Using Geospatial Reasoning in Institutional Research

morris_p.pdf

morris_p_Page_93.txt

morris_p_Page_66.txt

morris_p_Page_84.txt

morris_p_Page_04.txt

morris_p_Page_81.txt

morris_p_Page_24.txt

morris_p_Page_38.txt

morris_p_Page_05.txt

morris_p_Page_96.txt

morris_p_Page_76.txt

morris_p_Page_71.txt

morris_p_Page_12.txt

morris_p_Page_07.txt

morris_p_Page_53.txt

morris_p_Page_77.txt

morris_p_Page_08.txt

morris_p_Page_48.txt

morris_p_Page_35.txt

morris_p_Page_62.txt

morris_p_Page_72.txt

morris_p_Page_67.txt

morris_p_Page_50.txt

morris_p_Page_21.txt

morris_p_Page_92.txt

morris_p_Page_91.txt

morris_p_Page_31.txt

morris_p_Page_23.txt

morris_p_Page_65.txt

morris_p_Page_69.txt

morris_p_Page_19.txt

morris_p_Page_27.txt

morris_p_Page_58.txt

morris_p_Page_63.txt

morris_p_Page_57.txt

morris_p_Page_56.txt

morris_p_Page_28.txt

morris_p_Page_51.txt

morris_p_Page_45.txt

morris_p_Page_79.txt

morris_p_Page_36.txt

morris_p_Page_47.txt

morris_p_Page_02.txt

morris_p_Page_54.txt

morris_p_Page_59.txt

morris_p_Page_41.txt

morris_p_pdf.txt

morris_p_Page_82.txt

morris_p_Page_16.txt

morris_p_Page_60.txt

morris_p_Page_80.txt

morris_p_Page_68.txt

morris_p_Page_11.txt

morris_p_Page_74.txt

morris_p_Page_75.txt

morris_p_Page_20.txt

morris_p_Page_17.txt

morris_p_Page_13.txt

morris_p_Page_70.txt

morris_p_Page_34.txt

morris_p_Page_10.txt

morris_p_Page_06.txt

morris_p_Page_44.txt

morris_p_Page_87.txt

morris_p_Page_46.txt

morris_p_Page_64.txt

morris_p_Page_86.txt

morris_p_Page_88.txt

morris_p_Page_73.txt

morris_p_Page_39.txt

morris_p_Page_32.txt

morris_p_Page_22.txt

morris_p_Page_26.txt

morris_p_Page_18.txt

morris_p_Page_61.txt

morris_p_Page_83.txt

morris_p_Page_43.txt

morris_p_Page_09.txt

morris_p_Page_52.txt

morris_p_Page_03.txt

morris_p_Page_37.txt

morris_p_Page_85.txt

morris_p_Page_94.txt

morris_p_Page_90.txt

morris_p_Page_30.txt

morris_p_Page_25.txt

morris_p_Page_14.txt

morris_p_Page_29.txt

morris_p_Page_15.txt

morris_p_Page_42.txt

morris_p_Page_40.txt

morris_p_Page_55.txt

morris_p_Page_78.txt

morris_p_Page_89.txt

morris_p_Page_95.txt

morris_p_Page_49.txt

morris_p_Page_33.txt

morris_p_Page_01.txt

Over the last half century improvements in technology have allowed for tremendous

development of GIS using complex computer systems. Along with the development of GIS was

the development of the idea that geography could become explanatory in nature as well as

descriptive (Harvey 1969). On the rise of GIS Martin (2005, p. 491) writes:

the surge of this activity has been notable in geography, but it has been adopted
substantially and simultaneously in the adj acent social and environmental sciences, in
which speed and accuracy of data arrangement and delivery are of significance, spatial
relationships increase the complexity of statistical analysis, and heterogeneous behavior
makes computer based modeling essential. (Martin 2005)

Many other Hields can benefit from geo-spatial analysis as well as sub-disciplines within

geography. Education as a social science can benefit from the use of GIS methodology.

Business Geography: With the understanding that geography can go beyond the initial

descriptive phase of reasoning, the use of geospatial technology began to present itself in

business and industry and the sub discipline of geography, business geography, developed.

Geographic analysis has been used by businesses, especially the retail sector, since the early

1990s. A pioneer member of the discipline, Dr. Grant Thrall, offers a comprehensive definition:

Business Geography integrates geographic analysis, reasoning, and technology for the
improvement of the business judgmental decision... This differentiates business geography
from the traditional descriptive or explanatory objective of economic and urban geography.
(Thrall 2002)

Dr. Thrall, professor of geography and expert in the Hield of applied and business

geography, describes geospatial reasoning as a hierarchy of steps that allow for improved

decision making. The Hyve steps include: Description, Explanation, Prediction, Judgment,

Management and Implementation (Thrall 1995). These steps can be applied to decision making

in a variety of fields in the private and public sector.

Effective use of the Hyve steps of geo-spatial reasoning require the application of "best

practice' methods. The first is assessment of the trade area of the business. The trade area can











4-15 Spatial pattern of lifestyle segments one and seven for students within Pinellas
County. ................. ...............73._._._......

4-16 Student capture rate percentages by ZIP code. ............. ...............74.....

4-17 Positive and negative Black student capture rate throughout the county. .........................75

4-18 Positive and negative Asian student capture rate throughout the county. .........................76

4-19 Positive and negative Hispanic student capture rate throughout the county. ....................77

4-20 Positive and negative White student capture rate throughout the county. .........................78

4-21 Male / Female student capture rate throughout the county .........__. ....... ._.. ...........79

4-22 Age group capture rate throughout the county. ............. ...............80.....

4-23 Geographic distribution of education throughout the county. ............. .....................8

4-24 Industrial influence in the county. ............. ...............82.....

4-25 SPC obj ectives examined within this study ................. ...............83..............

4-26 Summary observations and suggestions. ............. ...............84.....

















E sena lmF- ia..E Fc1 dre] 'rl*4cr Sii S~~ Exaimt..rai', suppri;F ] a ii .y.
I IBplatsrllr B-esi :.i Edw311~ SaBE' rgrarni r.= Cr IuRirrRE1

CI~~ritcal Acadiem Cntical to College coe acadenue munsson; d iretupalct on
Minson Areen Expansla cad progn~rams elaroons. uraching al Learmag: New
?rllr~~X~t~i Erpercla iimijbutes anneddditslly r idiretiv. at enhanted attest


achsmy. atgthleric p8rograms. ere.
Pol:Anrln Inpsa +,~ plourelrl Irs ior utrs adderesse I
Q441.r~ Oiroan-arien orgaJntrat*al~l~ alky ia


Priority '


Sler ur Nam


Larne Dest indon


HIsfpanic Student Capture fr SPC
Ilt rerienhPerceri atDIapanc$MLItseersuith
PCeroc W npW Rew leuy Lwee=9


corners.d~n
% ap1 8



I P a


Us rruard is? rthdli 3D hla~b.11 fry).IaC: kdtlm(!r 1
Ima~Fsn on C.'llfie oper:mell


De-.r Ale00-some


I


9. Explore ways to enhance systems and services at SPC that wi ll ad to
ant icreaspe i Hrspanic student enrollment. ( iiose I... _e :ol


Distilbution of Male/Ferale
Student Capfture for SPC
Differnce in the Percen at Famale Studntsr
versue th Male Students byZIPCgde


mrlas mahdnb, oneonharamp alard snow
Ble overwpresented are

Parenest beyami OU, lul

-Fm1BtB~ 17 866-% -


r~lpnrSuaanty mcal stdn lumcI .* 2)


UWS~~FU WYYVII LVL


0 0 1% 5%
5 01%e 10
S10 01% 15%
15o0196209b


:1 5 4 R E
~Vrr


Figure 4-25. SPC objectives examined within this study.


I


,,
,,











































Clearwaterg






purg


Figure 4-7. Drive time zones for SPC learning sites.


Bsa Hwarte


:Illlllrhll










rlrrrirllrr


I I
II r






~lrrF~~



IICIPl'lr Ir


IrS;rs~Tnr;









been identified recently through research will be discussed here. This section of the study will

include issues that can affect institutions throughout the country, but because of the geographic

location of the study that will be examined in the next chapter, issues specific to the state of

Florida will also be addressed.

Inequality

Inequality in higher education has always been an issue. Following World War II the

Montgomery GI Bill challenged this issue and resulted in the many minorities having the

opportunity to go to college. Although the GI Bill allowed minorities the opportunity to attend

college it did not have the effect of creating equality in tertiary education (Greenberg 1997).

Since the GI Bill took affect state and federal governments have instituted policies and

procedures such as the FAFSA (Free Application for Federal Student Aid), Pell Grant, race

based admissions, affirmative action in faculty hiring, and other means designed to assist the

financially unprepared and racially underrepresented (Lucas 1994). Equal opportunity to attend

college is still an issue of importance to the U. S. Department of Education and is addressed in a

recent report published by this department.

In September 2006 a commission formed by the Secretary of Education, Margret

Spellings, released a report discussing the current state of higher education (Spellings 2006).

The report describes the need for a "new landscape [that] demands innovation and flexibility

from the institutions that serve the nation's learners", thus implying that the current system is not

adequately meeting the needs of students. Currently students are getting their education from a

variety of sources, all of which should offer services that accommodate needs for students from

all backgrounds and categories. The report mentions the "consumer driven environment" and

describes students as results driven. Students are receiving education from multiple avenues and

institutions and forty percent of the 14 million undergraduates in the U. S. are attending two year









reports on funding and figures surrounding financial aid. Currently the average institutional

grant aid for families earning over $100,000 annually is nearly $4,000, which is lower than the

amount granted students from families earning $40,000 or less (Gerald 2006). Another finding

from this study similar to reports by the Spellings' commission was that talented high income

students were four times as likely to end up at a highly selective university than a low income

students of equal talent (Gerald 2006). Similar findings in reports from differing organizations

indicate that this trend is apparent regardless of source or data bias.

Another study conducted by the Educational Testing Service discusses three distinct

concerns that will cause considerable changes in the future of higher education (Kirsch 2007).

The three forces include divergent skill distributions, the changing economy, and demographic

trends. An indicator that America' s skills are greatly varied is embodied by surveys that show

that the U. S. has a degree of inequality (a representation of the gap between the least and most

capable) that is among the highest in OECD (Organization for Economic Cooperation and

Development) countries (Kirsch 2007). Signs that the nation's economy is dramatically

changing are also offered in the report America' s Perfect Storm. Since 1950 the proportion of

manufacturing j obs has dropped from 33.1% to 18.2% in 1989 and by 2003 down to 10.7%.

Twenty of the thirty million j obs that have been created since 1984 j obs associated with college

level education. Not only is the U. S. losing low skill manufacturing j obs, it is at the same time

gaining a high proportion of jobs that require advanced education (Kirsch 2007), evidence of the

increasing need for further the education level of the workforce. In 1979, the expected lifetime

earnings of a male with a bachelor' s degree was 51% higher than a male without a degree, and

by 2004 this estimate stands at 96%. Without offering equal opportunity for further education the

increasing gap in skills will only serve to promote economic disparities within the country and































IC'


SpidE~r Graph of LSP Seven Students
Stig~qrits hirulg We hi~gh hop 5 141lghborhroods gold Itair mvrep(e QImus


