of particular importance though, are advances in mobility and technology, as they allow people to live further from areas of high concentration, and as they relate to changes in wealth production and economic restructuring. At the root of the change may be the increased ability to specialize without the need for spatial contiguity resulting in an urban ‘hierarchy’ having less meaning (Kristol, 1972). Other dispersion processes include attitudes regarding urbanity versus privacy and living in single family dwellings in countryside settings (Bunting, 1999). In describing the 'dispersed city form' of Kitchner Ontario, Bunting points to the amalgamation of small centers and the lack of a core as important contributors. He suggests that this form is not uncommon especially for smaller centers, while large agglomerations tend to maintain a strong central business district.

Though deconcentration may be considered to be a rather recent phenomena, others have pointed to a history of fluctuating degrees of expansion and contraction of urban centers (Robinson, 1989). What is important to consider in terms of counter-urbanization is the difference between maintaining an urban lifestyle, or adopting a rural lifestyle. The former refers to an ‘extension of the metropolis’ and vertical integration, while the later refers to a ‘deconcentration of the metropolis’ or vertical disintegration. Yet, while the role of the central place may be changing, the fact remains that there are places with high population concentrations and high degrees of social and economic integration. The rise of the service sector has provided some centers with a new primary function. With globalization, the 'traditional' industrial core/periphery relationship may be undergoing some fundamental changes. In some cities, opportunities for education and training as well as the availability of health services and improved living standards attract people and firms. Indeed, the decline in the use of the central business district and adjustments in manufacturing was replaced, in larger centers, with the influx of more white collar workers (Ley, 1988). On the other hand, access to education and health services via improved technologies reduces the need to move and improved living standards may not always be available in cores. The 'city effect' which attracts people may result in 'overburdened cities' (Cicerchia, 1999). Thus, mobility between the core and the periphery may be an attempt to seek an acceptable equilibrium.

With distances becoming less of a constraint on travel for consumers, one would expect an increase in demand limits. This relationship between the periphery and the core would only exist until intervening opportunities arose. Nagot and Schmitt suggest that unusually high, as well as unusually low transportation costs result in deconcentration. In the former case firms move to peripheral areas to address demands as it is cheaper to move products than it is to move people and in the later case people move further afield for lower rents (proving, the authors suggest, economic rationality). Therefore, spheres of influence can be affected by exogenous forces and these must be considered in relation to generalized movements over time (Nagot and Schmitt, 1998).

Rural areas also have varying landscapes which can be described in relation to core activities. Blunden developed five categories using a neural net with inputs summarizing access, settlement, population, socio-economics and telemetrics (wired and wireless technologies) to come up with :

    1. Urban imprint zone
    2. High amenity advantage
    3. Developed and Balanced areas
    4. Development potential
    5. Require restructuring

This approach implies that the urban influence reaches far beyond the immediate built up area "adding an important dimension to rural classification."

Definitions of central places have been tempered by changing agglomeration activities and counter-urbanization and therefore can be approached with a view to the potential importance of alternative measures. Certainly, size alone is not a measure of centrality (Gradman), but rather, measures of exchanges of goods and services must be considered (Christaller). Improvements in transportation and communications, as well as the rise in the emphasis on information have led to the consideration of new indices of integration (Carter). Spheres of influence of a central place are thought of in terms of the distance over which various goods and services are exchanged. The extent of each type of exchange from core to peripheral areas will likely vary. For example, items that are highly specialized and rarely acquired may attract people from further afield then items that are needed on a daily basis. In the latter case, the demand limits may be met by intervening opportunities. A gradient of demand limits for different elements can be drawn around core areas, following the example of Losch’s demand curves (Losch, 1954). Surrounding the core area, there maybe an intensive – inner – sphere, while further away there may be an extensive – outer – sphere. These gradients can be juxtaposed with degrees of specialization using Davies location coefficient (Davies, 1967). As well, different core functions may have very different spheres of influence, for example, university towns or convention centers.

Even the use of journey to work, though relatively quick and easy, is "not acceptable at a rigorous level of investigation" (Carter). Though Carter maybe all to ready to apply academic screws to ‘place of work’ as an indicator of social and economic integration (for he offers no alternatives), the restructuring of the workplace makes a survey of other potential indicators pertinent. The first difficulty in developing a survey then is finding appropriate elements. This exercise is at best subjective and arbitrary. Nonetheless, in such a reconsideration, it is helpful to refer to both physical exchanges of goods and services, as well as, cultural and information exchanges. Elements of both high order ‘intensive – inner’ limits to lower order ‘extensive – outer’ limits should be included.

