a .7

n I. 5;. .-

....u.....-si3.i. u..$
.. . . ..
.. !§.iifii....!i5.i
. .. . l. 2...}

9 a
.v Ii! 1....
,xItai. ¢
:

 

 

 

._ ,V

.x:
5.5
t: iv

. .
.213“.
711‘ :l.
h...“ a:
an
.23.
.2 3v..."
.93.:- .L...
up.
1...an .
gait!!!

Funny.
.9

. x3
zit: » 5&4}?
.Qubam‘mf‘m nupwum
Fuv.tsu_.ﬂni 1&9...“
3... :3.” (-

suing»... 3...
3.2.3.9.. a .11....

This is to certify that the
dissertation entitled

AN ANALYSIS OF THE PERFORMING ARTS CONSUMER:
DEVELOPING MARKET SEGMENTS BY USING CHI-
SQUARED AUTOMATIC INTERACTION DECETION (CHAID)

presented by

SUNG HEE PARK

has been accepted towards fulﬁllment
of the requirements for the

PhD. degree in Park, Recreation and Tourism
Resources

 

 

  
 

/.,

' * U

1 2/23/2009

 

Doctoral Dissertation

MS U is an Affirmative Action/Equal Opportunity Employer

 

 

 

 

.LIBRARY
Michigan State
UI iIVGI'SIty

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 IUProj/AccaPresIClRCIDateOtie.indd

 

 

AN ANALYSIS OF THE PERFORMING ARTS CONSUMER:
DEVELOPING MARKET SEGMENTS BY USING CHI-SQUARED
AUTOMATIC INTERACTION DETECTION (CHAID)

By

Sung Hee Park

A DISSERTATION

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Park, Recreation and Tourism Resources

2010

ABSTRACT
AN ANALYSIS OF THE PERFORMING ARTS CONSUMER:
DEVELOPING MARKET SEGMENTS BY USING CHI-SQUARED AUTOMATIC
INTERACTION DETECTION (CHAID)
By

Sung Hee Park

Performing arts organizations have long been considered as an important element
to provide enjoyable and educational experiences in societies and communities. Due to
the current economic climate and potential issues, most performing arts organizations
have experienced revenue shortfalls and ﬁnancial deﬁcits, along with declining audiences.

The purpose of this study was to examine the performing arts market by
identifying and proﬁling performing arts consumers. By categorizing these consumers
into segments using Chi-squared Automatic Interaction Detection (CHAID) analysis
(Kass, 1980), this study explored how the interactions between consumers’ characteristics
and art-related experiences were associated with: 1) number of tickets purchased; 2)
donation activity; and 3) out-of-state and/or out-of-country performing arts attendance.
Three deCision trees were developed using Exhaustive CHAID, which increas ed the
likelihood of ﬁnding optimal association between predictors and a dependent variable.

Data were collected through an online survey of E-club members of the Wharton
Center for Performing Arts at Michigan State University who had purchased a ticket at
least once from the Wharton Center and were on its email list. The E-club newsletter
announcing the online survey and providing the link to the survey was sent to a total of

39,011 valid email addresses. Approximately 13,503 of the E-club members opened the

electronic newsletter, and a total of 4,744 members responded to the survey for a
response rate of 35.1% during 12 days in January — February 2007.

This study focused on the heavy consumers (those who purchased 10 or more
tickets during the last 12 months), who represented approximately 30% of the performing
arts market and accounted for over 50% of actual spending on tickets. The results
identiﬁed important predictors of performing arts consumers, based on number of tickets
purchased, and distinguished ten segments of heavy consumers. In addition, viable sub-
segments from two different decision tree models were identiﬁed for ‘the propensity to
donate to arts, heritage or cultural organizations’ and ‘the inclination to travel out-of-state
and/or out-of-country to attend live performances’ among heavy consumers. Chi-square
tests and one-way ANOVAS were conducted to examine statistically signiﬁcant
differences among derived segments with regard to their characteristics and behaviors
toward the performing arts and cultural pleasure trips.

The results of this research support previous empirical studies indicating that
socioeconomic characteristics are important predictors of consumers’ behavior patterns
toward the performing arts. This study provides crucial information about performing arts
consumers, as well as developing a useful approach that could permit performing arts
organizations to identify valuable consumers effectively. The identiﬁed target segments
and their proﬁles will be essential in building effective communication and promotion

strategies for various marketing purposes.

Copyright by
Sung Hee Park
2010

This work is foremost dedicated to my Lord who led the way and makes me be strong.

ACKNOWLEDGMENTS

I thank Dr. Edward Mahoney, Dr. Richard Paulsen, Dr. Richard Spreng and Dr.
Regina McNally for supporting my research goals and providing invaluable intellectual
stimulation, guidance and mentorship. Most of all, I would like to thank my academic
advisor and the chairperson of my dissertation committee, Dr. Edward Mahoney. Without
his constructive criticism and ﬁnancial support, I would not have been able to complete
my degree. I would like to thank Dr. Richard Paulsen and Dr. Regina McNally for their
consistent support and encouragement throughout my Ph.D. journey. I would also like to
thank Dr. Richard Spreng for his numerous helpful comments and academic expertise.

I would also like to thank all of my family and friends for their sustained support
inspiration and friendship. Their conﬁdence in my ability to succeed has been a constant
source of encouragement. Most importantly, I would like to thank to my mother, Kyung
Hee You, for her support, prayers, encouragement and patience during this long process.
Without her love and support, this would not have been possible. I deeply appreciate her
unwavering love and understanding, and dedicate this dissertation in her honor.

Special thanks go to Dr. Chang Huh and Chi-Ming (Allen) Hsieh who were a
continuous source for support and advice. I am also indebted to many other friends and
colleagues: Rev. Jong Bem Won, Dr. Ariel Rodriguez, Drs. Frank & Katharine Dennis,
Dr. Joey C. Miller, Dr. Wesley Pollitte, Dr. Chi-0k Oh, Mi-Ran Kim, Chifumi Takagi,
Eun-Jeong Noh, Ju Hyoung Han, Eunseong Jeong, Sanghoon Kang, Jin Won Kim and
Seong Ok Lyu. I have not recognized all of the wonderful people who have expressed

their endless encouragement and deep friendship; I can never thank them enough.

vi

TABLE OF CONTENTS

LIST OF TABLES .............................................................................................................. x
LIST OF FIGURES ......................................................................................................... xiv
CHAPTER 1 ....................................................................................................................... 1
INTRODUCTION .............................................................................................................. 1
Performing Arts ............................................................................................................ 1
Deﬁnition ............................................................................................................. 2

History .................................................................................................................. 3

Current Status ....................................................................................................... 4

Wharton Center for Performing Arts ................................................................... 7

Problem Statement ........................................................................................................ 8
Purpose of the Study ................................................................................................... 13
Study Objectives and Hypotheses ...................................................................... 14
Framework of the Study .................................................................................... 16
CHAPTER 2 ..................................................................................................................... 18
LITERATURE REVIEW ................................................................................................. 18
Overview ..................................................................................................................... l 8
Performing Arts Consumer Studies ............................................................................ 18
Market Segmentation .................................................................................................. 27
Deﬁnition of Market Segmentation ................................................................... 27
Segmentation Variables ..................................................................................... 28
Segmentation Methods ....................................................................................... 29
Empirical Studies in the Arts and Cultural Consumer Market .......................... 31
Chi-squared Automatic Interaction Detection (CHAID) Approach ........................... 34
The Application of CHAID in Market Segmentation Studies ........................... 35
CHAPTER 3 ..................................................................................................................... 41
METHOD ......................................................................................................................... 41
Overview ..................................................................................................................... 41
Data Source ................................................................................................................. 41
Study Population ................................................................................................ 42

Data Collection Procedures ................................................................................ 43

Study Instrument ................................................................................................ 44
Measurements ............................................................................................................. 47
Data Selection and Procedures ........................................................................... 47
Dependent Variables .......................................................................................... 50
Independent Variables ....................................................................................... 50

Other Variables .................................................................................................. 52

Data Analysis .............................................................................................................. 58
Exhaustive CHAID Analysis ............................................................................. 59

vii

Chi-Square Test ................................................................................................. 62

One-way Analysis of Variance (ANOVA) ........................................................ 62
CHAPTER 4 ..................................................................................................................... 64
RESULTS ......................................................................................................................... 64

Overview ..................................................................................................................... 64
Exhaustive CHAID Analysis of Heavy and Light Consumers ................................... 64
Heavy and Light Consumers of the Performing Arts ................................................. 65
Decision Tree for Heavy and Light Consumers ................................................ 68
Gain Summary of the Heavy and Light Consumers .......................................... 70
Two Exhaustive CHAID Segmentations .................................................................... 75
Donation Activity to Arts Organizations .................................................................... 75
Decision Tree for Donation Activity to Arts Organizations .............................. 76
Gain Summary of the Donation Activity ........................................................... 78
Proﬁles of the Propensity to Donate to Arts Organizations Sub-Segments ................ 80

Socioeconomic Characteristics .......................................................................... 80

Type of Performing Arts Attendance ................................................................. 86

Spending on Tickets to the Performing Arts ...................................................... 90

Performing Arts Information Sources ................................................................ 92

Web-Based Purchases to Art- and Culture-Related Products ............................ 94

The Importance of Cultural Attractions and Activities in Selecting Pleasure Trip

Destinations ........................................................................................................ 94

Characteristics of Cultural Pleasure Trips ......................................................... 98

Summary of Behavior Proﬁles of the Propensity to Donate Six Sub-

Segments... ........................................................................................................ 98

Attandance at Live Peforrnances in a Different State or Country ............................ 103

Decision Tree for Attendance at Live Performances in a Different State or

Country ............................................................................................................ 105

Gain Summary of Attendance at Live Performances in a Different State or

Country ............................................................................................................ 106

Proﬁle of the Inclination to Travel Out of State or/and Out of Country to Attend Live
Performances Sub-Segments .................................................................................... 107

Socioeconomic Characteristics ........................................................................ 108

Type of Performing Arts Attendance ............................................................... 114

Spending on Tickets to the Performing Arts .................................................... 118

Performing Arts Information Sources .............................................................. 119

Web-Based Purchases to Art- and Culture-Related Products .......................... 121

The Importance of Cultural Attractions and Activities in Selecting Pleasure Trip

Destinations ...................................................................................................... 121

Characteristics of Cultural Pleasure Trips ....................................................... 125

Summary of Behavior Proﬁles of the Inclination to Travel Out of State or/and

Out of Country to Attend Five Sub-Segments. .. .............................................. 125

Summary of Results .................................................................................................. 129

viii

CHAPTER 5 ................................................................................................................... 140

DISCUSSION AND CONCLUSION ............................................................................ 140
Overview ................................................................................................................... 140
Summary of Findings ................................................................................................ 142

Heavy Consumers of the Performing Arts (objective 1) ................................. 142
Donation Activity to Arts Organizations (objective 2) .................................... 143
Differences of the Propensity to Donate to Arts Organizations Sub-Segments
(objective 3) ..................................................................................................... 145
Attendance at Live Performances in a Different State or Country (objective 4)
.......................................................................................................................... 148
Differences of the Inclination to Travel Out of State or/and Out of Country to
Attend Live Performances Sub-Segments (objective 5) .................................. 149
Implications ............................................................................................................... 1 5 1
Heavy Consumers of the Performing Arts ....................................................... 152
Propensity to Donate to Arts Organizations Sub-Segments ............................ 154
Inclination to Travel Out of State and/or Out of Country to Attend Live
Performances Sub-Segments ............................................................................ 155
Limitation of the Study ............................................................................................. 158
Future Research ........................................................................................................ 159

APPENDICES ................................................................................................................ 163
Appendix A. The Online Survey Questionnaire ....................................................... 164
Appendix B. Descriptive Statistics of Performing Arts Respondents ...................... 187

Appendix C. Comparison of Heavy and Light Consumer of the Performing Arts... 196

BIBLIOGRAPHY ........................................................................................................... 199

ix

Table 1.

Table 2.

Table 3.

Table 4.

Table 5.

Table 6.

Table 7.

Table 8.

Table 9.

Table 10.
Table 1 1.

Table 12.

Table 13.

Table 14.
Table 15.
Table 16.

Table 17.

Table 18.

LIST OF TABLES

U.S. Spending on Admission to Performing Arts from 2001 to 2005 .............. 6
Overview of Research on Performing Arts Consumer Research ................... 25
Taxonomy of Segmentation Bases ................................................................. 29
Classiﬁcation of Methods Used for Segmentation ......................................... 31

Overview of Previous CHAID Approach in Market Segmentation Studies .. 40
Sample Size and Response Rate ..................................................................... 42
Distribution of Respondents Who had Purchased Tickets for Themselves or

for Their Household to Attend Live Performances during the Last 12 Month: 3
Deﬁnition of Variables Used in CHAID Analysis ......................................... 53
Survey Construction and Core Statistical Analyses in the Proposed Model .. 56
Signiﬁcant Predictors in Decision Tree for the Heavy and Light Consumers 68

Results of the Misclassiﬁcation Matrix .......................................................... 7O

Gains Summary of the Heavy and Light Consumers of the Performing Arts

......................................................................................................................... 72
Summary of the Nodes for the Heavy Consumer (HC) Groups of the

Performing Arts .............................................................................................. 73
Signiﬁcant Predictors in the Decision Tree for Propensity to Donate ............ 76
Results of the Misclassiﬁcation Matrix .......................................................... 78
Gains Summary of the Donation Activity ...................................................... 79

Socioeconomic Characteristics of the Propensity to Donate Six Sub-Segments
......................................................................................................................... 81

Summary of the Socioeconomic Characteristics of the Propensity to Donate
Six Sub-Segments ........................................................................................... 84

Table 19.

Table 20.
Table 21.
Table 22.

Table 23.

Table 24.

Table 25.

Table 26.

Table 27.

Table 28.

Table 29.

Table 30.

Table 31.

Table 32.

Table 33.

Table 34.

Type of Attendance at Live Performances by the Propensity to Donate Six

Sub-Segments ................................................................................................. 87
Spending on Tickets by the Propensity to Donate Six Sub-Segments ........... 91
Information Sources by the Propensity to Donate Six Sub-Segments ............ 93

Web-Based Purchases by the Propensity to Donate Six Sub-Segments ......... 95

The Importance of Cultural Attractions and Activities in Selecting Pleasure

Trip Destinations by the Propensity to Donate Six Sub-Segments ................. 97
Characteristics of Cultural Pleasure Trips by the Propensity to Donate Six
Sub-Segments ................................................................................................. 99
Behavior Proﬁle toward the Performing Arts and Pleasure Trips of the
Propensity to Donate Six Sub-Segments ...................................................... 100
Signiﬁcant Predictors in the Decision Tree for Attendance of Live

in a Different State or Country ...................................................................... 103
Results of the Misclassiﬁcation Matrix ........................................................ 106

Gains Summary of the Attendance at Live Performances in a Different State
or Country ..................................................................................................... 107

Socioeconomic Characteristics of the Inclination to Travel Out of State and/or
Out of Country to Attend Live Performances Five Sub-Segments ............... 109

Summary of the Socioeconomic Characteristics of the Inclination to Travel
Out of State and/or Out of Country to Attend Live Performances Five Sub-
Segments ...................................................................................................... 112

Type of Performing Arts Attendance by the Inclination to Travel Out of State
and/or Out of Country to Attend Live Performances Five Sub-Segments
....................................................................................................................... 115

Spending on Tickets by the Inclination to Travel Out of State and/or Out of
Country to Attend Live Performances Five Sub-Segments .......................... 118

Performing Arts Information Sources by the Inclination Travel Out of State
and/or Out of Country to Attend Live Performances Five Sub-Segments 120

Web-Based Purchases by the Inclination to Travel out of State and/or Out of
Country to Attend Live Performances Five Sub-Segments .......................... 122

xi

Table 35.

The Importance of Cultural Attractions and Activities in Selecting Pleasure
Trip Destinations by the Inclination to Travel Out of State and/or Out of

Country to Attend Live Performances Five Sub-Segments .......................... 124
Table 36. Characteristics of Cultural Pleasure Trips by the Inclination to Travel Out of
State and/or Out of Country to Attend Live Performances Five Sub-Segments
....................................................................................................................... 126
Table 37. Behavior Proﬁle toward the Performing Arts and Pleasure Trips of the
Inclination to Travel Out of State and/or Out of Country to Attend Live
Performances Five Sub-Segments ................................................................ 127
Table 38. Market Relevant Proﬁles and Marketing Mix Strategies for the Heavy
Consumer Nodes ........................................................................................... 130
Table 39. Market Relevant Proﬁles and Marketing Mix Strategies for the Propensity to
Donate to Arts Organizations Six Sub-Segments ......................................... 134
Table 40. Market Relevant Proﬁles and Marketing Mix Strategies for the Inclination to
Travel Out-of-State and/or Out-of-Country to Attend Live Performances Five
Sub-Segments ............................................................................................... 1 3 7
Table 41. Results of the Testing of Three Hypotheses ................................................. 139
Appendix Bl. Socioeconomic Characteristics of the Respondents ............................. 187
Appendix BZ. Household Information for the Online Survey Respondents ............... 189
Appendix B3. Consumption Pattern of Live Performances during the Last 12 Months
.............................................................................................................. 190
Appendix B4. Formal Arts Education of Respondents ................................................ 191
Appendix B5. Current Art Education of the Respondents ........................................... 192
Appendix B6. Art Product Purchase Experience of the Respondents ......................... 193
Appendix B7. Cultural Membership Status of the Respondents ................................. 194
Appendix B8. Volunteering Status to Arts and Cultural Organizations of the
Respondents ......................................................................................... 1 94

Appendix B9. Donation Activity to Arts and Cultural Organizations and Causes of the

Respondents ......................................................................................... 1 95

xii

Appendix Cl. Comparison of Socioeconomics, Household Information and Art-related
Experiences between Heavy and Light Consumers of the Performing
Arts ....................................................................................................... 196

xiii

Figure 1.
Figure 2.
Figure 3.
Figure 4.

Figure 5.

LIST OF FIGURES

Framework of the Study ................................................................................. 17
Overview of the Analytic Procedures of the CHAID Method in the Study... 49
Decision Tree for Heavy and Light Consumers of the Performing Arts ........ 67
Decision Tree for Donation Activity to Arts Organizations .......................... 77

Decision Tree for Attendance at Live Performances in a Different State or
Country ......................................................................................................... 104

xiv

 

 

CHAPTER 1

INTRODUCTION

Performing Arts

Performing arts organizations have long been considered as an important element
to provide enjoyable and educational experiences in societies and communities. A
performing arts organization is a public or private, non-proﬁt or for-proﬁt entity that
facilitates exchanges between artists and audiences through creative, educational, and
performance opportunities (Hager & Pollak, 2002). Performing arts organizational
characteristics differ from their exploring variations among programs and managerial
styles. The primary objectives of performing arts organizations are to increase
proﬁtability and performance by increasing ticket sales and developing audience loyalty
(Rentschler et al., 2002). The concept of performing arts organization marketing is
explained as “achieving organizational goals based on the needs and wants of target
markets, and delivering the desired satisfactions better than competitors” (Kotler &
Armstrong, 2006, p.10). The application of marketing theories by arts organizations
implies an exchange relationship between consumers and organizations (MejOn et al.,
2004)

Today’s consumers have unlimited interests and partake of a variety of sectors of
cultural life. It is hard for performing arts organizations to survive if their service
offerings are based solely on the subjective motivations of loyal customers (Cuadrado &
Molla, 2000). Generally, performing arts organizations use attendance levels, venue
occupancy rates, subscription purchases and the number of members or friends as

important performance measures (Rentschler et al., 2002). However, this information is

not sufﬁcient to attract and satisfy performing arts consumers and build customer-
centered marketing management. Understanding the performing arts consumers—who
they are, what they purchase and why and how they process decision-making—is
essential for performing arts organization marketers and Operators. Thus, performing arts
organizations could make effective marketing decisions and develop efﬁcient strategies
to increase high levels of consumer satisfaction and retention (Clopton et al., 2006;
Colbert, 2003). This study widens understanding of performing arts consumers using a
relatively new method of market segmentation.
Deﬁnition

The term “performing arts” is described as “theater, music, opera, and dance from
the traditional high arts to the popular arts, including live arts performed in all venues and
non-live arts through all forms of mass media” (McCarthy & Jinnett, 2001, p. 5—6).
Performing arts include all combinations of these genres and can be referred to as the
high, popular, and folk arts aspects of the performing arts market. This market is aimed at
a wide range of interests, attracting consumers who expect enjoyable and educational
experiences. Performing arts primarily provide a show-like experience while fulﬁlling the
cultural and artistic goals of attendees, contributing to education, tourism, and the overall
economy and well-being of the community (Hume & Mort, 2008; Caldwell &Woodside,
2003)

According to the New Zealand Framework for Cultural Statistics (2003), the
performing arts can be broken into three sub-categories: (1) performing arts, which
include theatrical performance, dance, opera, theatrical music, and the performing arts of

other ethnic and cultural groups; (2) music, which includes primary music creation,

I — _.- —-—H—“"‘

 

popular music performance, classical music performance, and the recording, publishing,
and retailing of music; and (3) services to the performing arts, which include venues and
other support services. In sum, performing arts provide virtuosity, artistry, aesthetics, and
interpretation of art experiences in the most intimate way — through the bodies and
creative intelligences of performers (Royce, 2004).
History

The classical period of performing art began with the tragic poets starting in
Greece during the 6m-century B.C. During the Dark Ages, performing art was limited to
religious historical enactments and morality plays, which were organized by the Western
church. AS the Renaissance began, the arts were revived and spread throughout Europe.
European plays incorporating dance were performed, and the ballet form was created.
Several other types of art were introduced in the 16m-l7th centuries, including theatrical
companies, opera, and the Elizabethan masque, which featured music, dance and
elaborate costtunes. Also during this time, the traditional theater form was established
and formal dance instruction was introduced by the French. After performing arts
appeared in the English language during the 18‘h century, they were developed for all
social classes; this led to the establishment of several variety theaters, such as the Boston
Symphony Orchestra in the US during the 19th century.

In the 19305, Jean Rosenthal introduced modern stage lighting, changing the
nature of the stage Broadway musicals became a US. phenomenon. After World War II,
performing arts were highlighted by the resurgence of both ballet and opera in Europe
and in the United States. Since postmodernism dominated the performing arts in the

19603, there was an unprecedented proliferation in the number of performing arts

 

organizations, the sizes of their audiences, and the level of contributions to the arts.
Performing arts continued to grow in these aspects until the mid-19803: the number of
professional orchestras had increased from 58 to more than 1,000; the number of opera
companies increased from 27 to more than 110; the number of dance companies
increased from 37 to 250; and the number of professional resident theater companies
increased from 12 to more than 400 (Kotler & Scheff, 1997). By 1987, ticket sales to
non-proﬁt performing arts events had exceeded those of sport events.

Growth dropped in the mid-19903, and many performing arts organizations lost
ground in their efforts to attract and retain customers due to competition for leisure
activities and lifestyle changes. There were spiraling expenses, deﬁcits of critical
proportions, shrinking contributions, and stagnant audience sizes (Kotler & Scheff,
1997). Since the turn of the 21St century, performing arts organizations have devoted
themselves to offering high arts by broadening their socio-demographic base, improving
customer service, and making greater use of the intemet and marketing information
systems in order to attract more potential audiences (Colbert, 2003). Performing arts have
not only provided an opportunity for the local community and enthusiastic fans to attend
events, but have also attracted various cultural tourists willing to spend time and money
on cultural festivals and special arts events. In order to satisfy consumers’ needs and
preferences in a broader environment, the performing arts must continue to move and
change (Bernstein, 2007).

Current Status
According to the Arts & Economic Prosperity 111 Report (2007), the nonproﬁt arts

and culture industry grew by 24% during 2000-2005, from $134 billion to $166.2 billion

in economic activity: spending by nonproﬁt arts and cultural organizations grew by 19%
during the same period, from $53.2 billion to $63.1 billion. Arts and cultural attendees
generated an estimated $103.1 billion in event-related spending for local merchants and
their communities in 2005 — a 28% increase from $80.8 billion in 2000. Speciﬁcally, the
average spending of non-local attendees was $40.19 per person, compared to $19.53 per
local attendee in 2005.

The Performing ARTS Research Coalition’s (PARC) extensive 2003 research,
conducted in ten communities across the United States, indicated that attendance at
professional live performing arts events was an activity enjoyed at least occasionally by a
signiﬁcant majority of adults (61-78% of respondents) (Bumstein, 2007). According to
Nichols (2005), 65 million US. adults attended some type of live performing arts activity
in 2002. Speciﬁcally, 35 million attended a musical stage play or operetta, 25 million saw
a non-musical stage play, 24 million attended classical music performances, 22 million
attended jazz performances, 13 million saw other types of dance events, 8 million
attended ballets, and 7 million saw operas.

The Consumer Expenditure Survey reports that US households spent an average
of $2,218 on entertainment and paid $92 for admissions to movies, theater, opera, and
ballet in 2004. The same year, households spent an average of $36 on sports admissions,
of $18 on video game hardware and software, and $43 on videotapes and disks (Nichols,
2006a). According to a report of the National Endowment for the Arts (Nichols, 2006a),
consumer expenditures on arts performances reached $12.7 billion, and spending on
admissions to performing arts totaled $42.80 per person in 2005. When adjusted for

inﬂation, year-over—year spending held ﬁxed, and real spending on performing arts

admissions in 2004 moved slowly by 0.9%, followed by no growth in 2005 (Table 1).
Although consumers spent less on arts performances, total recreation spending grew, and
consumers spent amply on other forms of entertainment such as books, audio, and visual

goods.