Spider Graph of~ LSP One Students
Studelis~ Irvirig In high? agiggry neighborhoods grrid MhirF rspeKp~ctw campul


~~~I


LT High Hopes:
Young households
strivi ng fo r th e
"American Dream'


L1 High Society:
Affluent, well-
eduLcated, m carried
couple
homeow~ners


.


r
IL~ Y r -
?i s~~;B", I


I

C


Figure 4-15. Spatial pattern of lifestyle segments one and seven for students within Pinellas County.











MEDIA BUYS:
Once the precise location of your customer base is known, you can make cost effective
media buys that get the right message to the right target. For example,
by using the knowledge gained from mapping their customer base, a marketer
who wants to place inserts in a local newspaper can target the neighborhoods
with the highest level of customers.

Competitive analysis:
Plotting the location of the competition (direct and indirect) on a map is much
easier to understand than a list of locations on a report. You can adjust your
marketing strategy to fit the number of competitors in the immediate geographic
area.

Drive-time analysis:
Mapping your customer drive times is useful for analyzing cannibalization issues,
new store placement and competition. By plotting the drive-time area on a map,
you clearly see any overlap among various business locations and competitive
sites. Barriers such as military bases, airports, parks or college campuses all
influence your customer's drive time, but those barriers are not apparent on
reports.

Market-entry p3lanning:
Using demographic mapping for market-entry planning clearly identifies sales
potential in a region. For market entry, be sure to look at retail sales potential,
lifestyle segments, propensity to use or purchase products and services,
population, income, age, number of businesses and competition. Using a
thematic map, your marketing strategy can be tailored to fit the new market's
unique characteristics.

Budgeting:
For comparisons, demographic mapping is invaluable. Analyzing maps for each
geographic area enables marketers to determine where to allocate their budgets.
Areas with high growth potential or high sales potential are quickly spotted and
marketing strategies can be adjusted accordingly.

Lifestyle segmentation:
Used in conjunction with demographic mapping, segmentation is a great tool for
identifying quality prospects. Lifestyle segmentation systems use demographic
and aggregated consumer demand data to classify every household in the United
States into a unique market segment. Each segment consists of households that
share similar interests, purchasing patterns, financial behavior and demand for
specific products and services.
Figure 2-1. Advantages of demographics mapping and possible further analysis (Picillo 1999).









ACKNOWLEDGMENTS

To acknowledge those that have influenced my development of learning, I would first like

to thank my parents for always being supportive, caring and there to keep me on the right track.

I would also like to thank my colleagues and classmates who I continue to learn from. Lastly I

thank Dr. Grant Thrall for guiding me through the master's program and allowing me to explore

the field of spatial science as well as opening the door for me to pursue a doctoral degree in

higher education administration.









or 2000 census data to visualize the changes in demographics and student locations over time

gives state or institutional planners better information for forecasting and future planning

(Pennington 2002).

Using GIS for management of programs is also advantageous for institutional researchers

and college planners. Identification of market segments and particular neighborhoods for college

expansion or reduction can result from the use of GIS for institutional planning. A better

understanding of service areas gives institutional researchers more intuitive capabilities of

providing services. For example, resources can be allocated for courses such as English as a

second language (ESL) to campuses near neighborhoods dominated by non-English speaking

families (Pennington 2002). Although infrequently used geospatial analysis and GIS serve to

maximize decision making in higher education institutions.

The adoption of geographic technology and methodologies to the business realm resulted

in the formation of business geography. Business geography is a sub discipline that focuses on

identifying the needs of business and tailoring business to the client. Based on commonalties

with the business world, including the need to provide services to a client and base decisions on

trade area needs, higher education can benefit from utilizing geographic technologies and

analyses. Understanding geographic characteristics can assist in evaluating institutional

obj ectives, and identify constraints on implementing these obj ectives. Geography is an integral

component of decision making in business today, and should be incorporated in the decision

making of public institutions, including education.










































Figure 4-2. One and three year objectives for SPC. The first table "Priority Description Table"
describes what constitutes a level 2 priority, while the "Obj ectives Table" describes
the obj ective. Both level 2 and are significant in this study.










be determined by evaluating the location of existing customers. Applications of GIS provide

visual evidence of the trade area and allow the user to conduct experiments and draw

relationships from information about the customer base and other geographic characteristics. An

illustration of questions geospatial reasoning and technology can address when applied in the

business sector include:

* What are the average drive times for customers from their home to the location of service?

* Are service providers drawing from each others' market area? (termed cannibalization in
business geography)

* Do certain market areas need greater focus on recruiting and/or advertising?

* Are services provided based on geographic demand and are services offered in the correct
locations?

* Are customers clustered by neighborhood, or uniformly dispersed around the trade area?

* Where are recommended sites for future expansion or service area reduction?

* Do customers seek out services from the nearest service provider or do they skip one outlet
for another? If so, why is one location preferred to another? (Thrall 2002)

The above questions are general and could be asked by various types of service providers.

This thesis will ask these questions with reference to trade area analysis for higher education

institutions. Through the use of GIS and methods proven to be successful in business geography,

higher education will be shown to more effectively serve all segments of the population.

Market penetration calculates how and where services of a particular business or service

provider are reaching prospective consumers. Evaluation of the underlying demographics of

potential customers of the service provider can assist in revealing relationships and

characteristics of the market that can increase the level of market penetration. Psychographic

profiles are also used in business geography to examine customers and the population of the

trade area.









CHAPTER 2
LINKING GEO-SPATIAL REASONING TO HIGHER EDUCATION: LITERATURE
REVIEW

This chapter is a literature survey of geospatial reasoning and technology applied to

decision making in higher education. Trends and themes used in business geography that can be

applied in institutional research will be discussed along with other applied geographic

techniques.

Also included in this chapter are significant functional foundations of institutional

research. To give relevance and understanding to this investigation an identification of methods

traditionally used for decision making, marketing students and managing enrollment will be

included.

Advantages of GIS in Marketing

GIS Techniques: The use of geography to assist in market analysis can be traced back to

the early to mid 20th century. Evolution within the field has occurred with the introduction of

new technologies, particularly the automobile, which significantly modified the development of

cities. This evolution of city development has expanded the discipline to studies on traffic flow

as well as social and behavioral characteristics. Within business and industry these studies were

supplemented with people's lifestyle preferences. The growth in capabilities of the modern

computers and the introduction of GIS (Geographic Information Systems) enabled many

disciplines to significantly enhance data analysis, storage, and display with the added spatial

component that is lacking from all other data analysis technologies. In the business sector the

use of geospatial technologies along with lifestyle segmentation profiles has greatly increased

our understanding of trade areas and the demographic composition of those trade areas (Thrall

2002). Table 2-1shows the benefits of using GIS for marketing (Piccillo 1999). Prior to the










Table 2-1. Summary of steps to find socioeconomic status from block groups
Summary of Steps to Find Socioeconomic Status from Block Groups
1. Acquire student addresses.
2. "Geocode" student addresses by converting them into latitude and longitude points.
3. Create GIS data containing student latitude and longitude points.
4. Acquire Census geodata at block group level.

5. Match, or geographically intersect, student data points to block groups.
6. Assign SES variables from Census data at the block group level to each student.



Table 2-2. Summary of steps to involved in cluster analysis
Summary of Steps to Involved in Cluster Analysis
1. Decide geographic level.
2. Select variables.
3. Standardize variables.
4. Choose cluster methodology, and if necessary, distance metric and linkage method.
5. Run cluster algorithm.
6. Indentify the number of clusters to be defined.
Table 2-1, Table 2-2. Crosta, P., Leinbach, Timothy, and Jenkins, David (2006). "Using Census
Data to Classify Community College Students by Socioeconomic Status and Community
Characteristics." Community College Research Center: Research Tools No. 1: 12.










Petersburg College is specifically addressed and analyzed for spatial patterns based on various

character stics.









St. Petersburg College Analysis


Focus of Analysis

This study evaluates the geo-demographics of St. Petersburg College (SPC) in order to

address specific and general questions for the benefit of the college. Geographically, SPC

effectively serves the entire Pinellas County with higher education. This analysis draws

attention to opportunities for SPC to increase market penetration within the county by greater

targeting of particular population segments which are identified in this report. Also, several

geographic areas warrant monitoring due to high population growth, and the services provided

are not increasing in proportion to the population change. These geographic areas are identified

within this report.

This analysis completed for SPC is broken down into five sections. The five sections are

described below.

1. Data A description where the data came from and what needed to be done prior to
analysis

2. SPC Obj ectives A discussion of some of the outlined one and three year obj ectives of the
college.

3. Trade Area Assessment An overall look at the trade area of SPC and some underlying
patterns and observations.

4. Market Penetration An assessment of the proportion of the population captured by the
college

5. Program Need A look at industry and education indicators that may affect SPC planning.

Data

The data tables for this analysis include the fall 2005 student enrollment for St. Petersburg

College, as well as current demographic data tables for Pinellas County. The student

information was provided to Dr. Grant Thrall by SPC. Prior to commencement of the analysis,

Dr. Thrall restructured this database to be in a GIS (geographic information systems) suitable









format. The restructured database includes age, sex, race, and credit earned by campus. Geo-

coding is the process of converting address data into spatial data that can then be used for GIS

analysis. The student addresses included in the database were geo-coded to enable geo-

demographic analysis. Once the addresses were geo-coded they were deleted along with any

other personally identifying information. The data table reports were provided by Professor

Thrall in the summer of 2006. Tables 4-2 and 4-3 show the distributions of the age and race of

the fall 2005 SPC students, tabulated by the students' home campus and by degree sought.

The demographic data for the Pinellas County area was also provided by Professor Thrall.

Through the use of a site license for ESRI's Business Analyst demographic data was obtained.

Using this technology all of the student data records were assigned lifestyle segmentation

profiles (LSP) which are used in the analysis. This analysis also integrates SPC objectives with

the previously documented data tables (Kuttler 2006).

SPC objectives: In the fall of 2006 Dr. Grant Thrall introduced the St. Petersburg project

to his seminar in business geography course. Collaboration with SPC's Vice President Stan

Vittetoe allowed for the refinement of proj ect obj ectives. Recently the college had been

experiencing a decline in enrollment and thus SPC was interested in the type of services

geospatial analysis could provide to the college in the form of consulting. A descriptive and

analytical geo-demographic consulting proj ect was undertaken in an effort to assist SPC.

During the preliminary stages of the research project a proposal was put together for

review by the college. The proposal was developed partially from ideas taken from the St.

Petersburg College 2006-2009 Strategic Directions and 2006-2007 Institutional Obj ectives

(Kuttler 2006). The objectives and strategic directions outlined by the college early in 2006 all

have assigned priority designations, with one being the highest priority and five being the lowest.









the county (Figure 4-17). The average capture rate is 5.4% throughout the county. Two ZIP

codes that fall into the MSPO classification are 33760 and 33762 and both capture roughly 3% of

the college age population. SPC captures a high of seven to eight percent in the two

northernmost ZIP codes. The analysis shows that the total numbers of students attending from

this area is small but relative to the small number of college age population in this area, SPC

does relatively well enrolling students from this area.

Higher education strives for racial equality and public institutions value equivalent

representation amongst the population. SPC's institutional objectives and strategic directions

illustrate the importance for improvement in ethnic representation at the college (Figure 4-2).

Capture rate analysis is performed to evaluate the success of SPC of enrolling an equivalent

ethnic representation throughout the SPC trade area. Capture rate analysis differs from simple

race percentages in that capture rates compare the percent of students' enrollment from each race

in a ZIP code to the percent of the actual population of each race in a ZIP code. Capture rate

analysis reveals if a race is under represented, or overrepresented. The traffic light style color

ramp used in Figures 4-18 through 4-21 does an effective job of displaying under-represented

(red) ZIP codes to those that are very close equal (neutral) and the over-represented areas

(green).

Figure 4-18 shows the college capture rate for black populations. Most of the ZIP codes

fall within the -2.5 to 2.5 classification. The histogram shows a high frequency of zip codes near

zero, which indicates an equivalent representation by SPC throughout the county. Figure 4-18

also shows the ZIP codes that are over-represented, and only two have 5-10% more blacks

students enrolling. One ZIP code, 33760, is identified as an MSPO because the ratio is

approximately -12% here.










3~~L3 3467L8

J46tf (Cj~
34677
L_ i
'jl!
~*!rll~
M698 ~


Figure 4-23. Geographic distribution of education throughout the county.


3j~~6 I

~E~ je


I


Tampa BaY










p.1). To initiate the program existing merit scholarships in the state were combined and funding

was provided by the Florida Lottery system. A summary of the requirements for high school

graduates is listed in Table 4-1. The Pappas group found that the program has spent 1.6 billion

dollars since 1997, when the program was instituted, and most recipients were non-need based

students. Opposition for the BFP can be readily found as some researchers protest the use of

state funds for students that do not need assistance.

One study conducted by researchers at North Florida University provides evidence that

indicate lottery-funded merit scholarships redistribute income from lower income, non-white,

and less educated households to higher income, white, well-educated households (Borg 2004).

Within the sample chosen by the research team the authors find that high socioeconomic (SES)

households receive a net program benefit from Bright Futures while low SES households incur a

net program loss. Numerous studies show that low-income households pay more in lottery taxes.

The researchers from North Florida found that these low-income households are much less likely

to receive a BF scholarship. They also found that of the low-income households that do qualify

for BF, they are more likely to receive the FMS scholarship that only pays 75% than high-

income households (Borg 2004). Programs such as the BF in Florida can and should be replaced

by programs that place as much emphasis on need as is on merit. Identifying the segments of the

market that are in need can be done through the use of GIS.

Planning can be greatly enhanced with the use of information systems such as GIS.

Considering the demographic changes proj ected for Florida over the next 10 to 15 years planning

in higher education is vital. The Atlas of the State University System of Florida (Thrall 2005)

exemplifies the merits of the use of geographic technology for planning in higher education. In

2004 The Florida Board of Governors (FBOG) requested consulting service from Dr. Grant










map with ZIP codes shown for Pinellas County with designated LSPs as well as LifeMode

groups. The maps presented to SPC provide administrators with a descriptive as well as

predictive view of the community's population demographics and the student' s demographic

consistency.

Another intended outcome of the proj ect is the designation of segments of the population

that can be greater served by the college. The market segments of potential opportunity (MSPO)

are segments used to show potential areas of improvement throughout the stages of analysis.

The MSPOs listed with the various maps and are primarily applied to ZIP codes and designate

segments of the population based on the criteria used to produce the map.

Limitations and Possible Improvements

The SPC study revealed a few notable points for improvement or change when this type of

study is conducted in the future. For example, the study was broad in nature and from the

beginning of the study there were few specific questions identified to answer using geographic

technology. The Vice President of Economic Development for the SPC was general in requests

for the study to decrease the rate of decline in enrollment. Narrowing the scope of the analysis to

more specific issues was accomplished by viewing the one and three year obj ectives published

by the college. The initial objectives could have been more structured to establish detailed

criteria for investigation rather than using objectives delineated by the college. A general

geodemographic analysis of the landscape provided an appropriate starting point. Future research

that incorporates education planning and GIS should aim to identify key questions and concerns

that can be used to drive further analyses of preliminary results.

Another consideration to further the validity of the such research is the appropriateness of

the demographic classifications for the education institute in question. The zip codes, though

useful for SPC postal marketing, were a course scale dataset that did not allow for detail with









USINTG GEOSPATIAL REASONING INT INSTITUTIONAL RESEARCH:
ST. PETERSBURG COLLEGE GEO-DEMOGRAPHIC ANALYSIS





















By

PHILLIP MORRIS


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

2007










higher female capture rate than males. This is especially true in the Southern part of the county.

No significant ZIP codes have a higher percentage of males than females attending SPC.

To analyze the distribution of student age across the county, students were partitioned into

three age groups (Figure 4-23). The first group, aged 20 and below, indicates a few MSPO ZIP

codes with relatively low, zero to seven percent, capture rates. The second age group 20-25 has

a similar distribution throughout the county with many of the same MSPOs identified. The third

age group 26-45, shows a capture rate of 0-7% throughout all ZIP codes in the county. Even

though the mean age for SPC students is 28, the analysis shows that none of the ZIP codes in the

county capture more than seven percent of the population in this age group.

Program Need

Program need was also analyzed. Program need is used to describe the current demands

for education throughout the trade area of the college. The community college has multiple roles

within the community and to best serve these roles an understanding of the types of services in

demand is important. This section of the analysis includes looking at education levels and

industrial influences throughout Pinellas County. This analysis is important for recognition of

where residents of the county have specific educational and program needs. Examination of

Figure 4-24 can provide SPC information on where residents of the county are lacking in

education and areas where residents have various levels of education as specified by degree

level .

Figure 4-25 shows the breakdown of the industries in Pinellas County as well as the

number of workers in that industry. Areas with larger circles represent a greater number of

workers in that industry and ZIP code. These areas could be targeted to provide industry specific

services. An understanding of the influences of industries in the county and where specifically








































I Rnrles


-4 *2-


White Student Capture fo~r SPC
Drelfrence in the Pelrent of White Students versus the
Percent of White Residents by Zip Code~


H~tc~r;wr


r50



la0n



sao



0 0 -
-20.0


-1S.0


'i
'~Lt


-10 ,a -s~o
% Diffaretac


MSPOs
33716 33708
3j714 33707
3-7 11



Difference:
%White Students
-%~Whitre Residents
S2 51 3 00
1 .01 2.50
14.99 -1.00
-9.99 -5.001
S-15.23 --10.00

I Jegall a numbers corre Eate to an u nder
representationr while positive numbers
point to over representation


jt Cllpuses
a 2 4 8


12 ,


Figure 4-20. Positive and negative White student capture rate throughout the county.









CHAPTER 4
GEO-DEMOGRAPHIC ANALYSIS

St. Petersburg College

This chapter presents the geospatial analysis of St. Petersburg College. Having introduced

the history and importance of both geography and institutional research and described how the

tenets of Business Geography can relate to current issues higher education, this chapter will

discuss the core analysis of this research. The study exemplifies the use of GIS in education

planning. The chapter is organized as follows: first, information about the organization, history

and mission of the college will be discussed. Secondly, the analysis completed for the college

will be detailed along with an extensive appendix of figures and maps. Lastly the summary

findings for the study will be listed and discussed.

Organizational Structure

St. Petersburg College (SPC) is located in Pinellas County, Florida and offers services

throughout the county from various campuses and service locations. Geographically, Pinellas

County is a peninsula with the Gulf of Mexico to the west and Tampa Bay to the east.

SPC is broken down into eleven learning sites throughout the county. The term learning

site is used because not all of the sites are considered campuses. According to Stan Vittetoe, the

VP of Business Operations for St. Petersburg College, a campus must offer services that include

student services, libraries, counseling, bookstore, ect. (Phone interview with Stan Vittitoe, Vice

President of Economic Development, on October 9th, 2006) Four of the eleven learning sites are

considered campuses. The four campuses include St. Pete Gibbs, Tarpon Springs, Clearwater,

and Seminole along with the rest of the learning sites can be seen in Figure 4-1.

The administrative organization the college has a Board of Trustees that is a political sub-

division of the state and part of the state community college system. The Board governs the










LIST OF FIGURES


Figure page

1-1 Map of the cholera-infected area in Westminster region of London, England in 1854.....17

2-1 Advantages of demographics mapping and possible further analysis. ............. ................29

3-1 Summary Overview of The Atlas of the State University System of Florida. ................40

3-2 Summary Overview of The Atlas of the State University System of Florida. Nursing
degrees granted in Florida in the 2002-2003 school year. ................ ..................4

3-3 Summary Overview of The Atlas of the State University System of Florida.
Summarized conclusions for the The Atlas of the State University System of Florida ...42

4-1 Map of SPC campuses and learning sites in Pinellas County.............__ ........._ ......58

4-2 One and three year obj ectives for SPC.. .........._ ......... ...............59

4-3 Geographic dispersion of SPC college students throughout Florida. ............. ................ 60

4-4 Geo-coded address of SPC students living within Pinellas county aggregated to a
grid of 1.5KM square cells. ........._.._.. .......... ...............61..

4-5a Tarpon Springs and Clearwater campus geographic distribution of student
enroll m ent. ............. ...............62.....

4-5b Seminole and Gibbs campus geographic distribution of student enrollment. ...................63

4-6 Online student geographic distribution. .............. ...............64....

4-7 Drive time zones for SPC learning sites. ............. ...............65.....

4-8 Estimated population growth of the college age population within the county ........._......66

4-9 High school age percentage of change proj section. ........._.._.. ........._.._......._.. ...67

4-10 Initial consideration of the northeast corner of the county. ........._.._.. ........_.._.........68

4-11 Regional population outlook. ..........._ ..... ..__ ...............69...

4-12 Lifestyle segmentation patterns throughout the county. .................. ................7

4-13 Lifestyle group patterns throughout the county. .............. ...............71....

4-14 Spatial pattern of lifestyle segments five and ten for students within Pinellas County.....72









hence result in other social issues related to educational inequality (as discussed further in

(Chakravorty 2006)).

The demographic trends mentioned in the The Perfect Storm are also significant. By the

year 2030 the Hispanic population will exceed 20 percent of the total population. The authors of

this article cite The American Community Surveys as reporting that in 2004 approximately 57%

of the 16-64 year old Hispanic population in the U. S. is foreign born and around half of these

immigrants do not have a high school diploma.

The inability to close the existing skills gap and significantly enhance the literacy levels of

all Americans will result in demographic changes that leave the population in 2030 with "tens of

millions of adults unable to meet the requirements of a new economy" (Kirsch 2007). If these

tens of millions of low income people are going to narrow the skills gap and improve their

literacy levels enough to survive they will need the help of institutions such as the community

college.

Through examination of recent studies on higher education and trends that affect higher

education a pattern has been identified. An increase in the number of citizens in need of basic

skills, along with a trend in higher education for academic elitism has created a gap between

prepared and under-prepared, and this gap is yet widening. Efforts from the premier state

universities to elevate standings at the expense of equality of access have created a trend that

flows down the hierarchy of higher education. This drive for prestige has caused institutions to

abandon traditional values and roles in the community for a better name amongst peer

institutions. This effect, deemed mission creep, effects even community colleges and

community colleges and workforce training institutions have the highest potential for decreasing

the skills gap mentioned above (Campbell 2005).





I ~t
I


Estimatedi Increase iin
Popu latiion



61 -120

r 1 121 180

181 -260

26'1 430


MBPos
asse8 uses


SCampuses
0 2 4


I


8 '12 16
Miles


Figure 4-8. Estimated population growth of the college age population within the county.


Estimated Population Gro~wth 2010
Five Year Estimalted Growth of
College Age Population (Ages 15-45)


Hi gihest
estiim~atedi
growth in the
North
















~a


;LP


Number of Students
Top Four C~ampuses
i 10 to 224 (45)
I 57 to 109 (92)
O 23 to 57 (91)
O 4 to~ 23 (9;3)
O O to 4 r25zl


Figure 4-26. Summary observations and suggestions.


b
d .I
"*r g
F;A ,


51~rlrll i'L-T~ ~1~Psl ~5















































Figure 1-1. Map of the cholera infected area in Westminster region of London, England in 1854.
(Retrieved on March 5th from http ://en.wikipedia.org/wiki/Image: Snow-cholera-map-
1.jpg. This image is in the public domain because its copyright has expired in the
United States and those countries with a copyright term of life of the author plus 100
years or less)












































Figure 3-2. Summary Overview of The Atlas of the State University System of Florida. This map shows the supply of nursing degrees granted in
Florida in 2002-2003 school year. The shaded areas are major SUS trade areas. The darker shaded regions represent a higher projected
rate of change for the population age group 60 and over. (Thrall, G. 2005).


Niursiing Degreeps (MaIp 3.6a)



Th~e map depicts thep bacheolor--
derees~ in nursing reportedly
for 2002ir-2003. The map aIlso
shows by St S tr~ade area the
projected change in -the 60+ nluraninDgueren numerng e~gaa.
Atai nlrivers~ity Bystem Frame M.1lProfit
populnf ionICI in 2010).

l.a .. g


Total Nu(l sing Bachelor Degrees .um

State uiver~sirl System 126Nursing Deglrsee

Private, Not-forr-Profit 275 c, ,'

Private, Forr-Pro~fit 80 !.. x










LIST OF TABLES


Table page

1-1 Winning the site selection race ........... ..... .___ ...............18...

2-1 Summary of steps to find socioeconomic status from block groups .............. .................30

2-2 Summary of steps to involved in cluster analysis............... ...............30

3-1 List of required qualifications for Bright Futures scholarship ........_.._.. ... ......_.._.. .....43

4-1 Cumulative age distribution of SPC students within Pinellas County ........._...... ..............85

4-2 Fall 2005 distribution of student race by campus. ............. ...............86.....

4-3 2005 collegewide opening fall headcount enrollment by age and by division. .................87









LIST OF REFERENCES

Applebaum, W. (1966). Methods for determining store tradert~t~rt~t~rt~t~rt~ areas, marketing penetration, and
potential sales. Journal of Marketing Research 3: 127-141.

Bailey, T., & Morest, Vanessa, Ed. (2006). Defending the community college equity agenda in
the twenty first century. Baltimore, MD: The Johns Hopkins University Press.

Bolstad, P. (2002). GIS fundamentals:~dd~~~dd~~dd A first text on geographic information systems. White
Bear Lake, MN: Eider Press.

Borg, M. O., & Stranahan, Harriet A. (2004). Some futures are brighter than others: the net
benefits received by Florida bright futures scholarship recipients. Public Finance Review
32(1): 105-126.

Brawer, F., & Cohen, Arthur (2003). The American Community College. San Francisco, CA:
