<%BANNER%>

Geotemporal Trade Area Evaluation with Dynamic Gridding to Avoid the Modifiable Area Unit Problem (maup)

Permanent Link: http://ufdc.ufl.edu/UFE0041831/00001

Material Information

Title: Geotemporal Trade Area Evaluation with Dynamic Gridding to Avoid the Modifiable Area Unit Problem (maup)
Physical Description: 1 online resource (142 p.)
Language: english
Creator: Dietz, Ron
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: choropleth, fractal, geotemporal, gis, grid, gridding, map, maup
Geography -- Dissertations, Academic -- UF
Genre: Geography thesis, M.A.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Changes within a trade area of a retailer can change the profitable performance of the business. Among the most important changes to a trade area over which the retailer has no control are changes to the demographic composition of households, the entry and exit of competitors, and the relocation of competitors within the trade area. Geospatial technology provides for cost effective evaluation of changes in the phenomena that affect the trade area. This provides an opportunity for the retailer to document changes that have occurred and evaluate how those changes affect their enterprise. Important geotemporal measures are customer count change by location, transaction change by location, and revenue change by location. Sustainable best practices of a business firm requires awareness of the trajectory of change and thereby enable the firm to appropriately adjust to current and anticipated market conditions. A repeatable methodology is presented that measures changes in business activity across space and through time. A case study of a firm providing retail services in Alachua County, Florida, documents the methodology. This study indicates that further research on geotemporal analysis for business decisions is warranted, especially that which has focus on the Modifiable Areal Unit Problem (MAUP). MAUP reveals if the geographic container for the data is itself a source of a statistical bias radically influencing both qualitative and statistical results.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Ron Dietz.
Thesis: Thesis (M.A.)--University of Florida, 2010.
Local: Adviser: Thrall, Grant I.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041831:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041831/00001

Material Information

Title: Geotemporal Trade Area Evaluation with Dynamic Gridding to Avoid the Modifiable Area Unit Problem (maup)
Physical Description: 1 online resource (142 p.)
Language: english
Creator: Dietz, Ron
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: choropleth, fractal, geotemporal, gis, grid, gridding, map, maup
Geography -- Dissertations, Academic -- UF
Genre: Geography thesis, M.A.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Changes within a trade area of a retailer can change the profitable performance of the business. Among the most important changes to a trade area over which the retailer has no control are changes to the demographic composition of households, the entry and exit of competitors, and the relocation of competitors within the trade area. Geospatial technology provides for cost effective evaluation of changes in the phenomena that affect the trade area. This provides an opportunity for the retailer to document changes that have occurred and evaluate how those changes affect their enterprise. Important geotemporal measures are customer count change by location, transaction change by location, and revenue change by location. Sustainable best practices of a business firm requires awareness of the trajectory of change and thereby enable the firm to appropriately adjust to current and anticipated market conditions. A repeatable methodology is presented that measures changes in business activity across space and through time. A case study of a firm providing retail services in Alachua County, Florida, documents the methodology. This study indicates that further research on geotemporal analysis for business decisions is warranted, especially that which has focus on the Modifiable Areal Unit Problem (MAUP). MAUP reveals if the geographic container for the data is itself a source of a statistical bias radically influencing both qualitative and statistical results.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Ron Dietz.
Thesis: Thesis (M.A.)--University of Florida, 2010.
Local: Adviser: Thrall, Grant I.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041831:00001


This item has the following downloads:


Full Text





GEOTEMPORAL TRADE AREA EVALUATION WITH DYNAMIC
GRIDDING TO AVOID THE MODIFIABLE AREA UNIT PROBLEM (MAUP)















By

JAMES RONALD DIETZ


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF ARTS

UNIVERSITY OF FLORIDA

2010



























2010 James R. Dietz
































To my Mother and Father: I am eternally grateful for your loving support and generosity
throughout the years.









ACKNOWLEDGMENTS

I would like to acknowledge my advisor and mentor, Grant lan Thrall, for his

enduring patience, support, and direction during my pursuit of a Master of Arts degree.

He has been instrumental in guiding me toward this goal and can take full responsibility

for this timely accomplishment. He initiated my interest in the Marketing Geography

discipline and has supported my intellectual growth with a progressive movement

delving into the dynamic realms of the MAUP and Fractal Geometry. I could not ask for

a better advisor, as he has given me the academic guidance and moral support

necessary to make completion of this thesis a most rewarding and challenging

experience.

The business that provided the data used in this study prefers to remain

anonymous. Their cooperation allowed the topic of this thesis to be feasible. My

commitment to the firm was that their time and effort would be rewarded by the results

of the analysis of their data being both interesting to them, and relevant to their

managerial decisions. This commitment was important in designing the work and the

presentation of the completed analysis.

I would also like to acknowledge the rest of my committee-Timothy Fik and

Youliang Qiu-for their continued support and enthusiasm, I am truly grateful. With the

combined efforts of everyone involved, especially those of my committee members, I

am approaching the completion of this thesis. Their efforts have made it a truly

enjoyable and rewarding experience, for that, I am forever indebted.









TABLE OF CONTENTS

page

ACKNOW LEDGMENTS ........................................ ............... 4

LIST O F TA BLES ......... ................ ..................... ...... ............... 7

LIS T O F F IG U R E S .................................................................. 8

A BSTRA CT ........... ......... ..... ..... ..... ...................... ......... 11

CHAPTER

1 INTRODUCTION .................. ...... ......... ......... 13

2 TRADE AREA CALCULATION: AN OVERVIEW ............. ... ............... 17

Marketing Geography ...... .................. ......... ......... 17
Value and Justification ...... .................. .................. 21
Tool and Methodologies.......................... ......... ............... 22
Term inology .......................................................................... ........ .................. 22
Methods................................................ .. ........ 23
Customer spotting method ............................ ........... 23
Trade area delineation .................. ........................ 27
S u m m a ry .............. ..... ............ ................. ............................................... 2 8

3 TRADE AREA CALCULATION: AN INTEGRATED GIS APPROACH.................. 29

GIS as a Marketing Geography Research Tool ............... ...... .................. 29
Historical O verview .............................. ............... 29
Customer Spotting Method .................. ......................... 31
Demographic Application........................... ........ 33
S u m m a ry .............. ..... ............ ................. ............................................... 3 4

4 TRADE AREA CALCULATION: METHODOLOGY................................ 35

The Core Trade Area Radial Method ................................... ....... 38
The Grid Method ............... ......... ................. 42

5 TRADE AREA ALGORITHM................................ .................... 79

6 TRADE AREA CALCULATION: GEODEMOGRAPHICS.......................... .... 95

7 TRADE AREA CALCULATION: FUTURE CONJECTURE .............................. 99

8 TRADE AREA CALCULATION: CONCLUSION ............ ... ............... 102









APPENDIX: COMMUNITY TAPESTRY SEGMENTATION SUMMARY
D ESC R IPT IO N S .......... ................. ......... ...... ................ .............. 104

LIST OF REFERENCES ............... ........................................... 138

BIOGRAPHICAL SKETCH ............................................................ 142















































6









LIST OF TABLES

Table page

5-1 Total revenue statistics (1x1 mile grid cell overlay). ................. ................... 84

5-2 Average revenue statistics (1x1 mile grid cell overlay). .................... .......... 85

5-3 Customer count statistics (1x1 mile grid cell overlay) ................................ .. 86

5-4 Transaction statistics (1x1 m ile grid cell overlay)........................ ............ ... 87

5-5 Total revenue statistics (1.5x1.5 mile grid cell overlay). ............... ................ 88

5-6 Total rev. change stats (1x1 mile grid cell overlay offset foci 1,2, and 3)............ 89









LIST OF FIGURES


Figure page

4-1 Data manipulation and methods flow chart............ ......... ............... 43

4-2 Alachua county customers 2004 directional distribution (standard deviational
ellipse). .................... .... ...4......... ........ ..... .. ..........44

4-3 Alachua county customers 2004 standard distance. ............... ............... 45

4-4 Alachua county customers 2008 directional distribution (standard deviational
ellipse). .................... .... ...4......... ........ ..... .. ..........46

4-5 Alachua county customers 2008 standard distance. ............... ............... 47

4-6 Alachua county customers 2004+2008 directional distribution (standard
deviational ellipse) ...... .. ...... .. .......................... ............... 48

4-7 Alachua county customers 2004+2008 standard distance. ............................. 49

4-8 Alachua county customer growth and decay zones overlay of the directional
distribution (standard deviational ellipse) polygons by dataset year group......... 50

4-9 2004 and 2008 Customers in alachua county, fl. directional distribution
(standard deviation ellipse), symmetrical difference overlay. ........................... 51

4-10 Alachua county customer growth and decay zones overlay of the standard
distance polygons by dataset year group. .................. .. .............. 52

4-11 2004 and 2008 Customers in alachua county, fl. standard distance,
symmetrical difference overlay. ............................. .................. 53

4-12 2008 Publix supermarket locations (1.5 mile buffer) .......................................... 54

4-13 2004 Alachua county customers within 1.5 miles of the centroid.................... 55

4-14 2008 Alachua county customers within 1.5 miles of the centroid.................... 55

4-15 2004+2008 Alachua county customers within 1.5 miles of the centroid. ............ 56

4-16 2004+2008 All eight new competitor locations appearing from 2004. ............. 56

4-17 New competitor location number 1 appearing from 2004 to 2008. ................... 57

4-18 New competitor location number 2 appearing from 2004 to 2008. ................... 57

4-19 New competitor location number 3 appearing from 2004 to 2008. ................... 58









4-20 New competitor location number 4 appearing from 2004 to 2008. ....... ........ 58

4-21 New competitor location number 5 appearing from 2004 to 2008. ....... ........ 59

4-22 New competitor location number 6 appearing from 2004 to 2008. ....... ........ 59

4-23 New competitor location number 7 appearing from 2004 to 2008. ....... ........ 60

4-24 New competitor location number 8 appearing from 2004 to 2008. ....... ........ 60

4-25 First moving competitor appearing from 2004 to 2008, (2004 location)............. 61

4-26 First moving competitor appearing from 2004 to 2008, (2008 location)............. 61

4-27 Second moving competitor appearing from 2004 to 2008, (2004 location)......... 62

4-28 Second moving competitor appearing from 2004 to 2008, (2008 location)......... 62

4-29 Third moving competitor appearing from 2004 to 2008, (2004 location) ............ 63

4-30 Third moving competitor appearing from 2004 to 2008, (2008 location) ............ 63

4-31 Fourth moving competitor appearing from 2004 to 2008, (2004 location). ......... 64

4-32 Fourth moving competitor appearing from 2004 to 2008, (2008 location). ......... 64

4-33 2004 Total revenue (dollars), 1x1 mile grid cell. .......... ............. ............... 65

4-34 2008 Total revenue (dollars), 1x1 mile grid cell. .......... ............. ............... 66

4-35 2004 Average revenue per transaction (dollars), 1x1 mile grid cell ................ 67

4-36 2008 Average revenue per transaction (dollars), 1x1 mile grid cell................... 68

4-37 2004 Number of Transactions, 1x1 Mile Grid Cell. ...................... ................ 69

4-38 2008 Number of transactions, 1x1 mile grid cell. .......... ............ ................ 70

4-39 2004 Number of customers, 1x1 m ile grid cell................................................. 71

4-40 2008 Number of customers, 1x1 mile grid cell................................................. 72

4-41 Total revenue change from 2004 to 2008, 1x1 mile grid cell ............................. 73

4-42 Total revenue change from 2004 to 2008, 1.5x1.5 mile grid cell ...................... 74

4-43 1x1 Mile grid cell coordinate plane. ............ .............................. ............... 75

4-44 1x1 M ile grid cell overlay offset m ethod.................................. ..................... 75









4-45 Total revenue change from 2004 to 2008, 1x1 mile grid cell (first foci, origin).... 76

4-46 Total revenue change from 2004 to 2008, 1x1 mile grid cell (second foci, 1
m ile offset)................................................... ................... ........ ..... 77

4-47 Total revenue change from 2004 to 2008, 1x1 mile grid cell (third foci, 12 mile
offset). ................................................................ ....... .. ..... 78

5-1 Total revenue change from 2004 to 2008, hot spot analysis (getis-ord gi*)
1x1 m ile grid cell.............................................................. ......... 90

5-2 Total revenue change from 2004 to 2008, hot spot analysis (getis-ord gi*)
1.5x1 .5 m ile g rid ce ll............................................. ........ ... ........ 9 1

5-3 Total revenue change from 2004 to 2008, hot spot analysis (getis-ord gi*)
1x1 m ile grid cell (first foci, origin). ............................................ ............... 92

5-4 Total revenue change from 2004 to 2008, hot spot analysis (getis-ord gi*)
1x1 mile grid cell (second foci, 1/4 mile offset). ....................... ................ 93

5-5 Total revenue change from 2004 to 2008, hot spot analysis (getis-ord gi*)
1x1 m ile grid cell (third foci, 1/2 m ile offset)............. .. .......... .. ................. 94

6-1 Highest percentage tapestry profile group mean center (2008) and total
revenue change from 2004 to 2008, 1x1 mile grid cell overlay........................... 98

7-1 M odeling m aup flow chart ....................................................... ................... 101









Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Arts

GEOTEMPORAL TRADE AREA EVALUATION WITH
DYNAMIC GRIDDING TO AVOID THE MODIFIABLE AREA UNIT PROBLEM (MAUP)

By

James Ronald Dietz

August 2010

Chair: Grant lan Thrall
Major: Geography

Changes within a trade area of a retailer can change the profitable performance

of the business. Among the most important changes to a trade area over which the

retailer has no control are changes to the demographic composition of households, the

entry and exit of competitors, and the relocation of competitors within the trade area.

Geospatial technology provides for cost effective evaluation of changes in the

phenomena that affect the trade area. This provides an opportunity for the retailer to

document changes that have occurred and evaluate how those changes affect their

enterprise. Important geotemporal measures are customer count change by location,

transaction change by location, and revenue change by location. Sustainable best

practices of a business firm requires awareness of the trajectory of change and thereby

enable the firm to appropriately adjust to current and anticipated market conditions.

A repeatable methodology is presented that measures changes in business activity

across space and through time. A case study of a firm providing retail services in

Alachua County, Florida, documents the methodology. This study indicates that further

research on geotemporal analysis for business decisions is warranted, especially that

which has focus on the Modifiable Areal Unit Problem (MAUP). MAUP reveals if the









geographic container for the data is itself a source of a statistical bias radically

influencing both qualitative and statistical results.









CHAPTER 1
INTRODUCTION

"Real time" management decisions are critical to the long term success of large

and small businesses. Large multi-branch retail chains have for the past two decades

been beneficiaries of geospatial analysis and geospatial technology. Cost of

implementation and limited general public knowledge of "location based intelligence"'

has contributed to smaller locally owned businesses from reaping the same benefits as

large business. This thesis demonstrates that geospatial analysis provides management

of a small business, information that is critical to the business decision in "real time2"

This thesis integrates a triad of concepts, thereby adding time critical to mainstream

business geography literature:

retail market/service provider location trade area theory
geospatial technology
time

"Real time" geospatial analysis, hereafter referred to as geotemporal analysis,

identifies entry and exit of competitors, tracks customer interaction with the store, and

evaluates change in the non-customer population of the trade area. Geospatial changes

can lead to cannibalization of revenues, a change in market penetration, and prospects

to improve the market position.





1 Location Based Intelligence (LBI) allows for real time tracking of information to alert the operator and
provide an enhanced perspective of situational awareness that can help improve judgment decisions.
http://works.bepress.com/mgmichael/4/ ... http://www.encyclopedia.com/doc/l G1-199973425.html
2 Real time is when things respond to events as they occur, and may refer to: Real Time Locating
Systems (RTLS), are used to track and identify the location of objects in real time using simple,
inexpensive nodes (badges/tags) attached to or embedded in objects and devices (readers) that receive
the wireless signals from these tags to determine their locations. RTLS typically refers to systems that
provide passive (automatic) collection of location information.









Trade area calculations are primarily used in retailing to calculate expected store

performance before the stores "open their doors." The benefits of predicting store

performance include avoidance of financial losses, zeroing in on profitable

opportunities, and not to be underestimated is the ability to negotiate a more

advantageous long term lease or purchase price of a site.

After transaction, the choice of site is fixed. In the short run, managerial decisions

cannot be made on fixed costs. The store managers' degree of freedom subsequently

narrows to marketing and some variation in the value platform offered by the business.

"Real time" evaluation of a trade area has not entered the academic or practitioner

literature. This absence is not because of lack of importance; rather, the data has not

been available to academics to create a study of the type necessary for a real "real

time" demonstration. Businesses that employ the technology to be described here are

not sharing the benefits of the analysis, and not letting the world know about the

existence of their "Geotemporal War Room." Indeed, real time information is internally

recognized as critical to the success of the enterprise. Real estate has been said to be

"information arbitrage (Thrall, 2002)." The belief among stakeholders is that those with

the information can sustain their enterprise into the future. Those without the information

are "throwing dice with their future."

It is fortunate that a successful, locally owned medical service provider gave access

to the data used here. The data made available is comprised of two time periods, each

period being a full year, and with a four year interval separating the two time periods.

The methodology of this thesis can be extended to data sets with shorter time

differences: hourly, daily, and monthly. Knowledge gained from the two time period









analysis will contribute to further research on geotemporal business decisions. Alachua

County provides an ideal test site. Gainesville is a comparatively compact city. Traffic

congestion keeps the city compact. Most employment is at a central location. For these

reasons, the land use and urban form of Gainesville is not complex and mirrors the

general theory on urban spatial structure (Thrall, 1987). It is also fortunate that the

business that provided access to the data was a small locally owned enterprise. Large

multi-branch retail chains have benefited from geospatial analysis and technology. The

analysis here documents that geospatial analysis can also benefit the small business.

Adoption of geospatial analysis by small business is expected to occur, in large

measure by the ready and nearly ubiquitous acceptance of Google Earth and Google

Maps. Google at this time does not offer the geospatial functionality required for this

analysis; however, it is inevitable that the functionality is offered. In the meantime, the

necessary functionality is available via specialized software packages.

The technology employed in this thesis is ESRI Business Analyst, R-Statistics, and

Caliper Corporation's Maptitude GIS. The particular strength of Business Analyst is its

technology for geocoding and assignment of lifestyle segmentation profiles (this will be

explained a later part of this thesis). Maptitude GIS software is capable of carrying out

the spatial data operations, and spatial visual operations required in this thesis.

The customer data is from a locally owned, privately developed and owner managed

business. The business is a medical service provider. Service consumers make an

appointment, travel to and from the 'brick and mortar" facility. Familiarity with and trust of

the research team was a necessary precondition for this work to be performed. A needs

assessment was casually executed, without intruding with business operations. The









needs assessment revealed that the business had been maintaining a computerized

database as part of its scheduling and billing operations. However, the data had not

previously been considered an asset, and the management had never seen their own

data as geospatial. Google's geospatial capabilities wetted their appetites.

The data was received without any monetary transaction. A non disclosure

agreement was signed committing that the customer addresses and identities remained

anonymous, and that the business remained anonymous as well. The business did

however incur a cost due to disruption of normal routines, and allocation of resources

for data extraction from their transactions files. This effort was not insignificant, and an

indication of the importance in which the business management considered this study.

The quid-pro-quo is that the business facilitating this study by allowing access to its

data, receive a copy of the results in return.

The data covered the years 2004 and 2008. This time period too was fortunate for

the study though not for the economy. As the national economy entered a recession it

would have been expected that the decline would transfer into a decline in demand for

the goods and services offered by the case study business. In other words, as income,

expected income, and consumer confidence declined, it would be reasonable to expect

that the revenues would change as well. The more inelastic the consumers demand, the

less changes in the national economy would translate into local demand changes.

The data included address of the customer, count of each customer's transactions

(visits), revenue generated by each customer within the time frame. Customer

addresses gives rise to lifestyle segmentation profile derivation via ESRI Business

Analyst.









CHAPTER 2
TRADE AREA CALCULATION: AN OVERVIEW

Marketing Geography

The foundations of Marketing Geography can be traced to literature from the

early 1900s where practitioners first began utilizing the methodologies now being used

within this field. Geographers have long known that there are causal determinants to

many spatial distribution patterns. With a combination of quantitative and qualitative

analysis, the cause of spatial patterns can be detected, hypotheses formulated, and

statistically confirmed or refuted. These steps are important to predicting future spatial

distributions of phenomena, or spatial distributions of phenomena elsewhere. When the

phenomena is important to the success of a business, then the business must include

the present and predicted spatial distributions when making time critical operational

decisions and creating long term managerial strategy.

"All retailing operations are complex but most executives would agree that

location (and its associated attributes) contributes more to the long-term success of the

retail unit than does any other factor" (Ritchey, 1984). Additionally, the optimal location

of services and the significance of a location strategy is "...to help assure a successful

[business] undertaking" (Mercurio, 1984). The earliest attempts to apply geographic

research techniques to retail outlets date to the 1920s when the attempt was made to

determine, on behalf of large multi-branch retail firms, the relative value of one site,

compared to the relative advantages of other sites in the same region. These first

attempts and early studies employed methodologies that were mainly subjective and

based on factors believed to be significant (Goldstucker et al., 1978). Applebaum

suggested that the origins of Marketing Geography may be traced to the beginning of









the twentieth century when chain companies-especially tobacco shops-began to

conduct detailed surveys of pedestrian flows along streets in order to identify the most

desirable sites (highest foot traffic) within the main centers of towns (Davies, 1977).

These early studies were marginally effective at improving the business decision-

making process, but they were the first real use of the foundations used in today's

business geography techniques.

William Applebaum is widely recognized as a pioneer in the discipline of

Marketing Geography and is described by Davies (1977) as the chief architect of

Marketing Geography as a separate field of study in the United States. "Store location

research, as both an academic and practical area of inquiry, owes much to the

formative work of William Applebaum. From about 1945 until his death in 1978,

Applebaum was a colossus amongst those teaching the new subject of Marketing

Geography, lecturing at the Harvard Business school" (Davies and Rogers, 1984).

Although Applebaum had a degree in geography from the University of Minnesota, his

work was not widely accepted or valued within the discipline of geography.3 Turning a

bad thing into an opportunity, Applebaum gravitated to an audience that was highly

receptive business practitioners and business academics. Although his contributions,

as well as those other geographers from this era, were mainly cartographic

representations of market areas, Applebaum was instrumental in opening the avenues

of thought that created Marketing Geography as an accepted discipline within the field


3 It might be asked, "why did the Association of American Geographers" wait until 2008 to create its
Business Geography Specialty Group (see www.BusinessGeography.info ). Marketing Geography is a
subset of the larger emerging literature on Business Geography. The focus of Business Geography is
geospatial analysis to improve the business decision, while the focus of Marketing Geography is
essentially retail location. This thesis brings information relevant to Marketing Geography and analyzes
that information to be relevant to the large array of decisions that must be made in business.









of geography itself. Among the list of contributions made by Applebaum, he is well

known for the customer spotting method, the analog method, methods for determining

market penetration, methods for determining store rents, and store site evaluation. His

understanding of both the business/marketing perspective and the geographic concepts

allowed Applebaum to become recognized as the expert in his field during his own time.

Applebaum is regarded as one of the 'fathers' of Marketing Geography, and one of the

most important geographers of the 20th century.

Another pioneer in the advancement of the Marketing Geography discipline is

David Huff, whose work with spatial interaction models was a revolutionary concept for

the business community. Huff was a business student studying marketing at University

of Washington during the late 1950s. There he participated in "brown bag" lunches with

geography graduate students. Upon graduation he became a faculty member at

University of California at Los Angeles.4 There, and independent of his geography

colleagues, he formulated what is today regarded as "the Huff model." Huff's formulation

is the kingpin of spatial interaction models used in applications ranging from marketing

geography to transportation studies. Initially, Huff's work was not readily accepted in

mainstream geographic thought, but instead became one of the most studied

formulations in Regional Science which was at the interface between the disciplines of

economics and geography.5 Huff also purposely disseminated his work to business

practitioners by publishing examples of his work in trade publications. As Applebaum

did before him, Huff found an enthusiastic audience that was receptive to the value of


4 Personal communication between David Huff and Grant Thrall, narrated to Ron Dietz.
5 For more on the Huff model and spatial interaction models in general, see Kingsley Haynes and A.
Stewart Fotheringham, 1984, Gravity and Spatial Interaction Models, Sage Publications: Beverly Hills CA.









his work. Huff presented purely geographic based concepts and methodologies to this

audience which helped to diffuse these Marketing Geography techniques outside of the

confines of geographic circles. Today, David Huff is retired from the Department of

Marketing at University of Texas at Austin, and is a consultant to Environmental

Systems Research Institute (ESRI).6

Early use of site location methodologies were encumbered by the use of

'checklists;' decisions made with the use of mere checklists were in the final analysis

considered, and likely were, highly subjective (for examples see Nelson 1958;

Goldstucker et al., 1978; Applebaum and Cohen, 1960). The perceived absence of a

sound theoretical framework limited the academic acceptance of marketing geography

within the geographic discipline. The quantitative revolution of the 1960s would change

this. Marketing geographers were quick to apply regression analysis, and include the

"checklist" fields as independent variables. Revenues per square foot were one of the

more common dependent variables in the regression equation. Marketing Geography

subsequently gained high respect within the discipline of geography, and became

accepted as a distinct scientific subfield of geography. However, as the body of

marketing geography knowledge increased, and as more and more businesses adopted

marketing geography methodologies, the newly emerging geography academics were

drawn to Colleges of Business in which it had originated.7 Nevertheless, sound


6 Among Professor David Huffs consulting activities is overseeing the Business Analyst add-in to ESRI's
ArcMap. Business Analyst is a combination of data and software. In particular, Business Analyst includes
the Huff model, and the means to calibrate the Huff model with a firm's own data.
7 Avjit Ghosh is a very good example. He received his Ph.D. in Geography (1979), and MA in Geography
(1977), both from University of Iowa under the direction of Professor Gerard Rushton. Ghosh has been
one of the most important contributors to advancing the literature on spatial interaction models and their
use in business. He has been Dean of the College of Business at University of Illinois, and today remains
as a Professor of Business Administration at University of Illinois.









geographic analysis, geographic technology, geographic theory, when used to solve

actual business decisions, has evolved to form a body of knowledge of 'best practices'.

Value and Justification

Goldstucker et al. (1978) reasoned that retail market/service provider's trade

area size and shape is affected by: (1) the extent of product differentiation and the

relative effectiveness of brand promotion; (2) the range of choice in administering

pricing made possible by product differentiation, oligopoly, and other influences; (3) the

ratio of fixed to total costs; (4) the economies of scales of production at each center;

and (5) the availability of adequate markets within a radius of economical outreach.

Each of these contains an aspect of geography that is useful to the overall calculation of

a trade area, while others also include the intangible aspects of Marketing Geography.

"That shape has not been used more extensively in geography may result from the

earlier inability to measure it precisely. Many of the terms and measurements that have

been used to identify shape have been inadequate. Moreover, geographers have been

primarily concerned with an ideographic, rather than a nomothetic, approach to

geographical problems" (Boyce and Clark, 1964). Simmons (1984) states that if a store

is achieving a low share of trade, it is vital to establish the 'image' of the store compared

with selected competition to find out why people shop at on store rather than another.

The geographer must take these intangible aspects of marketing into account, but the

main focus for the geographer is on the locational attributes of a retail outlet that

determine and influence store patronage.

Location is crucial to the overall success of the retail outlet and can provide that

necessary competitive advantage, but retail demand consisting of the consumer's

demand for goods require a careful consideration of the positive and negative









externalities associated with doing business. This includes factors not normally taken

into account regarding the customer such as population distribution, household income,

race and ethnic characteristics, personal preferences and desires, purchasing habits,

etc. The successful marketing strategy attempts to describe how the individual retail

outlet performance rates in relation to the potential cannibalization and symbiotic

aspects of competition, the impact of the price of goods and quality of services

provided, as well as the influence of suitable location policy.

Retail performance is significantly shaped by the size and demographic

composition of an outlet's trade area; therefore, trade area is a spatial expression of the

limits to the firm's market potential (Ghosh and McLafferty, 1987). The marketing

geographer must use this information in completing an analysis as it is required to gain

a fully encompassing view of an inherently uneven spatial distribution of this retail

demand within a defined trade area in order to maximize the potential for future sales

and growth.

Tool and Methodologies

Marketing Geography has a body of established methodologies and techniques,

and some terminology that might be unique to the field. Therefore, in this section of the

thesis, a brief summary of terminologies used in Marketing Geography are presented.

Then, several methods illustrative of Marketing Geography, and used in this thesis, are

summarized.

Terminology

This section summarizes the various terms and concepts used by Marketing

Geographers and its associated literature.









MARKET SHARE is the ratio of a store's sales in a geographic region to the total sales
potential or otherwise defined as per capital sales divided by per capital sales
potential. Market share is also known as market penetration.

TRADE AREA of a retail market/service provider outlet is the spatial expression of its
market potential. It is the geographic area from which its customers are normally
drawn with an established benchmark of 80% capture of customers by location. It
is usually a function of distance, location of competitors, accessibility, and size
(square footage) of the retail establishment.

TRADE AREA DELINEATION is the process through which the trade are of a retail
market/service provider trade area of a store is spatially demarcated or the area
on a map where the majority of customers are drawn.

DIRECTIONAL DISTRIBUTION (STANDARD DEVIATIONAL ELLIPSE) measures whether a
distribution of features exhibits a directional trend (whether features are farther
from a specified point in one direction than in another direction).

STANDARD DISTANCE measures the degree to which features are concentrated or
dispersed around the geometric mean or median center.

SYMMETRICAL DIFFERENCE is a visual depiction of the area not covered when two
dissimilar polygons are overlaid on top of one another.

MEAN CENTER identifies the geographic center (or the center of concentration) for a set
of features.

CENTROID is the geometric center of mass for the polygon being studied.

Methods

Customer spotting method

With the intention that Marketing Geography become academically accepted, and

compete on a equal footing with other subfields of geography as well as economics,

Applebaum and Cohen (1960) promoted the use of repeatable analytical methods. Input

to their methodological approaches were measurements of trade area, site accessibility,

characteristics of the population within a trade area, competitive supply, economic base

and stability of that economic base, trade area penetration, store size and store

function, building costs, and operating costs. The outcome was a significant rise in the









academic stature of Marketing Geography, a rise in the demand for Marketing

Geography graduates in geography, allied academic fields, and practice.

While William Applebaum wrote extensively, his publications were mainly

targeted to practitioners. His publications in the trade press served to build his stature

among practitioners, served as an outlet for his highly creative mind, and served to

increase demand for his consultant work. Applebaum's publications in academic

journals which served to establish Marketing Geography as an academic discipline were

published 1960s. This was after Applebaum left Harvard, and after Applebaum had

worked for several decades in private consulting practice, learning and proving his

discipline first hand. Applebaum's publications therefore can be thought of as a

compilation of a consultant's "secret" methods, proven by application, and revealed by

paying clients to be highly valuable.8

Applebaum's work was a supreme example of the geographer at work in the

field. Among methods he innovated and put into practice were: customer spotting, trade

area delineation, site evaluation, and market penetration. His primary focus was on the

empirical study of store trade areas and on the market share captured from the trade

area. Among Applebaum's most important and steady clients were Kroger Company

(grocery stores) and Stop & Shop (small quick serve gasoline station oriented grocery

stores), among other grocery store chains in the USA Northeast and Midwest.

One of the methods Applebaum introduced in his own work in the 1930s, and

subsequently published in trade and academic outlets, was his customer spotting

8 The late Reg Golledge in a personal communication to Grant Thrall and subsequently communicated to
Ron Dietz, stated that one of Saul Cohen's great accomplishments was to convince William Applebaum to
publish his methods. Applebaum's response to "snubbing" by mainstream geography academics was to
disregard academic geographers. Applebaum's story appears to be itself an analogue in academic
geography toward its research publications whose focus is on improving business decisions.









method. Prior to the 1930s, the trade areas of grocery stores had been limited largely by

the distance people were willing to walk. However, as automobiles became affordable to

middle class families, trade areas became more complex. Trade areas could be larger

because of the automobile, and grocery stores needed parking lots. Applebaum

"spotted" automobile license plates in a grocery store's parking lot, then used that

information to retrieve the registered address of the automobile from the State's

Department of Transportation. Applebaum was also a pioneer early in the automobile

era on the in-store survey techniques by placing pins on a wall map indicating the

location customers' home addresses (namely their origin), as well as a pin for the

location of the grocery store (the destination). Hence, origin-destination models in

Marketing Geography were introduced. Today, of course, address match geocoding is

used in GIS software, instead of pins on a map. Through trial and error, Applebaum

concluded that a string encompassing 80% of the pins outlining a compact shape was a

useful estimate of a trade area. He reasoned that 20% of the customers "spotted" were

spurious. Today we still separate the core from spurious customers, while other criteria

are used for enveloping the boundary of a trade area (Patel, Fik, and Thrall, 2008).

Applebaum's customer spotting procedure has subsequently served as the

starting point for trade area calculations; indeed, the entire location based intelligence

literature owes a debt to Applebaum. Applebaum's work penetrated beyond the dense

walls of the academy, and showed the practitioner the importance of geography to their

business decisions, and provided a method for answering the question, where should

limited resources be allocated to advertising and location development. If a location is









developed, what is the impact on other stores under the same ownership; namely,

cannibalization?

Customer spotting is also important to location based service providers in that

with the data, the friction of space between the customer and the location can be

calculated.9 This is also known as the distance decay of the customer's demand

schedule. Each item in a store, and collection of all items within the store, creates or

overcomes a friction-of-distance. The choice of which goods are to be offered in the

store therefore affects the trade area. Cannibalization between stores can change

because of store offerings. While in the short run, location for most brick-and-mortar

cannot easily change, knowing the relationship between friction-of-distance and

stocking of goods can be useful in daily real-time business strategic decisions, including

advertising, marketing, goods stocked, and level of service.

Geospatial technology including GIS software and online geospatial databases

have improved the productivity of those performing Marketing Geography, and thereby

lowered its cost and increased its accessibility to smaller business firms. Sources of

geospatial databases used by Marketing Geographers may be comprised of bank data

(ATM records, check records, etc.), credit card data, or point of sale data collected by

the retail market/service provider, but each record has an attribute that can be used to

locate the activity on the landscape. Retail firms are in the business of making money

and as they operate within the competitive business environment they are very

protective of the information collected and techniques used, so they ensure that their

data is likewise protected. Although the means for customer spotting/plotting are greater

9 Customer spotting methods have been used by Ghosh and McLafferty (1987); Moloney (1989); and
Rogers and Green (1978), among others.









than they were in the past, some obstacles still exist such as the cost of obtaining

proprietary data monetarily or by entering into the contractual obligations and

protections afforded by a Non-Disclosure Agreement.

Trade area delineation

Using the information collected by customer spotting survey methods or point of

sale data, trade market areas can be accurately delineated and refined as primary,

secondary, or tertiary trade areas as introduced by Applebaum (1966). He has defined

the primary trade area as the geographic spatial core from which a store draws most of

its customer business and using his analysis of grocery store chains, Applebaum (1966)

found that this primary trade area encompassed about a 60 to 70% capture of store

customers. Applebaum found the secondary trade area to be comprised of the next

highest ratio of customers to population capture and drawing upon an additional 15, 20,

or 25% of overall sales. The tertiary trade area can be defined as that area from which

the residual, transitory, or spurious customers are drawn and usually exist on the outer

fringe of a trade area, often considered out of town sales.

The effects of agglomeration cannot be understated as retail stores can form

symbiotic relationships that serve to expand and increase the size of the market trade

area and subsequent market penetration. Applebaum (1965b) notes this effect in a case

study comparison of a discount food supermarket and a general discount merchandise

store located adjacent to each other with a stand-alone discount food supermarket and

no location association. The supermarket located adjacent to the merchandise store

formed a symbiotic relationship that strengthened the trade area for both businesses

that established more drawing power than the unassociated, stand-alone supermarket.









Summary

There are many ways, methods, and techniques available to the analyst to

calculate the value of location to a particular business, but it is up to the individual

analyst to apply those techniques that most effectively serve to enhance the business

decision. The models presented are only a small selection of the entire site

assessment/location selection measures commonly used and are significant because

they allow the analyst the means to objectively evaluate an existing site or a prospective

new site for a retail outlet. This objectivity facilitates solid analysis, improved judgment,

and timely recommendations that compliment the business decision allowing retail firms

the ability to methodically and systematically apply the results of these models with

confidence. For instance, knowing the customer origins and subsequent trade market

areas for a particular location may identify the presence of market overlap or market

gaps that indicate improper utilization of marketing practices prompting changes to

location configuration, marketing and advertising measures, or operating techniques.