Table 1. US. Spending on Admission to Performing Arts from 2001 to 2005

 

 

 

 

 

 

 

 

Year Spending on Admission to Per Capital Spending on Percentage change
Performing Arts Admissions to Performing Arts in Real GDP and
(in billions of dollars) Consumer
Spending on
Admissions to

Nominal Real Nominal Real Performing Arts
Events
2001 $10.9 $10.5 $38.2 $36.8 1.9%
2002 $11.7 $10.9 $40.6 $37.8 3.8%
2003 $11.9 $10.6 $40.9 $36.4 -2.8%
2004 $12.4 $10.7 $42.2 $36.4 0.9%
2005 $12.7 $10.7 $42.8 $36.1 0.0%

 

Source: US. Department of Commerce, Bureau of Economic Analysis and US. Census
Bureau (http://www.census.gov/svsd/www/services/sas/sas_data/7 l/2006_NA1CS71 .pdf)
Notes: “Real” refers to estimates measured in chained 2000 dollars to control for inﬂation.

On the other hand, The Economic Census shows that there were 9,303 performing
arts companies in the US in 2002 (Nichols, 2006b). Of these, 3,753 were classiﬁed as
individual musicians or music groups and 3,222 were theaters; together, these two groups
represented nearly 75% of all performing arts companies. Symphony orchestras and
chamber groups constituted 9% of the total, and dance and opera companies made up 6%
and 1.9% of the total, respectively. More than 500 organizations (5.5%) were other
performing arts companies (excluding circuses). There were also a noteworthy
geographical concentration of the performing arts companies, namely, - there were 1,659
and 1,330 such companies in California and New York, respectively; representing nearly

32.1% of all performing arts companies nationwide (Nichols, 2006b).

Wharton Center for Performing Arts

The Wharton Center for Performing Arts (Wharton Center) is a not-for-proﬁt
performing arts center that has been afﬁliated with Michigan State University since 1982.
Containing 2,500 seats, the Wharton Center is one of the largest performing arts venues
both in Michigan and among Big Ten institutions. The Center presents a diverse selection
of performing arts experiences, providing lifelong learning programs for people of all
ages. The mission of the Wharton Center is “to enrich the lives of Michigan residents and
strengthen the value of the arts in everyday life by serving as a leading resource for
renowned arts entertainment and education programs” (Wharton Center for Performing
Arts, 2008).

According to the Wharton Center’s 2007-2008 Annual Report (2008), there were
264 ticketed events during this time period, including approximately 259,000 in ticket
sales and $11.6 million in total revenue. Speciﬁcally, $7.9 million was generated from
ticket sales, $1.3 million from donations/sponsorship/grants, $1.2 million from rental/box
ofﬁce fees, and the remainder from other listed revenue sources. The center has ﬁve
major venues for selling tickets: intemet sales (44%), subscription sales (26%), phone
sales (13%), lobby sales (10%), and group sales (7%). During 2006-2007, intemet sales
increased by 6% and lobby sales decreased by 6%. Moreover, Wharton Center’s eClub
membership expanded dramatically from 39,700 members in 2007 to over 72,000
members in 2008. In addition, more than 1.4 million targeted e-commerce messages were
sent to consumers throughout the season (2007-2008).

In 2007-2008, the Wharton Center’s Advisory Council, whose members enabled

fundraising and contributions, was comprised of 35 mid-Michigan business and

 

community leaders committed to the arts. Furthermore, the Wharton Center received gifts
from individuals, corporations, and foundations, which accounted for 15% of the budget.
Private gifts (including identiﬁed bequests to the center) exceeded $5.2 million.
Speciﬁcally, annual giﬁs totaled nearly $1.4 million, endowment gifts totaled more than
$3.1 million, and gifts dedicated to facilities for the capital campaign equaled $720,000.
Furthermore, the center’s 323 volunteer members contributed more than 9,700 hours in
operating and serving as an integral part of the Wharton Center.

For marketing and promotion, the Wharton Center has implemented a
comprehensive multimedia advertising program and direct-mail campaign which target
subscribers, group leaders, and single-ticket buyers. In addition, these advertising
mechanisms facilitate creative promotions with television and radio partners, increasing
the value and exposure of the center’s media buys while creating a dynamic outdoor
advertising.

The question raised is whether it is possible to identify important variables to
predict ticket purchases of performing arts consumers at the Wharton Center. If so, what
is the most effective way to deﬁne the performing arts market segments? How can these
segments be linked to behaviors of arts consumption? Are there any signiﬁcant

differences among segments in their behaviors and characteristics?

Problem Statement
The performing arts industry currently faces several challenges and issues.
Performing arts organizations Often confront signiﬁcant barriers in attracting, developing,
and maintaining consumers (Roan et al., 2006). Attendees are becoming more diverse,

and organizations are sometimes unprepared for the wide variety of art consumers.

Performing arts organizations often try to market themselves to everyone in the general
market, failing to identify proper target segments for their products. Consruners are more
spontaneous in choosing their entertainment options, and many are unsatisﬁed with pre-
selected performance packages offered by organizations. Young people in particular are
less likely to conunit to an entire performance series or to speciﬁc dates months in
advance. As a result, the role of the arts as cultural capital is in decline (DiMaggio &
Mukhtar, 2004).

Furthermore, performing arts organizations is faced the signiﬁcant decrease in
audiences among traditional presentation genres, such as classical music and ballet
(Kotler & Scheff, 1997). This may be caused by changing consumer tastes, massive
demographic shifts, and long-term neglect in arts education. The idea of attending live
performances requires specialized knowledge, which is a barrier to attend for many
people. One factor is that potential consumers spend the same amount of money on other
entertainment, such as sports, movies, and cuisine — all of which are often more appealing
than performing arts (Hume et al., 2006). Without understanding customers’ needs and
preferences, or how the broader market drives satisfaction, value, and consumption,
performing arts organizations will continue to have difﬁculty attracting audiences.

A critical factor in sustaining performing arts organizations is the ability to
effectively meet revenue-generating goals. However, most organizations have
experienced revenue shortfalls and ﬁnancial deﬁcits, along with declining audiences
(Bernstein, 2007). According to the US Census Bureau (2008), the estimated year-to-
year percent change in revenue for performing arts companies decreased 1.3% from

2005-2006. US. economy is currently in decline, and this economic crash has directly

affected people’s leisure activities. Due to the current economic climate, participation in
the performing arts and their spending on entertainment have been signiﬁcantly turned
down. In this unstable and unpredictable environment, it is difﬁcult to estimate
attendance and ticket sales for performing arts organizations; attracting audiences has
become even a more complex problems.

In addition, increased costs have driven up high prices and risk (Roan et al., 2006).
A number of economic factors such as transportation costs, operational and labor costs,
and artist and equipment rental fees continue to rise. This has resulted in signiﬁcantly
higher costs for performing arts organizations, generating a vicious circle of ﬁnancial
difﬁculty. To avoid revenue shortfalls, organizations have raised ticket prices and
increased efforts to garner sponsorships and to raise donations. However, higher ticket
prices have made performing arts signiﬁcantly less accessible to certain socioeconomic
groups or rendered the arts unaffordable for some long-time audience members. Unless
the performing arts organizations have a less expensive and more convenient form of
entertainment to targeted segments, it will be difﬁcult to lead to attendance growth.

Another challenge faced by performing arts organizations is marketing and
promotion (Conway & Whitelock, 2007). Many performing arts organizations do not
fully understand their consumers and marketing-oriented activities. They continue to use
only traditional media such as television, radio, and newspapers for mass consumers.
However, most arts consumers are becoming less interested in general information and
not much attracted to only traditional media outlets. In particular, radio and newspapers
have declined in their effectiveness as informational and advertising venues to reach

young people. Young consumers are more willing to be exposed to advertising through

10

 

 

 

 

richer and more diverse high technology, such as 3-D virtual experiences via the intemet
or electronic devices. The intemet is becoming ever more powerful and efﬁcient means
of communication with performing arts consumers. Unfortunately, a number of
performing arts organizations have not fully realized the scope of opportunities or
provided by the intemet in reaching, maintaining and growing their consumer base.

Moreover, ticket purchasing behavior in the performing arts has changed
signiﬁcantly. Subscribers have decreased and online purchases are becoming increasingly
popular. For example, subscriptions at 100 trendy theatres surveyed by the Theatre
Communication Group declined by 5%, and the average subscription renewal rate
dropped from 73% to 63% between 2001 and 2005, as well as the total number of
subscriber-occupied theatre seats decreased by 10% (Bumstein, 2007).

The intemet is becoming the primary tool used to search for information and
purchase tickets. However, performing arts organizations tend not to provide detailed
information regarding the new online purchase segments because their marketing efforts
have concentrated on programs, venue management, and attendance. In addition, some
small performing arts organizations lack marketing personnel, making it difﬁcult for
them to identify the target market through scientiﬁc methods (Knobbe, 2003).

To solve these problems in performing arts, the organizations need a priori
knowledge of consumers’ preference and needs. “Meeting audience needs, pricing
tactics, and product development” (Rentschler et al., 2002) are the principles of arts
marketing, in which consumer information and loyalty are essential. They also need to
recognize the importance of market information in order to design tailored programs for

target consumers. Understanding how the performing arts industry can incorporate the

11

”—— ,2...

 

changing needs and demands of this marketing environment is vital. Several studies have
been done regarding consumers of the performing arts (McCarthy et al., 2001; Nichols,
2003; and Borgonovi, 2004). More studies are needed, however, because many questions
remain unanswered. Future studies should clearly deﬁne their target audiences and strive
to make meaningful connections with these audiences, because not all arts consumers are
alike, and their differences can affect their behavior toward the arts (Nielsen et al., 1974).

In order to understand consumer needs and behaviors, performing arts

organizations need to either redeﬁne the market for their present products, or create a

 

new marketplace for existing needs (Scheff & Kotler, 1996). Strategic marketing in
performing arts organizations is focused on identifying target segments of homogenous
groups sharing similar needs, behaviors, and identifying speciﬁc characteristics (Clopton
et al., 2006). By tailoring market offerings to separate groups, or by focusing a few target
segments, the organization is better suited to meet customer needs and to leverage its
marketing efforts (Kotler, 1991). These marketing strategies will lead to: better
understanding of different consumer segment needs; more careful shaping and launching
of new services; pruning of weak services; more effective methods of service delivery;
more ﬂexible pricing; and higher levels of client satisfaction (Kotler, 1979). The
information can be used to identify market uniqueness, and used as a marketing tool for
improving the positioning and planning for target groups. However, researchers and
marketers often face the problem of segmenting the market appropriately and packaging

differentiated services for targeted segments.

12

 

 

 

Purpose of the Study

The purpose of this study is to examine the performing arts market by identifying
and proﬁling performing arts consumers. By categorizing these consumers into segments
using Chi-squared Automatic Interaction Detection (CHAID) analysis (Kass, 1980), this
study explores how the interactions between consumers’ characteristics and art-related
experiences are associated with 1) number Of tickets purchased; 2) donation activity; and
3) out-of-state or out-of-country performing arts attendance.

The ﬁrst segmentation distinguishes heav y and light consumer groups of the
performing arts and identiﬁes the interaction among predictors of each heavy consumer
group. Later, two different sets Of market segments of heavy consumers are identiﬁed
based On donation activity and attendance at live performances in a different state or
country. Further analyses are conducted to proﬁle each segment and identify signiﬁcant
differences among segments in their characteristics and behaviors toward performing arts
and pleasure trips, including: 1) type of performing arts attendance; 2) spending on
tickets to the performing arts; 3) performing arts information sources; 4) web-based
purchases of art- and culture-related products; 5) the importance of cultural attractions
and activities in selecting pleasure trip destinations; and 6) characteristics of cultural
pleasure trips.

The CHAID analysis is a criterion-based decision tree model used to explore the
relationships between independent and dependent variables that can optimize the
performance of market segmentation (Magidson, 1994; Croes, 2008; Hsu & Kang, 2007;
Van Middelkoop et al., 2003; and Chen, 2003a). Each performing arts consumer is

assigned to one of several homogenous groups based on the signiﬁcant socioeconomic

l3

and household characteristics and arts-related experiences. This study will be useful for
the Wharton Center (and speciﬁc/performing arts organizations in general) to help them
identify target segments and assign selected marketing strategies. In addition, performing
arts organizations can use this information to promote their programs and services, as
well as develop strategic advertising.

This study is comprised of ﬁve chapters: following this introduction chapter,
literature reviews in chapter 2 discuss studies of performing arts consumers, theory of
market segmentation, and CHAID approach in market segmentation studies. Method in
chapter 3 follows with a discussion of data source, measurement and data analyses used
in this study. Chapter 4 presents the results of the analyses, including three CHAID
segmentation results of performing arts consumers, proﬁle of derived segments, and
comparison of derived segments in their characteristics and behaviors. Chapter 5
discusses the key ﬁndings, implications, limitations and future studies, and concludes the
study.

Study Objectives and Hypotheses

This study has the ﬁve research objectives:

1. To identify predictors which distinguish and portray heavy consumers of the
performing arts.

2. To segment heavy consumers of performing arts based on the propensity to donate to
arts, heritage or cultural organizations.

3. To investigate whether propensity to donate sub-segments differ in terms of:
types of performing arts attendance, spending on ticket purchases to the performing

arts; performing arts information sources; web-based purchases of art- and culture-

14

related products; the importance of cultural attractions and activities in selecting
pleasure trip destinations; and characteristics of cultural pleasure trips.

4. To segment heavy consumers of performing arts based on the inclination to travel
out of state or/and out of country to attend live performances.

5. To examine whether inclination to travel out of state or/and out of country to attend
live performances sub-segments differ in terms of: types of performing arts
attendance, spending on ticket purchases to the performing arts; performing arts
information sources; web-based purchases of art- and culture-related products; the
importance of cultural attractions and activities in selecting pleasure trip destinations;
and characteristics of cultural pleasure trips.

The followings are three research hypotheses related to the study objectives:

1) m: Persons who attend live performances (e.g., plays, folk dances, musicals and rock
concerts) can be classiﬁed into substantial], measurablez, accessible3 and
actionable4 market segments, based on the number of tickets purchased to the
performing arts.

2) _H_2: Heavy consumers of the performing arts can be further divided into substantial,
measurable, accessible and actionable market segments, based on the propensity
to donate to arts, heritage or cultural organizations.

3) H_3; Heavy consumers of performing arts can be further divided into substantial,
measurable, accessible and actionable market segments, based on the inclination

to travel out of state or/and out of country to attend live performances.

 

' The degree to which the segments are large or proﬁtable enough;

2 The degree to which the size and purchasing power of the segments can be measured;

3 The degree to which the segments can be reached and served;

4 The degree to which effective programs can be designed for attracting and serving the segments (Kotler &
Amstrong, 1987).

15

Framework of the Study

The study framework is divided into three parts. First, performing arts consumers
are classiﬁed using the CHAID method. Number of tickets purchased to live
performances is divided into two categories and utilized as the dependent variable.
Performing arts consumers’ characteristics including socioeconomics, household
information, and arts-related experiences are used to categorize consumers into mutually
exclusive homogenous subgroups. Next, to identify effective target segments, heavy
consumers are selected, and two distinct market segmentations are conducted through the
CHAID-tree diagrams. Donation activity, out-of-state or/and out-of-country travels to
attend live performances are used as the dependent variables for the distinct CHAID
segmentations. Their socioeconomics, household information and arts-related
experiences are employed as independent variables for both analyses. Figure 1 shows the

framework of this study.

16

 

 

 

PERFORMING ARTS CONSUMERS

 

 

 

 

 

 

 

 

 

 

 

 

 

1
NUMBER OF TICKETS PURCHASED

(Heavy & light consumers)
SOCIOECONOMICS HOUSEHOLD ART-RELATED EXPERIENCES

INFORMATION
- Gender - Attendance at live performances
- Age - Number of persons in in a different state or country
- Education household - Arts education during childhood
- Employment - Marital status - Recent arts education
- Annual income - Children’s status - Purchase ofart pI‘OdUCtS

- Membership status

 

 

 

 

 

- Volunteer status
- Donation activity

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

HEAVY CONSUMERS
l
l I
DONATION ATTENDANCE OF LIVE
PERFORMANCES IN A
ACTIVITY DIFFERENT STATE OR
COUNTRY
SOCIOECONOMICS HOUSEHOLD ART-RELATED EXPERIENCES
INFORMATION
- Gender - Attendance at live performances
- Age - Number of in a different state or country
- Education persons in - Arts education during childhood
- Employment household - Recent arts education
- Annual income - Marital status - Purchase of art products
- Children’s - Membership status
status - Volunteer status
- Donation activity

 

 

 

 

 

 

17

Figure 1. Framework of the Study

 

CHAPTER 2

LITERATURE REVIEW

Overview

Knowledge of performing arts consumers is a basis for segmenting the arts market
in order to develop effective marketing strategies. This chapter looks at previous research
that has been conducted on performing arts consumers and discusses strategies and
applications of market segmentation, such as Chi-squared Automatic Interaction
Detention (CHAID). The chapter begins with an overview of previous studies related to
performing arts consumers. Next, the deﬁnition of market segmentation, segmentation
variables, and various approaches to market segmentation are discussed, followed by a
review of empirical studies. Lastly, an overview of the CHAID approach to segmentation
modeling is given, and the applications of CHAID in market segmentation studies are

discussed.

Performing Arts Consumer Studies

Beginning with Baumol and Bowen’s pioneering 1966 study of the composition
of performing arts consumers, a number of studies have been conducted on the subject.
Most empirical studies offer descriptions of participating behaviors in the arts and
provide detailed explanations of socio-economic characteristics (Borgonovi, 2004). The
2002 Survey of Public Participation in the Arts (SPPA) notes that arts participants who
read literature, listened to classical or jazz radio, or attended a performing arts event
tended to be active people, and were more likely than non-attendees to engage in other

leisure activities (Nichols, 2005).

18

A lot of research demonstrates that socioeconomic status, household composition,
education, and previous arts experiences are key factors affecting behavior of performing
arts consumer. Market research has indicated that the “typical” performing art consumers
are mature, female and well-educated, with high income, and white-collar occupations
(Kastenholz et al., 2005; Colbert, 2003; McCarthy et al., 2001; Notzke, 1999; and
Bergonzi & Smith, 1996; Silberberg, 1995). People’s attendance at arts performances is
related to their childhood experiences with the arts. Moreover, people with occupations
related to cultural industries and education tend to be higher consumers at performing arts
(Borgonovi, 2004; Richards, 1996; and Urry, 1990).

Gainer (1993) ﬁnds that women attend performing arts events more often than do
men because girls tend to be more exposed to arts education and participate in the arts
more often. McArthur and Balasubramanian (2003) suggest that the gendered nature of
the performing arts in New Zealand is the result of gender expectations, childhood
experiences of performing arts and parental attitudes. Their study showed that the
predominance of girls’ participation in performing arts education at school was
reinforced by the parental belief that arts are feminine, causing boys not to be encouraged
to participate. Not only do art education and memorable childhood artistic experience
inﬂuence adults’ taste in the performing arts, they also play a much larger role than the
cost of tickets in the decision of whether to attend a performance (Borgonovi, 2004).
Mitchell (1984) examines the relationship between the amount of childhood experience
with the arts and the frequency of arts attendance as an adult. This connection to
childhood experience provides an explanation for the gendered nature of attendees at

performing arts.

19

Some researchers have developed participation models to describe arts consumers.
Kotler and Scheff (1997) identify a number of factors inﬂuencing a consumer’s
likelihood of participation in the arts, such as macro-environmental trends, cultural
factors and social factors. Andreasen and Belk (1980) investigate predictors of future
attendance at performing arts events. The authors suggested that lifestyle, attitude and
developmental experiences are variables that are conceptually useful for understanding
consumer behavior toward the performing arts and are more empirically predictive than
socioeconomic variables. McCarthy and Jinnett (2001) provide the RAND model of
audience development between likely and unlikely participants, differentiating among
several factors, such as sociodemographic, sociocultural and personality factors, past
experience, and personal beliefs and perceptions of social norms regarding participation
in the arts. Wiggins (2004) indicates the problems in the RAND model and re-
conceptualizes the model by drawing on the Motivation/Ability/Opportunity model,
which offers a different combination of motivation, ability and opportunity to participate,
therefore explaining a different segment of the art consumers.

Nielsen et a1. (1974) segment the performing arts audiences in Champaign and
Urbana, Illinois and discussed the importance of such segmentation analysis. The study
found that the core audience was primarily composed of students, professionals, teachers,
people with relatively high family incomes, Jews, and agnostics. The core audience
preferred modern music and dance events to classical music and dance or theatre events
in their leisure time. The authors conclude that performing arts events had different, small,
and relatively narrow audiences who preferred speciﬁc types of performing arts. In

addition, a relatively narrow and small core audience existed which was overrepresented;

20

however, a relatively broad and large potential audience also existed and was
underrepresented at most types of performing arts events.

Bernstein (2007) proposes that arts consumers can be segmented into four very
different groups according to their age and life-cycle: mature adults, baby boomers and
thirty- and forty-years old, who often have children at home, and the younger set, the
members of Generation X and Generation Y. He concluded that different life cycles and
levels of exposure to the arts affect people’s propensity to attend arts events and
signiﬁcantly inﬂuence the effectiveness of marketing messages and media. Brown (2007)
suggests a segmentation model for performing arts ticket buyers and donors to aid in
future marketing and fundraising efforts. Not only past experiences in the arts but also
values, beliefs, aspirations and motivations were helpful predictors of ticket purchasing
and donation behavior.

On the other hand, Caldwell and Woodside (2003) clarify the possibly complex
relationships involved in mediating the inﬂuence of cultural capital on performing arts
patronage. Demographic factors such as age, gender, self-concept, personality, stage in
family life cycle, pre-adult exposure, place of residence, job and vocational focus, degree
of social network, responsibility at work, self-identiﬁed epiphanies and consumption
motivations provide useful information for explaining the effect of cultural capital on
performing arts patronage.

Turrini (2006) employs Markov’s theory to describe audience attendance at live
performances at an Italian theatre. The study divided attendees into casual attendees,
frequent attendees, and aﬁcionados and predicted the composition of the arts audience

who moved through the different stages of appreciation of the arts. The results implied

21

that the Markov model was a useful tool: 1) for describing the process of audience
habituation to the arts; 2) for supporting the development of organizational arts policies;
and 3) for evaluating pre and post and implementation of an institution’s strategies.

Other studies have been conducted to investigate the habits of performing arts
consumers and their motivations for attendance (Bouder-Pailler, 1999; Cuadrado &
Molla, 2000). Highly-involved arts consumers fall into two categories of involvement.
The ﬁrst category describes consumers who bring an enduring level of personal relevance
to the performing arts. These patrons are committed and attached to the performing arts
(Jain & Srinivasan, 1990; Broderick & Mueller, 1999). The second category describes
consumers with a high level Of hedonistic involvement. These patrons are emotionally
attached and demonstrate a strong need for emotional attainment (Laurent & Kapferer,
1985; Broderick & Mueller, 1999).

Hume (2008) investigates consumers’ perception of performing arts and their
intention to repurchase. The study proposes a conceptual model that shows service
quality and emotion inﬂuence value, which is then mediated through satisfaction to the
intention to repurchase. Speciﬁcally, the perceived quality of core and peripheral services
were the main drivers of the repurchase of performing arts events, but goal-directed
emotion was a weak driver of repurchase intention. Entertainment experience comes from
the interaction with the main product, but also from the performance venue, which also
inﬂuences the purchase choice of performing arts among people with similar social
standing (Mencarelli & Pulh, 2006; Ouellet et al., 2008). d’Astous et a1. (2004)
investigates sales promotion in the performing arts and resulting consumer perception.

The study found that consumer reactions to sales promotions in performing arts are

22

affected by the type of promotion, the type of performance, the attractiveness of the
performance, and the combination of the promotion and the performance, all of which
have interactive effects on consumer reactions. Moreover, the pattern of effects is
different depending on the type of consumer response considered.

Performing arts patronage can take several forms, including: volunteer work by
employees of the sponsoring company, either at events or within cultural organizations;
donations of equipment or technology; in-house exhibitions; and performance tickets
offered to clients (Martorella, 1999). These donations and sponsorships account for
approximately 40% of arts organizations’ budgets in the US. (Rich, 2001). Arts and
cultural volunteers are generally characterized as being older (with a median age of 51),
better educated, and more giving of their time than any other type of volunteer (Nichols,
2006c). National Endowment for the Arts research in 2005 pointed out that more than 1.6
million Americans have volunteered with arts or cultural organizations with artistic
activities such as music, performance or other forms of artistic engagement, and selling
items to raise money or other forms of fundraising was conducted as a secondary part of
their service.