John Wiley and Sons Publishing.

Burrough, P., & McDonnell, R. (1998). Principles of Geoguraphical Information Systems_New
York: Oxford University Press.

Campbell, D. (2005). Florida: bdh11I Ithrll of the jitture for the community college baccalaureate ?
Community College Weekly. 17: 5.

Chakravorty, S. (2006). Fragments of inequality: Social, spatial, and evolutionary analyses of
income distribution. New York: Routledge.

Clark, B. (1990). Higher education American style: A structural model for the world.
Educational Record (Fall 1990): 41-44.

Crosta, P., Leinbach, T., & Jenkins, D. (2006). Using census data to cla~ssify/ community college
students by socioeconomic status and community characteristics. Community College
Research Center: Research Tools No. 1: 12.

Donhardt, G., & Keel, D. (2001). The analytical data warehouse: Empowering institutional
decision makers. Educause Quarterly 24(4): 56-58.

ESRI (2006). Community Tapestry, Environmental Science Research Institute. Retrieved
November 3, 2006, from http://www.esri. com/library/brochures/pdfs/community-tapety
handbook.pdf

Floyd, D., Skolnik, M., & Walker, K., Ed. (2005). The community college baccalaureate:
Emerging trends and policy issues. Sterling, VA: Stylus Publications.

Gerald, D., & Haycock, K., (2006). Engines of inequality: Diminishing equity in the nation's
premier public universities. Washington, DC, The Education Trust: 25.








Distribution of Students Attending
Clearwater by Zip Code


Distribution of Students At~tending
Tarrporn Springs bsy Zip Code


.~ To


SCampuses


# of Studenrts
3-24


# of Students


20-129


It~T"


51L


Figure 4-5a. Tarpon Springs and Clearwater campus geographic distribution of student enrollment. The students used to produce
these maps are taking most of their classes at that particular campus.










Psychographic profiles are a compilation of an individual's attributes relating to

personality, values, attitudes, interests, and lifestyles. In business geography these are used to

explain market forces and predict and judge current and future business or real estate

undertakings (Thrall 2002). Neighborhoods and individual households can be organized into

psychographic profiles and lifestyle groups. This type of analysis has proven to be effective in

predicting and understanding consumer behavior as well as guiding decision making for retail

and other business operations. ESRI' s Business Analyst and MapInfo' s TargetPro are leading

GIS software programs that use psychographic analysis for business applications.

Another focus of business geography is analysis of site location. Thompson Associates,

partners with the successful geospatial analysis software company MapInfo, have delineated ten

crucial steps for site selection (Table 1-1). Another effective method for site selection is the use

recorded data surrounding successful service locations in order to draw analogies for possible

future locations. This method is known as the analog method within business geography

(Applebaum 1966). The use of these site selection methods help regulate resource allocation,

provide intuitive learning site location strategy, and assist in providing appropriate services in

relevant locations.

Businesses have greatly benefited by having adopted geospatial technology and reasoning.

However, other institutions of our society have lagged behind adopting business geography

practices. Higher education institutes can benefit from the methods mentioned above.

Institutional Research

In any organization communication, research, and accountability are important for

effective leadership and efficient operation. Higher Education is no exception. Higher education

offices of institutional research provide these important functions within universities and

colleges. A simple and effective definition of institutional research is "research conducted












Table 1-1. Winning the site selection race
The following 10 items represent a proven framework for assessing site selection opportunities:
1. Collect demographic data
2. Build an inventory of competition
3. Generate market characteristics
4. Quantify market demand for financial products and services
5. Understand customer account information
6. Recognize customer behavior patterns
7. Identify potential site location opportunities
8. Conduct fieldwork
9. Identify site characteristics
10.Develop final recommendations for senior management
(Thompson 2005)






























5 as5 s 10 is 201


IRelative toth e
-small numberof
college age!
population in this
area, SP~does a
good job of


<
-
. d


Enrollment has the
potential to increase.


Figure 4-16. Student capture rate percentages by ZIP code.


Student Capture Rate for SPC

Percentage of College Age Population
Captured By SPC

W 2 5% 4%
/ 5%

| 5% -6%
i 16% -7%
S 7% -8%


MJSPos
as760 sa762
33715


~ a~4
~I


~iri
5jt









An integral component of the community college mission is providing unrestricted access

for citizens within the community (Brawer 2003). Access is directly rated to the cost of

transportation and increased distance equals increased costs for students. Because contemporary

transportation effectively requires commuting by car, analysis of drive times to each campus is

important for mitigating the costs of transportation and improving access. Figure 4-7 shows

drive time zones around each SPC learning site or campus. Each of the irregular polygons shows

the distance a person could drive in seven minutes from the campus. The software used for this

graph allows the user to determine the drive time and the speed of travel before creating the

polygon. Once the zones were created the number of students living within these seven minute

zones were tabulated. Approximately 35% of the total Pinellas County enrollment lives within

seven minutes of a learning site or campus. Seven minute zones were used here because more

than seven minutes would create overlapping zones that would be less visually effective difficult

to analyze. If the college were to continue with this reasoning, prospective new campuses would

be located at Palm Harbor, Dunedin, or Western Clearwater.

According to the population growth estimates included in the database provided by ESRI' s

business analyst each of the ZIP codes of Pinellas County will see an average increase of 144

people between the ages of 15-45 by the year 2010. Approximately 92% of SPC students fall

into the age group 15-45 (Table 3-1). Some ZIP codes will see an increase of up to 430 people

of the targeted group, college age population. The greatest estimated growth will be in the

northern part of the county served by the Tarpon Springs campus (Figure 4-9). Throughout the

analysis Market Segments of Potential Opportunity (MSPOs) are identified. Figure 4-9 has a

reference to zip codes that are designated in this reference as MSPO. The color ramp used in










making methods. Administrative functions in enrollment management, assessment, marketing,

and institutional obj ectives will be examined along with traditional methods of decision making

for these functions will be discussed.

Enrollment Management

The Association of Institutional Research (AIR) is an excellent source for information and

literature describing IR. A recent publication from the AIR, The Primer for Institutional

Research (Floyd 2005), aptly and comprehensively describes the key functions and duties related

to IR.

The Primer for Institutional Research analyses and reports on the current issues

surrounding institutional research in higher education (Knight 2003). The book is a compilation

of nine chapters written by practitioners within the field of institutional research. Each of the

chapters addresses a separate issue. According to the editor this book is "designed to provide an

introduction to some of the more common institutional research issues, methods, and resources

for newcomers and to provide means for veterans to update their capabilities." This volume is in

essence an update to previous publications of the same variety from the Association of

Institutional Research in Tallahassee, FL. One of the chapters of the The Primer for Institutional

Research is titled Enrollment Management.

Enrollment management is defined in The Primer for Institutional Research as "an

institutional research function that examines, and seeks to manage, the flow of students to,

through, and from the college." The educational pipeline is a term used to describe student

recruitment processes. The questions "Who does the institution want to educate?" and "Who is

available?" determine the educational pipeline. Once these questions are addressed "the next

step is to identify data sources that summarize how many pre-college students with these

characteristics exist in the pipeline" (Knight 2003). The use of geospatial techniques can greatly









According to the analysis the Asian populations are also proportionally represented at the

college (Figure 4-19). The ratio scale in Figure 4-19 varies from -1.5% to only 1.5% so

throughout all zip codes the Asian population is enrolling equally. Although there is little

variance for the Asian population Figure 4-19 does show the areas for improvement within

Pinellas County.

The Hispanic population analysis reveals four MSPOs (Figure 4-20). These four ZIP

codes fall within the -4% to -5.5% range. Overall the county has 19 orange and red zip codes

which indicate a limited proportion of Hispanics in total enrollment. Figure 4-20 can give

administrators at SPC an idea of where to emphasize effort for increasing Hispanic enrollment.

The last racial group analyzed was the White race (Figure 4-21). Of the four groups

examined, Whites are the least proportionally represented at SPC. The majority of the ZIP codes

are red and orange and a few zip codes (MSPO) are in the -10% to -15% range. One notable

observation is that the only ZIP code, 33760, that has a greater proportion of whites enrolling at

SPC than are living within that ZIP code is also the ZIP code that had the worst representation

amongst the black population. This is an issue that warrants further examination by institutional

researchers at SPC.

Figures 4-18 through 4-21 reveal that White and Hispanic populations were more extreme

in rates of capture compared to Blacks and Asians. SPC is doing a good j ob in capturing the

minority population when compared to each proportional representation of the population by ZIP

code. The distribution for the minority capture rate stays fairly close to zero.

The female capture rate throughout the county is an average of about 10% higher than

males (Figure 4-22). There are various ZIP codes within Pinellas County that have a 15-20%



























AllS~chools_Flonlda by MAIN_CAMPUS
* AC (342)
o CL (4.569)
* CT (27)
* DT (574)
o EC (50;67)
o EP (60)
* HEC (1193)
* MT (86)
* SE (1235)
* SPG (5787)
* TS (3068)


Figure 4-3. Geographic dispersion of SPC college students throughout Florida. Each dot
represents the geo-coded address of a SPC student enrolled at SPC in Fall 2005.
Yellow dots represent online learning as maj or campus designation. Distant non-
yellow dots likely represent true home locations, whereas the student' s campus
address is more proximate to St. Petersburg. The color of the dot represents the
campuses or learning site the student takes the most credit hours from. The number in
parenthesis represents the number of students taking a maj ority of their credits from
that particular campus.









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

USINTG GEOSPATIAL REASONINTG INT INSTITUTIONAL RESEARCH:
ST. PETERSBURG COLLEGE GEO-DEMOGRAPHIC ANALYSIS

By

Phillip Morris

August 2007

Chair: Grant Thrall
Major: Geography

Geographic analysis has been adopted by businesses, especially the retail sector, since the

early 1990s. Higher education can receive the same benefits as have businesses by adopting

business geography analysis and technology. The commonality between business geography and

institutional research for higher education is that both have trade areas, both provide services to

clients (students), and clients can be geographically identified by their addresses as well as their

psychographic profile. Among the valuable information that institutions of higher education can

create using business geography are psychographic profiles of the student body, commuting

patterns, and potential enrollment based upon the underlying demographics of the institution' s

trade area. A benefit of this analysis is the ability to anticipate the needs of the market.

Understanding these geographic characteristics can assist in evaluating institutional obj ectives,

and identify constraints on implementing these obj ectives.

My research is intended to provide a general guideline to geospatial reasoning in

institutional research, assessment, and evaluation. Through literary examination as well as

through the use actual data from a community college, the benefits of geospatial reasoning in

institutional research will be identified. Within this report the student population of St.









CHAPTER 1
GEOGRAPHY AND HIGHER EDUCATION: AN INTRODUCTION INTO GEOSPATIAL
PRACTICE AND INSTITUTIONAL RESEARCH

Foundations of Geospatial Reasoning

Geography has assisted in decision making throughout development of modern societies.

Early explorers relied on their knowledge of geography and landscape to survive and succeed in

global endeavors. Visual representations of the earth at various scales have been integral in

planning and problem solving throughout history. The understanding of the geography of the

earth has developed into many sub-disciplines of geography and other scientific disciplines,

some of which can be universal tools for decision making. Cartography, planning, remote

sensing, geographic information science, and even epidemiology use maps and spatial

information to solve problems and make advancements within society.

In 1854 Dr. John Snow, sometimes referred to as the father epidemiology, used geospatial

reasoning to discover the source of a cholera epidemic in a neighborhood of the Westminster

area of London, England (Kovalerchuk 2005). By mapping the infected individuals (Figure 1-1)

Dr. Snow was able to discover a direct correlation between the infection and the use of a single

water pump proximate to the clustering of the infections.

This type of spatial reasoning has led to development of Geographic Information Systems

(GIS). GIS can be defined as "a powerful set of tools for storing and retrieving at will,

transforming and displaying spatial data from the real world for a particular set of purposes."

(Burrough 1998). GIS is commonly described as having 5 main components; data, hardware,

software, procedures/goal s/plan, people, and a network. These components help describe the

what and where of particular features on the earth' s surface. GIS specializes in storing,

analyzing, compartmentalizing, and describing the properties and attributes of a particular

landmark or geographic occurrence (Bolstad 2002)









Higher Education in Florida

The study presented with this paper and discussed in the following chapter, deals with geo-

demographic analysis conducted for St. Petersburg College in Pinellas County, Florida. Giving

consideration to the study, it is pertinent to discuss issues in higher education that specifically

affect the state of Florida. Current issues along with a discussion of the implication of the use

of GIS will comprise this section of the paper.

The changing demographics and increasing population for the state of Florida is a concern

for the Florida Board of Governors (FBOG) for the State University System. In the last decade

the growth of population of 18 24 year olds has grown in Florida by 24.6% and will see an

increase in 19.5% (roughly 10.5% more than national proj sections) from 2004 to 2014 (Pappas

2007). This spike in college age population requires additional needs for access to higher

education. The Florida Board of Governors recognized this and hired the Pappas Consulting

Group to assist in planning for the future needs of college students within the state. The

consulting group concluded investigation and published a detailed report in January 2007.

Recommendations from the group included the use of the California model of the 3 tiered

system with clear delineation between research one, state college, and community college

systems. According to the consulting group the use of state resources for the additional openings

of medical schools, the additional emphasis of state universities on research and the granting of

baccalaureate degrees from community colleges constitutes "mission leap" (Pappas 2007).

Designation of clear mission parameters will limit the unnecessary use of valuable and

diminishing state funds.

Programs such as the Bright Futures Program (BFP) are also addressed by the consulting

group. Bright Futures was established to "reward any Florida high school graduate who merits

recognition of high academic achievement" (Florida Bright Futures Scholarship Program, 1997,

























orup.




geC



projea1 .led.


High School Age Gr
,. Peceintage of Chan
Le ~2005 -2010
I'stanar s based on Ie nulbi
p~T'- ~ I k&er np rqw age tS1*.d Ithe

., \ Rate ofChange
Patel unpubhshed
-14.99% -6%;


Figure 4-10. Initial consideration of the northeast corner of the county.









Another study recently completed by the Western Interstate Commission for Higher

Education (WICHE) discusses the financial aid and student success (Hauptman 2007). This

report again points to access as a maj or issue for higher education. This report by WICHE

makes some recommendations for improving access. One key recommendation related to this

paper is the recommendation for improvement in "collection, analysis, and presentation of data

on how well federal and state support of policies are targeted toward low income

students"(Hauptman 2007). The report goes on to discuss how currently aid programs are not

well targeted toward the poor. According to the commission, "this lack of targeting reinforces

chronic inequalities at each stage of the educational pipeline" (Hauptman 2007). Addressing

policy design and implementation, the author finds that availability of data, research, and

insightful analysis is limited prior to legislative decision making and program implementation.

GIS provides the tools that have insightfully informative analysis and presentation capabilities.

State and Federal education agencies stand to improve decision making through the use of

geospatial analysis.

Prior to descriptive analysis the commission identifies three obj ectives, which are listed

below:

* Strengthen educational opportunities for students through expanded access to programs.

* Assist policy makers in dealing with higher education and human resource issues through
research and analysis.

* Foster cooperative planning, especially that which targets the sharing of resources.

These three obj ectives are worth mention here because the topic of this paper, using

geographic techniques to improve education decision making, is related. These objectives along

with the commission's recommendations can be affected and improved upon through the use of

GIS for targeting segments of the population that are underserved.










Table 3-1. Florida Bright Futures award requirements.


(Florida Bright Futures Scholarship Program, 1997, p.1)


Award Award Level *GPA


Required Credits Community Service Test Scores


75 hours to be
approved by the
college


No requirement



No requirement


100% tuition
FAS
and fees


75% tuition
FMS
and fees

75% tuition
GSV
and fees


15 various college
prep courses


15 various college
prep courses (same)

15 various college
prep courses (same)


1270 SAT
28 ACT


970 SAT
20 ACT

840 SAT









Once the geographic level has been established the variables are clustered together based

on similarities and a set number of clusters, or lifestyle segments are produced and one LSP is

assigned to each geographic area. Once customer addresses are geocoded they can be appended

to an LSP. These techniques will be illustrated and expanded upon in chapter four.

Spatial Information in Education

Support for GIS in Higher Education: Presently, there is a limited supply of resources

that document research using GIS in institutional research. However, those that contribute

significantly to the field oflIR will be discussed in this segment of the paper. Topics that relate

to the use of GIS in education will also be discussed. One such topic is equity in the

administration of financial aid.

A problematic issue for college administrators striving for financial aid equity is

determination of student socioeconomic status (SES). SES is currently difficult to measure for

college students (Bailey 2006). Unless a student files for financial aid administrators have found

no easy way to determine his or her status, which determines if the student needs to be awarded a

grant or other funding. However this problem can be addressed using GIS. Using the student' s

address and what can be determined from neighborhood analysis GIS researchers can determine

a students' SES (Crosta 2006).

The Community College Research Center (CCRC), regarded as one of the leaders in

research on community colleges, is based at Columbia University in New York. Recently a

study was conducted by the CCRC that details the use of student addresses to determine

socioeconomic status (Crosta 2006). The study was conducted using data from the Washington

state Board of Community and Technical Colleges. The goal of the study was to provide the

board with accurate information on student SES for the purpose of making informed policy

decisions and improve service to residents of the state through the various community and












Table 4-2. Fall 2005 distribution of student race by campus.

Campus White Black Hispanic Asian Am Indian Unknown Total


Clearw~ter


St PetersburgiGibbs


Sennnole


Tarpon Springs


Lower Division Total

UPPER DI1TSION BY' PROGRAM
Dental HIerlene


Educatbion


International Bus~iness


Nllring


Orthotics &R Prosthetics


Public Safety


Technology Managemet


Veterinary Technology


EPyImpct


Non-Degree Seeking


Upper Division Tota

COLLEGE TOTAL


(SPC Factbook 2005)


5219 514 492


247 41 211 6;7Z4


77.646 7.16%

6674 1802
67.9% 13..3 ,

2324 102
84.7% 3.7%

3462 103
Sil%~ 2.6%

17679 2521


84 6
78.5% 5.16%

425 41
86.4%4 8.3%

35 4
72.9% 8.3%

174 30
78.7% 13.16%

18 0
73.3%4 0.0%

54 6
81.846 9.1%

304 44
710%~l 11.1% 8

48 0
90.6% 0.0%

50 1
90.9%4 I.8%

49 0
90.7% 0.0%

1241 132

18920 21653


7.3%

514
3.2% ,

121
4.-1* e

190
4.8%

1317


9
8.4%

11
2.2%

5
10.4%

9
4.1%

2
S."**O

3
4.5**

25
6.3' o

1
I.9%

2
3.6% i

0
0.0%

167

1384


3.7%


4.0%

52
1.9%

75
1.9%

768


4
3.7%

5
1.0%

4
8.3%

5
2.3%

3
13.0%

0
0.00 a

15
3.8%

1
1.9%

1
1.8%

1
1.9%

39

807


0.6%

5;6
0.6%

15
0.5%

12
0.3%




D
0.0%

4
0.8% p

0
0.0%

1
0.5%

0
0.0%

1
1.5%

0
0.0%

1
1.9%

D
0.0%

1
1.9%

8

132


3. 1%

3&2
3.9%

130
4.7%i

132
3.3%$

855


4
3.7%

6
1.2%

0
0.0%

2
0.9%

0
0.0%

2
J.0%

7
1.8%

2
3.8%i

I
1.8%

3
5.6%

27

882


9822


2744


3974


23264


107


492


4B


221


23


66d


395


53


55


54


1514

24778









within an institution of higher education in order to provide information which supports

planning, policy formation, and decision making"(Saupe 1981). The office of institutional

research provides important information for academic administrators, accrediting agencies and

governmental officials. This information allows individuals within these organizations to make

informed decisions.

Linking Geography to Higher Education. Higher education can receive the same

benefits as have businesses by adopting business geography analysis and technology. The

commonality is that both have trade areas, both provides services to clients (students), and clients

can be geographically identified by their addresses. Among the valuable information that

institutions of higher education can create using business geography are psychographic profiles

of the student body, commuting patterns, and potential enrollment based upon the underlying

demographics of the institution's trade area. A benefit of this analysis is the ability to anticipate

the needs of the market. Understanding these geographic characteristics can assist in evaluating

institutional obj ectives, and identify constraints on implementing these objectives.

My research is intended to provide a general guideline to geospatial institutional research,

assessment and evaluation. Through relevant literary examination as well as through the use of

data from a community college the benefits of geospatial reasoning in institutional research can

be identified.












Black Student Capture for SPC *~
Difference in the Percent of Bfack Students versus thre
Percent of Black Residents by Zip Code


MSP~s _
33760




Difference:
%GBlack Students
ul I % Black Residents
5.01 -10.00
S2.5"1- 5.003so~
1 -2.49 -2.50
S-9_99 -2 50 c
-11 68 --10.00 meji
-15.0O -1110 -5_D 00 !it 10.0'
%DClfferane erG
Neg~ativre nu mbe rs coure ate tor an u nder ~Cmue
reprelsentationt while positive numbers Cmpse
point to over rep resentation 1 2 1


Figure 4-17. Positive and negative Black student capture rate throughout the county.









Thrall for analysis of the geographic access to the state university system (SUS). Thrall's

analysis provides immense planning potential for the SUS, and can be duplicated by other

university systems elsewhere. Thrall's (2005) report uses a variety of methods for delineating

trade areas for the universities in the SUS and addresses both the needs for key fields of

education around the state and the supply of these degree programs and graduates with these

degrees. Examples of the work provided to the FBOG (Florida Board of Governors) can be seen

in Tables 3-1, -2, and -3. This analysis provides further evidence of the utility in the use of GIS

in planning for higher education at the highest level.

Chapter Conclusion

The important role has been established for geospatial analysis at the university system

level and at the community college level, therefore justification is given to the importance of

effective resource allocation. Workforce development programs and other fundamentals

characteristics of the community college are very important for advancement of society as a

whole. This paper provides evidence for use of geo-spatial applications in improving the

administration of these fundamental operations in the community college and other realms of

higher education. By examining the current issues in higher education a gap has been identified

that geo-spatial science and techniques can begin to fill.

Using data and collaboration with St. Petersburg College in St. Petersburg, FL, the

following chapter will illustrate specific benefits of using geospatial reasoning for planning,

implementation of institutional obj ectives, and enrollment management.









these industries are located can give SPC management an idea of what programs and courses are

in demand.

Summary Findings

Based on SPC's obj ectives (Figure 4-2), my analysis has revealed locations within Pinellas

County where SPC can put marketing and recruiting emphasis (Figure 4-26). Overall

conclusions and revealed market segments of potential opportunity are shown in Figure 4-27.

The northeast part of the county will have high growth and should be monitored for

neighborhood change for timing of a high visibility information center. Demographics in this

area are well suited for a community college.

SPC should monitor areas in the central county that have a low minority capture rate.

Drive time analysis shows that prospective new campuses or learning sites could be located at

Palm Harbor, Dunedin, and Western Clearwater to improve accessibility. An industrial centered

campus might be considered as an intervening opportunity for workers in commercial areas in

the south central area of the county. Several market segments have been identified by ZIP code

according to age, race, gender, capture rate, and maj or industry that have promise for potential

enrollment (Figure 4-26). The following chapter discusses the implications for this study along

with possible recommendations for further analysis.










Saupe, J. (1981). The functions of institutional research. Tallahassee, FL: The Association of
Institutional Research.

St. Petersburg College (2006). 2006 St. Petersburg College Student Handbook.

Spellings, M. (2006). A Test of leadership: Charting the future of U.S. higher education. A
report of the commission appointed by secretary of education Margret Spellings.
Washington DC: U. S. Department of Education.

Teodorescu, D., Ed. (2003). Using geographic information systems in institutional research, New
directions for institutional research, San Francisco, CA: Jossey-Bass.

Terenzini, P. (1993). On the nature of institutional research and knowledge and skills it requires.
The Journal ofResearch in Higher Education 34(1): 1-10.

Thompson, A. (2005). Winning the site selection race: White paper. Ann Arbor, MI: MapInfo /
Thompson 1-7. Retrieved March 5th, 2006 from www.mapinfo.com.

Thrall, G. (1995). The stages of GIS reasoning GeolInfo Systems. 5: 46-51.

Thrall, G. (2002). Business geography and new real estate market analysis. NY: Oxford
University Press.

Thrall, G. (2005). Summary overview of the atlas of the state university system of Florida.
Gainesville, FL: Florida Board of Governors: 19. Retrieved on January 4th, from
http://www.clas.ufl.edu/users/thrall/fbog/neht

Thrall, G., & Mecoli, N. (2003). Spatial analysis, political support, and higher education funding,
GeoSpatial Sohutions 13(7): 44-47.

Walleri, D. (2003). The role of institutional research in the comprehensive community college.
Journal ofApplied Research in Conanunity College 11(1): 49-56.










Figure 4-9 is also used at various times in this analysis. The traffic light pattern of green and red

was chosen to designate the highs and lows, or hot and cold areas of the county.

According to Dr. Carol Weideman, Director of Institutional Research at St. Petersburg

College, there has been a gradual decrease in high school age population enrollment at SPC

(Interview with Carol Weideman on March 6th, 2007). Figure 4-10 shows a high school age

population proj section from 2005-2010 for Pinellas County. These areas can be targeted for

recruitment to perhaps reverse the trend recently experienced by SPC. The percentage scale of

proj ected change for the county's high school population is -25% to +30% broken down by zip

codes. By multiplying the numbers of current population in the age group 10-15 by the proj ected

growth per ZIP code the proj sections were developed. The northern part of the county will have

the greatest growth of high school aged population, while the central county is proj ected to have

decreasing numbers of high school age population. SPC does not capture a high number of

students from the Northeast corner of the county while at the same time the analysis shows the

growth rate of high school age population as high in the northeastern area (Figure 4-1 1). The

northeast is a target of opportunity for SPC.

Being an older, more established, and densely populated area, Pinellas County is not

proj ected to have as great a percentage of population change as surrounding counties.

Nevertheless, the change in demographic composition of the county will significantly affect

SPC. Pinellas County is a desirable destination, therefore it is reasonable to expect an ongoing

process of densification in the county. Because of these factors examination in the overall

population trends throughout Pinellas and the surrounding counties is important (Figure 4-12).

On average, each ZIP code within Pinellas County is expected to increase in population 0.5%

during the next five years annually. Figure 4-12 illustrates that the greatest increase is expected












Section II .............. ...............48....
Market Penetration .............. ...............53....

Program Need ................. ...............56.......... ......
Summary Findings ................. ...............57.................


5 IMPLICATIONS AND CONCLUSIONS .............. ...............88....


St. Petersburg College Study: After Action Review .............. ...............88....