CHAPTER 3
TRADE AREA CALCULATION: AN INTEGRATED GIS APPROACH

GIS as a Marketing Geography Research Tool

Historical Overview

GIS (geographic information systems) is a combination of computer hardware,

software programs, analysis methods, and the proper application of these components

together by competent and knowledgeable analysts.10 GIS allows the user to create,

display, analyze, and manage spatially referenced data consisting of a wide range of

categories from customer location to supply nodes to lines of distribution to competitor

locations to population demographic characteristics and more. Researchers realized the

potential value that computer mapping software programs could provide to the

Marketing Geographer in performing retail market analysis. Wolf (1969) describes the

Synagraphic Mapping System (SYMAP), developed in 1964 by Howard T. Fisher at

Northwestern University, as one of the first programs that allowed for the creation of a

visual representation of spatially distributed data on a digital map. SYMAP was

revolutionary in that it allowed the user the ability to analyze really distributed data and

create choropleth maps with a thematic function for shading polygons. With this

technology, Marketing Geographers could integrate this early and first known GIS

application software with the business decision-making process to calculate areal sales

volume potential, sales territory mapping, and determining industrial plant or warehouse

locations (Wolf 1969).

Businesses have been using GIS technology since its inception on mainframe

computers dating back to the mid-1960s although published literature does not account

10 See Peuquet and Marble, (1990), editors, for seminal discussion of the meaning of GIS.









for this use in applied research in the private sector until the 1980s (King, 1993). This

could be due to the fact that GIS was not widely available, the technology was cost

prohibitive to operate, the functions provided limited and time-consuming application,

and that businesses utilizing the technology were not willing to broadcast the existence

and use of technologies that provided a competitive advantage over rivals. "Emerging

standards for the transfer of spatial data and for the specification of spatial process

models will provide greater inter-operability between components developed at different

times, for different purposes, and by different people using different hardware and

software" (Heikkila, 1998). Corporations are very protective of closely guarded

proprietary data, business decision-making processes, techniques, and methodologies

to the point that in comparison with industrial firms, retailers tend to be oversensitive to

confidentiality with many facts of operation being needlessly suppressed (Davies,

1977). There are many open source websites where useable data can be found and

downloaded, but the fact remains that the Marketing Geographer often has to deal with

the costs associated with the acquisition, protection, and use of proprietary business

datasets. This is detrimental to the analyst and impedes the improvement of findings

and conclusions where transparency could improve the application of techniques and

methodologies with the open sharing of ideas and information. While Marketing

Geography literature is vast and has been improved with expansion through recent

years, "its structure resembles an iceberg 90% submerged. Geographers engaged in

intelligence work are not the only ones with disclosure problems" (Epstein, 1978). It

seems that human nature and the competitive advantage within this field are overriding

factors contributing to the historically slow emergence of GIS, as the first study explicitly









incorporating the use of this technology, as noted by Ghosh and Craig (1986), did not

emerge in published literature until the 1980s.

Customer Spotting Method

The computing power, size of memory storage files, and advances in GIS programs

available today have enhanced the applicability of the methods introduced by William

Applebaum. His work with customer spotting and the means through which customer

addresses are plotted on a map, the basis for trade area delineation, have been

improved with the efficiency of GIS in terms of time saved by the analyst. GIS facilitates

the use of raw data collected and allows for the display of spatial data visually,

especially with the use of Geocoding functions. Geocoding is the procedure through

which GIS applies coordinate data in the form of latitudes and longitudes to individual

records based on a field in the data consisting most commonly of address, zip code, or

zip code + 4 location information. When an individual record has coordinate data, GIS is

able to position the object on the correct location on the planet (Thrall, del Valle, and

Thrall, 1995). GIS software geocodes data based on the existence of two attribute files

that are joined together to present the data spatially. The first attribute file contains the

individual record of transactions for each customer and the second attribute file must

contain the coordinate data used to present the data spatially on a map.

The U.S. Census has created a set of data, collectively referred to as

Topologically Integrated Geographic Encoding and Referencing (TIGER) Line files.

TIGER files can be accessed via the Internet and freely downloaded from the US

Census www.census.gov web site. While not perfect, the intent is that the TIGER files

include the streets in the U.S., their names, type of street, and addresses, all within a

standardized digital format that can be readily integrated with other databases within a









GIS environment. Databases of addresses then can be matched with the TIGER line

files in a manner similar to relational database management (RDM). Since the TIGER

line latitude-longitude coordinates are known, and the address range for each line

segment is known, then RDM can be used with spatial interpolation to estimate the

geographic coordinate of the address within the external database. This point is then

displayed on the map in a position corresponding to the street address number, on the

correct side of the street, and set back 50 feet from the street segment line. This

procedure is repeated by the GIS software for each record in the database. Such

geocoding allows spatial visualization of external database addresses within the GIS.

The external databases can be customer addresses, store locations, event locations

such as the occurrence of a crime, and so on.

There is a match accuracy and error associated with geocoding because TIGER

Line files do not cover all of the streets in the U.S., especially in the rural areas, and

because the quality of this data is only as good as the technician that digitized the line

and entered the street address values. Geocoding to a zipcode or zip code + 4 polygon

has less error because the match is based on a larger polygon area than the street

address which is represented by a single point along a line. Thrall and Thrall (1994)

have documented that street level geocoding has a lower 'hit rate' than geocoding to the

zip code + 4 level. This is due to the arrangement of zip code and zip + 4 areas, which

are depicted with polygons used to geocode match records located within the polygon

boundaries and are assigned to the centroid of the individual polygon. The analyst must

determine the level of precision required by the analysis and whether or not the level of









error introduced by TIGER Line files is an acceptable sacrifice to geocode at the street

level and a specific locational point.

The use of GIS facilitates the process of plotting customer location in relation to

the retail outlet store and the calculation of trade areas. This is done by determining the

distance of each customer location from the retail outlet store and attaching that

distance to the customer record data file. The analyst can then query these distances to

find the primary, secondary, or tertiary trade areas by finding those records that are

closest to the store and comprise 70% of store sales.

Demographic Application

Marketing Geographers utilize Geodemographics to gain a better understanding

of the customer base being served by the retail market/service provider outlet store.ll

ESRI's Business Analyst add-in to ArcMap includes "Community Coder" geocoding

application. Community Coder was formerly known as CACI/Coder Plus. Community

Coder appends demographic information to the data record based upon its address.

Community Coder has 60 demographic attribute traits utilized by the U.S. Census

Bureau specifically focusing on population and household characteristics to create

customer cluster profiles encompassing over 200,000 neighborhoods. The assignment

of a demographic profile by Community Coder divides and separates the customer

dataset into 12 Life Mode groups and 65 Residential Segments each individually known



11 Goss, J.D. 1995a, Marketing the new marketing: the "strategic" discourse of advertising for
Geodemographic Information Systems. In Pickles, J. (ed) Representations in an electronic age:
geography, GIS and democracy. New York: Guildford Press. And, Goss, J.D. 1995b, "We
know where you are and we know where you live": the instrumental rationality of
geodemographic information systems. Economic Geography 71,2: 171-198









as a Tapestry (Appendix A). These Tapestries have also formerly been known as A

Classification of Residential Neighborhoods (ACORNs) or Lifestyle Segmentation

Profiles (LSPs). "LSPs provide a composite measurement that summarizes population

characteristics by location" (Thrall and Mecoli, 2003). ESRI BIS Coder Business Analyst

uses a clustering of customers, based on the consumption history and anticipated

spending profile of the area where they live, in a process of assigning a Tapestry that

groups these consumers at the zip code, zip code + 4, or street level. "The basic

principle that underlies the creation and use of an LSP is Tobler's first law of geography:

all things are related, but near things are more related than distant things" (see Goss,

1995; Thrall and Mecoli, 2003). These profiles, or Tapestries, are assigned to residential

neighborhoods to reveal the spending habits of an area in the theory that people with

similar interests, similar spending habits, and making similar choices will choose to

reside within close proximity of each other spatially.

Summary

Geospatial technology has increased the productivity and precision of analysis

created by Marketing Geographers (see Pickles, 1995). "The geospatial technological

innovations include remote GIS, spatial data transfer standards, open GIS

specifications, digital spatial data libraries, object-oriented GIS, and network-resident

programming such as JAVA" (Heikkila, 1998). GIS has proven to be a valuable tool that

enhances the business decision-making process through timely and accurate spatial

analysis that offers the retail market/service provider a competitive advantage that

competitors may or may not be using.









CHAPTER 4
TRADE AREA CALCULATION: METHODOLOGY

This section presents a methodological algorithm that can be followed by a retail

market/service provider to calculate a trade area. The methodology integrates concepts

from retail location analysis, geospatial statistics, and GIS technology. The algorithm

presented here would not have been practical before the creation of contemporary

geospatial technology and powerful computers. This methodology has its foundations in

the retail location assessment literature summarized in Chapters 2 and 3. The algorithm

presented here is demonstrated using actual data collected from a single outlet retail

service provider in Alachua County, Florida.

The first objective of this study is to present the algorithm as in step-by-step

fashion. The second objective of this section is to provide a descriptive overview to

increase the understanding of the procedure so that so that it can be repeated by other

analysts. The benefits of the algorithm arise because the success of a retail service

provider depends upon location; the algorithm offers the entrepreneur to better evaluate

the changing market landscape, and thereby realize a competitive advantage.

The retail business that provided the data for this study requires anonymity. The

retail business will therefore be referred to as the "Client." The "Client" provided

customer data for two time periods, 2004 and 2008. Both databases included customer

address, transaction date, and amount paid for the transaction. The datasets were

received in comma-delimited format (CSV). Data entry errors were corrected or

eliminated. The CSV file was exported as an Excel XLS file. ESRI Community Coder

can read XLS data files. The XLS file is processed by Community Coder and then

imported as a data layer into the GIS environment. A second dataset was created by









joining the 2004 and 2008 datasets that is comprised of the customers with accounts

showing up in both the 2004 and 2008 datasets. The second database was also

geocoded. Therefore, three GIS data layers (shapefiles) were used:

* customers in 2004
* customers in 2008
* customers in both 2004 and 2008

Geographically spurious customers were eliminated by including only customers

with home addresses within Alachua County, Florida. Geocoding had revealed

customers with addresses in Iowa and Minnesota, as well as other countries. The

results of each dataset selection refinement are as follows:

* 2004 Alachua Clients (1423)/ 2004 All Clients (1745) = 81.55%
* 2008 Alachua Clients (1459)/ 2008 All Clients (1761) = 82.85%
* 2004+2008 Alachua Clients (645) / 2004+2008 All Clients (664) = 97.14%

The presence of selected customers, as a percentage in relation to all customers,

represent a ratio for each year group dataset that are all above 80% and can be

considered a statistically significant study group.

Using the ESRI ArcMap spatial statistics function, the Centroid, Directional

Distribution (Standard Deviation Ellipse), and Standard Distance for each year group

dataset was calculated with the process shown graphically in the flow chart below

(Figure 4-1). It should be noted that the Directional Distribution (Standard Deviation

Ellipse) and Standard Distance shown below were calculated using the 1st Standard

Deviation and Euclidean Distance centered around the Centroid calculated for the

individual year group dataset being used. The figures below are organized by dataset

year group used and, as much as possible, are color coded representations with 2004

customers depicted in red (Figures 4-2 and 4-3), 2008 customers depicted in blue









(Figures 4-4 and 4-5), and 2004+2008 customers depicted in green (Figures 4-6 and 4-

7). Overlaying the 2004, 2008, and 2004+2008 shapefile functions by year group, the

growth and decay areas are further revealed (Figures 4-8 and 4-10). Additionally, the

Symmetrical Difference between the Directional Distribution (Standard Deviation

Ellipse) and the Standard Distance was calculated using a comparison of the 2004 and

2008 datasets (Figures 4-9 and 4-11). The area shown represents the customer

transition area or change in time for the respective functions.

By using the spatial statistics functions provided by ArcMap, and visually shown

on the Figures (4-2 through 4-11), the following customer capture statistics are shown in

the list below.

Customer Capture Statistics.

Client's Customer List 2004
Directional Distribution (Standard Deviational Ellipse) 1074/1423 = 75.47%
capture of 2004 customers.
Standard Distance 1060/1423 = 74.49% capture of 2004 customers.

Client's Customer List 2008
Directional Distribution (Standard Deviational Ellipse) 1090/1459 = 74.71%
capture of 2008 customers.
Standard Distance 1080/1459 = 74.02% capture of 2008 customers.

Client's Customer List 2004+2008
Directional Distribution (Standard Deviational Ellipse) 489/645 = 75.81% capture
of 2004+2008 customers.
Standard Distance 486/645 = 75.35% capture of 2004+2008 customers.

The measurements shown above indicate a very stable and solid market

penetration and customer capture. There appears to be little change or variation within

the three different datasets throughout the encompassing five year period being

analyzed.









The Core Trade Area Radial Method

In applying the core trade area radial method, the centroid of the spatial

distribution of the data was used as the starting point. From the centroid, radial

distances of 1.5 and 1.0 miles were used to create a circle polygon and calculate the

number of customers captured within these distances. These distances were not

chosen arbitrarily, and for the size of the study area were appropriate using

conventional marketing geography practices.

Publix supermarket corporate headquarters utilize the competitive edge benefit of

a Marketing Geography division that employs analysts to perform the techniques and

methodologies associated with retail market/service provider location and trade area

calculation.12 The location, by address, of each Publix store was found using

BellSouth's The Real Yellow Pages Gainesville, Florida phonebook for 2008, obtaining

the addresses of Publix locations, geocoding those addresses, and plotting each

location on the map using GIS. By reverse engineering the known locations of Publix

supermarkets for the Gainesville, Florida area, the established trade area can be

roughly estimated. This is done by buffering each Publix supermarket location until the

buffered areas become tangent to each other where possible and, in contiguous areas,

where market overlap is minimized while maximizing the market penetration for each

site (Figure 4-12). The estimated radius of 1.5 miles revealed the target core trade area

for existing Publix supermarket locations and will be used as a benchmark

representative trade area.


12 Communication to Ron Dietz by Professor Grant Thrall. Publix headquarters has employed students
from University of Florida that have completed the Business Geography Program under the direction of
Professor Thrall. Professor Thrall has also presented invited Friday Afternoon Seminars to senior Publix
management at their headquarters in Lakeland Florida.









Customer spotting and the benchmark trade area distance of 1.5 mile radius are

applied to the "Client's" database. The number and percent of "Client" customers

located within a 1.5 mile radius from the centroid are thereby determined. The results

are shown in Figures 4-13 to 4-15, and in the list below.

Market Penetration Within The Core Trade Area.

* 2004 Alachua County, FL Customers 511/1423 = 35.91% (Figure 4-13).
* 2008 Alachua County, FL Customers 504/1459 = 34.54% (Figure 4-14).
* 2004+2008 Alachua County, FL Customers 216/645 = 33.49% (Figure 4-15).

(Based upon the number of customers within 1.5 miles from the client's location, divided
by the total number of client's customers within Alachua County, Florida.)

This list reveals that the customer capture rate is constant, with the difference in

percentage within the core trade area from 2004 to 2008 changing by only 1.37%.

During the five year period covered by this study, there were eight new retail

market/service provider outlet stores that appeared on the landscape. This was

determined by using BellSouth's The Real Yellow Pages Gainesville, Florida phonebook

from 2004 (32 competitor listings) and 2008 (43 competitor listings), obtaining the

addresses of competitor locations, creating a spreadsheet of these locations, geocoding

those addresses, and plotting each location on the map using GIS. The appearance of

these eight new competitor locations are important because of the uncertainty that they

introduce to the existing business equilibrium and the potential impact on current

customer strength through cannibalization and market penetration. Figures 4-16 to 4-24

show the number and percent of the Client's customer dataset captured within a 1.5

mile radius from all eight new competitors (Figure 4-16) and each new competitor

location (Figure 4-17 to 4-24). In no particular order, new competitor 1 through 8

potential impact by customer capture revealed in the following list:









New Competitor Location Analysis.

* All eight new competitor locations appearing from 2004 to 2008, Alachua County,
FL Customer capture 326/1459 = 22.34% (Figure 4-16).

* New competitor location number 1 appearing from 2004 to 2008, Alachua
County, FL No Customer capture 0/1459 = 0% (Figure 4-17).

* New competitor location number 2 appearing from 2004 to 2008, Alachua
County, FL Customer capture 13/1459 = 0.89% (Shared buffer trade area with
new competitor 3) (Figure 4-18).

* New competitor location number 3 appearing from 2004 to 2008, Alachua
County, FL Customer capture 14/1459 = 0.96% (Shared buffer trade area with
new competitor 2) (Figure 4-19).

* New competitor location number 4 appearing from 2004 to 2008, Alachua
County, FL Customer capture 68/1459 = 4.66% (Shared buffer trade area with
new competitor 5) (Figure 4-20).

* New competitor location number 5 appearing from 2004 to 2008, Alachua
County, FL Customer capture 150/1459 = 10.28% (Shared buffer trade area with
new competitor 4) (Figure 4-21).

* New competitor location number 6 appearing from 2004 to 2008, Alachua
County, FL Customer capture 87/1459 = 5.96% (Figure 4-22).

* New competitor location number 7 appearing from 2004 to 2008, Alachua
County, FL Customer capture 6/1459 = 0.41% (Figure 4-23).

* New competitor location number 8 appearing from 2004 to 2008, Alachua
County, FL Customer capture 39/1459 = 2.67% (Figure 4-24).

Additionally, during the five year period covered by this study, there were four

existing retail market/service provider outlet stores that moved, changing their physical

location within the landscape. This was determined by using BellSouth's The Real

Yellow Pages Gainesville, Florida phonebook from 2004 and 2008, obtaining the

addresses of competitor locations, creating a spreadsheet of these locations, geocoding

those addresses, and plotting each location on the map using GIS. The movement of

these four existing competitor locations is important because of the uncertainty that they









introduce to the existing business equilibrium and the potential impact on current

customer strength through cannibalization and market penetration. These moving

competitor locations are more important than the new outlets because they have a

customer base and could increase the market penetration. The following list depicts the

number and percent of the Client's customer dataset captured within a 1.5 mile radius of

each moving competitor location for 2004 and 2008 listed first through fourth:

Existing Competitor Change in Location Analysis

* First moving competitor appearing from 2004 to 2008, (2004 Location) Alachua
County, FL Customer capture 451/1423=31.69% of 2004 customers,
444/1459=30.43% of 2008 customers (Figure 4-25).

* First moving competitor appearing from 2004 to 2008, (2008 Location) Alachua
County, FL Customer captured 69/1459 = 30.43% of 2008 customers (Figure 4-
26).

* Second moving competitor appearing from 2004 to 2008, (2004 Location)
Alachua County, FL Customer captured 67/1423 = 4.71% of 2004 customers and
65/1459 = 4.46% of 2008 customers (Figure 4-27).

* Second moving competitor appearing from 2004 to 2008, (2008 Location)
Alachua County, FL Customer captured 63/1459 = 4.32% of 2008 customers
(Figure 4-28).

* Third moving competitor appearing from 2004 to 2008, (2004 Location) Alachua
County, FL Customer captured 345/1423=24.24% of 2004 customers,
367/1459=25.15% of 2008 customers (Figure 4-29).

* Third moving competitor appearing from 2004 to 2008, (2008 Location) Alachua
County, FL Customer captured 346/1459 = 23.71% of 2008 customers (Figure 4-
30).

* Fourth moving competitor appearing from 2004 to 2008, (2004 Location) Alachua
County, FL Customer captured 67/1423=4.71% of 2004 customers,
65/1459=4.46% of 2008 customers (Figure 4-31).

* Fourth moving competitor appearing from 2004 to 2008, (2008 Location) Alachua
County, FL Customer captured 477/1459 = 32.69% of 2008 customers (Figure 4-
32).









The results of an analysis of the competitor's moving physical outlet store location

from 2004 to 2008 revealed that the First competitor moved to a more distant location

and lost market penetration influence on the Client's customer base. The Second and

Third competitor's had very little location change and, therefore, had very little change in

market penetration influence on the Client's customer base, remaining at consistent

potential capture levels. The Fourth competitor moved closer to the Client's location

from 2004 to 2008, significantly increasing potential market penetration influence and

could be a future threat to potential customer capture levels.

The Grid Method

The next method of analysis was to apply a 1 x 1 and then a 1.5 x 1.5 mile grid

layer that covered Alachua County. The two different sized grid cells are used to detect

existence of MAUP. The respective data layer was then was then combined with the

2004 and 2008 customer datasets using the overlay function. The cells with customer

data were selected and exported as an individual layer to be used in the analysis and

creation of the choropleth maps (Figures 4-33 to 4-47). The choropleth maps were

created using the Jenks Method consisting of natural breaks and classified into five

distinct classes. Each breaking point threshold by class was then rounded to a nearest

whole number and applied to other similar category choropleth maps showing different

year group data results.






































Figure 4-1. Data manipulation and methods flow chart.







43












2004 Customers in Alachua County
Directional Distribution (Standard Deviational Ellipse)


S


Latitude/Longitude Coordinate System


Figure 4-2. Alachua county customers 2004 directional distribution (standard
deviational ellipse).













2004 Customers in Alachua County
Standard Distance


N


S


Latitude/Longitude Coordinate System


Figure 4-3. Alachua county customers 2004 standard distance.










2008 Customers in Alachua County
Directional Distribution (Standard Deviational Ellipse)


4-- E


Latitude/Longitude Coordinate System


Figure 4-4. Alachua county customers 2008 directional distribution (standard
deviational ellipse).













2008 Customers in Alachua County
Standard Distance


N


S


Latitude/Longitude Coordinate System


Figure 4-5. Alachua county customers 2008 standard distance.










2004+2008 Customers in Alachua County
Directional Distribution (Standard Deviational Ellipse)


4-- E


Latitude/Longitude Coordinate System


Figure 4-6. Alachua county customers 2004+2008 directional distribution (standard
deviational ellipse).









2004+2008 Customers in Alachua County
Standard Distance


4-- E


Latitude/Longitude Coordinate System


Figure 4-7. Alachua county customers 2004+2008 standard distance.












Alachua County, FL Customer Growth and Decay Zones
Directional Distribution (Standard Deviational Ellipse)
Polygon Overlay by Dataset Year Group


N
W E


Latitude/Longitude Coordinate System


Figure 4-8. Alachua county customer growth and decay zones overlay of the directional
distribution (standard deviational ellipse) polygons by dataset year group.












2004 and 2008 Customers in Alachua County
Directional Distribution (Standard Deviational Ellipse)
Symmetrical Difference


N
.-AF


Latitude/Longitude Coordinate System


Figure 4-9. 2004 and 2008 Customers in alachua county, fl. directional distribution
(standard deviation ellipse), symmetrical difference overlay.












Alachua County, FL Customer Growth and Decay Zones
Standard Distance Polygon Overlay by Dataset Year Group


N
W E


Latitude/Longitude Coordinate System


Figure 4-10. Alachua county customer growth and decay zones overlay of the standard
distance polygons by dataset year group.













2004 and 2008 Customers in Alachua County
Standard Distance Symmetrical Difference


N
.-AF


Latitude/Longitude Coordinate System


Figure 4-11. 2004 and 2008 Customers in alachua county, fl. standard distance,
symmetrical difference overlay.












2008 Publix Supermarket Locations (1.5 Mile Buffer)
Alachua County, FL


N

w^-
s


Latitude/Longitude Coordinate System


Figure 4-12. 2008 Publix supermarket locations (1.5 mile buffer).























- H

U ':


'K~

1 -


I3_ IQ


- --~;I.


/


I ..
r

'.
"
a.



'--


v -


-"" -


Figure 4-13. 2004 Alachua county customers within 1.5 miles of the centroid.


i

( ,


- .


IC' ''~


Si-i


.5
, IF


/

a.
'/


""
.; I i

yblpc;rpp
'"

'' "~
'"


- "t~


Figure 4-14. 2008 Alachua county customers within 1.5 miles of the centroid.














55


~rrre


4


;ixrs~


-N II


*


- ~ -Lh~ru.hlnn








"=


,,

c-
1 =


,-" i






















-


C,


J-- i -._.


-'~-V ..*LV.r


" 1 /


*WI -t8


Figure 4-15. 2004+2008 Alachua county customers within 1.5 miles of the centroid.


'c
f' ".


S ,' .


f 11 J 11i


a.* ~
t,3.
ar r




















S I


B .


Figure 4-16. 2004+2008 All eight new competitor locations appearing from 2004.









56


* ..I:


-jir- I-'


,i--


."'


'i


ALALi


-"
'L~' ~r~Ci.FZI
---c. ---
--~- ~~_L~-


,_, '
,.i i
;--~: "
:


:.a



O .ji:" -~ \I_
























ft*


'A
-

'f *


_- .-.*
_.- ^

a:


9


a f
S


Figure 4-17. New competitor location number 1 appearing from 2004 to 2008.


--

-r,
':a


'. ,I',m


8
-


. *t


Figure 4-18. New competitor location number 2 appearing from 2004 to 2008.


ir ~i~r


i




























,a

Vi


\--


., I


t 0*
S


__ K-- -


Figure 4-19. New competitor location number 3 appearing from 2004 to 2008.


te.


* ,-


* .
(4


tic


,'



- 16*8


K-


Figure 4-20. New competitor location number 4 appearing from 2004 to 2008.


S.'

U,,
I-'


;,'
'.,*


1...:


6 i-
-; ~























U r '1


* *4,


H--


* 0* 4
i *


I I.,.


a.


Figure 4-21. New competitor location number 5 appearing from 2004 to 2008.


U5 rl


C *,

to.


KI


*

.3sc


\ 1


*~ a
S


Figure 4-22. New competitor location number 6 appearing from 2004 to 2008.





59
















~------\ -u
*o o











: : ^ :..':


3
"' n't





II.
l '-1 i1


5 "


Figure 4-23. New competitor location number 7 appearing from 2004 to 2008.


t--
Og

* *4.
;;ih


--a

*

e/








7, -
S"*"


Figure 4-24. New competitor location number 8 appearing from 2004 to 2008.


60


* ; Sb


^- 9


* '


6 i'


.i


-i
L"



8


r~ i-
















----- --
-t


~c~u~.~p

tIs e


.--


<'3j












1 |


S I I i
.^'


.2
"* 4 't-e



.. *
-7.


4I- 4 ,,
S S" .

-." q f "
W.. *w 3%.-





is I

n
-'** a1 i ;A I


a
...: *. *


% .mgL. p* *.* Cu:_-




I.- lea LJ~lr-...

S LA5 *r]
'a-u g *


t -


Figure 4-25. First moving competitor appearing from 2004 to 2008, (2004 location).


N


a,.


..,.. -


SI `b
a,.


*t L .1
- ,




*



*
SN




aS. .


-.*
'I
em


e


U* S *
i, *
e *


m- :



.. ...... ..*..." .*
S -


Figure 4-26. First moving competitor appearing from 2004 to 2008, (2008 location).






61


ea'inv


1*
S.
F,


4^
1f


*'

a ,l


I
.4 S
S


-5

ilr


-- -. -- ...


dl"













S.. -* -* .
ir *.-. a a, 1* i .-


-: S4 "
S as '41 I '



- .a.-: r I. .,






eL aI


1


n ;




1


Figure 4-27. Second moving competitor appearing from 2004 to 2008, (2004 location).


"- i0
.; 1..
*~ a %


a


* *


Sr
I


* A ,-.
" 16,i

V.-
'.^ +T- *. .



> I
a
.0 r

li

.4. *




9
rrrnloe' S
Da


a
a.


G "'-"s' r o* o*'
A.



,t p-a rt. ::41


-P A
i *' *ew* s .

~ -- ~*


Cri 9.


*. *
-e
irj


S
a
* a*
S-


*'
\, C; I Iq
, ,,. ,:,,, ..

\ .- --
'K .aprt.


a.
S Iil


Figure 4-28. Second moving competitor appearing from 2004 to 2008, (2008 location).


S
a
a
uer e
ar


* w
I
S.


w


S





ei




* *ii
Umm~~


-





*
i n



4. w "''
* *- *- ,


. *
S


, .


S
a







U *
. *
a
4


, I


U1





















em. usi-ti


*I U t
'a
0r
a1
S.~


-t
'S
* *


h.





I -


a


t 4


e
S
*i
*n*4C8*--,'
U


* ad


0
a- ._?.
~ -B-,ra ^ -
a a *


es
a-


Figure 4-29. Third moving competitor appearing from 2004 to 2008, (2004 location).


...... ..







a* e. *:
-Ir -


-Vt


42' 4t4 ^-*: *

L '* a .

-.-" .? .:.j-1e_..r.- a .. .



h a a s e -- .
"\ '. .-. --,, .."..*:j? .. .. _,



.. r : .




Figure 4-30. Third moving competitor appearing from 2004 to 2008, (2008 location).







63


F


Si



ma


S
U,


6 *
*t.


4


S
a~ r


J





















-*





m4
S. ..





S". .
'*' -. l ...I I. ,









0* "
-\
.i i
1+ r m. II


I--'


I~

Sc
*- 1-*
;I,,.


.e -






S
G- ............


.5-,


-._-. .


Figure 4-31. Fourth moving competitor appearing from 2004 to 2008, (2004 location).


a









4 t1









*


I


, '


- 0


a'l + .


A* di i 0 Anl &. .

*





a Sl f *.






o o. *
*.1 :
* S '-^ fl e5
e4 r
sjA~~a4' ~it









.r *, ai~~lC~
: r


*
S


. ,
*


S -.'
\ C
'a 1 i






a"0


o,
* *.;;
a* i


Figure 4-32. Fourth moving competitor appearing from 2004 to 2008, (2008 location).






64


V.1..


5 S
S

















a.


S


0-; i


c
o
r-
r,
14.




,,















1x1 Mile Grid Overlay Choropleth Map
2004 Total Revenue in Alachua County, Florida


N




Latitude/Longitude Coordinate System


Figure 4-33. 2004 Total revenue (dollars), 1x1 mile grid cell.


Total Revenue
S$000 $50000
$500.01 $1,000.00
I $1,000.01 -$1,500 00
$1,500.01 -$3,000.00
$3000.01 $7,000 00















1x1 Mile Grid Overlay Choropleth Map
2008 Total Revenue in Alachua County, Florida


N




Latitude/Longitude Coordinate System


Figure 4-34. 2008 Total revenue (dollars), 1x1 mile grid cell.


Total Revenue
S$000 $50000
$500.01 $1,000.00
I $1,000.01 -$1,500 00
$1,500.01 -$3,000.00
$3000.01 $7,000 00














1x1 Mile Grid Overlay Choropleth Map
2004 Average Revenue per Transaction
Alachua County, Florida


N




Latitude/Longitude Coordinate System


Figure 4-35. 2004 Average revenue per transaction (dollars), 1x1 mile grid cell.


Average Revenue per Transaction
S$0.00 $50 00
$50 01 -$10000
S$100.01 $200.00
S$200 01 $350 00
$350 01 -$700 00














1x1 Mile Grid Overlay Choropleth Map
2008 Average Revenue per Transaction
Alachua County, Florida


N




Latitude/Longitude Coordinate System


Figure 4-36. 2008 Average revenue per transaction (dollars), 1x1 mile grid cell.


Average Revenue per Transaction
S$0.00 $50 00
$50 01 -$10000
S$100.01 $200.00
S$200 01 $350 00
$350 01 -$700 00














1x1 Mile Grid Overlay Choropleth Map
2004 Transactions in Alachua County, Florida


N



Latitude/Longitude Coordinate System


Figure 4-37. 2004 Number of transactions, 1x1 mile grid cell.


Number of Transactions
-- 3
1 4-8
99-15
16-27
S28-44














1x1 Mile Grid Overlay Choropleth Map
2008 Transactions in Alachua County, Florida


N



Latitude/Longitude Coordinate System


Figure 4-38. 2008 Number of transactions, 1x1 mile grid cell.


Number of Transactions
-- 3
1 4-8
99-15
16-27
S28-44














1x1 Mile Grid Overlay Choropleth Map
2004 Customers in Alachua County, Florida


N



Latitude/Longitude Coordinate System


Figure 4-39. 2004 Number of customers, 1x1 mile grid cell.


Number of Customers
11-4
75-16
- 17-35
S36 75
76-115














1x1 Mile Grid Overlay Choropleth Map
2008 Customers in Alachua County, Florida


N



Latitude/Longitude Coordinate System


Figure 4-40. 2008 Number of customers, 1x1 mile grid cell.


Number of Customers
11-4
75-16
- 17-35
S36 75
76-115












1x1 Mile Grid Overlay Choropleth Map
Total Revenue Change
Alachua County, Florida


M


s


Latitude/Longitude Coordinate System


Figure 4-41. Total revenue change from 2004 to 2008, 1x1 mile grid cell.


Total Revenue Change
S($3,074 11)-($1,000 00)
S($999.99)- $0.00
$0 01 -$1,000 00
$1,000 01 $2,000 00
$2,000.01 -$4,100.00










1.5x1.5 Mile Grid Overlay Choropleth Map
Total Revenue Change
Alachua County, Florida


N

FQ


Latitude/Longitude Coordinate System


Figure 4-42. Total revenue change from 2004 to 2008, 1.5x1.5 mile grid cell.


Total Revenue Change
S($3,074 11)-($1,000.00)
S($999 99)-$0 00
- $0.01 -$500.00
S$500 01 -$1,000.00
$1,000.01 -$4,100.00









The Coordinate Plane


Figure 4-43. 1x1 Mile grid cell coordinate plane.



Grid Overlay Offset


Third Foci (1/2 Mile Offset)
Second Foci
(1/4 Mile Offset)


First Foci (Origin)


Figure 4-44. 1x1 Mile grid cell overlay offset method.


----A -----


----4-- -------
II I


-X+
- 3 2 1


~I IV
-- --~ ~-- -

ZjnZ^-.tr










1x1 Mile Grid Overlay Choropleth Map
Total Revenue Change (First Foci, Origin)
Alachua County, Florida


-7


4
1-s-

s


Latitude/Longitude Coordinate System


Figure 4-45. Total revenue change from 2004 to 2008, 1x1 mile grid cell (first foci,
origin).


Total Revenue Change
S($2,279 51)- ($500.00)
($499.99) $0.00
- $0.01 -$500.00
S$500 01 $1,000 00
$1,000.01 -$4,600 00


---- M











1x1 Mile Grid Overlay Choropleth Map
Total Revenue Change (Second Foci, 1/4 Offset)
Alachua County, Florida


e


n7. E

Ni U E


N-


III 1W


N

A "2hY-


Latitude/Longitude Coordinate System


Figure 4-46. Total revenue change from 2004 to 2008, 1x1 mile grid cell (second foci,
mile offset).


Total Revenue Change
| ($7,333.06)- ($500.00)
($499 99)- $0 00
- $0 01 -$500 00
$500.01 -$1,000.00
$1,000 01 $4,600 00


7

M *JI 1 :D^












1x1 Mile Grid Overlay Choropleth Map
Total Revenue Change (Third Foci, 1/2 Offset)
Alachua County, Florida


M


s


Latitude/Longitude Coordinate System


Figure 4-47. Total revenue change from 2004 to 2008, 1x1 mile grid cell (third foci, 12
mile offset).


Total Revenue Change
S($7,400 00)- ($1,500 00)
S($1,499.99)- ($500.00)
S($499 99) -$500 00
$500.01 $1,500 00
$1,500.01 $4,600.00









CHAPTER 5
TRADE AREA ALGORITHM

A layer of same-sized grid cells are used as a container for data within each cell's

area, with the added benefit that geography is held constant. The grid provides a static

visual display of the dataset year group values applied across the landscape in an area

that maintains a consistent level of geospatial parity. Placing dataset year groups in a

grid and comparing the results of the mapped areas together reveal the dynamic

geotemporal change in phenomena being studied. An example of a dynamic

geotemporal application of unemployment rates by county throughout the nation,

(http://cohortl 11.americanobserver.net/latoyaegwuekwe/multimediafinal.html), can be

seen in the example created by Latoya Egwuekwe. In Egwuekwe's geotemporal study,

Bureau of Labor Statistics' monthly unemployment statistics from January 2007 to

December 2009 are displayed by country. While Egwuekwe's geotemporal "movie" is

visually appealing, the interpretation of the space-time trend is biased because

geography is not held constant each county has a different size and shape. The larger

the county, the greater is the inferred visual weight given to the value of unemployment.