Colbert et a1. (2005) investigate consumer evaluations of sponsorships based on
the type of cultural product being sponsored. Four products are classiﬁed by combining
the degree of artistic complexity with the cultural context in which the event takes place:
performing/popular arts; performing/high arts; heritage/popular arts; and heritage/high
arts. The results show that consumers rate philanthropic and performing arts sponsorship

more highly than commercial and heritage arts sponsorship. These forms of patronage are

23

related, with performing arts organizations relying heavily on social networking. An
overview of important consumer research in the performing arts is presented in Table 2.
Within performing arts literature, various aspects of consumer behaviors have
been studied, including predictors of attendance, participation models, segmentation,
perception and repurchase. However, performing arts organizations still ﬁnd it difﬁcult to
establish effective marketing strategies. The organizations need more efﬁcient
information about their consumers so that they can achieve better market deﬁnition.
Segmentation strategy can be the most effective approach to deﬁne the current

performing arts market.

24

Table 2. Overview of Research on Performing Arts Consumer Research

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Author(s) Year Subject
Developing a conceptual model for repurchase
Hume 2008 intention in the performing arts: the roles of
emotion, core service and service delivery
Hume & Mort 2008 Satisfaction in performing arts: the role of value
Reevaluating segmentation practices and public
Huntington 2007 policy in classical performing arts marketing: A
macro approach
Brown 2007 A segmentatron model for performrng arts ticket
buyers
Bernstein 2007 Arts marketing insights
Turrini 2 006 Measurrng audience addrctron to the arts: the case of
an Italian Theatre
McCarthy 2 006 Burldrng and sustaining trust In networks: Lessons
from the performrng arts
Hm... Essie:2%:strategists;52°32
Liesch & Winzar . '
servrce management
Mont ome & Take me out to the opera: are sports and arts
. g ry 2006 complements? Evidence from the performing arts
Robinson . .
research coalrtron data
. Consumer spending on performing arts: outlays ﬂat
NIChOIS 2006a for 2005; non-spectator categories show growth
Colbert, d’Astous 2005 Consumer perception of private versus public
& Parrnentier sponsorship of the arts
. Strategic positioning of the Venice Biennial:
Bagdadlr & . . .
. . 2005 analyzrng the market for periodic contemporary art
Arrrgom . . .
exhrbrtrons
Motivation, ability and opportunity to participate: a
Wiggins 2004 reconceptualization of the RAND Model of audience
development
d’Astous, Legoux 2004 Consumer perceptions of promotional offers in the

& Colbert

performing arts: an experimental approach

 

25

 

Table 2 (cont’d).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Author(s) Year Subject
DiMaggio & 2004 Arts participation as cultural capital in the United
Mukhtar States, 1982-2002: Signs of decline?
Borgonovi 2004 Performing arts attendance: an economic approach
McArthur & 2003 A theoretical explanation for the gendered nature of
Balasubramanian performing arts in New Zealand
Caldwell & The role of cultural capital in performing arts
. 2003
Woodsrde patronage
. A measure of culture: cultural experiences and
Pmk & Matthews 2003 cultural spending in New Zealand
Rentschler, . . . . .
Ra dboume, Carr & 2002 Relationship marketing, audrenceoretentron and
Rickard performrng arts organization vrabrlrty
McCarthy & Jinnett 2001 ﬁngew framework for burldrng participation 1n the
Kotler & Scheff 1997 Standing room only
Colbert 1997 Changes In marketing envrronment and therr Impact
on cultural polrcy
Gainer 1993 The importance of gender to arts marketing
Mi tchell 1984 The professronal performrng arts: attendance
patterns, preferences and motrves
Andreasen & Belk 1980 Predictors of attendance at the performing arts
Nielsen, McQueen . .
& Nielsen 1974 Performing arts audience segments
Baumol & Bowen 1966 Performing arts - the eConomic dilemma

 

26

 

Market Segmentation
Deﬁnition of Market Segmentation

Market segmentation is a methodological process originated by Smith (1956) that
has been one of the most powerful and fashionable concepts in marketing. Market
segmentation separates a market into mutually exclusive and exhaustive subgroups of
people (Reid & Bojanic, 2005; Venugopal & Baets, 1994) who differ with regard to some
criterion in order to achieve competitive advantage and superior ﬁnancial performance
(Hunt, 2002). Market segmentation is used to develop a better understanding of
consumers’ distinct characteristics and to improve marketing strategies, since this can be
used as a strategic marketing tool for deﬁning markets and allocating resources. Thus,
members of each group respond in similar ways to given sets of marketing efforts (Kotler
and Armstrong, 2006). Segmentation can achieve great efﬁciency in the supply of
products or services to meet identiﬁed demand and increase cost effectiveness in the
marketing process (Park & Yoon, 2009). Therefore, market segmentation is an effective
and valuable instrument that divides a total market or market sector for the purpose of
planning appropriate marketing strategies.

Market segmentation is an important strategic concept contributed by the
marketing discipline to business ﬁrms and other types of organizations; segmentation’s
importance is highlighted in terms of strategic marketing decisions (Bowen, 1998; Myers,
1996). Moreover, market segmentation is a powerful tool used by successful consumer
product companies that can build and maintain proﬁtable relationships with their target
segments. Market segmentation is one of the cornerstones of marketing management

approach (Guiltinan & Paul, 1991). Best (2000) indicates that market segmentation is

27

based on multiple market-based strategies, resulting in greater marketing efﬁciency. The
effectiveness of market segmentation strategies can be evaluated with four criteria:
whether the segments are (1) large enough for attention (substantiability); (2) able to be
measured (measurability); (3) accessible through a variety of marketing communication
vehicles (accessibility); and (4) sharing certain characteristics and responding similarly to
the marketing mix (actionability) (Bojanic, 2007 and Kotler, 1972).
Segmentation Variables

According to Mazanec (1984), segmentation occurs in two steps: (1) determining
the most important variables that separate groups of consumers from each other; and (2)
choosing consumer segments to target with marketing. Various segmentation techniques
and methods have traditionally used a number of variables for market segmentation
strategy, including: (1) demographic variables such as gender, age, family size, education
and race; (2) socioeconomic variables like occupation and income; (3) motivations; (4)
geographic variables such as region and degree of urbanization; (5) speciﬁc beneﬁts
sought by consumers; (6) behavioral patterns such as on occasions and characteristics of
purchase behaviors; (7) psychological characteristics including attitudes, opinions, and
lifestyles; and (8) involvement proﬁle (Park et al., 2002). Of these primary bases for
segmentation, demographic and socioeconomic variables are commonly used and popular
bases for segmenting consumer groups (Lu, 2005; Palakurthi & Parks, 2000; Kotler &
Armstrong, 1987). However, there is no simple way to determine which of the methods
provides the best approach (Levy & Weitz, 1992).

Clopton et a1. (2006) adapt the 2 x 2 taxonomy of segmentation based on Wedel

and Kamakura (2000). The criteria for segment building are either observable or

28

unobservable variables. Other segmentation bases are independent of general- and
product-speciﬁc goods or services. These segmentation variables have been used as
important criteria depending on the purpose of the given studies. Table 3 shows the

taxonomy of segmentation bases.

Table 3. Taxonomy of Segmentation Bases

 

 

 

General Product-speciﬁc
Observable Cultural, geographic, User status, usage frequency,
demographic, and store loyalty and patronage, and
socioeconomic variables situational variables
Unobservable Psychographic, values, Psychographic, beneﬁt,
personality and lifestyle perception, attribute, preference
variables and intention variables

 

 

 

 

Source: Clopton at el. (2006) adapted from Wedel and Kamakura, 2000.

Segmentation Methods

Different methodological aspects of segmentation have been applied in” much
consumer behavior and marketing research. Among these aspects are factor-cluster
segmentation, cluster segmentation, criterion segmentation (Magidson, 1994), neural
network models (Mazanec, 1992), and a combination of neural network and criterion
segmentation (Yu & Wang, 2007). According to Wedel and Kamakura (2000), market
segmentation methods can be largely classiﬁed based on two criteria for the four
categories: a priori or post hoc, and descriptive or predictive statistical methods. The a
priori approach is used when the type and number of segments are determined in advance

by the researcher, while a post hoc approach is used when the type and number of

29

segments are determined based on the results of data analyses. Descriptive methods are
associated with a Single set of segmentation bases that has no distinction between
dependent and independent variables. Predictive methods are applied when one set
consists of dependent variables to be explained or predicted by a set of independent
variables.

A priori descriptive segmentation methods, using cross-tabulation or log-linear
analysis to describe two or more separate strata in a population, decide the type and
number of segments before data collection. Post hoc descriptive segmentation generally
identiﬁes segments using clustering methods, forming homogeneous consumer groups
with regard to measured characteristics such as their values, activities, interests, and
opinions. A priori predictive segmentation approaches deﬁne a priori descriptive
segments based on one set of criteria, such as sociodemographics and psychographics,
and subsequently employ predictive models to describe the relation between segment
membership and a set of independent variables. Post hoc predictive segmentation
methods, mainly using CHAID, automatic interaction detection (AID), and clusterwise
regression, identify categorical segments based on the interaction of independent
predictors and a dependent variable. This method divides a sample into segments that
differ maximally according to a dependent variable, such as purchase behavior, based on
a set of independent variables, which are often socioeconomic and demographic

characteristics. Table 4 presents the classiﬁcation of methods used for segmentation.

30

Table 4. Classiﬁcation of Methods Used for Segmentation

 

 

A priori Post__hoc
Descriptive Contingency tables and Clustering methods:
Log-linear models Nonoverlapping, Overlapping,

Fuzzy techniques, ANN, and

mixture models

 

Predictive Cross-tabulation, AID, CHAID, CART,
Regression, Logit, and Clusterwise Regression, ANN,
Discriminant analysis Mixture Model

 

 

 

 

Source: Wedel & Kamakura (2000). Market segmentation: conceptual and methodological foundations.

Empirical Studies in the Arts and Cultural Consumer Market

Several studies identify market segments related to the arts and the cultural
consumer market (Molera & Albaladejo, 2007; Clopton et al., 2006; Stepchenkova &
Morrison, 2006; Kastenholz et al., 2005; Prentice & Andersen, 2003; Gursoy and Chen,
2000). According to Nielsen et a1. (1974), arts audience segmentation analysis is a
systematic method of studying: (1) how art forms are related to consumers’
characteristics, needs, wants, beliefs, and preferences; and (2) how art forms ﬁt into the
arts, entertainment, and leisure markets in the process of satisfying what people need in
relation to what they want, and what good art is.

Clopton et al. (2006) segment four groups of cultural arts patrons along their
preferences and developed proﬁles of these segments. First, reluctant arts patrons express
low preferences for almost all art forms. Second, heritage arts patrons have high levels of
preference for ceramics, crafts, and pottery. Third, visual and performing arts patrons
prefer dance, drawing, ﬁlm, painting, photography, sculpture and theatre and are not

interested in heritage arts. Fourth, comprehensive arts patrons express a high level of

31

preference for all art forms. The study shows a signiﬁcant relationship between arts
preferences and personal characteristics, which explains why patrons of different arts
segments exhibit different behaviors with regard to attendance, spending, venue
preferences and information sources.

Kastenholz et a1. (2005) identify two segments of cultural tourists based on
education level, income, and age. The results revealed signiﬁcant differences in terms of
other socio-demographic variables and travel behavior. One segment, representing
relatively Older, wealthier and more educated respondents, tended to be mostly foreign
tourists, traveling with their partner or family, and revealed higher expenditure levels.
The other group was relatively younger, poorer and less educated and included many
students traveling either alone or with friends.

Prentice & Andersen (2003) identify three different consumption styles of
participants at the Edinburgh festival: consuming an historic city, consuming Scottish
performing arts in a Scottish context, and consuming international performing arts. The
authors found seven distinct segments based on tourism styles: serious consumers of
international culture; British drama-going socializers; Scots performing arts attendees;
Scottish experience tourists; gallery-goers; incidental festival-goers; and accidental
festival-goers. The festival attracted arts segments that focused largely on contemporary
performing arts, of either an international or regional production. The authors concluded
that the Edinburgh Festival needs to continuously offer a product mix for market-based
cultural products.

Robbins & Robbins (1981) segment museum-goers into three distinct target

categorizations describing their propensities for attending museum exhibits. Present

32

attendance as a usage pattern was used as the segmentation descriptor and to categorize
respondents into low, moderate, and high frequencies of museum attendance. One-way
analysis of variance and stepwise discriminant were used to differentiate the
characteristics of the three groups of museum attendees. The results indicated that all
three groups of attendees differed signiﬁcantly from each other in their mean responses to
six present attendance variables, one future attendance variable, and two demographic
variables. The identiﬁcation of highly frequent attendees pinpointed the primary target
market for museums, and moderate attendees were a secondary target market that could
be developed as a new-user of existing products and services of the museum by
differentiating and extending the product life cycle. The museum should require greater
promotional effort and use different strategies to attract low attendees.

Overall, market segmentation studies have been conducted in a wide-range of
disciplines, including the arts and the cultural market where marketers seek to penetrate
and build effective marketing strategies for their target markets. Much research has
evidently showed that various segmentation variables and segmentation approaches are
conducted according to their market characters. In this study, the CHAID approach, used
in the criterion-based segmentation, is used to effectively segment the performing arts

market.

33

Chi-squared Automatic Interaction Detection (CHAID) Approach

Chi-squared Automatic Interaction Detection (CHAID) analysis, an offshoot of
the Automatic Interaction Detection (AID) method, was ﬁrst proposed by Kass (1980)
and further utilized by Magidson (1994). The CHAID, using of decision tree algorithms,
is an exploratory method for segmenting a population into two or more exclusive and
exhaustive subgroups by maximizing the signiﬁcance of the chi-square, based on
categories of the best predictor of the dependent variable (Kass, 1980).

Segments obtained from CHAID analysis are different from cluster type models
because the CHAID method, which is derived to be predictive of a criterion variable, is
deﬁned by combinations of predictor variables (Magidson, 1994). CHAID technique
depends on interactions among the independent variables, ﬁnding those that explain the
greatest differences within the dependent variable. Thus, a CHAID decision tree
demonstrates how the predictors are differently formed and predicts a dependent variable
that shows nominal scaling (continuous variable may be used as well but the ﬁrst step of
the analysis is to categorize them).

CHAID analysis is also useﬁrl for data mining, which is deﬁned as the extraction
of hidden predictive information from a large dataset. Moreover, CHAID analysis is able
to formulate predictive segments with regard to a meaningful criterion. Researchers
choose the criterion, which is explained by a combination of independent variables such
as demographic, behavioral, and attitudinal depending on the purpose of the research
(Chen, 2003a). Marketers can identify the key inﬂuencers or signiﬁcant drivers in certain
markets using CHAID analysis, which results in a treelike diagram commonly called a

decision tree.

34

Decision trees are a sequential approach used to generate rules for the
classiﬁcation of the dataset (Biggs et al., 1991). Decision trees have several advantages in
a data mining environment (Bakken, 2005). The type of representation makes the
resulting classiﬁcation model easy to use. Moreover, decision trees are suited for
exploratory knowledge discovery because they are non-parametric and make no
assumptions about the underlying probability distribution. Decision trees are also
efﬁcient and superior to non-linear covariate effects and higher-order interactions. They
are relatively quickly 0 onstructed for large datasets compared to other classiﬁcation
models (Magidson & Vermunt, 2005). The classiﬁcation trees approach has been used in
a variety of disciplines such as marketing, medicine, computer science, botany and
psychology. Usefulness and effectiveness of the CHAID-tree method in marketing has
been shown in the variety of market segmentation studies (Gan et al., 2008; Gomez &
Benito, 2008; Ruﬁn, 2007; and Chen, 2003b). The next section reviews the application of
CHAID chieﬂy focused on market segmentation studies.

The Application of CHAID in Market Segmentation Studies

Croes (2008) uses CHAID to identify the characteristics of visitors that
contributed to their choice of a type of accommodation and thus to market differentiation
in Aruba. Among three independent variables (length of stay, country of origin, purpose
of visit), the length of stay was found to be the strongest predictor of accommodation
choice. The procedure was repeated for each of the groups and the result was a tree-like
partitioning into ﬁve mutually exclusive, exhaustive segments that strongly described the
choice of a type of accommodation. The CHAID tree provided a more detailed analysis

that also uncovered the niche markets for the various types of accommodations. The

35

author concluded that the various types of accommodations are associated with different
tourist markets. Thus, diversifying the portfolio can lead to a differentiated market mix.

Gan et a1. (2008) analyzes Singapore’s credit cardholders in search of variations
among demographics and card usage patterns as well as the cardholders’ perceptions
regarding credit card ownership and use. This study offered a more in-depth study
through a decision tree analysis using the CHAID algorithm and Answer Tree to examine
the association between the number of credit cards held by individuals and the
independent variables. Annual income and gender were more important positive drivers
of the number of credit card accounts and credit card usage. This research identiﬁed four
potential card users to target and provided useful marketing strategies for consumer
behaviors related to credit card issues.

Using CHAID and multiple regression analyses, Worsely & Lea (2008) examines
the relationships between consumers’ concerns regarding food and health issues and their
demographic characteristics and personal values. The research indicated that personal
values, gender, and age were strong predictors of consumers’ speciﬁc food concerns.

Hsu and Kang (2007) employ CHAID to identify market segments among
international travelers to Hong Kong, based on traveler characteristics. In their analysis,
travel purpose, age, repeat visit. status and income were respectively found to be the best
predictors of the likelihood of a revisit. Subsequently, a total of six segments were
formed according, to their index scores. The higher index scores indicated that
respondents in the segment were more likely to return to Hong Kong for pleasure travel.
In addition, Chi-square analyses and ANOVAS were conducted to proﬁle trip

characteristics among the six segments. This study demonstrated the successful

36

application of CHAID for deriving market segments and the usefulness of the CHAID
technique with monotonic variables, which explain the multidimensional nature of
various traveler behaviors.

Legohérel and Wong (2006) apply CHAID when exploring the pertinence of
segmentation based on Hong Kong visitors’ direct expenditures, using it to deﬁne
segment characteristics. A CHAID analysis was performed with independent variables,
which were related to the trip proﬁle, and demographic variables used in the
segmentation analysis of consumers’ total expenditures and consumers’ daily
expenditures during their stay in Hong Kong. The length of stay was the most signiﬁcant
predictor of both outcomes for describing the “big spenders” segment. The authors then
used multivariate regression analysis to explain consumers’ behavior regarding their level
of expenditures and to validate the use of the CHAID technique. This study demonstrated
that CHAID analysis can be used as a precursor to a more parametric technique.
Moreover, the segmentation tree identiﬁed the most relevant predictor variables for
explaining the changeability of the dependent variables.

Galguera et al. (2006) employ a two-step procedure to investigate whether age,
education and location of residence would inﬂuence loyalty card possession. First, a
logistic regression was conducted to identify key signiﬁcant predictors of the dependent
variable, and then CHAID was used to segment the loyalty card market in two different
countries. The CHAID algorithm, which found the best segmentation scheme based on
these variables, provides non-binary classiﬁcation trees where the resulting segments are

mutually exclusive, and it permits prediction of whether certain segments are more likely

37

to engage in the target behavior. This study showed that CHAID method is a reliable
segmentation procedure.

Van Middelkoop et al. (2003) use the Exhaustive CHAID algorithm to identify
heuristic principles for transportation mode choices. Given the conceptual representation
of the tourist decision-making process, four types of factors conditioning the choice of
travel mode were used, such as personal and household characteristics, characteristics of
the annual trip program, conditions reﬂecting previous decisions regarding the tourist trip
under consideration and tourist trips that are more important than the trip under
consideration. Whether the vacation under consideration was a domestic or foreign
destination was the most important conditioning factor for the choice of transport mode.
The study results indicated that CHAID methodology could be applied to better
understand tourist choice behavior.

Chen (2003a) provides a step-by-step review of the critical procedures pertaining
to the proposed CHAID model and includes a case study using the CHAID analysis to
segment college students’ spring-break trip decision-making. The study identiﬁed four
segments, two of which are regarded as actionable segments and the others as
nonactionable segments. Subsequently, Chi-square, logit analysis, and ANOVA were
used as major analytical tools to compare the characteristics of the two actionable
segments. The study concluded that CHAID would be a useﬁrl tool to advance the
segmentation methodology and detect valuable market segments effectively in the study
of hospitality and tourism.

Chen (2003b) employs a CHAID analysis to identify respondents in four mutually

exclusive segments. A segment with an index score of 100 or above was considered an

38

actionable market, and a segment with an index score below 100 was a non-actionable
market. Each segment was proﬁled by demographic and trip characteristics and two of
the segments were deﬁned as actionable segments, based on the likelihood that positive
recommendations would be made about a destination. Destination satisfaction, pricing
and prompt assistance were identiﬁed as the most important factors inﬂuencing
respondents’ recommendations of destination to others.

In summary, CHAID appears to be a more powerful and applicable statistical
technique to effectively segment a consumer market into homogeneous groups using
large data sets. Research indicates that demographic or socioeconomic characteristics are
signiﬁcant predictors alongside any kind of criterion for segmenting particular markets in
most research. Thus, these variables are a good starting point for understanding the multi-
dimensional nature of performing arts consumer behaviors. Table 5 presents an overview
of previous market segmentation studies using CHAID Approach. There is no study
using the CHAID method in performing arts consumer research. Therefore, this study
contributes effective market segmentation to understand performing arts consumers in
terms of identifying important predictors, targeting segments and identifying their

proﬁles and behaviors.

39

Table 5. Overview of Previous CHAID Approach in Market Segmentation Studies

 

 

 

 

 

 

Author(s) Dependent Variable Independent Variables
(Criterion) (Predictor)
Croes (2008) Choice of a type of Length of stay
accommodation in Aruba Purpose of visit
Country of origin
Gan, Maysami & Number of credit cards owned Demographic characteristics
Koh (2008) Usage patterns
Perception of credit card
ownership
Worsely (2008) Food and health concerns Personal Values
Demographic variables
Hsu & Kang Likelihood of return to Hong Country of residence
(2007) Kong Purpose of visit
Repeat visit status
Socioeconomic variables
Legohérel & Consumers’ direct total & Trip proﬁle
Wong daily expenditures during their Demographic variables
(2006) visiting to Hong Kong

 

Galguera, Luna &

Loyalty card possession

Client characteristics (sex, age,

 

 

 

 

Méndez (2006) urban versus suburban
dwelling, education etc.)

Van Middelkoop, Dutch tourist decisions of the Characteristics of the annual

Borgers & transport mode trip program

Timmerrnans Personal and household

(2003) characteristics
Conditions reﬂecting previous
decisions regarding the tourist
trip under consideration
Conditions reﬂecting previous
decisions regarding tourist
trips

Chen Destination loyalty 17 perceptions of travel

(2003a) destinations

Chen Likelihood of make positive 20 attributes of the tourists’

(2003b) recommendations about the sentiment toward marketing

 

destination they previously
visited

 

 

40

 

CHAPTER 3

METHOD

Overview

The purposes of this study are to explore performing arts consumers with a
predictive segmentation method and to identify signiﬁcant differences among market
segments in their characteristics and behaviors. Chi-Squared Automatic Interaction
Detection (CHAID) is used to identify performing arts market segments based on
‘number of tickets purchased,’ ‘donation activity’ and ‘out-of-state or out-of-country
travels to attend live performances’ explained by combinations of predictors including
socioeconomics and household information and arts-related experiences.

This chapter provides an overview of the data source, measurements and analyses.
First, the study population, data collection procedures and instrument are explained.
Second, the data selection and the procedures of the CHAID method (including
modiﬁcation of the variables) are discussed. Finally, the statistical data analyses used in

the study are discussed.

Data Source
To achieve the purpose of this study, data developed and collected through an
online survey in partnership with the Wharton Center for Performing Arts (Wharton
Center) and the Recreational Industries Research Center (RIRC) at Michigan State
University in 2007 were used for the study. The main purpose of this online survey to
understand patrons’ proﬁles and identify the basis for market segments including their

lifestyle, consumption and participation in the arts, culture and tourism.

41

Study Population

The study population consisted of performing arts patrons of the Wharton Center
at Michigan State University. The Wharton Center had a total of 39,684 E-club members
on its email list in January 2007. To qualify, performing arts patrons had to have
purchased a ticket at least once from the Wharton Center and who had provided their
email addresses before January 2007 were included. Table 6 indicates sample size and

response rate.

Table 6. Sample Size and Response Rate

 

 

 

 

 

 

Sample size n
Total number of emails received 39,684
Wrong/bounced emails 673
Valid emails 39,011
Opened emails 13,503

Number of respondents n
After invitation email 4,405
Alter reminder 339
Total 4,744

Response rate %
After invitation email 32.6
After reminder 2.5
Total 35.1

 

 

 

The respondents were predominantly female (73%); and the majority ranged in
age from 35 to 64 (72%); More than 90% of the respondents were straight/heterosexual
(91%) and Caucasian or white (95%); In addition, a majority had a 4-year college degree

or above (66%); full time employment (64%); and were married (72%). Annual income

42

was reported as less than $50,000 (22%), $50,000-$99,999 (43%), and $100,000 or more
(35%).
Data Collection Procedures

The web-based survey was created and conducted in partnership between the
Wharton Center and the RIRC at Michigan State University. The recruitment of
participants was executed via Wharton Center E-club newsletter that was emailed to all
E-club members on January 22, 2007. Speciﬁcally, the newsletter contained the
announcement of the survey and the link to the online questionnaire. Once participants
clicked on the survey link, they accessed the main page of the survey, in which the
purpose of the study, the incentive, and the conﬁdentiality procedures were detailed.

A total of 39,011 valid invitations were delivered to valid email addresses,
approximately 13,503 of the E-club members opened the electronic newsletter, and 4,405
members responded to the survey in the ﬁrst week. After a week, a reminder with the link
to the survey was placed in another E-club news letter and emailed to non-respondents on
January 29, 2007. Another 339 members completed the survey, which was closed on
February 2, 2007.