Expected Outcomes ................ ............ ...............89.......
Limitations and Possible Improvements .............. ...............90....
Implications for Higher Education Planning ................. ......... ......... ...........9

LIST OF REFERENCES ................. ...............93................


BIOGRAPHICAL SKETCH .............. ...............96....









Distribution of Students Attending
Online S~chool by Zip Cod~e


SCampuses
# of S~tudents







Figure 4-6. Online student geographic distribution. Online students show greater geographic dispersal of home addresses than
traditional campuses. The students used to produce this map are taking most of their classes online.













Table 4-3. 2005 College wide opening fall headcount enrollment by age and by division.


19 or less 20-24


25-29 30-39


40-49 50-59 60&over


Unknown Total


LO11TR DIVISION BY CAMPUS
Cleanwater


St Perer-bu~ rP Gi!-bs


Sennnole


Taurpou 5pr~ing

Lower Division Total

UPPER DIVISION BY PROGRAMS\



Education


International Business


Nursing

Orthotics & Pro~sthetics


Pubhec ibler:


Technology Management


Veterinr Technology


EPL72npact


Non-Degree Seeking


Upper Division Total

COLLEGE TOTAL


143;0 2129 1024 1154
21.3% 31.7%/ 15.2%/ 17.2%
2018 2967 1523 1738
20.5%/ 30.2% 15.5% 17.7%
869 836 297 376
31.7% 30.5%/ 10.8%/ 13.7%
1199 1394 471 490
30.2%0/ 35.1%/ 11.9% 12.3% (
5516 7326 3315 3758


0 6 23 36
0.0% 5.6%O/ 21.5%/ 33.6%
1 166 96 139
0.2%/ 33.7% 19.5% 28.3%
0 24 9 8
0.0% 50.0%/ 18.8%/ 16.7% 0
0 11 31 84
0.0%/ 5.0%/ 14.0% 38.0% 6
0 3 6 8
0.0% 13.0%~ 26.1%~ 34.8% o
0 14 12 27
0.0%/ 21.2%/ 18.2% 40.9% 6
0 50 67 137
0.0% 12.7%~6 17.0%/ 34.7% o
0 6 13 18
0.0%/ 11.3%/ 24.5% 34.0% 6
0 1 6 20
0.0%/ I.8% 10.9% 36.4%
0 2 4 13
0.0% 3.7% 7.4%i 24.1%
I 283 267 490

5517 7609 3582 4248


659
9.8%
959
9.8%
241
8.8%
310
7.8% 6
2169


30
28.0%
66
13.4%
6
12.5% 0
71
32.1%
6
26.1% o
12
18.2%
102
25.8% 0
14
26.4% ~
23
41.8%
23
42.6%
353

2522


267
4.0% b
359
3.7%
110
4.0% b
80
2.0% /
816


12
11.2% b


4.7%
1
2.1% b
24
10.9% 0
0
0.0%
1
1.5% 0
35
8.9%
2
3.8%
4
7.3%
11
20.4% 6
113

929


41
0.6% i
63
0.6% /
14
0.5% i
23
0.6% /
141


0
0.0% i
1
0.2% /
0
0.0% ;
0
0.0% /
0
0.0% ~
0
0.0% ~S
2
0.5% ~
0
0.0% ~6
1
I.8% ~
1
I.9% /6
5

146


20
0.3%/
195
2.0%
1
0.0%~
7
0.2%/
223


0
0.0%~
0
0.0%
0
0.0%~(
0
0.0%/
0
0.0%~
0
0.0%/
2
0.5%~
0
0.0%/
0
0.0%
0
0.0%;
2

225


6724


9822


2744


3974


23264


107

492


48


221


23


66


395


53


55


54


1514

24778


(SPC Factbook 2005)









regards to local populations. Use of census blocks as the core geographic unit of analysis would

increase the resolution of data and thus increase the detail and accuracy of Eindings. However,

considering the marketing value of using ZIP codes, a combination of ZIP codes, ZIP +4 codes,

and census blocks would provide the most useful data for both analysis and marketing purposes.

Finally, the student population included in the analysis only accounted for students living

within Pinellas County because SPC marketing is limited to within the county borders.

Limitations to the extent of marketing is the result of state rules used to maximize the resource

allocation to community colleges by discouraging competition between state funded higher

education institutions. This means that although analysis looks at a large portion of the student

body, in fact the analysis relies on a sample of the study body and could be expanded (though not

for marketing purposes) to include students traveling from outside of Pinellas County.

Implications for Higher Education Planning

Despite limitations, my research demonstrates that the geographic landscape is very

important consideration for higher education for a variety of reasons. As discussed in chapter

two GIS can be used as proficient determinant of socioeconomic status (SES) (Pennington 2002).

Maps can then be generated to display the SES of distribution of students throughout the trade

area of the college or even the state. Once student SES is mapped various components of the

landscape such as demographic characteristics, industry influences, environmental

characteristics, and even lifestyle segmentations can be overlayed to examine geographic

relationships (Pennington 2002).

As described by Bailey 2006, SES is used for determination of Einancial aid distribution

and for many students SES is unknown. Determining SES through the use of GIS can improve

reporting for tuition and Einancial aid policy. Tracking the changes in the landscape of student

SES or LSP over time is also a valuable use of GIS for higher education. Looking back at 1990













Hispanic Student Capture for SPC
Dilffernce in the rPerent osf Hispanic; Srtudents versus the
Percent of Hispanic Re~sidlents by Zip Code


MSPOs
33755 33756
33765 33787
Mslogam


Difference:
%Hispanic Students
-% d"Hispanic Residents 'au


1 1.01 2 00




-60 -3;6 121,2 36 SD
~%Dieran
kligal;el nu mbe rs corre late to an u nder Cmue
repre~sentation rrbhle positive nUmbeTS Cmue
point to over repr4 semation
6 2 4 8 12 16


Figure 4-19. Positive and negative Hispanic student capture rate throughout the county.
























111 07 0 **


rs.C~ C __ i 1a *i




I 11"E
a~~~ sus*
0~~~~~~ 30 *s u *A C







e a- ag 3 Tn

A I S II "
22 Me


32 Ruo


Legend
us 'Locations




urbanlrite
style
rer and GokI6
te~rprising Profession
?tropoillins
stbel t Retires
tinremnt Commmu nities
sthelt T~rditions
~ifle Junceionr
d and Newcmers
ung andI Roesless
e Eders
ear Espctarlnon
n~ior Sun Seekers
rdest Ilncome Homes


6--
r~.


,' r

a a


Figure 4-12. Lifestyle segmentation patterns throughout the county.









BIOGRAPHICAL SKETCH

Phillip Allen Morris was born and raised in southwest Virginia. He graduated from

Auburn High School in 1997 and immediately enlisted in the U.S. Army. After serving a two

and a half year enlistment he was honorably discharged and began undergraduate study at

Concord University in Athens, WV. While studying at Concord, Phillip was a member of the

West Virginia Army National Guard and a member of the university basketball team.

Upon graduation from Concord in 2003 Phillip deployed with his National Guard unit as

part of Operation Iraqi Freedom II. Phillip finished his enlistment with the National Guard in

2005, prior to enrollment in the Department of Geography at the University of Florida. Upon

graduating with his MA from the University of Florida, Phillip will pursue a Ph.D. degree from

the department of Education Administration and Policy.



















Growth Percentag e

-0 3%6 2 5%

2 501%6 5%

5 01%6 10%

10.01% 15%6

S15.01% 2096


.'
I
~i


1P 3 G 1? 161 ~24
~ IIdBr


Figure 4-11. Regional population outlook.
















Area of high population growth of all age groups
~pr-oviding pressure for additional higher
-1 education opportunities.


IComparatively adequate supply of

higher educational opportunities I~l~~~ Potentiay increased demand for
serving local markets. I aI higher education in areas of
critical need and economic

Comparatively modest population Ppressure w~ithin thiS Z LI development.
region domlinated by the State's only Assocation
of Am~erican Universities member. ETF draws
students from across the state. thereby limitingPpltinredidit nd
local opportunities for those unable to meet UFls A fr additional education
Shigh admission standards. 'Illopruii


Elder populations growth, distant from existing or plammed
major SUS presence.


Growing elder populations indirectly pressure higher
~growth in all age categories place! direct pressure
an higher education to meet its needs. ;







Figure 3-3. Summary Overview of The Atlas of the State University System of Florida. This map shows the summarized conclusions for the The
Atlas of the State University System of Florida. Gainesville. This type of analysis can be done by institutional researchers within
colleges and universities. (Thrall, G. 2005)









This level of intelligence requires good communication and organizational skills along with the

ability to integrate methodologies and Imowledge fiom multiple disciplines to design a study and

interpret results. The highest tier, the contextual intelligence, encompasses the ability to effect

change through the use of both quantitative and qualitative information (Terenzini 1993).

Understanding these three levels of intelligence can help understand the implicated role for the

use of GIS in IR. The use of GIS, a methodology from an external discipline, can assist in both

the technical and issues levels of intelligence through capabilities of data storage and analysis as

well as the ability to design a study and interpret results (Howard 2001).

The use of GIS in education planning is can assist the persons responsible for the

summarized process that leads to decision making presented below(Howard 2001).

1 The initial collection and storage of data.

2 The use of the resulting information to make decisions.

3 Effective communication of useful information for decision making.

Depending on the nature of the decision, all three steps could benefit from the visual and

analytical potential of geospatial analysis. Though not frequently used for most decision making

models in the higher education system in the US, GIS is a data system accompanied by spatially

explicit methodologies that would be useful for the implementation of key components of

decision making.


































0 2.5 6 10 15 20


,


D iffe re nce:
Percent beyond 50%
of F~erale Students

-17.86% 0%

0.01l% 5%~


~j


5.01% 10%

"10.01%S~ 1 5%

15,0"1% 20%


Figure 4-21. Male / Female student capture rate throughout the county.


Distri bution of Male/Female
Student Capture for SPC

Diffe rence in the Percrent of Female Students
versus the Male Students by ZipJ Codre
Negative numbers (symbolized as green)
show zip codes with under-represented
female students wthille the orange to red show
the: over-represented areas


usPos

3371 33701
3333M 1




































III


SPC Age Group Capture Rate
Capture Rate:
The percentage of the total college age
population enrolled at SPC.


Age 26-45


21-25


and Below


I
capture Rate

00%~- 7%/


MSP Os
33760 33714
33711 33716
33705 33701

o 4.s a


S7.01% 10%


a 14.01% 1 8%


SCampuses


Figure 4-22. Age group capture rate throughout the county.


MBSPos

337167



































Number of Students
Grid cells are -1.5 Top Four Camnpuses
** Kilometers on a side. ar9c2 seem
g 57 (10 l (92) 5,
O 23 ro 57 (91)
Only students from th~e top O Oor 4 (2se
four campuses are represented.

Figure 4-4. Geo-coded address of SPC students living within Pinellas county aggregated to a grid of 1.5KM square cells. The
displayed data only includes the students enrolled in each of the top four SPC campuses in Fall 2005


Student Distribution in Pinellas
Cou nty









assist in this process by tailoring institutional programs to the currently and future enrolled

students.

There are certain measures that are effective in predicting student success in the collegiate

setting (Knight 2003). High school GPA in combination with standardized test scores is

typically best at predicting college GPA and retention in the first year of college. However, the

authors of the chapter on enrollment management show that this combination does not do a

particularly good job predicting success and retention after the first year and does a relatively

poor j ob of predicting which students will graduate after four to six years. This is evidence for

putting less emphasis on merit based financial aid. If merit is not a good predictor of who will

succeed and graduate, then more emphasis should be put into need based aid.

Decision Making

Traditionally the Institutional Research office has relatively few direct responsibilities, but

this is changing. Some institutions combine IR with planning (Walleri 2003). Two planning

considerations from the IR office that can be assisted with GIS are demographic and population

proj sections as well as enrollment analysis and forecasting. Most IR offices do not incorporate a

spatial component in the planning process, so to better understand the benefit of using a GIS in

planning an examination of current considerations for decision making in post-secondary

education is presented below.

Decision support may take many forms and various levels of analytic complexity (Howard

2001). According to Terenzini (Terenzini 1993) there are 3 levels of intelligence in

administration in higher education. The first tier, technical intelligence would include abilities in

conducting surveys, cost benefit analysis, and data collection, manipulation, and analysis. This

typifies the primary day to day operations of an IR office and consists of numeric tabulated data.

The second tier, issues intelligence, includes a understanding of institutional decision making.



























O 3 16 12 18 24
i Mlllr~


Highi School Age Group
Percentage of C~hange
2005 2010O
Estimate is based on the unumbers of
children now age 10)-15 and the proqjegscte
change in population.

Rate of Chlange


Area of high
H.S. age
growth


-25%~~ 1~5%
S-'14.99% -5%
I ~ -4.99% 5%
S5.01% ~15%
M 15. O1% 30%


r
~1V


Figure 4-9. High school age percentage of change proj section.


seass



Area of
D ecreasi nlg
H.S. ag e population
































O 2007 Phillip Allen Morris









student takes the most credits from determines the color of their dot. The online campuses have

the potential to draw from areas outside Pinellas County. Moreover, online offerings can serve

to attract more Pinellas County residents to SPC. SPC online students are for the large part

clustered around Tampa Bay. Distant students might be registering with their parents' addresses,

versus their own near campus address. Figure 4-3 shows the extent of the student enrollment and

reveals potential for marketing and recruitment from various clusters within the state.

Student geographic distribution within the boundaries of Pinellas County was evaluated by

dividing the county into 1.5 kilometer cells. The number of students within each cell was

calculated. Figure 4-4 shows student geography within Pinellas County. The darker colored and

elevated cells have more students than the lighter lower cells. The geographic distribution of

SPC students is clustered in the southern perimeter of the county, with SPC serving fewer

students in the northeast. Also, areas of the southeastern region of the county shows enrollment

dropping to below 50 students per 1.5 KM cell. These areas may be commercial, and the

database did not include addresses for place of work.

There is a strong correlation between enrollment of students and proximity to the campus.

SPC does well in enrolling students within close proximity to the campus. Online students

however show greater geographic dispersal of home addresses than traditional campuses. Figure

4-5 shows student enrollment separately for the Tarpon Springs and Clearwater campuses, by

ZIP code. The Seminole and online enrollment are shown in Figure 4-6. Each area of the county

is revealed to be well served by at least one of the SPC campuses, with the exception of the

northeast. The northeast is revealed to be part of the Tarpon Springs trade area, but

comparatively few students are enrolling at SPC from this area.









it is address, ZIP code, or ZIP +4) can be programmed into a digital map using geospatial

technology, including software and specialized databases. This process, called geo-coding,

allows for the demarcation of trade areas as well as a variety of other spatial analysis. Knowing

the location of the customer base gives the service provider the ability to describe the

relationship between store and customer with other descriptive variables. David Loudon

(Loudon 1979) describes the collection of cognitive information by consumer researchers below.

Consumer researchers who desire to know about their market more than just
demographic characteristics may attempt to collect cognitive information; that is,
information about consumers' knowledge, attitudes, motivations, and perceptions.
Merely observing consumers cannot fully explain why they behave as they do, and
questioning often does not provide reliable answers because of consumers' inability
or reluctance to reveal true feelings to an interviewer. Thus, researchers attempt to
explore intervening variables potentially useful in explaining consumer behavior by
utilizing other techniques (Loudon 1979).

Knowing the locations of the customers is a beginning step for market researchers. The

intervening variables mentioned above can be summarized through the use of lifestyle

segmentation profiles. Lifestyle segmentation profies (LSPs) are classifications of a

neighborhood that incorporates many different variables such as family status, income, consumer

spending behaviors, media and advertising influences, and even leisure and recreational

activities. These variables collectively represent the households in a particular neighborhood.

The first step in creating LSPs is defining the scale at which the profies will be used. The scale

can range from a Census block group to ZIP+4 designations. The smaller the geographic area

the more accurate the assigned LSP is likely to be. Census blocks are the smallest geographic

area for which the Census reports data. The shapes of census blocks follow the geographic

pattern of the streets, and are generally in between intersections. Residences included are usually

back to back rather than on opposing sides of the streets (for further explanation on U. S. Census

Bureau reporting units see (Thrall 2002) or (Hamilton and Thrall, 2004)).












Asian Student Capture for SPC*
Different ce in the Percent of Asian Students versus the
Percent of~sian Residents b~y Zip GCode p ~




p~~ONE ~Hstocgam -

10.0

Difference :
o;; %AsianStudents I
% Asian Residents "7

S1.51 -2.~64--
S0.51 -1.50
1 -0.49 0.50 1iIIITI""~~




IlJegative numbers corre~late to anr unders Cmue
represtentation while positive? numbers Cmue
point to oveor repre~sentation
0 2 4 8 112 16


Figure 4-18. Positive and negative Asian student capture rate throughout the county.









CHAPTER 3
CURRENT ISSUES INT HIGHER EDUCATION

This chapter will bring current and common problems that exist in higher education into

context with the notion that geographic techniques can benefit institutional research. By

examining relevant reports and published literature, issues will be identified that can be

addressed or improved upon through the use of geo-demographic analysis and geo-spatial

reasoning. This survey of literature will establish grounds for the use of geography in analytical

reasoning in higher education and examine existing studies and literature that address the use of

geographic information systems in education.

United States Higher Education

Education is considered the hard core of human capital theory (Sahota 1978). With global

specialization human capital has become one of the most important factors of production from

the local to global scale. With the focus on specialization of production and particularly on the

technology industry, education and the development of specialized skills across the entire

population is of increasing importance. Of particular interest for this study is the distribution of

access to higher education opportunities. This distribution has far reaching effects. Unequal

distribution can lead to an insufficiently educated population and thus affect the economic and

social structure of society.

Relative to most countries the United States has one of the most unique and successful

higher education systems in the world (Clark 1990). The success of the higher education system

in the US is partially dependent on the structure of the system. The de-fragmented

organizational structure, along with many other factors, has allowed U.S. tertiary education

systems to adapt and grow to the point where they are now some of the best in the world (Lucas

1994). The success of higher education in the U. S. is far from flawless. Current flaws that have










ubiquitous use of GIS in marketing and business some groundwork was established by highly

influential theorists such as William Applebaum.

William Applebaum, a geographer who was able to apply the techniques of geography to

business and marketing, is regarded as a founding father of business geography. Numerous

publications in academic journals as well as business and marketing volumes document his

success at merging the discipline of geography to business practices (Ghosh 1987).

Applebaum's contributions to business geography include the customer spotting method, the

analog method, market penetration methods, and methods for determining store location (Thrall

2002). The methods developed by Applebaum that most impact this study are the customer

spotting and market penetration techniques.

The customer spotting technique involved surveying of the customers or recording

information from license plates or cars in the parking lot in order to obtain customer addresses.

This technique allowed Applebaum to determine the trade area for a particular establishment. A

trade area is the geographic region for which a business draws most of its customers (Ghosh

1987). In discussions about customer spotting Applebaum delineates trade areas into three

distinct zones:

1. Primary Trade Area: This area is comprised of 60-70% of the customer base
(analysts now most frequently consider 80% capture of the customers as the primary
trade area).

2. Secondary Trade Area: This area is comprised of 15-25% of the customer base.

3. Tertiary Trade Area: This accounts for the residual customers. This consists of
sporadic customers or out of town customers (Thrall 2002).

The method used by Applebaum for customer spotting has given way to the much more

efficient practices of collecting bank data, credit card data, and point of sale data such as ZIP

codes (Thrall 2002). Once this descriptive customer data has been collected, locations (whether









Maybe the most significant source of literature discussing the use of GIS in higher

education is the Winter 2003 edition of the quarterly sourcebook sponsored by the Association of

Institutional Research, New Directions in Institutional Research. The individual chapters each

detail a different utility of using GIS in institutional research (IR) and the benefits associated.

Typical applications offered by the book include:

Using Census data and existing knowledge of student demographics to inform decisions
with consideration of planning and implementation of recruitment strategies and tactics.

Building student databases and displaying enrollment trends with maps for visualization
and analysis.

Use of GIS in survey research for design, analysis, and reporting of survey data.

Campus planning and facilities management; GIS in a traditional planning capacity.

Mapping and analyzing alumni donation patterns. Through the uses of address data and
demographic data alumni associations can better understand where to focus future
campaign activity (Teodorescu 2003).

Resource allocation in higher education is directly linked to state legislative bodies.

Geographic techniques can be used by institutions to inform state legislators of the influence of

their university on the legislative districts throughout the state (Thrall and Mecoli 2003). The

University of Florida' s student address database was recently used for this purpose. By spatially

j oining the addresses of the students to the state legislative districts, the University of Florida

was able to show the impact of the school per district, and therefore the value of the university

throughout the state (Thrall and Mecoli 2003).

Institutional Research and Decision Making

The examples described in this chapter exemplify significant justification for incorporation

of geospatial methods in college and university planning and decision making. The following

section of this chapter focuses on key functions of institutional research and existing decision









CHAPTER 5
IMPLICATIONS AND CONCLUSIONS

St. Petersburg College Study: After Action Review

The analysis completed for St. Petersburg College represents the overall advantages gained

from a geodemographic analysis report. Geodemographic measurements are descriptive

characteristics of a population, arranged and ordered by scales of geography that is meaningful to

the analysis (Thrall 2002). In the SPC analysis the most meaningful geographic scale is ZIP

codes because of the functional nature of postal code designation for advertising and marketing

campaigns. The use of maps for geodemographic analysis provides a graphical representation of

the landscape. Demonstration of the benefits of market analysis can be seen in the applications

of business geography. Thrall 2002, discusses the importance of research in real estate market

analysis:

With this information in hand, an analyst then wants to predict the answers to questions
such as, what will be the geography on the city in five or ten years? If a retail outlet is
built here, will it be successful? In the future where will the population that is expected to
be consumers or products live? Which neighborhoods will be on the decline, and which
will be on the rise? Knowing the answers to such questions creates opportunity for the
investor, and is the raison d' etre of the market analyst. (Thrall 2002)

This reasoning is typically used for business analysis of all sorts, but can be transformed

for the use of higher education. For example, multiple branch retail outlets use this type of

analysis as standard operating procedure and parallels can be drawn between multiple branch

retail and multi-branch learning institutions. Both have customers with trade areas and both can

benefit from market and trade area analysis. According to a recent report by the Secretary of

Education for the U.S., higher education is becoming increasingly consumer market driven and

students care less about whether a college is a for profit or public institution, a predominantly

online or brick and mortar instructional system, and more about absolute results (Spellings

2006). This evolution of students into consumers of education along with the need for





I I


Locatrr ionr s


.-. .Jh.:I Pll 11p 7t.~I: rri:


Pinellals Comet


- 1


21


0ie


0 2.5


Figure 4-1. Map of SPC campuses and learning sites in Pinellas County

















The gr~owrb of undergroaduate~ aged (15-25
yerars old) population ir rimuinlar o thej
concentration of growth of glnllrdut aIged I
popnulaion.

The armsn ihade~d in darkL redI areC projeCtedI to rci.LeP~ the gl~~reatest
increase in runderglRaduate aged~ p~opulationu between 20051 nan 20103.
UNIVERSITY
.A rui ng h ,iger edul. Ca I tion iis pIresen utls. oper nr1~ing at f uU iCa pa1cit ;-VL
errr ing existing undergl'rladue age~d populnfionr in the da;rk redl nee' s-c
p fb~the threre arear identify~ locations where~ popubl~lion surrger mayB '_ sLu
Sindientte need for additionnl eduication~ errvi~cs.
m:IC






1.2d { hanug In UndelP~rgraduatl e Aged Population




Figure 3-1. Summary Overview of The Atlas of the State University System of Florida. This map shows the proj ected increase in undergraduate
aged population in Florida by the year 2010. (Thrall, G. 2005)










i"


Legend
SPC Atrea FOOD ANDa BEVERAGE, STORES
Nu mber of Em ployees [iiiiiil FO3OD SiVCS & D RINKIN G PLACES
oo ~ 67 -3011 HEALTH- & PER SONrL. CARE STriFES
O 3012 -6057 1 HOSPITALS
O 60Esa ?271 JUSTICE PUBIC ORDEFsSAFETY
Q 41:2- ~i16: ~MAdHCHIERY MANrUFACTUR N G
SMERCH, WHFOLESALERS DURABLE GIS
O 14068 22131 I MOTOR bEHIcLE PARTS DA~LERS
I rURslrG RSI* RE51;~E FACI~LIT
torpi n dusry TOPJNDUST
M OTHER INFORMATION SERVICES
M AICCO)MMO DA;TION
I PR~OF; SCIENTIFIC; & TECH SVICS
ADMIN. HUMAN R ESOU RCE iPROGRAM~S
I REAL ESTATE
III AMIBULcfO RY HEALTH CARE SVCS
I RELIG GRANT; CIVIC; PROF ORG
I CREDIT INITER IEDI;ATIOON & RELATD SPILTRECOTAOS
I EDUCATkIrfjAlL SERVICES
I E EC LEGIS.; &OTHE SUPPORT


ic4' 11

t
Is~-

r


17 .1
~1

jr


ig


Figure 4-24. Industrial influence in the county.









In the proposal development stage an intention of the class was finding SPC objectives that were

both high priority and approachable from a geo-spatial perspective. Figure 4-2 shows an

example of the college obj ectives that are focused on in this analysis.

Trade Area Assessment

Section I

In order to get an overall idea of where the students are coming from the analysis requires

assessment of the trade area. The trade area can be described as an area encompassing 80% of

the customer base (Thrall 2002). In a community college this standard can be applied, but in the

case of SPC the trade area is bound by the Pinellas County boundary. Conversations with SPC

administration determined that the community college system in Florida prohibits SPC from

marketing outside of Pinellas County. Using this constraint the analysis for of this research

proj ect is for the most part limited to Pinellas County. Of the 22,456 students included in the

database tables provided for the proj ect, 18,03 5 are identified as having an address within

Pinellas County. Coincidentally, the Pinellas located students represent approximately 80% of

the total student population which is the same benchmark used in business geography.

Section II

In this section of the analysis student geographic distribution, drive time analysis, age

distribution of students and psychographic composition of enrolled students will be examined.

Based on these measures a better understanding of the population of St. Petersburg College

students can be developed as well as an understanding of what can be done by the college to

better represent this population.

The maj ority of SPC students reside within the Tampa Bay area, most within Pinellas

County, but many throughout the State of Florida. Figure 4-3 shows the home address of every

SPC student. The learning sites are represented by a different color dots and the site that the









Distribution of Students Attending
Seminole by Zip Code


Distribution of Stuidents Attending
Gibbs by Zip Code


jlCampuses
# of Students




S126 -194


# of Students
sPG_ ToT




379 57


Y

1~J
'i


Figure 4-5b. Seminole and Gibbs campus geographic distribution of student enrollment. The students used to produce these maps are
taking most of their classes at that particular campus.











Spider Graph of~ LSP Five Studen~ts
Students livin is senior stle neighborhoods ad their respe6ive cEampus





usep see saw seveles -
makec up of 32%b at the
..c~ rrebrssn nmw


Spider Grcaplh o~f LSP Ten Students;
Sitisadents Ilvin in trojiegonpl living nesgqhitesr~rilood and their IreSEctive camPlr


.---- L10Traditio nal
-- n 19Living: Middle
-aged, middle
.InCOme,
Middle Ame rica


I
Y,


B


LS Senior Styles:
Senior lifestyles t~y
income, age, and
housing ty e


3


twP Ten Tre~sthonbl l uno -
runs up 20i ar trke sausanls
en.-alled in rh+t REp four
44'"PW='**


.5


LSP 5 Legend


Frfg~f Sp~fg


.*' .. .~'


Figure 4-14. Spatial pattern of lifestyle segments five and ten for students within Pinellas County.









the locations of these areas, SPC management can better understand their trade area and develop

recruitment strategies to appeal to their desired segment.

The study also sought evidence of clustering of specific socio-demographics by campuses.

This question is addressed by producing a series of spider maps (Figures 4-15, 4-16). Spider

maps draw lines from the students' residence to their respective campuses. Almost 75% of all

SPC students are contained within only four lifestyle segmentation profies (LSP). LSP 5, or

"senior sun seekers", makes up 32% of the total enrollment. These students are classified as the

senior population based on income, age, and housing type. LSP 10, or the "traditional living",

makes up 20% of the enrollment. This profile is classified as middle aged, middle income -

middle America. LSP 7, or the "high hopes", makes up 14% of the enrollment and is classified

as young households striving for the "American Dream". Lastly, LSP 1 students, or, the "high

society" makes up 7% of total enrollment. This profile is classified as affluent, well-educated,

married couple homeowners.

After analyzing these student profiles and their respective campuses, it is difficult to