Since the unemployment value is calculated as a ratio and reported at the state or

county level, it is easy to apply and show these values visually using these arbitrary

legal boundaries that are not consistent and often defined by topography. Going a step

further, by applying the unemployment values to the centroid of the state or county

boundary polygon and then overlaying a grid, the projection and areal weighting

inference disparity would be eliminated by using standard area polygons that the grid

presents. Application of the grid would also reveal clustering trends and trajectories that

would otherwise remain transparent and undetected using dissimilar polygons.









Keeping size and scale constant does not always ensure that statistical analysis of

a spatial area will show results that are unbiased. Different areal units can have large

impact on the end result of aggregated values. The MAUP is highly sensitive to the

direction of investigation (0 to 360 degrees) progression on the landscape, the cell size,

the offset distance, and the time frame being studied. Therefore, it is critically important

to realize that the depiction of descriptive statistics on a map may not be truly

representative of the actual variation of a particular variable displayed across the map

as shown.

Application of the grid in conjunction with accepted methodology allows for the

spatial variation to become consistent across the landscape. By using a grid size that is

appropriate to the study area, the analyst is able to focus on the variable being studied.

By keeping geography constant in this way, it allows for the application of other

measures to identify significant occurrences of trends and phenomena.

The calculated statistics by grid cell category can be found in Table 5-1 through

Table 5-6, and are summarized by approximate calculated arithmetic mean in the

following list:

* 1x1 Mile Grid, Total Revenue Change (Table 5-1), $203 per cell.
* 1x1 Mile Grid, Average Revenue Change (Table 5-2), $52 per cell.
* 1x1 Mile Grid, Customer Count Change (Table 5-3), 0.429 per cell.
* 1x1 Mile Grid, Transaction Change (Table 5-4), 0.273 per cell.
* 1.5x1.5 Mile Grid, Total Revenue Change (Table 5-5), 115 per cell.
* 1x1 Mile Grid, Total Revenue Change (Table 5-6), Foci 1 $20 per cell.
* 1x1 Mile Grid, Total Revenue Change (Table 5-6), Foci 2 $75 per cell.
* 1x1 Mile Grid, Total Revenue Change (Table 5-6), Foci 3 $130 per cell.

Figures 4-41 and 4-42 show how the Total Revenue values change on a

choropleth map when applied across a 1 x 1 mile grid cell overlay and a 1.5 x 1.5 mile

grid cell overlay respectively. The values are clearly dependent upon grid cell size as









their importance becomes more diffused (Figure 4-41) or less diffused (Figure 4-42)

with areal change. This can be problematic as the proper areal size adjustment of a grid

cell can serve to either highlight or bypass noteworthy clusters as shown in different grid

cell overlay sizes.

Figure 4-43 is a depiction of the coordinate plane where the X and Y axes

emanate from an origin dividing the grid plane into four quadrants. Figure 4-44 shows

how the grid overlay was shifted from the origin into the first quadrant using a direction

of 45 degrees and a distance of 14 and /2 mile. Figure 4-45 is a choropleth map showing

total revenue values across a 1 x 1 mile grid overlay, with the First foci lower left grid

cell co-located at the origin. Figure 4-46 is a choropleth map showing Total Revenue

values across a 1 x 1 mile grid overlay, with the Second foci lower left grid cell located a

14 mile distance and 45 degrees from the origin. Figure 4-47 is a choropleth map

showing total revenue values across a 1 x 1 mile grid overlay, with the Third foci lower

left grid cell located a /2 mile distance and 45 degrees from the origin. The choropleth

map clearly shows how small changes in distance and direction impact the values

represented by the choropleth map. The Second foci (Figure 4-46) has a similar spatial

distribution to the First foci (Figure 4-45), but the Third foci (Figure 4-47) has a distinctly

different appearance with clustered neighborhoods dominated by the middle range of

values.

The Hot Spot Analysis tool calculates the Getis-Ord Gi* statistic for each feature in

a weighted set of features. The Gi* statistic tells you whether features with high values

or features with low values tend to cluster in a study area with the output being a Z

score for each feature which represents the statistical significance of clustering for a









specified distance. This method works by looking at each feature within the context of

neighboring features. If a feature's value is high, and the values for all of its neighboring

features is also high, it is a part of a hot spot. The higher (or lower) the Z score, the

stronger the association. For statistically significant positive Z scores, the larger the Z

score, the more intense the clustering of high values. For statistically significant

negative Z scores, the smaller the Z score, the more intense the clustering of low

values. A Z score near zero indicates no apparent concentration (neighbors have a

range of values). The local sum for a feature and its neighbors is compared

proportionally to the sum of all features; when the local sum is much different than the

expected local sum, and that difference is too large to be the result of random chance, a

statistically significant Z score is the result. The use of this statistical method reveals the

presence of neighborhoods that are saturated and others that could be mined using

more robust marketing applications.

The Hot Spot Analysis (Getis-Ord Gi*) choropleth map values were created using

a fixed distance band and a Euclidean distance. The Local G statistics are used to test

for spatial clustering in group-level data making it possible to assess the spatial

association of a variable within a particular distance of each observation. The spatial

clusters show areas with high and low attribute values graphically within the choropleth

overlay layer. The 1 x 1 mile grid cell overlay (Figure 5-1) of the Hot Spot Analysis

choropleth map clearly shows a more diverse spatial distribution than the 1.5 x 1.5 mile

grid cell overlay (Figure 5-2) which appears to become more spatially diffused with

small area clusters disappearing and values approaching equivalence.









The Hot Spot Analysis of the 1 x 1 grid cell overlay choropleth maps shows a

distinct cluster of cells, with values less than -2.58 standard deviations, appearing in the

Second foci (Figure 5-4) spatial distribution that can be described as a submarket. In

the application of dynamic gridding shown here, there was still an instance where a

submarket was revealed through this process and identifies the existence of tapestry

clusters (Appendix A). There are a total of 26 cells in this category submarket with the

majority of cells labeled by the College Towns tapestry (7 cells), Prosperous Empty

Nesters (5 cells), Enterprising Professionals (2 cells), and In Style (2 cells). The

remaining 10 cells from this submarket each garnered enough weight to warrant an

individual tapestry label not mentioned above. This submarket cluster did not appear in

either the First foci (Figure 5-3) or the Third foci (Figure 5-5) spatial distribution

choropleth maps, whose values appeared to be much more closely distributed to each

other spatially.










Table 5-1. Total revenue statistics (1x1 mile grid cell overlay).
1x1 Mile Grid Total Revenue 2004 Total Revenue 2008 Total Revenue Change

N of Cases 161 161 161
Minimum 0 0 -3,074.11
Maximum 3,103.61 6,986.81 3,998.20
Range 3,103.61 6,986.81 7,072.31
Sum 71,921.23 104,639.82 32,718.59
Median 228.32 388.57 90.04
Arithmetic Mean 446.716 649.937 203.221
Standard Error of Arithmetic 47.373 66.429 68.248
Mean
95.0% Lower Confidence Limit 353.158 518.746 68.439
95.0% Upper Confidence Limit 540.273 781.127 338.003
Trimmed Mean (10%, Two Sided) 325.096 484.932 148.415
No. of Observations Trimmed Out 34 34 34
Standard Deviation 601.101 842.888 865.964
Variance 361,321.97 710,459.51 749,894.46
Coefficient of Variation 1.346 1.297 4.261
Skewness(G1) 2.18 3.61 0.682
Standard Error of Skewness 0.191 0.191 0.191
Kurtosis(G2) 5.784 20.546 4.58
Standard Error of Kurtosis 0.38 0.38 0.38
Shapiro-Wilk Statistic 0.742 0.67 0.892
Shapiro-Wilk p-value 0 0 0
Anderson-Darling Statistic 11.541 12.81 5.657
Adjusted Anderson-Darling 11.595 12.87 5.684
Statistic
p-value <0.01 <0.01 <0.01










Table 5-2. Average revenl
1x1 Mile Grid

N of Cases
Minimum
Maximum
Range
Sum
Median
Arithmetic Mean
Standard Error of Arithmetic
Mean
95.0% Lower Confidence Limit
95.0% Upper Confidence Limit
Trimmed Mean (10%, Two
Sided)
No. of Observations Trimmed
Out
Standard Deviation
Variance
Coefficient of Variation
Skewness(G1)
Standard Error of Skewness
Kurtosis(G2)
Standard Error of Kurtosis
Shapiro-Wilk Statistic
Shapiro-Wilk p-value
Anderson-Darling Statistic
Adjusted Anderson-Darling
Statistic
p-value


ue statistics (1x1 mile grid cell overlay).
Average Revenue Average Revenue
2004 2008
161 161
0 0
341.74 696.51
341.74 696.51
10,900.25 19,237.22
60.147 97.897
67.703 119.486
5.47 8.3


56.9


78.506
57.094


103.095
135.877
102.995


69.408
4,817.44
1.025
1.541
0.191


105.312
11,090.71
0.881
2.432
0.191


3.364
0.38
0.845
0
5.289
5.315

<0.01


8.503
0.38
0.79
0
7.295
7.329

<0.01


Average Revenue
Change
161
-243.14
640.41
883.55
8,336.98
33.445
51.782
9.094

33.823
69.741
39.413

34

115.385
13,313.67
2.228
1.787
0.191
6.24
0.38
0.865
0
5.209
5.234

<0.01










Table 5-3. Customer court
1x1 Mile Grid

N of Cases
Minimum
Maximum
Range
Sum
Median
Arithmetic Mean
Standard Error of Arithmetic
Mean
95.0% Lower Confidence Limit
95.0% Upper Confidence Limit
Trimmed Mean (10%, Two
Sided)
No. of Observations Trimmed
Out
Standard Deviation
Variance
Coefficient of Variation
Skewness(G1)
Standard Error of Skewness
Kurtosis(G2)
Standard Error of Kurtosis
Shapiro-Wilk Statistic
Shapiro-Wilk p-value
Anderson-Darling Statistic
Adjusted Anderson-Darling
Statistic
p-value


It statistics (1x1 mile grid cell overlay).
Customer Count Customer Count
2004 2008
161 161
0 1
109 115
109 114
1,390.00 1,459.00
3 3
8.634 9.062
1.329 1.347


6.008
11.259
4.472


6.401
11.723
4.591


16.866
284.459
1.954
3.7
0.191
16.15
0.38
0.529
0
25.074
25.193


17.097
292.296
1.887
3.59
0.191
15.436
0.38
0.521
0
27.196
27.325


<0.01


<0.01


Customer Count
Change
161
-16
14
30
69
0
0.429
0.274

-0.112
0.969
0.331

34

3.473
12.059
8.103
0.229
0.191
5.586
0.38
0.849
0
8.831
8.873

<0.01










Table 5-4. Transaction statistics (1x1 mile grid cell overlay).
1x1 Mile Grid Transactions 2004 Transaction 2008 Transaction Change

N of Cases 161 161 161
Minimum 0 0 -33
Maximum 44 29 24
Range 44 29 57
Sum 847 891 44
Median 3 4 1
Arithmetic Mean 5.261 5.534 0.273
Standard Error of Arithmetic Mean 0.55 0.442 0.57
95.0% Lower Confidence Limit 4.175 4.662 -0.853
95.0% Upper Confidence Limit 6.347 6.406 1.4
Trimmed Mean (10%, Two Sided) 3.756 4.48 0.583
No. of Observations Trimmed Out 34 34 34
Standard Deviation 6.977 5.602 7.239
Variance 48.682 31.388 52.4
Coefficient of Variation 1.326 1.012 26.487
Skewness(G1) 2.182 1.997 -0.753
Standard Error of Skewness 0.191 0.191 0.191
Kurtosis(G2) 6.37 4.475 4.019
Standard Error of Kurtosis 0.38 0.38 0.38
Shapiro-Wilk Statistic 0.748 0.784 0.927
Shapiro-Wilk p-value 0 0 0
Anderson-Darling Statistic 12.003 9.769 3.403
Adjusted Anderson-Darling Statistic 12.06 9.816 3.419
p-value <0.01 <0.01 <0.01










Table 5-5. Total revenue statistics (1.5x1.5 mile grid cell overlay).
1.5x1.5 Mile Grid Total Revenue 2004 Total Revenue 2008 Total Revenue Change

N of Cases 128 128 128
Minimum 0 0 -3,074.11
Maximum 3,103.61 5,130.88 4,094.81
Range 3,103.61 5,130.88 7,168.92
Sum 53,694.99 68,438.11 14,743.12
Median 151.28 293.805 46.955
Arithmetic Mean 419.492 534.673 115.181
Standard Error of Arithmetic 54.764 66.592 81.115
Mean
95.0% Lower Confidence Limit 311.124 402.899 -45.332
95.0% Upper Confidence Limit 527.86 666.447 275.693
Trimmed Mean (10%, Two Sided) 278.404 373.291 100.17
No. of Observations Trimmed Out 26 26 26
Standard Deviation 619.584 753.406 917.714
Variance 383,884.47 567,620.99 842,199.22
Coefficient of Variation 1.477 1.409 7.968
Skewness(G1) 2.304 2.764 0.42
Standard Error of Skewness 0.214 0.214 0.214
Kurtosis(G2) 5.756 11.093 4.148
Standard Error of Kurtosis 0.425 0.425 0.425
Shapiro-Wilk Statistic 0.699 0.703 0.915
Shapiro-Wilk p-value 0 0 0
Anderson-Darling Statistic 12.224 10.618 3.659
Adjusted Anderson-Darling 12.297 10.681 3.68
Statistic
p-value <0.01 <0.01 <0.01










Table 5-6. Total rev. change stats (1x1 mile grid cell overlay offset foci 1,2, and 3).
1x1 Mile Grid Total Revenue Change Total Revenue Change Total Revenue Change
Foci 1 Foci 2 Foci 3
N of Cases 193 194 183
Minimum -2,279.51 -7,333.06 -2,643.91
Maximum 3,029.86 4,159.85 4,584.89
Range 5,309.37 11,492.91 7,228.80
Sum 3,895.04 14,678.11 23,714.39
Median 10 28.42 0
Arithmetic Mean 20.182 75.66 129.587
Standard Error of 55.214 82.327 69.213


Arithmetic Mean
95.0% Lower Confidence
Limit
95.0% Upper Confidence
Limit
Trimmed Mean (10%, Two
Sided)
No. of Observations
Trimmed Out
Standard Deviation
Variance
Coefficient of Variation
Skewness(G1)
Standard Error of
Skewness
Kurtosis(G2)
Standard Error of Kurtosis
Shapiro-Wilk Statistic
Shapiro-Wilk p-value
Anderson-Darling Statistic
Adjusted Anderson-
Darling Statistic
p-value


-88.721

129.085

27.362


-86.716

238.037


84.49


767.051
588,366.81
38.008
0.197
0.175


3.116
0.348
0.911
0
6.363
6.388

<0.01


1,146.68
1,314,884.45
15.156
-1.298
0.175


11.417
0.347
0.82
0
9.795
9.834

<0.01


-6.975

266.149

65.981

38

936.291
876,640.35
7.225
1.294
0.18

6.025
0.357
0.846
0
9.117
9.155

<0.01













lx1 Mile Grid Overlay Choropleth Map

Hot Spot Analysis (Getis-Ord Gi*)

Total Revenue Change Alachua County, FL


N




Latitude/Longitude Coordinate System


Figure 5-1. Total revenue change from 2004 to 2008, hot spot analysis (getis-ord gi*)
1x1 mile grid cell.


GiZScore
<-2 58 Std Dev.
-2 58 -1 96 Std. Dev
-1.96 -1.65 Std. Dev.
-1 65- 1 65 Std. Dev.
1 65- 1.96 Std Dev
1 96- 2.58 Std Dev
> 2.58 Std. Dev.












1.5x1.5 Mile Grid Overlay Choropleth Map

Hot Spot Analysis (Getis-Ord Gi*)

Total Revenue Change Alachua County, FL


N




Latitude/Longitude Coordinate System


Figure 5-2. Total revenue change from 2004 to 2008, hot spot analysis (getis-ord gi*)
1.5x1.5 mile grid cell


GiZScore
S<-2 58 Std Dev.
-2.58 -1.96 Std. Dev.
-1 96 -1 65 Std. Dev
-1- 1- 1 65 Std. Dev.
1.65- 1.96 Std. Dev.
1 96- 2.58 Std Dev
S2.58 Std. Dev.









1x1 Mile Grid Overlay Choropleth Map
Hot Spot Analysis (Getis-Ord Gi*)
Total Revenue Change (First Foci, Origin)
Alachua County, FL


E I


I.


U I


El


.eE ,,%


N
W F E


Latitude/Longitude Coordinate System


Figure 5-3. Total revenue change from 2004 to 2008, hot spot analysis (getis-ord gi*)
1x1 mile grid cell (first foci, origin).


GiZScore
SI <-2 58 Std. Dev.
I -2.58- -1 96 Std. Dev.
S-1.96- -1 65 Std. Dev.
-1.65- 1.65 Std. Dev.
1 65- 1 96 Std Dev
1 96- 258 Std Dev
> 2.58 Std. Dev.












1x1 Mile Grid Overlay Choropleth Map
Hot Spot Analysis (Getis-Ord Gi*)
Total Revenue Change (Second Foci, 1/4 Offset)
Alachua County, FL


U.


I .


E E


U-.


*1


N


LatitudeLongitude Coordinate Sytem

Latitude/Longitude Coordinate System


Figure 5-4. Total revenue change from 2004 to 2008, hot spot analysis (getis-ord gi*)
1x1 mile grid cell (second foci, 1/4 mile offset).


GiZScore
I -2 58 Std. Dev.
S-2.58- -1.96 Std. Dev.
S-1.96- -1 65 Std. Dev.
-1.65- 1.65 Std Dev
165-1 96Std Dev
1 96- 2 58 Std Dev
> 2.58 Std. Dev.


E MM












1x1 Mile Grid Overlay Choropleth Map
Hot Spot Analysis (Getis-Ord Gi*)
Total Revenue Change (Third Foci, 1/2 Offset)
Alachua County, FL


U El



GiZScore
SI <-2 58 Std. Dev.
I |-2.58- -1 96 Std. Dev.
-1.96- -1 65 Std. Dev.
-1.65- 1.65 Std. Dev.
1 65- 1 96 Std Dev
1 96- 258 Std Dev
S>2.58 Std. Dev.


W I E


Latitude/Longitude Coordinate System


Latitude/Longitude Coordinate System


Figure 5-5. Total revenue change from 2004 to 2008, hot spot analysis (getis-ord gi*)
1x1 mile grid cell (third foci, 1/2 mile offset).









CHAPTER 6
TRADE AREA CALCULATION: GEODEMOGRAPHICS

The importance of knowing what comprises a customer base and the particular

spending habits of those customers cannot be understated. Marketing Geographers

utilize selective profiling methods to group customers into classifications based on their

consumption history and Tobler's Law. The commonly accepted terminology of this

method of grouping customers by geodemographics is as a Tapestry profile.

When the Geodemographics are studied, the customer base for this particular

retail market/service provider reveal the following results by Tapestry:

Top 5 Tapestry groups by customer count comprised 65% and the Top 10 Tapestry
groups by customer count comprised 87% of the total customer base in 2004.

* 19.4% were 13-In Style
* 17.8% were 14-Prosperous Empty Nesters
* 11.9% were 7-Exurbanites
* 9.3% were 16-Enterprising Professionals
* 6.5% were 33-Midlife Junction
* 5.9% were 28-Aspiring Young Families
* 5.0% were 63-Dorms to Diplomas
* 4.9% were 26-Midland Crowd
* 4.4% were 55-College Towns
* 2.0% were 46-Rooted Rural

Top 5 Tapestry groups by customer count comprised 66% and the Top 10 Tapestry
groups by customer count comprised 88.4% of the total customer base in 2008.

* 20.9% were 13-In Style
* 17.8% were 14-Prosperous Empty Nesters
* 12.5% were 7-Exurbanites
* 8.1% were 16-Enterprising Professionals
* 6.7% were 28-Aspiring Young Families
* 6.0% were 33-Midlife Junction
* 5.6% were 63-Dorms to Diplomas
* 4.6% were 26-Midland Crowd
* 4.2% were College Towns
* 2.0% were Rooted Rural









Top 5 Tapestry groups by customer count comprised 55% of the Alachua County, FL
customer base in 2004.

* 18.4% were 13-In Style
* 19.7% were 14-Prosperous Empty Nesters
* 0.0% were 7-Exurbanites
* 10.8% were 16-Enterprising Professionals
* 6.4% were 33-Midlife Junction

Top 5 Tapestry groups by customer count comprised 54% of the Alachua County, FL
customer base in 2004.

* 17.7% were 13-In Style
* 19.3% were 14-Prosperous Empty Nesters
* 0.0% were 7-Exurbanites
* 9.6% were 16-Enterprising Professionals
* 7.8% were 28-Aspiring Young Families

The geodemographics show a very consistent overall Tapestry group capture as well as

an individual Tapestry group capture as a percent of the total customer base. There is a

slight variation among some of the groups, but this is negligible and for the most part,

the percent of overall customer Tapestry capture remained constant throughout the five

year study period. When we examine the Alachua County, FL customer Tapestry

capture, it is interesting to note that the Exubanites Tapestry group drops to a value of

zero for both time periods. This is indicative of a mobile customer Tapestry group

coming into Alachua County, FL specifically for this good/service or, if already here,

consuming this good/service, but reporting a home address outside of the county. This

is typically the case for a University town, such as Gainesville, FL, where students often

report their home of record as being collocated with the parent's address, are the

parents themselves visiting their children in college, are alumni visiting for a University

event, are snowbirds passing through, or frequent this retail/service provider due to

other intangible positive externalities provided only at this location. This is important









because these Exurbanite Tapestry consumers do not live locally, but visit this retail

outlet and spend enough to make it into the Top 5 list by customer count.

The individual Tapestry groups were selected from the entire 2008 year group

dataset and were used to create the mean center from that group. Recall, from earlier

discussion, that the mean center is the gravitational center of a spatially distributed

dataset. Figure 6-1 is a choropleth map showing the total revenue change and the

Alachua County, FL Tapestry profile group mean center displayed as an overlay. The

map indicates that the correlation indicates that the Top 5 Tapestry groups are spending

within the mid range of values calculated for total revenue change. For a detailed

description of each Tapestry group, refer to Appendix A.












1x1 Mile Grid Overlay Choropleth Map
Total Revenue Change and Tapestry Mean Center
Alachua County, Florida


E


F 1


Tapestry Profile Mean Center
In Style
Prosperous Empty Nesters
O Enterprising Professionals
O Aspiring Young Families
Total Revenue Change
($3,074 11)- ($1,000 00)
S($999.99) $0.00
$0.01 $1,000.00
S$1,000 01 -$2,000 00
$2,000 01 -$4,100 00


L




Latitude/Longitude Coordinate System


Figure 6-1. Highest percentage tapestry profile group mean center (2008) and total
revenue change from 2004 to 2008, 1x1 mile grid cell overlay.


I I


t~i









CHAPTER 7
TRADE AREA CALCULATION: FUTURE CONJECTURE

This analysis has demonstrated the importance to Marketing Geography of the

geographic object as a container of attribute data. MAUP has been demonstrated to

exist with customer data. This thesis is the first to demonstrate MAUP using customer

data. The implication is that trade area analyses need to both hold geography constant,

and test for the existence of bias attributable to the manner in which geography is held

constant.13 This should be integral with best business practices, but has not been

previously recognized by those practicing marketing geography.

Trade area analysis is affected by MAUP. Because of recent advances in

geospatial technology, GIS automation can be used to detect and correct for MAUP.

See Figure 7-1. It is technically feasible and pragmatically important to report time

sensitive data within the GIS environment. For instance, with an automated process

using GIS and an hourly input of consumption data, it is realistic to expect that the

computer aided execution of the methods described above will provide high value

added to the business decision. However, to minimize visual and statistical bias, the

conceptualization of MAUP and the associated algorithmic framework introduced here

should be applied; and that application should be executed at various scales with

multiple variables. These applications add value to business, and have relevance to

other disciplines studying spatially distributed phenomena, such as tracking of infectious

disease, package delivery shipments, terrorist threat/incident reporting, watch/no-fly

lists, and home sales data. By adding GIS geospatial processes to available data and

13 By the phrase "holding geography constant" I mean the necessity of using same sized and same
shaped containers for the geographic data. The containers should be packable leaving no area excluded.
The literature on Central Place Theory has discussed similar requirements. For further discussion, see
King, L. 1984. Central Place Theory, Sage Publications: Beverly Hills, CA.









computer generation efficiency, there is value added to the business decision-making

process and the formation of a Common Operating Picture (COP)14 that is applicable in

"real time".


14A common operating picture (COP) is a single identical display of relevant (operational)
information that facilitates collaborative planning and assists all echelons to achieve situational
awareness. Traditionally, headquarters prepares maps electronically with various symbols to
show the locations of significant phenomena and other relevant information.


100















n-1 ... lx1


1.5 x 1.5 ... n+1


Figure 7-1. Modeling maup flow chart.


101


Foci

Fl

F2

F3

Fn









CHAPTER 8
TRADE AREA CALCULATION: CONCLUSION

The next step for the work presented in this thesis is to create geospatial

mapping software that detects the MAUP variables (time, area, scale, size, shape,

direction, distance, offset foci, etc.) and calculates the threshold values that tip the

results into MAUP. Ideally, this software would scan the study area applying the

created template every 45 degrees (8 directions), through 4 foci offsets, and 3 different

grid cell sizes resulting in 96 snapshots of the trade area. Of course, this is purely

subjective and would be dependent upon the variables being studied as well as the

expertise and judgment of the analyst. The proposed software would incorporate the

functionality of GIS, tapestry profiles, and fractal geometry to calculate a potential

strength of trade area value that has a marketable identity. "While the issue of scale has

been widely examined in various aspects of physical geography, the MAUP has been

largely ignored despite its presence in various types of large scale spatial data analysis"

(Dark and Bram, 2007). Conceptually, the proposed software would be an integration of

existing open-source GIS software and open-source "R" statistical software. As a side

note to this conjecture, it would be advantageous to have the tapestry data updated on

an annual basis for more real time applications while also having access to historical

tapestry data. This thesis points to a direction for further development of the body of

business geographic knowledge and procedures; it merely scratches the surface when

considering the potential importance of MAUP to the business decision.

This thesis brings to light that traditional business education including education

for various professional degrees such as pharmacy and medicine, does not prepare the

entrepreneur practitioner to succeed in business. It is essential for entrepreneurs to


102









realize that cannibalization and market penetration may be difficult to attain in

geographic regions in which competitors are already established. New entering

competitors might only be able to compete on the basis of price competition or intense

advertising which raises costs and lowers profit margins. The bias then is for new

entrants to the market to locate in newly developed areas or by providing a good or

service previously not offered in the market.

The value of this thesis extends beyond academic contributions. The results of

the thesis are also of value to the "Client" who provided the data. Benefits to the "Client"

include the provision of a geospatial overview of the Client's trade area, and sensitivity

(or non sensitivity) to the dynamics within the trade area. Those dynamics include

changing competition and customer change.


103









APPENDIX
COMMUNITY TAPESTRY SEGMENTATION SUMMARY DESCRIPTIONS

1. Top Rung

Top Rung is the wealthiest consumer market, representing less than 1% of all U.S.

households. The median household income of $179,000 is three and a half times that of

the national median, and the median net worth of $556,400 is more than five times that

of the national level. The median home value is approximately $1,014,600. These highly

educated residents are in their peak earning years, 45-64, in married-couple

households, with or without children. The median age is 42.3 years. With the purchasing

power to indulge any choice, Top Rung residents travel in style, both domestically and

overseas. This is the top market for owning or leasing a luxury car; residents favor new

imported vehicles, especially convertibles. Exercise and community activities are part of

their busy lifestyle. Avid readers, these residents find time to read two or more daily

newspapers and countless books.

2. Suburban Splendor

These successful suburbanites are the epitome of upward mobility, just a couple of

rungs below the top, situated in growing neighborhoods of affluent homes, with a

median value of $408,100. Most households are comprised of two-income, married-

couple families with or without children. The population is well-educated and well

employed, with a median age of 40.5 years. Home improvement and remodeling are a

main focus of Suburban Splendor residents. Their homes feature the latest amenities

and reflect the latest in home design. Residents travel extensively in the U.S. and

overseas for business and pleasure. Leisure activities include physical fitness, reading,


104









and visiting museums, or attending the theater. This market is proactive for tracking

investments, financial planning, and holding life insurance policies.

3. Connoisseurs

Second in wealth to Top Rung, but first for conspicuous consumption, Connoisseurs

residents are well-educated and somewhat older, with a median age of 45.4 years.

Although residents appear closer to retirement than child rearing, many of these married

couples have children who still live at home. Their neighborhoods tend to be older

bastions of affluence where the median home value is $664,500. Growth in these

neighborhoods is slow. Residents spend money for nice homes, cars, clothes, and

vacations. Exercise is a priority; they work out weekly at a club or other facility, ski, play

golf, snorkel, play tennis, practice yoga, and jog. Active in the community, they work for

political candidates or parties, write or visit elected officials, and participate in local civic

issues.

4. Boomburbs

The newest additions to the suburbs, Boomburbs communities are home to younger

families who live a busy, upscale lifestyle. The median age is 33.8 years. This market

has the highest population growth, at 4.6% annually, more than four times that of the

national figure. The median home value is $308,700, and most households have two

workers and two vehicles. This is the top market for households to own projection TVs,

MP3 players, scanners, and laser printers, as well as owning or leasing full size SUVs.

It's the second ranked market for owning flat-screen or plasma screen TVs, video game

systems, and digital camcorders, as well as owning or leasing minivans. Family

vacations are a top priority. Popular vacation destinations are Disney World and


105









Universal Studios in Florida. For exercise, residents play tennis and golf, ski, and go

jogging.

5. Wealthy Seaboard Suburbs

Wealthy Seaboard Suburbs neighborhoods are established quarters of affluence,

located in coastal metropolitan areas, primarily along the California, New York, New

Jersey, and New England coasts. Neighborhoods are older and slow to change, with a

median home value that exceeds $444,600. Households consist of married-couple

families. Approximately half of employed persons are in management and professional

occupations. The median age is 41.7 years. Residents enjoy traveling and shopping.

They prefer to shop at Lord & Taylor, Macy's, and Nordstrom, as well as Costco

Wholesale, their favorite club store. They also purchase many items online or by phone.

Residents take nice vacations, traveling in the U.S. and abroad. Europe, Hawaii,

Atlantic City, Las Vegas, and Disneyland are popular destinations. Leisure activities

include going to the beach, skiing, ice skating, and attending theater performances.

6. Sophisticated Squires

Sophisticated Squires residents enjoy cultured country living in newer home

developments with low density and a median home value of $244,500. These urban

escapees are primarily married-couple families, educated, and well-employed. They

prefer to commute to maintain their semi-rural lifestyle. The median age is 37.4 years.

They do their own lawn and landscaping work, as well as home improvement and

remodeling projects, such as installing carpet or hardwood floors, and interior painting.

They like to barbeque on their gas grills and make bread with their bread-making

machines. This is the top market for owning 3 or more vehicles. Vehicles of choice are


106









minivans and full size SUVs. Family activities include playing volleyball, bicycling,

playing board games and cards, going to the zoo, and attending soccer and baseball

games.

7. Exurbanites

Open areas with affluence define these neighborhoods. Empty nesters comprise 40% of

these households; married couples with children occupy 32%. Half of the householders

are between the ages of 45 and 64 years. The median age is 43.6 years. Approximately

half of those who work hold professional or managerial positions. The median home

value is approximately $255,900; the median household income is $83,200. Financial

health is a priority for the Exurbanites market; they consult with financial planners and

track their investments online. They own a diverse investment portfolio, and hold long-

term care and substantial life insurance policies. Residents work on their homes, lawns,

and gardens. Leisure activities include boating, hiking, kayaking, playing Frisbee,

photography, and birdwatching. Many are members of fraternal orders and participate in

civic activities.

8. Laptops and Lattes

The most eligible and unencumbered market, Laptops and Lattes residents are affluent,

single, and still renting. They are highly educated, professional, and partial to city life,

preferring major metropolitan areas such as New York, Los Angeles, San Francisco,

Boston, and Chicago. The median household income is $91,000; the median age is

38.1 years. Technologically savvy, this is the top market for owning a laptop or

notebook PC; they use the Internet on a daily basis, especially to shop. Their favorite

department store, by far, is Banana Republic. Leisure activities include going to the


107









movies, rock concerts, shows, museums, and nightclubs. These residents exercise

regularly and take vitamins. They enjoy yoga, jogging, skiing, reading, watching foreign

films on video tape/DVD, dining out, and foreign travel. They embrace liberal

philosophies and work for environmental causes.

9. Urban Chic

Urban Chic residents are well-educated professionals living an urban, exclusive

lifestyle. Most own expensive single-family homes with a median value of $633,000.

Married-couple families and singles comprise most of these households. The median

age is 41.4 years. Urban Chic residents travel extensively, visit museums, attend dance

performances, play golf, and go hiking. They use the Internet frequently to trade or track

investments or to shop, buying concert and sports tickets, clothes, flowers, and books.

They appreciate a good cup of coffee while reading a book or newspaper, and prefer to

listen to classical music, all-talk, or public radio programs. Civic-minded, they would

probably work as volunteers.

10. Pleasant-Ville

Prosperous domesticity distinguishes the settled homes of Pleasant-Ville

neighborhoods. Most residents live in single-family homes with a median value of

$326,500; approximately half were built in the 1950s and 1960s. Located in the

Northeast and California primarily, these households are headed by middle-aged

residents, some nearing early retirement. The median age is 39.4 years. Approximately

40% of households include children. Home remodeling is a priority for residents who live

in older homes. Shopping choices are eclectic, ranging from upscale department stores,


108









to warehouse, or club stores. Sports fanatics, they attend ball games, listen to sports

programs and games on the radio, and watch a variety of sports on TV.

11. Pacific Heights

Pacific Heights neighborhoods are found in the high-rent districts of California and

Hawaii. The median home value is approximately $573,600; residents prefer single-

family homes or townhomes. This market is small but affluent; one in two households

earns approximately $76,000 annually. The median age is 38.4 years. Distance does

not deter Pacific Heights residents from keeping in touch with family living overseas, as

they make frequent phone calls and travel overseas to visit. Many households own 3 or

more cell phones. Residents generally visit Disneyland or Las Vegas during the year,

and enjoy playing chess, reading history books, and renting classics on DVD to watch

on their giant screen or projection TVs. This is the top market for owning an Apple iMac

brand PC.

12. Up and Coming Families

Up and Coming Families represents the second highest household growth market and

with a median age of 31.9 years, the youngest of the affluent family markets. The profile

for these neighborhoods is young, affluent families with young children. Approximately

half of the households are concentrated in the South, with another half in the West and

Midwest. Neighborhoods are located in suburban outskirts of midsized metropolitan

areas. The homes are newer, with a median value of $185,500. Because family and

home priorities dictate their consumer purchases, they frequently shop for baby and

children's products and household furniture. Leisure activities include playing softball,

going to the zoo, and visiting theme parks (generally Sea World or Disney World).


109









Residents enjoy watching science fiction, comedy, and family-type video tapes or

DVDs.

13. In Style

In Style residents live in affluent neighborhoods of metropolitan areas. More suburban

than urban, they nevertheless embrace an urban lifestyle. Townhome ownership is

more than double that of the national level; however, more than half of the households

live in traditional single-family homes. Labor force participation is high and professional

couples predominate. The median household income is $67,800. Nearly one-third of

these households include children. The median age is 39.3 years. In Style residents are

computer savvy; they use the Internet daily to research information, track investments,

or shop. They own a diverse investment portfolio, contribute to retirement savings plans,

and hold long-term care and life insurance policies. They enjoy going to the beach,

snorkeling, playing golf, casino gambling, and domestic travel.