Overall, the survey lasted for 12 days, and a total of 4,744 E-club members
responded to the survey which amounts to a response rate of 35.1%. As an incentive,
upon completion of the survey, the respondents had a chance to be entered in a drawing
for a $100 Wharton Center gift certiﬁcate and dinner for two at the Kellogg Center’s
State Room.

The online system automatically recorded answers and made it easy to lead

respondents to questions applicable only to them based on the responses they provided

43

earlier in the survey. The coded data were automatically sent to an Access database,
eliminating the potential errors associated with manual data entry and coding. The data
were then converted into an appropriate format and analyzed using the Statistical Package
of Social Sciences (SPSS).
Study Instrument

The online questions were developed from a review of previous research,
industrial reports and other literatures pertaining to consumers of performing and cultural
arts, as well as from discussions with the Wharton Center marketing staff. The number of
questions that participants answered depended on their responses to various sub-questions
throughout the survey. The entire online questionnaire was comprised of 84 questions in
ten different categories, including (1) attendance at live performances and ticket purchase
behavior; (2) participation in arts and culture-related activities or disciplines; (3) purchase
of arts and cultural products and employment of artists, performers and culture-related
professionals; (4) importance of selecting cultural destinations during pleasure trips; (5)
frequency and nature of activities participated in during recent pleasure trips; (6)
participation in cultural membership, donation and volunteer activities; (7) sources of
information and intemet use; (8) artistic and cultural experiences of the respondent during
childhood; (9) artistic and cultural experiences of children living at home; and (10)
socioeconomic and household characteristics. Appendix A presents the questionnaire of
the online survey used for this study. The speciﬁc questions of each category are
summarized.

First, respondents were asked for information about the live performances they

attended such as frequency, type, place and whether they paid and for information about

44

purchased tickets including frequency, average price, highest price, and total price during
the last 12 months.

Second, respondents were asked to indicate the nature of their participation in
performing arts, visual arts, and art- and cultural-related disciplines in each type and, for
each type, they indicated whether they were professionally engaged during the last 3
years. Furthermore, if they had checked that they were professionally engaged,
respondents were asked whether or not they considered themselves professional artists,
received formal education/training and were currently a member of an arts-related guild,
union or association/organization, as well as how much they had earned during the last 12
months.

Third, respondents were asked for the purchase information of visual arts and
cultural products related to purchasing decisions made during last 12 months. For visual
arts, what types and where to purchase, and how much in total and highest price to pay
were asked. Moreover, they were questioned whether they hired any performing artists,
and if so, what types and purpose, and how much they had spent during the last 12
months. They were also asked whether they hired architects or designers, and historic
preservation/restoration, archaeological or genealogical services.

Fourth, respondents were asked about the importance of selecting a set of cultural
destinations associated with their pleasure trips, including museums; cultural
fairs/festivals/events; performing arts; places to buy local arts/crafts; historical/heritage
attractions; sites and districts; gardens, zoos and aquariums; agricultural attractions and
events; purchasing products grown or processed locally; architecture and buildings;

customs and ways of living; and libraries, literary events and bookstores. In addition, they

45

were asked whether they had taken any pleasure trips to destinations more than 50 miles
away during the last 12 months.

Fifth, respondents were also asked about the frequency of the pleasure trips, and
the proportion of overnight stays, out-of-state trips, and overseas trips and the primary
purpose to participate in the arts and culture related activities.

Sixth, respondents were asked whether they have any cultural memberships,
donations and volunteering, if so, they were then asked the primary mission and purpose
for the memberships, the amount of the donations, and what the donations supported and
volunteering covered.

Seventh, respondents were asked how often they used various information
sources, including TV; magazines; newspapers; radio; websites specialized for online
ticketing; websites of performing art organizations or centers; acquaintances, friends,
family or relatives; art-related publications; and leaﬂets/brochures for live theater, dance
performances, and music concerns. Furthermore, they were asked whether they use
intemet to research or purchase the art and culture related products, and, if so, how often
to research or purchase through intemet.

Eighth, respondents were asked about their childhood experiences related to arts,
regarding frequency of attending arts or cultural performances and attendance and types
of arts classes. In addition, they were asked whether they had taken any arts classes
during the last 12 months and, if so, they were asked what types they attended and how
much they paid.

Ninth, respondents were asked about art and culture related experiences of their

children during the last 12 months. It the response was afﬁrmative, they were asked for

46

the number of performances attended, whether they joined performances or exhibits and
arts classes or lessons, and what types were attended and how much they paid.

Last, respondents were asked about socioeconomics including gender, age, sexual
orientation, education, employment status, race, annual income and permanent residence,
and household information, such as number in household, family status and number and

age of children.

Measurements

Data Selection and Procedures

To address the purpose of this study, a number of speciﬁc variables were selected
and used for the CHAID method and further analyses based on the proposed model. The
study population consisted of performing arts consumers who were E-club members of
the Wharton Center. With the CHAID procedure, number Of tickets purchased was used
as criterion of market segmentation in this study. Respondents who had purchased tickets
to live performances during the last 12 months were included; however, respondents who
had attended but free live performances were excluded. A sub-sample of 4,138
respondents who provided information on the number of purchased tickets to live
performances during the last 12 months was selected from the total of 4,744 respondents.
The distribution of the respondents in the sub-sample is provided in Table 7.

Using the CHAID method for two more decision trees, the heavy consumers were
selected and analyzed. Donation activiy and attendance of live performances in a

different state or country were used as criteria in two other analyses.

47

Table 7. Distribution of Respondents Who had Purchased Tickets for Themselves or for
Their Household to Attend Live Performances during the Last 12 Months

 

 

 

 

Number of tickets purchased n %
l 106 2.6
2 579 14.0
3 343 8.3
4-5 964 23.3
6-9 879 21.2
10 or more 1,267 30.6
Total 4,138 100.0

 

 

The CHAID is a technique that recursively partitions a population into separate
and distinct subgroups so that the variation of the criterion variable is minimized within
the segments, and maximized among the segments. While both the CHAID and cluster
techniques divide a population into subgroups, only CHAID makes use of an explicit
dependent variable criterion in forming the subgroups (Magidson, 1994). The criterion
variable can be nominal, ordinal, or continuous, and the predictor variables can be
categorical or continuous (Magidson & Vermunt, 2005). The nodes are deﬁned by
predictor variables that pass through an algorithm for partitioning. To perform a CHAID
analysis, a dependent variable and a set of independent variables were ﬁrst selected and
deﬁned. Figure 2 illustrates the overview of the analytic procedures of the CHAID

method used in the study.

48

 

 

Start

 

 

 

V

 

 

Compute criterion variable response

 

 

 

 

Merge levels of

on the criterion

_ predictors which have
similar response patterns

. . . a ----------------- a
drstrrbutron for “parent” group
I
For each predictor, determine how
A many splits are required to result in

 

subgroups that are homogeneous with
respect to the criterion distribution

 

 

 

 

 

 

 

 

f

 

 

 

’ Split merged groups 5
if necessary :
V

Test signiﬁcance of best predictor with

.- -------------------- bonferroni adjusted chi-squared values

 

 

 

 

: Signiﬁcant
v

 

Split “parent” group response on (new)
levels of the best predictor

 

 

 

6
Determine if additional parent groups
are available for splitting

 

Next parent

I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
b

 

 

 

U
I

: No qualiﬁed groups
1

 

Finish

 

 

 

 

Source: Perreault, W. D. (Nov. 1980). “A Model- free Approach for Analysis of Complex Contingency
Data rn Survey Research.” Journal of Marketing Research, 17 (4).

Figure 2. Overview of the Analytic Procedures of the CHAID Method in the Study

49

 

Dependent Variables

Number of tickets purchased The number of tickets purchased was re-
categorized into two groups by heavy and light consumers and used as dependent
variable in the ﬁrst CHAID analysis. The respondents who purchased between one and
nine tickets to live performances were considered to be light consumers, who recoded as
1. Respondents who purchased 10 or more tickets to live performances during the last 12
months were considered to be heavy consumers, who recoded as 2.

Donation activity Donation activity was re-categorized into two groups (donor
and art non-donor) and used as a dependent variable in the second CHAID analysis. The
respondents who donated money to arts organizations were considered to be donor and
recoded as l, and those who did not donated money to arts organizations during the last
12 months were considered to be non-donor and recoded as 2.

Attendance at live performances in a different state or countg Attendance at live
performances in a different state or country was re-categorized into two groups (cultural
tourist and non-cultural tourist) and used as a dependent variable in the third CHAID
analysis. The respondents who attended live performances in other states or countries
were considered to be cultural tourists, and recoded as 1 and those who did not attend live
performances other than where they resided were considered to be non-cultural tourists,
and recoded as 2.

Independent Variables

While a number of studies demonstrate that socioeconomic and household

characteristics and previous arts experiences and education are key factors affecting the

behavior of performing arts consumers, the independent variables of this study were

50

divided into three categories: socioeconomic characteristics, household information, and
arts-related experience variables.

Socioeconomic characteristics Socioeconomics included gender, age, education,
employment and annual household income. These are easy to measure and used as initial
predictors in most market segmentation analyses (Hsu & Kang, 2007; Legohérel &
Wong, 2006; Hudson, 2000; Muller & Cleaver, 2000). The age was re-categorized into
ﬁve groups (1 = under 35, 2 = 35-44, 3 = 45-54, 4 = 55-64, 5 = 65 or above). Education
was re-categorized into six groups (1 = 2-year college degree or less, 2 = completed 4-
year college degree, 3 = completed some graduate courses or master’s degree, and 4 =
doctoral degree). Employment was re-categorized into two groups by employed ﬁrll-time
= 1, others =2. Annual household income was also re-categorized into four groups (1 =
less than $50,000, 2 = $50,000-$99,000, 3 = $100,000 - $149,999, 4 = $150,000 or
more).

Household information Household variables included three predictors which were
the number in household, marital status, and children’s status for the CHAID analysis.
The number in household variable was divided into 5 categories (1 = live alone, 2 = 2
people, 3 = 3 people, 4 = 4 people or more). The marital status variable was re-
categorized into two groups (1 = single and 2 = married). Children’s status variable was
re-categorized into three groups (1 = no child, 2 = children living at home, and 3 =
children no longer living at home).

Arts-related experience variables Arts-related experience variables included
attendance to live performances in a different state or country in the last 12 months, arts

education during childhood, recent arts education, art product purchase experiences in the

51

last 12 months, membership status in arts organizations in the last 12 months, volunteer
status to arts organizations in the last 12 months, and donation activity to arts
organizations in the last 12 months. The category for the variables was dichotomous (yes
= 1 and no = 2).

The deﬁnition of the variables used in the CHAID Analysis is detailed in Table 8.
Other Variables

Further analyses were conducted to identify statistically signiﬁcant differences
among market segments in type of attendance; spending on tickets; searching information
sources; online research and purchase; the importance of attractions and activities in
selecting pleasure trip destinations; and pleasure trip behaviors are investigated.

Type of performing arts attendance These variables divided into sixteen categories:
play, musical, opera, ballet modern dance ethnic dance, folk dance, folk and ethnic
concert, jazz concert, blues concert, symphony, country concert, new and experimental
concert, rock concert, hip hop concert, and world concert. Each item was recoded into a
dichotomous scale (1 = attended and 2= not attended).

Spending on tickets to the performing arts Spending on tickets to live performances
was classiﬁed according to average ticket price, highest ticket price, and total spending
on live performances during the last 12 months. These were measured with continuous

variables.

52

Table 8. Deﬁnition of Variables Used in CHAID Analysis

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Value of
Construct Variable recoded Deﬁnition of category
category
Dependent variable
Heavy & light 1 Heavy consumer
consumers 2 Light consumer
Art donor & art 1 Art donor
non-donor 2 Art non-donor
Attendance at live 1 Attended live performances in a
performances in a different state or country
different state or 2 Did not attend live performances in
country a different state or country
Independent variables
Socioeconomics Gender 1 Male
2 Female
Age 1 Under 35
2 35-44
3 45-54
4 55-64
5 65 or old
Education 1 Completed 2-year college degree or
less
2 Completed 4-year college degree
3 Completed some graduate courses
or master’s degree
4 Doctoral degree
Employment 1 Employed full-time
2 Other employment status
Annual household 1 Less than $50,000
income 2 $50,000 - $99,999
3 $100,000 - $149,999
4 $150,000 or more
Household Number of persons 1 I live alone
lnfonnation in household 2 2 persons
3 3 persons
4 4 persons or more
Marital status 1 Single
2 Married

 

 

 

 

53

 

Table 8 (cont’d)

 

 

 

 

 

 

 

 

 

 

 

 

 

Value of
Construct Variable recoded Deﬁnition of category
category
Independent variables
Children’s status 1 No child
2 Children living at home
3 Children no longer living at home

Art-related Attendance at live 1 Yes

experiences performances in a 2 No
different state or
country
Art education 1 Yes
during childhood 2 No
Recent art 1 Yes
education 2 NO
Purchase of art 1 Yes
products 2 No
Membership 1 Yes
status in arts 2 No
organizations
Volunteer 1 Yes
status for arts 2 No
organizations
Donation 1 Yes
activity to arts 2 No
organizations

 

 

 

 

 

 

Performing arts information sources Information sources used to learn about live
performance were associated with nine items: TV; magazines; newspapers; radio;
websites specialized in online ticketing; websites of performing art organizations or

centers; acquaintances, friends, family or relatives; art-related publications; and

54

leaﬂets/brochures. Each item was recoded into a dichotomous scale (1 = low and 2=
high) from using a four-point scale (1 = never and 4 = always).

Web-based purchases of art- and culture-related products Web-based purchases
of art- and culture-related products were related ﬁrst to the question; “Do you use the
intemet to research and/or purchase products?” which was measured with a dichotomous
response of l = yes and 2 = no. Three detailed questions were asked about both the online
research and online purchase regarding: art; tickets for performing arts; and antiques and
collections. These variables were recorded and categorized by dichotomous scale (1 =
low and 2= high) from the four-point scale (1 = never and 4 = always).

The importance of culturaL attractions and activities in selectirpg pleasure trip
destinations Also investigated is the importance of the eleven cultural attractions and
activities in selecting pleasure trips: museums; cultural fairs/festivals/events; performing
arts; places to buy local arts/crafts; historical/heritage attractions, sites and districts;
gardens, zoos and aquariums; agricultural attractions and events; purchasing products
grown or processed locally; architecture and buildings; customs and ways of living; and
libraries, literary events, and bookstores). Each destination was measured with a ﬁve-
point scaling range from 1 = not important at all, to 5 = extremely important.

Characteristics of cultural pleasure trips Characteristics of pleasure trips are
related to three questions. The respondents were asked the proportion of the pleasure trips
involving an overnight stay, out-of-state, and out-of-country in the last 12 months
respectively. Three variables related to travel choice were measured using ﬁve-point
scales, ranging from 1 = none, to 5 = all.

All items and scales are presented in Table 9.

55

Table 9. Survey Construction and Core Statistical Analyses in the Proposed Model

 

 

 

 

 

Construct 33:23:: Survey items Scale 32:11:21:
Type of q3_play Play 1: Attended Chi-
perforrning arts q3_music Musical 2: Not square
attendance q3_opera Opera attended test
q3_balle Ballet
q3_modan Modern dance
q3_edanc Ethnic dance
q3_fdanc Folk dance
q3_fecon Folk and ethnic concert
q3_jconc Jazz concert
q3_bconc Blues concert
q3_symph Symphony
q3_cconc Country concert
q3_necon New and experimental concert
q3_rconc Rock concert
q3_hhcon Hip hop concert
q3_wconc World concert
Spending on pric_avr Average price paid for tickets continuous ANOVA
tickets to the pric_hig Highest price paid for a ticket
performing arts pric_tot Total spending tickets
Performing arts inf_tv TV 1: Low Chi-
inforrnation inf_mag Magazines 2: High square
sources inf_new Newspapers test
inf_rad Radio
inf_webl Websites specialized in online
ticketing
inf_web2 Websites of performing art
organizations or centers
inf_fam Acquaintances, friends, family
or relatives
inf _pub Art-related publications
inf_bro Leaﬂets/brochures

 

 

 

 

56

 

 

Table 9 (cont’d)

 

 

 

 

 

 

 

 

 

 

 

to an out-of-country destination

 

 

Construct Observed Survey items Scale Analysrs
variables methods
Web-based intemet Use the intemet to research or 1: Yes Chi-square
purchases to purchase products 2: No test
art- and
culture- web1_art Art (e.g., paintings, photos) 1: Low
related webl _per Tickets to performing arts (e.g., 2: High
products concerts)
web1_ant Antiques and collections
web2_art Art (e.g., paintings, photos) 1: Low
web2_per Tickets to performing arts (e.g., 2: High
concerts)
web2_ant Antiques and collections
The trim_mus Museums 1: not ANOVA
importance of trim_fai Cultural fairs/festivals events important
cultural trim _per Performing arts at all
attractions trim_cra Places to buy local arts/crafts 3: neutral
and activities trim_his Historical/heritage attractions, 5: extremely
in selecting sites and districts important
pleasure trip trim_zoo Gardens, zoos, aquariums
destinations trim_agr Agricultural attractions and
events
trim_loc Purchasing products grown or
processed locally
trim_arg Architecture and buildings
trim_cus Customs and ways of living
trim_lib Libraries, literary events and
bookstores
Characteris— tur_nite The proportion of pleasure trips 1: none ANOVA
tics of involving an overnight stay away 2: few
pleasure trips from home more than 50miles 3: about half
4: most
5: all
tur_stat The proportion of pleasure trips
to an out-of-state destination
tur_intr The proportion of pleasure trips

 

57

 

Data Analysis

The analysis focuses on the number of tickets purchased to live performances
rather than number of performances attended. This is because the primary emphasis was
on the market that purchases tickets. Some persons attend many free performances (e.g.,
school or community events) but do not purchase tickets. A primary problem with these
data is that a person who purchased ten tickets for one event (i.e., bought tickets for
family members or friends) and a person who purchased ten tickets for ten different
events are both treated as ten ticket purchasers. However, their involvement is very
different. Analysis revealed that a 0.586 correlation between number of performances
attended and number of tickets purchased.

The data analysis procedure was divided into three parts: Exhaustive CHAID
analysis for heavy and light consumers; Exhaustive CHAID analysis for donor and non-
donors, and further analyses to identify signiﬁcant differences among segments; and
Exhaustive CHAID analysis for cultural tourists and non-cultural tourists, and further
analyses to indentify signiﬁcant differences among segments. The Exhaustive CHAID
analysis is brieﬂy introduced, and a systematic algorithm through classiﬁcation trees is
discussed. Exhaustive CHAID is selected over the original CHAID algorithm because it
has a better chance of ﬁnding optimal association between dependent variable and
predictors. Finally, the use of Chi-square and One-way analysis of variance (ANOVA)
are discussed. Answer Tree 3.1 and SPSS version 16.0 for Windows were used to analyze

the data.

58

Exhaustive CHAID Analysis

CHAID modeling is a statistical algorithm that shows how the variables of the
model work together in contributing to the predicted probability of response. The CHAID
analysis uses chi-square tests to create multi-way splits which results in a treelike
diagram. The CHAID method explores data quickly and efﬁciently and builds segments
and proﬁles with respect to the desired outcome.

Exhaustive CHAID is a modiﬁcation of CHAID that examines all possible splits
for each predictor. The Exhaustive CHAID can be more accurate and comprehensive
method which explores data exhaustively (Byrd, 2003; Biggs et al., 1991). The ﬁrst
Exhaustive CHAID analysis is implemented to segment the performing arts consumers
into several mutually exclusive and exhaustive subgroups based on the interactions of the
predictors (socioeconomic and household information, and art-related experiences) with
heavy and light consumers according to their ticket purchases. To begin the Exhaustive
CHAID analysis, the overall respondents who indicated the number of tickets purchased
are considered to make up the parent group. The second Exhaustive CHAID analysis
segments heavy consumers of the performing arts based on the interactions between the
predictors (socioeconomic and household information, and art-related experiences) and
donors and non-donors according to their donation activity to arts organizations. The
third Exhaustive CHAID analysis is utilized to segment heavy consumers of the
performing arts based on the interactions of the predictors (socioeconomic and household
information, and art-related experiences) with cultural tourists and non-cultural tourists

according to their attendance at live performances in another different state or country.

59

Heavy consumers are opinion leaders and more involved, innovative and
knowledgeable than light consuemrs (Litvin, 2000; Goldsmith & Litvin, 1999; Goldsmith
et al., 1994). Since heavy consumers represent the greatest volume and proﬁt for the
market, although they are smaller in number than light consumers (Reynolds & Olson,
2001), last two Exhaustive CHAID analyses are conducted using the heavy consumers
who purchased ten or more live performance tickets during the last 12 months. The
analyses were conducted using SPSS Answer Tree 3.0 statistical software (SPSS, 2001).

The CHAID technique essentially involves automatically constructing many
cross-tabulations and working out statistical signiﬁcance of the proportions. A series of
predictor variables are analyzed to determine if splitting the sample leads to a statistically
signiﬁcant effect on the dependent variable. The split is based on a variable which has the
lowest p value or the highest F value if two or more variables have same value (Byrd,
2003). CHAID analysis shows that different variables affect the subgroups in a different
manner from the ﬁrst level downwards. It begins with the entire data split into
appropriate groups based on a most signiﬁcant predictor. A cross-tabulation is associated
with this part and related to the “Chi-squared” part of the name. Bonferroni adjustment is
used in multiple comparison procedures to calculate an adjusted probability of
comparison-wise type I error from the Chi-square distribution function.

The data can be sequentially and selectively split into mutually exclusive
homogeneous groups, where each group is then individually examined. New branches are
then created for these groups with the most important predictor. The degrees of
differentiation between the levels of predictors and the outcome variable display the

optimally split predictors until all predictors are exhausted. The stopping step checks if

60

the tree growing process should be stopped according to the following stopping rules in
the CHAID algorithms: (1) if a node becomes pure; that is, all cases in a node have
identical values of the dependent variable, the node will not be split; (2) if all cases in a
node have identical values for each predictor, the node will not be split; (3) if the current
tree depth reaches the user-speciﬁed maximum tree depth limit value, the tree growing
process will stop; (4) if the size of a node is less than the user-speciﬁed minimum node
size value, the node will not be split; and (5) if the split of a node results in a child node
whose size is less than the user-speciﬁed minimum child node size value, child nodes that
have minimum will merge with the most similar child node as measured by the largest of
the p-values. However, if the resulting number of child nodes is 1, the node will not be
split.

This method is based on principles underpinning a large number of decision tree
procedures, which show how certain sets formed from the predictor variables
differentially predict the dependent variable (Galguera et al., 2006). Homogeneous
segments of performing arts market can be identiﬁed in terms of combinations of the
socioeconomic, household and arts-related experience variables on the heavy and light
consumers of the performing arts. Predictors used Bonferroni adjusted p-value of less
than .05 which was eligible to use for segmentation (Hoare, 2004). The adjusted p-value
is calculated as the p-value times a Bonferroni multiplier which adjusts for multiple tests.
For two other CHAID analyses, variables with a Bonferroni signiﬁcance level of less
than 0.05 were qualiﬁed for selection as a predictor of each dependent variable.

Moreover, given condition was that parent node and child node should contain at least

61

150 and 100 respondents respectively in either of the decision trees. The terminal nodes
or ﬁnal subgroups of each tree diagram were referred to as segments.
Chi-Square Test

Pearson’s chi-square is used to test if a sample of data comes from a population
with a speciﬁc distribution (Snedecor & Cochran, 1989). The chi-square test of goodness
of ﬁt which investigates signiﬁcant differences between the expected and observed
results across the market segments in the: 1) type of performing arts attendance, with
respect to plays; musicals; opera; ballet; modern dance; ethnic dance; folk dance;
folk/ethnic concerts; jazz concerts; blues concerts; symphony; country concerts; new and
experimental concerts; rock concerts; hip-hop concert; world concerts; and other, 2)
performing arts information sources, with respect to TV; magazines; newspapers; radio;
websites specialized for online ticketing; websites of performing art organizations or
centers; acquaintances, friends, family or relatives; art-related publications; and
leaﬂets/brochures, and 3) web-based purchases to art- and culture-related products, with
respect to arts; tickets for performing arts; and antiques and collections.

One-way Analysis of Variance (ANO VA)

ANOVA is employed to examine the dependent variable mean differences across
the segments (Hair et al., 1998). ANOVA identiﬁes the signiﬁcant differences among the
performing arts heavy market segments in the: 1) spending on tickets to live
performances, with respect to the average ticket price; highest ticket price; and total
spending on tickets during the last 12 months, 2) importance of cultural attractions and
activities in selecting pleasure trip destinations with respect to museums; cultural

fairs/festivals/events; performing arts; places to buy local arts/crafts; historical/heritage

62

attractions Sites and districts; gardens, zoos and aquariums; agricultural attractions and
events; purchasing products grown or processed locally; architecture and buildings;
customs and ways of living; and libraries, literary events and bookstores, and 3)
characteristics of pleasure trips, with respect to overnight stays; out-of-state trips; and
out-of-country trips. If signiﬁcant differences across the market segments and selected
variables were found based upon calculated F statistics, then follow-up analysis of
Tukey’s Honestly Signiﬁcant Difference (HSD) method of post hoc tests is further
utilized to examine all possible combinations in order to identify statistically signiﬁcant

mean differences among the segments.