Following the suggestion that a study of social and economic integration indicators requires a flexible approach to the development of indices, an investigation of integration between urban ‘core’ areas and rural ‘peripheral’ areas can be approached as a study of the areas serviced by the core or as a study of the use of the core by residents of the countryside.

A consumer behavior approach (increasingly popular today) can be satisfied by a survey of the people living in areas surrounding urban cores. Smailes trait complex described the extent of use of various core elements. As mentioned, these elements change as core and peripheral activities change. For example, banking was an element of centrality in the recent past (as was the church) but access to banking machines (ATM's), or even more recently, to internet banking has reduced the demand limits for basic banking services by providing intervening opportunities. Other activities continue to have relevance as core related elements, such as hospital services, newspaper circulation, theater, etc., though for how long is left to speculation. It is not difficult to see that there may soon be changes spatially in these elements as well. Still, more specialized types of banking services, such as commercial loans, have a greater degree of centrality than do basic services.

The delineation between town and country has never been absolute. With growing affluence and mobility, access to the city margins has increased and the spread of urbanites into villages is pervasive. Second homes 'in the country' are no longer the domain of the rich. As well, it is not uncommon for people to look outside of the core for goods and services. People in small towns may visit smaller towns for things not found elsewhere. Mobility has increasingly allowed for specialization without the need for spatial ordering. In the past, it would seem that differences between areas in terms of population concentration were more easily recognizable. It may be more difficult to find alternative explanatory variables during a time of accelerated change. Social indicators are less tangible and more qualitative than economic measures. Being less tangible, they include the possibility of different perceptions of social activities. For example, work is very much a social activity, yet many people would be inclined to look increasingly to recreation, the arts, and entertainment as social indicators. Certainly, it is never easy to justify an ‘eyeballing’ of the various indicators to be considered.

Nonetheless, it is relatively easy to test different possibilities using a telephone survey, and the use of economic and social indicators remains relevant. Broad concepts tend to nominalize to the exclusion of variation, while continued differentiation lacks reference. Yet the orientation towards marrying concepts with data is provides an opportunity to examine conditions and perhaps explain changes through time and between places. Therefore, a study of integration indicators must be broad enough to be inclusionary across a large area with diverse cultures, as well as being applicable through time.

 

3. Methodology

An attempt to analyze social and economic integration requires finding appropriate indicators, as well as determining the area of study while keeping in mind methods of collecting data and of measurement. The choice of indicators should follow guidelines that are developed locally as the value given to various elements may be spatially differentiated. Since these are not readily available, arbitrary choices are made. Once these are selected, there are several ways of conducting the study. In this case, a telephone survey was conducted of the people residing in selected census subdivisions (CSD) surrounding the core of interest. The percentage of the people in each area who are using the core for various reasons can then be compared with the place of work percentages. Each of these are described below beginning with the orientation that the study will take.

 

3.1 Orientation

While it is important to consider the core use of the periphery, this study attempts to look primary at the use of the urban core by residents in peripheral areas. This orientation neglects the extent of the use of the periphery by people living in the core. For example, cottages in surrounding areas, as well as other recreational areas used, indicate a two way relationship. This outward view follows the example of the 25 % ‘Reverse Commuting Rule’ (rule 3) found in the geography section of the 1996 Census Dictionary. Further studies may investigate this orientation perhaps using alternative indicators.

 

3.2 Spatial Elements

3.2 a) Urban-Rural Site Selection

The cities of primary interest are those which have approximately 100,000 people, and therefore are very near census metropolitan area (CMA) status. As well, city size makes a difference as studies show that smaller centers have different degrees of influence on surrounding areas. Larger centers tend to require labor from outside for functioning and also provide a high degree of specialized goods and services. In this study, Kingston, Ontario is chosen because it meets the population and density requirements, and its proximity reduces survey costs.

 

3.2 b) Areas surveyed

The selection of specific places to test indicators is meant to be comparable to the existing use of place of work in defining CMA's. Therefore, place of work data is examined for Kingston (see map 1). A tangent is drawn outwards from the core through census subdivisions which show a gradual decrease in the percentage of residents working in the core area. We are particularly interested in the change from above fifty percent to below fifty percent as well as further incremental decreases (see map 2).

Three principle jurisdictions are surveyed; one at 20 km, the second at 40 km, and the third at approximately 60 kilometers from Kingston. Telephone exchange prefixes (the first three of the seven digits in a phone number) are found for each of the distances along a vector. Each telephone exchange covers a varying amount of area but each is progressively further from the core. An approximation of the telephone 'catchment area is provided on map 2. These tend to grow in area as the population density drops.