14. Prosperous Empty Nesters

Prosperous Empty Nesters are well-established neighborhoods located throughout the

U.S.; approximately one-third are on the eastern seaboard. The median age is 47.2

years. More than half of the householders are aged 55 or older. Approximately 40% of

household types are married couples with no children living at home. Educated and

experienced, residents are enjoying the lifestage transition from child-rearing to

retirement. The median household income is $66,200. Residents place a high value on

their physical and financial well-being, and take an active interest in their homes and

communities. They travel extensively, both at home and abroad. Leisure activities

include refinishing furniture, playing golf, attending sports events, and reading


110









mysteries. Civic participation includes joining civic clubs, engaging in fundraising, and

working as volunteers.

15. Silver and Gold

Silver and Gold residents are the second oldest of the Tapestry segments and the

wealthiest seniors, with a median age of 58.5 years; most are retired from professional

occupations. Their affluence has allowed them to move to sunnier climates. More than

60% of the households are in the South (mainly in Florida); 25% reside in the West,

primarily in California and Arizona. Neighborhoods are exclusive, with a median home

value of $326,600 and a high proportion of seasonal housing. Residents enjoy traveling,

woodworking, playing cards, birdwatching, target shooting, salt water fishing, and power

boating. Golf is more a way of life than a mere leisure pursuit; they play golf, attend

tournaments, watch golf on TV, and listen to golf programs on the radio. They are avid

readers, but allow time to watch their favorite TV shows and a multitude of news

programs.

16. Enterprising Professionals

This fast-growing market is home to young, educated, working professionals, with a

median age of 32.4 years. Single or married, they prefer newer neighborhoods with

townhomes or apartments. The median household income is $66,000. This segment is

ranked second of all the Tapestry markets for labor force participation, at 75%. Their

lifestyle reflects their youth, mobility, and growing consumer clout. Residents rely on cell

phones and PCs to stay in touch. They use the Internet to find the next job or home,

track their investments, and shop. They own the latest electronic gadgets. Leisure

activities include yoga, playing Frisbee and football, jogging, going to the movies, and


111









attending horse races and basketball games. These residents also travel frequently,

domestically and overseas.

17. Green Acres

A "little bit country", Green Acres residents live in pastoral settings of developing

suburban fringe areas, mainly in the Midwest and South. The median age is 39.9 years.

Married couples with and without children comprise most of the households, live in

single-family dwellings. This upscale market has a median household income of

$62,300 and a median home value of $179,700. These do-it-yourselfers maintain and

remodel their homes, painting, installing carpet, or adding a deck, and own all the

necessary tools to accomplish these tasks. They also take care of their lawn and

gardens, again with the right tools. Vehicles of choice are motorcycles and full-size

pickup trucks. For exercise, residents ride their bikes and go water skiing, canoeing,

and kayaking. Other activities include birdwatching, power boating, target shooting,

hunting, and attending auto races.

18. Cozy and Comfortable

Cozy and Comfortable residents are settled, married, and still working. Many couples

are still living in the pre-1970s, single-family homes in which they raised their children.

Households are located primarily in suburban areas of the Midwest, Northeast, and

South. The median age is 41.0 years and the median home value is $164,000. Home

improvement and remodeling are important to Cozy and Comfortable residents.

Although some work is contracted, homeowners take an active part in many projects,

especially painting and lawn care. They play softball and golf, attend ice hockey games,

watch science fiction films on video tapes/DVDs, and gamble at casinos. Television is


112









significant; many households have four or more sets. Preferred cable stations include

QVC, Home & Garden Television, and The History Channel.

19. Milk and Cookies

Milk and Cookies households are comprised mainly of young, affluent married-couple

families. Approximately half of the households include children. The median age for this

market is 33.5 years. Residents preferred single-family homes in suburban areas,

chiefly in the South, particularly in Texas. Smaller concentrations of households are

located in the West and Midwest. The median home value is $131,900. Families with

two or more workers, more than one child, and two or more vehicles is the norm for this

market. Residents are well-insured for the future. The presence of children drives their

large purchases of baby and children's products, and timesavers such as fast food. For

fun, residents play video games, chess, backgammon, basketball, and football, or fly

kites. Favorite cable channels include Toon Disney, The Discovery Health Channel,

ESPNews, and Lifetime Movie Network.

20. City Lights

City Lights are diverse neighborhoods, situated primarily in the Northeast. This dense

urban market is a mixture of housing, household types, and cultures, sharing the same

city walks. Housing types include single-family homes, townhomes, and apartments.

Approximately 35% of households are apartments in buildings with two to four units,

almost four times the national level. Approximately two-thirds of the housing units were

built before 1960. Households include both families and singles. The median age of

37.8 years is slightly older than the U.S. median. City Lights residents are more likely to

spend for household furnishings than home maintenance. They shop at a variety of


113









stores, especially Macy's, Lord & Taylor, The Disney Store, The Gap, and BJ's

Wholesale Club. They favor foreign travel. Being conservative investors, they own U.S.

savings bonds.

21. Urban Villages

Urban Villages neighborhoods are multicultural enclaves of young families, unique to

U.S. gateway cities, located primarily in California. The median age is 30.7 years. All

family types dominate this market. The average family size of 4.12 is the second highest

of all the Tapestry segments. Many households have two wage earners, chiefly

employed in the manufacturing, health care, retail trade, construction, and educational

services industries. The median household income is $56,200. Most residents own

older, single-family homes with a median value of $355,600, and multiple vehicles.

Family and home dictate purchases. To maintain their older homes, time and money are

spent on home remodeling and repairs. Leisure activities include playing soccer and

tennis, renting foreign films, listening to Hispanic and variety radio, and visiting

Disneyland, Sea World, or Six Flags.

22. Metropolitans

Metropolitans residents favor city living in older neighborhoods. Approximately half of

the households are comprised of singles who live alone or with others. However,

married-couple families are 40% of the households. The median age is 37.1 years. Half

of employed persons hold professional or management positions. These neighborhoods

are an eclectic mix of single-family homes and multiunit structures, with a median home

value of $194,100. The median household income is $57,600. Residents lead busy,

active lifestyles. They travel frequently and participate in numerous civic activities. They


114









enjoy going to museums and zoos, and listening to classical music and jazz on the

radio. Refinishing furniture and playing a musical instrument are favorite hobbies.

Exercise includes yoga, roller blading, and hiking/backpacking.

23. Trendsetters

These neighborhoods are located primarily on the West Coast. On the cutting edge of

urban style, Trendsetters residents are young, diverse, mobile, educated professionals

with substantive jobs. The median age is 35.0 years. More than half of the households

are single-person or shared. Most still rent, preferring upscale, multiunit dwellings in

older city districts. The median household income is $56,700. Residents are spenders;

they shop in stores, online, and via the phone. They own the latest laptop computers,

cell phones, and MP3 players, and use the Internet daily. Exercise includes playing

tennis, volleyball, baseball, and golf, as well as ice skating, snorkeling, and yoga.

Leisure activities include traveling, attending rock concerts, and reading biographies.

Residents also enjoy syndicated TV shows such as Access Hollywood and Seinfeld.

24. Main Street, USA

Main Street, USA neighborhoods are a mix of single-family homes and multiunit

dwellings, found in the suburbs of smaller metropolitan cities, mainly in the Northeast,

West, and Midwest. This market is similar to the U.S. when comparing household type,

age, race, educational attainment, housing type, occupation, industry, and household

income type distributions. The median age of 36.3 years matches that of the U.S.

median. The median household income is a comfortable $51,200. Home

homeownership is at 66% and the median home value is $190,200. Active members of

the community, residents participate in local civic issues and work as volunteers. They


115









take care of their lawns and gardens, and work on small home projects. They enjoy

going to the beach and visiting theme parks, as well as playing chess, going bowling or

ice skating, and participating in aerobic exercise.

25. Salt of the Earth

A rural or small town lifestyle best describes the Salt of the Earth market. The median

age is 40.4 years. Labor force participation is higher than the U.S. level and

unemployment is lower. Above-average numbers of employed residents work in the

manufacturing, construction, mining, and agricultural industries. The median household

income is $48,800. Households are dominated by married-couple families who live in

single-family dwellings, with homeownership at 86%. Approximately 28% of the

households own three or more vehicles. Most homes own a truck; many own a

motorcycle. Residents are settled, hardworking, and self-reliant, taking on small home

projects, as well as vehicle maintenance. Families often own two or more pets, usually

dogs or cats. Residents enjoy fishing, hunting, target shooting, attending country music

concerts and auto races, and flying kites.

26. Midland Crowd

Approximately 10.8 million people represent Midland Crowd, Tapestry's largest market.

The median age of 36.3 years parallels the U.S. median. Most households are

comprised of married-couple families, half with children and half without. The median

household income is $48,200. Housing developments are generally in rural areas

throughout the U.S. (more village or town than farm), mainly in the South. Home

ownership is at 84%. Two-thirds of households are single-family structures; 28% are

mobile homes. This is a somewhat conservative market politically. These do-it-


116









yourselfers take pride in their homes, lawns, and vehicles. Hunting, fishing, and

woodworking are favorite pursuits. Pet ownership, especially birds or dogs, is common.

Many households have a satellite dish, and TV viewing includes various news

programs, as well as shows on CMT and Outdoor Life Network.

27. Metro Renters

Metro Renters residents are young (approximately 30% are in their twenties), well-

educated singles, beginning their professional careers in some of the largest U.S. cities

such as New York City, Chicago, and Los Angeles. The median age is 33.6 years; the

median household income is $52,300. As the name Metro Renters implies, most

residents are renting apartments in high-rise buildings, living alone or with a roommate.

Their interests include traveling, reading two or more daily newspapers, listening to

classical music and public radio programs, and surfing the Internet. For exercise, they

work out regularly at clubs, play tennis and volleyball, practice yoga, ski, and jog. They

enjoy dancing, attending rock concerts, going to museums or the movies, and throwing

a Frisbee. Painting and drawing are favorite hobbies. Politically, this market is liberal.

28. Aspiring Young Families

Aspiring Young Families neighborhoods are located in large, growing metropolitan

areas in the South and West, with the highest concentrations in California, Florida, and

Texas. Mainly comprised of young, married-couple families or single parents with

children, the median age for this segment is 30.4 years. Half of the households are

owner-occupied single-family dwellings or townhomes, and half are occupied by renters,

many living in newer multiunit buildings. Residents spend much of their discretionary

income on baby and children's products and toys, as well as home furnishings. Recent


117









electronic purchases include cameras and video game systems. Leisure activities

include dining out, dancing, going to the movies, attending professional football games,

fishing, weight lifting, and playing basketball. Vacations would probably include visits to

theme parks. Internet usage mainly involves chat room visits.

29. Rustbelt Retirees

Most Rustbelt Retirees neighborhoods can be found in older, industrial cities, in the

Northeast and Midwest, especially in Pennsylvania and other states surrounding the

Great Lakes. Households are mainly occupied by married couples with no children and

singles who live alone. The median age is 43.8 years. Although many residents are still

working, labor force participation is below average. More than 40% of the households

receive Social Security benefits. Most residents live in owned, single-family homes, with

a median value of $118,500. Unlike many retirees, these residents are not inclined to

move. They are proud of their homes and gardens, and participate in community

activities. Some are members of veterans' clubs. Leisure activities include playing

bingo, gambling in Atlantic City, going to the horse races, working crossword puzzles,

and playing golf.

30. Retirement Communities

Retirement Communities neighborhoods are found mostly in cities scattered across the

U.S. The majority of households are multiunit dwellings. Congregate housing, which

commonly includes meals and other services in the rent, is a trait of this segment,

dominated by singles who live alone. This educated, older market has a median age of

50.7 years. A third of residents are aged 65 years or older. Although the median

household income is a modest $45,100, the median net worth is $172,000. Good health


118









is a priority; residents visit their doctors regularly, diet and exercise, purchase low-

sodium food, and take vitamins. They spend their leisure time working crossword

puzzles, playing bingo, gardening indoors, canoeing, gambling, and traveling overseas.

They like to spend time with their grandchildren and spoil them with toys. Home

remodeling projects are usually in the works.

31. Rural Resort Dwellers

Favoring milder climates and pastoral settings, Rural Resort Dwellers live in rural

nonfarm areas. These small, growing communities mainly consist of single-family and

mobile homes, with a significant inventory of seasonal housing. This somewhat older

market has a median age of 46.0 years. Most households consist of married-couples

with no children living at home or singles who live alone. A higher than average

proportion of residents are self-employed and work from home. The median household

income is $45,600. Modest living and simple consumer tastes describe this market. The

rural setting calls for more riding lawn mowers and satellite dishes. Lawn maintenance

and gardening is a priority, and households own a plethora of tools and equipment.

Many households own or lease a truck. Residents enjoy boating, hunting, fishing,

snorkeling, canoeing, and listening to country music.

32. Rustbelt Traditions

Rustbelt Traditions neighborhoods are the backbone of older, industrial cities in states

bordering the Great Lakes. Most employed residents work in the service,

manufacturing, and retail trade industries. Most residents own and live in modest single-

family homes that have a median value of $97,000. Households are primarily a mix of

married-couple families, single-parent families, and singles who live alone. The median


119









age is 35.9 years; the median household income is $45,300. Residents prefer to use a

credit union and invest in certificates of deposit. They use coupons regularly, especially

at Sam's Club, work on home remodeling or improvement projects, and buy domestic

vehicles. Favorite leisure activities include hunting, bowling, fishing, and attending auto

races, country music shows, and ice hockey games (in addition to listening to games on

the radio).

33. Midlife Junction

Midlife Junction communities are found in suburbs across the country. Residents are

phasing out of their child-rearing years. Approximately half of the households are

comprised of married-couple families; 31% are singles who live alone. The median age

is 40.5 years; the median household income is $43,600. A third of the households

receive Social Security benefits. Nearly two-thirds of the households are single-family

structures; most of the remaining dwellings are apartments in multiunit buildings. These

residents live quiet, settled lives. They spend their money prudently and do not

succumb to fads. They prefer to shop by mail or phone from catalogs such as J.C.

Penney, L.L. Bean, and Lands' End. They enjoy yoga, attending country music concerts

and auto races, refinishing furniture, and reading romance novels.

34. Family Foundations

Family is the cornerstone of life in Family Foundations communities. A family mix of

married couples, single parents, grandparents, and young and adult children populate

these small, urban neighborhoods, located in large metropolitan areas, primarily in the

South and Midwest. This market represents stability. Hardly any household growth has

occurred since 2000; these neighborhoods experience little turnover. The median age is


120









38.1 years; the median household income is $42,100. Most households are single-

family structures, built before 1970, occupied by owners. Many residents are members

of church boards or religious clubs, and participate in fundraising. Basketball is a

favorite sport; residents play it, attend professional games, watch games on TV, and

listen to games on the radio. They watch court TV shows, sports, and news programs

on TV, and listen to gospel, urban, and jazz radio formats.

35. International Marketplace

Located primarily in cities in coastal gateway states, International Marketplace

neighborhoods are developing urban markets with a rich blend of cultures and

household types. Approximately 70% of households are occupied by families. Married-

couples with children and single parents with children represent 44% of households. A

typical family rents an apartment in an older, multiunit structure. Most of the households

are located in California and Northeastern states. The median age is 30.4 years and the

median household income is $42,600. Top purchases include groceries and children's

clothing. Residents shop at stores such as Marshalls and Costco Wholesale, but for

convenience, they stop at AM/PM or 7-Eleven. They are loyal listeners of Hispanic radio

programs, and prefer to watch movies and sports on TV.

36. Old and Newcomers

Old and Newcomers neighborhoods are in transition, populated by those who are

starting their careers, or are retiring. The proportion of householders in their twenties or

aged 75 years or older is higher than the national level. The median age is 36.6 years.

Spread throughout metropolitan areas of the U.S., these neighborhoods have more

single-person and shared households than families. Many residents have moved in the


121









last five years. Approximately 60% of households are occupied by renters;

approximately half live in mid-rise or high-rise buildings. Residents have substantial life

insurance policies and investments in certificates of deposit, bonds, and annuities.

Leisure activities include roller skating, roller blading, playing golf, gambling at casinos,

playing bingo, and attending college ball games. They listen to classic hits on the radio.

Many residents are members of fraternal orders or school boards.

37. Prairie Living

Agriculture plays an important part of the Prairie Living economy; small, family-owned

farms dominate this stable market, located mainly in the Midwest. Two-thirds of the

households are married-couple families; the median age is 40.5 years. Homeownership

is at 81%; the median home value is $96,300. Although single-family dwellings are

characteristic of these communities, 11% of the households live in mobile homes.

Approximately 36% of the housing units were built before 1940. These residents are big

country music fans, and enjoy hunting, fishing, target shooting, and horseback riding.

They work on their vegetable gardens, vehicles, and home projects. Many are members

of church boards or civic clubs, and get involved in civic issues. Because cable TV can

be unavailable in these rural areas, many households have a satellite dish. Families

with pet cats or dogs are common.

38. Industrious Urban Fringe

Industrious Urban Fringe neighborhoods are found on the fringe of metropolitan cities.

Approximately half of these households are located in the West; 40% are in the South.

Most employed residents work in the manufacturing, construction, retail trade, and

service industries. Family is central, and children are present in more than half of the


122









households. Many live in multigenerational households. The median age is 28.5 years;

the median household income is $40,200. Two-thirds of the households own their

single-family dwellings, with a median value of $131,400. Necessities for babies and

children are among their primary purchases along with toys and video games. Big

movie fans, residents visit the cinema several times a month and watch movies at home

frequently. They prefer to watch syndicated TV and listen to Hispanic radio.

39. Young and Restless

Change is the constant in this diverse market. With a median age of 28.9 years, the

population is young and on the go. About 85% of householders moved in the last five

years. Young and Restless householders are primarily renters, living in apartments in

multiunit buildings. Almost 60% of households are single-person or shared. This

educated market has the highest labor force participation among all the Tapestry

segments, at 75%, and the highest female labor force participation, at 73%. The median

household income is $40,900. Residents use the Internet daily, to visit chat rooms, play

games, obtain the latest news, and search for employment. They read computer and

music magazines, and listen to public radio. They watch movies in the theater and on

video/DVD, attend rock concerts, play pool, go dancing, and exercise weekly at a

facility.

40. Military Proximity

Military Proximity communities depend upon the military for their livelihood. More than

75% of the labor force is in the Armed Forces, while others work in civilian jobs on

military bases. The median household income is $40,100 and the median age is 22.5

years. Two-thirds of the households are composed of married couples with children.


123









Housing types are mainly townhomes and apartments in small multiunit buildings; 93%

are occupied by renters. Residents participate in civic activities and are members of

business clubs. Many homes have a pet, most likely a dog. Residents use the Internet

to trade stocks and make purchases. For exercise, they snorkel, play tennis, practice

yoga, and jog. Families visit theme parks and the zoo, throw Frisbees, and go bowling.

Recent purchases include MP3 players, digital cameras, video game systems, cell

phones, apparel, and jewelry.

41. Crossroads

Young families living in mobile homes typify Crossroads neighborhoods, found in small

towns throughout the South, Midwest, and West. These growing communities are home

to married-couple and single-parent families. The median age is 31.9 years.

Homeownership is at 77% and the median home value is $60,300. More than half of the

householders live in mobile homes; 36% live in single-family dwellings. Employment is

chiefly in the manufacturing, construction, retail trade, and service industries. Many

homes have dogs. Residents generally shop at discount stores, but also frequent

convenience stores. They prefer domestic cars and trucks, often buying and servicing

used vehicles. Residents go fishing, attend auto races, participate in auto racing, and

play the lottery. An annual family outing to Sea World is common. Outer Limits is a

favorite weekly TV show.

42. Southern Satellites

Southern Satellites neighborhoods are rural settlements found primarily in the South,

with employment chiefly in the manufacturing and service industries. Married-couple

families dominate this market. The median age is 37.1 years and the median household


124









income is $37,700. Most housing is newer single-family dwellings or mobile homes with

a median value of $81,400, occupied by owners,. Residents enjoy country living. They

listen to gospel and country music on the radio, and attend country music concerts.

They participate in fishing, hunting, and auto racing. Favorite TV stations are CMT and

Outdoor Life Network. Satellite dishes are popular in these rural locations. Households

own older, domestic vehicles, particularly trucks and 2-door sedans. Residents invest

time in vegetable gardening, and households are likely to own riding mowers, garden

tractors, and tillers.

43. The Elders

The Elders' median age of 73.4 years represents Tapestry's oldest market. The highest

concentration of retiree residents prefer communities designed for senior living,

primarily in warm climates. Half of these households are located in Florida, and 30% are

situated in Arizona or California. Approximately 80% of households collect Social

Security benefits; 48% receive retirement income. These residents are members of

veterans' clubs and fraternal orders. Health-conscious, they take vitamins, visit doctors

regularly, and watch their diets. Leisure activities include traveling, working crossword

puzzles, fishing, attending horse races, gambling at casinos, going to the theater, and

dining out. They play golf, listen to golf on the radio, and watch tournaments on The

Golf Channel. Their daily routine includes watching TV and reading newspapers.

44. Urban Melting Pot

The ethnically rich Urban Melting Pot neighborhoods are made up of recently settled

immigrants; more than half of whom were born abroad. Half of the foreign-born

residents immigrated to the U.S. in the last 10 years. Most rent apartments in high-


125









density urban canyons of large cities, primarily in New York and California.

Approximately half of the housing units were built before 1950. The median age is 35.7

years and the median household income is $37,400. Fashion- and cost-conscious,

these residents love to shop, from upscale retailers to warehouse/club stores. Leisure

activities include going to the beach, visiting theme parks and museums, playing

football, ice skating, and roller blading. Distance does not deter these residents from

contacting family living outside the U.S. They keep in touch with phone calls and foreign

travel.

45. City Strivers

City Strivers are urban denizens of densely settled neighborhoods in major metropolitan

areas, such as New York City and Chicago. Most households are composed of a mix of

family types. The median age is 32.1 years and the median household income is

$36,800. Employment is concentrated in the city, with half of employed residents

working in the service industry, particularly in health care. Approximately 22% are

government workers. Unemployment is twice that of the U.S. level. Housing is mostly

older, rented apartments in smaller multiunit buildings. Primary spending is for

groceries, baby products, and children's essentials. Residents enjoy going to dance

performances, football and basketball games, and Six Flags theme parks. They listen to

urban, all-news, and jazz radio formats, and watch lots of TV, especially movies,

sitcoms, news programs, courtroom TV and talk shows, tennis, and wrestling.

46. Rooted Rural

Rooted Rural neighborhoods are located in rural areas throughout the country;

however, more than three-fifths of the households are located in the South. Households


126









are dominated by married-couple families; approximately one-third of whom already

receive Social Security benefits. The median age is 41.0 years. Housing is

predominantly single-family dwellings, with a strong presence of mobile homes and

some seasonal housing. The median home value is $89,900. Stable and settled,

residents tend to move infrequently. They are do-it-yourselfers, constantly working on

their homes, gardens, and vehicles. Many families have pets. Residents enjoy hunting,

fishing, target shooting, boating, attending country music concerts, and listening to

country music on the radio. Many households have a satellite dish; favorite stations

include Outdoor Life Network and CMT.

47. Las Casas

Las Casas residents are the latest wave of western pioneers. Settled primarily in

California, approximately half were born outside the United States. Young, Hispanic

families dominate these households; 63% include children. This market has the highest

average household size (4.27) among all the Tapestry segments. The median age is

25.4 years and the median household income is $35,400. Most households are

occupied by renters, although homeownership is at 42%. The median home value is

$278,400. Housing is a mix of older apartment buildings, single-family homes, and

townhomes. This is a strong market for purchase of baby and children's products.

Residents enjoy listening to Hispanic radio, reading adventure stories, and playing

soccer. Many treat their children to a family outing at a theme park, especially

Disneyland. When taking a trip, Mexico is a popular destination.

48. Great Expectations


127









Great Expectations neighborhoods are located throughout the country, with higher

proportions found in the Midwest and South. Young singles and married-couple families

dominate. The median age is 33.0 years. Labor force participation is high.

Manufacturing, retail, and service industries are the primary employers. Approximately

half of the households are owners living in single-family dwellings, with a median value

of $100,600; the other half are renters, mainly living in apartments in low-rise or mid-rise

buildings. Most of the housing units in these older suburban neighborhoods were built

before 1960. Residents enjoy a young and active lifestyle. They go out to dinner, to the

movies, to bars, and to nightclubs. They enjoy roller skating, roller blading, playing

Frisbee, chess and pool, and attending auto races. They read music magazines and

listen to rock music on the radio.

49. Senior Sun Seekers

The Senior Sun Seekers market is one of the faster growing markets, located mainly in

the South and West, especially in Florida. Escaping from cold winter climates, many

residents have permanently relocated to warmer areas; others are "snowbirds" who

move south for the winter. Most residents are retired or are anticipating retirement. The

median age is 51.4 years; 62% of the householders are aged 55 years or older. Most

households are single-family dwellings or mobile homes, with a median value of

$107,500. There is a high proportion of seasonal housing. Many residents are members

of veterans' clubs or fraternal orders. They own lots of insurance and consult with a

financial advisor. Leisure activities include dining out, reading (especially boating

magazines), watching TV, fishing, playing backgammon and bingo, working crossword

puzzles, and gambling at casinos.


128









50. Heartland Communities

Heartland Communities neighborhoods are preferred by approximately 6 million people.

These neighborhoods can be found primarily in small towns, primarily in the Midwest

and South. More than 75% of the households are single-family dwellings, with a median

home value of $74,400. Most homes are older, built before 1960. The median age is

41.3 years; nearly one-third of the householders are aged 65 years or older. The

distinctly country lifestyle of these residents is reflected in their interest in hunting,

fishing, woodworking, playing bingo, and listening to country music. In addition to

working on home improvement projects, they are avid gardeners and read gardening

magazines. They participate in civic activities and take an interest in local politics.

Residents order items from catalogs, QVC, and from Avon sales representatives.

51. Metro City Edge

Metro City Edge residents live in older suburban neighborhoods of large metropolitan

cities, primarily in the Midwest and South. This market is home to married-couple,

single-parent, and multi-generational families. The median age is 29.1 years and the

median household income is $30,200. Nearly half of employed residents work in the

service industry. Most households live in single-family dwellings; 14% live in buildings

with 2 to 4 units, many duplexes. Homeownership is at 56% and the median home

value is $74,100. Prudent shoppers, residents buy household and children's items at

superstores and wholesalers. They enjoy watching TV (especially sitcoms and

courtroom TV shows), going to the movies, visiting theme parks, roller skating, and

playing basketball. They read music, gardening, and baby magazines, and listen to

urban and gospel radio.


129









52. Inner City Tenants

Inner City Tenants neighborhoods are a microcosm of urban diversity, located primarily

in the South and West. This multicultural market is young, with a median age of 27.8

years. Households are a mix of singles and families. Most residents rent economical

apartments in mid- or high-rise buildings. Recent household purchases by this market

include video game systems, baby food, baby products, and furniture. Internet access at

home is not typical; those who have no access at home surf the Internet at school or at

the library. Playing games and visiting chat rooms are typical online activities. Residents

frequently eat at fast-food restaurants. They enjoy going to the movies, attending

football and basketball games, water skiing, and playing football, basketball, and

soccer. Some enjoy the nightlife, visiting bars and nightclubs to go dancing.

53. Home Town

These low-density, settled neighborhoods, located chiefly in the Midwest and South,

rarely change. Home Town residents stay close to their home base. Although they may

move from one house to another, they rarely cross the county line. Household types are

a mix of singles and families. The median age is 33.7 years. Single-family homes

predominate in this market. Homeownership is at 61% and the median home value is

$61,800. The manufacturing, retail trade, and service industries are the primary sources

of employment. Residents enjoy fishing and playing baseball, as well as playing bingo,

backgammon, and video games. Favorite cable TV stations include CMT, Nick at Nite,

Game Show Network, and TV Land. When shopping, Belk and Wal-Mart are favorite

stops, but residents also purchase items from Avon sales representatives.

54. Urban Rows


130









With about 1.2 million people, Urban Rows is the smallest Tapestry segment. Row

houses are characteristic of these neighborhoods found primarily in large Northeastern

cities, with much smaller concentrations in the South. Two-thirds of the households are

in Pennsylvania; one-fifth are in Maryland. Homeownership is 62% and the median

home value is $81,300. Most housing was built before 1950. Households are a mix of

family types. Nearly half of the households do not own a vehicle. The median age is

32.9 years. These residents rarely eat out. They prefer BJ's Wholesale Club for general

shopping; preferred grocery stores are Acme, Pathmark, and Giant. Residents enjoy

roller skating; playing baseball; attending basketball games; listening to urban, variety,

and jazz radio programs; and watching sitcoms and sports on TV. Many households do

not subscribe to cable.

55. College Towns

Education is the key focus for College Towns residents. College and graduate school

enrollment is approximately 41%. The median age for this market is 24.5 years, with a

high concentration of 18-24-year-olds. One out of eight residents lives in a dorm on

campus. Students in off-campus housing live in low-income apartment rentals.

Approximately 31% of the households are typically town residents who live in owner-

occupied, single-family dwellings. The median home value is $132,900. Convenience is

the primary consideration for food purchases; residents frequently eat out, order in, or

eat easy-to- prepare food. Many own a laptop computer. In their leisure time, they jog,

go horseback riding, practice yoga, play tennis, rent videos, play chess or pool, attend

concerts, attend college football or basketball games, and go to bars. They listen to

classical music and public radio programs.


131









56. Rural Bypasses

Open space, undeveloped land, and farmland are found in Rural Bypasses

neighborhoods, located almost entirely in the South. This market is home to families

who live in small towns along country back roads. The median age is 37.1 years.

Higher-than-average proportions of employed residents work in the agricultural, mining,

manufacturing, and construction industries. Labor force participation is low and

unemployment is high. Although most households are single-family dwellings, 32% are

mobile homes. Homeownership is at 78% and the median home value is $58,500.

Residents save money by maintaining their homes, gardens, and vehicles themselves.

They enjoy hunting, reading fishing and hunting magazines, and listening to gospel

radio. They prefer to watch courtroom TV and talk shows, as well as cartoons. Recent

purchases include baby products, clothes, and toys.

57. Simple Living

Simple Living neighborhoods are found throughout the U.S., in urban outskirts or

suburban areas. Half of the households are singles who live alone or share housing,

and 32% consist of married-couple families. The median age is 40.1 years.

Approximately one-third of householders are aged 65 years or older; 19% are aged 75

years or older. Housing is a mix of single-family dwellings and multiunit buildings of

varying stories. Some seniors live in congregate housing (assisted living).

Approximately 55% of households are occupied by renters. Approximately 40% of

households receive Social Security benefits. Younger residents enjoy going out

dancing, while seniors prefer going to bingo night. To stay fit, residents play softball and


132









volleyball. Many households do not own a PC, cell phone, or DVD player. Residents

watch a lot of TV, especially sitcoms and science fiction shows.

58. NeWest Residents

Most NeWest Residents rent apartments in mid- or high-rise buildings in primarily in

major western and southern cities. California has the largest concentration of these

households, followed by Texas. Families dominate this market. Children reside in 54%

of the households, either in married-couple or single-parent families. Approximately half

of the population is foreign-born. This young market has a median age of 25.3 years.

Most of the employed residents work in service and skilled labor occupations. These

residents lead a strong family-oriented lifestyle. Budget constraints restrict their

purchases to essentials such as baby food, equipment, and products, as well as

children's clothing. For fun, families go to the movies, visit theme parks, and play

soccer. They like to watch sports on TV, especially wrestling and soccer, and listen to

Hispanic radio.

59. Southwestern Families

These families are the bedrock of the Hispanic culture in the Southwest, more with

children than without. Two-thirds of the households live in owner-occupied, single-family

dwellings with a median home value is $52,100. Most employed residents work in blue-

collar or service occupations. Southwestern Families is an ethnically diverse market,

with a median age of 28.2 years and a median household income of $26,600. Recent

purchases include baby and children's products. Households generally own or lease a

2-door sedan. The grocery store of choice is H.E. Butt. When eating fast food,

Whataburger is a favorite stop. Residents enjoy fishing, water skiing, playing soccer,


133









and going to the movies. They read gardening and parenthood magazines, and listen to

Hispanic and urban radio formats. Typical TV viewing includes comedies, as well as

wrestling and boxing.

60. City Dimensions

Diversity in household type and ethnicity characterize City Dimensions neighborhoods

that are located in large urban cities. Population density remains high, with

approximately 2,900 people per square mile. This market is young, with a median age

of 29.0 years. Nearly 63% of households rent; more than half are apartments in

multiunit structures. Most of the real estate is older. Approximately 70% of the housing

units were built before 1960, 42% of which were built before 1940. Many households

lease their vehicles, preferring Mercury or Ford models. Residents shop at BJ's

Wholesale Club, Kmart, Marshalls, and T.J. Maxx. They enjoy roller skating, playing

soccer and chess, attending auto races and shows, going to the movies, and renting

movies on DVD (especially classics, horror, and science fiction). Video game systems

are quite popular also.

61. High Rise Renters

This segment has the highest percentage of renters among all of the Tapestry

segments; more than nine in ten households are renters in these densely populated

neighborhoods. Approximately 41% rent in buildings with 50 or more units. High Rise

Renters communities are located almost entirely in the Northeast; 86% of the

households are in New York. Residents represent a diverse mix of cultures; many

speak a language other than English. The median age is 29.6 years. Household types

are mainly single-parent and single-person. Part-time work is just as common as full-


134









time. Residents do aerobics and play soccer. They enjoy dancing, attending basketball

and football games, watching movies on video tapes/DVDs, and listening to all-news,

urban, and Hispanic radio. They watch a variety of news programs and are avid viewers

of daytime TV.

62. Modest Income Homes

Modest Income Homes neighborhoods are found primarily in the older suburbs of

metropolitan areas. Single-family dwellings represent more than two thirds of the

housing; 15% are duplexes. The median home value is $52,800. Household types are

mainly single-person and single-parent. However, approximately 64% of households are

family types. The median age is 35.0 years. Slightly more employed residents work

part-time than full-time, mainly in service and blue-collar occupations. At 20%,

unemployment is high. These frugal residents shop at discount stores, do not pay for

Internet access, and rarely eat out. They are content to wait for movies to be shown on

TV instead of going to the theater. They watch daytime and primetime TV, especially

courtroom TV shows and sitcoms, and listen to urban and gospel radio. A favorite cable

channel is BET.

63. Dorms to Diplomas

Dorms to Diplomas is Tapestry's youngest market, with a median age of 21.8 years.

College and graduate school enrollment is approximately 81%. Nearly three-fourths of

employed residents work part-time in low-paying service industry jobs. Approximately

43% of residents live in on-campus dormitories; the remainder rent apartments in off-

campus multiunit buildings. Approximately 90% of households are renters. PCs are a

necessity, and the Internet is easily accessible to research assignments, search for


135









jobs, obtain the latest news, and keep in touch with family. For exercise, residents

participate in a variety of sports. They enjoy going to college football and basketball

games, rock concerts, the movies, and bars, as well as dancing, playing pool, and

renting video tapes/DVDs. They listen to classic hits, public, and rock radio programs.

64. City Commons

City Commons neighborhoods are found in cities of large metropolitan areas, mainly in

the South and Midwest. This younger market has a median age of 24.2 years. Single-

parent families and singles dominate these households, and children abound.

Approximately 77% of the households are renters; approximately 63% of the rentals are

apartments in multiunit buildings, primarily with fewer than 20 units. More residents work

part-time instead of full-time. This market has the highest unemployment rate among all

of the Tapestry segments. Baby and children's products are the major purchases.

Residents enjoy playing basketball, softball, and backgammon. A yearly family outing to

a theme park is common. They prefer courtroom TV shows when watching television;

listen to gospel, urban, and jazz programs on the radio; and read music, baby,

parenthood, and fashion magazines.

65. Social Security Set

Four in ten residents in the Social Security Set segment are aged 65 years or older; the

median age is 44.6 years. Most of these residents live alone. Located in large cities

scattered across the U.S., these communities are dispersed among business districts

and around city parks. The service industry provides more than half of the jobs held by

residents who will work. Households subsist on very low fixed incomes. Most residents

rent apartments in low-rent, high-rise buildings. Many rely on public transportation,


136









because more than half of these households do not own a vehicle. Limited resources

somewhat restrict the purchases and activities of these residents, although many have

invested their savings in stock. They enjoy going to the movies and soccer games, and

reading science fiction. Many households subscribe to cable TV; residents particularly

enjoy watching game shows, sports, and entertainment news shows.