63

CHAPTER 4

RESULTS

Overview

This chapter presents an overview of the survey respondents and the results of the
statistical analyses of the data that contribute to better understanding of performing arts
market by 1) identifying the important predictors of performing arts consumers, and by 2)
proﬁling the heavy-purchase market segments. There are three sections containing
summaries of the results of: l) the ﬁrst Exhaustive CHAID analysis of the number of
tickets purchased; 2) the second Exhaustive CHAID analysis of the donation activity of
heavy consumers; and 3) the third Exhaustive CHAID analysis of the out-of- state or out-

of-country travels to attend live performances among heavy consumers.

Exhaustive CHAID Analysis of Heavy and Light Consumers

The ﬁrst decision tree was created by a systematic algorithm using the Exhausitve
CHAID method, to explore the interactions between predictors and number of tickets
purchased. This decision tree is named “Decision Tree for the Heavy and Light
Consumers of the Performing Arts” (Figure 3). The analysis generated a misclassiﬁcation
matrix and a gain summary of heavy and light consumers. To acquire an optimal model
in a decision tree format, all of the predictors were used and the most signiﬁcant
predictors were identiﬁed among the socioeconomic and household information, and art-
related experiences. Each split was created by determining the strongest combinations
between predictors and outcome variables at each level until all predictors were

exhausted. A Bonferroni adjustment was used in multiple comparison procedures to

64

calculate an adjusted probability of comparison type I errors from the Chi-square
distribution function. Categories of a predictor variable were merged if there was no
statistical distinction between categories with respect to the dependent variable. Variables
with a Bonferroni signiﬁcance level p < 0.05 were qualiﬁed for selection as a predictor of
each parent group. A general summary of the decision tree and a description of the
subgroups that were identifed, based on the CHAID-based decision tree results, are

presented.

Heavy and Light Consumers of the Performing Arts

The results of the ﬁrst Exhaustive CHAID analysis performed to test the
hypothesis are presented: Hypothesis 1: Persons who attend live performances (eg.,
plays, folk dances, musicals and rock concerts) can be classified into substantial,
measurable, accessible and actionable market segments based on the number of tickets
purchased to live performances.

The decision tree resulting from the Exhaustive CHAID analysis grew to six
levels, and the sample was divided into 34 nodes (subgroups). The most signiﬁcant
predictor in this model was donation activiy; other important predictors were income,
age, gender, employment, marital status, children’s status, attendance at live
performances in a different state or country, volunteer status, purchase of arts products
and membership status in arts organizations. These predictors explained the variance in
the heavy and light consumers of the performing arts (Table 10).

The resulting decision tree was divided into two main branches based on donation

activity, and then six sub-branches according to attendance at live performances in a

65

different state or country and annual household income level. Each node in Figure 3

presents the number and percentage of heavy and light consumers, and total respondents.

66

 

Heavy & light consumer ofthe performing arts
Node 0

Heavy consumers (H): 1,267, 30.6%
Light consumers (L): 2,871, 69.4%
Total (T): 4,138, 100.0%

Donation activity
P-value=0.0000. Chi-square=269.5954
l

 

 

 

 

 

 

 

Yes No
Node 1 Node 2
H: 491, 20.6%
L: 1,898, 79.4%
T: 2,389, 57.7%

     
 

| H: 776, 44.4%

L: 973, 55.6%
T: 1,749, 42.3% I

Attendance at live performances in a different state or country Annual household income
P—value=0.0000, Chi-square=6l .0646 P-value=0.0000. Clhr-square=50.7926
l

 
   

 

 

 

 

 

   

  
 

    
 

   
  

  
 

  
 

   
 

 
       
 

   

 

 

     

 

 

 

 

Yes N0 <$50,000 $50,000-$99,999 $100,000-$149,999 Z$150,000
Node 3 Node 4 Node 5 Node 6 Node 7 Node 8
H: 329, 57.7% H: 447, 37.9% H; 82, 134% H: 204, 19.7% H: 129, 24.7% H: 76, 34.4%
L: 241, 42.3% L: 732, 62.1% L: 529, 86.6% L: 830, 80.3% L: 394, 75.3% L: 145. 65.6%
T1570, 13.8% T: 1,179, 28.5% T: 611, 14.8% T: 1,034, 25.0% T: 523, 12.6% T: 221, 5.3%
Age Annual household income Volunteer status . Attendance in a Children’s status Purchase Of art products
P-value=0.0000, P—value=0.0000, P—value:0_0236s diffegentlstat%%rlcg%unﬂy P-value=0.0393, P-value=0.0217,
~ = ._ = . _ -va ue= . , ~_ 2 ' -_ =
Chl—squarrle 21.6358 ChI square I28.9830 Chr—square»5.1211 Chi-square=8.0051 Chl square 6 4710 cm square 5.2682
__I_ J“ —J_
<45 :45 <$50,000 $50,000-$149,999 28150,ooo Yes NO Yes No Yes No
Node 9 Node 10 Node 1 1 Node 12 Node 13 Node 14 Node 15 Node 16 Node 17 Node 21 Node 22

 

   

   
 

 

  

  
 

   

     
 

H: 54,40.3% Hi22,25-3%
L: 80, 59.7% L265,74.7%
T: 134, 3.2% T: 87. 2.1%

 
 

H: 58, 26.5%
L: 161, 73.5%

H: 146, 17.9%
L: 669, 82.1%

H: 265, 37.0% H: 92, 55.4% H: 17, 21.5% H: 65,12.2%
L: 452, 63.0% L: 74, 44.6% L: 62, 78.5% L: 467, 87.8%

H: 64, 41.8%
L: 89, 58.2%

H: 90, 30.4%
L: 206, 69.6%

H: 265, 63.6%
L: 152, 36.4%

 
 

 
   
 

   
   

 
 

 

 

 

 

 

             

 

 

 

 

 

T2153, 3.7% T:417,10,1% T: 296, 7.16% T: 717, 17.3% T: 166, 4.0% T: 79, 1.9% Tz532, 12.9% T: 219, 5.3% T:815, 19.7%
Membership status Marital status Employment Purchase of art products
P-value=0.0001, P-value=0.0496, P—value=0.026 1 , P-value=0.0036,
Chi-square=14.3951 Chi-square=3 .8559 Chi—square=7.40 1 6 Chi-square=8.467l
l I __l_ _l__-
. . , . d .
Yes No SIngle MarrIed Others Employed full-tune Yes No No children Iii/\irggcgilhbihle liviggliictirlfclime
Node 23 Node 24 Node 25 Node 26 Node 27 Node 28 Node 29 Node 30 Node 18 Node 19 Node 20

   

 

   

   

 

   

  

      
 

  
 

H: 69, 24.2%
L: 216, 75.8%
T: 285, 6.9%

H: 40, 37.4% H: 50, 26.5%
L: 67, 62.6% L: 139, 73.5%
T: 107, 2.6% T: 189, 4.6%

H: 108, 43.7%
L: 139, 56.3%
T: 247, 6.0%

H: 45, 30.8%
L: 101, 69.2%
T: 146, 3.5%

H: 111, 54.4%
L: 93, 45.6%
T: 204, 4.9%

H: 157, 33.4% H: 43,16.4% H: 22, 8.2%
L: 313, 66.6% L: 219, 83.6% L: 248, 91.9%
T: 470,11.4% T: 262, 6.3% T: 270, 6.5%

H: 15, 16.3%
L: 77, 83.7%
T: 92, 2.2%

       

 
   
 

 
 

 
 

 
 

 
 

 

    

     

 

 

    

 

 

Children’s status Gender
P-value=0.0319. Chl-square=7-0393 P-value=0.0013, Chi-square=12.9043
_1‘
No children, Children living at home Children no longer living at home Female Male
Node 31 ode 32 Node 33 Node 34

   

          
 

   
 

 
 

     
 

    
  

 
 

 
 

H: 24, 20.0% H: 26, 37.7% H: 8, 4.2% H: 14, 17.3%
L: 96, 80.0% L: 43, 62.3% L: 181, 95.8% L: 67, 82.7%
T: 120, 2.9% T: 69, 1.7% T: 189, 4.6% T: 81, 2.0%

 

       

Figure 3. Decision Tree for Heavy and Light Consumers of the Performing Arts

67

 

 

Decision

1
4.138 re
Light cr
consume

based or

Table 1C

 

I

N

Predict

Donatii
:' Attend;
I in ad
I Annual

Age
Annual

 

I
I
I Mernbr
I

I

. Gender
' PUICha
I

Attend

In a (
Emplo
Childrr
Childh-
Phrcha
I VOIum

I Ma -
w

 

Decision Tree for Heavy and Light Consumers

The root node at the top of the decision tree (Figure 3) represents the sample of

4,138 respondents, who were divided into two categories (heavy vs. light consumers).

Light consumer (n = 2,871, 69.4%) was distributed higher than that of the heavy

consumer (n = 1,267, 30.6%). This node is divided into two child nodes (subgroups),

based on donation activities during the last 12 months.

Table 10. Signiﬁcant Predictors in Decision Tree for the Heavy and Light Consumers

 

 

 

Predictor Donation )6 df Bonferroni
activity adjusted
p-value
Donation activity 269.60 1 0.0000****
Annual household income No 50.79 3 0.0000****
Age Yes 21.64 1 0.0000****
Annual household income Yes 28.98 2 0.0000****
Membership status Yes 14.40 1 0.0001 *"
Gender No 12.90 1 0.0013M
Purchase of art products No 8.47 1 0.0036"
Tearsrrszrtfirrs No 1
Employment Yes 7.40 1 00261“
Children’ 3 status Yes 7.04 1 0.0319*
Children’s status No 6.47 2 0.0393“
Purchase of art products No 5.27 1 0.0217*
Volunteer status No 5.12 1 0.0236*
Marital status Yes 3.86 1 0.0496*

 

*, **. ***, **** Signiﬁcant at p<0.05, 0.01, 0.001 and 0.0001 respectively.

68

 

Based on the ﬁrst split in Figure 3, respondents who donated money to arts
organizations were more likely to be heavy consumers. This donor group was further
broken down to reveal that people who attended live performances in a different state or
country were more likely to be heavy consumers than others, and older respondents (245)
were more likely to be heavy consumers than younger people. Further analysis showed
that members of arts organizations were more likely to be heavy consumers than non-
members.

Respondents most likely to be heavy consumers in the tree model were those: 45
or older, who donated money, had membership in arts organizations, and attended live
performances in a different state or country during the last 12 months (node 23). On the
other hand, the people least likely to be heavy consumers were female with annual
household incomes of less than $50,000, who did not donate money or volunteer in arts
organizations, and did not purchase art products during the last 12 months (node 33).

Certain categories were combined when there was no statistical distinction. For
example, respondents who did not attend live performances in a different state or country
in node 4 divided into three groups by annual household income. People who had annual
household income of $50,000 - $99,999 were grouped with those having an annual
household income of $100,000 - $149,999.

The estimated risk of misclassiﬁcation, based on the tree and the numbers of cases
corresponding to speciﬁc prediction errors, is presented in Table 11. Results of the
Exhaustive CHAID analysis revealed a 27.5% risk of false classiﬁcation: a total of 910 of
3,555 respondents were predicted to be light consumers, but were actually heavy

consumers, and 226 of 583 respondents who were predicted to be heavy consumers, were

69

actually light consumers; there was a cross-validation risk of 28.5%. The risk summary
indicates that the overall correct classiﬁcation accuracy of 72.5% was a substantial
improvement over the base rate of 69.4%. Overall, the predictors were better at predicting
light consumers (92.1% accuracy, with 2,645 of the 2,871 respondents) than heavy
consumers (28.2% accuracy, with 357 of the 1,267 respondents). The ability to capture
variation in heavy and light consumers using the current tree model is less than optimal,

but this tree diagram shows a more detailed parametric model for these data.

Table 11. Results of the Misclassiﬁcation Matrix

 

 

 

 

 

 

 

 

 

 

Actual category
Light consumer Heavy consumer Total
. Light consumer 2,645 910 3,555
:gfgglgtrid Heavy consumer 226 357 583
Total 2,871 1,267 4,138 -
Risk statistics Cross-validation
Risk estimate 0.2745 0.2847
SE of risk estimate 0.0069 0.0070

 

 

 

Gain Summary of the Heavy and Light Consumers

The gain summary presents all of the terminal nodes that have the highest and
lowest proportions of the target category (heavy consumer) within them. The gain index
derived from the analysis can be used as a tool to decide which clusters have a more
positive relationship with the heavy consumers than with the overall sample (Chen,
2003a). An index score of 100% is the crossover point; an index score of more than
100% shows that the corresponding node has a higher than expected proportion of people

with a certain characteristic or behavior. Table 12 reveals 20 nodes with index scores of

70

greater and less than 100% - 10 nodes for heavy consumers and 10 nodes for light
consumers. These nodes are sorted according to the index score, from highest to lowest.
The index score gives a measure of how the score of heavy consumers in this node
compares lwith that for the overall sample. For example, the ﬁrst node in Table 12, node
23, contains 154 heavy consumers out of 213 respondents, or a 72.3% heavy consumer
rate. The index score is about 236%. The proportion of heavy consumers for this node is
over twice the heavy consumer rate for the overall sample. Therefore, the higher the
index score, the more likely respondents would be heavy consumers of performing arts.
The nodes with index scores over 100% are those, 72.3%, 55.4%, 54.4%, 43.7%,
41.8%, 40.3%. 37.7%, 37.4%, 33.4% and 30.8%, containing more than 30.6% heavy
consumers. These nodes represent 67.2% of heavy consumers by targeting 46.1% of the
total market (Table 12). These heav y consumer nodes can be considered actionable
segments that can be valuable target markets for the performing arts (Chen, 2003a). The
gains chart shows important information about which segments to target and which to
avoid. On the other hand, the nodes with index scores of less than 100% consist of light
consumer groups who have a signiﬁcantly lower response rate than the average heavy
consumer rate, as compared with the overall sample. Thus, these light consumer nodes
are considered less important than the heavy consumer nodes. Table 13 presents a

summary of the heavy consumer groups ranked by the index score.

71

Table 12. Gains Summary of the Heavy and Light Consumers of the Performing Arts

 

 

 

Node No. of Proportion No. of Proportion Proportion Index
respondents of all the heavy 0f overall of heavy scoreb
(n) respondents consumers heavy consumer (%)
W») (n) consumer8| response
(%) rate in the
node
Wt!)
23 213 5.1 154 12.2 72.3 236.3° .
13 166 4.0 92 7.3 55.4 181.0
24 204 4.9 111 8.8 54.4 177.8
27 247 6.0 108 8.5 43.7 142.8
Heavy 9 153 3.7 64 5.0 41.8 136.6
consumers 21 134 3.2 54 4.3 40.3 131.7
32 69 1.7 26 2.0 37.7 123.2
25 107 2.6 40 3.2 37.4 122.2
28 470 11.4 157 12.4 33.4 109.2
19 146 3.5 45 3.5 30.8 100.7
Sub-total 1 ,909 46. l 85 1 67.2

16 219 5.3 58 4.6 26.5 86.6

22 87 2.1 22 1.7 25.3 82.7

20 285 6.9 69 5.5 24.2 79.1

14 79 1.9 17 1.3 21.5 70.3

31 120 2.9 24 1.9 20.0 65.4

Light

consumers 17 815 19.7 146 11.5 17.9 58.5
34 81 2.0 14 1.1 17.3 56.5

29 262 6.3 43 3.4 16.4 53.6

18 92 2.2 15 1.2 16.3 53.3

33 189 4.6 8 0.6 4.2 13.7

Sub-total 2,229 53.9 416 32.8

 

 

 

a Heavy consumer in a particular node to all of the heavy consumers in total market.
b Proportion of heavy consumer rates in a particular node compared with the overall proportion of heavy
consumer rate.

c 236.3% = 72.3% (heavy consumer rate in node 23) / 30.6% (the overall proportion of heavy consumers).

72

.0: .5 03:35...— x 5:05:53: .5 38:35:

a

 

 

 

 

 

 

 

 

3.3 8%: - aoeoaeeao 8855.: -

3%.me :5 53» 3.9m - :23: 5 89cm $5 05.3:— 25535: _:=::< - mg «.9. N». _m
eXNéD 03:5,: - "2:353:35 $5 8 335:0: oz -
$3.53 86.2.; - 833 do 6585 - 2o coo» a. 55 so: -

@368 505.5550 2:545: - .9550 5 88m 59:55 a E m3:a::o.to: o>= .5 3:25:32 - m; ”.3 hm o
3..de 65:5."— - 35:35:95 wt: 8 525:8 -
@3de 52.52 - 5255559 055:5 oz -

@835 8&2 .5 8&8 82.8 sore - 58.3.; as 832 838 do 385 228:2 .252 - em :3 3 R
Agwdﬁ :35 5 E3» 2. - .9550 5 88m 595$: a E mooﬁscoto: 3: 5 33:23 oz -
3%.th 05:5“— - 55:35:35 85 c: 53559 -
3%.va cor—52 - 255559: 8.5 5 5553505 :2 -

@153 35.52% - 8:63 .5 2:35 - .52: 5 53> me - EN vém 56 em
:3?th 5&5 5 3&3 owe—56 50.34 - .953 5 28m 529%: a E 30558.5: o3. 3:25:83. -
$3ng 05:5..— - 3.533535 35 8 52359 -
3%. SC 5522 - 5&2 5 856m _ m5 2:85 2283: 325.: -

3..qu :5 E3» 3-3 - .953 5 88m 595%: a :_ m3:a::c.toa o>= 5 3:533 52 - Nd v.3 oé .m—
Ao\eo.eov 2:50.: - 52:35.55 35 o: 525:8 -
3%va 25:35: 05 5 $568 053: 95... - 22:35:30 95 E 5583502 -

Aston 852 a c3855 2:85 - dose a as: a. - 5. Md. 3 mm
@3de SAME 5 3:3: owe—_oo 59?“. - .953 5 88m 595$: a :_ 305585: 3: 5 3:25:82 -
G35. _D 2:th— - m:o_5N_:aw.5 35 S 525:8 -

a o: .5
0.8% 3:3
«cu—.88 DE 1:88.—
65 05: 05 5 35553.5 @5558: .5 5:550 .5 ..\o .5 ..\o .5 o\o 052

 

 

 

 

 

 

 

mg wEEotom 2: .5 3:90 am: 55350 .93: 05 :8 352 2.: .5 585.5% .2 033.

73

.03 .5 05535: x 3:050:00. .5 055005..

a

 

 

 

 

 

 

35.38 “.0282 - 0:5: 8 w:.>= .052 o: 5.259 -

3.5.2.8 5055550 05.25... - 53.2.3 58 25.83 :0035: .5 0:50:— Eo:0m.5: 8.5:... - _._ wdm m.m :—
3.5.38 20 080» 3-9m - 50.858»... 38 0. 0:88:20 02 -
3.5.58 50:82 .. 5055550 05.25... -

3..m.m8 v.00— 5 350: 0w0=o0 80»-.. - 53.3; :8 8ch 303.0: 5 05005 25:025.: 855.. - m. m 3mm v.2 mm
3... .8 20 0.80» 3-9:. - D550 5 08.... 50.08: 8 :_ 03:85.85: 0>= .5 05050.8 02 -
3.958 0850...— - 50.8.2555 3.8 8 50.855 -
3..»«8 0:58 02. 50.0: 0:0 - 0_w:_m -

3..m.N8 00.0. 5 0030: 0w0=50 80»-.. - :8.on :85 0.00— 5 0:505 20:03:: 8.55.. - o: v.5 9m mm
3.5.98 .020 5 we 09.. - .3500 5 08.... 50.08:. 8 :_ 03.85.85: 0>= .5 00:85:05. :2 -
3.5.me 0850: - 53.358»... 8.8 8 0:28:50 -
@130 22.3.2... 2.. a. 0228. 2.8.. as. - 8.5.2 -
3.868 505.5550 05...:5 oz - 0:5: 8 w:.>= .050. o: 5.250 -

25.0%. a... 8 8.0.... 002.8 .8»-.. - 08.8.. s... .8. .o 2:8... 0.2.3.5.. .255 - 0... EM 2 NM
3.0.3 .35 5 0.80» m: - .9550 5 08.... 50.08:. 8 :_ 3085.080: 0>= .5 00850.8 02 -
3.6.2.8 0550.: - 0:355:55 0.8 9 55.855 -

a D: .5
0.8:... 050:
«00—58 U: .558.
55 0:0: 0:. .5 m0_.m_.0.008:0 83050.5 .5 5550a 5 3. .5 .x. .5 3.. 052

 

 

 

 

 

 

 

.228. .m. 0......

74

Two Exhaustive CHAID Segmentations

This section presents results of two different decision trees using a dataset of the
heavy performing arts consumers who purchased 10 or more tickets for live performances
during the last 12 months. Two distinctive Exhaustive CHAID segmentations were
conducted; their market proﬁles were based on characteristics and behaviors of: (1)
donation activity to arts organizations; and (2) attendance at live performances in a
different state or country. A tree diagram, misclassiﬁcation matrix and gain summary
were generated for both analyses.

“Decision Tree for Donation Activity to Arts Organizations” (Figure 4) was used
to identify propensity to donate sub-segments of heavy consumers. Previous results of
CHAID analysis showed that donation activity was the most signiﬁcant predictor of
heavy performing arts consumers. “Decision Tree for Attandance at Live Peformances In
a Different State or Country” (Figure 5) was used to identify inclination to be cultural
tourist sub-segments of heavy consumers. The respondents who had attended live

performances in a different state or country were considered to be cultural tourists.

Donation Activity to Arts Organizations
The results of the second Exhaustive CHAID analysis performed to test the
second hypothesis are presented: Hypothesis 2: Heavy consumers of the performing arts
are further divided into substantial, measurable, accessible and actionable market
segments based on the propensity to donate to arts, heritage or cultural organizations.
The Decision Tree for Donation Activity to Arts Organizations resulting consists
of four levels and the donation activity of heavy consumers is segmented into a total of

six terminal nodes. Membership status in arts organizations, age, volunteer status, and

75

purchase of art products were identiﬁed as the most signiﬁcant predictors in explaining
the variance in the heavy consumers and donation activity in the decision tree (Table 14).

The resulting decision tree was divided into two main branches containing nine
nodes. Branch 1 consists of nodes 1, 3 and 4; and branch 2 of node 2, 5, 6, 7, 8 and 9. In
the CHAID tree diagram (Figure 4), the number and percentage of donors, non-donors
and total respondents deﬁnes each node. The predictor, chi-square value and Bonferroni
adjusted p-value shown below distinctive node divide respondents into mutually

exclusive groups.

Table 14. Signiﬁcant Predictors in the Decision Tree for Propensity to Donate

 

 

 

Predictor Chi-square df a djﬁgtretdelpr-(itrzlue
Membership status 218.12 1 0.0000"
Age 50.11 2 0.0000"
Purchase of art products 21.05 1 0.0000"
Volunteer status 4.18 1 0.0409*

 

 

*, ** Signiﬁcant at p < 0.05 and 0.001, respectively.

Decision Tree for Donation Activity to Arts Organizations

The root node at the top of the decision tree (Figure 4) represents the sample of
1,267 heavy consumers, who were divided into two categories (donor and non-donor).
There were more donors (n = 776, 61.3%) than non-donors (n = 491, 38.7%) in the study
sample. This root node was split, based on the membership status in arts organizations:
members (90.3%) were more likely to donate money to arts organizations than non-
members (47.2%). An additional split on members showed that volunteers (93.3%) were

more likely to donate money to arts organizations than non-volunteers (87.3%).

76

Therefore, the most likely to be donors in the decision tree were node 3 (segment 1), who
were members and volunteered in arts organizations during the last 12 months. However,
people in node 5 (segment 6) were the least likely to be donors; they were under 35 years

old who were not members in arts organizations during the last 12 months.

Donation activities
Node 0

Yes (Y): 776, 61.3%
No (N): 491, 38.7%
Total (T): 1,267, 100.0%

 

 

 

 

Membership status
P-value=0.0000, Chi-square=218.1228
l

 

 

 

 

Yes No
Node 1 Node 2
Y: 373, 90.3% Y: 403, 47.2%

N: 40, 9.7%
T:4l3.32.6%

 

 

Volunteer status

P-value=0.0409, Chi-square=4. 1 8 l 6
1

 

 

N: 451, 52.8%
T: 854. 67.4%

 

 

Age

P-value=0.0000, Chi-square=50.1 121
|

 

 

 

Yes No <35 35-54 255
Node 3 Node 4 Node 5 Node 6 Node 7
Y: 194, 93.3% Y: 179, 87.3% Y: 35, 26.1% Y: 204, 44.5% Y: 164, 62.6%
N: 14, 6.7% N: 26, 12.7% N: 99, 73.9% 1 N: 254, 55.5% N: 98, 37.4%

T: 208, 16.4%

 

 

T: 205, 16.2%

 

 

T: 134, 10.6%

 

 

T: 458, 36.1%

 

 

T: 262, 20.7%

 

Segment 1 Segment 2 16mm Purchase of art products Segment 3
P-value=0.0000, Cl'ii-square=2 l .0501
Yes No
Node 8 Node 9
Y: 156, 52.4% Y: 48, 30.0%

 

 

 

N: 142, 47.6%
T: 298, 23.5%

 

 

 

N: 112, 70.0%
T: 160, 12.6%

 

 

Segment 4

Segment 5

Figure 4. Decision Tree for Donation Activity to Arts Organizations

The estimated risk of misclassiﬁcation, based on the tree and the number of cases
corresponding to speciﬁc prediction errors is presented in Table 15. Results of the
Exhaustive CHAID analysis revealed a 28.7% risk of false classiﬁcation; 280 of the 973
respondents predicted to be donors had not donated, and 83 of the 294 respondents
predicted to be non-donors had actually donated in the last 12 months. The analysis of a
risk on cross-validation was 28.7%. The risk summary indicates that the overall correct
classiﬁcation accuracy of 71.3% was a substantial improvement over the base arts donor
rate of 61.3%. Overall, the predictors were better for donors (89.3% accuracy; with 693

of the 776 respondents) than for non-donors (43.0% accuracy; with 211 of the 491).