identify significant clustering within the largest LifeMode group, Senior Styles. The senior sun

seekers are enrolling at campuses throughout the county and there is a group that travels beyond

their closest campus. This could be explained by the offerings of certain maj ors and classes at

specific campuses. Clustering increases as the percent of enrollment from LSPs decrease.

Market Penetration

Capture rate is defined as the percentage of the total college age population enrolled at St.

Petersburg College. The market penetration section of the analysis shows the student capture

rate throughout county as well as the minority capture rate, age group capture rate, and the

gender capture rate. For this analysis the market penetration capture rate is assessed by zip code.

SPC enrolls from five to six percent of the college age population in 17 of 47 ZIP codes within












TABLE OF CONTENTS




page

ACKNOWLEDGMENTS .............. ...............4.....


LIST OF TABLES .........__.. ..... .__. ...............7....


LIST OF FIGURES .............. ...............8.....


AB S TRAC T ............._. .......... ..............._ 10...


CHAPTER


1 GEOGRAPHY AND HIGHER EDUCATION: AN INTRODUCTION INTO
GEOSPATIAL PRACTICE AND INSTITUTIONAL RESEARCH .................. ...............12


Foundations of Geospatial Reasoning .........__... .....__ ...............12...
Institutional Research .............. ...............15....


2 LINKING GEO-SPATIAL REASONING TO HIGHER EDUCATION: LITERATURE
REVIEW ................. ...............19.................


Advantages of GIS in Marketing ............ ..... ._ ...............19..
Spatial Information in Education............... ...............2
Institutional Research and Decision Making .....__.....___ ..........._ ...........2
Enroll ment Management .............. ...............26....
Decision M making .............. ...............27....

3 CURRENT IS SUES IN HIGHER EDUCATION ................. ...............31........... ..


United States Higher Education ................. ...............31................
Inequality ............... .... .......... ...............3.. 2....
Higher Education in Florida ................ ...............37................
Chapter Conclusion .............. ...............39....

4 GEO-DEMOGRAPHIC ANALY SI S............... ...............4


St. Petersburg College .............. ...............44....
Organizational Structure................ ..............4
Historical Development............... ..............4
Values ................. ... ........ ...............45.......
St. Petersburg College Analysis .............. ...............46....
Focus of Analy si s .......................... ......... .................. ................46
Data............... ..... .............4
Trade Area Assessment ................ ...............48........... ....
Section I.............. ..............48....






























To my parents










Ghosh, A., & McLafferty, S. (1987). Location strategies for retail and service firms. Lexington,
MA: Lexington Books.

Greenberg, M. (1997). The GIBill: The law that changed America. New York, NY: Lickle
Publications.

Harvey, D. (1969). Explanation in geography, New York: St Martin' s Press

Hauptman, A. (2007). Strategies for improving student success in postsecondary education.
Changing Direction. Boulder: CO, Western Interstate Commission for Higher Education
(WICHE) 24.

Howard, R. (2001). Conceptual Models for creating useful decision support. New Directions for
Institutional Research (Winter 2001) Volume 112: 45-55.

Kirsch, I., Braun, H., & Yamamoto, K. (2007). America's perfect storm: Three forces changing
our nation's jitture. Princeton, NJ: Educational Testing Service.

Knight, W., Ed. (2003). The primer for institutional research: Resources in institutional
research. Tallahassee, FL: The Association of Institutional Research.

Kovalerchuk, B., & Schwing, J., Ed. (2005). Visual and spatial analysis: Advances in data
mining, reasoning and problem solving, NY: Springer.

Kuttler, C. (2006). St. Petersburg College 2006-2009 strategic directions and 2006-2007
institutional objectives. St. Petersburg, FL: 1-20.

Loudon, D. (1979). Consumer behavior: concepts and applications, New York: McGraw Hill.

Lucas, C. (1994). American higher education: A history. New York, NY: St. Martin Press.

Martin, G. (2005). All possible worlds: A history of geographical ideas. New York: Oxford
University Press.

Pappas, A. T. (2007). Proposing a blueprint for higher education in Florida: Outlining the way to
a long term master plan for higher education in Florida. Tallahassee, FL, Pappas
Consulting Group Inc. Retrieved on March, 5th, 2007 from www.uff-
fsu. org/art/Papp asB OGStructureRep ort.p df

Pennington, K., & Williams, M. (2002). Community college enrollment as a function of
economic indicators. Community college journal of research and practice 26: 43 1-43 7.

Piccillo, S. (1999). How marketers benefit from mapping demonstrations. Marketing News. 33:
15-16.

Sahota, G. (1978). Theories of personal income distribution: A survey. Journal of Economic
Literature 16: 1-55.









in ZIP code 34688, bordering Pasco and Hillsborough Counties; ZIP 34688 has an expected

estimated growth of 4.5%. Both Pasco and Hillsborough Counties will experience high growth

rates, particularly in areas adj acent to Pinellas County. While Pinellas County may be an

effective trade area for SPC, SPC can better serve the state' s population by carrying some of the

burden of student enrollment in nearby adj acent areas. The adj acent areas shown in Figure 4-12

are MSPOs.

Student demographics were also analyzed using market segmentation profiling. The

reasoning of market segmentation profiling is that people with similar tastes, lifestyles, and

behaviors seek others with the same tastes. These behaviors can be measured, predicted, and

targeted (Thrall 2002). By examining these profiles with student data, we can better understand

the economic landscape that SPC must draw their students from.

Lifestyle Segmentation Profiles of SPC students were calculated using ESRI's Tapestry

LifeMode groups (ESRI 2006). Figure 4-13 shows ZIP codes color coded to dominant

LifeMode group. LifeMode groups can be further decomposed into 64 more detailed segments

(Figure 4-14). SPC trade area has 17 ZIP codes dominated by the LifeMode group Senior

Styles, which is comprised of Rustbelt Retirees, Senior Sun Seekers, The Elders, and others.

The majority of SPC students are from areas with aging populations. Projections indicate

that the recent trend will continue over the next five to seven years. However, Figures 4-13 and

4-14 document that some areas within the SPC trade area are dominated by younger households

with college age students that are potential SPC students. Examples of MSPO are high hopes,

young and restless, solo acts, metropolis, great expectations, global roots. Figures 4-13 and 4-14

shows the locations of MSPO. By understanding the differences in these groups and knowing









to determine the level of market penetration of particular demographics throughout the state and

" in identifying particular communities where students may face barriers to college access"

(Crosta 2006). A summary of the steps involved in the cluster analysis is shown in Table 2-2.

Further recommendations from this study will be discussed in the concluding chapter of this

paper.

Further demonstration of the use of GIS in higher education can be seen by examination of

the office of IR at the University of Memphis (U of M). Memphis has used GIS to map students

and alumni since 2001. The webpage for the IR office has a collection of interactive maps

(http://oirmaps. memphi s. edu/maps/map~index.htm) Thi s collection includes:

* U of M Undergraduate Retention and New Enrollment, Fall 2005 (U. S.)

* Current Off-campus Class Locations and Underrepresented Areas (Shelby County, TN)

* U of M Undergraduate Applicants / Acceptees / New Enrollees, Fall 2005 (Tennessee)

* U of M Graduate Applicants / Acceptees / New Enrollees, Fall 2005 (Tennessee)

* U of M Students by Home Zip Code Area, Spring 2006 (Shelby County and nearby
counties TN, AR, MS)

* Undergraduate Students by High School Attended, Spring 2006 (Shelby County, TN)

* Foreign Graduate Students Attending U of M, Spring 2006 (World)

* Foreign Undergraduate Students Attending U of M, Spring 2006 (World)

* U of M Active Alumni (by zip code area), May 2004 (continental U.S.)

The maps displayed on the University of Memphis website demonstrate the capability of

GIS to track and manage students and alumni. This gives university administrators the ability to

easily see where students are coming from and as well as areas to target for alumni donations or

areas with underrepresented segments of the population (Donhardt 2001).









college as a whole. The president of the college is the secretary of the Board and is charged

with the day to day operational management. A Provost, Vice President, or an Executive

Officer supervises each campus or learning site.

Historical Development

In 1927 St. Petersburg Junior College opened as Florida's first two-year institution of

higher learning. The college was based out of a wing of the new St. Petersburg High School.

The initial enrollment was 102 students with a faculty of 14. Historical dates for the

development of the college are listed below.

* 1931 Gained full accreditation
* 1948 Private college became public
* 1965 SPJC merged with African-American Gibbs Junior College
* 1990s SPJC occupied a dozen sites throughout the county
* June 2001 SPJC became St. Petersburg College, a four-year institution
* August 2002 SPC began offering fully accredited baccalaureate programs
(SPC 2006)

Values

According to the information found on the college website the purpose of the college is to

provide access for:

Students pursuing selected baccalaureate degrees, associate degrees, technical certificates,
applied technology diplomas and continuing education within our service area as well as in
the State of Florida... As a comprehensive, multi-campus postsecondary institution, St.
Petersburg College seeks to be a creative leader and partner with students, communities,
and other educational institutions to deliver enriched learning experiences and to promote
economic and workforce development.

(http://www. spcollege.edu/webcentral/catalog/Current/msingashm

This research has significance for SPC's mission statement because of the ability to use

geospatial tools to geographically organize data related to student profiles, community

characteristics, and other institutions that will assist in promoting economic and workforce

development.









technical colleges. The CCRC used two methods to achieve this goal. The first process is

determining SES from student addresses and census block groups, and the second process

involves using cluster analysis to create segments of the population.

The first process, deriving SES by address location, is achieved through the use of GIS.

The process, summarized in table 2-1, defines student SES by using household income,

education, and occupation as indicators. This use of appending census data to individual

addresses is important because it gives decision makers specific information about characteristics

of residents living at these addresses and comparisons of tabulated data can identify patterns

within a particular population. For this study the population is community college students of

Washington State.

The second method used to better understand the relationship of the population to

characteristics of students is called cluster analysis. Cluster analysis is the use of multiple

algorithms to group objects with similar characteristics into categories. In this study cluster

analysis was used for grouping census block groups that have similar traits into "community

clusters". Cluster analysis is very useful at grouping large datasets, but the composition of the

resulting groups depends on variables used and parameters designated. In the Washington study

the variables used were demographic variables "relevant to community college educators"

(Crosta 2006).

Using clusters, according to the authors, is more descriptive and effective in determining

SES than using the three indicators (income, education, and occupation) mentioned in the section

described earlier. In the CCRC study the Washington Census block groups were divided into 15

individual clusters with distinctive characteristics. Once the clusters are determined, students

within the state can be assigned a cluster based on home address. These clusters can also be used









innovative methods of reaching the disadvantaged segments of the population give validation to

the methods described in the SPC study. Based on the Eindings, SPC will have a better

understanding of the demographic characteristics of the county and a better understanding of

which markets and submarkets to target in order to meet the institutional obj ectives and goals.

Typically an after action review, a term borrowed from the US Army that describes a post

training exercise review, is designed to 1) investigate and discuss expected outcomes, 2)

examine unexpected outcomes and 3) determine what could have been done to improve the

sequence or lend strength or relevance to the action in question. The first and third components

of this review are useful for this study, however unexpected outcomes will not be discussed.

Expected Outcomes

The SPC study started as a proj ect designed to investigate declining enrollment of students

at the college. A primary intention of the proj ect was to give the college a better understanding

of the student population based on the physical location of the students and the characteristics

and relationships that can be determined based on location. Using Geospatial technology the

students of SPC were assigned geographic coordinates and plotted on map that contained ZIP

codes and demographic data for the people within these ZIP codes. With the knowledge of how

many students are living within these ZIP codes in Pinellas County, analytical relationships were

asserted and recommendations about marketing and recruitment were made to the college.

Along with ZIP code boundaries the analysis also used ZIP + 4 boundaries and with the

use of methods and tools of business geography lifestyle segmentation profiles (LSP) were

assigned to all the students. This allowed for analysis of the student population based on

lifestyle profies. ZIP codes were then labeled based on the maj ority or dominant lifestyle

population. Specific recommendations were not given to the college based on the geographic

distribution of student lifestyles, but a more general landscape view is presented in the form of a













~ --I

ams a


n


"~
d


e ase asLegend
A cinamps Locations


** LifMOde Groups



SH~gh Socialy


MSIlerlct r~Sty

*" ** *Tradstroina Ltvn
W Ulet~ rie Averkets








Figure 4-13. Lifestyle group patterns throughout the county.


SUls utis
r 5


D~ranbauerunpublished









community colleges (Spellings 2006). This emphasizes the need for a focus on the services

provided by two year institutions.

One section of the Spellings report discusses access as a concern. On access for minorities

the commission writes, "we are especially troubled by gaps in college access for low-income

Americans and ethnic and racial minorities." The population in the U.S. is growing the fastest in

the groups of people that fall into the category of non-traditional and underserved students. This

shift in demographics will constitute an ever increasing proportion of the workforce, a workforce

that needs proper education and training. The community college can play a great role in

training and educating these groups of people, and according to the report "provide a place to

begin for many of these students" (Spellings 2006).

Evidence from the Spellings' commission report supports the fact that eligible young

people from low income families are far less likely to attend college than young people with

similar qualities from high income families. The percentage of whites that obtain a bachelors

degree by the age of 29 is around 34%, compared to 17% of blacks and 1 1% of Latinos. The

most alarming finding is that low income high school graduates that score in the top quartile on

standardized tests attend college at the same rate as high-income high school graduates in the

bottom quartile on the same tests. This statistic suggests that although low income graduates are

capable of being successful in higher education, other factors (for the most part financial

restraints) limit these students and they are not afforded the same opportunities as high income

students to further their education.

This occurrence is further evidenced by a study published recently by an independent

nonprofit organization called The Education Trust. The study discusses the nation's premier

public institutions and shortcomings in regards with equal access. One section of the study












Age Below 20 21-25 26-30 31-35 36-40 41-45 46-50 51-55 Over 55 Total
Total 5285 5128 2321 1547 1236 1026 734 443 313 18033
Percent by Age 29% 28% 13% 9% 7% 6% 4% 2% 2% 100%
Cum. % by age 29% 58% 71% 79% 86% 92% 96% 98% 100%


Table 4-1. Cumulative age distribution of SPC students within Pinellas County

PAGE 1

1 USING GEOSPATIAL REASONING IN INSTITUTIONAL RESEARCH: ST. PETERSBURG COLLEGE GEODEMOGRAPHIC ANALYSIS By PHILLIP MORRIS A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS UNIVERSITY OF FLORIDA 2007

PAGE 2

2 2007 Phillip Allen Morris

PAGE 3

3 To my parents

PAGE 4

4 ACKNOWLEDGMENTS To acknowledge those that have influenced my development of learning, I would first like to thank my parents for always being supportive, caring and there to keep me on the right track. I would also like to thank my co lleagues and classmates who I cont inue to learn from. Lastly I thank Dr. Grant Thrall for guiding me through the masters program and allowing me to explore the field of spatial science as well as opening the door for me to pursue a doctoral degree in higher education administration.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........7 LIST OF FIGURES................................................................................................................ .........8 ABSTRACT....................................................................................................................... ............10 CHAPTER 1 GEOGRAPHY AND HIGHER EDUCATIO N: AN INTRODUCTION INTO GEOSPATIAL PRACTICE AND INSTITUTIONAL RESEARCH....................................12 Foundations of Geospatial Reasoning....................................................................................12 Institutional Research......................................................................................................... ....15 2 LINKING GEO-SPATIAL RE ASONING TO HIGHER EDUC ATION: LITERATURE REVIEW......................................................................................................................... ........19 Advantages of GIS in Marketing............................................................................................19 Spatial Information in Education............................................................................................22 Institutional Research and Decision Making..........................................................................25 Enrollment Management.................................................................................................26 Decision Making.............................................................................................................27 3 CURRENT ISSUES IN HIGHER EDUCATION..................................................................31 United States Higher Education..............................................................................................31 Inequality..................................................................................................................... ....32 Higher Education in Florida............................................................................................37 Chapter Conclusion............................................................................................................. ...39 4 GEO-DEMOGRAPHIC ANALYSIS.....................................................................................44 St. Petersburg College........................................................................................................ ...44 Organizational Structure..................................................................................................44 Historical Development...................................................................................................45 Values......................................................................................................................... .....45 St. Petersburg College Analysis.............................................................................................46 Focus of Analysis............................................................................................................46 Data........................................................................................................................... .......46 Trade Area Assessment...................................................................................................48 Section I....................................................................................................................48

PAGE 6

6 Section II..................................................................................................................48 Market Penetration..........................................................................................................53 Program Need..................................................................................................................56 Summary Findings............................................................................................................... ...57 5 IMPLICATIONS AND CONCLU SIONS.............................................................................88 St. Petersburg College Study: After Action Review.............................................................88 Expected Outcomes.........................................................................................................89 Limitations and Possible Improvements.........................................................................90 Implications for Higher Education Planning..........................................................................91 LIST OF REFERENCES............................................................................................................. ..93 BIOGRAPHICAL SKETCH.........................................................................................................96

PAGE 7

7 LIST OF TABLES Table page 1-1 Winning the site selection race..........................................................................................18 2-1 Summary of steps to find soci oeconomic status from block groups.................................30 2-2 Summary of steps to invo lved in cluster analysis..............................................................30 3-1 List of required qualifications for Bright Futures scholarship...........................................43 4-1 Cumulative age distribution of SPC students within Pinellas County...............................85 4-2 Fall 2005 distribution of student race by campus..............................................................86 4-3 2005 collegewide opening fall headcount enrollment by age and by division..................87

PAGE 8

8 LIST OF FIGURES Figure page 1-1 Map of the cholera-infected area in Westminster region of London, England in 1854.....17 2-1 Advantages of demographics ma pping and possible further analysis...............................29 3-1 Summary Overview of The Atlas of th e State University System of Florida...................40 3-2 Summary Overview of The Atlas of the St ate University System of Florida. Nursing degrees granted in Florida in the 2002-2003 school year..................................................41 3-3 Summary Overview of The Atlas of th e State University System of Florida. Summarized conclusions for the The Atlas of the State University System of Florida ...42 4-1 Map of SPC campuses and lear ning sites in Pinellas County............................................58 4-2 One and three year objectives for SPC..............................................................................59 4-3 Geographic dispersion of SPC co llege students throughout Florida................................ 60 4-4 Geo-coded address of SPC students living within Pinellas county aggregated to a grid of 1.5KM square cells.................................................................................................61 4-5a Tarpon Springs and Clearwater cam pus geographic distribution of student enrollment..................................................................................................................... .....62 4-5b Seminole and Gibbs campus geographic distribution of student enrollment....................63 4-6 Online student geogr aphic distribution..............................................................................64 4-7 Drive time zones for SPC learning sites............................................................................65 4-8 Estimated population growth of the co llege age population w ithin the county.................66 4-9 High school age percentage of change projection.............................................................67 4-10 Initial consideration of the northeast corner of the county................................................68 4-11 Regional population outlook..............................................................................................69 4-12 Lifestyle segmentation pa tterns throughout the county.....................................................70 4-13 Lifestyle group pattern s throughout the county.................................................................71 4-14 Spatial pattern of lifestyle segments five and ten for students within Pinellas County.....72

PAGE 9

9 4-15 Spatial pattern of lifestyle segments one and seven for students within Pinellas County......................................................................................................................... .......73 4-16 Student capture rate percentages by ZIP code...................................................................74 4-17 Positive and negative Black student capture rate throughout the county..........................75 4-18 Positive and negative Asian studen t capture rate throughout the county..........................76 4-19 Positive and negative Hispanic student capture rate throughout the county.....................77 4-20 Positive and negative White student capture rate throughout the county..........................78 4-21 Male / Female student capt ure rate throughout the county................................................79 4-22 Age group capture rate throughout the county..................................................................80 4-23 Geographic distribution of education throughout the county............................................81 4-24 Industrial influence in the county......................................................................................82 4-25 SPC objectives examined within this study.......................................................................83 4-26 Summary observati ons and suggestions............................................................................84

PAGE 10

10 Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Arts USING GEOSPATIAL REASONING IN INSTITUTIONAL RESEARCH: ST. PETERSBURG COLLEGE GEODEMOGRAPHIC ANALYSIS By Phillip Morris August 2007 Chair: Grant Thrall Major: Geography Geographic analysis has been adopted by busines ses, especially the retail sector, since the early 1990s. Higher education can receive the sa me benefits as have businesses by adopting business geography analysis and technology. Th e commonality between business geography and institutional research for higher ed ucation is that both have trade ar eas, both provide services to clients (students), and clients can be geographically id entified by their addresses as well as their psychographic profile. Among the valuable informa tion that institutions of higher education can create using business geography are psychogra phic profiles of the student body, commuting patterns, and potential enrollment based upon the underlying demographics of the institutions trade area. A benefit of this analysis is the ability to anticipate th e needs of the market. Understanding these geographic characteristics can assist in evaluating institutional objectives, and identify constraints on implementing these objectives. My research is intended to provide a ge neral guideline to geos patial reasoning in institutional research, assessment, and evalua tion. Through literary examination as well as through the use actual data from a community colle ge, the benefits of ge ospatial reasoning in institutional research will be identified. With in this report the st udent population of St.

PAGE 11

11 Petersburg College is specifically addressed an d analyzed for spatial patterns based on various characteristics.

PAGE 12

12 CHAPTER 1 GEOGRAPHY AND HIGHER ED UCATION: AN INTRODUCTION INTO GEOSPATIAL PRACTICE AND INSTITUTIONAL RESEARCH Foundations of Geospatial Reasoning Geography has assisted in decision making thr oughout development of m odern societies. Early explorers relied on their knowledge of geography and landscape to survive and succeed in global endeavors. Visual represen tations of the earth at various scales have been integral in planning and problem solving throughout histor y. The understanding of the geography of the earth has developed into many s ub-disciplines of geography and ot her scientific disciplines, some of which can be universal tools for decision making. Cartogr aphy, planning, remote sensing, geographic information science, and even epidemiology use maps and spatial information to solve problems and make advancements within society. In 1854 Dr. John Snow, sometimes referred to as the father epidemiology, used geospatial reasoning to discover the source of a cholera epidemic in a neighborhood of the Westminster area of London, England (Kovalerchuk 2005). By ma pping the infected indi viduals (Figure 1-1) Dr. Snow was able to discover a direct correla tion between the infection and the use of a single water pump proximate to the clus tering of the infections. This type of spatial reasoning has led to development of Geographic Information Systems (GIS). GIS can be defined as "a powerful se t of tools for storing and retrieving at will, transforming and displaying spatial data from the real world for a particular set of purposes. (Burrough 1998). GIS is commonly described as having 5 main components; data, hardware, software, procedures/goals/plan, people, and a network. These components help describe the what and where of particular features on the earths surface. GIS specializes in storing, analyzing, compartmentalizing, a nd describing the properties and attributes of a particular landmark or geographic occurrence (Bolstad 2002)

PAGE 13