Towns were also chosen which were of near equal size and near equal distance from major arteries to Kingston.

 

3.3 Indicators and Survey Questions

In order to see to what areal extent people use Kingston, several different indicators have been chosen. The two main elements to be considered are social and economic integration. In addition, reference to the core via information and communication exchange is included. In each case, a value is arbitrarily chosen as an estimator of inner and outer demand limits (examples are given below). The areal extent of the themes considered is important to this study but so are the frequency and durations of use. Since this survey is meant to be short and simple, and in an attempt to reduce imprecision, respondents were asked to recall the most recent use of each element.

 

3.3 a) Economic indicators

Economic indicators are relatively easy to find in comparison with social indicators. In this survey, economic indicators are divided into goods and services. The two questions covering goods are:

The choice of a thousand dollars is arbitrary but meant to recognize 'big ticket' items, and therefore help define an outer demand limit which would extend further from the core than a lower cut off might capture. As well, it is rounded off in order to be an easy reference for respondents.

For services, health care is chosen though there is not a direct economic exchange involved. Perhaps this indicator reaches more towards that of social integration, yet it is still an important core service which deserves attention. The questions are:

Health care with or without hospital services helps define inner and outer demand limits respectively and is hoped to be easy for respondents to remember.

 

3.3 b) Social indicators

For social integration, two main questions are also chosen. Since there are many ways to define 'social', specific indicators were avoided and instead the following questions were used:

By leaving the choice of what constitutes a social activity to the respondent major difficulties are avoided. The possibility of using a similar test across the country requires a generalized approach to the subject. In this manner, regional differences can be incorporated more easily. If specific indicators were selected, they may accurately characterize one area and completely fail in another. Nonetheless, for a social element the emphasis remains on human to human interaction and therefore an outer and inner demand limit has been arbitrarily set at groups of more or less than twenty people.

Information and communication exchange is also included in the survey as they might help define levels of integration. Phone calls to and from the core are examined as well as newspaper readership and radio listening habits:

For these questions, there is no demand limit required. Each is simply meant to indicate the amount of reference made to the core. Referring to Hagerstrands diffusion studies, telephone calls to the core tend to exceed those emanating from the core. It is thought that rural residents may call the core for information on goods while return calls are fewer.

 

3.4 Measures

In order to make a survey of spheres of influence easily tabulated, Rowley (1967) used five questionnaires per square kilometer with twenty questions on each giving a total of 100 percent. Cut offs were set at 50% for loss in dominance and 1% as the outer limit of the influence.

 

Figure 1. Measuring integration.

 

While it is likely that many people play outside of the core, the example is meant to show the possible spatial variation in extent of types of use of the core. If a cross-section were made through the core outwards, the levels of use of the core might be measured in relation to distance. If the level of use of the core varies between indicators, the results might suggest that there is integration of other sorts.

 

Figure 2. Scattergram of integration indicators in relation to distance to the core.

Using the example above, it is possible to assert that work is not as good an indicator of integration as is play. In order to simplify the survey further and to avoid the detail required in employing a full cross-sectional approach, CSD’s are chosen which are just below the 50% commuter rate and outward, therefore meeting the requirement highlighted in Figure 1.

 

3.5 Distances and Mapping

The distances between the selected sites were calculated using a UTM projection of the Kingston area (section 18). The map of the area with csd's (map 2) uses this projection with a scale bar extending outwards from the outer edge of the urban area. Map 3 is a scanned image of a road map which is not as reliable for distances but shows the selected sites in relation to the closest arteries. Sites along two cross-sections are indicated.

 

3.6 The Survey

3.6a) Respondents

In order to be comparable with the 'place of work' data, the respondent must be 15 years of age and over and be a non-institutionalized resident. The importance of reducing the under-representation of individuals in large households is somewhat offset by the reduction of calls to apartments with fewer people, as it is most likely that single family dwellings will be chosen outside of the core area.

 

3.6 b) Telephone Procedures

A telephone survey is chosen as it is relatively easy to perform and can be done in a short period of time. While it is quick, it is not thorough, and therefore requires many inputs to help reduce spurious results. Telephone numbers were randomly chosen from the selected municipalities outside of the urban core area. The survey was conducted in the evening to catch the day workers. An attempt was made to get a minimum of twenty five respondents per area.

 

3.6 c) Question ordering

As mentioned earlier, the questions cover a range of topics meant to measure social and economic integration. In order to allow the respondent to have an adequate comfort level with the types of questions, they were ordered in a manner thought to be the least offensive, and or, worrisome. Therefore, questions begin with newspaper and radio source and leave health and purchase information towards the end. The questions move quickly as they require simple yes and no answers. The last queries regard social activities. The questions are wide ranging enough and yet appropriate for a graduate studies research topic.