Table 15. Results of the Misclassiﬁcation Matrix

 

 

 

 

 

 

 

 

 

 

 

 

Actual category
Arts donor Arts non-donor Total
Arts donor 693 280 973
giggiggd Arts non-donor 83 211 294
Total 776 491 1,267
Risk statistics Cross-validation
Risk estimate 0.2865 0.2865
SE of risk estimate 0.0127 0.0127

 

Gain Summary of the Donation Activity

The gain summary is represented by six terminal nodes, which were sorted by an
index score listing the donor rates from their highest to lowest proportions (Table 16).
The index column gave a measure of how the score of donors in this node compared with
the overall sample (61.3%). The higher the index score, the more likely the heavy

consumers were to donate money to arts organizations. Nodes 3, 4 and 7 which have an

78

index score of greater than 100%, have a higher response rate of donations to arts
organizations than the overall sample. For example, node 3 contains 194 art donors out of
208 heavy consumers. This index score is about 152%. The proportion of donors for this
node is more than one-half of the donor rates for the overall sample.

The nodes with index scores greater than 100% represented 69.2% of the donors
by targeting 53.3% of the heavy market. Nodes 8, 9 and 5, which had an index score of
less than 100 % indicating art non-donor groups, had signiﬁcantly lower response rates
compared to the overall sample. These nodes are mutually exclusive respondent groups
that represent effective segments of the performing arts heavy consumers and donation

market.

Table 16. Gains Summary of the Donation Activity

 

Node No. of Proportion No. of Proportion Proportion of lndex scoreb
respondents of all the donors 0f overall donor response (%)
(n) respondents (n) donorsa rate in the
(%) (%) node
(%)

    

8 298 156 20.1 52.3 85.3
9 160 12.6 48 6.2 30.0 48.9
5 134 10.6 35 4.5 26.1 42.6

 

' Donors in a particular node to all of the donors in the heavy market.
Proportion of donor rates in the node compared to the overall sample.

c 152.3% = 93.3% (donor rate in node 3) / 61.3% (donor rate in the overall sample).

79

Proﬁles of the Propensity to Donate to Arts Organizations Sub-Segments

The propensity to donate six sub-segments of heavy consumers are proﬁled based
on socioeconomic and household characteristics, and other behaviors toward the
performing arts and pleasure trips, with respect to: (1) the type of performing arts
attendance; (2) spending on tickets to the performing arts; (3) performing arts information
sources; (4) web-based purchases to art- and culture-related products; (5) the importance
of cultural attractions and activities in selecting pleasure trip destinations; and (6)
characteristics of pleasure trips. Chi-square tests and ANOVA were the analytical tools
used in comparisons of these six segments.

Socioeconomic Characteristics

Cross-tabulations were performed to provide socioeconomic characteristics of the
propensity to donate six sub-segments of the heavy consumers with regards to: (1)
gender; (2) age; (3) sexual orientation; (4) race; (5) education; (6) annual household
income; (7) employment; (8) number of persons in household; and (9) household
composition. Statistically signiﬁcant differences among the segments were found in age,
sexual orientation, education, annual household income and employment (p<0.001), and
in gender and race (p<0.01) (Table 17).

Regardless of the segment, a majority in the performing arts donation activity of
heavy consumers were female, heterosexual and Caucasian, had a 4-year college degree
or higher, an annual household income of $50,000 or higher, and two persons or more
residing in their households. A summary of the socioeconomic characteristics of the six

donation market segments are presented in Table 18.

80

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5.on .8.ch _odvm “350$ch
Smwm n ax m _ .3: u "x 23% u «x 222% .89
m._ QM mg. 0.x» o o o o odo— vée 9mm 06— v2 0
ed w.m a; ”do 0 o ode adv c 643 v. _ m gm. of m
5.0 w.w vé _dw o o .60 QMM 0 New ham m.mN mam v
o m.» «.0 56$ 0.: va o o o 9N0 vﬁm EON New. m
m6 #6 ON 13 «:2 odm mdm Q: vé oSw 9mm N.o_ mom N
We 52 md wa WV— m.mm YNM 92 up v.2. 9mm v.3 wow _
r m m
e ( \l/
m m m m
m m m .6. m m
m M. W H n W m
n s a to m m e m w
m a my .m m M M M. w m m m m
or. m a. m a ‘5. .5. u u n M m. m ..
m m m
E g Ase m m a
cocﬁcoto _asxom uw< 5950 P N S

 

 

 

 

 

 

 

maaoﬁwombsm 5m 08:09 9 banana; 05 mo moumtouoﬂaao omanoooomoom .2 2an

81

 

£3. 2:5» N ~ 072:.

Table 17. (cont’d)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

,_, "2 ". V! “l ‘9 ‘0.
:: Others o as N <1- N o~ o
o tn M tn N N v m .—
Exa v: 8
.2 E, a :5
g- n \A
. “. "2 V‘. °°. V. V.
to Employed-full trme m o h m l‘ c N?"
Vl' \O w l‘ l‘ W
. °‘. ‘0. °9 “l V‘: ‘0.
$150,000 or higher m o o r~ m o
N N u—u N — '—
O
E
8
'5 $100 000 $149 999 V’ °°' '7' "i “i ‘Q
- l‘
g ’ ’ § Si N 83 a 52 Q _
4: q; 2 8
saw - c.»-
I1
.8 ~. 9: ~92 6. st. ‘0. «x a
3 $50,000 - $99,999 as m «— In r~ co
= M M V M V M
V“. 00 0‘. to m (‘i
Under $50,000 53 ,5 :3 .5 ,5 R
-- ‘9 "I ‘12 o N
Doctoral degree a; r: g : .,; ,5
Completed some graduate courses V. o! O: "I O. ‘9
a , -— N on as c b as
o or Master 3 degree n v v N M N g S
‘5 Q «i c,
O Q, 0‘ O
:1 n V
3?: o. 0: m. "2 N. «2 «x a
4-year college degree N h an o —- In
N N .— M M M
“2 "‘2 ‘0. O. “t ‘t
2-year college degree or less o N m as m 00
F‘ — N N M N
v n v F —- un
Others v —. o N m' s g
o v—
u A - o
s s e 9'
11 g-
Caucasran or White tn co as b o N
O‘ O‘ O5 O\ O\ O\
‘1‘. (‘i 1‘. V? ‘0. ‘0.
Percentage of respondents (%) :2 :2 g R N e:
co V) N 00 O V
O O \O ('1
Number of respondents (n) N N N g; :3 __ :3), 3
to 5‘3
Segment ~ N M V W ‘0 i5 Eh
[- m

 

 

 

 

 

 

 

 

82

 

Table 17. (cont’d)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Married/Partnered with children no 1‘. ": '1 ‘1: V.

. . -— r~ m o <- 0
longer lrvmg at home M m w —- -
Married/Partnered with children 1‘. "I V‘. 1‘. 9. ‘9

. . M l‘ V M C '—
lrvrng at home to N - m o N
C
02
a:
m . . . N ‘0. °°. "I N ‘12
8. Married/Partnered wrthout children O5 :2 :3 : ,5 a 8
E ’3‘ ° 8
0 Q\, 3 <5
2 . . . u y,
_2 Single wrth children no longer a! q m. ix. —: o "x
8 livin th °° ‘9 °* -: *9
z, ga ome
o
:1:
Smgle wrth children lrvmg at home v’ r,; v‘ ,0- h- ,,;
. . . 0‘. “i as ". co 9.
Single wrthout children m o (o- c 06 m
1— —I '— V}
V. ‘9. <1- 9. “i 0‘.
4 persons or more v as V; oo —- \o
E m — m In N
o
J:
8
S
,2 3 <2 «2 a: 0: sr. ~O.
rsons so as — co v 00
.5 pe .... ._. ._ _ ._ .— Q _
2 A l\ O
as" 2 2-
iv n V
.,_ a: or «a h. ~o. 9: “k Q
o 2 persons 3 m l‘ <2- nn ax
'- W \O M N N
O
.D
E
3
Z ' "'3 l‘- O‘- v oo ‘0.
l lrve alone 3 2 E 06 06 g
4, N. h. w. 0. «2
Percentage of respondents (%) o o a (:3 N a
co m m 00 o v
C O \O O\ \0
Number of respondents (n) N N N N __ r2 :3: g
'5
U
*3 SE
Segment — N m v m \o g .5)
[— m

 

 

 

 

 

 

 

 

83

 

.225: he owﬁcouuom x 85238: :0 03:33“.

 

.§..$. use. a was: 3:2 2. 5.22. 5.3 neogaeuraz -
$.38 22.38.. 2.. a was... 282 93 -

.Agmdwv aadim U5 686mm Guggen— mo 05005 Eon—025: 3355‘ .
62%di 8&2. .5sz 8 82:8 226an 2:8 :22an0 -

$8: 2.3 3 e8» 2 -

£98. 225: -

2o .5 28» mm -
8282.83: 3:. 5 2:23:58 02 -

9m—

9N0

ﬁom

 

.3... .5 use. a was: 39.2 2. 86.2“. 5.3
s 3.2.... 2:2. a was: 5.2.5 5.3 .22. £2.58 .o 8.52 -
2.2.23 22.2.3 2.. a 2.28. 282. S; -
§§c .0ng 8 25.8.20 .585 22.352. .252 -
.A..\ow.omv came: :23... 8 82:8 22.2.3 088 :BoEEoU -
25.8. 52.. a $2 a. -
3&8 one“... -

828259». 8.3 o. 32:23 02 -
EBEanS at: 5 2:23:52 -

:1

QB

N6—

 

.Ao\oh. _ 8 2:3 8 w£>= 32:: o: coho—Eu .23
:o Giana 2:0: “a 9.3: coho—Eu .23 55:. 682:8: .5 3:82 -
.A..\..o.v3 20:23:: 5 $528: 2:8: 93. -
A2323. 2&2 S 25.8.» mo 2:85 20:32.: 3:52. -
6.2.003 3:qu 2..sz .5 82:8 3268» 2:8 :22an0 -
éxﬁdmv 320 ho we?» we -
6.2.208 225..“ -

238259: 8.8 S cohoo.§_o> -
822.259: 9.8 5 2:23:52 -

m.m_

Wm:

v.2

 

 

83:88:20

 

20860.5 .3 .835qu

 

«22.2.
a:
22m
8th
.3 .x.

 

288
MO ..\..

 

3:2.
:88:

«0.x.

 

Eon—wow

 

 

3:“...Ewombsm 2% 82:5 9 bﬁggoi 2: mo 3:388:20 omEo:oooomoom 05 mo haEEdm .E 2an

84

.225: wﬁpcota .3 023.820.: x 8:02.280: .3 022—020: a

 

 

 

 

 

.218:
5:2... sea? 8:588:35 :8 2.8.8: 82:: sea? 28% -
éxﬁsﬂ 2563 :2: $0. .3 2:52: 25:00.25: 8:52 -
Ao\om.v:v 80. .5 00&0: 0w0=oo 80.3. - 2282.8w 5 288$.anme ”0:“:MD - wN Gm 2o. :
8:22:55 002 8:3 28 .833. 5 8w 8:2: 3028—02 - . . . .: 0: Z

.28: 2.. $2 8 .08: -
3°33: 22:0..— -
2.8.8: 28.. a was: :28: 23 83588252 -
AXE. _ m: 20:025.: 2: E 2:28.. 0:2: .5 0:50: :5”: -
A2568 80. .5 8&0: 0w0=8 80%-: -

2282.895 28 2 2:20:22: 52 -
.28: Ea $2 8.8. -
3023: 22:0,: -
.3063: 2:0: 8 w2>= 5:220 :23 Boga—32:82 -

38.3 2535 88.88 - 30:8: 5 88 as: -

6.80:3 8:2: 5 25.853 0:52: 20:02.0: 8:22 - : o .980» :m-mm - m. _ Wmm mam :
85:. 2.: 5:2: 5 3&0: 0w0=50 80%-: - _ v N
328282: 3.8 2 2:20:22: 02 -
2.8: 20 52 8-8 - . . . .
A2835 22:0“— -
8225:
.3
085 2:0:
828.: 2259 -2500:
85:20:82.0 22030.5 .3 525.85 3 o\.. .3 2. .3 .x. Eon—wow

 

 

 

 

 

 

 

6.28: .2 29¢

85

Type of Performing Arts Attendance

Cross-tabulations were performed to provide the type of attendance at live
performances during the last 12 months among the propensity to donate sub-segments,
with regards to: (1) plays; (2) musicals; (3) opera; (4) ballet; (5) modern dance; (6) ethnic
dance; (7) folk dance; (8) folk/ethnic concerts; (9) jazz concerts; (10) blues concerts; (11)
symphony; (12) country concerts; (13) new and experimental concerts; (14) rock
concerts; (15) hip-hop concert; (16) world concerts; and (17) other. The results of chi-
square tests indicated that statistical signiﬁcant differences among the segments existed
for attendance at: plays, musicals, opera, ballet, modern dance, folk/ethnic concerts, jazz
concerts, symphony, new and experimental concerts and rock concerts (p<0.001); ethnic
dance, blues concerts, hip hop concerts and. world concerts (p<0.01); and folk dance
(p<0.05). However, differences were not signiﬁcant for attendance at country concerts
and others (Table 19).

In all six segments, respondents attending musicals and plays showed the highest
attendance during the last 12 months among the live performances, while folk dance, new
and experimental concerts, hip hop concerts and world concerts were the least attended of
the live performances across segments. Regardless of the segment, more than half of the
respondents had attended plays and musicals in the last 12 months; attendance was higher

for musicals than for plays.

86

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sex. .89.. .83.. Siva 5.3.. .89.. 85.2.35
3%. u «x 3.9% u «x .32 u "x 81% u «x 3.3. u "x 3...; u «x 2.3% an;
S «.8 Z: 93 2.. 3... 2 «.8 3K .8 9% 9:. 2: x. o
2 3.. N... «.8 3 «.8 ca 93 an. 3. a. G 2:. gm. 8. m
2 3.. Z. a. a a... 3.. .5 3.. f... E: ..$ 3m 08 ”a v
3 3.. no. 2... 3. 3.. 2.. an. .2 a... a: EN EN SN m
3 58 S. 2... 3. «.8 m. . N 2: 3.. 3. .ﬁ 3m 3. mg m
B: 3... 3. 2... SN 03 a: .ﬁ 3s .3 3... «.2 .3. mom .
0W
5 \m
n (
d d d d d d e m
m m m m m .n. m m
n u n... n n... ﬂ 0 d
d a .d a d a d a d a d a w n
m. m .m m m m m m 4.". m m m m m
m n m n m n m n m n m n f a
u ..m n ..m n m n M u ..m n M o h
A D A D A D A D A D A D m... o
m a m
m b e
.s. .s. 3.. es. 3... Q... m m m
3:3 025m 85% E332 8.3m 280 .86...)— me P N S

 

 

 

 

 

 

 

 

 

 

mEoEwombzm cam Bacon— 8 5.3255 2.: 3 mooqmgofom 2,5 3 352.83. mo 33. d— 2an

87

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

$8.1... .8...v.. 8.9.. .89.. .8...v.. 8.3.. 858.85
83 u N. 8.8 u m. 8:. u «x .88 u a. 3.3 n m. .8. . u «x 0.8.5... an;
2.. m. .8 N... 3.. .8. 3... Z. 3.. o... o. a w. 2... 8.... X. o
.8. 8.8 2.. 2.. 3. o. .8 m... «.8 9m «.8 3 m8 3. 8. m
2.. 8.3 98 2.. «.m. on. 2.. m... ..... an. E «.8 m8 .8 v
3. 3.. m. . m 8.8 5.. m. 3. .8 8.8 .8. 9m. .2 8.8 28 SN m
2: m8 .8 8.8 8.: «.8 To... «:8 n8. n8 ...~ 8.8 3. m8. m
n: 3.. a? New 2.. m8 8.8 8.8 «.3 m: .3 m8 .8. 8m .
m
m m
d d d d d d m m
n n n... n n n n
m m m m m m m m
d a d a d a d a d a d a m. n
.m m m m w m m m w m. u m m m
n n n n n n n n n n n n f a
m m m m m m m m m w m w o n
A D A D A D A D A D A D n. o t
m m m
.8 .8 .8 .8 .8 .8 m m. m
.3280 .3550 Accnaﬁxm .8280 82m .5050 ~NE. 50:8 35:03.8... 35% 5o... P N S

 

 

 

 

 

 

 

 

 

 

8.2.8. .8. 2.3

88

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

00N.0u.. 8.0x. 8.0x. .8...v.. .8...v.. 85.8.8.0
00N.0 n N0. 000.0. u N0. 0N3. u 0.. 0.0.8 u «0. 0N...VN u «0. 0.00050 08.0
SN 0.8 0.0 0.0.. 0.0 0.0.. 0.00 0.00 N. . 0.00 0.... 0
..0N a. . N 0.. .00 N. 0.00 0.00 0.00 0N 0.8 0N. 0
0.0N 0. .0 0.0 0.00 0.. 0.0.. N00 0.00 0N 0.8 0.0N ..
ZN 0.8 0.N 0.8 0... N8 0... 0.00 0.. .00 SN 0
N.00 0.00 0.0 0.0 0N 0.8 0.8 .00 0... 0.00 N.0. N
.00 0.00 0.0 S... 0.. 0.00 0.00 0.00 0.0 N00 0.0. .
d d d d d
n n n n n
C C e e e
H u H ﬂ 9....
d a d a d a d a d a
e t e t e t e t e t
d 0 d 0 d 0 d 0 d 0
n n n n n n n n n n
m m m m m m .0... m .u. w
A D A D A D A D A D
A0\ov Ao\ov m
0:00:00 3:020:00: e
.8 .8 .8 .8 352035 .o m
0050 0:00:00 28>» 50:00 no: 9: 50:00 0.03— .05... 302 03200000— S

 

 

 

 

 

 

 

 

5680 .0. 0.000

89

Spending on Tickets to the Performing Arts

The results of one-way ANOVA indicated statistically signiﬁcant differences
among segments in the average ticket price and total spending on tickets (p< 0.001)
(Table 20). The highest price for a ticket for live performances, however, was not
signiﬁcantly different across six segments.

Further analysis using Tukey’s HSD post_h0c test revealed signiﬁcant mean
differences: (1) in the average ticket price for tickets purchased between segment 1 and
segments 3 and 4; between segment 2 and segment 6; between segment 3 and segment 6,
between segment 4 and segment 6, and between; and between segment 5 and segment 6;
and (2) in the total spending on tickets between segment 1 and segment 6, between
segment 2 and segment 6, between segment 3 and segments 5 and 6, and between
segment 4 and segment 6.

The average ticket price was the highest in segments 3 and 4 while the highest
ticket price was the greatest in segment 2 and the total spending on tickets was the
highest in segment 3. On the other hand, the average ticket price, the highest ticket price,

and the total spending on tickets were all the lowest in segment 6.

9O

Table 20. Spending on Tickets by the Propensity to Donate Six Sub-Segments

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Segment Test
Ticket 1 2 3 4 5 6 statistic/
Price (16.4%) (16.2%) (20.7%) (23.5%) (12.6%) (10.6%) Signiﬁ-
paid cance
Avera e a
ticket g Mea“ (b5) 44.00 49.00 51.00 51.00 49.00 43.00
rice 48.00 F = 7.621
p Range 8.00- 8.00- 8.00- 12.00- 5.00- 10.00- p<0-00|
(s) 100.00 1 1 1.00 100.00 125.00 100.00 100.00
Highest Mean (3)
ticket 86.00 86.00 93.00 88.00 88.00 81.00 80.00 F = 1.386
Pm" Range 20.00- 26.00- 12.00- 15.00- 10.00- 10.00- p=0-227
(3) 350.00 500.00 1000.00 750.00 300.00 300.00
Total Mean ($)
Spending 740.00 775.00 775.00 791.00 752.00 714.00 633.00 F = 8‘38]
32ka Range 128.00- 90.00- 100.00- 120.00- 100.00- 150.00- p<0-00|
($) 1000.00 1000.00 1000.00 1000.00 1000.00 1000.00

 

a Mean values rounded to the nearest dollar.
b Mean score for average ticket price in the six segments.

91

 

Performing Arts Information Sources

Chi-square tests were used to compare the six segments as to information sources
used to search for live theater, dance performances and music concerts (Table 21).
Signiﬁcant differences among the six segments existed for: newspapers, websites
specialized for online ticketing, acquaintances, friends, family or relatives, art-related
publications and leaﬂets/brochures (p<0.001); and websites of performing art
organizations or centers (p<0.01). TV, magazines and radio, however, no signiﬁcant
differences among the segments were existed.

Regardless of the segment, newspapers; websites that specialized in online
ticketing; websites of performing art organizations or centers; and acquaintances, ﬁiends,
family or relatives were the most important information sources. Three-quarters or more
of the respondents in segments 1, 2, 3 and 4 usually or always acquired information about
live performances ﬁ'om newspapers; three-quarters of respondents in segments 4, 5 and 6
usually or always acquired information from websites specializing in online ticketing or
from the websites of performing art organizations or centers; and three-quarters of the
respondents in segment 1, which showed the highest percentage among segments, usually
or always obtained information from acquaintances, friends, family or relatives. More
than one-third of the respondents in segments 1 and 2 usually or always obtained
information from art-related publications, while the majority of those in segments 3, 4, 5
and 6 seldom or never used them. Over half of the respondents in segments 1 and 2
usually or always gained information from leaﬂets or brochures, while more than two-

thirds of respondents in segments 4, 5 and 6 seldom or never used these sources.

92

Table 21. Information Sources by the Propensity to Donate Six Sub-Segments

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

segment 1 2 3 4 5 6 Test
. (16.4%) (16.2%) (20.7%) (23.5%) (12.6%) (10.6%) statistic/
lsrlfsmngatlon % % % % % % Signiﬁcance
TV
31:33; 63.2 67.3 63.2 54.9 63.7 59.0 x2 - 9493
(1:32;? 36.8 32.7 36.8 45.1 36.3 41.0 p=0.09l
Ma azines
g 5:166:12: 69.7 75.4 71.5 72.1 79.5 75.8 x2 = 5922
2:32;? 30.3 24.6 28.5 27.9 20.5 24.2 p=0.314
News a ers
p p 51:35:: 16.1 25.1 14.9 22.6 31.4 43.3 x’= 51.804
2:52;? 83.9 74.9 85.1 77.4 68.6 56.7 p<0.001
Radio
81:35:; 48.5 53.8 50.2 41.0 45.2 48.1 x2 = 9365
2:32;}! 51.5 46.2 49.8 59.0 54.8 51.9 p=0.095
Websites Never/
specialized seldom 40.0 42.6 38.2 28.0 23.3 22.1 X2 = 32812
for online Usuall /
ticketing alway: 60.0 57.4 61.8 72.0 76.7 77.9 p<0.001
Websites of Never/
perfoming seldom 23.3 29.6 28.2 15.6 20.3 22.6 x2 = 18.73,
artorganiza- U ll /
tions amy: 76.7 70.4 71.8 84.4 79.7 77.4 p<0.01
01' centers
Acquaintan- Never/
ces, friends, seldom 25.0 45.0 40.9 32.9 42.1 31.8 x2 = 25.123
family Usuall /
or relatives away: 75.0 55.0 59.1 67.1 57.9 68.2 p<0.001
Art-related Never/
publications seldom 59.5 66.5 79.9 83.3 92.3 87.2 x’ = 34144
3:32;? 40.5 33.5 20.1 16.7 7.7 12.8 p<0.001
Leaflets/
brochures :fc‘l’oeg 41.5 44.7 53.7 67.9 74.7 69.4 x2 = 75663
2:32;? 58.5 55.3 46.3 32.1 25.3 30.6 p<0.001

 

 

93

 

Web-Based Purchases to Art- and Culture-Related Products

Chi-square tests were also used to compare: (1) Internet use; (2) on-line research;
and (3) the purchase of art- and culture-related products (Table 22). Statistically
signiﬁcant differences were found among the segments for research of the arts;
researching tickets for the performing arts; researching antiques and collections; the
purchase of arts; and the purchase of tickets for the performing arts (p<0.001); Internet
use (p<0.01); and the purchase of antiques and collections (p<0.05).

Regardless of the segment, almost all of the respondents used the Internet to
research or purchase art- and culture-related products. A majority in all six segments
usually or always used the lntemet to research and purchase tickets for the performing
arts, but they seldom or never used the lntemet to research and purchase for arts, antiques
and collections. In particular, respondents in segments 4, 5 and 6 had higher percentages
of usually or always researching and purchasing tickets for performing arts than did other
segments. On the other hand, those in segments l, 4 and 6 had higher percentages of
usually or always researching and purchasing arts, antiques and collections than did other
segments.