66. Unclassified

Unclassified neighborhoods include unpopulated areas such as parks, golf courses,

open spaces, or other types of undeveloped land. Institutional group quarters, such as

prisons, juvenile detention homes, mental hospitals, or any area with insufficient data for

classification are also included in this category.


137









LIST OF REFERENCES


Aalberts, R., and D. Bible. 1992 Geographic Information Systems: Applications for the
Study of Real Estate. The Appraisal Journal 60: 483-492.

Applebaum W. 1965a. Can Store Location be a Science? Economic Geography 41: 234-
237.

Applebaum W. 1965b. Measuring Retail Market Penetration for a Discount Food
Supermarket a Case Study. Journal of Retailing 41: 1-47.

Applebaum, W. 1966. Methods for Determining Store Trade Areas, Market Penetration
and Potential Sales. Journal of Marketing Research 3: 127-141.

Applebaum, W. and S. Cohen. 1960. Evaluating Store Sites and Determining Store
Rents. Economic Geography 36:1-35.

Baker, S. and K. Baker. 1993. Market Mapping: How to Use Revolutionary New
Software to Find, Analyze, and Keep Customers. New York, NY: McGraw-Hill,
Inc.

Boyce, R. and W. Clark. 1964. The Concept of Shape in Geography. Geographical
Review 54, 4: 561-572.

Dark, S. and D. Bram. 2007. The Modifiable Areal Unit Problem (MAUP) in Physical
Geography. Progress in Physical Geography 31,5: 471-479.

Davies, R. and D. Rogers. 1984. Store Location and Store Assessment Research. New
York, NY: John Wiley & Sons.

Davies, R. 1977. Marketing Geography: With Special Reference to Retailing. London,
England: Methuen.

Epstein, B. 1978. Marketing Geography: A Chronice of 45 Years. Proceedings of the
Applied Geography Conference 1: 372-379.

Fik, T., B. Sidman, and R, Swett. 2005. A Methodology for Delineating a Primary
Service Area For Recreational Boaters Using a Public Access Ramp: A Case
Study of Cockroach Bay. The Florida Geographer 36. 23-40.

Fotheringham, S. 1981. Spatial Structure and Distance Decay Parameters. Annals of
the Association of American Geographers 71. 425-436.

Ghosh, A. and C. Craig. 1986. An Approach to Determining Optimal Locations for New
Services. Journal of Marketing Research 23: 354-362.


138









Ghosh, A. and S. McLafferty. 1987 Location Strategies for Retail and Service Firms.
Lexington, KY: D.C. Heath and Company.

Goldstucker, J., D. Bellenger, T. Stanley, and R, Otte. 1978. New Developments in
Retail Trading Area Analysis and Site Selection. College of Business
Administration, Georgia State University, Atlanta, GA.

Goss, J. 1995a. Marketing the New Marketing: The Strategic Discourse of
Geodemographic Information Systems. Ground Truth. New York, NY: The
Guilford Press.

Goss, J. 1995b. We Know Who You Are and We Know Where You Live: The
Instrumental Rationality of Geodemographics Systems. Economic Geography
71,2: 171-198.

Haynes, K. and S. Fotheringham. 1984. Gravity and Spatial Interaction Models.
Scientific Geography Series. West Virginia University, Morgantown, WV: Sage
Publications.

Heikkila, E. 1998. GIS is Dead; Long Live GIS! Journal of the American Planning
Association 64, 3: 350-360.

King, L. 1984. Central Place Theory. Scientific Geography Series. West Virginia
University, Morgantown, WV:. Sage Publications.

King, L. 1993. Financial institutions Profiting from a Geographic Information System.
Castle G (ed.) Fort Collins: GIS World 57-74.

Maantay, J. 2007. Asthma and Air Pollution in the Bronx: Methodological and Data
Considerations in Using GIS for Environmental Justice and Health Research.
Health and Place 13, 1: 32-56.

Mercurio, J. 1984. Store Location and Store Assessment Research. Davies R and
Rogers D (eds.) New York, NY: John Wiley and Sons 237-262.

Moloney, T. 1989. A Case Study Using a Geographic Information System in Food
Retailing. Operational Geographer 7:23-27.

Nelson, R. 1958. The Selection of Retail Locations. New York, NY: FW Dodge.

Patel, A., T. Fik, and G. Thrall. 2008. Direction Sensitive Wedge-Casting for Trade Area
Delineation. Journal of Real Estate Portfolio Management 14, 2: 125-139.

Pickles, J. (ed.) 1995. Representations in an Electronic Age: Geography, GIS and
Democracy. Ground Truth: The Social Implications of Geographic Information
Systems. New York, NY: Guilford Press.


139









Peuquet, D. and D. Marble. 1990. Introductory Reading in Geographic Information
Systems. London, England: Taylor & Francis.

Ritchey, J. 1984. Developing a Strategic Planning Data Base. Store Location and Store
Assessment Research. Davies, R., and Rogers, D. (eds.). New York, NY: John
Wiley & Sons.

Rogers, D. and H. Green. 1978. A New Perspective on Forecasting Store Sales:
Applying Statistical Models and Techniques in the Analog Approach.
Geographical Review 69: 449-458.

Simmons, M. 1984. Store Assessment Procedures. Store Location and Store
Assessment Research. Davies, R., and Rogers, D. (eds.). New York, NY: John
Wiley & Sons.

Swift, A., L. Liu, and J. Uber. 2008. Reducing MAUP Bias of Correlation Statistics
Between Water Quality and GI Illness. Computers, Environment and Urban
Systems 32, 2: 134-148.

Thomas, I., P. Frankhauser, and M. DeKeersmaeker. 2007. Fractal Dimension Versus
Density of Built-Up Surfaces in the Periphery of Brussels. Papers in Regional
Science 86, 2: 287-308.

Thrall, G. 1987. Land Use and Urban Form. London, England: Methuen.

Thrall, G. 1988. Statistical and Theoretical Issues in Verifying the Population Density
Function. Urban Geography 9: 518-537.

Thrall, G. and S. Thrall. 1994. Business GIS Data, Part Three: ZipPlus 4 Geocoding.
Geo Info Systems 4: 57-60.

Thrall, G. 1995. The Stages of GIS Reasoning. Geospatial Solutions 5, 2: 46-51.

Thrall, G., J. del Valle, and S. Thrall. 1995. Ninety Years of Urban Growth as Described
with GIS: A Historic Geography. Geo Info Systems 5: 20-45.

Thrall, G. and S. McMullin. 2000. Trade Area Analysis: The Buck Starts Here.
Geospatial Solutions 10, 6: 45-49.

Thrall, G. 2000. The Future of GIS in Public Health Management and Practice.
Geospatial Solutions 10, 9: 2-7.

Thrall, G. 2002. Business Geography and New Real Estate Market Analysis. Oxford,
England: Oxford Press.


140









Thrall, G., E. Borden, and S. Thrall. 2002. Delineating Hospital Trade Areas: It's
Practically Brain Surgery. Geospatial Solutions 10, 8: 38-46.

Thrall, G. and N. Mecoli. 2003. Spatial Analysis, Political Support, and Higher Education
Funding. Geospatial Solutions 10, 7: 46-49.

Wofford, L. and G. Thrall. 1997. Real Estate Problem Solving and Geographic
Information Systems: A Stage Model of Reasoning. Journal of Real Estate
Literature 5: 177-201.

Wolf, J. 1969. SYMAP: Computer Graphics for Marketing Management. Journal of
Marketing Research 6: 357-358.


141









BIOGRAPHICAL SKETCH

James Ronald Dietz is the son of Glenn and Myrian Dietz. Born in Coral Gables,

Florida where he lived with his family until graduation from High School in 1989. He

subsequently joined the U. S. Navy for Active Duty service and, upon completion of that

obligation, moved to Jacksonville, Florida where his studies in geography began at

Jacksonville University under the mentorship of Raymond Oldakowski. James Ronald

Dietz received his Bachelor of Arts degree with Honors in geography from Jacksonville

University in 1994, and began his graduate studies at the University of Florida,

Department of Geography, in 1994 while also serving as a Commissioned Officer in the

U. S. Navy. He is currently working towards his Master of Arts degree in geography with

special emphasis on retail market trade area calculation.


142





PAGE 1

1 GEOTEMPORAL TRADE AREA EV ALUATION WITH DYNAMIC GRIDDING TO AVOID THE MODIFIAB LE AREA UNIT PROBLEM (MAUP) By JAMES RONALD DIETZ A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORID A IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 James R. Dietz

PAGE 3

3 To my Mother and Father: I am eternally gr ateful for your loving support and generosity throughout the years.

PAGE 4

4 ACKNOWLEDGMENTS I would like to acknowledge my advisor and mentor, Grant Ian Thrall, for his enduring patience, support, and direction during my pursuit of a Master of Arts degree. He has been instrumental in guiding me towa rd this goal and can take full responsibility for this timely accomplishment. He initiat ed my interest in the Marketing Geography discipline and has supported my i ntellectua l growth with a progressive movement delving into the dynamic realms of the MAUP and Fractal Geometry. I could not ask for a better advisor, as he has given me the academic guidance and moral support necessary to make completion of this thesis a most rewarding and challenging experience. The business that provided the data used in this study prefers to remain anonymous. Their cooperation allowed the topic of this thesis to be feasible. My commitment to the firm was that their time and effort w ould be rewarded by the results of the analysis of their data being both inte resting to them, and relevant to their managerial decisions. This commitment was important in designing the work and the presentation of the completed analysis. I would also like to acknowledge the re st of my committeeTimothy Fik and Youliang Qiufor their continued support and ent husiasm, I am truly grateful. With the combined efforts of everyone in volved, especially those of my committee members, I am approaching the completion of this thes is. Their efforts have made it a truly enjoyable and rewarding experience, for that, I am forever indebted.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDG MENTS .................................................................................................. 4 LIST OF TABLES............................................................................................................ 7 LIST OF FIGURES .......................................................................................................... 8 ABSTRACT................................................................................................................... 11 CHA PTER 1 INTRODUCTION.................................................................................................... 13 2 TRADE AREA CALCULAT ION: AN O VERVIEW .................................................... 17 Marketing Geography ............................................................................................. 17 Value and Just ification ............................................................................................ 21 Tool and Met hodologies .......................................................................................... 22 Terminology ...................................................................................................... 22 Methods............................................................................................................ 23 Customer spo tting me thod ......................................................................... 23 Trade area de lineati on ............................................................................... 27 Summary................................................................................................................ 28 3 TRADE AREA CALCULATION: AN INTEGRATED GIS APPROACH .................... 29 GIS as a Marketing G eography Resear ch Tool ...................................................... 29 Historical Overview ........................................................................................... 29 Customer Spo tting Me thod ............................................................................... 31 Demographic A pplication.................................................................................. 33 Summary................................................................................................................ 34 4 TRADE AREA CALCULAT ION: M ETHODOLOGY................................................. 35 The Core Trade Area Radial Method ...................................................................... 38 The Grid Method..................................................................................................... 42 5 TRADE AREA ALGORITHM ................................................................................... 79 6 TRADE AREA CALCULATION : GEODEMOGR APHICS ........................................ 95 7 TRADE AREA CALCULATION: FUTURE CO NJECTU RE ..................................... 99 8 TRADE AREA CALCULAT ION: CONCL USION ................................................... 102

PAGE 6

6 APPENDIX: COMMUNITY TAPESTRY SEGMENTATION SUMMARY DESCRIPTIONS ................................................................................................... 104 LIST OF RE FERENCES ............................................................................................. 138 BIOGRAPHICAL SKETCH .......................................................................................... 142

PAGE 7

7 LIST OF TABLES Table page 5-1 Total revenue statistics (1x1 mile grid ce ll overla y). ........................................... 845-2 Average revenue statistics (1x1 mile grid ce ll overla y)....................................... 855-3 Customer count statistics (1 x1 mile grid cell over lay)......................................... 865-4 Transaction statistics (1x1 mile grid ce ll overla y)................................................ 875-5 Total revenue statistics (1.5x1 .5 mile grid cell over lay)...................................... 885-6 Total rev. change stats (1x1 mile grid cell overlay offset fo ci 1,2, and 3)............ 89

PAGE 8

8 LIST OF FIGURES Figure page 4-1 Data manipulation and methods fl ow chart. ........................................................ 434-2 Alachua county customer s 2004 directional distribut ion (standard deviational ellips e)................................................................................................................ 444-3 Alachua county custom ers 2004 standard distance........................................... 454-4 Alachua county customer s 2008 directional distribut ion (standard deviational ellips e)................................................................................................................ 464-5 Alachua county custom ers 2008 standard distance........................................... 474-6 Alachua county cust omers 2004+2008 directional distribution (standard deviational ellipse).............................................................................................. 484-7 Alachua county customer s 2004+2008 standard distance................................. 494-8 Alachua county customer growth and decay zones ov erlay of the directional distribution (standard deviati onal ellipse) polygons by dataset year group......... 504-9 2004 and 2008 Customers in alachua c ounty, fl. directional distribution (standard deviation ellipse), symmetrical differ ence overlay.............................. 514-10 Alachua county customer growth and decay zones overlay of the standard distance polygons by dataset y ear group........................................................... 524-11 2004 and 2008 Customers in alac hua county, fl. standard distance, symmetrical diffe rence ov erlay........................................................................... 534-12 2008 Publix supermarket lo cations (1.5 mile buffer)........................................... 544-13 2004 Alachua county customers with in 1.5 miles of the centroid........................ 554-14 2008 Alachua county customers with in 1.5 miles of the centroid........................ 554-15 2004+2008 Alachua county customers wit hin 1.5 miles of the centroid............. 564-16 2004+2008 All eight new competit or locations appearing from 2004................. 564-17 New competitor location num ber 1 appearing fr om 2004 to 2008...................... 574-18 New competitor location num ber 2 appearing fr om 2004 to 2008...................... 574-19 New competitor location num ber 3 appearing fr om 2004 to 2008...................... 58

PAGE 9

9 4-20 New competitor location num ber 4 appearing fr om 2004 to 2008. ..................... 584-21 New competitor location num ber 5 appearing fr om 2004 to 2008...................... 594-22 New competitor location num ber 6 appearing fr om 2004 to 2008...................... 594-23 New competitor location num ber 7 appearing fr om 2004 to 2008...................... 604-24 New competitor location num ber 8 appearing fr om 2004 to 2008...................... 604-25 First moving competitor appearing from 2004 to 2008, (2004 location).............. 614-26 First moving competitor appearing from 2004 to 2008, (2008 location).............. 614-27 Second moving competitor appearing fr om 2004 to 2008, (2004 location)......... 624-28 Second moving competitor appearing fr om 2004 to 2008, (2008 location)......... 624-29 Third moving competitor appearing fr om 2004 to 2008, (2004 location)............. 634-30 Third moving competitor appearing fr om 2004 to 2008, (2008 location)............. 634-31 Fourth moving competitor appearing from 2004 to 2008, (2004 location).......... 644-32 Fourth moving competitor appearing from 2004 to 2008, (2008 location).......... 644-33 2004 Total revenue (dollars ), 1x1 mile grid cell.................................................. 654-34 2008 Total revenue (dollars ), 1x1 mile grid cell.................................................. 664-35 2004 Average revenue per transaction (dollars), 1x1 mi le grid cell.................... 674-36 2008 Average revenue per transacti on (dollars), 1x1m ile grid cell..................... 684-37 2004 Number of Transactions, 1x1 Mile Grid Cell.............................................. 694-38 2008 Number of transacti ons, 1x1 mile grid cell................................................. 704-39 2004 Number of customer s, 1x1 mile grid cell.................................................... 714-40 2008 Number of customer s, 1x1 mile grid cell.................................................... 724-41 Total revenue change from 2004 to 2008, 1x1 mile grid ce ll ............................. 734-42 Total revenue change from 2004 to 2008, 1.5x1.5 mile grid cell ....................... 744-43 1x1 Mile grid cell coordinat e plane..................................................................... 754-44 1x1 Mile grid cell overlay offs et me thod.............................................................. 75

PAGE 10

10 4-45 Total revenue change from 2004 to 2008, 1x1 mile grid cell (first foci, origin). ... 764-46 Total revenue change from 2004 to 2008, 1x1 mile grid cell (second foci, mile o ffset).......................................................................................................... 774-47 Total revenue change from 2004 to 2008, 1x1 mile grid cell (third foci, mile offset)................................................................................................................. 785-1 Total revenue change from 2004 to 2008, hot spot analysis (getis-ord gi*) 1x1 mile gr id cell................................................................................................. 905-2 Total revenue change from 2004 to 2008, hot spot analysis (getis-ord gi*) 1.5x1.5 mile grid cell........................................................................................... 915-3 Total revenue change from 2004 to 2008, hot spot analysis (getis-ord gi*) 1x1 mile grid cell (f irst foci, origin )...................................................................... 925-4 Total revenue change from 2004 to 2008, hot spot analysis (getis-ord gi*) 1x1 mile grid cell (second foci, 1/4 mile offset)................................................... 935-5 Total revenue change from 2004 to 2008, hot spot analysis (getis-ord gi*) 1x1 mile grid cell (third foci, 1/2 mile offset)........................................................ 946-1 Highest percentage tapestry profil e group mean center (2008) and total revenue change from 2004 to 2008, 1x 1 mile grid cell over lay........................... 987-1 Modeling maup flow chart................................................................................. 101

PAGE 11

11 Abstract of Thesis Pres ented to the Graduate School of the University of Florida in Partial Fulf illment of the Requirements for the Degr ee of Master of Arts GEOTEMPORAL TRADE AREA EVALUATION WITH DYNAMIC GRIDDING TO AVOID THE MO DIFIABLE AREA UNIT PROBLEM (MAUP) By James Ronald Dietz August 2010 Chair: Grant Ian Thrall Major: Geography Changes within a trade area of a reta iler can change the profitable performance of the business. Among the mo st important changes to a tr ade area over which the retailer has no control are changes to the demographic composition of households, the entry and exit of competitors, and the relocation of compet itors within the trade area. Geospatial technology provides for cost effective evaluation of changes in the phenomena that affect t he trade area. This provides an opportunity for the retailer to document changes that have occurred and eval uate how those changes affect their enterprise. Important geotemporal measures are customer count change by location, transaction change by location, and revenue change by location. Sustainable best practices of a business firm requires awaren ess of the trajectory of change and thereby enable the firm to appropriately adjust to current and anticipated market conditions. A repeatable methodology is presented that measures changes in business activity across space and through time. A case study of a firm providing retail services in Alachua County, Florida, documents the methodology. This study indicates that further research on geotemporal analysis for business decisions is warranted, especially that which has focus on the Modifiable Areal Unit Problem (MAUP). MAUP reveals if the

PAGE 12

12 geographic container for the data is itself a source of a statistical bias radically influencing both qualitative and statistical results.

PAGE 13

13 CHAPTER 1 INTRODUCTION "Real time" management decisions are crit ical to the long term success of large and small businesses. Large multi-branch reta il chains have for the past two decades been beneficiaries of geospatial analysis a nd geospatial technology. Cost of implementation and limited general public k nowledge of "location based intelligence1" has contributed to smaller locally owned bus inesses from reaping the same benefits as large business. This thesis demonstrates that geospatial analysis provides management of a small business, information that is crit ical to the business decision in "real time2". This thesis integrates a triad of concepts, thereby adding time critical to mainstream business geography literature: retail market/service provider location trade area theory geospatial technology time "Real time" geospatial analysis, hereafter referred to as geotemporal analysis, identifies entry and exit of comp etitors, tracks cu stomer interaction with the store, and evaluates change in the non-cu stomer population of the trade area. Geospatial changes can lead to cannibalization of revenues, a change in market penetration, and prospects to improve the market position. 1 Location Based Intelligence (LBI) allows for real ti me tracking of information to alert the operator and provide an enhanced perspective of situational aw areness that can help improve judgment decisions. http://works.bepress.com/mgmichael/4/ ... http://www.encyclopedia.c om/doc/1G1-199973425.html 2 Real time is when things respond to events as they occur, and may refer to: Real Time Locating Systems (RTLS), are used to track and identify the lo cation of objects in real time using simple, inexpensive nodes (badges/tags) attached to or em bedded in objects and devices (readers) that receive the wireless signals from these tags to determine thei r locations. RTLS typically refers to systems that provide passive (automatic) collection of location information.

PAGE 14

14 Trade area calculations are pr imarily used in retailing to calculate expected store performance before the stores "open their doors." The benefits of predicting store performance include avoidance of financia l losses, zeroing in on profitable opportunities, and not to be underestimated is the ability to negotiate a more advantageous long term lease or purchase price of a site. After transaction, the choice of site is fixed. In the sh ort run, managerial decisions cannot be made on fixed costs. The store managers' degree of freedom subsequently narrows to marketing and some variation in the value platform offered by the business. "Real time" evaluation of a trade area has not entered the academic or practitioner literature. This absence is not because of la ck of importance; rat her, the data has not been available to academics to create a study of the type necessary for a real "real time" demonstration. Businesses that empl oy the technology to be described here are not sharing the benefits of the analysis, and not letting the world know about the existence of their "Geotemporal War Room." I ndeed, real time information is internally recognized as critical to the success of the ent erprise. Real estate has been said to be "information arbitrage (Thrall, 2002)." The be lief among stakeholders is that those with the information can sustain their enterprise in to the future. Those without the information are "throwing dice with their future." It is fortunate that a succe ssful, locally owned medical service provider gave access to the data used here. The data made available is comprised of two time periods, each period being a full year, and with a four year interval sepa rating the two time periods. The methodology of this thesis can be ex tended to data sets with shorter time differences: hourly, daily, and monthly. K nowledge gained from t he two time period

PAGE 15

15 analysis will contribute to further research on geotemporal business decisions. Alachua County provides an ideal test site. Gainesv ille is a comparatively compact city. Traffic congestion keeps the city compact. Most employ ment is at a central location. For these reasons, the land use and urban form of Gaine sville is not complex and mirrors the general theory on urban spatial st ructure (Thrall, 1987). It is also fortunate that the business that provided access to the data wa s a small locally owned enterprise. Large multi-branch retail chains hav e benefited from geospatial analysis and technology. The analysis here documents that geospatial analysis can also benefit the small business. Adoption of geospatial analys is by small business is expected to occur, in large measure by the ready and near ly ubiquitous acceptance of Google Earth and Google Maps. Google at this time does not offer th e geospatial functionality required for this analysis; however, it is inevitable that the functionality is offered. In the meantime, the necessary functionality is available via specialized software packages. The technology employed in this thesis is ESRI Business Analyst, R-Statistics, and Caliper Corporations Maptitude GIS. The particular str ength of Business Analyst is its technology for geocoding and assignment of lif estyle segmentation profiles (this will be explained a later part of this thesis). Mapt itude GIS software is capable of carrying out the spatial data operations, and spatial vis ual operations required in this thesis. The customer data is from a locally owne d, privately developed and owner managed business. The business is a medical servic e provider. Service consumers make an appointment, travel to and from the 'brick and mo rtar" facility. Familiarity with and trust of the research team was a necessary preconditi on for this work to be performed. A needs assessment was casually executed, without intruding with business operations. The

PAGE 16

16 needs assessment revealed that the busi ness had been maintaining a computerized database as part of its schedu ling and billing o perations. However, the data had not previously been considered an asset, and the management had never seen their own data as geospatial. Google's geospatial capabilities wetted their appetites. The data was received without any m onetary transaction. A non disclosure agreement was signed committing that the customer addresse s and identities remained anonymous, and that the business remained anonymous as well. The business did however incur a cost due to di sruption of normal routines, and allocation of resources for data extraction from their tr ansactions files. This effort was not insignificant, and an indication of the importance in which t he business management considered this study. The quid-pro-quo is that the business facilita ting this study by a llowing access to its data, receive a copy of the results in return. The data covered the years 2004 and 2008. Th is time period too was fortunate for the study though not for the economy. As the national economy entered a recession it would have been expected that the decline would transfer in to a decline in demand for the goods and services offered by the case st udy business. In other words, as income, expected income, and consumer confidence declined, it would be reasonable to expect that the revenues would change as well. T he more inelastic the consumers demand, the less changes in the national economy woul d translate into local demand changes. The data included address of the customer, count of each customers transactions (visits), revenue generated by each custom er within the time frame. Customer addresses gives rise to lifestyle segmentat ion profile derivatio n via ESRI Business Analyst.

PAGE 17

17 CHAPTER 2 TRADE AREA CALCULATION: AN OVERVIEW Marketing Geography The foundations of Marketing Geography can be traced to literature from the early 1900 s where practitioners first began utilizing the methodologies now being used within this field. Geographers have long known that there are causal determinants to many spatial distribution patterns. With a combination of quantitat ive and qualitative analysis, the cause of spatial patterns c an be detected, hypotheses formulated, and statistically confirmed or refu ted. These steps are important to predicting future spatial distributions of phenomena, or spatial dist ributions of phenomena elsewhere. When the phenomena is important to the success of a business, then the business must include the present and predicted spatial distributions when making time critical operational decisions and creating l ong term managerial strategy. All retailing operations are complex but most executives would agree that location (and its associated attributes) contri butes more to the long-term success of the retail unit than does any other factor (Ritchey 1984). Additionally, the optimal location of services and the significance of a location strategy is to help assure a successful [business] undertaking (Merc urio, 1984). The earliest atte mpts to apply geographic research techniques to retail outlets date to the 1920s when the attempt was made to determine, on behalf of large multi-branch reta il firms, the relative value of one site, compared to the relative advantages of other sites in the same region. These first attempts and early studies employed methodol ogies that were mainly subjective and based on factors believed to be signific ant (Goldstucker et al., 1978). Applebaum suggested that the origins of Marketing Geography may be traced to the beginning of

PAGE 18

18 the twentieth century when chain compani esespecially tobacco shopsbegan to conduct detailed surveys of pedestrian flows al ong streets in order to identify the most desirable sites (highest foot traffic) within the main centers of towns (Davies, 1977). These early studies were marginally effe ctive at improving the business decisionmaking process, but they were the first r eal use of the foundations used in todays business geography techniques. William Applebaum is widely recogniz ed as a pioneer in the discipline of Marketing Geography and is described by Davi es (1977) as the chief architect of Marketing Geography as a separate field of st udy in the United States. Store location research, as both an academic and practica l area of inquiry, owes much to the formative work of William Applebaum. From about 1945 until his death in 1978, Applebaum was a colossus amongst those teaching the new subject of Marketing Geography, lecturing at the Harvard Busi ness school (Davies and Rogers, 1984). Although Applebaum had a degree in geography from the University of Minnesota, his work was not widely accepted or val ued within the discipline of geography.3 Turning a bad thing into an opportunity, Applebaum grav itated to an audience that was highly receptive business practitioners and busi ness academics. Although his contributions, as well as those other geographers from this era, were mainly cartographic representations of market areas, Applebaum was instrument al in opening the avenues of thought that created Market ing Geography as an accepted discipline within the field 3 It might be asked, "why did the Association of American Geographers" wait until 2008 to create its Business Geography Specialty Group (see www.Busi nessGeography.info ). Marketing Geography is a subset of the larger emerging literature on Busine ss Geography. The focus of Business Geography is geospatial analysis to improve the business decisi on, while the focus of Marketing Geography is essentially retail location. This thesis brings in formation relevant to Marketing Geography and analyzes that information to be relevant to the large arra y of decisions that must be made in business.

PAGE 19

19 of geography itself. Among the list of cont ributions made by Applebaum, he is well known for the customer spotting method, the analog method, methods for determining market penetration, methods fo r determining store rents, and store site evaluation. His understanding of both the busi ness/marketing perspective and the geographic concepts allowed Applebaum to become re cognized as the expert in hi s field during his own time. Applebaum is regarded as one of the father s of Marketing Geography, and one of the most important geographers of the 20th century. Another pioneer in the advancement of the Marketing Geography discipline is David Huff, whose work with spatial interaction models was a revolutionary concept for the business community. Huff was a business student studying marketing at University of Washington during the late 1950s. There he participated in "brown bag" lunches with geography graduate students. U pon graduation he became a faculty member at University of California at Los Angeles.4 There, and independent of his geography colleagues, he formulated what is today regar ded as "the Huff model." Huff's formulation is the kingpin of spatial interaction models used in applications ranging from marketing geography to transportation studies. Initially, Huff's work was not readily accepted in mainstream geographic thought, but inst ead became one of the most studied formulations in Regional Science which was at the interface between the disciplines of economics and geography.5 Huff also purposely disse minated his work to business practitioners by publishing examples of his work in trade publications. As Applebaum did before him, Huff found an ent husiastic audience that was receptive to the value of 4 Personal communication between David Huff and Grant Thrall, narrated to Ron Dietz. 5 For more on the Huff model and spatial interaction models in general, see Kingsley Haynes and A. Stewart Fotheringham, 1984, Gravity and Spatial Interaction Models Sage Publications: Beverly Hills CA.

PAGE 20

20 his work. Huff presented purely geographic based concepts and methodologies to this audience which helped to diffuse these Ma rketing Geography techniques outside of the confines of geographic circles. Today, Davi d Huff is retired from the Department of Marketing at University of Texas at Aust in, and is a consultant to Environmental Systems Research Institute (ESRI).6 Early use of site location methodolog ies were encumbered by the use of checklists; decisions made wit h the use of mere checklists were in the final analysis considered, and likely were, highly subjective (for examples see Nelson 1958; Goldstucker et al., 1978; Applebaum and Cohen, 1960). The perceived absence of a sound theoretical framework limited the academic acceptanc e of marketing geography within the geographic di scipline. The quantitative revo lution of the 1960s would change this. Marketing geographers were quick to apply regression analysis, and include the "checklist" fields as independent variables. Revenues per square foot were one of the more common dependent variables in the regression equation. Marketing Geography subsequently gained high respect within t he discipline of geography, and became accepted as a distinct scient ific subfield of geography. However, as the body of marketing geography knowledge increased, and as more and more businesses adopted marketing geography methodologies, the ne wly emerging geography academics were drawn to Colleges of Business in which it had originated.7 Nevertheless, sound 6 Among Professor David Huff's consulting activities is overseeing the Business Analyst add-in to ESRI's ArcMap. Business Analyst is a combination of data and software. In particular, Bu siness Analyst includes the Huff model, and the means to calibrate the Huff model with a firm's own data. 7 Avjit Ghosh is a very good example. He received his Ph.D. in Geography (1979), and MA in Geography (1977), both from University of Iowa under the dire ction of Professor Gerard Rushton. Ghosh has been one of the most important contributors to advancing th e literature on spatial interaction models and their use in business. He has been Dean of the College of Busi ness at University of I llinois, and today remains as a Professor of Business Administra tion at University of Illinois.

PAGE 21

21 geographic analysis, geographic technology, geographic theory, when used to solve actual business decisions, has evolved to form a body of knowledge of best practices. Value and Justification Goldstucker et al. (1978) reasoned that retail market/service providers trade area size and shape is affected by: (1) the ex tent of product differentiation and the relative effectiveness of brand promotion; (2) the range of choi ce in administering pricing made possible by product differentiation, oligopoly, and other influences; (3) the ratio of fixed to total cost s; (4) the economies of scales of production at each center; and (5) the availability of adequate markets within a radius of economical outreach. Each of these contains an aspec t of geography that is useful to the overall calculation of a trade area, while others also include the intangible aspects of Marketing Geography. That shape has not been used more extensively in geography may result from the earlier inability to measure it precisely. M any of the terms and m easurements that have been used to identify shape have been inade quate. Moreover, geographers have been primarily concerned with an ideographic, ra ther than a nomothetic, approach to geographical problems (Boyce and Clark, 1964). Simmons (1984) states that if a store is achieving a low share of trade, it is vital to establish th e image of the store compared with selected competition to find out why people shop at on store rather than another. The geographer must take these intangible aspects of marketing into account, but the main focus for the geographer is on the locati onal attributes of a retail outlet that determine and influence store patronage. Location is crucial to the overall success of the retail outlet and can provide that necessary competitive advantage, but reta il demand consisting of the consumers demand for goods require a careful consideration of the positive and negative

PAGE 22

22 externalities associated with doing business. This includes factors not normally taken into account regarding the customer such as population distributi on, household income, race and ethnic characteristics, personal pr eferences and desires, purchasing habits, etc. The successful marketing strategy atte mpts to describe how the individual retail outlet performance rates in relation to t he potential cannibaliz ation and symbiotic aspects of competition, the impact of the price of goods and quality of services provided, as well as the influenc e of suitable location policy. Retail performance is significantly shaped by the size and demographic composition of an outlets trade area; therefore, trade area is a spatial expression of the limits to the firm's market potential (Ghosh and McLafferty, 1987). The marketing geographer must use this information in completi ng an analysis as it is required to gain a fully encompassing view of an inherently un even spatial distribution of this retail demand within a defined trade area in order to maximize the potential for future sales and growth. Tool and Methodologies Marketing Geography has a body of est ablished methodologies and techniques, and some terminology that might be unique to the field. Therefor e, in this section of the thesis, a brief summary of terminologies used in Marketing Geography are presented. Then, several methods illustrative of Marketing Geography, and used in this thesis, are summarized. Terminology This section summari zes the various terms and concepts used by Marketing Geographers and its associated literature.

PAGE 23

23 MARKET SHARE is the ratio of a stores sales in a geographic region to the total sales potential or otherwise defined as per capita sales divided by per capita sales potential. Market share is also known as market penetration. TRADE AREA of a retail market/service provider outlet is the spatia l expression of its market potential. It is t he geographic area from which its customers are normally drawn with an established benchmark of 80% c apture of customers by location. It is usually a function of distance, locati on of competitors, accessibility, and size (square footage) of the retail establishment. TRADE AREA DELINEATION is the process through which the trade are of a retail market/service provider trade area of a st ore is spatially demarcated or the area on a map where the majority of customers are drawn. DIRECTIONAL DISTRIBUTION (STANDARD DEVIATIONAL ELLIPSE) measures whether a distribution of features ex hibits a directional trend (w hether features are farther from a specified point in one dire ction than in another direction). STANDARD DISTANCE measures the degree to which features are concentrated or dispersed around the geometri c mean or median center. SYMMETRICAL DIFFERENCE is a visual depiction of the area not covered when two dissimilar polygons are over laid on top of one another. MEAN CENTER identifies the geographic center (or t he center of concentration) for a set of features. CENTROID is the geometric center of ma ss for the polygon being studied. Methods Customer spotting method With the intention that Marketing G eography become academically accepted, and compete on a equal footing with other subfields of geography as well as economics, Applebaum and Cohen (1960) promot ed the use of repeatable analytical methods. Input to their methodological approache s were measurements of tra de area, site accessibility, characteristics of the population within a tr ade area, competitive supply, economic base and stability of that economic base, tr ade area penetration, store size and store function, building costs, and oper ating costs. The outcome was a significant rise in the

PAGE 24

24 academic stature of Marketing Geography a rise in the demand for Marketing Geography graduates in geography, allied academic fields, and practice. While William Applebaum wrote extensiv ely, his publications were mainly targeted to practitioners. His publications in the trade press served to build his stature among practitioners, served as an outlet for his highly creative mind, and served to increase demand for his consultant work Applebaum's publications in academic journals which served to establish Marketi ng Geography as an academic discipline were published 1960s. This was after Applebaum left Harvard, and after Applebaum had worked for several decades in private cons ulting practice, learning and proving his discipline first hand. Appl ebaum's publications theref ore can be thought of as a compilation of a consultant's "secret" method s, proven by application, and revealed by paying clients to be highly valuable.8 Applebaum's work was a supreme exampl e of the geographer at work in the field. Among methods he innovated and put into practice were: customer spotting, trade area delineation, site evaluation, and market penetration. His primary focus was on the empirical study of store trade areas and on the market share c aptured from the trade area. Among Applebaum's most important and steady cli ents were Kroger Company (grocery stores) and Stop & Shop (small quick serve gasoline stati on oriented grocery stores), among other grocery store chai ns in the USA Northeast and Midwest. One of the methods Appl ebaum introduced in his own work in the 1930s, and subsequently published in trade and academic outlets, was his customer spotting 8 The late Reg Golledge in a personal communicati on to Grant Thrall and subsequently communicated to Ron Dietz, stated that one of Saul Cohen's great accomplishments was to convince William Applebaum to publish his methods. Applebaum's response to "snubb ing" by mainstream geography academics was to disregard academic geographers. Applebaum's st ory appears to be itself an analogue in academic geography toward its research publications whose focus is on improving business decisions.