13 Over the last half century improvements in technology have allowed for tremendous development of GIS using complex computer syst ems. Along with the development of GIS was the development of the idea that geography coul d become explanatory in nature as well as descriptive (Harvey 1969). On the rise of GIS Martin (2005, p. 491) writes: the surge of this activity has been notab le in geography, but it has been adopted substantially and simultaneously in the adjace nt social and environmental sciences, in which speed and accuracy of data arrangement and delivery are of significance, spatial relationships increase the complexity of sta tistical analysis, and he terogeneous behavior makes computer based modeling essential. (Martin 2005) Many other fields can benefit fr om geo-spatial analysis as we ll as sub-disciplines within geography. Education as a social science can benefit from the use of GIS methodology. Business Geography: With the understanding that geography can go beyond the initial descriptive phase of reasoning, the use of geospatial technology began to present itself in business and industry and the sub discipline of geography, business geography, developed. Geographic analysis has been used by businesses, especially the retail sector, since the early 1990s. A pioneer member of the discipline, Dr. Gr ant Thrall, offers a comprehensive definition: Business Geography integrates geographic an alysis, reasoning, and technology for the improvement of the business judgmental deci sionThis differentia tes business geography from the traditional descriptive or explanat ory objective of economic and urban geography. (Thrall 2002) Dr. Thrall, professor of geography and e xpert in the field of applied and business geography, describes geospatial reasoning as a hierarchy of steps that allow for improved decision making. The five steps include: De scription, Explanation, Prediction, Judgment, Management and Implementation (Thrall 1995). These steps can be app lied to decision making in a variety of fields in the private and public sector. Effective use of the five steps of geo-spat ial reasoning require th e application of best practice methods. The first is assessment of the trade area of the business. The trade area can

PAGE 14

14 be determined by evaluating the location of existing customers. Applications of GIS provide visual evidence of the trade area and allow the user to conduct experiments and draw relationships from information about the customer base and other geographic characteristics. An illustration of questions geospatial reasoning a nd technology can address when applied in the business sector include: What are the average drive times for customers from their home to the location of service? Are service providers drawing from each others market area? (termed cannibalization in business geography) Do certain market areas n eed greater focus on recrui ting and/or advertising? Are services provided based on geographic dema nd and are services offered in the correct locations? Are customers clustered by neighborhood, or uni formly dispersed around the trade area? Where are recommended sites for future expansion or serv ice area reduction? Do customers seek out services from the neares t service provider or do they skip one outlet for another? If so, why is one locatio n preferred to another? (Thrall 2002) The above questions are general and could be asked by various types of service providers. This thesis will ask these questi ons with reference to trade area analysis for higher education institutions. Through the use of GIS and method s proven to be successful in business geography, higher education will be shown to more eff ectively serve all segments of the population. Market penetration calculates how and where se rvices of a particular business or service provider are reaching prospective consumers. Evaluation of the underlying demographics of potential customers of the service provider can assist in reveal ing relationships and characteristics of the market th at can increase the level of ma rket penetration. Psychographic profiles are also used in business geography to examine customers and the population of the trade area.

PAGE 15

15 Psychographic profiles are a compilation of an individuals attr ibutes relating to personality, values, attitudes, in terests, and lifestyles. In bus iness geography these are used to explain market forces and predict and judge current and future business or real estate undertakings (Thrall 2002). Neighborhoods and i ndividual households can be organized into psychographic profiles and lifestyle groups. This type of analysis has proven to be effective in predicting and understanding consumer behavior as well as guiding decision making for retail and other business operations. ESRIs Business Analyst and MapInfos TargetPro are leading GIS software programs that use psychographi c analysis for business applications. Another focus of business geography is analys is of site location. Thompson Associates, partners with the successful geospatial analysis software company MapInf o, have delineated ten crucial steps for site selection (T able 1-1). Another effective met hod for site selection is the use recorded data surrounding successful service loca tions in order to draw analogies for possible future locations. This method is known as the analog method within business geography (Applebaum 1966). The use of these site selectio n methods help regulate resource allocation, provide intuitive learning site location strategy, a nd assist in providing appropriate services in relevant locations. Businesses have greatly benefited by havi ng adopted geospatial te chnology and reasoning. However, other institutions of our society have lagged behind adopting business geography practices. Higher education inst itutes can benefit from the methods mentioned above. Institutional Research In any organization communication, resear ch, and accountability are important for effective leadership and efficient operation. Hi gher Education is no ex ception. Higher education offices of institutional research provide these important functions within universities and colleges. A simple and effective definition of institutional research is research conducted

PAGE 16

16 within an institution of higher education in order to provide information which supports planning, policy formation, and decision making(S aupe 1981). The office of institutional research provides important information for acad emic administrators, accrediting agencies and governmental officials. This information allows individuals within these organizations to make informed decisions. Linking Geography to Higher Education. Higher education can receive the same benefits as have businesses by adopting bus iness geography analysis and technology. The commonality is that both have trad e areas, both provides services to clients (students), and clients can be geographically identifie d by their addresses. Among the valuable information that institutions of higher educati on can create using business ge ography are psychographic profiles of the student body, commuting patterns, and po tential enrollment based upon the underlying demographics of the institutions trade area. A bene fit of this analysis is the ability to anticipate the needs of the market. Understanding these ge ographic characteristics ca n assist in evaluating institutional objectives, and identify constr aints on implementing these objectives. My research is intended to provide a general guideline to geospatial institutional research, assessment and evaluation. Through relevant litera ry examination as well as through the use of data from a community college the benefits of geospatial reasoning in in stitutional research can be identified.

PAGE 17

17 Figure 1-1. Map of the choler a infected area in Westminste r region of London, England in 1854. (Retrieved on March 5th from http://en.wikipedia.org/wiki/Image:Snow-cholera-map1.jpg This image is in the public domain because its copyright has expired in the United States and those countries with a copyr ight term of life of the author plus 100 years or less) Broad street water pump (source of outbreak) Infected household Study area

PAGE 18

18 Table 1-1. Winning the site selection race The following 10 items represent a proven framewor k for assessing site se lection opportunities: 1. Collect demographic data 2. Build an inventory of competition 3. Generate market characteristics 4. Quantify market demand for financial products and services 5. Understand customer account information 6. Recognize customer behavior patterns 7. Identify potential s ite location opportunities 8. Conduct fieldwork 9. Identify site characteristics 10.Develop final recommendations for senior management (Thompson 2005)

PAGE 19

19 CHAPTER 2 LINKING GEO-SPATIAL RE ASONING TO HIGHER EDUC ATION: LITERATURE REVIEW This chapter is a literature survey of geospatial reasoning and technology applied to decision making in higher education. Trends an d themes used in business geography that can be applied in institutional research will be di scussed along with other applied geographic techniques. Also included in this chapter are signif icant functional foundations of institutional research. To give relevance and understanding to this investigation an identification of methods traditionally used for decision making, marketin g students and managing enrollment will be included. Advantages of GIS in Marketing GIS Techniques: The use of geography to assist in ma rket analysis can be traced back to the early to mid 20th century. Evolution within the field has occurred with the introduction of new technologies, particularly the automobile, wh ich significantly modifi ed the development of cities. This evolution of city development has expanded the discipline to studies on traffic flow as well as social and behavioral characteristics. Within business and industry these studies were supplemented with peoples lifest yle preferences. The growth in capabilities of the modern computers and the introduction of GIS (Ge ographic Information Systems) enabled many disciplines to significantly enhan ce data analysis, storage, and display with the added spatial component that is lacking from all other data analysis technologi es. In the business sector the use of geospatial technologies al ong with lifestyle segmentation profiles has greatly increased our understanding of trade areas and the demogra phic composition of those trade areas (Thrall 2002). Table 2-1shows the benefits of using GIS fo r marketing (Piccillo 1999). Prior to the

PAGE 20

20 ubiquitous use of GIS in marketing and busine ss some groundwork was established by highly influential theorists such as William Applebaum. William Applebaum, a geographer who was able to apply the techniques of geography to business and marketing, is regarded as a f ounding father of business geography. Numerous publications in academic journals as well as business and marketing volumes document his success at merging the discipline of ge ography to business practices (Ghosh 1987). Applebaums contributions to business geograp hy include the customer spotting method, the analog method, market penetration methods, and me thods for determining store location (Thrall 2002). The methods developed by Applebaum that most impact this study are the customer spotting and market penetration techniques. The customer spotting techni que involved surveying of th e customers or recording information from license plates or cars in the park ing lot in order to obtain customer addresses. This technique allowed Applebaum to determine the trade area for a particular establishment. A trade area is the geographic region for which a business draws most of its customers (Ghosh 1987). In discussions about customer spottin g Applebaum delineates trade areas into three distinct zones: 1. Primary Trade Area: This area is comp rised of 60-70% of the customer base (analysts now most frequently consider 80% capture of the customers as the primary trade area). 2. Secondary Trade Area: This area is comp rised of 15-25% of the customer base. 3. Tertiary Trade Area: This accounts for the residual customers. This consists of sporadic customers or out of town customers (Thrall 2002). The method used by Applebaum for customer spotting has given way to the much more efficient practices of collecting bank data, credit card data, and poi nt of sale data such as ZIP codes (Thrall 2002). Once this descriptive customer data has been collected, locations (whether

PAGE 21

21 it is address, ZIP code, or ZIP +4) can be programmed into a digital map using geospatial technology, including software a nd specialized databases. Th is process, called geo-coding, allows for the demarcation of trade areas as well as a variety of other sp atial analysis Knowing the location of the customer base gives the service provider the ability to describe the relationship between store and customer with other descriptive va riables. David Loudon (Loudon 1979) describes the collection of cognitive information by consumer researchers below. Consumer researchers who desire to know about their market more than just demographic characteristics may attempt to collect cognitive information; that is, information about consumers knowledge, attitudes, motivations, and perceptions. Merely observing consumers cannot fully e xplain why they behave as they do, and questioning often does not provide reliable an swers because of consumers inability or reluctance to reveal true feelings to an interviewer. Thus, researchers attempt to explore intervening variables potentially us eful in explaining c onsumer behavior by utilizing other techniques (Loudon 1979). Knowing the locations of the customers is a beginning step for market researchers. The intervening variables mentioned above can be summarized through th e use of lifestyle segmentation profiles. Lifestyle segmentati on profiles (LSPs) are classifications of a neighborhood that incorporates many different variables such as family status, income, consumer spending behaviors, media and advertising in fluences, and even leisure and recreational activities. These variables collectively repres ent the households in a particular neighborhood. The first step in creating LSPs is defining the s cale at which the profiles will be used. The scale can range from a Census block group to ZIP+4 designations. The sma ller the geographic area the more accurate the assigned LSP is likely to be. Census blocks are the smallest geographic area for which the Census reports data. The shapes of census blocks follow the geographic pattern of the streets, and are ge nerally in between intersections. Residences included are usually back to back rather than on opposing sides of the streets (for further expl anation on U.S. Census Bureau reporting units see (Thrall 2002) or (Hamilton and Thrall, 2004)).

PAGE 22

22 Once the geographic level has be en established the variables are clustered together based on similarities and a set number of clusters, or lifestyle segments are produced and one LSP is assigned to each geographic area. Once customer addresses are geocoded they can be appended to an LSP. These techniques will be illustrated and expanded upon in chapter four. Spatial Information in Education Support for GIS in Higher Education: Presently, there is a lim ited supply of resources that document research using GIS in institutional research. However, those that contribute significantly to the field of IR will be discussed in this segment of the paper. Topics that relate to the use of GIS in education will also be discussed. One such topic is equity in the administration of financial aid. A problematic issue for college administrato rs striving for financial aid equity is determination of student socioeconomic status (S ES). SES is currently difficult to measure for college students (Bailey 2006). Un less a student files for financia l aid administrators have found no easy way to determine his or her status, which de termines if the student needs to be awarded a grant or other funding. However this problem can be addressed using GIS. Using the students address and what can be determined from ne ighborhood analysis GIS researchers can determine a students SES (Crosta 2006). The Community College Research Center ( CCRC), regarded as one of the leaders in research on community colleges, is based at Co lumbia University in New York. Recently a study was conducted by the CCRC that details th e use of student addresses to determine socioeconomic status (Crosta 2006). The study was conducted using data from the Washington state Board of Community and Technical Colleges The goal of the study was to provide the board with accurate information on student SE S for the purpose of making informed policy decisions and improve service to residents of the state through the various community and

PAGE 23

23 technical colleges. The CCRC used two methods to achieve this goal. The first process is determining SES from student addresses and census block groups, and the second process involves using cluster analysis to cr eate segments of the population. The first process, deriving SES by address lo cation, is achieved thr ough the use of GIS. The process, summarized in table 2-1, defi nes student SES by using household income, education, and occupation as indicators. This use of appending census data to individual addresses is important because it gives decision makers specific information about characteristics of residents living at these addr esses and comparisons of tabulated data can identify patterns within a particular population. For this study the population is co mmunity college students of Washington State. The second method used to better unders tand the relationship of the population to characteristics of students is called cluster anal ysis. Cluster analysis is the use of multiple algorithms to group objects with similar characteri stics into categories. In this study cluster analysis was used for grouping census block grou ps that have similar traits into community clusters. Cluster analysis is very useful at grouping large da tasets, but the composition of the resulting groups depends on variables used and parameters designated. In the Washington study the variables used were demographic variable s relevant to community college educators (Crosta 2006). Using clusters, according to th e authors, is more descriptive and effective in determining SES than using the three indicato rs (income, education, and occupa tion) mentioned in the section described earlier. In the CCRC study the Washi ngton Census block groups were divided into 15 individual clusters with distin ctive characteristics. Once the clusters are determined, students within the state can be assigned a cluster based on home address. These clusters can also be used

PAGE 24

24 to determine the level of market penetration of particular demographics throughout the state and in indentifying particular communities where students may face barriers to college access (Crosta 2006). A summary of the steps involved in the cluster analysis is shown in Table 2-2. Further recommendations from this study will be discussed in the concl uding chapter of this paper. Further demonstration of the use of GIS in hi gher education can be se en by examination of the office of IR at the Univers ity of Memphis (U of M). Mem phis has used GIS to map students and alumni since 2001. The webpage for the IR office has a collection of interactive maps (http://oirmaps.memphis.edu/maps/map_index.htm ). This collection includes: U of M Undergraduate Re tention and New Enrollment, Fall 2005 (U.S.) Current Off-campus Class Locations and U nderrepresented Areas (Shelby County, TN) U of M Undergraduate Applicants / Accept ees / New Enrollees, Fall 2005 (Tennessee) U of M Graduate Applicants / Acceptees / New Enrollees, Fall 2005 (Tennessee) U of M Students by Home Zip Code Ar ea, Spring 2006 (Shelby County and nearby counties TN, AR, MS) Undergraduate Students by High School A ttended, Spring 2006 (Shelby County, TN) Foreign Graduate Students Attend ing U of M, Spring 2006 (World) Foreign Undergraduate Students Atte nding U of M, Spring 2006 (World) U of M Active Alumni (by zip code area), May 2004 (continental U.S.) The maps displayed on the University of Me mphis website demonstrate the capability of GIS to track and manage students and alumni. This gives university administrators the ability to easily see where students are coming from and as well as areas to target for alumni donations or areas with underrepresented segments of the population (Donhardt 2001).

PAGE 25

25 Maybe the most significant source of literatu re discussing the use of GIS in higher education is the Winter 2003 edition of the quart erly sourcebook sponsored by the Association of Institutional Research, New Directions in Institutional Research The individual chapters each detail a different utility of using GIS in institutio nal research (IR) and the benefits associated. Typical applications offered by the book include: Using Census data and existing knowledge of student demographics to inform decisions with consideration of planning and implementati on of recruitment strategies and tactics. Building student databases and displaying enroll ment trends with maps for visualization and analysis. Use of GIS in survey research for desi gn, analysis, and reporting of survey data. Campus planning and facilities managemen t; GIS in a traditional planning capacity. Mapping and analyzing alumni donation patterns. Through the uses of address data and demographic data alumni associations can better understand where to focus future campaign activity (Teodorescu 2003). Resource allocation in higher education is di rectly linked to stat e legislative bodies. Geographic techniques can be used by institutions to inform state legislators of the influence of their university on the le gislative districts throughout the st ate (Thrall and Me coli 2003). The University of Floridas student address database wa s recently used for this purpose. By spatially joining the addresses of the students to the state legislative districts, the University of Florida was able to show the impact of the school per di strict, and theref ore the value of the university throughout the state (Thrall and Mecoli 2003). Institutional Research and Decision Making The examples described in this chapter exemp lify significant justification for incorporation of geospatial methods in college and univers ity planning and decision making. The following section of this chapter focuses on key functions of institutional research and existing decision

PAGE 26

26 making methods. Administrativ e functions in enrollment management, assessment, marketing, and institutional objectives will be examined along with traditional methods of decision making for these functions will be discussed. Enrollment Management The Association of Institutiona l Research (AIR) is an excelle nt source for information and literature describing IR. A recen t publication from the AIR, The Primer for Institutional Research (Floyd 2005), aptly and comprehensively descri bes the key functions and duties related to IR. The Primer for Institutional Research analyses and reports on the current issues surrounding institutional research in higher ed ucation (Knight 2003). The book is a compilation of nine chapters written by practitioners within the field of institutional research. Each of the chapters addresses a separate issu e. According to the editor this book is designed to provide an introduction to some of the more common instit utional research issues, methods, and resources for newcomers and to provide means for veterans to update their capabilities This volume is in essence an update to previous publications of the same variety from the Association of Institutional Research in Tallahassee, FL. One of the chapters of the The Primer for Institutional Research is titled Enrollment Management. Enrollment management is defined in The Primer for Institutional Research as an institutional research function that examines, and seeks to manage, the flow of students to, through, and from the college. The educational pi peline is a term used to describe student recruitment processes. The que stions Who does the institution want to educate? and Who is available? determine the educat ional pipeline. Once these ques tions are addressed the next step is to identify data sources that summa rize how many pre-college students with these characteristics exist in the pipe line (Knight 2003). The use of ge ospatial techniques can greatly

PAGE 27

27 assist in this process by tailoring institutional programs to the currently and future enrolled students. There are certain measures that are effective in predicting student success in the collegiate setting (Knight 2003). High school GPA in combination with st andardized test scores is typically best at predicting college GPA and retention in the first year of college. However, the authors of the chapter on enrollment manageme nt show that this combination does not do a particularly good job predicting success and reten tion after the first year and does a relatively poor job of predicting which students will graduate af ter four to six years. This is evidence for putting less emphasis on merit based financial aid. If merit is not a good predictor of who will succeed and graduate, then more emphasis s hould be put into need based aid. Decision Making Traditionally the Institutional Research office has relatively few direct responsibilities, but this is changing. Some institutions combine IR with planning (Walleri 2003). Two planning considerations from the IR office that can be assisted with GIS are de mographic and population projections as well as enrollment analysis and fo recasting. Most IR offi ces do not incorporate a spatial component in the planning process, so to better understand the benefit of using a GIS in planning an examination of current considera tions for decision making in post-secondary education is presented below. Decision support may take many forms and various levels of analytic complexity (Howard 2001). According to Terenzini (Terenzini 1993) there are 3 levels of intelligence in administration in higher education. The first tier, technical intellig ence would include abilities in conducting surveys, cost benefit analysis, and data collection, manipulatio n, and analysis. This typifies the primary day to day ope rations of an IR office and consis ts of numeric tabulated data. The second tier, issues intelligence, includes a understanding of institutional decision making.

PAGE 28

28 This level of intelligence requires good communica tion and organizational skills along with the ability to integrate methodologies and knowle dge from multiple disciplines to design a study and interpret results. The highest tier, the contextual intelligence, encompasses the ability to effect change through the use of both qua ntitative and qualitative info rmation (Terenzini 1993). Understanding these three levels of intelligence can help unders tand the implicated role for the use of GIS in IR. The use of GIS, a methodology fr om an external discipline, can assist in both the technical and issues levels of intelligence through capabilities of data storage and analysis as well as the ability to design a study a nd interpret result s (Howard 2001). The use of GIS in education planning is can assist the persons responsible for the summarized process that leads to deci sion making presented below(Howard 2001). 1 The initial collection and storage of data. 2 The use of the resulting information to make decisions. 3 Effective communication of useful information for decision making. Depending on the nature of the decision, all three steps could benefit from the visual and analytical potential of geospatial analysis. Th ough not frequently used for most decision making models in the higher education system in the US GIS is a data system accompanied by spatially explicit methodologies that would be useful for the implementation of key components of decision making.

PAGE 29

29 MEDIA BUYS: Once the precise location of your customer base is known, you can make cost effective media buys that get the right message to the right target. For example, by using the knowledge gained from m apping their customer base, a marketer who wants to place inserts in a loca l newspaper can target the neighborhoods with the highest level of customers. Competitive analysis: Plotting the location of the competition (direct and indirect) on a map is much easier to understand than a list of locati ons on a report. You can adjust your marketing strategy to fit the number of competitors in the immediate geographic area. Drive-time analysis: Mapping your customer drive times is us eful for analyzing c annibalization issues, new store placement and competition. By plotting the drive-time area on a map, you clearly see any overlap among vari ous business locations and competitive sites. Barriers such as military bases airports, parks or college campuses all influence your customer's drive time, but those barriers are not apparent on reports. Market-entry planning: Using demographic mapping for market-ent ry planning clearly identifies sales potential in a region. For market entry, be sure to look at retail sales potential, lifestyle segments, propensity to use or purchase products and services, population, income, age, number of bu sinesses and competition. Using a thematic map, your market ing strategy can be tailored to fit the new market's unique characteristics. Budgeting: For comparisons, demographic mapping is in valuable. Analyzing maps for each geographic area enables marketers to deter mine where to allocate their budgets. Areas with high growth potential or hi gh sales potential are quickly spotted and marketing strategies can be adjusted accordingly. Lifestyle segmentation: Used in conjunction with dem ographic mapping, segmentati on is a great tool for identifying quality prospect s. Lifestyle segmentati on systems use demographic and aggregated consumer demand data to cl assify every household in the United States into a unique market segment. Each segment consists of households that share similar interests, purchasing pa tterns, financial behavior and demand for specific products and services. Figure 2-1. Advantages of dem ographics mapping and possible fu rther analysis (Picillo 1999).

PAGE 30

30 Table 2-1. Summary of step s to find socioeconomic status from block groups Summary of Steps to Find Socio economic Status from Block Groups 1. Acquire student addresses. 2. Geocode student addresses by converting th em into latitude and longitude points. 3. Create GIS data contai ning student latitude and longitude points. 4. Acquire Census geodata at block group level. 5. Match, or geographically intersect, st udent data points to block groups. 6. Assign SES variables from Census data at the block group level to each student. Table 2-2. Summary of steps to involved in cluster analysis Summary of Steps to Involved in Cluster Analysis 1. Decide geographic level. 2. Select variables. 3. Standardize variables. 4. Choose cluster methodology, and if necessar y, distance metric and linkage method. 5. Run cluster algorithm. 6. Indentify the number of clusters to be defined. Table 2-1, Table 2-2. Crosta, P., Leinbach, Ti mothy, and Jenkins, David (2006). "Using Census Data to Classify Community College Student s by Socioeconomic Status and Community Characteristics." Community College Res earch Center: Research Tools No. 1: 12.

PAGE 31

31 CHAPTER 3 CURRENT ISSUES IN HIGHER EDUCATION This chapter will bring current and common problems that exist in higher education into context with the notion that ge ographic techniques can benefit institutional research. By examining relevant reports and published litera ture, issues will be identified that can be addressed or improved upon through the use of geo-demographic analysis and geo-spatial reasoning. This survey of literatu re will establish grounds for the use of geography in analytical reasoning in higher education and examine existing studies and liter ature that address the use of geographic information systems in education. United States Higher Education Education is considered the ha rd core of human capital theo ry (Sahota 1978). With global specialization human capita l has become one of the most im portant factors of production from the local to global scale. With the focus on sp ecialization of production and particularly on the technology industry, education and the developm ent of specialized sk ills across the entire population is of increasing importance. Of particular interest for this stu dy is the distribution of access to higher education opportunities. This distribution has far reaching effects. Unequal distribution can lead to an in sufficiently educated population a nd thus affect the economic and social structure of society. Relative to most countries the United States has one of the most unique and successful higher education systems in the world (Clark 199 0). The success of the higher education system in the US is partially dependent on the st ructure of the system. The de-fragmented organizational structure, along w ith many other factors, has allo wed U.S. tertiary education systems to adapt and grow to the point where they are now some of the best in the world (Lucas 1994). The success of higher education in the U.S. is far from flawless. Current flaws that have