The Importance of Cultural Attractions and Activities in Selecting Pleasure Trip
Destinations

Statistically signiﬁcant differences were found among segments in the importance
ratings of: museums; cultural fairs/festivals/events; performing arts; places to buy local
arts/crafts; historical/heritage attractions, sites and districts; architecture and buildings;
and customs and ways of living (p<0.001); and libraries, literary events and bookstores

(p<0.01). However, no statistical signiﬁcant difference existed among the six segments

94

for gardens, zoos and aquariums; agricultural attractions and events; and purchasing

products grown or processed locally (Table 23).

Table 22. Web-Based Purchases by the Propensity to Donate Six Sub-Segments

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Segment Test
1 2 3 4 5 6 statistic/
(16.4%) (16.2%) (20.7%) (23.5%) (12.6%) ( 10.6%) . . _
lntemet % % % % % % Slgmﬁ
Use cance
lntemet use
Use 94.2 93.2 92.0 97.7 98.8 98.5 X2 =20_521
N‘” 5.8 6.8 8.0 2.3 1.2 1.5 p<0.01
use
Research arts Never/
seldom 68.1 73.0 75.0 63.9 85.3 70.5 X2 = 5.625
usually’ 31.9 27.0 25.0 36.1 14.7 29.5 p<0.001
always
Research Never/
tickets for seldom 13.3 17.4 15.8 5.5 9.6 4.5 X2 =29.243
performing
ans Usua'ly/ 86.7 82.6 84.2 94.5 90.4 95.5 p<0-001
always
Research for Never/
antiques and seldom 67.2 78.3 82.1 70.1 84.4 74.8 X2 =24.325
collections Usuall /
V 32.8 21.7 17.9 29.9 15.6 25.2 p<0.001
always
Purchase arts Never/
seldom 87.3 92.6 92.9 86.6 96.8 85.6 X2 =20.023
usually/ 12.7 7.4 7.1 13.4 3.2 14.4 p=0.001
always
Purchase Never/
tickets for seldom 22.4 23.7 31.1 9.6 10.3 9.2 X2 =61 _27]
performing
arts “wall“ 77.6 76.3 68.9 90.4 89.7 90.8 p<0.001
always
Purchase Never/
antiques and seldom 85.2 93.1 92.4 88.6 94.9 88.5 X2 =13.728
collections Usuall /
y 14.8 6.9 7.6 11.4 5.1 11.5 p<0.05
always

 

 

95

 

Results of the Tukey’s HSD post_hoc test indicated signiﬁcant mean differences:
(1) for museums between segment 1 and segments 3, 4, 5 and 6; between segment 2 and
segments 4, 5 and 6; and between segment 3 and segment 5; (2) for cultural
fairs/festivals/events between segment 1 and segments 4, 5 and 6; (3) for performing arts
between segment 1 and segments 2, 4, 5 and 6; and between segment 3 and segments 4
and 5; (4) for places to buy local arts/crafts between segment 1 and segments 5 and 6;
between segment 2 and segments 5 and 6; between segment 3 and segments 5 and 6; and
between segment 4 and segments 5 and 6; (5) for historical/heritage attractions, sites and
districts between segment 1 and segments 3, 4, 5 and 6; between segment 2 and segments
5 and 6; and between segment 3 and segment 6; (6) for architecture and buildings
between segment 1 and segments 3, 4 and 5; between segment 2 and segment 5; between
segment 3 and segment 5; and between segment 5 and segment 6; (7) for customs and
ways of living between segment 1 and segments 4 and 5; and between segment 3 and
segments 4 and 5; (8) for libraries, literary events and bookstores between segment 1 and
segments 5 and 6.

“Historical/heritage attractions, sites and districts,” “gardens, zoos and
aquariums,” “performing arts,” and “museums” were the most important attractions and
activities among the six segments when respondents were selecting destinations for
pleasure trips. On the other hand, “agricultural attractions and events,” “purchasing
products grown or processed locally,” and “libraries, literary events and bookstores” were
the least important attractions and activities in all six segments. People in Segment 1
indicated almost all of the cultural attractions and activities were more important for

pleasure trip destinations than that of other segments.

96

Table 23. The Importance of Cultural Attractions and Activities in Selecting Pleasure

Trip Destinations by the Propensity to Donate Six Sub-Segments

 

 

 

 

 

 

 

 

 

 

 

 

 

Segment Test
1 2 3 4 5 6 statistic/
(164%) (16.2%) (20.7%) (23.5%) (12.6%) (10.6%) . .

cultural 3 8c 3 7 3 7 3 6 3 4 3 5 S‘gmﬁ'
attraction & activity —'- —'—' "— ”“ —" —‘— cance
Museums Meant: F =12.360

3 9b 4.2 4.1 3.9 3.8 3.6 3.7 p<0.001
Cultural fairs/ Mean F=5.454
festivals/ events 3 8 4.0 3.9 3.9 3.8 3.6 3.7 p<0.001
Performing arts Mean F =8.459

3 9 4.2 3.9 4.1 3.8 3.7 3.9 p<0.001
Places to buy Mean F =9.3 85
local arts/crafts 3 6 3.8 3.7 3.7 3.7 3.3 3.6 p<0.001
Historical/ F= 10.863
heritage Mean p<0.001
attractions, sites 4_J 4.4 4.3 4.2 4.1 3.9 3.9
and districts
Gardens, zoos Mean F = 1.339
and aquariums 4 0 4.0 4.1 3.9 4.0 4.0 4.0 p=0.245
Agricultural Mean F = 0.085
attractions and 2 7 2.7 2.7 2.8 2.8 2.7 2.7 p=0.995
events —"
Purchasing F = 1.520
products grown Mean 3 3 3 2 3 3 3 3 3 1 3 1 p=0.180
or processed 32 ' ' ' ' ' '
locally
Architecture and Mean F =6.494
buildings 3 5 3.8 3.5 3.5 3.4 3.2 3.6 p<0.001
Customs and Mean F = 4.856
ways of living 3 7 3.9 3.7 3.8 3.6 3.5 3.7 p<0.001
Libraries, Mean F = 4.233
literary events 3.4 3.3 3.3 3.2 3.0 3.0 p<0.01

3.;
and bookstores

 

 

 

 

 

 

 

 

 

a Mean values measured on the basis of a 5-point Likert scale (1=not important at all, 3=neutral,
5= extremely important).
Mean score for museums in the ﬁve segments.
c Mean score for segment 1 in the overall cultural attractions and activities.

97

 

C1

as

Characteristics of Cultural Pleasure Trips

Statistically signiﬁcant differences among six segments were found for out-of-
country trips (p<0.001), and out-of-state trips (p<0.05) (Table 24). However, overnight
stays were not signiﬁcantly different among the six segments. Further analysis using
Tukey’s HSD post_hoc test indicated signiﬁcant mean differences: for out-of-country
trips between segment 1 and segments 4, 5 and 6; between segment 2 and segment 5; and
between segment 3 and segment 5. In addition, another Tukey’s HSD post_hoc test
showed signiﬁcant mean differences for out-of-state trips between segment 4 and
segment 6.

Regardless of the segment, a majority of respondents indicated that most or all of
their trips had involved overnight stays during the last 12 months. Respondents in
segment 6 took the most trips out-of-state while those in segment 4 took the least
pleasure trips to another state. A majority of the people in all segments indicated that
none or few had taken pleasure trips to another country. For segment 1 and 2 took more
trips out-of-country than other segments.

Summary of Behavior Proﬁles of the Propensity to Donate Six Sub-Segments
Based on the results of the behaviors regarding performing arts and pleasure trips,

a summary of the proﬁle of each segment is provided in Table 25.

98

Table 24. Characteristics of Cultural Pleasure Trips by the Propensity to Donate Six Sub-

 

 

 

 

 

 

 

 

 

 

 

 

 

Segments
Segment l 2 3 4 5 6 staltiessttic/
O 0 0 0 0
Proportion (16.4%) (16.2 A) (20.7 /o) (23.5 /o) (12.6 /o) (10.6 /o) Signiﬁ-
of pleasure trips(°/o) cance
Overnight Meana F = 0.575
stays Mb 4.4 4.3 4.4 4.4 4.4 4.3 =0_719
None/
few (%) 4.9 6.5 6.3 4.2 5.3 8.7
About half
(%) 8.8 11.4 7.1 8.8 9.3 5.5
Most/
all (%) 86.3 82.1 86.6 87.0 85.4 85.8
Out-Of- Meana F = 2.568
state trips .M 3.4 3.3 3.4 3.2 3.3 3.5 p<0.05
None/
few (%) 21.2 26.7 22.4 31.2 24.7 24.4
Abm“ half 31.5 28.2 30.6 27.7 30.0 18.9
(%)
Most/
all (%) 47.3 45.1 47.0 41.1 45.3 56.7
Out-of- Meana r = 7.856
99mm L6 1.8 1.6 1.6 1.5 1.3 1.5 p<0.001
mp5 “my 83 3 90 5 89 4 93 7 98 0 92 9
few (%) . . . . . .
About half
(%) 11.3 6.0 8.2 4.5 1.3 3.9
Most/
all (%) 5.4 3.5 2.4 1.8 0.7 3.2

 

 

 

 

 

 

 

 

 

 

a Mean values measured on the basis of a 5-point scale (1=none, 3=about half, 5=all).
b Mean score for overnight stays in the ﬁve segments.

99

 

Table 25. Behavior Proﬁle toward the Performing Arts and Pleasure Trips of the
Propensity to Donate Six Sub-Segments

 

Behavior proﬁle

 

Segmentl

3, 6‘ 3, 6"

jazz concerts

9, 66 ’9 ‘6

symphonies, rock

Mainly attended “musicals, plays,

concerts.”

Spent an average of $44.00 for purchasing tickets, with $86.00 as the
average highest price paid, and $775.00 as the average total spending on
tickets for the performing arts.

9’ 6‘

Mostly used “newspapers, websites of performing art organizations or
centers,” “acquaintances, friends, family or relatives,” and “websites
specialized for online ticketing” as the information sources.

Always or usually used the lntemet to research and purchase tickets for
performing arts.

9’ 66

“Historical/heritage attractions, sites and districts, performing arts,” and
“museums” were the most important attractions and activities in selecting
pleasure trip destinations.

Most or all of their pleasure trips involved overnight stays and more than
half were related to out-of-state trips.

The highest mean score of out-of-country trips, among the segments.

 

Segment 2

 

 

9, 6‘ 9’ 6‘ ,9 6"

Largely attended “musicals, symphonies, jazz concerts” and

“opera.”

plays,

Spent an average of $49.00 for purchasing tickets, with $93.00 as the
average highest price paid, and $775.00 as the average total spending on
tickets for the performing arts.

The highest average amounts for the highest price paid for tickets among all
of the segments.

,9 ‘6

Chieﬂy used “newspapers, websites of performing art organizations or
centers,” “websites specialized for online ticketing,” and “leaflets/
brochures” as their information sources.

Always or usually used the lntemet to research and purchase tickets for the
performing arts.

,9 6‘

“Historical/heritage attractions, sites and districts, gardens, zoos and
aquariums,” and “museums” were the most important attractions and
activities when they selected destinations for pleasure trips.

Most or all of their pleasure trips involved overnight stays, and more than
half were related to out—of-state trips.

 

100

 

Table 25. (cont’d)

 

Behavior proﬁle

 

Segment 3

’9 6‘ ’3 ‘6'

symphonies, jazz concerts” and

’9 ‘6

Mostly attended “musicals,
“folk/ethnic concerts.”

plays,

Spent an average of $5 1 .00 for purchasing tickets, with $88.00 as the
average highest price paid, and $791.00 as the average total spending on
tickets for the performing arts.

Spent the largest average total amount for tickets purchases, compared to
other segments.

99 ‘6

Mainly used “newspapers, websites of performing art organizations or
centers,” “websites specialized for online ticketing” and “acquaintances,
friends, family or relatives” as information sources.

Always or usually used the lntemet to research and purchase tickets for the
performing arts.

9’ 6‘

“Historical/heritage attractions, sites and districts, performing arts,” and
“gardens, zoos and aquariums” were the most important attractions and
activities when they selected pleasure trip destinations.

Most or all of their pleasure trips involved overnight stays, and more than
half were related to out-of-state trips.

 

Segment 4

 

 

,9 6‘ 9’ 6‘

Frequently attended live performances of “musicals, rock

concerts, “symphonies,” and “jazz concerts.”

plays,

Spent an average of $51.00 for purchasing tickets, with $88.00 as the
average highest price paid, and $752.00 as the average total spending on
tickets for the performing arts.

Chieﬂy used “websites of performing art organizations or centers,”
“newspapers,” “websites specialized for online ticketing,” “acquaintances,
friends, family or relatives,” and “radio” as their information sources.

Always or usually used the lntemet to research and purchase tickets for
performing arts.

,9 6‘

“Historical/heritage attractions, sites and districts, gardens, zoos, and
aquariums,” and “performing arts” were the most important attractions and
activities when they selected pleasure trip destinations.

The lowest mean score for out-of-state trips among the six segments.

 

101

 

Table 25. (cont’d)

 

Behavior proﬁle

 

Segment 5

9, “ 9’ “ ,9 ‘6 3,

Frequently attended “musicals, rock concerts,

and “symphonies.”

plays, country concerts

Spent an average of $49.00 for purchasing tickets, with $81.00 as the
average highest price paid, and $714.00 as the average total spending on
tickets for the performing arts.

Primarily used “websites of performing art organizations or centers,”
“websites specialized for online ticketing,” “newspapers” and
“acquaintances, friends, family or relatives” as information sources.

The highest lntemet users for researching and purchasing art- and culture-
related products.

“Gardens, zoos and aquariums,” “historical/heritage attractions, sites and
districts,” and “performing arts” were the most important attractions and
activities when they selected destinations for their pleasure trips.

Most or all of their pleasure trips involved overnight stays, and more than
half were related to out-of-state trips.

The lowest mean score for out-of-country trips among the segments.

 

Segment 6

 

 

,3 “

A higher percentage of attendance at “rock concerts, country concerts,”
“new and experimental concerts” and “hip hop concerts” than other
segments.

Spent an average of $43.00 for purchasing tickets, with $80.00 as the
average highest price paid, and $633.00 as the average total spending on
tickets for the performing arts.

The lowest spending on ticket purchases among all of the segments.

’9 “

Largely used “websites specialized for online ticketing, websites of
performing art organizations or centers,” “acquaintances, friends, family or
relatives” and “newspapers” as their information sources.

The highest ticket purchases for the performing arts through the Internet.

“Gardens, zoos and aquariums,” “historical/heritage attractions, sites and
districts,” and “performing arts” were the most important attractions and
activities when they selected pleasure trip destinations.

The highest mean score for out-of-state trips among the six segments.

 

102

 

Attandance at Live Peforrnances in a Different State or Country

The results of the third Exhaustive CHAID analysis performed to test the third
hypothesis are presented: Hypothesis 3: Heavy consumers of the performing arts are
further divided into substantial, measurable, accessible and actionable market segments
based on the inclination to travel out of state or/and out of country to attend live
performances.

This section shows another segmentation result for heavy consumers of the
performing arts. The decision tree resulting from the Exhaustive CHAID analysis grew to
four levels and respondents were segmented into ﬁve terminal homogeneous subgroups.
The most signiﬁcant predictors for attendance at live performances in a different state or
country were membership status in arts organizations, purchase of art products, education

and donation activity to arts organizations (Table 26).

Table 26. Signiﬁcant Predictors in the Decision Tree for Attendance of Live
Performances in a Different State or Country

 

 

 

 

 

Predictor Chi-square df Bonferroni
adjusted p-value
Membership status 18.76 1 0.0000***
Purchase of art products 7.77 1 0.0053“
Donation activity 6.78 1 0.0092"
Education 1 0.52 1 0.0248*
*, **, *** indicates signiﬁcance levels at 0.05, 0.01 and 0.0001, respectively.

The resulting decision tree is divided into two main branches of eight nodes.
Branch 1 consists of nodes 1, 3 and 4. Branch 2 consists of nodes 2, 5, 6, 7 and 8. The

nodes are presented by the number and percentage of cultural tourists and non-cultural

103

tourists, and the total respondents. The predictors are divides respondents into mutually
exclusive groups with the Chi-square value, and the Bonferroni adjusted p-value reported

in the CHAID tree diagram (Figure 5).

Attended live performances in a different state or country
Node 0

Yes (Y) 480, 38.2%
No (N) 775, 61.8%
Total (T) 1,255, 100.0%

 

 

 

 

Membership status
P-value=0.0000, Chi-square=l 8.7608
1

 

Yes No
Node 1 Node 2

v: 220, 53.3% 1 Y: 287, 34.1%
N: 193, 46.7% N: 555, 65.9%
T: 413, 32.9% r: 842, 67.1%

 

 

 

 

 

 

 

 

Education
P-value=0.0248, Clhi-square=10.5181

Purchase of art products
P-value=0.0053, (lihi-square=7.7717

 

 

 

 

 

SBachelor degree >Bachelor degree Yes No
Node 3 Node 4 Node 5 Node 6
Y: 61, 37.0% Y: 132, 53.2% Y: 208, 37.3% Y: 79, 27.7%

N: 104, 63.0%
T: 165. 13.1%

 

 

N: 116, 46.8%

 

T: 248. 19.8%

 

Segment 3

Segment 1

 

N: 349, 62.7%
T: 557. 44.4%

 

 

N: 206, 72.3%
T: 285. 22.7%

 

Donation activity

Segment 5

P-value=0.0092, Chi-square=6.78 15

l

 

 

 

Yes No
Node 7 Node 8
Y: 125, 42.4% Y: 83, 31.7%

N: 170, 57.6%
T: 295. 23.5%

 

 

N: 179, 68.3%
T: 262. 20.9%

 

 

Segment 2

Country

104

Segment 4

Figure 5. Decision Tree for Attendance at Live Performances in a Different State or

Decision Tree for Attendance at Live Performances in a Different State or Country

The root node at the top of the decision tree represents the sample of 1,255 heavy
performing arts consumers who indicated their attendance at live performances in a
different state or country during the last 12 months (Figure 5). These respondents were
divided into two categories -- cultural tourists (those who attended live performances in a
different state or country) and non-cultural tourists (those who did not attend live
performances in a different state or country). There were more non-cultural tourists (n =
775, 61.8%) than cultural tourists (n = 480, 38.2%) in the sample. This root node breaks
down into two child nodes, based on the membership status in arts organizations during
the last 12 months.

Based on the ﬁrst breakdown in Figure 5, the more people who had membership
in arts organizations, the more likely they were to be cultural tourists. When members
were further divided by education, people with a degree higher than a bachelor’s degree
were more likely to be cultural tourists. These people in node 4 were the most likely to be
cultural tourists in the decision tree, based on statistical distinction. The education
category was re-categorized into two groups — bachelor’s degree or less, and higher than
bachelor’s degree. On the other hand, respondents in node 6 were the least likely to be
cultural tourists; they were not members of arts organizations and did not purchase art
products during the last 12 months.

The estimated risk of misclassiﬁcation, based on the tree and the numbers of cases
corresponding to speciﬁc prediction errors, are presented in Table 27. Results of the
Exhaustive CHAID analysis revealed a 37.0% risk of false classiﬁcation; 348 of 1,007

respondents predicted to be non-cultural tourists were cultural tourists, and 116 of 248

105

respondents predicted to be cultural tourists were non-cultural tourists. The risk on cross-
validation was 40.6%. The risk summary indicates that the overall correct classiﬁcation
accuracy of 63.0% was a moderate improvement over the base rate of 61 .8%. Overall, the
predictors were better for non-cultural tourists (85.0% accuracy, with 659 of the 775

respondents) than for cultural tourists (27.5% accuracy, with 132 of the 480 respondents).

Table 27. Results of the Misclassiﬁcation Matrix

 

 

 

 

 

 

 

 

 

 

Actual category
Non-cultural Cultural Total
tourist tourist
Non-cultural tourist 659 348 1,007
Predicted .
category Cultural tourlst 1 16 132 248
Total 775 480 1,255
Risk statistics Cross-validation
Risk estimate 0.3697 0.4064
SE of risk estimate 0.0136 0.0139

 

 

 

Gain Summary of A ttendance at Live Performances in a Different State or Country
The gain summary (Table 28) is represented by ﬁve nodes, which were sorted by
an index score. The higher the index score, the more likely the respondents (heavy
consumers) in the node were to be cultural tourists for the performing arts. Nodes 4 and
7, which have an index score of greater than 100%, have higher response rates, while
nodes 3, 8 and 6, which have an index score of less than 100% for cultural tourists, have
signiﬁcantly lower response rates compared to the overall sample. The nodes with index
scores greater than 100% represent 53.5% of cultural tourists by targeting 43.3% of the
heavy consumer market. The gains chart shows important information about who needs

to be targeted as cultural tourists among the heavy consumers of the performing arts.

106

These nodes are effective segments of the cultural tourists for the performing arts that are
mutually exclusive, large enough for attention, measurable, accessible through marketing

communication tools, and share certain characteristics (Kotler, 1972).

Table 28. Gains Summary of the Attendance at Live Performances in a Different State or

 

     
       

Country
Node No. of % of all the No. of % 0f % of cultural Index scoreb
respondents respondents cultural cultural tourist (%)
(n) (%) tourists tourist response rate

 
   
   
 

(11) sample11 In the node
(‘79) (%)

 
   

 

 
 
 
   

 
  
 

3 165 13.1 61 12.7 37.0 96.9
8 262 20.9 83 17.3 31.7 83.0
6 285 22.7 79 16.5 27.7 72.5

  
   

 

a Cultural tourists in a particular node to all of the cultural tourists in the heavy market.

b

C

Proportion of cultural tourist rates in the node compared to the overall sample.
139.3% = 53.2% (cultural tourist rate in Node 6) / 38.2% (cultural tourist rate in the overall sample).

Profile of the Inclination to Travel Out of State or/and Out of Country to Attend
Live Performances Sub-Segments

Heavy performing arts consumers were divided into ﬁve sub-segments of the
inclination to travel out of state or/and out of country to attend live performances and
proﬁled based on socioeconomic and household characteristics, and other behaviors
toward the performing arts and pleasure trips, with respect to: (1) the type of performing
arts attendance; (2) spending on tickets to the performing arts; (3) performing arts
information sources; (4) web-based purchases to art- and culture-related products; (5) the
importance of cultural attractions and activities in selecting pleasure trip destinations; and

(6) characteristics of pleasure trips. A series of chi-square and one-way ANOVA analyses

107

were employed to determine statistically signiﬁcant differences among the ﬁve sub-
segments.
Socioeconomic Characteristics

A chi-square test was performed to determine whether statistically signiﬁcant
differences existed among the ﬁve segments in the respondents’ characteristics,
including: (1) gender; (2) age; (3) sexual orientation; (4) race; (5) education; (6) annual
household income; (7) employment; (8) number of people in household; and (9)
household composition. Statistically signiﬁcant differences among segments were found
in gender, age, education, annual household income, number of people in household and
household composition (p < 0.001), and in employment (p < 0.01) (Table 29).

Regardless of the segment, a majority of respondents were female, aged 45 years
or older, straight or heterosexual and Caucasian or White, had an annual household
income of $50,000 or higher, had two people or more people residing in their household,
and were married or partnered, either with children living at home or children no longer
living at home. A summary of the socioeconomic characteristics of the inclination to
travel out of state and/or out of country to attend live performances ﬁve sub-segments is

presented in Table 30.

108

 

$38 .896 Sodva 88565

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5.2 u «x Sam. 1 «x 83m 1 4x 22333

m.m “no ad ”3 b6 m._N 9mm QNN Q: Q; ﬁwm WNN mwN m

in N._ ad 93 mg 54mm hem <2 0. _N New ”.mm adm mom v

mg. 96 N._ Eco wd— NMN wen v.2 ad adw _d— _.m_ mg M

No C _.m hdo NS _.Nm «.mm ﬁb— ws m.w© mAM WMN mom N

m6 v6 N._ _dw o6. Wow mdm Nd— w.N Name .613... mm. wvm _
m a m m
m m a .e
m m m .6. m m
.0. Nu. W .0. n m. n
n 6 W M e m m.
r m L .anu r 4 4 t m f w
a...» e W; .1 0 M 5 4 S m P» O r.
e .hr. 3 m 5 ﬁx (.5 5 ace. 6 a e f.

mu 0 G S 6 5 4 3 L F. M We m t

n M w

w m

8 A5 3 a m a.

ceaseﬁre 338m ow< #650 P N S

mucoﬁwombsm 03m

moocmnnotom 2,5 @534 3 3560 mo 30 8:28 88m .«o 50 .35; 8 539505 05 mo 33338850 ogocoooomoom .am 033.