PAGE 25

25 method. Prior to the 1930s, t he trade areas of grocery stores had been limited largely by the distance people were willing to walk. Howe ver, as automobiles became affordable to middle class families, trade areas became more complex. Trade areas could be larger because of the automobile, and grocery stores needed parking lots. Applebaum "spotted" automobile license plates in a gr ocery store's parking lot, then used that information to retrieve the registered addr ess of the automobile from the State's Department of Transportation. Applebaum wa s also a pioneer early in the automobile era on the in-store survey techniques by placing pins on a wall map indicating the location customers' home addresses (namely their origin), as well as a pin for the location of the grocery stor e (the destination). Hence, or igin-destination models in Marketing Geography were introduced. Today, of course, address match geocoding is used in GIS software, instead of pins on a map. Through trial and error, Applebaum concluded that a string encompassing 80% of the pins outlining a compact shape was a useful estimate of a trade area. He reasoned that 20% of the cust omers "spotted" were spurious. Today we still separate the core from spurious customers, while other criteria are used for enveloping the boundary of a trade area (Patel, Fik, and Thrall, 2008). Applebaum's customer spotting procedure has subsequently served as the starting point for trade area calculations; indeed, the entire location based intelligence literature owes a debt to Applebaum. A pplebaum's work penetrated beyond the dense walls of the academy, and showed the practi tioner the importance of geography to their business decisions, and provided a method for answering the question, where should limited resources be allocated to advertising and location development. If a location is

PAGE 26

26 developed, what is the im pact on other stores under t he same ownership; namely, cannibalization? Customer spotting is also important to location based service providers in that with the data, the friction of space between the customer and the location can be calculated.9 This is also known as the dist ance decay of the customer's demand schedule. Each item in a stor e, and collection of all items within the store, creates or overcomes a friction-of-distanc e. The choice of which goods are to be offered in the store therefore affects t he trade area. Cannibalization between stores can change because of store offerings. While in the short run, location for most brick-and-mortar cannot easily change, knowing the relati onship between friction-of-distance and stocking of goods can be useful in daily real -time business strategi c decisions, including advertising, marketing, goods stocked, and level of service. Geospatial technology including GIS software and online geospatial databases have improved the productivity of those performing Mark eting Geography, and thereby lowered its cost and increased its accessibilit y to smaller business firms. Sources of geospatial databases used by Marketing G eographers may be comprised of bank data (ATM records, check records, etc.), credit ca rd data, or point of sale data collected by the retail market/service provider, but each record has an attribute that can be used to locate the activity on the landscape. Retail firms are in the business of making money and as they operate within t he competitive business environment they are very protective of the information collected and te chniques used, so they ensure that their data is likewise protected. Al though the means for customer spotting/plotting are greater 9 Customer spotting methods have been used by Ghosh and McLafferty (1987); Moloney (1989); and Rogers and Green (1978), among others.

PAGE 27

27 than they were in the past, some obstacles still exist such as the cost of obtaining proprietary data monetarily or by entering into the contractual obligations and protections afforded by a Non-Disclosure Agreement. Trade area delineation Using the information collected by customer spotting survey methods or point of sale data, trade market areas can be accu rately delineated and refined as primary, secondary, or tertiary trade areas as in troduced by Applebaum (1966). He has defined the primary trade area as the geographic spatial core from which a store draws most of its customer business and using his analysis of grocery store chains, Applebaum (1966) found that this primary tr ade area encompassed about a 60 to 70% capture of store customers. Applebaum found the secondary trade area to be comprised of the next highest ratio of customers to population capture and drawing upon an additional 15, 20, or 25% of overall sales. The tertiary trade area can be defined as that area from which the residual, transitory, or spurious customers are drawn and usually exist on the outer fringe of a trade area, often c onsidered out of town sales. The effects of agglomeration cannot be understated as retail stores can form symbiotic relationships that serve to expand and increase the size of the market trade area and subsequent market penetration. Applebaum (1965b) notes this effect in a case study comparison of a disc ount food supermarket and a general discount merchandise store located adjacent to each other with a stand-alone di scount food supermarket and no location association. The supermarket located adjacent to the merchandise store formed a symbiotic relationship that st rengthened the trade area for both businesses that established more drawing power t han the unassociated, stand-alone supermarket.

PAGE 28

28 Summary There are many ways, methods, and techniques available to the analyst to calculate the value of location to a particula r business, but it is up to the individual analyst to apply those techniques that most effectively serve to enhance the business decision. The models presented are only a small selection of the entire site assessment/location selection measures commonly used and are significant because they allow the analyst the means to objectively evaluate an existing site or a prospective new site for a retail outlet. This objectivity facilitates solid analysis, improved judgment, and timely recommendations that compliment the business decision allowing retail firms the ability to methodically and systematically apply the results of these models with confidence. For instance, knowing the cu stomer origins and subsequent trade market areas for a particular location may identify the presence of market overlap or market gaps that indicate improper utilization of marketing practices prompting changes to location configuration, marketing and adver tising measures, or operating techniques.

PAGE 29

29 CHAPTER 3 TRADE AREA CALCULATION: AN INTEGRATED GIS APPROACH GIS as a Marketing Geography Research Tool Historical Overview GIS (geographic information systems) is a combination of computer hardware, software programs, analysis methods, and the proper application of these components together by competent and knowledgeable analysts.10 GIS allows the user to create, display, analyze, and manage spatially refer enced data consisting of a wide range of categories from customer location to supply nodes to lines of distribution to competitor locations to population demographic characteri stics and more. Researchers realized the potential value that comput er mapping software programs could provide to the Marketing Geographer in performing retail market analysis. Wolf (1969) describes the Synagraphic Mapping System (SYMAP), developed in 1964 by Howard T. Fisher at Northwestern University, as one of the first programs that allowed for the creation of a visual representation of spatially di stributed data on a digi tal map. SYMAP was revolutionary in that it allowed the user t he ability to analyze areally distributed data and create choropleth maps with a thematic f unction for shading polygons. With this technology, Marketing Geographers could in tegrate this early and first known GIS application software with the bus iness decision-making process to calculate areal sales volume potential, sales territo ry mapping, and determining in dustrial plant or warehouse locations (Wolf 1969). Businesses have been using GIS technol ogy since its inception on mainframe computers dating back to t he mid-1960s although published liter ature does not account 10 See Peuquet and Marble, (1990), editors, for se minal discussion of the meaning of GIS.

PAGE 30

30 for this use in applied research in the priv ate sector until the 1980s (King, 1993). This could be due to the fact that GIS was not widely available, the technology was cost prohibitive to operate, the functions prov ided limited and time-consuming application, and that businesses utilizing the technology were not willing to broadcast the existence and use of technologies that provided a competitive advantage over rivals. Emerging standards for the transfer of spatial data and for the specification of spatial process models will provide greater inter-operability between components developed at different times, for different purposes, and by diffe rent people using different hardware and software (Heikkila, 1998). Corporations are very protective of closely guarded proprietary data, business decision-making pr ocesses, techniques, and methodologies to the point that in comparison with industria l firms, retailers tend to be oversensitive to confidentiality with many fa cts of operation being needl essly suppressed (Davies, 1977). There are many open source websites where useable data can be found and downloaded, but the fact rema ins that the Marketing Geogr apher often has to deal with the costs associated with the acquisition, pr otection, and use of proprietary business datasets. This is detrimental to the analyst and impedes the improvement of findings and conclusions where transparency could impr ove the application of techniques and methodologies with the open sharing of i deas and information. While Marketing Geography literature is vast and has been improved with expansion through recent years, its structure resembles an iceber g 90% submerged. Geographers engaged in intelligence work are not the only ones with disclosure problems (Epstein, 1978). It seems that human nature and the competitive advantage within th is field are overriding factors contributing to the histor ically slow emergence of GIS, as the first study explicitly

PAGE 31

31 incorporating the use of this technology, as noted by Ghosh and Craig (1986), did not emerge in published liter ature until the 1980s. Customer Spotting Method The computing power, size of memory st orage files, and advances in GIS programs available today have enhanced the applicability of the methods intr oduced by William Applebaum. His work with customer spotti ng and the means through which customer addresses are plotted on a map, the bas is for trade area delineation, have been improved with the efficiency of GIS in terms of time saved by the analyst. GIS facilitates the use of raw data collected and allows for the display of spatial data visually, especially with the use of G eocoding functions. Geocoding is the procedure through which GIS applies coordinate data in the form of latitudes and longitudes to individual records based on a field in the data consisti ng most commonly of addr ess, zip code, or zip code + 4 location information. When an individual record has coordinate data, GIS is able to position the object on the correct lo cation on the planet (Thrall, del Valle, and Thrall, 1995). GIS software geocodes data based on the existence of two attribute files that are joined together to present the data spatially. The firs t attribute file contains the individual record of transactions for each customer and the second attribute file must contain the coordinate data used to pr esent the data spatially on a map. The U.S. Census has created a set of data, collectively referred to as Topologically Integrated Geographic Enco ding and Referencing (TIGER) Line files. TIGER files can be accessed via the Inte rnet and freely downloaded from the US Census www.census.gov web site. While not per fect, the intent is that the TIGER files include the streets in the U.S., their names type of street, and addresses, all within a standardized digital format that can be readily integrated wit h other databases within a

PAGE 32

32 GIS environment. Databases of addresse s then can be matched with the TIGER line files in a manner similar to relational dat abase management (RDM ). Since the TIGER line latitude-longitude coordi nates are known, and the address range for each line segment is known, then RDM can be used with spatial interpolation to estimate the geographic coordinate of the addr ess within the external database. This point is then displayed on the map in a position corres ponding to the street address number, on the correct side of the street, and set back 50 f eet from the street segment line. This procedure is repeated by the GIS software for each record in the database. Such geocoding allows spatial visu alization of external database addresses within the GIS. The external databases can be customer addr esses, store locations, event locations such as the occurrence of a crime, and so on. There is a match accuracy and error associated with geocoding because TIGER Line files do not cover all of the streets in the U.S., especially in the rural areas, and because the quality of this data is only as good as the technician that digitized the line and entered the street address values. Geocodi ng to a zipcode or zip code + 4 polygon has less error because the match is based on a larger polygon area than the street address which is represented by a single po int along a line. Thrall and Thrall (1994) have documented that street le vel geocoding has a lower hit rate than geocoding to the zip code + 4 level. This is due to the arr angement of zip code and zip + 4 areas, which are depicted with polygons used to geocode ma tch records located within the polygon boundaries and are assigned to the centroid of the individual polygon. The analyst must determine the level of precision required by the analysis and whether or not the level of

PAGE 33

33 error introduced by TIGER Line files is an a cceptable sacrifice to geocode at the street level and a specific locational point. The use of GIS facilitates the process of plotting customer location in relation to the retail outlet store and the ca lculation of trade areas. This is done by determining the distance of each customer location from t he retail outlet store and attaching that distance to the customer record data file. The analyst can then query these distances to find the primary, secondary, or tertiary trade areas by fi nding those records that are closest to the store and comprise 70% of store sales. Demographic Application Marketing Geographers utilize Geodemogr aphics to gain a better understanding of the customer base being served by the re tail market/service provider outlet store.11 ESRI's Business Analyst add-in to ArcM ap includes "Community Coder" geocoding application. Community Coder was formerl y known as CACI/Coder Plus. Community Coder appends demographic information to t he data record based upon its address. Community Coder has 60 demographic attribut e traits utilized by the U.S. Census Bureau specifically focusing on population and household characteristics to create customer cluster profiles encompassing ov er 200,000 neighborhoods. The assignment of a demographic profile by Community Coder divides and separates the customer dataset into 12 Life Mode groups and 65 Resi dential Segments each individually known 11 Goss, J.D. 1995a, Marketing the new marketing: the "strategic" discours e of advertising for Geodemographic Information Systems. In Pickles, J. (ed) Representations in an electronic age: geography, GIS and democracy New York: Guildford Press. And, Goss, J.D. 1995b, "We know where you are and we know where you live ": the instrumental rationality of geodemographic information systems. Economic Geography 71,2: 171-198

PAGE 34

34 as a Tapestry (Appendix A). These Tapestr ies have also formerly been known as A Classification of Resident ial Neighborhoods (ACORNs) or Lifestyle Segmentation Profiles (LSPs). LSPs provide a composit e measurement that summarizes population characteristics by location (Thrall and Me coli, 2003). ESRI BIS Coder Business Analyst uses a clustering of customers, based on the consumption hist ory and anticipated spending profile of the area wher e they live, in a process of assigning a Tapestry that groups these consumers at the zip code, zip code + 4, or street level. The basic principle that underlies the creat ion and use of an LSP is Toblers first law of geography: all things are related, but near things are more related than distant things (see Goss, 1995; Thrall and Mecoli, 2003). These profiles, or Tapestries, are assigned to residential neighborhoods to reveal the spending habits of an area in the theory that people with similar interests, similar spending habits, and making similar choices will choose to reside within close proximit y of each other spatially. Summary Geospatial technology has increas ed the productivity and precision of analysis created by Marketing Geogr aphers (see Pickles, 1995). The geospatial technological innovations include remote GIS, s patial data transfer standards, open GIS specifications, digital spatia l data libraries, object-ori ented GIS, and network-resident programming such as JAVA (Heikkila, 1998). GI S has proven to be a valuable tool that enhances the business decision-making process through timely and accurate spatial analysis that offers the retail market/servi ce provider a compet itive advantage that competitors may or may not be using.

PAGE 35

35 CHAPTER 4 TRADE AREA CALCULATION: METHODOLOGY This section presents a me thodological algor ithm that can be followed by a retail market/service provider to calculate a tr ade area. The methodology integrates concepts from retail location analysis, geospatial statistics, and GIS technology. The algorithm presented here would not have been practical before the creation of contemporary geospatial technology and powerfu l computers. This methodol ogy has its foundations in the retail location assessment literature summarized in Chapt ers 2 and 3. The algorithm presented here is demonstrated using actual data collected from a single outlet retail service provider in Alachua County, Florida. The first objective of this study is to present the algorithm as in step-by-step fashion. The second objective of this section is to provide a descriptive overview to increase the understanding of the procedure so that so that it can be repeated by other analysts. The benefits of the algorithm arise because the success of a retail service provider depends upon location; the algorithm offers the ent repreneur to better evaluate the changing market landscape, and thereby realize a competitive advantage. The retail business that provided the data for this study requi res anonymity. The retail business will therefore be referred to as the "Client." The "Client" provided customer data for two time periods, 2004 and 2008. Both databases included customer address, transaction date, and amount paid fo r the transaction. The datasets were received in comma-delimited format (CSV). Data entry errors were corrected or eliminated. The CSV file was exported as an Excel XLS file. ESRI Community Coder can read XLS data files. The XLS file is processed by Community Coder and then imported as a data layer into the GIS env ironment. A second dataset was created by

PAGE 36

36 joining the 2004 and 2008 datasets that is co mprised of the customers with accounts showing up in both the 2004 and 2008 datas ets. The second database was also geocoded. Therefore, th ree GIS data layers (shapefiles) were used: customers in 2004 customers in 2008 customers in both 2004 and 2008 Geographically spurious customers were e liminated by including only customers with home addresses within Alachua County, Florida. Geocoding had revealed customers with addresses in Iowa and Minnesota, as well as other countries. The results of each dataset selection refinement are as follows: 2004 Alachua Clients (1423) / 2004 All Clients (1745) = 81.55% 2008 Alachua Clients (1459) / 2008 All Clients (1761) = 82.85% 2004+2008 Alachua Clients (645) / 2004+200 8 All Clients (664) = 97.14% The presence of selected customers, as a percentage in relation to all customers, represent a ratio for each year group dat aset that are all above 80% and can be considered a statistically significant study group. Using the ESRI ArcMap spatial statisti cs function, the Centroid, Directional Distribution (Standard Deviat ion Ellipse), and Standard Dist ance for each year group dataset was calculated with the process s hown graphically in the flow chart below (Figure 4-1). It should be not ed that the Directional Distri bution (Standard Deviation Ellipse) and Standard Distance shown below were calculated using the 1st Standard Deviation and Euclidean Distance centered around the Cent roid calculated for the individual year group dataset being used. T he figures below are organized by dataset year group used and, as much as possible, are color coded representations with 2004 customers depicted in red (Figures 4-2 and 4-3), 2008 cu stomers depicted in blue

PAGE 37

37 (Figures 4-4 and 4-5), and 2004+2008 customer s depicted in green (Figures 4-6 and 47). Overlaying the 2004, 2008, and 2004+2008 shapef ile functions by year group, the growth and decay areas are further revealed (Figures 4-8 and 4-10). Additionally, the Symmetrical Difference between the Direc tional Distribution (Standard Deviation Ellipse) and the Standard Distance was calculated using a comparison of the 2004 and 2008 datasets (Figures 4-9 and 4-11). The area shown represent s the customer transition area or change in time for the respective functions. By using the spatial statistics functions provided by ArcMap, and visually shown on the Figures (4-2 through 4-11) the following customer capt ure statistics are shown in the list below. Customer Captur e Statistics. Clients Customer List 2004 Directional Distribution (Standard De viational Ellipse) 1074/1423 = 75.47% capture of 2004 customers. Standard Distance 1060/1423 = 74.49% capture of 2004 customers. Clients Customer List 2008 Directional Distribution (Standard De viational Ellipse) 1090/1459 = 74.71% capture of 2008 customers. Standard Distance 1080/1459 = 74.02% capture of 2008 customers. Clients Customer List 2004+2008 Directional Distribution (Standard Deviat ional Ellipse) 489/645 = 75.81% capture of 2004+2008 customers. Standard Distance 486/645 = 75.35% capture of 2004+2008 customers. The measurements shown above indicate a very stable and solid market penetration and customer captur e. There appears to be little change or variation within the three different datasets throughout the encompassing five year period being analyzed.

PAGE 38

38 The Core Trade Area Radial Method In applying the core trade area radial me thod, the centroi d of the spatial distribution of the data was used as the starting point. From the centroid, radial distances of 1.5 and 1.0 miles were used to create a circle polygon and calculate the number of customers captur ed within these distances. These distances were not chosen arbitrarily, and for the size of the study area were appropriate using conventional marketing geography practices. Publix supermarket corporate headquarters utilize the competitive edge benefit of a Marketing Geography division that emplo ys analysts to perform the techniques and methodologies associated with retail market/s ervice provider location and trade area calculation.12 The location, by address, of each Publix store was found using BellSouths The Real Yellow Pages Gainesv ille, Florida phonebook for 2008, obtaining the addresses of Publix locations, geocodi ng those addresses, and plotting each location on the map using GIS. By reverse en gineering the known locations of Publix supermarkets for the Gainesville, Florida area, the established trade area can be roughly estimated. This is done by buffering each Publix supermarket location until the buffered areas become tangent to each other where possible and, in contiguous areas, where market overlap is minimized while maximizing the market penetration for each site (Figure 4-12). The estimated radius of 1.5 miles revealed the target core trade area for existing Publix supe rmarket locations and will be used as a benchmark representative trade area. 12 Communication to Ron Dietz by Professor Grant Thrall. Publix headquarters has employed students from University of Florida that have completed t he Business Geography Program under the direction of Professor Thrall. Professor Thrall has also presented invited Friday Afternoon Seminars to senior Publix management at their headquarters in Lakeland Florida.

PAGE 39

39 Customer spotting and the benchmark trade area distance of 1.5 mile radius are applied to the "Client's" database. The num ber and percent of "Client" customers located within a 1.5 mile radius from the ce ntroid are thereby determined. The results are shown in Figures 4-13 to 4-15, and in the list below. Market Penetration Within The Core Trade Area. 2004 Alachua County, FL Custom ers 511/1423 = 35.91% (Figure 4-13). 2008 Alachua County, FL Custom ers 504/1459 = 34.54% (Figure 4-14). 2004+2008 Alachua County, FL Custom ers 216/645 = 33.49% (Figure 4-15). (Based upon the number of cust omers within 1.5 miles from t he client's location, divided by the total number of client's customers within Alachua County, Florida.) This list reveals that the customer captur e rate is constant, with the difference in percentage within the core trade area from 2004 to 2008 changing by only 1.37%. During the five year period covered by this study, there were eight new retail market/service provider outlet stores that appeared on the landscape. This was determined by using BellSouths The Real Yellow Pages Gainesville, Florida phonebook from 2004 (32 competitor listings) and 2008 (43 competitor listings), obtaining the addresses of competitor locations, creating a spreadsheet of these locations, geocoding those addresses, and plotting each location on the map using GIS. The appearance of these eight new competitor locations are impor tant because of the uncertainty that they introduce to the existing business equilibrium and the potentia l impact on current customer strength through canni balization and market penetration. Figures 4-16 to 4-24 show the number and percent of the Clients customer da taset captured within a 1.5 mile radius from all eight new competitors (Figure 416) and each new competitor location (Figure 4-17 to 4-24). In no par ticular order, new com petitor 1 through 8 potential impact by customer captur e revealed in the following list:

PAGE 40

40 New Competitor Location Analysis. All eight new competitor locations appearing from 2004 to 2008, Alachua County, FL Customer capture 326/ 1459 = 22.34% (Figure 4-16). New competitor location number 1 appearing from 2004 to 2008, Alachua County, FL No Customer capt ure 0/1459 = 0% (Figure 4-17). New competitor location number 2 appearing from 2004 to 2008, Alachua County, FL Customer c apture 13/1459 = 0.89% (Shar ed buffer trade area with new competitor 3) (Figure 4-18). New competitor location number 3 appearing from 2004 to 2008, Alachua County, FL Customer c apture 14/1459 = 0.96% (Shar ed buffer trade area with new competitor 2) (Figure 4-19). New competitor location number 4 appearing from 2004 to 2008, Alachua County, FL Customer c apture 68/1459 = 4.66% (Shar ed buffer trade area with new competitor 5) (Figure 4-20). New competitor location number 5 appearing from 2004 to 2008, Alachua County, FL Customer c apture 150/1459 = 10.28% (Shar ed buffer trade area with new competitor 4) (Figure 4-21). New competitor location number 6 appearing from 2004 to 2008, Alachua County, FL Customer captur e 87/1459 = 5.96% (Figure 4-22). New competitor location number 7 appearing from 2004 to 2008, Alachua County, FL Customer captur e 6/1459 = 0.41% (Figure 4-23). New competitor location number 8 appearing from 2004 to 2008, Alachua County, FL Customer captur e 39/1459 = 2.67% (Figure 4-24). Additionally, during the five year period covered by this study, there were four existing retail market/service provider outlet stores that moved, changing their physical location within the landscape. This was determined by using BellSouths The Real Yellow Pages Gainesville, Florida phonebook from 2004 and 2008, obtaining the addresses of competitor locations, creating a spreadsheet of these locations, geocoding those addresses, and plotting each location on the map using GIS. The movement of these four existing com petitor locations is important bec ause of the uncertainty that they

PAGE 41

41 introduce to the existing business equilibr ium and the potential impact on current customer strength through c annibalization and market penetration. These moving competitor locations are mo re important than the new outlets because they have a customer base and could increase the market penetration. The following list depicts the number and percent of the Clients customer dataset captured wit hin a 1.5 mile radius of each moving competitor location for 2004 and 2008 listed first through fourth: Existing Competitor Change in Location Analysis First moving competitor appearing from 2004 to 2008, (2004 Location) Alachua County, FL Customer capture 451/1423=31.69% of 2004 customers, 444/1459=30.43% of 2008 cust omers (Figure 4-25). First moving competitor appearing from 2004 to 2008, (2008 Location) Alachua County, FL Customer c aptured 69/1459 = 30.43% of 2 008 customers (Figure 426). Second moving competitor appearing fr om 2004 to 2008, (2004 Location) Alachua County, FL Cust omer captured 67/1423 = 4.71% of 2004 customers and 65/1459 = 4.46% of 2008 customers (Figure 4-27). Second moving competitor appearing fr om 2004 to 2008, (2008 Location) Alachua County, FL Cu stomer captured 63/1459 = 4.32% of 2008 customers (Figure 4-28). Third moving competitor appearing from 2004 to 2008, (2004 Location) Alachua County, FL Customer captured 345/1423=24.24% of 2004 customers, 367/1459=25.15% of 2008 cust omers (Figure 4-29). Third moving competitor appearing from 2004 to 2008, (2008 Location) Alachua County, FL Customer c aptured 346/1459 = 23.71% of 2008 customers (Figure 430). Fourth moving competitor appearing from 2004 to 2008, (2004 Location) Alachua County, FL Customer captured 67/1423=4.71% of 2004 customers, 65/1459=4.46% of 2008 customers (Figure 4-31). Fourth moving competitor appearing from 2004 to 2008, (2008 Location) Alachua County, FL Customer c aptured 477/1459 = 32.69% of 2008 customers (Figure 432).

PAGE 42

42 The results of an analysis of the competitors moving physical outlet store location from 2004 to 2008 revealed that the First com petitor moved to a more distant location and lost market penetration influence on the Clients customer base. The Second and Third competitors had very little location change and, therefore, ha d very little change in market penetration influence on the Clients cu stomer base, remaining at consistent potential capture levels. The F ourth competitor moved closer to the Clients location from 2004 to 2008, significantly increasing potential market penetration influence and could be a future threat to potential customer capture levels. The Grid Method The next method of analysis was to apply a 1 x 1 and then a 1.5 x 1.5 mile grid layer that covered Alachua Count y. The two di fferent sized grid cells are used to detect existence of MAUP. The respective data layer was then was then combined with the 2004 and 2008 customer datasets using the overlay function. The cells with customer data were selected and export ed as an individual layer to be used in the analysis and creation of the choropl eth maps (Figures 4-33 to 4-47) The choropleth maps were created using the Jenks Method consisting of natural breaks and classified into five distinct classes. Each breaking point thre shold by class was then rounded to a nearest whole number and applied to other similar ca tegory choropleth maps showing different year group data results.

PAGE 43

43 Figure 4-1. Data manipulat ion and methods flow chart.

PAGE 44

44 Figure 4-2. Alachua county customers 2004 directional distribution (standard deviational ellipse).

PAGE 45

45 Figure 4-3. Alachua county cu stomers 2004 standar d distance.

PAGE 46

46 Figure 4-4. Alachua county customers 2008 directional distribution (standard deviational ellipse).

PAGE 47

47 Figure 4-5. Alachua county cu stomers 2008 standar d distance.

PAGE 48

48 Figure 4-6. Alachua county customers 2004+ 2008 directional distribution (standard deviational ellipse).

PAGE 49

49 Figure 4-7. Alachua county cu stomers 2004+2008 standard distance.

PAGE 50

50 Figure 4-8. Alachua county customer growth and decay zones overlay of the directional distribution (standard deviati onal ellipse) polygons by dataset year group.

PAGE 51

51 Figure 4-9. 2004 and 2008 Customers in alachua county, fl. directional distribution (standard deviation ellipse), symmetrical difference overlay.

PAGE 52

52 Figure 4-10. Alachua county cu stomer growth and decay zones overlay of the standard distance polygons by dataset year group.

PAGE 53

53 Figure 4-11. 2004 and 2008 Customers in al achua county, fl. standard distance, symmetrical difference overlay.

PAGE 54

54 Figure 4-12. 2008 Publix supermarket locations (1.5 mile buffer).

PAGE 55

55 Figure 4-13. 2004 Alachua county customer s within 1.5 miles of the centroid. Figure 4-14. 2008 Alachua county customer s within 1.5 miles of the centroid.

PAGE 56

56 Figure 4-15. 2004+2008 Alachua county custom ers within 1.5 miles of the centroid. Figure 4-16. 2004+2008 All eight new competitor locations appearing from 2004.

PAGE 57

57 Figure 4-17. New competitor locati on number 1 appearing from 2004 to 2008. Figure 4-18. New competitor locati on number 2 appearing from 2004 to 2008.

PAGE 58

58 Figure 4-19. New competitor locati on number 3 appearing from 2004 to 2008. Figure 4-20. New competitor locati on number 4 appearing from 2004 to 2008.

PAGE 59

59 Figure 4-21. New competitor locati on number 5 appearing from 2004 to 2008. Figure 4-22. New competitor locati on number 6 appearing from 2004 to 2008.

PAGE 60

60 Figure 4-23. New competitor locati on number 7 appearing from 2004 to 2008. Figure 4-24. New competitor locati on number 8 appearing from 2004 to 2008.

PAGE 61

61 Figure 4-25. First moving competitor appear ing from 2004 to 2008, (2004 location). Figure 4-26. First moving competitor appear ing from 2004 to 2008, (2008 location).

PAGE 62

62 Figure 4-27. Second moving competitor appearing from 2004 to 2008, (2004 location). Figure 4-28. Second moving competitor appearing from 2004 to 2008, (2008 location).

PAGE 63

63 Figure 4-29. Third moving competitor appearing from 2004 to 2008, (2004 location). Figure 4-30. Third moving competitor appearing from 2004 to 2008, (2008 location).

PAGE 64

64 Figure 4-31. Fourth moving competitor appearing from 2004 to 2008, (2004 location). Figure 4-32. Fourth moving competitor appearing from 2004 to 2008, (2008 location).

PAGE 65

65 Figure 4-33. 2004 Total revenue (do llars), 1x1 mile grid cell.

PAGE 66

66 Figure 4-34. 2008 Total revenue (do llars), 1x1 mile grid cell.

PAGE 67

67 Figure 4-35. 2004 Average revenue per transaction (dollars), 1x1 mile grid cell.

PAGE 68

68 Figure 4-36. 2008 Average revenue per transaction (dollars), 1x1mile grid cell.

PAGE 69

69 Figure 4-37. 2004 Number of transactions, 1x1 mile grid cell.

PAGE 70

70 Figure 4-38. 2008 Number of transactions, 1x1 mile grid cell.

PAGE 71

71 Figure 4-39. 2004 Number of customers, 1x1 mile grid cell.

PAGE 72

72 Figure 4-40. 2008 Number of customers, 1x1 mile grid cell.

PAGE 73

73 Figure 4-41. Total revenue change from 2004 to 2008, 1x1 mile grid cell

PAGE 74

74 Figure 4-42. Total revenue change from 2004 to 2008, 1.5x1.5 mile grid cell

PAGE 75

75 Figure 4-43. 1x1 Mile gr id cell coordinate plane. First Foci (Origin) Second Foci (1/4 Mile Offset) Third Foci (1/2 Mile Offset)Grid Overlay Offset Figure 4-44. 1x1 Mile grid cell overlay offset method.

PAGE 76

76 Figure 4-45. Total revenue change from 2004 to 2008, 1x1 m ile grid cell (first foci, origin).

PAGE 77

77 Figure 4-46. Total revenue change from 2004 to 2008, 1x1 mile grid cell (second foci, mile offset).

PAGE 78

78 Figure 4-47. Total revenue chan ge from 2004 to 2008, 1x1 mile grid cell (third foci, mile offset).

PAGE 79

79 CHAPTER 5 TRADE AREA ALGORITHM A layer of same-sized grid cells ar e used as a container for data within each cell's area, with the added benefit that geography is held constant. T he grid provides a static visual display of the dataset year group values applied across the landscape in an area that maintains a consistent level of geospatial parity. Placing dataset year groups in a grid and comparing the results of the mapped areas together reveal the dynamic geotemporal change in phenomena being studi ed. An example of a dynamic geotemporal application of unemployment rates by county throughout the nation, ( http://cohort11.americanobserver.net/l atoyaegwuek we/multimediafinal.html ), can be seen in the example created by Latoya Egwuek we. In Egwuekwe's geotemporal study, Bureau of Labor Statistics' monthly unempl oyment statistics fr om January 2007 to December 2009 are displayed by country. Wh ile Egwuekwe's geotemporal "movie" is visually appealing, the interpretation of the space-time trend is biased because geography is not held constant each county has a different size and shape. The larger the county, the greater is the inferred visual weight given to the value of unemployment. Since the unemployment value is calculated as a ratio and reported at the state or county level, it is easy to apply and show t hese values visually using these arbitrary legal boundaries that ar e not consistent and often defined by topography. Going a step further, by applying the unemployment values to the centroid of the state or county boundary polygon and then overlaying a grid the projection and areal weighting inference disparity would be eliminated by using standard area polygons that the grid presents. Application of the grid would also re veal clustering trends and trajectories that would otherwise remain transparent and undetected using dissimilar polygons.

PAGE 80

80 Keeping size and scale constant does not a lways ensure that statistical analysis of a spatial area will show results that are unbi ased. Different areal units can have large impact on the end result of aggregated values. The MAUP is highly sensitive to the direction of investigation (0 to 360 degrees ) progression on the landscape, the cell size, the offset distance, and the time frame being studied. Therefore, it is critically important to realize that the depiction of descrip tive statistics on a map may not be truly representative of the actual variation of a particular variable displayed across the map as shown. Application of the grid in conjunction with accepted methodology allows for the spatial variation to become consistent across the landscape. By using a grid size that is appropriate to the study area, the analyst is able to focus on the variable being studied. By keeping geography constant in this way, it allows for the application of other measures to identify significant occurrences of trends and phenomena. The calculated statistics by grid cell category can be found in Table 5-1 through Table 5-6, and are summarized by approxim ate calculated arithmetic mean in the following list: 1x1 Mile Grid, Total Revenue Change (Table 5-1), $203 per cell. 1x1 Mile Grid, Average Revenue Change (Table 5-2), $52 per cell. 1x1 Mile Grid, Customer Count Change (Table 5-3), 0.429 per cell. 1x1 Mile Grid, Transaction Change (Table 5-4), 0.273 per cell. 1.5x1.5 Mile Grid, Total Revenue Change (Table 5-5), 115 per cell. 1x1 Mile Grid, Total Revenue Cha nge (Table 5-6), Foci 1 $20 per cell. 1x1 Mile Grid, Total Revenue Cha nge (Table 5-6), Foci 2 $75 per cell. 1x1 Mile Grid, Total Revenue Cha nge (Table 5-6), Foci 3 $130 per cell. Figures 4-41 and 4-42 show how the Total Revenue values change on a choropleth map when applied acro ss a 1 x 1 mile grid cell ov erlay and a 1.5 x 1.5 mile grid cell overlay respectively. The values are clearly dependent upon grid cell size as

PAGE 81

81 their importance becomes more diffused (Fi gure 4-41) or less diffused (Figure 4-42) with areal change. This can be problematic as the proper areal size adjustment of a grid cell can serve to either highlight or bypass not eworthy clusters as shown in different grid cell overlay sizes. Figure 4-43 is a depiction of the coordinate plane where the X and Y axes emanate from an origin dividing the grid pl ane into four quadrants. Figure 4-44 shows how the grid overlay was shifted from the origin into the first quadrant using a direction of 45 degrees and a distance of and mile. Figure 4-45 is a choropleth map showing total revenue values across a 1 x 1 mile grid overlay, with the First foci lower left grid cell co-located at the origin. Figure 4-46 is a choropleth map showing Total Revenue values across a 1 x 1 mile grid overlay, with the Second foci lower left grid cell located a mile distance and 45 degrees from the or igin. Figure 4-47 is a choropleth map showing total revenue values across a 1 x 1 mile grid overlay, with the Third foci lower left grid cell located a mile distance and 45 degrees from the origin. The choropleth map clearly shows how small changes in di stance and direction impact the values represented by the choropleth m ap. The Second foci (Figure 4-46) has a similar spatial distribution to the First foci (F igure 4-45), but the Third foci (Figure 4-47) has a distinctly different appearance with cluste red neighborhoods dominat ed by the middle range of values. The Hot Spot Analysis tool ca lculates the Getis-Ord Gi* st atistic for each feature in a weighted set of features. T he Gi* statistic tells you whet her features with high values or features with low values tend to cluster in a study area with the output being a Z score for each feature which represents the statistical significance of clustering for a

PAGE 82

82 specified distance. This method works by looking at each feature within the context of neighboring features. If a feature's value is high, and the values for all of its neighboring features is also high, it is a part of a hot spot. The higher (or lower) the Z score, the stronger the association. For statistically si gnificant positive Z scores, the larger the Z score, the more intense the clustering of hi gh values. For statistically significant negative Z scores, the smaller the Z score, the more intense the clustering of low values. A Z score near zero indicates no apparent concentration (neighbors have a range of values). The local sum for a feature and its neighbors is compared proportionally to the sum of all features; w hen the local sum is much different than the expected local sum, and that difference is too large to be the result of random chance, a statistically significant Z score is the result. The use of this statistical method reveals the presence of neighborhoods that are satura ted and others that could be mined using more robust marketing applications. The Hot Spot Analysis (Getis-Ord Gi*) c horopleth map values were created using a fixed distance band and a Euclidean distance. The Local G statistics are used to test for spatial clustering in group-level data making it possible to assess the spatial association of a variable within a particular distance of each observation. The spatial clusters show areas with high and low attribute values graphically within the choropleth overlay layer. The 1 x 1 mile grid cell over lay (Figure 5-1) of the Hot Spot Analysis choropleth map clearly shows a more diverse s patial distribution than the 1.5 x 1.5 mile grid cell overlay (Figure 5-2) which appears to become more spatially diffused with small area clusters disappearing and values approaching equivalence.