 

4. Results and Discussion

4.1 Response

Though the survey was short and easy to conduct, response rates were not as high as hoped. The telephone exchange numbers used for the random selection process were out dated (1997). This resulted in many numbers no longer in use. As well, many of those called were not interested in participating in a survey. Rates of non interest ranged

considerably. A rate of response was calculated for each study area, i.e.

 

with the following variation

Table 1. Rate of Response

Town

Rate of Response

Distance to Kingston (km)

Camden

0.6

20

Enterprise

0.9

40

Arden

1.2

60

 

The decision to not participate seemed to hinge on the subject. People not interested in discussing Kingston, said as much. A future survey should reword the introduction to not specify the core area of interest.

 

4.2 Significance levels

With only twenty five respondents in each area, the validity of the survey is questionable. The coefficient of variance was high for all sites, from thirty percent to Sharbot Lake at nearly sixty percent. This is not unexpected as there may be respondents who do not refer to Kingston, while others are regular users. Indeed, variance should increase with distance from 'the core'. The variability may be great, but the average value should indicate a trend in relation to distance to the urban center. It is difficult to compare the sample size with that of the population for each exchange area, as the number of residents in the telephone catchment area requires more time then is available to calculate. As well, some respondents use the core often and may even have a high preference for Kingston's elements but were missed by the survey since Kingston was simply not the most recent place that they utilized. This point highlights the need for a large survey sample. Another option is to conduct another random survey of the same sites at a later date. If this were to be done, the respondents who participate should be warned of the eminent possibility.

The coefficient of variance were as follows:

Table 2. Coefficient of Variance

Camden East

Enterprise

Arden

0.30

0.41

0.36

Sydenham

Godfrey

Sharbot Lake

0.32

0.40

0.59

 

4.3 Areal Extent of Various Indicators

The average response for each of the indicators was calculated and entered into a summary table and a graph was generated (see Figure 3). The average use of the core was highest for those living closest Kingston and dropped off gradually for most of the indicators tested. Within each area, there was a considerable amount of variation, from those who seldom use the core to those who regularly 'go to town'. The extent of the radio listening was revised for the first cross section. Many responces had not been tabulated correctly originally, but the CBC does have a broadcasting station in Kingston.

This section is made up of two parts - the first refers to the Graphs in Figure 3, while the second refers to Map 4(Parts A & B) which estimates the extent of dominance of Kingston for the tested indicators.

 

4.3 a) Information and Communication

The results of the first cross-section (Camden East etc.) seem to be the most well behaved. All the percentages drop off nicely further from the core area. Yet curiously, newspaper readership extends farther than does radio listenership. As well, the number of 'telephone calls from' is higher than the number of 'telephone calls to' the urban core. In the second cross-section these results were reversed. An explanation for the change in the radio results has been offered in previous section. Still, the results from Godfrey seem unusually high.

'Phone calls to' and 'from' may be for any number of reasons. For example, planning social activities, getting information about purchasing goods and services, as well as health care etc. are apparent uses. It would seem therefore, that phone call patterns would be a good indicator of social and economic integration. Yet the results seem to show that phone calls of both types are similar to social extent.

4.3 b) Health Services

The activity with the greatest extent of reference to the core area was for major health care services. The type of service, whether hospital or tertiary, was not mentioned in the question starting first with the Godfrey calls yet the results seem very similar through out. The use of the hospital in Napanee was mentioned by some of the respondents in the Camden East area and may explain the lower percentages there. Here, the major service does extend much further than does the minor service. As well, the spacing between major and minor seems to be fairly constant.

For those respondents who were 'disgustingly healthy' and chose to register a Not Applicable answer, a no answer was entered instead.

The most recent major health service may have been received many years ago. Frequency seems to be negatively correlated with extent. For those activities which are more frequent the extent of reference to the core is considerably reduced.

4.3 c) Goods

For both of the cross-sections this activity seems to be similar except for the high results from Godfrey. The only variation in the method of conducting the survey which occurred for this area, was the timing of the calls. They were made principally in the afternoon. Perhaps, shopping in Kingston had occurred in the mornings of the days called?

In both cross-sections, the percentage of lower priced purchases was higher than the more expensive items. Still, they were for the most part very similar except for the notable difference in the Godfrey results. Shopping in the core may not have been measured here though. In fact, the proliferation of box store on the fringe of large urban areas tends to attract people from both within and outside. Many respondants did mention box stores as having been the last place that a purchase was made.