PAGE 32

32 been identified recently through research will be discussed here. This section of the study will include issues that can affect institutions throughout the countr y, but because of the geographic location of the study that will be examined in the next chapter, issues specific to the state of Florida will also be addressed. Inequality Inequality in higher educati on has always been an issue. Following World War II the Montgomery GI Bill challenged this issue and resulted in the many minorities having the opportunity to go to college. Although the GI Bill allowed minor ities the opportunity to attend college it did not have the effect of creating equa lity in tertiary education (Greenberg 1997). Since the GI Bill took affect state and fede ral governments have instituted policies and procedures such as the FAFSA (Free Applicatio n for Federal Student Ai d), Pell Grant, race based admissions, affirmative action in faculty hiring, and other means designed to assist the financially unprepared and raci ally underrepresented (Lucas 1994) Equal opportunity to attend college is still an issue of importance to the U.S. Department of Education and is addressed in a recent report published by this department. In September 2006 a commission formed by the Secretary of Education, Margret Spellings, released a report disc ussing the current state of hi gher education (Spellings 2006). The report describes the need for a new landscape [that] demands innovation and flexibility from the institutions that serve the nations learners thus implying that the current system is not adequately meeting the needs of students. Curr ently students are getting their education from a variety of sources, all of which should offer se rvices that accommodate needs for students from all backgrounds and categories. The report mentions the consumer driven environment and describes students as results driven. Students ar e receiving education from multiple avenues and institutions and forty percent of the 14 million unde rgraduates in the U.S. are attending two year

PAGE 33

33 community colleges (Spellings 2006). This em phasizes the need for a focus on the services provided by two year institutions. One section of the Spellings report discusses a ccess as a concern. On access for minorities the commission writes, we are especially troub led by gaps in college access for low-income Americans and ethnic and racial mi norities. The population in the U. S. is growing the fastest in the groups of people that fall into the category of non-traditional and unders erved students. This shift in demographics will constitute an ever in creasing proportion of the workforce, a workforce that needs proper education and training. The community college can play a great role in training and educating these groups of people, and according to the report provide a place to begin for many of these students (Spellings 2006). Evidence from the Spellings commission repo rt supports the fact that eligible young people from low income families are far less li kely to attend college than young people with similar qualities from high income families. The percentage of whites that obtain a bachelors degree by the age of 29 is around 34%, compared to 17% of blacks and 11% of Latinos. The most alarming finding is that low income high school graduates that score in the top quartile on standardized tests attend college at the same rate as high-inco me high school graduates in the bottom quartile on the same tests. This statisti c suggests that although low income graduates are capable of being successful in higher educati on, other factors (for th e most part financial restraints) limit these students and they are not afforded the same opportunities as high income students to further their education. This occurrence is further evidenced by a study published recently by an independent nonprofit organization called The Ed ucation Trust. The study di scusses the nations premier public institutions and shortcom ings in regards with equal a ccess. One section of the study

PAGE 34

34 reports on funding and figures surrounding financia l aid. Currently the average institutional grant aid for families earning over $100,000 annually is nearly $4,000, which is lower than the amount granted students from families earning $40,000 or less (Gerald 2006). Another finding from this study similar to reports by the Spel lings commission was that talented high income students were four times as likel y to end up at a highly selectiv e university than a low income students of equal talent (Gerald 2006). Similar findings in repor ts from differing organizations indicate that this trend is apparent regardless of source or data bias. Another study conducted by the Educational Te sting Service discu sses three distinct concerns that will cause consider able changes in the future of higher education (Kirsch 2007). The three forces include divergent skill distri butions, the changing economy, and demographic trends. An indicator that Americas skills are greatly varied is embodied by surveys that show that the U.S. has a degree of inequality (a re presentation of the gap between the least and most capable) that is among the highest in OECD (Organization for Economic Cooperation and Development) countries (Kirsch 2007). Signs that the nations economy is dramatically changing are also offered in the report Ameri cas Perfect Storm. Since 1950 the proportion of manufacturing jobs has dr opped from 33.1% to 18.2% in 1989 and by 2003 down to 10.7%. Twenty of the thirty million jobs that have been created since 1984 jobs associated with college level education. Not only is the U.S. losing low skill manufacturing jobs, it is at the same time gaining a high proportion of jobs that require advanced education (Kirsch 2007), evidence of the increasing need for further the education level of the workforce. In 1979, the expected lifetime earnings of a male with a bachelors degree wa s 51% higher than a male without a degree, and by 2004 this estimate stands at 96%. Without offe ring equal opportunity for further education the increasing gap in skills will only serve to prom ote economic disparities within the country and

PAGE 35

35 hence result in other social issues related to educational inequality (as discussed further in (Chakravorty 2006)). The demographic trends mentioned in the The Pe rfect Storm are also significant. By the year 2030 the Hispanic population will exceed 20 percent of the total population. The authors of this article cite The American Community Survey s as reporting that in 2004 approximately 57% of the 16-64 year old Hispanic population in the U.S. is foreign born and around half of these immigrants do not have a high school diploma. The inability to close the existing skills gap a nd significantly enhance the literacy levels of all Americans will result in demographic change s that leave the populatio n in 2030 with tens of millions of adults unable to meet the requireme nts of a new economy (Kirsch 2007). If these tens of millions of low income people are goi ng to narrow the skills gap and improve their literacy levels enough to survive th ey will need the help of ins titutions such as the community college. Through examination of recent studies on higher education and trends that affect higher education a pattern has been identified. An increa se in the number of citizens in need of basic skills, along with a trend in hi gher education for academic elit ism has created a gap between prepared and under-prepared, and this gap is ye t widening. Efforts from the premier state universities to elevate standings at the expense of equality of access have created a trend that flows down the hierarchy of higher education. This drive for presti ge has caused institutions to abandon traditional values and roles in the community for a better name amongst peer institutions. This effect, deemed missi on creep, effects even community colleges and community colleges and workforce training institu tions have the highest potential for decreasing the skills gap mentione d above (Campbell 2005).

PAGE 36

36 Another study recently completed by the We stern Interstate Commission for Higher Education (WICHE) discusses the financial ai d and student success (Hauptman 2007). This report again points to access as a major issu e for higher education. This report by WICHE makes some recommendations for improving access. One key recommendation related to this paper is the recommendation for improvement in collection, analysis, and presentation of data on how well federal and state support of po licies are targeted toward low income students(Hauptman 2007). The report goes on to discuss how currently aid programs are not well targeted toward the poor. According to the commission, this lack of targeting reinforces chronic inequalities at each st age of the educational pipelin e (Hauptman 2007). Addressing policy design and implementation, the author find s that availability of data, research, and insightful analysis is limited prior to legisl ative decision making and program implementation. GIS provides the tools that have insightfully info rmative analysis and presentation capabilities. State and Federal education agencies stand to improve decision making through the use of geospatial analysis. Prior to descriptive analysis the commission identifies three objectives, which are listed below: Strengthen educational opport unities for students through expa nded access to programs. Assist policy makers in dea ling with higher education and human resource issues through research and analysis. Foster cooperative planning, es pecially that which targets the sharing of resources. These three objectives are wort h mention here because the to pic of this paper, using geographic techniques to improve education decision making, is related. These objectives along with the commissions recommendations can be affected and improved upon through the use of GIS for targeting segments of the population that are underserved.

PAGE 37

37 Higher Education in Florida The study presented with this paper and discu ssed in the following chapter, deals with geodemographic analysis conducted for St. Petersburg College in Pinellas County, Florida. Giving consideration to the study, it is pertinent to disc uss issues in higher educ ation that specifically affect the state of Florida. Current issues along with a discussion of th e implication of the use of GIS will comprise this section of the paper. The changing demographics a nd increasing population for the st ate of Florida is a concern for the Florida Board of Governors (FBOG) for the State University System. In the last decade the growth of population of 18 24 year olds has grown in Flor ida by 24.6% and will see an increase in 19.5% (roughly 10.5% more than national projections) from 2004 to 2014 (Pappas 2007). This spike in college age population requires additional needs for access to higher education. The Florida Board of Governors recognized this and hired the Pappas Consulting Group to assist in planning for the future needs of college students within the state. The consulting group concluded investigation and pub lished a detailed report in January 2007. Recommendations from the group included the use of the California model of the 3 tiered system with clear delineation between research one, state college, and community college systems. According to the consulting group the us e of state resources fo r the additional openings of medical schools, the additional emphasis of st ate universities on research and the granting of baccalaureate degrees from community colleges constitutes mission leap (Pappas 2007). Designation of clear mission parameters will limit the unnecessary use of valuable and diminishing state funds. Programs such as the Bright Futures Program (BFP) are also addre ssed by the consulting group. Bright Futures was established to rewar d any Florida high school graduate who merits recognition of high academic achievement (Flori da Bright Futures Scholarship Program, 1997,

PAGE 38

38 p.1). To initiate the program existing merit schola rships in the state were combined and funding was provided by the Florida Lottery system. A summary of the requirements for high school graduates is listed in Table 41. The Pappas group found that th e program has spent 1.6 billion dollars since 1997, when the program was institut ed, and most recipients were non-need based students. Opposition for the BFP can be readily found as some researchers protest the use of state funds for students that do not need assistance. One study conducted by researchers at North Fl orida University provides evidence that indicate lottery-funded merit scholarships redistribute income from lower income, non-white, and less educated households to higher income white, well-educated households (Borg 2004). Within the sample chosen by the research team the authors find that high socioeconomic (SES) households receive a net program benefit from Br ight Futures while low SES households incur a net program loss. Numerous studies show that low-income households pay more in lottery taxes. The researchers from North Florida found that th ese low-income households are much less likely to receive a BF scholarship. They also found th at of the low-income households that do qualify for BF, they are more likely to receive the FMS scholarship that onl y pays 75% than highincome households (Borg 2004). Programs such as the BF in Florida can and should be replaced by programs that place as much emphasis on need as is on merit. Identifying the segments of the market that are in need can be done through the use of GIS. Planning can be greatly enhanced with the us e of information systems such as GIS. Considering the demographic changes projected fo r Florida over the next 10 to 15 years planning in higher education is vital. The Atlas of the State University System of Flor ida (Thrall 2005) exemplifies the merits of the use of geographi c technology for planning in higher education. In 2004 The Florida Board of Governors (FBOG) re quested consulting service from Dr. Grant

PAGE 39

39 Thrall for analysis of the geographic access to the state university system (SUS). Thralls analysis provides immense planning potential for the SUS, and can be duplicated by other university systems elsewhere. Thrall's (2005) re port uses a variety of methods for delineating trade areas for the universities in the SUS and addresses both the needs for key fields of education around the state and th e supply of these degree programs and graduates with these degrees. Examples of the work provided to the FBOG (Florida Board of Governors) can be seen in Tables 3-1, -2, and -3. This analysis provides further evidence of the utility in the use of GIS in planning for higher educati on at the highest level. Chapter Conclusion The important role has been established for geospatial analysis at the university system level and at the community college level, theref ore justification is given to the importance of effective resource allocation. Workforce de velopment programs and other fundamentals characteristics of the community college are ve ry important for advancement of society as a whole. This paper provides evidence for us e of geo-spatial applic ations in improving the administration of these fundamental operations in the community college and other realms of higher education. By examining the current issues in higher education a gap has been identified that geo-spatial science and techniques can begin to fill. Using data and collaboration with St. Peters burg College in St. Petersburg, FL, the following chapter will illustrate sp ecific benefits of using geos patial reasoning for planning, implementation of institutional objec tives, and enrollment management.

PAGE 40

40 Figure 3-1. Summary Overview of The Atlas of the State University System of Florida. This map shows the pr ojected increase in u ndergraduate aged population in Florida by the year 2010. (Thrall, G. 2005)

PAGE 41

41 Figure 3-2. Summary Overview of The Atlas of the State University System of Florida. This map shows the supply of nursing degr ees granted in Florida in 2002-2003 school year. The shaded areas are major SUS trade areas. The dark er shaded regions re present a higher pr ojected rate of change for the population age group 60 and over. (Thrall, G. 2005).

PAGE 42

42 Figure 3-3. Summary Overview of The Atlas of the State University System of Florida. This map shows the summarized conclusions for the The Atlas of the State University Sy stem of Florida. Gainesville. Th is type of analysis can be done by institutional researchers wi thin colleges and universities. (Thrall, G. 2005)

PAGE 43

43 Table 3-1. Florida Bright Futures award requirements. Award Award Level *GPA Required Credits Community Service Test Scores FAS 100% tuition and fees 3.5 15 various college prep courses 75 hours to be approved by the college 1270 SAT 28 ACT FMS 75% tuition and fees 3.0 15 various college prep courses (same) No requirement 970 SAT 20 ACT GSV 75% tuition and fees 3.0 15 various college prep courses (same) No requirement 840 SAT (Florida Bright Futures Sc holarship Program, 1997, p.1)

PAGE 44

44 CHAPTER 4 GEO-DEMOGRAPHIC ANALYSIS St. Petersburg College This chapter presents the geospatial analysis of St. Petersburg Co llege. Having introduced the history and importance of both geography and institutional research and described how the tenets of Business Geography can relate to curr ent issues higher education, this chapter will discuss the core analysis of this research. Th e study exemplifies the use of GIS in education planning. The chapter is organized as follows: firs t, information about the organization, history and mission of the college will be discussed. Se condly, the analysis completed for the college will be detailed along with an extensive appendi x of figures and maps. Lastly the summary findings for the study will be listed and discussed. Organizational Structure St. Petersburg College (SPC) is located in Pinellas County, Florida and offers services throughout the county from various campuses and service locations. Geographically, Pinellas County is a peninsula with the Gulf of Mexico to the west and Tampa Bay to the east. SPC is broken down into eleven learning si tes throughout the county. The term learning site is used because not all of the sites are cons idered campuses. According to Stan Vittetoe, the VP of Business Operations for St. Petersburg College, a campus mu st offer services that include student services, libraries, counseling, bookstore, ect. (Phone interview with Stan Vittitoe, Vice President of Economic Development, on October 9th, 2006) Four of the eleven learning sites are considered campuses. The four campuses includ e St. Pete Gibbs, Tarpon Springs, Clearwater, and Seminole along with the rest of the le arning sites can be seen in Figure 4-1. The administrative organization the college has a Board of Trustees that is a political subdivision of the state and part of the state co mmunity college system. The Board governs the

PAGE 45

45 college as a whole. The president of the coll ege is the secretary of the Board and is charged with the day to day operational management. A Provost, Vice Presid ent, or an Executive Officer supervises each cam pus or learning site. Historical Development In 1927 St. Petersburg Junior College opened as Florida's first two-year institution of higher learning. The college was based out of a wing of the new St. Petersburg High School. The initial enrollment was 102 students with a faculty of 14. Historical dates for the development of the college are listed below. 1931 Gained full accreditation 1948 Private college became public 1965 SPJC merged with African-American Gibbs Junior College 1990s SPJC occupied a dozen sites throughout the county June 2001 SPJC became St. Petersburg College, a four-year institution August 2002 SPC began offering fully accredited baccalaureate programs (SPC 2006) Values According to the information found on the colleg e website the purpose of the college is to provide access for: Students pursuing selected baccalaureate degrees associate degrees, technical certificates, applied technology diplomas and co ntinuing education within our service area as well as in the State of FloridaAs a comprehensive, multi-campus postsecondary institution, St. Petersburg College seeks to be a creative l eader and partner with students, communities, and other educational institutions to deliver enriched learning experiences and to promote economic and workforce development. (http://www.spcollege.edu/webcentr al/catalog/Current/mission_goals.htm ) This research has significance for SPCs missi on statement because of the ability to use geospatial tools to geographi cally organize data related to student profiles, community characteristics, and other inst itutions that will assist in promoting economic and workforce development.

PAGE 46

46 St. Petersburg College Analysis Focus of Analysis This study evaluates the geo-demographics of St. Petersburg College (SPC) in order to address specific and general questions for the benefit of the college. Geographically, SPC effectively serves the entire Pinellas County with higher education. This analysis draws attention to opportunities for SPC to increase market penetrati on within the county by greater targeting of particular population segments which are identifie d in this report. Also, several geographic areas warrant monitoring due to hi gh population growth, and the services provided are not increasing in proportion to the population change. These geographic areas are identified within this report. This analysis completed for SPC is broken down into five sections. The five sections are described below. 1. Data A description where the data came from and what needed to be done prior to analysis 2. SPC Objectives A discussion of some of the outlined one and three year objectives of the college. 3. Trade Area Assessment An overall look at the trade area of SPC and some underlying patterns and observations. 4. Market Penetration An assessment of th e proportion of the population captured by the college 5. Program Need A look at industry and educatio n indicators that may affect SPC planning. Data The data tables for this analysis include th e fall 2005 student enrollment for St. Petersburg College, as well as current demographic data tables for Pinellas County. The student information was provided to Dr. Grant Thrall by SPC. Prior to commencement of the analysis, Dr. Thrall restructured this data base to be in a GIS (geographic information systems) suitable

PAGE 47

47 format. The restructured database includes ag e, sex, race, and credit earned by campus. Geocoding is the process of converting address data into spatial data that can then be used for GIS analysis. The student addresse s included in the database we re geo-coded to enable geodemographic analysis. Once the addresses were geo-coded they were deleted along with any other personally identifying information. The data table reports were provided by Professor Thrall in the summer of 2006. Tables 4-2 and 4-3 show the distributions of the age and race of the fall 2005 SPC students, tabulated by the st udents home campus and by degree sought. The demographic data for the Pinellas County area was also provided by Professor Thrall. Through the use of a site license for ESRIs Bu siness Analyst demographic data was obtained. Using this technology all of the student data records were assigned lifestyle segmentation profiles (LSP) which are used in the analysis. Th is analysis also integr ates SPC objectives with the previously documented da ta tables (Kuttler 2006). SPC objectives: In the fall of 2006 Dr. Grant Thrall in troduced the St. Petersburg project to his seminar in business geography course. Collaboration with SPCs Vice President Stan Vittetoe allowed for the refinement of project objectives. Recently the college had been experiencing a decline in enrollment and thus SPC was interested in the type of services geospatial analysis could provide to the college in the form of consulting. A descriptive and analytical geo-demographic consulting project wa s undertaken in an effort to assist SPC. During the preliminary stages of the resear ch project a proposal was put together for review by the college. The proposal was develo ped partially from ideas taken from the St. Petersburg College 2006-2009 St rategic Directions and 2006-2007 Institutional Objectives (Kuttler 2006). The objectives and strategic di rections outlined by the college early in 2006 all have assigned priority designations with one being the highest priori ty and five being the lowest.

PAGE 48

48 In the proposal development stage an intention of the class was finding SPC objectives that were both high priority and approachable from a ge o-spatial perspective. Figure 4-2 shows an example of the college objectives that are focused on in this analysis. Trade Area Assessment Section I In order to get an overall idea of where the st udents are coming from the analysis requires assessment of the trade area. The trade area can be described as an area encompassing 80% of the customer base (Thrall 2002). In a community co llege this standard can be applied, but in the case of SPC the trade area is bound by the Pi nellas County boundary. Conversations with SPC administration determined that the community co llege system in Florida prohibits SPC from marketing outside of Pinellas County. Using this constraint the analysis for of this research project is for the most part limited to Pine llas County. Of the 22,456 students included in the database tables provided for the project, 18,035 are identified as havi ng an address within Pinellas County. Coincidentally, the Pinellas located students represent approximately 80% of the total student population which is the sa me benchmark used in business geography. Section II In this section of the anal ysis student geographic distri bution, drive time analysis, age distribution of students and psyc hographic composition of enrolled students will be examined. Based on these measures a better understandi ng of the population of St Petersburg College students can be developed as we ll as an understanding of what can be done by the college to better represent th is population. The majority of SPC students reside within the Tampa Bay area, most within Pinellas County, but many throughout the State of Florida. Figure 4-3 show s the home address of every SPC student. The learning sites ar e represented by a different colo r dots and the site that the

PAGE 49

49 student takes the most credits from determines the color of their dot. The online campuses have the potential to draw from areas outside Pinella s County. Moreover, online offerings can serve to attract more Pinellas County residents to SPC SPC online students are for the large part clustered around Tampa Bay. Distant students might be registering wi th their parents addresses, versus their own near campus a ddress. Figure 4-3 shows the exte nt of the student enrollment and reveals potential for marketing and recruitment from various clusters within the state. Student geographic distribution within the boundaries of Pinellas C ounty was evaluated by dividing the county into 1.5 kilo meter cells. The number of students within each cell was calculated. Figure 4-4 shows stude nt geography within Pinellas C ounty. The darker colored and elevated cells have more students than the light er lower cells. The ge ographic distribution of SPC students is clustered in the southern peri meter of the county, with SPC serving fewer students in the northeast. Also, areas of the s outheastern region of the county shows enrollment dropping to below 50 students per 1.5 KM cell. These areas may be commercial, and the database did not include addr esses for place of work. There is a strong correlation be tween enrollment of students a nd proximity to the campus. SPC does well in enrolling student s within close proximity to the campus. Online students however show greater geographic dispersal of ho me addresses than traditional campuses. Figure 4-5 shows student enrollment separately fo r the Tarpon Springs and Clearwater campuses, by ZIP code. The Seminole and online enrollment ar e shown in Figure 4-6. Each area of the county is revealed to be well served by at least one of the SPC campuses, with the exception of the northeast. The northeast is revealed to be part of the Tarpon Springs trade area, but comparatively few students are enrol ling at SPC from this area.

PAGE 50

50 An integral component of the community co llege mission is providi ng unrestricted access for citizens within the community (Brawer 2003). Access is directly rated to the cost of transportation and increased dist ance equals increased costs for students. Because contemporary transportation effectively requires commuting by car, analysis of drive times to each campus is important for mitigating the costs of transpor tation and improving access. Figure 4-7 shows drive time zones around each SPC lear ning site or campus. Each of the irregular polygons shows the distance a person could drive in seven minutes from the campus. The software used for this graph allows the user to determine the drive tim e and the speed of travel before creating the polygon. Once the zones were created the number of students living with in these seven minute zones were tabulated. Approximately 35% of th e total Pinellas County enrollment lives within seven minutes of a learning site or campus. Se ven minute zones were used here because more than seven minutes would create overlapping zones that would be less visually effective difficult to analyze. If the college were to continue with this reasoning, prospective new campuses would be located at Palm Harbor, D unedin, or Western Clearwater. According to the population growth estimates included in the databa se provided by ESRIs business analyst each of the ZIP codes of Pine llas County will see an average increase of 144 people between the ages of 15-45 by the year 20 10. Approximately 92% of SPC students fall into the age group 15-45 (Table 3-1). Some ZIP codes will see an increase of up to 430 people of the targeted group, college age population. Th e greatest estimated growth will be in the northern part of the county served by the Tar pon Springs campus (Figure 4-9). Throughout the analysis Market Segments of Potential Opportu nity (MSPOs) are identif ied. Figure 4-9 has a reference to zip codes that are designated in th is reference as MSPO. The color ramp used in

PAGE 51

51 Figure 4-9 is also used at various times in this analysis. The tra ffic light pattern of green and red was chosen to designate the highs and lows, or hot and cold areas of the county. According to Dr. Carol Weideman, Director of Institutional Research at St. Petersburg College, there has been a gra dual decrease in high school ag e population enrollment at SPC (Interview with Carol Weideman on March 6th, 2007). Figure 4-10 shows a high school age population projection from 2005-2010 for Pinellas C ounty. These areas can be targeted for recruitment to perhaps reverse the trend recently experienced by SPC. The percentage scale of projected change for the countys high school population is -25% to +30% broken down by zip codes. By multiplying the numbers of current population in the age group 10-15 by the projected growth per ZIP code the projecti ons were developed. The northern part of the county will have the greatest growth of high school aged population, while the central county is projected to have decreasing numbers of high school age populatio n. SPC does not capture a high number of students from the Northeast corner of the county while at the same time the analysis shows the growth rate of high school age population as high in the northeastern area (Figure 4-11). The northeast is a target of opportunity for SPC. Being an older, more established, and dens ely populated area, Pi nellas County is not projected to have as great a percentage of population change as surrounding counties. Nevertheless, the change in demographic com position of the county will significantly affect SPC. Pinellas County is a desirable destination, therefore it is reasonabl e to expect an ongoing process of densification in the county. Because of these fact ors examination in the overall population trends throughout Pinellas and the su rrounding counties is important (Figure 4-12). On average, each ZIP code within Pinellas C ounty is expected to increase in population 0.5% during the next five years annuall y. Figure 4-12 illustrates that the greatest increase is expected