PAGE 83

83 The Hot Spot Analysis of the 1 x 1 grid cell overlay choropleth maps shows a distinct cluster of cells, with values less than -2.58 standard deviations, appearing in the Second foci (Figure 5-4) spatial distributi on that can be described as a submarket. In the application of dynamic gridding show n here, there was still an instance where a submarket was revealed through this proce ss and identifies the existence of tapestry clusters (Appendix A). There are a total of 26 cells in this category submarket with the majority of cells labeled by the College Towns tapestry (7 cells), Prosperous Empty Nesters (5 cells), Enterprising Professional s (2 cells), and In Style (2 cells). The remaining 10 cells from this submarket each garnered enough weight to warrant an individual tapestry label not mentioned above. This submar ket cluster did not appear in either the First foci (Figure 5-3) or the Third foci (Figure 5-5) spatial distribution choropleth maps, whose values appeared to be much more closely distributed to each other spatially.

PAGE 84

84 Table 5-1. Total revenue statistics (1x1 mile grid cell overlay). 1x1 Mile Grid Total Revenue 2004 Total Revenue 2008 Total Revenue Change N of Cases 161 161 161 Minimum 0 0 -3,074.11 Maximum 3,103.61 6,986.81 3,998.20 Range 3,103.61 6,986.81 7,072.31 Sum 71,921.23 104,639.82 32,718.59 Median 228.32 388.57 90.04 Arithmetic Mean 446.716 649.937 203.221 Standard Error of Arithmetic Mean 47.373 66.429 68.248 95.0% Lower Confidence Limit 353.158 518.746 68.439 95.0% Upper Confidence Limit 540.273 781.127 338.003 Trimmed Mean (10%, Two Sided) 325.096 484.932 148.415 No. of Observations Trimmed Out 34 34 34 Standard Deviation 601.101 842.888 865.964 Variance 361,321.97 710,459.51 749,894.46 Coefficient of Variation 1.346 1.297 4.261 Skewness(G1) 2.18 3.61 0.682 Standard Error of Skewness 0.191 0.191 0.191 Kurtosis(G2) 5.784 20.546 4.58 Standard Error of Kurtosis 0.38 0.38 0.38 Shapiro-Wilk Statistic 0.742 0.67 0.892 Shapiro-Wilk p-value 0 0 0 Anderson-Darling Statistic 11.541 12.81 5.657 Adjusted Anderson-Darling Statistic 11.595 12.87 5.684 p-value <0.01 <0.01 <0.01

PAGE 85

85 Table 5-2. Average revenue statistics (1x1 mile grid cell overlay). 1x1 Mile Grid Average Revenue 2004 Average Revenue 2008 Average Revenue Change N of Cases 161 161 161 Minimum 0 0 -243.14 Maximum 341.74 696.51 640.41 Range 341.74 696.51 883.55 Sum 10,900.25 19,237.22 8,336.98 Median 60.147 97.897 33.445 Arithmetic Mean 67.703 119.486 51.782 Standard Error of Arithmetic Mean 5.47 8.3 9.094 95.0% Lower Confidence Limit 56.9 103.095 33.823 95.0% Upper Confidence Limit 78.506 135.877 69.741 Trimmed Mean (10%, Two Sided) 57.094 102.995 39.413 No. of Observations Trimmed Out 34 34 34 Standard Deviation 69.408 105.312 115.385 Variance 4,817.44 11,090.71 13,313.67 Coefficient of Variation 1.025 0.881 2.228 Skewness(G1) 1.541 2.432 1.787 Standard Error of Skewness 0.191 0.191 0.191 Kurtosis(G2) 3.364 8.503 6.24 Standard Error of Kurtosis 0.38 0.38 0.38 Shapiro-Wilk Statistic 0.845 0.79 0.865 Shapiro-Wilk p-value 0 0 0 Anderson-Darling Stat istic 5.289 7.295 5.209 Adjusted Anderson-Darling Statistic 5.315 7.329 5.234 p-value <0.01 <0.01 <0.01

PAGE 86

86 Table 5-3. Customer count statistics (1x1 mile grid cell overlay). 1x1 Mile Grid Customer Count 2004 Customer Count 2008 Customer Count Change N of Cases 161 161 161 Minimum 0 1 -16 Maximum 109 115 14 Range 109 114 30 Sum 1,390.00 1,459.00 69 Median 3 3 0 Arithmetic Mean 8.634 9.062 0.429 Standard Error of Arithmetic Mean 1.329 1.347 0.274 95.0% Lower Confidence Limit 6.008 6.401 -0.112 95.0% Upper Confidence Limit 11.259 11.723 0.969 Trimmed Mean (10%, Two Sided) 4.472 4.591 0.331 No. of Observations Trimmed Out 34 34 34 Standard Deviation 16.866 17.097 3.473 Variance 284.459 292.296 12.059 Coefficient of Variation 1.954 1.887 8.103 Skewness(G1) 3.7 3.59 0.229 Standard Error of Skewness 0.191 0.191 0.191 Kurtosis(G2) 16.15 15.436 5.586 Standard Error of Kurtosis 0.38 0.38 0.38 Shapiro-Wilk Statistic 0.529 0.521 0.849 Shapiro-Wilk p-value 0 0 0 Anderson-Darling Stat istic 25.074 27.196 8.831 Adjusted Anderson-Darling Statistic 25.193 27.325 8.873 p-value <0.01 <0.01 <0.01

PAGE 87

87 Table 5-4. Transaction statistics (1x1 mile grid cell overlay). 1x1 Mile Grid Transactions 2004 Transaction 2008 Transaction Change N of Cases 161 161 161 Minimum 0 0 -33 Maximum 44 29 24 Range 44 29 57 Sum 847 891 44 Median 3 4 1 Arithmetic Mean 5.261 5.534 0.273 Standard Error of Arithmetic Mean 0.55 0.442 0.57 95.0% Lower Confidence Limit 4.175 4.662 -0.853 95.0% Upper Confidence Limit 6.347 6.406 1.4 Trimmed Mean (10%, Two Sided) 3.756 4.48 0.583 No. of Observations Trimmed Out 34 34 34 Standard Deviation 6.977 5.602 7.239 Variance 48.682 31.388 52.4 Coefficient of Variation 1.326 1.012 26.487 Skewness(G1) 2.182 1.997 -0.753 Standard Error of Skewness 0.191 0.191 0.191 Kurtosis(G2) 6.37 4.475 4.019 Standard Error of Kurtosis 0.38 0.38 0.38 Shapiro-Wilk Statistic 0.748 0.784 0.927 Shapiro-Wilk p-value 0 0 0 Anderson-Darling Stat istic 12.003 9.769 3.403 Adjusted Anderson-Darling Statistic 12.06 9.816 3.419 p-value <0.01 <0.01 <0.01

PAGE 88

88 Table 5-5. Total revenue statistics (1 .5x1.5 mile grid cell overlay). 1.5x1.5 Mile Grid Total Revenue 2004 Total Revenue 2008 Total Revenue Change N of Cases 128 128 128 Minimum 0 0 -3,074.11 Maximum 3,103.61 5,130.88 4,094.81 Range 3,103.61 5,130.88 7,168.92 Sum 53,694.99 68,438.11 14,743.12 Median 151.28 293.805 46.955 Arithmetic Mean 419.492 534.673 115.181 Standard Error of Arithmetic Mean 54.764 66.592 81.115 95.0% Lower Confidence Limit 311.124 402.899 -45.332 95.0% Upper Confidence Limit 527.86 666.447 275.693 Trimmed Mean (10%, Two Sided) 278.404 373.291 100.17 No. of Observations Trimmed Out 26 26 26 Standard Deviation 619.584 753.406 917.714 Variance 383,884.47 567,620.99 842,199.22 Coefficient of Variation 1.477 1.409 7.968 Skewness(G1) 2.304 2.764 0.42 Standard Error of Skewness 0.214 0.214 0.214 Kurtosis(G2) 5.756 11.093 4.148 Standard Error of Kurtosis 0.425 0.425 0.425 Shapiro-Wilk Statistic 0.699 0.703 0.915 Shapiro-Wilk p-value 0 0 0 Anderson-Darling Stat istic 12.224 10.618 3.659 Adjusted Anderson-Darling Statistic 12.297 10.681 3.68 p-value <0.01 <0.01 <0.01

PAGE 89

89 Table 5-6. Total rev. change stats (1x1 mile grid cell overlay offset foci 1,2, and 3). 1x1 Mile Grid Total Revenue Change Foci 1 Total Revenue Change Foci 2 Total Revenue Change Foci 3 N of Cases 193 194 183 Minimum -2,279.51 -7,333.06 -2,643.91 Maximum 3,029.86 4,159.85 4,584.89 Range 5,309.37 11,492.91 7,228.80 Sum 3,895.04 14,678.11 23,714.39 Median 10 28.42 0 Arithmetic Mean 20.182 75.66 129.587 Standard Error of Arithmetic Mean 55.214 82.327 69.213 95.0% Lower Confidence Limit -88.721 -86.716 -6.975 95.0% Upper Confidence Limit 129.085 238.037 266.149 Trimmed Mean (10%, Two Sided) 27.362 84.49 65.981 No. of Observations Trimmed Out 40 40 38 Standard Deviation 767.051 1,146.68 936.291 Variance 588,366.81 1,314,884.45 876,640.35 Coefficient of Variation 38.008 15.156 7.225 Skewness(G1) 0.197 -1.298 1.294 Standard Error of Skewness 0.175 0.175 0.18 Kurtosis(G2) 3.116 11.417 6.025 Standard Error of Kurtosis 0.348 0.347 0.357 Shapiro-Wilk Statistic 0.911 0.82 0.846 Shapiro-Wilk p-value 0 0 0 Anderson-Darling Stat istic 6.363 9.795 9.117 Adjusted AndersonDarling Statistic 6.388 9.834 9.155 p-value <0.01 <0.01 <0.01

PAGE 90

90 Figure 5-1. Total revenue change from 2004 to 2008, hot spot analysis (getis-ord gi*) 1x1 mile grid cell.

PAGE 91

91 Figure 5-2. Total revenue change from 2004 to 2008, hot spot analysis (getis-ord gi*) 1.5x1.5 mile grid cell

PAGE 92

92 Figure 5-3. Total revenue change from 2004 to 2008, hot spot analysis (getis-ord gi*) 1x1 mile grid cell (first foci, origin).

PAGE 93

93 Figure 5-4. Total revenue change from 2004 to 2008, hot spot analysis (getis-ord gi*) 1x1 mile grid cell (second foci, 1/4 mile offset).

PAGE 94

94 Figure 5-5. Total revenue change from 2004 to 2008, hot spot analysis (getis-ord gi*) 1x1 mile grid cell (third foci, 1/2 mile offset).

PAGE 95

95 CHAPTER 6 TRADE AREA CALCULATION: GEODEMOGRAPHICS The importance of knowing what compri ses a customer base and the particular spending habits of those customers cannot be understated. Marketing Geographers utilize selective profiling met hods to group customers into classifications based on their consumption history and Toblers Law. T he commonly accepted terminology of this method of grouping custom ers by geodemographics is as a Tapestry profile. When the Geodemographics are studied, the customer bas e for this particular retail market/service provider reveal the following results by Tapestry: Top 5 Tapestry groups by customer count comprised 65% and the Top 10 Tapestry groups by customer count comprised 87% of the total customer base in 2004. 19.4% were 13-In Style 17.8% were 14-Prosperous Empty Nesters 11.9% were 7-Exurbanites 9.3% were 16-Enterprising Professionals 6.5% were 33-Midlife Junction 5.9% were 28-Aspiring Young Families 5.0% were 63-Dorms to Diplomas 4.9% were 26-Midland Crowd 4.4% were 55-College Towns 2.0% were 46-Rooted Rural Top 5 Tapestry groups by customer count comprised 66% and the Top 10 Tapestry groups by customer count comprised 88.4% of the total cust omer base in 2008. 20.9% were 13-In Style 17.8% were 14-Prosperous Empty Nesters 12.5% were 7-Exurbanites 8.1% were 16-Enterprising Professionals 6.7% were 28-Aspiring Young Families 6.0% were 33-Midlife Junction 5.6% were 63-Dorms to Diplomas 4.6% were 26-Midland Crowd 4.2% were College Towns 2.0% were Rooted Rural

PAGE 96

96 Top 5 Tapestry groups by customer count comprised 55% of the Alachua County, FL customer base in 2004. 18.4% were 13-In Style 19.7% were 14-Prosperous Empty Nesters 0.0% were 7-Exurbanites 10.8% were 16-Enterprising Professionals 6.4% were 33-Midlife Junction Top 5 Tapestry groups by customer count comprised 54% of the Alachua County, FL customer base in 2004. 17.7% were 13-In Style 19.3% were 14-Prosperous Empty Nesters 0.0% were 7-Exurbanites 9.6% were 16-Enterprising Professionals 7.8% were 28-Aspiring Young Families The geodemographics show a very consistent overall Tapestry group capture as well as an individual Tapestry group capture as a percent of the total custom er base. There is a slight variation among some of the groups, but this is negligible and for the most part, the percent of overall customer Tapestry c apture remained constant throughout the five year study period. When we examine t he Alachua County, FL customer Tapestry capture, it is interesting to note that the Exubanites Tapestry group drops to a value of zero for both time periods. This is indica tive of a mobile customer Tapestry group coming into Alachua County, FL specifically for this good/service or, if already here, consuming this good/service, but reporting a home address outsi de of the county. This is typically the case for a University town, such as Gainesville, FL, where students often report their home of record as being collo cated with the parents address, are the parents themselves visiting their children in co llege, are alumni visi ting for a University event, are snowbirds passing through, or fre quent this retail/service provider due to other intangible positive externalities provided only at this location. This is important

PAGE 97

97 because these Exurbanite Tapestry consumers do not live locally, but visit this retail outlet and spend enough to make it into the Top 5 list by customer count. The individual Tapestry groups were se lected from the entire 2008 year group dataset and were used to create the mean center from that group. Recall, from earlier discussion, that the mean center is the gravitational cent er of a spatially distributed dataset. Figure 6-1 is a choropleth map s howing the total revenue change and the Alachua County, FL Tapestry profile group mean center disp layed as an overlay. The map indicates that the correlation indicates that the Top 5 Tapestr y groups are spending within the mid range of values calculated for total revenue change. For a detailed description of each Tapestry gr oup, refer to Appendix A.

PAGE 98

98 Figure 6-1. Highest percentage tapestry pr ofile group mean center (2008) and total revenue change from 2004 to 2008, 1x 1 mile grid cell overlay.

PAGE 99

99 CHAPTER 7 TRADE AREA CALCULATION: FUTURE CONJECTURE This analysis has demonstrated the impor tance to Marketing Geography of the geographic object as a contai ner of attribute data. MA UP has been demonstrated to exist with customer data. This thesis is t he first to demonstrate MAUP using customer data. The implication is t hat trade area analyses need to both hold geography constant, and test for the existence of bias attributable to the manner in which geography is held constant.13 This should be integral with best business practices, but has not been previously recognized by those practicing marketing geography. Trade area analysis is affected by MAUP. Because of recent advances in geospatial technology, GIS automation can be used to detect and correct for MAUP. See Figure 7-1. It is techni cally feasible and pragmatically important to report time sensitive data within the GIS environment. For instance, with an automated process using GIS and an hourly input of consumption data, it is realistic to expect that the computer aided execution of the methods described abov e will provide high value added to the business decision. However, to minimize visual and statistical bias, the conceptualization of MAUP and the associated algorithmic framework introduced here should be applied; and that application should be executed at various scales with multiple variables. These applications add value to business, and have relevance to other disciplines studying spatially distributed phenomena, such as tracking of infectious disease, package delivery ship ments, terrorist threat/inc ident reporting, watch/no-fly lists, and home sales data. By adding GIS geospatial processes to available data and 13 By the phrase "holding geography constant" I m ean the necessity of using same sized and same shaped containers for the geographic data. The containers should be pac kable leaving no area excluded. The literature on Central Place Theory has discussed similar requirements. For further discussion, see King, L. 1984. Central Place Theory Sage Publications: Beverly Hills, CA.

PAGE 100

100 computer generation efficien cy, there is value added to the business decision-making process and the formation of a Common Operating Picture (COP)14 that is applicable in real time. 14A common operating picture ( COP ) is a single identical displa y of relevant (operational) information that facilitates collaborative planning and assists all echelons to achieve situational awareness. Traditionally, headqua rters prepares maps electronical ly with various symbols to show the locations of significant phenomena and other relevant information.

PAGE 101

101 Figure 7-1. Modeling maup flow chart.

PAGE 102

102 CHAPTER 8 TRADE AREA CALCULATION: CONCLUSION The next step for the work presented in this thesis is to create geos patial mapping software that detects the MAUP variables (time, area, scale, size, shape, direction, distance, offset foci, etc.) and ca lculates the threshold values that tip the results into MAUP. Ideally, this software would scan the study area applying the created template every 45 degrees (8 directions ), through 4 foci offsets, and 3 different grid cell sizes resulting in 96 snapshots of the trade area. Of course, this is purely subjective and would be depen dent upon the variables being studied as well as the expertise and judgment of the analyst. Th e proposed software would incorporate the functionality of GIS, tapestry profiles, and fractal geo metry to calculate a potential strength of trade area val ue that has a marketable identity. While the issue of scale has been widely examined in vari ous aspects of physical geography, the MAUP has been largely ignored despite its presence in various types of large scale spatial data analysis (Dark and Bram, 2007). Conceptually, the propos ed software would be an integration of existing open-source GIS software and open-source "R" statistical software. As a side note to this conjecture, it would be adv antageous to have the tapestry data updated on an annual basis for more real time applications while also having access to historical tapestry data. This thesis points to a direct ion for further development of the body of business geographic knowledge and procedures; it merely scr atches the surface when considering the potential importance of MAUP to the business decision. This thesis brings to light that tradit ional business education including education for various professional degrees such as pharmacy and medicine, does not prepare the entrepreneur practitioner to succeed in busi ness. It is essential for entrepreneurs to

PAGE 103

103 realize that cannibalization and market penet ration may be difficult to attain in geographic regions in which competitors are already es tablished. New entering competitors might only be able to compete on t he basis of price competition or intense advertising which raises costs and lowers pr ofit margins. The bias then is for new entrants to the market to locate in newly developed areas or by providing a good or service previously not offered in the market. The value of this thesis extends beyond academic contributions. The results of the thesis are also of value to the "Client" who provided the data. B enefits to the "Client" include the provision of a geos patial overview of the Client 's trade area, and sensitivity (or non sensitivity) to the dynamics wit hin the trade area. Those dynamics include changing competition and customer change.

PAGE 104

104 APPENDIX COMMUNITY TAPESTRY SEGMENTATION SUMMARY DESCRIPTIONS 1. Top Rung Top Rung is the wealthiest consumer market, representi ng less than 1% of all U.S. households. The median household income of $179, 000 is three and a half times that of the national median, and the m edian net worth of $556,400 is more than five times that of the national level. The median home value is approximat ely $1,014,600. These highly educated residents are in their peak earning years, 45-64, in married-couple households, with or without children. The median age is 42.3 years. With the purchasing power to indulge any choice, Top Rung residents travel in style, both domestically and overseas. This is the top market for owning or leasing a luxury car; residents favor new imported vehicles, especially co nvertibles. Exercise and comm unity activities are part of their busy lifestyle. Avid readers, these resi dents find time to read two or more daily newspapers and countless books. 2. Suburban Splendor These successful suburbanites are the epitome of upward mobility, just a couple of rungs below the top, situated in growi ng neighborhoods of affl uent homes, with a median value of $408,100. Most households are comprised of two-income, marriedcouple families with or wit hout children. The population is well-educated and well employed, with a median age of 40.5 years. Home improv ement and remodeling are a main focus of Suburban Splendor residents. Their homes feat ure the latest amenities and reflect the latest in home design. Resi dents travel extensiv ely in the U.S. and overseas for business and pleasure. Leisure acti vities include physical fitness, reading,

PAGE 105

105 and visiting museums, or attending the theater This market is pr oactive for tracking investments, financial planning, a nd holding life insurance policies. 3. Connoisseurs Second in wealth to Top Rung but first for conspicuous consumption, Connoisseurs residents are well-educated and somewhat older, with a median a ge of 45.4 years. Although residents appear closer to retirement than child rearing, many of these married couples have children who still live at home. Their neighborhoods tend to be older bastions of affluence where the median home value is $664,500. Growth in these neighborhoods is slow. Residents spend m oney for nice homes, cars, clothes, and vacations. Exercise is a priority; they work out weekly at a club or other facility, ski, play golf, snorkel, play tennis, practice yoga, and j og. Active in the community, they work for political candidates or parties, write or visit elected officials, and participate in local civic issues. 4. Boomburbs The newest additions to the suburbs, Boomburbs communities are home to younger families who live a busy, upscale lifestyle. T he median age is 33.8 years. This market has the highest population growth at 4.6% annually, more than four times that of the national figure. The median home value is $308,700, and most households have two workers and two vehicles. This is the top ma rket for households to own projection TVs, MP3 players, scanners, and laser printers, as well as owning or leasing full size SUVs. Its the second ranked market for owning flat-screen or plasma screen TVs, video game systems, and digital camcorders, as well as owning or leasing minivans. Family vacations are a top priority. Popular vacation destinations are Disney World and

PAGE 106

106 Universal Studios in Florida. For exerci se, residents play tennis and golf, ski, and go jogging. 5. Wealthy Seaboard Suburbs Wealthy Seaboard Suburbs neighborhoods are established quarters of affluence, located in coastal metropolitan areas, primarily along the California, New York, New Jersey, and New England coasts. Neighborhoods are older and slow to change, with a median home value that ex ceeds $444,600. Households cons ist of married-couple families. Approximately half of employed pe rsons are in management and professional occupations. The median age is 41.7 years. Residents enjoy traveling and shopping. They prefer to shop at Lord & Taylor, Ma cys, and Nordstrom, as well as Costco Wholesale, their favorite club store. They also purchase many items online or by phone. Residents take nice vacations, traveling in the U.S. and abroad. Europe, Hawaii, Atlantic City, Las Vegas, and Disneyland are popular destinations. Leisure activities include going to the beach, skiing, ice ska ting, and attending theater performances. 6. Sophisticated Squires Sophisticated Squires residents enjoy cultured country living in newer home developments with low density and a medi an home value of $244,500. These urban escapees are primarily married-couple fam ilies, educated, and well-employed. They prefer to commute to maintain their semi-rural lifestyle. The median age is 37.4 years. They do their own lawn and landscaping work, as well as home improvement and remodeling projects, such as installing carpet or hardwood floors, a nd interior painting. They like to barbeque on their gas grills and make bread with their bread-making machines. This is the top market for owning 3 or more vehicles. Vehi cles of choice are

PAGE 107

107 minivans and full size SUVs. Family activities include playing volleyball, bicycling, playing board games and cards, going to the zoo, and attending soccer and baseball games. 7. Exurbanites Open areas with affluence define these neighbor hoods. Empty nesters comprise 40% of these households; married couples with children occupy 32%. Half of the householders are between the ages of 45 and 64 years. The median age is 43.6 y ears. Approximately half of those who work hold professional or managerial positions. The median home value is approximately $255, 900; the median household inco me is $83,200. Financial health is a priority for the Exurbanites market; they consult with financial planners and track their investments online. They own a diverse investment portfolio, and hold longterm care and substantial life insurance polic ies. Residents work on their homes, lawns, and gardens. Leisure activities include boati ng, hiking, kayaking, playing Frisbee, photography, and birdwatching. M any are members of fraternal orders and participate in civic activities. 8. Laptops and Lattes The most eligible and unencumbered market, Laptops and Lattes residents are affluent, single, and still renting. They are highly educated, professiona l, and partial to city life, preferring major metropolitan areas such as New York, Los Angeles, San Francisco, Boston, and Chicago. The median household income is $91,000; the median age is 38.1 years. Technologically savvy, this is the top market for owning a laptop or notebook PC; they use the Internet on a daily basis, especially to shop. Their favorite department store, by far, is Banana Republic. Leisure acti vities include going to the

PAGE 108

108 movies, rock concerts, shows, museums, and nightclubs. These residents exercise regularly and take vitamins. They enjoy yoga, jogging, skiing, reading, watching foreign films on video tape/DVD, dining out, and fo reign travel. They embrace liberal philosophies and work for environmental causes. 9. Urban Chic Urban Chic residents are well-educated professionals living an urban, exclusive lifestyle. Most own expensive single-family homes with a median value of $633,000. Married-couple families and singles comprise most of these households. The median age is 41.4 years. Urban Chic residents travel extensively, visit museums, attend dance performances, play golf, and go hiking. They use the Internet frequently to trade or track investments or to shop, buying concert and sports tickets, clothes, flowers, and books. They appreciate a good cup of coffee while r eading a book or newspaper, and prefer to listen to classical music, all-talk, or pub lic radio programs. Civic-minded, they would probably work as volunteers. 10. Pleasant-Ville Prosperous domesticity distinguishes the settled homes of Pleasant-Ville neighborhoods. Most residents live in single-family homes with a median value of $326,500; approximately half were built in the 1950s and 1960s. Located in the Northeast and California primarily, these households are headed by middle-aged residents, some nearing early retirement. T he median age is 39.4 y ears. Approximately 40% of households include children. Home rem odeling is a priority for residents who live in older homes. Shopping choices are eclectic, ranging from upscale department stores,

PAGE 109

109 to warehouse, or club stores. Sports fanatics, they attend ball games, listen to sports programs and games on the radio, and wa tch a variety of sports on TV. 11. Pacific Heights Pacific Heights neighborhoods are found in the high-r ent districts of California and Hawaii. The median home value is approx imately $573,600; residents prefer singlefamily homes or townhomes. This market is small but affluent; one in two households earns approximately $76,000 ann ually. The median age is 38.4 years. Distance does not deter Pacific Heights residents from keeping in touch with family living overseas, as they make frequent phone calls and travel overseas to visit. Many households own 3 or more cell phones. Residents generally visit Dis neyland or Las Vegas during the year, and enjoy playing chess, reading history books, and renting classics on DVD to watch on their giant screen or projection TVs. This is the top market for owning an Apple iMac brand PC. 12. Up and Coming Families Up and Coming Families represents the second highest household growth market and with a median age of 31.9 years, the youngest of the affluent fa mily markets. The profile for these neighborhoods is young, affluent families with young children. Approximately half of the households are concentrated in the South, with another half in the West and Midwest. Neighborhoods are located in suburban outskirts of midsized metropolitan areas. The homes are newer, with a median value of $185,500. Because family and home priorities dictate their consumer pur chases, they frequently shop for baby and childrens products and household furniture. Leisu re activities include playing softball, going to the zoo, and visiting theme parks (generally Sea World or Disney World).

PAGE 110

110 Residents enjoy watching science fiction, comedy, and family-type video tapes or DVDs. 13. In Style In Style residents live in affluent neighborhoods of metropolitan areas. More suburban than urban, they nevertheless embrace an urban lifestyle. Townhome ownership is more than double that of the national level; however, mo re than half of the households live in traditional single-family homes. Labor fo rce participation is high and professional couples predominate. The median household in come is $67,800. Nearly one-third of these households include children. The median age is 39.3 years. In Style residents are computer savvy; they use the In ternet daily to research information, track investments, or shop. They own a diverse investment portfo lio, contribute to reti rement savings plans, and hold long-term care and life insurance policies. They enjoy going to the beach, snorkeling, playing golf, casino gambling, and domestic travel. 14. Prosperous Empty Nesters Prosperous Empty Nesters are well-established neighbor hoods located throughout the U.S.; approximately one-thir d are on the eastern seaboard. The median age is 47.2 years. More than half of the householders are aged 55 or older. Approximately 40% of household types are married couples with no children living at home. Educated and experienced, residents are enjoying the lifes tage transition from child-rearing to retirement. The median household income is $66,200. Residents place a high value on their physical and financial well-being, and ta ke an active interest in their homes and communities. They travel extensively, both at home and abroad. Leisure activities include refinishing furniture, playing golf, attending sports events, and reading

PAGE 111

111 mysteries. Civic participation includes joining civic clubs engaging in fundraising, and working as volunteers. 15. Silver and Gold Silver and Gold residents are the second oldest of the Tapestry segments and the wealthiest seniors, with a median age of 58.5 years; most are retired from professional occupations. Their affluence has allowed them to move to sunnier climates. More than 60% of the households are in the South (mainly in Florid a); 25% reside in the West, primarily in California and Arizona. Neighb orhoods are exclusive, with a median home value of $326,600 and a high proportion of seasona l housing. Residents enjoy traveling, woodworking, playing cards, birdwatching, ta rget shooting, salt water fishing, and power boating. Golf is more a way of life than a mere leisure pursuit; they play golf, attend tournaments, watch golf on TV, and listen to gol f programs on the radio. They are avid readers, but allow time to watch their fa vorite TV shows and a multitude of news programs. 16. Enterprising Professionals This fast-growing market is home to y oung, educated, working professionals, with a median age of 32.4 years. Si ngle or married, they pref er newer neighborhoods with townhomes or apartments. The median househol d income is $66,000. This segment is ranked second of all the Tapestry markets for labor force participation, at 75%. Their lifestyle reflects their youth, mobility, and gr owing consumer clout. Residents rely on cell phones and PCs to stay in touch. They use the Internet to find the next job or home, track their investments, and shop. They own the latest electronic gadgets. Leisure activities include yoga, playing Frisbee and football, jogging, going to the movies, and

PAGE 112

112 attending horse races and basketball games. These residents also travel frequently, domestically and overseas. 17. Green Acres A little bit country, Green Acres residents live in pastoral settings of developing suburban fringe areas, mainly in the Midwest and South. The median age is 39.9 years. Married couples with and without children comp rise most of the households, live in single-family dwellings. This upscale ma rket has a median household income of $62,300 and a median home val ue of $179,700. These do-it-yourselfers maintain and remodel their homes, painting, installing carpet, or adding a deck, and own all the necessary tools to accomplish these tasks. T hey also take care of their lawn and gardens, again with the right tools. Vehicles of choice are motorcycles and full-size pickup trucks. For exercise, residents ride th eir bikes and go water skiing, canoeing, and kayaking. Other activities include birdwa tching, power boating, target shooting, hunting, and attending auto races. 18. Cozy and Comfortable Cozy and Comfortable residents are settled, married, a nd still working. Many couples are still living in the pre-1970s, single-family homes in which they raised their children. Households are located primarily in suburban areas of the Midwest, Northeast, and South. The median age is 41.0 years and the median home value is $164,000. Home improvement and remodeli ng are important to Cozy and Comfortable residents. Although some work is contracted, homeowners take an active part in many projects, especially painting and lawn care. They play softball and golf, attend ice hockey games, watch science fiction films on video tapes/D VDs, and gamble at casinos. Television is

PAGE 113

113 significant; many households have four or mo re sets. Preferred cable stations include QVC, Home & Garden Television, and The History Channel. 19. Milk and Cookies Milk and Cookies households are comprised mainly of young, affluent married-couple families. Approximately half of the household s include children. The median age for this market is 33.5 years. Re sidents preferred single-fam ily homes in suburban areas, chiefly in the South, particularly in Texa s. Smaller concentrations of households are located in the West and Midwest. The medi an home value is $131,900. Families with two or more workers, more t han one child, and two or more v ehicles is the norm for this market. Residents are well-insur ed for the future. The presence of children drives their large purchases of baby and chil drens products, and timesavers such as fast food. For fun, residents play video games, chess, back gammon, basketball, and football, or fly kites. Favorite cable channels include To on Disney, The Discovery Health Channel, ESPNews, and Lifetime Movie Network. 20. City Lights City Lights are diverse neighborhoods, situated primar ily in the Northeast. This dense urban market is a mixture of housing, house hold types, and cultures, sharing the same city walks. Housing types include singlefamily homes, townhomes, and apartments. Approximately 35% of household s are apartments in buildings with two to four units, almost four times the national level. Approximately two-thir ds of the housing units were built before 1960. Households include both fa milies and singles. The median age of 37.8 years is slightly ol der than the U.S. median. City Lights residents are more likely to spend for household furnishings than home maintenance. They shop at a variety of

PAGE 114

114 stores, especially Macys, Lord & Taylor The Disney Store, The Gap, and BJs Wholesale Club. They favor fore ign travel. Being conservative investors, they own U.S. savings bonds. 21. Urban Villages Urban Villages neighborhoods are multicultural encla ves of young families, unique to U.S. gateway cities, located primarily in Ca lifornia. The median age is 30.7 years. All family types dominate this market. The average family size of 4.12 is the second highest of all the Tapestry segments. Many hous eholds have two wage earners, chiefly employed in the manufacturing, health care, retail trade, construction, and educational services industries. The median household income is $56,200. Most residents own older, single-family homes with a median value of $355,600 and multiple vehicles. Family and home dictate purchases. To maintain their older homes, time and money are spent on home remodeling and repairs. Leisure activities include playing soccer and tennis, renting foreign film s, listening to Hispanic and variety radio, and visiting Disneyland, Sea World, or Six Flags. 22. Metropolitans Metropolitans residents favor city living in olde r neighborhoods. Approximately half of the households are comprised of singles who live alone or with others. However, married-couple families are 40% of the households. The median age is 37.1 years. Half of employed persons hold professional or management positions. These neighborhoods are an eclectic mix of single-family homes and multiuni t structures, with a median home value of $194,100. The median household in come is $57,600. Residents lead busy, active lifestyles. They travel frequently and par ticipate in numerous civic activities. They

PAGE 115

115 enjoy going to museums and zoos, and list ening to classical music and jazz on the radio. Refinishing furniture and playing a musical instrument are favorite hobbies. Exercise includes yoga, roller blading, and hiking/backpacking. 23. Trendsetters These neighborhoods are locate d primarily on the West C oast. On the cutting edge of urban style, Trendsetters residents are young, diverse, mobile, educated professionals with substantive jobs. The m edian age is 35.0 years. More than half of the households are single-person or shared. Mo st still rent, preferring upsca le, multiunit dwellings in older city districts. The median household income is $56,700. Residents are spenders; they shop in stores, online, and via the phone. They own t he latest laptop computers, cell phones, and MP3 players, and use the Inte rnet daily. Exercise includes playing tennis, volleyball, baseball, and golf, as we ll as ice skating, snorkeling, and yoga. Leisure activities include traveling, a ttending rock concerts, and reading biographies. Residents also enjoy syndi cated TV shows such as Access Hollywood and Seinfeld 24. Main Street, USA Main Street, USA neighborhoods are a mix of singl e-family homes and multiunit dwellings, found in the suburbs of smaller me tropolitan cities, mainly in the Northeast, West, and Midwest. This market is similar to the U.S. when comparing household type, age, race, educational attainment, housing ty pe, occupation, industry, and household income type distributions. The median age of 36.3 years matches that of the U.S. median. The median household income is a comfortable $51,200. Home homeownership is at 66% and the median home value is $190,200. Active members of the community, residents participat e in local civic issues and work as volunteers. They

PAGE 116

116 take care of their lawns and gardens, and wo rk on small home projects. They enjoy going to the beach and visiting theme parks, as well as playing chess, going bowling or ice skating, and participating in aerobic exercise. 25. Salt of the Earth A rural or small town lifestyle best describes the Salt of the Earth market. The median age is 40.4 years. Labor force participation is higher than the U.S. level and unemployment is lower. Above-average num bers of employed residents work in the manufacturing, construction, mining, and agric ultural industries. The median household income is $48,800. Households are dominated by married-couple families who live in single-family dwellings, with homeownership at 86%. Approxim ately 28% of the households own three or more vehicles. Most homes own a truck; many own a motorcycle. Residents are settled, hardworki ng, and self-reliant, taking on small home projects, as well as vehicle maintenance. Fa milies often own two or more pets, usually dogs or cats. Residents enjoy fishing, hunting, target shooting, attending country music concerts and auto races, and flying kites. 26. Midland Crowd Approximately 10.8 m illion people represent Midland Crowd, Tapestrys largest market. The median age of 36.3 years parallels the U.S. median. Most households are comprised of married-couple families, half with children and half without. The median household income is $48,200. Housing dev elopments are generally in rural areas throughout the U.S. (more village or town th an farm), mainly in the South. Home ownership is at 84%. Two-thirds of househol ds are single-family structures; 28% are mobile homes. This is a somewhat conservative market politically. These do-it-