 

4.3 d) Social Activities

Of all the indicators surveyed, social activities has the lowest extent. The larger the group, the greater the extent, except for Enterprise, where the responses were the same.

The question was refined some what to include the suggestion of 'social activities of any kind' but this does not appear to have had an influence on the results. No other indication of what constituted a social activity was offered, yet this question was received very well. Some people asked what a 'postal activity' was, not having heard the question clearly. The reference to the cut off number of twenty seems to have helped people focus on what they thought was a social activity. Nonetheless, it would be interesting to find out what types of activities people had considered to be 'social'.

 

4.4) Spheres of Influence

A line, extending across the graphs at the 50% mark were used to estimate the point along the cross-section where loss of dominance of the core occurred. The results, though obviously estimates, do show some interesting results. Within 10 to 30 km from Kingston lay the extents for minor health care, phone calls of both types, and all social activities. Outside of the forty km line are found goods purchases above and below $1000, newspaper and radio, and major health care.

The apparent relationship between newspaper and radio and goods purchasing is interesting. Obviously, for the media provider, the cost of delivery should not exceed the use of advertising by too great a degree.

Telephone calls and social activity seem also to be closely matched. There maybe a strong relationship between these activities. A quick survey of family and friends suggests that social phone calls are at least twice as frequent as others.

The map shows that the 'phone calls to' and 'from' change in extent between the two cross-sections. 'Phone calls to' would be expected to excede that of 'phone calls from' the core area as retailers and services may phone out less often then do people phone in. Yet for the first cut, 'phone calls from' excede those 'to' Kingson. Two respondants mentioned that they had just received calls from Kingston from a fund raising drive.

The extent of minor health care resembles that of phone calls and social activities. For the second cross-section, this indicator reachs further afield. Again, there was mention of the use of Napanee for those in the Camden East area.

Perhaps, between 30 - 40 kilometers out from the core there is a threshold where frequency of use begins to influence the results. Shopping, may be done daily but it may also be done weekly or longer, while newspaper and radio may also vary in frequency. As well, the influx of large whole sale type 'box stores' located on the fringe of larger cities such as Kingston may be influencing the purchasing spheres.

Major health care services are the least frequent and also the furthermost. Yet, minor health care is not necessarily a frequently used service. Nontheless, local practioneers may be able to effectively compete with Kingston providers.

Local provides of goods do not seem to be as capable of competing with Kingston retailers though. This indicator , along with radio and newspaper, reaches far afield and raises the question of the adequacy of 'place of work' as an apropriate measure for deliniating a central metropolitan area. Of course, the place of work data is from the 1996 census while this survey shows results from the year 2000. Nonetheless, there is a difference in the extent of use of core facilities.

Interestingly, the inner extents (phone, social, minor health care) match the 50% place of work data from 1996. Therefore, these may be good surogates for the 'place of work' data.

 

 5. Summary

The chosen socio-economic integration indicators show that there is a difference in the extent to which core facilities and services are utilized by residents in outlying areas. The place of work variable may be the strongest indicator of social and economic integration, but the results of this survey suggests that social activities (with both small and large groups) are satisfied much more locally than are the purchase of goods (for both low and high values). The map of the results of the extent of use of the core suggests that the place of work is a conservative estimate of the sphere of influence. But economic integration is not simply made up of the purchase of goods; the exchange of labor is also important to consider. As well, work may not have been included in the rationale for deciding what constituted a social activity. Yet clearly, work is a social and an economic activity. A diagram based on 'Christaller's rural networks spheres of influence' (see Figure 4) helps show the difference in the extent of the use of the core for the indicators tested (see Figure 5).

The use of major health care services had the greatest range. The use of this variable for delineating a CMA would likely not be appropriate as the frequency of use of the core is very low. Radio and newspaper use could also be employed but they represent more of a use of the core rather than as an indicator of integration. Many of the residents in the rural areas also read the Toronto papers. An arguement could be made for the utilization of the purchase of goods, but if the prerequisite is socio-economic integration, then the purchase of goods would appear to fall short as it is a use of the core rather than an integration indicator. An equally strong arguement could be made for the use of social activities (as defined by the respondants) but this alone would not satisfy the prerequisite either.

The indicators used were meant to break open the terms social and economic integration. The survey looked at the use of the core area by residents living at varying distances from the center. The extent of the indicators differed and the results were fairly evenly matched in both cross-sections. Yet, it is possible that the indicators used could be broken open once again to more specific activities. It is likely that further refinement in this regard would result in further variation in extent. But for an indicator of both social and economic integration, place of work appears to be inclusive and therefore an entirely justifiable compromise. Nonetheless, there are gradations between urban and rural which must take into consideration distance and time.