PAGE 52

52 in ZIP code 34688, bordering Pasco and Hillsbo rough Counties; ZIP 34688 has an expected estimated growth of 4.5%. Both Pasco a nd Hillsborough Counties will experience high growth rates, particularly in areas adjacent to Pine llas County. While Pinellas County may be an effective trade area for SPC, SPC can better serv e the states population by carrying some of the burden of student enrollment in nearby adjacent areas. The adjacent areas shown in Figure 4-12 are MSPOs. Student demographics were also analyzed using market segmentation profiling. The reasoning of market segmentation profiling is that people with similar tastes, lifestyles, and behaviors seek others with the same tastes. These behaviors can be measured, predicted, and targeted (Thrall 2002). By examining these profile s with student data, we can better understand the economic landscape that SPC must draw their students from. Lifestyle Segmentation Profiles of SPC student s were calculated using ESRIs Tapestry LifeMode groups (ESRI 2006). Figure 4-13 shows ZIP codes color coded to dominant LifeMode group. LifeMode groups can be further decomposed into 64 more detailed segments (Figure 4-14). SPC trade area has 17 ZIP codes dominated by the LifeMode group Senior Styles, which is comprised of Rustbelt Retirees, Senior Sun Seekers, The Elders, and others. The majority of SPC students are from areas with aging populations. Projections indicate that the recent trend will conti nue over the next five to seven years. However, Figures 4-13 and 4-14 document that some areas within the SPC trade area are dominated by younger households with college age students that are potential SPC students. Examples of MSPO are high hopes, young and restless, solo acts, metropolis, great exp ectations, global roots. Figures 4-13 and 4-14 shows the locations of MSPO. By understandi ng the differences in th ese groups and knowing

PAGE 53

53 the locations of these areas, SPC management can better understand their trade area and develop recruitment strategies to appeal to their desired segment. The study also sought evidence of clustering of specific socio-demographics by campuses. This question is addressed by pr oducing a series of spider maps (Figures 4-15, 4-16). Spider maps draw lines from the students residence to their respective campuses. Almost 75% of all SPC students are contained within only four lif estyle segmentation profiles (LSP). LSP 5, or senior sun seekers, makes up 32% of the total en rollment. These students are classified as the senior population based on income, age, and housi ng type. LSP 10, or the traditional living, makes up 20% of the enrollment. This profile is classified as middle aged, middle income middle America. LSP 7, or the high hopes, ma kes up 14% of the enrollment and is classified as young households striving for th e American Dream. Lastly, LSP 1 students, or, the high society makes up 7% of total enrollment. This profile is classified as affluent, well-educated, married couple homeowners. After analyzing these student profiles and their respective campuses, it is difficult to identify significant clustering wi thin the largest LifeMode group, Senior Styles. The senior sun seekers are enrolling at campuses throughout the county and there is a gr oup that travels beyond their closest campus. This could be explained by the offerings of certain majors and classes at specific campuses. Clustering increases as the percent of enrollment from LSPs decrease. Market Penetration Capture rate is defined as the percentage of the total college age popul ation enrolled at St. Petersburg College. The market penetration sec tion of the analysis sh ows the student capture rate throughout county as well as the minority capture rate, age group capture rate, and the gender capture rate. For this analysis the market penetration capture rate is assessed by zip code. SPC enrolls from five to six percent of the co llege age population in 17 of 47 ZIP codes within

PAGE 54

54 the county (Figure 4-17). The average capture rate is 5.4% throughout the county. Two ZIP codes that fall into the MSPO classification are 33760 and 33762 and both capture roughly 3% of the college age population. SPC captures a hi gh of seven to eight percent in the two northernmost ZIP codes. The analysis shows that the total numbers of students attending from this area is small but relative to the small numb er of college age population in this area, SPC does relatively well enrolling st udents from this area. Higher education strives for r acial equality and public in stitutions value equivalent representation amongst the population. SPCs ins titutional objectives and strategic directions illustrate the importance for improvement in ethnic representation at the college (Figure 4-2). Capture rate analysis is perfor med to evaluate the success of SPC of enrolling an equivalent ethnic representation throughout the SPC trade area. Capture rate analysis differs from simple race percentages in that capture rates compare the percent of students enrollment from each race in a ZIP code to the percent of the actual popula tion of each race in a ZIP code. Capture rate analysis reveals if a r ace is under represented, or overrepresen ted. The traffic light style color ramp used in Figures 4-18 through 4-21 does an effective job of displaying under-represented (red) ZIP codes to those that are very close equal (neutral) and the over-represented areas (green). Figure 4-18 shows the college capture rate fo r black populations. Most of the ZIP codes fall within the -2.5 to 2.5 classification. The hi stogram shows a high frequency of zip codes near zero, which indicates an equiva lent representation by SPC th roughout the county. Figure 4-18 also shows the ZIP codes that are over-repre sented, and only two have 5-10% more blacks students enrolling. One ZIP code, 33760, is iden tified as an MSPO because the ratio is approximately -12% here.

PAGE 55

55 According to the analysis the Asian populations are also proportionally represented at the college (Figure 4-19). The ratio scale in Figure 4-19 varies from -1.5% to only 1.5% so throughout all zip codes the Asia n population is enrolling equa lly. Although there is little variance for the Asian population Figure 4-19 does show the areas for improvement within Pinellas County. The Hispanic population analysis reveals f our MSPOs (Figure 4-20). These four ZIP codes fall within the -4% to -5.5% range. Over all the county has 19 orange and red zip codes which indicate a limited proporti on of Hispanics in total enrollment. Figure 4-20 can give administrators at SPC an idea of where to emphasi ze effort for increasing Hispanic enrollment. The last racial group analyzed was the White race (Figure 4-21). Of the four groups examined, Whites are the least proportionally repres ented at SPC. The majority of the ZIP codes are red and orange and a few zi p codes (MSPO) are in the -10% to -15% range. One notable observation is that the only ZIP code, 33760, that has a greater proportion of whites enrolling at SPC than are living within that ZIP code is also the ZIP code that had the worst representation amongst the black population. This is an issue th at warrants further examination by institutional researchers at SPC. Figures 4-18 through 4-21 reveal that White and Hispanic populations were more extreme in rates of capture compared to Blacks and As ians. SPC is doing a good job in capturing the minority population when compared to each proportional representation of the population by ZIP code. The distribution for the minority captu re rate stays fairly close to zero. The female capture rate throughout the county is an average of about 10% higher than males (Figure 4-22). There are various ZIP c odes within Pinellas County that have a 15-20%

PAGE 56

56 higher female capture rate than males. This is es pecially true in the Southern part of the county. No significant ZIP codes have a higher percenta ge of males than females attending SPC. To analyze the distribution of student age ac ross the county, students were partitioned into three age groups (Figure 4-23). The first group, aged 20 and below, indicates a few MSPO ZIP codes with relatively low, zero to seven percent, captu re rates. The second age group 20-25 has a similar distribution throughout th e county with many of the same MSPOs identified. The third age group 26-45, shows a capture rate of 0-7% throughout all ZIP codes in the county. Even though the mean age for SPC students is 28, the anal ysis shows that none of the ZIP codes in the county capture more than seven percent of the population in this age group. Program Need Program need was also analyzed. Program n eed is used to describe the current demands for education throughout the trade area of the coll ege. The community college has multiple roles within the community and to best serve these ro les an understanding of the types of services in demand is important. This section of the anal ysis includes looking at education levels and industrial influences throughout Pinellas County. This analysis is important for recognition of where residents of the county have specific educ ational and program needs. Examination of Figure 4-24 can provide SPC info rmation on where residents of the county are lacking in education and areas where residents have various levels of education as specified by degree level. Figure 4-25 shows the breakdown of the industries in Pinell as County as well as the number of workers in that indus try. Areas with larger circle s represent a greater number of workers in that industry and ZIP code. These areas could be targ eted to provide industry specific services. An understanding of the influences of industries in the county and where specifically

PAGE 57

57 these industries are located can give SPC manage ment an idea of what programs and courses are in demand. Summary Findings Based on SPCs objectives (Figure 4-2), my analys is has revealed locations within Pinellas County where SPC can put marketing and re cruiting emphasis (Figure 4-26). Overall conclusions and revealed market segments of potential opportunity are shown in Figure 4-27. The northeast part of the c ounty will have high growth and should be monitored for neighborhood change for timing of a high visibility information cen ter. Demographics in this area are well suited for a community college. SPC should monitor areas in th e central county that have a low minority capture rate. Drive time analysis shows that prospective new campuses or learning site s could be located at Palm Harbor, Dunedin, and Western Clearwater to improve accessibility. An industrial centered campus might be considered as an intervening opportunity for workers in commercial areas in the south central area of the count y. Several market segments have been identified by ZIP code according to age, race, gender, capture rate, and major industry that have promise for potential enrollment (Figure 4-26). The following chapte r discusses the implications for this study along with possible recommendations for further analysis.

PAGE 58

58 Figure 4-1. Map of SPC campuses and learning sites in Pinellas County

PAGE 59

59 Figure 4-2. One and three year objectives for SPC The first table Priority Description Table describes what constitutes a level 2 priorit y, while the Objectives Table describes the objective. Both level 2 and are significant in this study.

PAGE 60

60 Figure 4-3. Geographic dispersion of SPC college student s throughout Florida. Each dot represents the geo-coded a ddress of a SPC student enro lled at SPC in Fall 2005. Yellow dots represent onlin e learning as major campus designation. Distant nonyellow dots likely represent true home lo cations, whereas the students campus address is more proximate to St. Peters burg. The color of the dot represents the campuses or learning site the student takes the most credit hours from. The number in parenthesis represents the number of student s taking a majority of their credits from that particular campus.

PAGE 61

61 Figure 4-4. Geo-coded address of SPC students living within Pinellas county aggr egated to a grid of 1.5KM square cells. The displayed data only includes the students enrolled in each of the top four SPC campuses in Fall 2005 Campuses

PAGE 62

62 Figure 4-5a. Tarpon Springs and Cl earwater campus geographic distri bution of student enrollment. The students used to produce these maps are taking most of their classes at that particular campus. Campuses

PAGE 63

63 Figure 4-5b. Seminole and Gibbs campus geogra phic distribution of student enrollment. Th e students used to produce these maps a re taking most of their classes at that particular campus. Campuses

PAGE 64

64 Figure 4-6. Online student geographic distri bution. Online students show greater geog raphic dispersal of home addresses than traditional campuses. The students used to produce th is map are taking most of their classes online. Campuses

PAGE 65

65 Figure 4-7. Drive time zone s for SPC learning sites.

PAGE 66

66 Figure 4-8. Estimated population growth of the college age populati on within the county. Campuses

PAGE 67

67 Figure 4-9. High school age percen tage of change projection.

PAGE 68

68 Figure 4-10. Initial cons ideration of the northeas t corner of the county.

PAGE 69

69 Figure 4-11. Regional population outlook.

PAGE 70

70 Figure 4-12. Lifestyle segmenta tion patterns thro ughout the county.

PAGE 71

71 Figure 4-13. Lifestyle group patterns throughout the county.

PAGE 72

72 Figure 4-14. Spatial pattern of lif estyle segments five and ten fo r students within Pinellas County.

PAGE 73

73 Figure 4-15. Spatial pattern of lif estyle segments one and seven fo r students within Pinellas County.

PAGE 74

74 Figure 4-16. Student capture ra te percentages by ZIP code.

PAGE 75

75 Figure 4-17. Positive and negative Black st udent capture rate throughout the county. Campuses

PAGE 76

76 Figure 4-18. Positive and negative Asian st udent capture rate throughout the county. Campuses

PAGE 77

77 Figure 4-19. Positive and negative Hispanic st udent capture rate throughout the county. Campuses

PAGE 78

78 Figure 4-20. Positive and negative White st udent capture rate throughout the county. Campuses

PAGE 79

79 Figure 4-21. Male / Female student capture rate throughout the county.

PAGE 80

80 Figure 4-22. Age group capture rate throughout the county. Campuses

PAGE 81

81 Figure 4-23. Geographic distribution of educa tion throughout the county.

PAGE 82

82 Figure 4-24. Industrial in fluence in the county.

PAGE 83

83 Figure 4-25. SPC objectives examined within this study.

PAGE 84

84 Figure 4-26. Summary obser vations and suggestions.

PAGE 85

85Table 4-1. Cumulative age di stribution of SPC students within Pinellas County Age Below 20 21-25 26-30 31-35 36-40 41-45 46-50 51-55 Over 55 Total Total 5285 5128 2321 1547 1236 1026 734 443 313 18033 Percent by Age 29% 28% 13% 9% 7% 6% 4% 2% 2% 100% Cum. % by age 29% 58% 71% 79% 86% 92% 96% 98% 100%

PAGE 86

86 Table 4-2. Fall 2005 distributi on of student race by campus. Campus White Black Hispanic Asian Am Indian Unknown Total (SPC Factbook 2005)

PAGE 87

87 Table 4-3. 2005 College wide opening fall h eadcount enrollment by age and by division. 19 or less 20-24 25-29 30-39 40-49 50-59 60&over Unknown Total (SPC Factbook 2005)

PAGE 88

88 CHAPTER 5 IMPLICATIONS AND CONCLUSIONS St. Petersburg College Study: After Action Review The analysis completed for St. Petersburg Colle ge represents the overall advantages gained from a geodemographic analysis report. Ge odemographic measurements are descriptive characteristics of a population, arranged and ordere d by scales of geography th at is meaningful to the analysis (Thrall 2002). In the SPC analysis the most meaningful geographic scale is ZIP codes because of the functional nature of posta l code designation for advertising and marketing campaigns. The use of maps for geodemographic an alysis provides a graphical representation of the landscape. Demonstration of the benefits of market analysis can be seen in the applications of business geography. Thrall 2002, discusses the importance of res earch in real estate market analysis: With this information in ha nd, an analyst then wants to pr edict the answers to questions such as, what will be the geography on the city in five or ten years? If a retail outlet is built here, will it be succe ssful? In the future where will th e population that is expected to be consumers or products live? Which neighborhoods will be on the decline, and which will be on the rise? Knowing the answers to such questions creates opportunity for the investor, and is the raison detre of the market analyst. (Thrall 2002) This reasoning is typically used for business an alysis of all sorts, but can be transformed for the use of higher education. For example, mu ltiple branch retail outle ts use this type of analysis as standard operating procedure and parallels can be drawn between multiple branch retail and multi-branch learning institutions. Both have customers with trade areas and both can benefit from market and trade area analysis. Ac cording to a recent re port by the Secretary of Education for the U.S., higher education is beco ming increasingly consumer market driven and students care less about whether a college is a for profit or pub lic institution, a predominantly online or brick and mortar instructional system and more about absolute results (Spellings 2006). This evolution of stude nts into consumers of educa tion along with the need for

PAGE 89

89 innovative methods of reaching the disadvantaged segments of the population give validation to the methods described in the SPC study. Base d on the findings, SPC will have a better understanding of the demographic characterist ics of the county and a better understanding of which markets and submarkets to target in order to meet the institutional objectives and goals. Typically an after action review a term borrowed from the US Army that describes a post training exercise review, is desi gned to 1) investigate and di scuss expected outcomes, 2) examine unexpected outcomes and 3) determine what could have been done to improve the sequence or lend strength or rele vance to the action in question. The first and third components of this review are useful for this study, however unexpected outcomes will not be discussed. Expected Outcomes The SPC study started as a project designed to investigate declining en rollment of students at the college. A primary inte ntion of the project was to give the college a better understanding of the student population based on the physical lo cation of the students and the characteristics and relationships that can be determined base d on location. Using Geospatial technology the students of SPC were assigned geographic coordinates and plo tted on map that contained ZIP codes and demographic data for the people within these ZIP codes. With the knowledge of how many students are living with in these ZIP codes in Pinellas C ounty, analytical relationships were asserted and recommendations a bout marketing and recruitment we re made to the college. Along with ZIP code boundaries the analysis also used ZIP + 4 boundaries and with the use of methods and tools of business geogra phy lifestyle segmentati on profiles (LSP) were assigned to all the students. This allowed for analysis of the student popul ation based on lifestyle profiles. ZIP codes were then labele d based on the majority or dominant lifestyle population. Specific reco mmendations were not given to th e college based on the geographic distribution of student li festyles, but a more general landscape vi ew is presented in the form of a

PAGE 90

90 map with ZIP codes shown for Pinellas County w ith designated LSPs as well as LifeMode groups. The maps presented to SPC provide ad ministrators with a de scriptive as well as predictive view of the communitys population demographics and the students demographic consistency. Another intended outcome of th e project is the designation of segments of the population that can be greater served by the college. The market segments of potential opportunity (MSPO) are segments used to show potential areas of improvement throughout the stages of analysis. The MSPOs listed with the various maps and are primarily applied to ZI P codes and designate segments of the population based on the criteria used to produce the map. Limitations and Possible Improvements The SPC study revealed a few notable points for improvement or change when this type of study is conducted in the future. For example, the study was broad in nature and from the beginning of the study there were few specific qu estions identified to answer using geographic technology. The Vice President of Economic Deve lopment for the SPC was general in requests for the study to decrease the rate of decline in enrollment. Narrowi ng the scope of the analysis to more specific issues was accomplished by viewi ng the one and three year objectives published by the college. The initial objectives could have been more structured to establish detailed criteria for investigation rather than using objectives delineated by the college. A general geodemographic analysis of the landscape provided an appropriate starting point. Future research that incorporates education pl anning and GIS should aim to iden tify key questions and concerns that can be used to drive further analyses of preliminary results. Another consideration to furthe r the validity of the such rese arch is the appropriateness of the demographic classifications for the educat ion institute in questi on. The zip codes, though useful for SPC postal marketing, were a course s cale dataset that did not allow for detail with

PAGE 91

91 regards to local populations. Use of census blocks as the core ge ographic unit of analysis would increase the resolution of data and thus increa se the detail and accuracy of findings. However, considering the marketing value of using ZIP code s, a combination of ZIP codes, ZIP +4 codes, and census blocks would provide the most useful data for both analysis and marketing purposes. Finally, the student population included in the analysis on ly accounted for students living within Pinellas County because SPC marketing is limited to within the county borders. Limitations to the extent of marketing is the resu lt of state rules used to maximize the resource allocation to community colleges by discourag ing competition between state funded higher education institutions. This means that although an alysis looks at a large portion of the student body, in fact the analysis relies on a sample of the study body and could be expanded (though not for marketing purposes) to include students tr aveling from outside of Pinellas County. Implications for Higher Education Planning Despite limitations, my research demonstrates that the geographic landscape is very important consideration for higher education for a variety of reas ons. As discussed in chapter two GIS can be used as proficient determinant of socioeconomic status (SES) (Pennington 2002). Maps can then be generated to display the SE S of distribution of st udents throughout the trade area of the college or even the state. Once student SES is mapped various components of the landscape such as demographic characterist ics, industry influences, environmental characteristics, and even lifestyle segmenta tions can be overlayed to examine geographic relationships (Pennington 2002). As described by Bailey 2006, SES is used for de termination of financial aid distribution and for many students SES is unknown. Determin ing SES through the use of GIS can improve reporting for tuition and financia l aid policy. Tracking the change s in the landscape of student SES or LSP over time is also a valuable use of GIS for higher education. Looking back at 1990

PAGE 92

92 or 2000 census data to visualize the changes in demographics and student locations over time gives state or institutional pla nners better information for fo recasting and future planning (Pennington 2002). Using GIS for management of programs is also advantageous for institutional researchers and college planners. Identifica tion of market segments and pa rticular neighborhoods for college expansion or reduction can result from the us e of GIS for institutional planning. A better understanding of service areas gi ves institutional rese archers more intuitive capabilities of providing services. For example, resources can be allocated for courses such as English as a second language (ESL) to campuses near ne ighborhoods dominated by non-English speaking families (Pennington 2002). Although infrequently used geospatial analysis and GIS serve to maximize decision making in higher education institutions. The adoption of geographic technology and me thodologies to the business realm resulted in the formation of business geography. Business geography is a sub discipline that focuses on identifying the needs of business and tailoring business to the client. Based on commonalties with the business world, including the need to provide services to a client and base decisions on trade area needs, higher education can benef it from utilizing geographic technologies and analyses. Understanding geographic characteri stics can assist in evaluating institutional objectives, and identify constraints on implemen ting these objectives. Geography is an integral component of decision making in business today, and should be incorporated in the decision making of public institu tions, including education.

PAGE 93

93 LIST OF REFERENCES Applebaum, W. (1966). Methods for determining store trade areas, marketing penetration, and potential sales Journal of Marketing Research 3: 127-141. Bailey, T., & Morest, Vanessa, Ed. (2006). Defending the community college equity agenda in the twenty first century Baltimore, MD: The Johns Hopkins University Press. Bolstad, P. (2002). GIS fundamentals: A first text on geographic information systems White Bear Lake, MN: Eider Press. Borg, M. O., & Stranahan, Harriet A. (2004). Some futures are brighter than others: the net benefits received by Florida bri ght futures scholarship recipients Public Finance Review 32 (1): 105-126. Brawer, F., & Cohe n, Arthur (2003). The American Community College San Francisco, CA: John Wiley and Sons Publishing. Burrough, P., & McDonnell, R. (1998). Principles of Geographical Information Systems New York: Oxford University Press. Campbell, D. (2005). Florida: bellwether of the future fo r the community college baccalaureate? Community College Weekly 17 : 5. Chakravorty, S. (2006). Fragments of inequality: Social, sp atial, and evolutionary analyses of income distribution New York: Routledge. Clark, B. (1990). Higher education American style: A structural model for the world. Educational Record (Fall 1990): 41-44. Crosta, P., Leinbach, T., & Jenkins, D. (2006). Using census data to classify community college students by socioeconomic status and community characteristics Community College Research Center: Research Tools No. 1: 12. Donhardt, G., & Keel, D. (2001). The analytical data warehouse: Empowering institutional decision makers. Educause Quarterly 24(4): 56-58. ESRI (2006). Community Tapestr y, Environmental Science Research Institute. Retrieved November 3, 2006, from http://www.esri.com/lib rary/brochures/pdfs/community-tapestryhandbook.pdf Floyd, D., Skolnik, M., & Walker, K., Ed. (2005). The community college baccalaureate: Emerging trends and policy issues. Sterling, VA: Stylus Publications. Gerald, D., & Haycock, K., (2006). Engines of inequality: Dimini shing equity in the nation's premier public universities. Washington, DC, The Education Trust : 25.

PAGE 94

94 Ghosh, A., & McLafferty, S. (1987). Location strategies for retail and service firms. Lexington, MA: Lexington Books. Greenberg, M. (1997). The GI Bill: The law that changed America New York, NY: Lickle Publications. Harvey, D. (1969). Explanation in geography, New York: St Martins Press Hauptman, A. (2007). Strategies for improving student success in pos tsecondary education. Changing Direction. Boulder: CO, Western Interstate Commission for Higher Education (WICHE) 24. Howard, R. (2001). Conceptual Models for creating useful decision support. New Directions for Institutional Research (Winter 2001) Volume 112: 45-55. Kirsch, I., Braun, H., & Yamamoto, K. (2007). America's perfect storm: Three forces changing our nation's future Princeton, NJ: Educatio nal Testing Service. Knight, W., Ed. (2003). The primer for institutional research: Resources in institutional research Tallahassee, FL: The Association of Institutional Research. Kovalerchuk, B., & Schwing, J., Ed. (2005). Visual and spatial analysis: Advances in data mining, reasoning and problem solving NY: Springer. Kuttler, C. (2006). St. Petersburg College 2006-2009 strategic directions and 2006-2007 institutional objectives St. Petersburg, FL : 1-20. Loudon, D. (1979). Consumer behavior: concepts and applications New York: McGraw Hill. Lucas, C. (1994). American higher education: A history New York, NY: St. Martin Press. Martin, G. (2005). All possible worlds: A hist ory of geographical ideas New York: Oxford University Press. Pappas, A. T. (2007). Proposing a blueprint for highe r education in Florida: Outlining the way to a long term master plan for higher edu cation in Florida. Tallahassee, FL, Pappas Consulting Group Inc Retrieved on March, 5th, 2007 from www.ufffsu.org/art/PappasBOGStructureReport.pdf Pennington, K., & Williams, M. ( 2002). Community college en rollment as a function of economic indicators. Community college journal of research and practice 26: 431-437. Piccillo, S. (1999). How marketers bene fit from mapping demonstrations. Marketing News. 33 : 15-16. Sahota, G. (1978). Theories of person al income distribution: A survey. Journal of Economic Literature 16: 1-55.

PAGE 95

95 Saupe, J. (1981). The functions of institutional research Tallahassee, FL: The Association of Institutional Research. St. Petersburg College (2006). 2006 St. Petersburg College Student Handbook. Spellings, M. (2006). A Test of leadership: Charting th e future of U.S. higher education A report of the commission appointed by s ecretary of education Margret Spellings. Washington DC: U.S. Department of Education. Teodorescu, D., Ed. (2003). Using geographic info rmation systems in institutional research, New directions for institutional research San Francisco, CA: Jossey-Bass. Terenzini, P. (1993). On the nature of institutiona l research and knowledge and skills it requires. The Journal of Research in Higher Education 34(1): 1-10. Thompson, A. (2005). Winning the site selection race: White paper. Ann Arbor, MI: MapInfo / Thompson 1-7. Retrieved March 5th, 2006 from www.mapinfo.com Thrall, G. (1995). The stages of GIS reasoning Geo Info Systems 5 : 46-51. Thrall, G. (2002). Business geography and new real estate market analysis NY: Oxford University Press. Thrall, G. (2005). Summary overvie w of the atlas of the state uni versity system of Florida. Gainesville, FL: Florid a Board of Governors : 19. Retrieved on January 4th, from http://www.clas.ufl.edu/us ers/thrall/fbog/index.htm Thrall, G., & Mecoli, N. (2003). Spatial analysis, political suppor t, and higher education funding, GeoSpatial Solutions 13(7): 44-47. Walleri, D. (2003). The role of institutional res earch in the comprehensive community college. Journal of Applied Research in Community College 11(1): 49-56.

PAGE 96

96 BIOGRAPHICAL SKETCH Phillip Allen Morris was born and raised in southwest Virginia. He graduated from Auburn High School in 1997 and immediately enliste d in the U.S. Army. After serving a two and a half year enlistment he was honorably discharged a nd began undergraduate study at Concord University in Athens, WV. While study ing at Concord, Phillip was a member of the West Virginia Army National Guard and a memb er of the university basketball team. Upon graduation from Concord in 2003 Phillip deployed with his National Guard unit as part of Operation Iraqi Freedom II. Phillip fi nished his enlistment with the National Guard in 2005, prior to enrollment in the Department of Geography at the University of Florida. Upon graduating with his MA from the University of Florida, Phillip will pursue a Ph.D. degree from the department of Education Administration and Policy.