PAGE 117

117 yourselfers take pride in their homes, la wns, and vehicles. Hunting, fishing, and woodworking are favorite pursuits. Pet owner ship, especially birds or dogs, is common. Many households have a satellite dish, and TV viewing includes various news programs, as well as shows on CMT and Outdoor Life Network. 27. Metro Renters Metro Renters residents are young (approximately 30% are in their twenties), welleducated singles, beginning their prof essional careers in some of the largest U.S. cities such as New York City, Chicago, and Los An geles. The median age is 33.6 years; the median household income is $52,300. As the name Metro Renters implies, most residents are renting apartments in high-rise buildings, living alone or with a roommate. Their interests include traveling, reading tw o or more daily newspapers, listening to classical music and public radi o programs, and surfing the In ternet. For exercise, they work out regularly at clubs, play tennis and vo lleyball, practice yoga, ski, and jog. They enjoy dancing, attending rock concerts, going to museums or the movies, and throwing a Frisbee. Painting and drawing are favorite hobbies. Politically, this market is liberal. 28. Aspiring Young Families Aspiring Young Families neighborhoods are located in large, growing metropolitan areas in the South and West, wit h the highest concentrations in California, Florida, and Texas. Mainly comprised of young, marr ied-couple families or single parents with children, the median age for th is segment is 30.4 years. Half of the households are owner-occupied single-family dwellings or to wnhomes, and half are occupied by renters, many living in newer multiunit buildings. Residents spend much of their discretionary income on baby and childrens products and toys, as well as home furnishings. Recent

PAGE 118

118 electronic purchases include cameras and video game systems. Leisure activities include dining out, dancing, going to the mo vies, attending professional football games, fishing, weight lifting, and playing basketball. Vacations would probably include visits to theme parks. Internet usage mainly involves chat room visits. 29. Rustbelt Retirees Most Rustbelt Retirees neighborhoods can be found in older, industrial cities, in the Northeast and Midwest, especially in Pennsylv ania and other states surrounding the Great Lakes. Households are mainly occupied by married couples with no children and singles who live alone. The median age is 43. 8 years. Although many residents are still working, labor force participation is below average. More than 40% of the households receive Social Security benefits. Most resid ents live in owned, single-family homes, with a median value of $118,500. Unlike many reti rees, these residents are not inclined to move. They are proud of their homes and gardens, and participate in community activities. Some are members of veterans clubs. Leisure activities include playing bingo, gambling in Atlantic City, going to t he horse races, working crossword puzzles, and playing golf. 30. Retirement Communities Retirement Communities neighborhoods are found mostly in cities scattered across the U.S. The majority of households are multi unit dwellings. Congregate housing, which commonly includes meals and other services in the rent, is a trait of this segment, dominated by singles who live alone. This educated, older ma rket has a median age of 50.7 years. A third of residents are aged 65 years or older. Although the median household income is a modest $45,100, the median net worth is $172,000. Good health

PAGE 119

119 is a priority; residents visit their doctors regularly, diet and exercise, purchase lowsodium food, and take vitamins. They s pend their leisure time working crossword puzzles, playing bingo, gardening indoors, c anoeing, gambling, and traveling overseas. They like to spend time with their grandc hildren and spoil them with toys. Home remodeling projects are usually in the works. 31. Rural Resort Dwellers Favoring milder climates and pastoral settings, Rural Resort Dwellers live in rural nonfarm areas. These small, growing communiti es mainly consist of single-family and mobile homes, with a signific ant inventory of seasonal hous ing. This somewhat older market has a median age of 46.0 years. Mo st households consist of married-couples with no children living at home or singles who live alone. A higher than average proportion of residents are self-employed a nd work from home. The median household income is $45,600. Modest living and simple c onsumer tastes describe this market. The rural setting calls for more riding lawn mo wers and satellite dishes. Lawn maintenance and gardening is a priority, and households own a plethora of tools and equipment. Many households own or lease a truck. Re sidents enjoy boating, hunting, fishing, snorkeling, canoeing, and listening to country music. 32. Rustbelt Traditions Rustbelt Traditions neighborhoods are the back bone of older, industrial cities in states bordering the Great Lakes. Most employed residents work in the service, manufacturing, and retail trade industries. Most residents own and live in modest singlefamily homes that have a median value of $97,000. Households are primarily a mix of married-couple families, single-parent familie s, and singles who live alone. The median

PAGE 120

120 age is 35.9 years; the median household income is $45,300. Residents prefer to use a credit union and invest in certificates of deposit. They use coupons regularly, especially at Sams Club, work on home remodeling or improvement projects, and buy domestic vehicles. Favorite leisure activities include hunting, bowling, fishing, and attending auto races, country music shows, and ice hockey games (in addition to listening to games on the radio). 33. Midlife Junction Midlife Junction communities are found in suburbs across the country. Residents are phasing out of their child-rear ing years. Approximately half of the households are comprised of married-couple families; 31% are singles who live alone. The median age is 40.5 years; the median household income is $43,600. A third of the households receive Social Security benefits. Nearly tw o-thirds of the households are single-family structures; most of the remain ing dwellings are apartments in multiunit buildings. These residents live quiet, settled lives. They spend their money prudently and do not succumb to fads. They prefer to shop by mail or phone from catalogs such as J.C. Penney, L.L. Bean, and Lands End. They enjoy yoga, attending country music concerts and auto races, refinishing furniture, and reading romance novels. 34. Family Foundations Family is the cornerstone of life in Family Foundations communities. A family mix of married couples, single parent s, grandparents, and young an d adult children populate these small, urban neighborhoods, located in large metropolitan areas, primarily in the South and Midwest. This market represents st ability. Hardly any household growth has occurred since 2000; these neighborhoods experi ence little turnover. The median age is

PAGE 121

121 38.1 years; the median household income is $42,100. Most households are singlefamily structures, built bef ore 1970, occupied by owners. Many residents are members of church boards or religious clubs, and participate in fundraising. Basketball is a favorite sport; residents play it, attend professional games, watch games on TV, and listen to games on the radio. They watch court TV shows, sports, and news programs on TV, and listen to gospel, urban, and jazz radio formats. 35. International Marketplace Located primarily in cities in coastal gateway states, International Marketplace neighborhoods are developi ng urban markets with a rich blend of cultures and household types. Approximately 70% of househol ds are occupied by families. Marriedcouples with children and single parents wit h children represent 44% of households. A typical family rents an apartment in an older, multiunit struct ure. Most of the households are located in California and Northeastern stat es. The median age is 30.4 years and the median household income is $42,600. Top pur chases include groceries and childrens clothing. Residents shop at stores such as Marshalls and Costco Wholesale, but for convenience, they stop at AM/PM or 7-Eleven. They are loyal listeners of Hispanic radio programs, and prefer to watc h movies and sports on TV. 36. Old and Newcomers Old and Newcomers neighborhoods are in transiti on, populated by those who are starting their careers, or ar e retiring. The proportion of hous eholders in their twenties or aged 75 years or older is higher than the nati onal level. The median age is 36.6 years. Spread throughout metropolitan areas of th e U.S., these neighborhoods have more single-person and shared households than fam ilies. Many residents have moved in the

PAGE 122

122 last five years. Approximately 60% of households are occ upied by renters; approximately half live in mid-rise or high-rise buildings. Residents have substantial life insurance policies and investments in certificates of deposit, bonds, and annuities. Leisure activities include roller skating, rolle r blading, playing golf, gambling at casinos, playing bingo, and attending coll ege ball games. They listen to classic hits on the radio. Many residents are members of fr aternal orders or school boards. 37. Prairie Living Agriculture plays an im portant part of the Prairie Living economy; small, family-owned farms dominate this stable market, located ma inly in the Midwest. Two-thirds of the households are married-couple families; the median age is 40.5 years. Homeownership is at 81%; the median home value is $96,300. Although si ngle-family dwellings are characteristic of these communities, 11% of the households live in mobile homes. Approximately 36% of the hous ing units were built before 1940. These residents are big country music fans, and enjoy hunting, fishin g, target shooting, and horseback riding. They work on their vegetable gardens, vehicl es, and home projects. Many are members of church boards or civic clubs, and get invo lved in civic issues. Because cable TV can be unavailable in these rural areas, many households have a satellite dish. Families with pet cats or dogs are common. 38. Industrious Urban Fringe Industrious Urban Fringe neighborhoods are found on the fri nge of metropolitan cities. Approximately half of these households are loca ted in the West; 40% are in the South. Most employed residents work in the manufacturing, construction, retail trade, and service industries. Family is central, and ch ildren are present in more than half of the

PAGE 123

123 households. Many live in multi generational households. The median age is 28.5 years; the median household income is $40,200. Tw o-thirds of the households own their single-family dwellings, with a median value of $131,400. Necessities for babies and children are among their pr imary purchases along with toys and video games. Big movie fans, residents visit the cinema severa l times a month and watch movies at home frequently. They prefer to watch syndica ted TV and listen to Hispanic radio. 39. Young and Restless Change is the constant in this diverse ma rket. With a median ag e of 28.9 years, the population is young and on the go. About 85% of householders moved in the last five years. Young and Restless householders are primarily rent ers, living in apartments in multiunit buildings. Almost 60% of households are single-person or shared. This educated market has the highest labor forc e participation among all the Tapestry segments, at 75%, and the highe st female labor force participation, at 73%. The median household income is $40,900. Residents use the In ternet daily, to visit chat rooms, play games, obtain the latest news, and search for employment. They read computer and music magazines, and listen to public radio. They watch movies in the theater and on video/DVD, attend rock concerts, play poo l, go dancing, and exercise weekly at a facility. 40. Military Proximity Military Proximity communities depend upon the military fo r their livelihood. More than 75% of the labor force is in the Armed Forces, while others work in civilian jobs on military bases. The median household income is $40,100 and the median age is 22.5 years. Two-thirds of the households are composed of married couples with children.

PAGE 124

124 Housing types are mainly townhomes and apar tments in small multiunit buildings; 93% are occupied by renters. Resi dents participate in civic activities and are members of business clubs. Many homes have a pet, most likely a dog. Residents use the Internet to trade stocks and make purchases. For exercise they snorkel, play tennis, practice yoga, and jog. Families visit theme parks and the zoo, throw Frisbees, and go bowling. Recent purchases include MP3 players, digital cameras, video game systems, cell phones, apparel, and jewelry. 41. Crossroads Young families living in mobile homes typify Crossroads neighborhoods, found in small towns throughout the South, Midwest, and We st. These growing communities are home to married-couple and single-parent fa milies. The median age is 31.9 years. Homeownership is at 77% and the median home value is $60,300. More than half of the householders live in mobile homes; 36% live in single-family dwellings. Employment is chiefly in the manufacturing, construction, retail trade, and service industries. Many homes have dogs. Residents generally shop at discount stores, but also frequent convenience stores. They pref er domestic cars and trucks, often buying and servicing used vehicles. Residents go fishing, attend aut o races, participate in auto racing, and play the lottery. An annual family outing to Sea World is common. Outer Limits is a favorite weekly TV show. 42. Southern Satellites Southern Satellites neighborhoods are rural settlements found primarily in the South, with employment chiefly in the manufacturi ng and service industries. Married-couple families dominate this market. The median age is 37.1 years and the median household

PAGE 125

125 income is $37,700. Most housing is newer singl e-family dwellings or mobile homes with a median value of $81,400, occupied by owner s,. Residents enjoy country living. They listen to gospel and country music on the radio, and attend country music concerts. They participate in fishing, hunting, and auto racing. Favorite TV stations are CMT and Outdoor Life Network. Satell ite dishes are popular in these rural locations. Households own older, domestic vehicles, particularl y trucks and 2-door sedans. Residents invest time in vegetable gardening, and households are likely to own riding mowers, garden tractors, and tillers. 43. The Elders The Elders median age of 73.4 years represents Tapestrys oldest market. The highest concentration of retiree residents prefer communities designed for senior living, primarily in warm climates. Ha lf of these households are loca ted in Florida, and 30% are situated in Arizona or California. Approx imately 80% of househo lds collect Social Security benefits; 48% receive retirement income. These residents are members of veterans clubs and fraternal orders. Health-c onscious, they take vitamins, visit doctors regularly, and watch their diets. Leisure activities include traveling, working crossword puzzles, fishing, attending horse races, gam bling at casinos, going to the theater, and dining out. They play golf, listen to golf on the radio, and watch tournaments on The Golf Channel. Their daily routine incl udes watching TV and reading newspapers. 44. Urban Melting Pot The ethnically rich Urban Melting Pot neighborhoods are made up of recently settled immigrants; more than half of whom were born abroad. Half of the foreign-born residents immigrated to the U.S. in the last 10 years. Mo st rent apartments in high-

PAGE 126

126 density urban canyons of large cities, primarily in New York and California. Approximately half of the housing units were built before 1950. The median age is 35.7 years and the median household income is $37,400. Fashionand cost-conscious, these residents love to shop, from upscale retailers to warehouse/club stores. Leisure activities include going to the beach, vi siting theme parks and museums, playing football, ice skating, and roller blading. Dis tance does not deter these residents from contacting family living outside the U.S. They keep in touch with phone calls and foreign travel. 45. City Strivers City Strivers are urban denizens of densely settled neighborhoods in major metropolitan areas, such as New York City and Chicago. Most households are composed of a mix of family types. The median age is 32.1 y ears and the median household income is $36,800. Employment is concentrated in the city, with half of employed residents working in the service industry, particular ly in health care. Approximately 22% are government workers. Unemployment is twice that of the U.S. level. Housing is mostly older, rented apartments in smaller multi unit buildings. Primary spending is for groceries, baby products, and childrens essentials. Residents enjoy going to dance performances, football and basketball games, and Six Flags theme parks. They listen to urban, all-news, and jazz radio formats, and watch lots of TV, especially movies, sitcoms, news programs, courtroom TV and talk shows, tennis, and wrestling. 46. Rooted Rural Rooted Rural neighborhoods are located in rura l areas throughout the country; however, more than three-fifths of the households are located in the South. Households

PAGE 127

127 are dominated by married-couple families; approximately one-third of whom already receive Social Security benefits. The median age is 41.0 years. Housing is predominantly single-family dwellings, wit h a strong presence of mobile homes and some seasonal housing. The median home value is $89,900. Stable and settled, residents tend to move infrequently. They ar e do-it-yourselfers, constantly working on their homes, gardens, and vehicles. Many fa milies have pets. Residents enjoy hunting, fishing, target shooting, boating, attendi ng country music concerts, and listening to country music on the radio. Many households have a satellite dish; favorite stations include Outdoor Life Network and CMT. 47. Las Casas Las Casas residents are the latest wave of we stern pioneers. Settled primarily in California, approximately half were born outside the United States. Young, Hispanic families dominate these households; 63% include children. This market has the highest average household size (4.27) among all t he Tapestry segments. The median age is 25.4 years and the median household income is $35,400. Most households are occupied by renters, although homeownership is at 42%. The median home value is $278,400. Housing is a mix of older apartm ent buildings, single-family homes, and townhomes. This is a strong market for purchase of baby and childrens products. Residents enjoy listening to Hispanic radio, reading adventure stories, and playing soccer. Many treat their children to a fa mily outing at a theme park, especially Disneyland. When taking a trip, Me xico is a popular destination. 48. Great Expectations

PAGE 128

128 Great Expectations neighborhoods are located throughout the country, with higher proportions found in the Midwest and South. Young singles and married-couple families dominate. The median age is 33.0 years. Labor force participation is high. Manufacturing, retail, and service industries are the primary employ ers. Approximately half of the households are owners living in si ngle-family dwellings, with a median value of $100,600; the other half are rent ers, mainly living in apartment s in low-rise or mid-rise buildings. Most of the housing units in t hese older suburban neighborhoods were built before 1960. Residents enjoy a young and active lifestyle. They go out to dinner, to the movies, to bars, and to nightclubs. They enjoy roller skating, roller blading, playing Frisbee, chess and pool, and attending auto ra ces. They read music magazines and listen to rock music on the radio. 49. Senior Sun Seekers The Senior Sun Seekers market is one of the faster gro wing markets, located mainly in the South and West, especially in Florida. Escaping from cold winter climates, many residents have permanently relocated to wa rmer areas; others ar e snowbirds who move south for the winter. Mo st residents are retired or ar e anticipating retirement. The median age is 51.4 years; 62% of the householders are aged 55 years or older. Most households are single-family dwellings or mobile homes, with a median value of $107,500. There is a high proportion of seas onal housing. Many residents are members of veterans clubs or fraternal orders. They own lots of insurance and consult with a financial advisor. Leisure activities include dining out, reading (especially boating magazines), watching TV, fishing, playi ng backgammon and bingo, working crossword puzzles, and gambling at casinos.

PAGE 129

129 50. Heartland Communities Heartland Communities neighborhoods are preferred by approximately 6 million people. These neighborhoods can be found pr imarily in small towns, primarily in the Midwest and South. More than 75% of the households are single-family dwellings, with a median home value of $74,400. Most homes are ol der, built before 1960. The median age is 41.3 years; nearly one-third of the householders are aged 65 years or older. The distinctly country lifestyle of these residents is reflected in their interest in hunting, fishing, woodworking, playing bingo, and lis tening to country music. In addition to working on home improvement projects, t hey are avid gardeners and read gardening magazines. They participate in civic activiti es and take an interest in local politics. Residents order items from catalogs, QV C, and from Avon sales representatives. 51. Metro City Edge Metro City Edge residents live in older suburban neigh borhoods of large metropolitan cities, primarily in the Midwest and South. This market is home to married-couple, single-parent, and multi-generational families. The median age is 29.1 years and the median household income is $30,200. Nearly half of employed residents work in the service industry. Most households live in sing le-family dwellings; 14% live in buildings with 2 to 4 units, many duplexes. Homeowner ship is at 56% and the median home value is $74,100. Prudent shoppers, resi dents buy household and childrens items at superstores and wholesalers. They enjoy watching TV (especially sitcoms and courtroom TV shows), going to the movies, visiting theme parks, roller skating, and playing basketball. They read music, gardening, and baby magazines, and listen to urban and gospel radio.

PAGE 130

130 52. Inner City Tenants Inner City Tenants neighborhoods are a micr ocosm of urban diversit y, located primarily in the South and West. This multicultural market is young, with a median age of 27.8 years. Households are a mix of singles and families. Most residents rent economical apartments in midor high-rise buildings. Recent household purchases by this market include video game systems, ba by food, baby products, and furniture. Internet access at home is not typical; those who have no access at home surf the Inter net at school or at the library. Playing games and visiting chat rooms are typical online activities. Residents frequently eat at fast-food re staurants. They enjoy going to the movies, attending football and basketball games, water skiing, and playing football, basketball, and soccer. Some enjoy the nightlife, vi siting bars and nightclubs to go dancing. 53. Home Town These low-density, settled neighborhoods, loca ted chiefly in the Midwest and South, rarely change. Home Town residents stay close to thei r home base. Although they may move from one house to another, they rarely cross the county line. Household types are a mix of singles and families. The median age is 33.7 years. Single-family homes predominate in this market. Homeownership is at 61% and the median home value is $61,800. The manufacturing, retail trade, and service industries are the primary sources of employment. Residents enjoy fishing and play ing baseball, as well as playing bingo, backgammon, and video games. Favorite cable TV stations include CMT, Nick at Nite, Game Show Network, and TV Land. When shopping, Belk and Wal-Mart are favorite stops, but residents also purchase items from Avon sales representatives. 54. Urban Rows

PAGE 131

131 With about 1.2 million people, Urban Rows is the smallest Tapestry segment. Row houses are characteristic of these neighbor hoods found primarily in large Northeastern cities, with much smaller concentrations in the South. Two-thirds of the households are in Pennsylvania; one-fifth are in Maryl and. Homeownership is 62% and the median home value is $81,300. Most housing was bu ilt before 1950. Households are a mix of family types. Nearly half of the households do not own a vehicle. The median age is 32.9 years. These residents rarely eat out. T hey prefer BJs Wholesale Club for general shopping; preferred grocery stores are Acme, Pathmark, and Giant. Residents enjoy roller skating; playing baseball; attending baske tball games; listening to urban, variety, and jazz radio programs; and watching sitcom s and sports on TV. Many households do not subscribe to cable. 55. College Towns Education is the key focus for College Towns residents. College and graduate school enrollment is approximately 41%. The median age for this market is 24.5 years, with a high concentration of 18-24-y ear-olds. One out of eight residents lives in a dorm on campus. Students in off-campus housing liv e in low-income apartment rentals. Approximately 31% of the hous eholds are typically town residents who live in owneroccupied, single-family dwellings. The medi an home value is $132,900. Convenience is the primary consideration fo r food purchases; residents frequently eat out, order in, or eat easy-toprepare food. Many own a laptop computer. In t heir leisure time, they jog, go horseback riding, practice yoga, play tenni s, rent videos, play chess or pool, attend concerts, attend college football or basket ball games, and go to bars. They listen to classical music and public radio programs.

PAGE 132

132 56. Rural Bypasses Open space, undeveloped land, and farmland are found in Rural Bypasses neighborhoods, located almost entirely in the South. This market is home to families who live in small towns along country ba ck roads. The median age is 37.1 years. Higher-than-average proportions of employed residents work in the agricultural, mining, manufacturing, and construction industries. Labor force participation is low and unemployment is high. Although most households are single-family dwellings, 32% are mobile homes. Homeownership is at 78% and the median home value is $58,500. Residents save money by maintaining their homes, gardens, and vehicles themselves. They enjoy hunting, reading fishing and hunting magazines, and listening to gospel radio. They prefer to watch courtroom TV and talk shows, as well as cartoons. Recent purchases include baby products, clothes, and toys. 57. Simple Living Simple Living neighborhoods are found throughout t he U.S., in urban outskirts or suburban areas. Half of the households are singles who live alone or share housing, and 32% consist of married-couple fam ilies. The median age is 40.1 years. Approximately one-third of householders are aged 65 years or older; 19% are aged 75 years or older. Housing is a mix of singlefamily dwellings and multiunit buildings of varying stories. Some seniors live in congregate housin g (assisted living). Approximately 55% of househo lds are occupied by renter s. Approximately 40% of households receive Social Security benefit s. Younger residents enjoy going out dancing, while seniors prefer going to bingo nigh t. To stay fit, residents play softball and

PAGE 133

133 volleyball. Many households do not own a PC, cell phone, or DVD player. Residents watch a lot of TV, especially sitcoms and science fiction shows. 58. NeWest Residents Most NeWest Residents rent apartments in midor high-rise buildings in primarily in major western and southern cities. Californi a has the largest concentration of these households, followed by Texas. Families dominate this market. Children reside in 54% of the households, either in married-couple or single-parent families. Approximately half of the population is foreign-born. This young market has a median age of 25.3 years. Most of the employed residents work in service and skilled labor occupations. These residents lead a strong family-oriented lifestyle. Budget constraints restrict their purchases to essentials such as baby food, equipment, and products, as well as childrens clothing. For fun, families go to the movies, visit theme parks, and play soccer. They like to watch sports on TV, es pecially wrestling and soccer, and listen to Hispanic radio. 59. Southwestern Families These families are the bedrock of the Hispan ic culture in the Southwest, more with children than without. Two-thirds of the househol ds live in owner-occupied, single-family dwellings with a median home value is $52, 100. Most employed residents work in bluecollar or service occupations. Southwestern Families is an ethnically diverse market, with a median age of 28.2 years and a median household income of $26,600. Recent purchases include baby and childrens products. Households generally own or lease a 2-door sedan. The grocery stor e of choice is H.E. Bu tt. When eating fast food, Whataburger is a favorite stop. Residents enjo y fishing, water skiing, playing soccer,

PAGE 134

134 and going to the movies. They read gar dening and parenthood magazines, and listen to Hispanic and urban radio formats. Typical TV viewing includes comedies, as well as wrestling and boxing. 60. City Dimensions Diversity in household type and ethnicity characterize City Dimensions neighborhoods that are located in large urban cities. Population density remains high, with approximately 2,900 people per square mile. This market is young, with a median age of 29.0 years. Nearly 63% of households rent; more than half are apartments in multiunit structures. Most of the real estate is older. Approximatel y 70% of the housing units were built before 1960, 42% of which were built before 1940. Many households lease their vehicles, preferring Mercury or Ford models. Residents shop at BJs Wholesale Club, Kmart, Marshalls, and T.J. Maxx. They enjoy roller skating, playing soccer and chess, attending auto races and s hows, going to the movies, and renting movies on DVD (especially classics, horror, and science fiction). Video game systems are quite popular also. 61. High Rise Renters This segment has the highest percentage of renters among all of the Tapestry segments; more than nine in ten households are renters in these densely populated neighborhoods. Approximately 41% rent in buildings with 50 or more units. High Rise Renters communities are located almost ent irely in the Northeast; 86% of the households are in New York. Residents repres ent a diverse mix of cultures; many speak a language other than English. The me dian age is 29.6 years. Household types are mainly single-parent and si ngle-person. Part-time work is just as common as full-

PAGE 135

135 time. Residents do aerobics and play soccer. They enjoy dancing, attending basketball and football games, watching movies on video tapes/DVDs, and listening to all-news, urban, and Hispanic radio. They watch a vari ety of news programs and are avid viewers of daytime TV. 62. Modest Income Homes Modest Income Homes neighborhoods are found primarily in the older suburbs of metropolitan areas. Single-family dwellings represent more than two thirds of the housing; 15% are duplexes. The median home value is $52,800. Household types are mainly single-person and single-parent. However, approxim ately 64% of households are family types. The median age is 35.0 years. Slightly more employed residents work part-time than full-time, main ly in service and blue-collar occupations. At 20%, unemployment is high. These frugal resident s shop at discount stores, do not pay for Internet access, and rarely eat out. They are content to wait for movies to be shown on TV instead of going to the t heater. They watch daytime and primetime TV, especially courtroom TV shows and sitcoms, and listen to urban and gospel radio. A favorite cable channel is BET. 63. Dorms to Diplomas Dorms to Diplomas is Tapestrys youngest market, wit h a median age of 21.8 years. College and graduate school enrollment is approximately 81%. Nearly three-fourths of employed residents work part-time in low-paying service industry jobs. Approximately 43% of residents live in on-campus dormitori es; the remainder rent apartments in offcampus multiunit buildings. Approximately 90% of households are renters. PCs are a necessity, and the Internet is easily accessibl e to research assignments, search for

PAGE 136

136 jobs, obtain the latest news, and keep in touch with family. For exercise, residents participate in a variety of sports. They enjoy going to college football and basketball games, rock concerts, the movies, and bars, as well as dancing, playing pool, and renting video tapes/DVDs. They listen to classic hits, public, and rock radio programs. 64. City Commons City Commons neighborhoods are found in cities of large metropolitan areas, mainly in the South and Midwest. This younger market has a median age of 24. 2 years. Singleparent families and singles dominate these households, and children abound. Approximately 77% of the househ olds are renters; approximat ely 63% of the rentals are apartments in multiunit buildings primarily with fewer than 20 units. More residents work part-time instead of full-time. This market has the highest unemployment rate among all of the Tapestry segments. Baby and child rens products are the major purchases. Residents enjoy playing basketball, softball, and backgammon. A yearly family outing to a theme park is common. They prefer cour troom TV shows when watching television; listen to gospel, urban, and jazz programs on the radio; and read music, baby, parenthood, and fashion magazines. 65. Social Security Set Four in ten residents in the Social Security Set segment are aged 65 ye ars or older; the median age is 44.6 years. Mo st of these residents live al one. Located in large cities scattered across the U.S., these communities are dispersed among business districts and around city parks. The servic e industry provides more than half of the jobs held by residents who will work. Households subsist on very low fixed incomes. Most residents rent apartments in low-rent, high-rise build ings. Many rely on public transportation,

PAGE 137

137 because more than half of these households do not own a vehicle. Limited resources somewhat restrict the purchas es and activities of these re sidents, although many have invested their savings in stock. They enjoy going to the movies and soccer games, and reading science fiction. Many households subscribe to cable TV; residents particularly enjoy watching game shows, sports and entertainment news shows. 66. Unclassified Unclassified neighborhoods include unpopulat ed areas such as parks, golf courses, open spaces, or other types of undeveloped land. Institutional group quarters, such as prisons, juvenile detention homes, mental hospitals, or any area with insufficient data for classification are also included in this category.

PAGE 138

138 LIST OF REFERENCES Aalberts, R ., and D. Bible. 1992 Geographic Info rmation Systems: App lications for the Study of Real Estate. The Appraisal Journal 60: 483-492. Applebaum W.1965a. Can Store Location be a Science? Economic Geography 41: 234237. Applebaum W. 1965b. Measuring Retail Mark et Penetration for a Discount Food Supermarket a Case Study. Journal of Retailing 41: 1-47. Applebaum, W. 1966. Methods for Determining Store Trade Areas, Market Penetration and Potential Sales. Journal of Marketing Research 3: 127-141. Applebaum, W. and S. Cohen. 1960. Evaluati ng Store Sites and Determining Store Rents. Economic Geography 36:1-35. Baker, S. and K. Baker. 1993. Market Mapping: How to Use Revolutionary New Software to Find, Analyze, and Keep Cust omers. New York, NY: McGraw-Hill, Inc. Boyce, R. and W. Clark. 1964. The Concept of Shape in Geography. Geographical Review 54, 4: 561-572. Dark, S. and D. Bram. 2007. The Modifiable Areal Unit Problem (MAUP) in Physical Geography. Progress in Physical Geography 31,5: 471-479. Davies, R. and D. Rogers. 1984. Store Location and Store Assessment Research. New York, NY: John Wiley & Sons. Davies, R. 1977. Marketing Geography: With S pecial Reference to Retailing. London, England: Methuen. Epstein, B. 1978. Marketing Geography: A Chronice of 45 Years. Proceedings of the Applied Geography Conference 1: 372-379. Fik, T., B. Sidman, and R, Swett. 2005. A Methodology for Delineating a Primary Service Area For Recreational Boaters Us ing a Public Access Ramp: A Case Study of Cockroach Bay. The Florida Geographer 36. 23-40. Fotheringham, S. 1981. Spatial Structure and Distance Decay Parameters. Annals of the Association of American Geographers 71. 425-436. Ghosh, A. and C. Craig. 1986. An Approach to Determining Optimal Locations for New Services. Journal of Marketing Research 23: 354-362.

PAGE 139

139 Ghosh, A. and S. McLafferty. 1987 Location Strategies for Retail and Service Firms. Lexington, KY: D.C. Heath and Company. Goldstucker, J., D. Bellenger T. Stanley, and R, Otte. 1978. New Developments in Retail Trading Area Analysi s and Site Selection. College of Business Administration, Georgia Stat e University, Atlanta, GA. Goss, J. 1995a. Marketing the New Mark eting: The Strategic Discourse of Geodemographic Informati on Systems. Ground Trut h. New York, NY: The Guilford Press. Goss, J. 1995b. We Know Who You Are and We Know Where You Live: The Instrumental Rationality of Geodemographics Systems. Economic Geography 71,2: 171-198. Haynes, K. and S. Fotheringham. 1984. Grav ity and Spatial Interaction Models. Scientific Geography Series. West Virgin ia University, Morgantown, WV: Sage Publications. Heikkila, E. 1998. GIS is Dead; Long Live GIS! Journal of the American Planning Association 64, 3: 350-360. King, L. 1984. Central Place Theory. Scient ific Geography Series. West Virginia University, Morgantown, WV:. Sage Publications. King, L. 1993. Financial instit utions Profiting from a G eographic Information System. Castle G (ed.) Fort Collins: GIS World 57-74. Maantay, J. 2007. Asthma and Air Pollution in the Bronx: Methodological and Data Considerations in Using GIS for Envi ronmental Justice and Health Research. Health and Place 13, 1: 32-56. Mercurio, J. 1984. Store Location and St ore Assessment Research. Davies R and Rogers D (eds.) New York, NY: John Wiley and Sons 237-262. Moloney, T. 1989. A Case Study Using a Geographic Information System in Food Retailing. Operational Geographer 7:23-27. Nelson, R. 1958. The Selection of Reta il Locations. New York, NY: FW Dodge. Patel, A., T. Fik, and G. Thrall. 2008. Direc tion Sensitive Wedge-Casting for Trade Area Delineation. Journal of Real Estate Portfolio Management 14, 2: 125-139. Pickles, J. (ed.) 1995. Representations in an Electronic Age: Geography, GIS and Democracy. Ground Truth: The Social Im plications of Geographic Information Systems. New York, NY: Guilford Press.

PAGE 140

140 Peuquet, D. and D. Marble. 1990. Introductory Reading in Geograp hic Information Systems. London, England: Taylor & Francis. Ritchey, J. 1984. Developing a Strategic Planning Data Base Store Location and Store Assessment Research. Davies, R., and R ogers, D. (eds.). New York, NY: John Wiley & Sons. Rogers, D. and H. Green. 1978. A New Perspective on Forecasting Store Sales: Applying Statistical Models and Techniques in the Analog Approach. Geographical Review 69: 449-458. Simmons, M. 1984. Store Assessment Pr ocedures. Store Location and Store Assessment Research. Davies, R., and R ogers, D. (eds.). New York, NY: John Wiley & Sons. Swift, A., L. Liu, and J. Uber. 2008. Reducing MAUP Bias of Correlation Statistics Between Water Quality and GI Illness. Computers, Environment and Urban Systems 32, 2: 134-148. Thomas, I., P. Frankhauser, and M. DeKeersm aeker. 2007. Fractal Dimension Versus Density of Built-Up Surfaces in the Periphery of Brussels. Papers in Regional Science 86, 2: 287-308. Thrall, G. 1987. Land Use and Urban Form. London, England: Methuen. Thrall, G. 1988. Statistical and Theoretical I ssues in Verifying the Population Density Function. Urban Geography 9: 518-537. Thrall, G. and S. Thrall. 1994. Business GIS Da ta, Part Three: ZipPlus 4 Geocoding. Geo Info Systems 4: 57-60. Thrall, G. 1995. The Stages of GIS Reasoning. Geospatial Solutions 5, 2: 46-51. Thrall, G., J. del Valle, and S. Thrall. 1995. Ninety Years of Urban Growth as Described with GIS: A Historic Geography. Geo Info Systems 5: 20-45. Thrall, G. and S. McMullin. 2000. Trade Ar ea Analysis: The Buck Starts Here. Geospatial Solutions 10, 6: 45-49. Thrall, G. 2000. The Future of GIS in Public Health Management and Practice. Geospatial Solutions 10, 9: 2-7. Thrall, G. 2002. Business Geography and New Re al Estate Market Analysis. Oxford, England: Oxford Press.

PAGE 141

141 Thrall, G., E. Borden, and S. Thrall. 2002. Delineating Hospital Trade Areas: Its Practically Brain Surgery. Geospatial Solutions 10, 8: 38-46. Thrall, G. and N. Mecoli. 2003. Spatial Analysis, Political Support, and Higher Education Funding. Geospatial Solutions 10, 7: 46-49. Wofford, L. and G. Thrall. 1997. Real Estate Problem Solving and Geographic Information Systems: A Stage Model of Reasoning. Journal of Real Estate Literature 5: 177-201. Wolf, J. 1969. SYMAP: Computer Graphics for Marketing Management. Journal of Marketing Research 6: 357-358.

PAGE 142

142 BIOGRAPHICAL SKETCH James Ronald Dietz is the son of Glenn and Myrian Dietz. Born in Coral Gables, Florida where he lived with his family until graduation from High Sc hool in 1989. He subsequently joined the U. S. Navy for Active Duty service and, upon completion of that obligation, moved to Jacksonville, Flori da where his studies in geography began at Jacksonville University under the mentorsh ip of Raymond Oldakowski. James Ronald Dietz received his Bachelor of Arts degree with Honors in geography from Jacksonville University in 1994, and began his graduate studi es at the University of Florida, Department of Geography, in 1994 while also serving as a Commissioned Officer in the U. S. Navy. He is currently working toward s his Master of Arts degree in geography with special emphasis on retail market trade area calculation.