109

Table 29. (cont’d)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

._. "I ‘0. “. V2 V?
:: Others 3 co m V b :
O ('3 V1 M ('1 ON '—
5"? 00' q
o o\ "‘ o
‘0‘." ll ‘5.
E . ". st. "2 0. V2
U-l Employed full-ume tn — so m N “R
m o v o so
. V. '1 V. N. ":
$150,000 or higher o o m h c:
N M N —- v—
0
E
8
o o o o m
E $100,000-$149,999 re; 00 :5 o m 2
a s s °-
=3 V n 9
.8 V. <2 0: M. O. °-
7.: $50,000 - $99,999 o N so 0‘ no “N
5 M M M M V
o m "I ‘9. ‘0.
Under $50,000 M- a; e S 2
“l ‘0. ca l‘
Doctoral degree a : O a; ‘5
Completed some graduate courses 2 0! O :3 °‘. N
'5' or Master’s degree F v m m ‘°. _
.. o
H A If) 8
8 o\° v d
a V
1: II V
m v, sr. h. -: a ‘3‘
4-year college degree O tn m o co x
N \O N N
“2 ‘9. ‘9 "I
2-year college degree or less 9 —- o c m
N M M M
V} — \O 00 M
Others <13 hi 6 «5' hi 3 M
o l‘. V!
0 Q o v-:
N o
a: V II E
' ' V! O‘- q: h! 6! pk
Caucasnan or Whnte m l\ ox so 5
ON O\ O\ 0‘ OS
°°. V‘. "i 0‘. l‘.
Percentage of respondents (%) ax to m o N
.— N .— N N
00 V1 V) N ‘0
ON 00
Number of respondents (n) g; N :2 g N g 3
.5 5
S 0
m E
Segment '- N M <- W 1""; E)
l- U)

 

 

 

 

 

 

 

110

 

Table 29. (cont’d)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Married/Partnered with children no I‘. V. O. V‘! '1
I I w a w v o
longer hung at home M N N N N
Married/Partnered with children 0. 0‘. G. N. O.
. . l\ 00 \O M '—
livmg at home N M M M v
I:
.2
H
'7: . . . M. V. V. '1 0*.
8. Married/Partnered Without children 2 m 2 2: 2 g;
g I; m' 8
2 v I I I II V
% Single With children no longer q v, :5 <3: <1; ”x °'
3 living at home ‘° " ‘° M V
o
32
o u . . . o w— b (N m
Single With children livmg at home v‘ v‘ c.; .n- o-
. . . "2 co “l "I "2
Single Without children o a; m 3 xx
‘0. "i V. V? V?
.0 4 persons or more xo v o so m
.—. '— N M N M
O
.2
0
(I)
5
.8 —~. 6. «a °°. h.
s 3Persons : 2 :2 z z a
a A 53- §
0 °\° m .
12 V II 9
a h. —: °°. or N 0. °-
“5 2 persons <r i~ Ox n o x
t— V) V M V M
E
:3
Z . ‘0. °9 9. °°. ‘9
1 live alone v o m <2- m
°9 W. "t O‘. '5
Percentage of respondents (%) as m m o m
'— N — N N
00 V1 V3 N W
ON \0 \O 00
Number of respondents (n) g N __ N N g 3
g 5
w .‘E
Segment —- N M v M 373 Q
[- U)

 

 

 

 

 

 

 

111

 

.3889 88:3 .8 owﬂcoeom x 358883.. .8 “88:88; a

 

Afavwv 0:8: 8 w£>= 898.
0: 5.8.20 525 8 0:8: 8 w£>= 58:8 5:: 8205838882 -
.38.th 22828: 05 E magma» 088 8 3888 03,—. -

m8. 8 8.8% owe—.8 89¢». -

 

 

 

 

 

 

 

.880me 83 min - 95 ca .8 0885 22828..— 828< - 828583“. 38 E aim—3:82 - Nu— hN— _.m_ m
éxéc: man. 8 3&8 owe—80 893‘ - . . .
8x883 820 8 80» mv -
3.8de 08th— -
.888de 0:8..— 8 w£>= cow—8—
c: 88.20 5? 8 0:8: 8 w£>= 58:8 .53 Bunyan—82.82 -
.315; 20:88; 05 5 wEEmB 888g 03... - 828.8838 38 9 8288C -
.888.me 8&2 8 8985.8 0885 22.8.8: 8::=< - 30:88 .8 “8.888; - _8 oém m.m~ N
.38.an .3: 8 88% 08:8 893. - 8285588 3.8 5 9:83.58 oz -
88¢.me 82o 8 8% 3.2“ -
$.38 ease“. -
.3868 0:8: 8 w£>= 882
o: 58:8 .23 8 2.8: 8 w£>= 3.8.20 5:: 385888082 -
. .31..va 10:88: E 9:28» 8888 95... - 8&8 owe—.8 80.3» .85 8.3: .. . .
Ago c8 8&2 8 coo 85.8 282: 22.88: 8:52 - £85388» 8 38 E a. w. :8 - Wm m nm a o. _
.358: 3&8 8:»:— 8 388 08:58» 088 38.an0 - . . . h on 2
£33 5% a as» a. -
31.68 08:8“— ..
«38:8
88:3
.8
08% 3.0.89 3:8
8:88 .9523 -882
83.0.888an @888er .8 .825qu .8 X. .8 .x. .8 ..\o Eoﬁwom

 

 

 

 

maeoﬁwombsm 03m moonmacotom 2’5

.883. 8 .3580 .8 80 8:28 88m .8 SO 88d. 9 88585 05 .8 8:338:30 omaocoooomoom 05 .8 85.56 .8 28,—.

112

.883 .8380 8 08888: x 8:82:82 .8 08888..

 

 

 

 

£6. E 2:2. a 85>:
58...."v 5:3 808888088 8:8 3oz: 58.2”v :88; o.w:.m -
. c.8888 21.38.. 2.” 5 wwmwuohwEMooHMeasﬁh - 8:35 5 83:22. oz - . 3. Z w
Cam m5 3: 3.5 - coo cm: . 8. a _. 8 < 828888: 8.8 E 988882: :2 - h m N
8x8. .8 m8. 8 8&8 &o..8 .893. -
.283. 3% s 8% 8.2 -
33.3 8.5“. -
.3620 2:2. a 83.
5.2.6 5.3 835380.55 Ea 31.5 56:8 :55? 8:5 -
.38.va 8.2.082. 2: :. 887.8 88.8: :3... - 80.8883: 8.8 2 828:8 -
8x868 38.3.5 - 8on .8 088:: 8.2.3.8.. 8:::< - $8.88 8 80822:: - 9m ms. 8.2.. v
4x868 gm... 8 8&8 09:8 893‘ - 80.8888: 38 E £58988: :2 -
$8.5 2o 5% mm 55 m8: -
Afamdt 82:0“. -
«38:3
88:8
.8
«8% 8:88 3:8
“8.8:: .8380 82.8..
8.8888850 £88898 .8 .55...ch .8 .x. .8 ..\o .8 .x. :5:&&

 

 

 

 

 

 

 

 

6.258 .8 22¢

113

Type of Performing Arts Attendance

Cross-tabulations were performed to provide the type of performing arts
attendance during the last 12 months among the ﬁve sub-segments, with regards to: (1)
plays; (2) musicals; (3) opera; (4) ballet; (5) modern dance; (6) ethnic dance; (7) folk
dance; (8) folk/ethnic concerts; (9) jazz concerts; (10) blues concerts; (11) symphony;
(12) country concerts; (13) new and experimental concerts; (14) rock concerts; (15) hip-
hop concerts; (16) world concerts; and (17) other. Chi-square tests indicated that
signiﬁcant differences among segments existed for: plays, opera, ballet, folk/ethnic
concerts, jazz concerts and symphony (p<0.001); modern dance and rock concerts
(p<0.01); and blues concerts and world concerts (p<0.05). However, differences were not
signiﬁcant for musicals, ethnic dance, folk dance, country concerts, new and
experimental concerts, hip-hop concerts, and other (Table 31).

For respondents in all ﬁve segments, attending musicals and plays showed the
highest frequency among the live performances, while hip-hop concerts, folk dance, new
and experimental concerts, and world concerts were the least attended. Approximately
90% of all respondents in the ﬁve segments attended musicals, and more than two-thirds
had attended plays. Attendance of respondents in segment 1 was the highest for most live
performances, while attendance of those in segments 4 and 5 was lower than that of other

segments.

114

 

 

 

 

 

 

 

 

 

 

 

 

 

 

22an Save _oodva Sodva ~3ch Siva 8863a
~26 u «x 3...: u ”x 5.2 u «a 23m u «x 3.... u «x 323 n ”x oﬁesﬁmﬁ
2 is 2 q; 2 «a 2 as A: 3. 3% :2 EN 2m m
3. v.8 G a? 3.. :3 E v.8 0.8 v.2 93 3% a2 man u
3 nos 3. .8 SN 3: «.2 98 n: n: QR 9mm 1: a: m
3. N9 3. a: 2. Z” 3. E» 25 we 92 3m 08 am N
2 3o 3: 2m 2m «.3 EN mg :a a.» a: 3: ”.2 «a _
m...
..... m
d d d d d d m B
.n. m m m m m .m .n.
n H n n ﬂ ﬂ 0 d
d a d a d a d a d a d a m. n
.m m a m r...“ m w m w m w m m m.
n n n n n n n n n n n n f a
m .d m m m m m m m u m .d o r
.. .. 1 e f
A D A D A D A D A D A D g o
m r m
m m e
g g g g g 3 m m m.
85% 035m 85% E252 B=~m £30 .8632 ~85 P N S

 

 

 

 

 

 

 

 

 

 

 

“Eon/x 9 3550 mo 50 SEE“ 38m mo 50 .26; 3 gag—=05 05 3 maggot/w mg mummy—8.5m mo 09C. Am 2an

mwcoEwumbsm ASE moogncotom 2,5

115

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sack Siva 8.9a Siva 83.x. 8».ng 858$:me
Ed u «x 3.? u «x 39: n M Sean n ”x 892 u x ”2.. n M 055%“;
«:2 63 w: 0% ad N8 3.. o8 3 3a w.m NE SN 2n n
3. a» S: m; n: 03 v.2 can .3 :5 mm Ea mom SN v
am. 3% 3m 30 9: an” Ex <8 98 v.3 3H «.3 <2 9: m
«:2 2% 2m 23 a. _3 v.8 can 3: v.5 «a 98 02 RN N
T: we» 93 or} «.2 New .3 Q8 3m 3» 3 98 ”.2 wvm _
m
m m
d d d d d d m m
n n n n n n d n
m m m m m. m m m
d a d a d a d a d a d a :0... n
a m m m a m m m. w m m m m w
.n. n m n m n m n m n m n f a
n m u m .t.. ..m u ..m u m u m o h
A D A D A D A D A D A D me o t
m a n
m b e
3 g 3 g g g m m m.
too—.8 .9560 mace—aim €3.50 35 50:8 N53. 5280 25:023..— 358 ion P N am

 

 

 

 

 

 

 

 

 

 

23:88 . _ m 293‘

116

 

 

 

 

 

 

 

 

 

 

 

 

 

 

9:.ch 8.9g among 8.3g ".8.ng 8:85am;
:3 u «x a? u «x «as u «x 3.: u Ax 23 u «x 239389
In «.9 2 QR _.N a: 3m a? 3. 35 EN 2a m
EN 3 ”A 0.3 mm Ea 3% c8 Tm 9: QB SN e
Z... A? S a? ”A 0.3 S». as 3 N3 1: m2 m
No». ”do 3, 98 3 33 .3 q: 3. 93 3N m8 N
am :3 3 2a 3 33 «MN 92 3 $5 ”.2 ”a _
m
d d d d d B \m/
n n n n n n (
e e e e e e .6
.t.. n n u n d n
d a d a d a d a d a n e
w m a m w m m m w m w m
n n n n n n n n n n S O
m m m m m w m w m m m w
A D A D A D A D A D .m m
e f
93 mg m m
50:8 m .m .m.
3 3 3 A 3 55.5.3.3 m m m...
350 toocoo 283 rouge no: am: .5250 :03. can 302 P N S

 

 

 

 

 

 

 

 

 

528$ .N m £an

117

Spending on Tickets to the Performing Arts

One-way ANOVA revealed statistically signiﬁcant differences among the ﬁve

sub-segments in the total spending on tickets (p<0.01) and in an average ticket price and

the highest ticket price (p<0.05) (Table 32). Further analysis using Tukey’s HSD

post_hoc test indicated signiﬁcant mean differences: (1) in the total spending on tickets

between segment 1 and segment 5; and between segment 2 and segment 5; (2) in the

average ticket price between segment 2 and segment 3; (3) in the highest ticket price for

tickets purchased between segment 2 and segment 5. The average ticket price for tickets

and the highest price were the greatest for persons in segment 2 while the total spending

on tickets was the greatest for those in segment 1. On the other hand, the average ticket

price was the lowest in segment 3 while the average highest ticket price and the average

total spending on tickets had the lowest values in segment 5.

Table 32. Spending on Tickets by the Inclination to Travel Out of State and/or Out of
Country to Attend Live Performances Five Sub-Segments

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Segment
Ticket 1 2 3 4 5 Test statistic/
Price (19.8%) (23 5%) (13.1%) (20.9%) (22.7%) Signiﬁcance
paid
Average price Meana (s) F _._ 2 462
for tickets 48.00b 47.00 50.00 46.00 49.00 48.00 p<0.05
Range 10.00- 8.00- 8.00- 10.00- 5.00-
(S) 100.00 125.00 1 l 1.00 100.00 100.00
Highest price Mean ($) F = 2.461
for a ticket 87.00 91.00 92.00 87.00 83.00 81 .00 p<0.05
Range 20.00- 25.00- 26.00- 10.00- 10.00-
($) 500.00 750.00 350.00 300.00 1000.00
Total spending Mean (3) F = 3.994
on tickets 752.00 792.00 772.00 749.00 733.00 712.00 p<0.01
Range 150.00- 120.00- 90.00- 1 10.00- 100.00-
($) 1000.00 1000.00 1000.00 1000.00 1000.00

 

a Mean values rounded to the nearest dollar.

Mean score for average price for tickets in the ﬁve segments.

118

 

Performing Arts Information Sources

Chi-square tests were used to compare the ﬁve sub-segments as to information
sources used to search for live theater, dance performances and music concerts (Table 33).
Differences among the segments were signiﬁcant for: art-related publications and
leaﬂets/brochures (p<0.001); newspapers and websites specialized for online ticketing
(p<0.01); and websites of performing art organizations or centers (p<0.05). TV,
magazines, radio and acquaintances, friends, family or relatives, however, no statistically
signiﬁcant differences were found for.

A majority of the respondents in all of the segments either usually or always used
newspapers, websites specializing in online ticketing, websites of performing art
organizations or centers, acquaintances, friends, family or relatives and magazines as
information sources for live performances, but never or seldom acquired information
about live performances from art-related publications and magazines.

More than 80% of the respondents in segments 1 and 2 either usually or always
obtained information related to live performances from newspapers, whereas
approximately 70% of the respondents in segments 4 and 5 either usually or always
obtained information from websites specializing in online ticketing. Higher percentages
of people in segments 2 and 4 either usually or always used websites of performing art
organizations or centers, while more people in segments 1 and 3 obtained the information
from art-related publications and leaﬂets/brochures, leaﬂets or brochures, than did those

in other segments.

119

Table 33. Performing Arts Information Sources by the Inclination Travel Out of State
and/or Out of Country to Attend Live Performances Five Sub-Segments

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Segment 1 2 3 4 5 Test statistic/
Information (19.8%) (23.5%) (13.1%) (20.9%) (22.7%) Si niﬁcance
Source % % % % % g
Never/
seldom 67.9 61.7 61.2 55.2 61.9 X2 = 8.620
TV U 11 /
5"“ V 32.1 38.3 38.8 44.8 38.1 p=0.07l
always
Never/
. seldom 72.6 72.4 72.3 71.7 77.1 X2 = 2.618
Magazmes U 11 /
5‘” y 27.4 27.6 27.7 28.3 22.9 p=0.624
always
Never/
seldom 17.8 18.4 25.0 27.5 29.6 x2 = 17.134
Newspapers U 11 /
5‘” y 82.4 81.6 75.0 72.5 70.4 p<0.01
always
Never/
' seldom 52.7 45.1 48.8 43.8 48.9 X2 = 4.969
Rad1o U 11 /
“'3 V 47.3 54.9 51.2 56.2 51.1 p=0.290
always
Never/
Websites. . seldom 42.6 30.3 39.4 28.6 28.4 X2 = 18.803
spec1ahzmg 1n
online ticketing usually’ 57.4 69.7 60.6 71.4 71.6 p<0.01
always
Websites of New” 28 5 16 3 23 5 21 5 25 3 2 ..
performingart seldom ' ' . ' ' X _ 12'468
organizations or Usually/
centers always 71.5 83.7 76.5 78.5 74.7 p<0.05
Never/
fnends, family or U 11 /
relatives “’3 y 64.7 68.3 65.6 63.7 57.1 p=0.080
always
Never/
62.1 78.9 64.1 85.3 88.9 2 =
Art-related seldom X 78' ‘93
publ'camns ”sual'y/ 37.9 21.1 35.9 14.7 11.1 p<0.001
always
Never/
Leaﬂets/ seldom 44.0 59.7 41.7 67.2 68.2 X2 = 57.72
mm“ usually 56.0 40.3 58.3 32.8 31.8 p<0.001
always

 

 

120

 

Web-Based Purchases to Art- and Culture-Related Products

Chi-square tests were also used to compare: (1) lntemet use; (2) on-line research;
and (3) the purchase of art- and culture-related products (Table 34). Signiﬁcant
differences were found among segments for researching of the arts (p<0.001);
researching antiques and collections, and the purchase of arts (p<0.01); and researching
tickets for the performing arts, and the purchase of tickets for the performing arts
(p<0.05).

Regardless of the segment, almost all of the respondents used the lntemet to
research or purchase art- and culture-related products. A majority in all ﬁve segments
usually or always used the lntemet to research and purchase tickets for the performing
arts. Respondents in segments 2, 4 and 5 had higher percentages of usually or always
researching and purchasing tickets for performing arts than did other segments. On the
other hand, a majority in the segments seldom or never used the lntemet to research and
purchase for arts, antiques and collections. Respondents in segments 2 and 4 had higher
percentages of usually or always researching and purchasing arts.

The Importance of Cultural Attractions and Activities in Selecting Pleasure Trip
Destinations

Signiﬁcant differences among the sub-segments were found in the importance
ratings of: museums; cultural fairs/festivals/events; performing arts; places to buy local
arts/crafts; historical/heritage attractions; architecture and buildings; and libraries, literary
events and bookstores (p<0.001); customs and ways of living (p<0.01); and purchasing
products grown or processed locally (p<0.05). There were no signiﬁcant differences
among the ﬁve segments, however, for: gardens, zoos and aquariums; and agricultural

attractions and events (Table 35).

121

Table 34. Web-Based Purchases by the Inclination to Travel out of State and/or Out of
Country to Attend Live Performances Five Sub-Segments

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Segment 1 2 3 4 5 Test statistic/
lntemet (19.8%) (23.5%) (13.1%) (20.9%) (22.7%) S' .f
Use % % % % % 'g'" ”an“
Use 92.7 95.6 95.1 97.7 95.4 X2 = 7.194
lntemet use N
0‘ 7.3 4.4 4.9 2.3 4.6 p=0-126
use
New” 69 3 65 5 72 4 68 8 83 0 2
seldom - . - 2 - x = 24.053
Research arts U 11 /
W V 30.7 34.5 27.6 31.2 17.0 p<0.001
always
, Never/
Research t1ckets seldom 15.7 8.2 14.7 8.2 10.7 X2 = 11.709
to the performing U 11 /
arts 5‘” y 84.3 91.8 85.3 91.8 89.3 p<0-05
always
Never/
Research seldom 74.3 73.3 70.4 72.9 84.7 7(2 = 16.255
antiques and U 11 /
collections 5‘“ y 25.7 26.7 29.6 27.1 15.3 p<0.01
always
New" 90 7 89 0 88 8 85 9 95 5 2
seldom ' ' ' ' - x = 14.798
Purchase arts U 11 /
5”“ y 9.3 11.0 11.2 14.1 4.5 p<0.01
always
, Never/
Purchase tickets seldom 21.7 18.1 25.0 13.3 16.3 x2 = 11.558
to the performing U 11 /
arts “3 V 78.3 81.9 75.0 86.7 83.7 p<0.05
always
Never/
Purchase seldom 88.5 90.4 90.1 89.0 93.3 X2 = 4.155
antiques and
collections US““"” 1 1.5 9.6 9.9 1 1.0 6.7 p=0.3 85
always

 

 

 

 

 

 

 

 

122

 

Results of the Tukey’s HSD posthoc test indicated signiﬁcant mean differences
for attraction to: (1) museums between segment 1 and segments 2, 4 and 5; between
segment 2 and segments 4 and 5; and between segment 3 and segments 4 and 5; (2)
cultural fairs/festivals/events between segment 1 and segment 5; and between segment 3
and segment 5; (3) performing arts between segment 1 and segments 4 and 5; between
segment 2 and segments 4 and 5; and between segment 3 and segment 4; (4) places to
buy local arts/crafts between segment 1 and segment 5; between segment 2 and segment
5; between segment 3 and segment 5; and between segment 4 and segment 5; (5)
historical/heritage attractions, sites and districts between segment 1 and segments 2, 4
and 5; between segment 2 and segment 5; between segment 3 and segment 5; and
between segment 4 and segment 5; (6) architecture and buildings between segment 1 and
segment 5; between segment 2 and segment 5; between segment 3 and segment 5; and
between segment 4 and segment 5; (7) libraries, literary events and bookstores between
segment 1 and segments 4 and 5; between segment 2 and segments 4 and 5; and between
segment 3 and segments 4 and 5; (8) purchasing products grown or processed locally
between segment 2 and segment 5.

“Historical/heritage attractions, sites and districts,” “gardens, zoos and

9, 66

aquariums, performing arts,” and “museums” were the most important attractions and

activities among the ﬁve segments when respondents were selecting destinations for

97 ‘6

pleasure trips. On the other hand, “agricultural attractions and events, purchasing
products grown or processed locally,” and “libraries, literary events and bookstores” were

the least important attractions and activities in all ﬁve segments. People in Segment 1

123

indicated almost all of the cultural attractions and activities were more important for

pleasure trip destinations than that of other segments.

Table 35. The Importance of Cultural Attractions and Activities in Selecting Pleasure
Trip Destinations by the Inclination to Travel Out of State and/or Out of
Country to Attend Live Performances Five Sub-Segments

 

 

 

 

 

 

 

 

 

 

 

 

 

Segment 1 2 3 4 5 . .
Cultural (19.8%) (23.5%) (13. 1%) (20.9%) (22.7%) E?“ igt‘St'C/
attraction & activity 3_.§° L7. 3+7 12.-é 2:1 lgm 1cance
Museums 8 F = 17.801
“33:3 4.2 3.9 4.1 3.7 3.6 p<0.001
Cultural fairs/ Mean F = 5.489
festivals/ events 3 8 3.9 3.8 4.0 3.8 3.7 p<0.001
Performing arts Mean F = 7.383
4 0 4.1 4.1 4.0 3.8 3.8 p<0.001
Places to buy local Mean F = 12.279
arts/crafts 3 6 3.8 3.7 3.7 3.7 3.3 p<0.001
Historical/ heritage Mean F = 15.1 12
attractions, sites & 4 2 4.4 4.2 4.3 4.1 3.9 p<0.001
districts é
Gardens, zoos & Mean F = 1.129
aquariums 4 0 4.0 4.0 4.1 4.0 3.9 p=0.341
Agricultural Mean F = 1.698
attractions and 2 7 2.6 2.8 2.9 2.7 2.7 p=0.l48
events "—
Purchasing Mean F = 3.093
products grown or 3 3 3.3 3.4 3.3 3.3 3.1 p<0.05
processed locally é
Architecture and Mean F = 9.006
buildings 3 5 3.7 3.6 3.6 3.5 3.2 p<0.001
Customs and ways Mean F = 4.189
of living 3 7 3.9 3.7 3.7 3.6 3.5 p<0.01
Libraries, literary Mean F = 7.259
events and 3 2 3.4 3.4 3.4 3.0 3.1 p<0.001
bookstores é

 

 

 

 

 

 

 

 

a Mean values measured on the basis of a 5-point Likert scale (1=not important at all, 3=neutral,
5= extremely important).
Mean score for museums in the ﬁve segments.
c Mean score for segment 1 in the overall cultural attractions and activities.

124

 

Characteristics of Cultural Pleasure Trips

Statistically signiﬁcant differences among segments were found for out-of-
country trips at p<0.001 (Table 36). Overnight stays and out-of-state trips were not
signiﬁcantly different among the ﬁve segments. Further analysis using the Tukey’s HSD
post_hoc test indicated signiﬁcant mean differences: for out-of-country trips between
segment 1 and segments 3, 4 and 5; and between segment 2 and segment 5.

Regardless of the segment, a majority of respondents indicated that most or all of
their trips had involved overnight stays, and most or all of pleasure trips of approximately
half of respondents had involved out-of-state trips. A majority of the people in all ﬁve
segments indicated that none or few had taken pleasure trips to another country. For
segments 1 and 2 took more trips out-of-country than other segments.

Summary of Behavior Proﬁles of the Inclination to Travel Out of State or/and Out
of County to Attend Five Sub-Segments

Based on the results of the behaviors toward the performing arts and pleasure

trips, a summary of the behavior proﬁle of each segment is presented in Table 37 .

125

Table 36. Characteristics of Cultural Pleasure Trips by the Inclination to Travel Out of

State and/or Out of Country to Attend Live Performances Five Sub-Segments

 

Segment

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Proportion l 2 3 4 5 Test statistic/
0 0 o o 0 . .
of pleasure trips (19.8 /o) (23.5 /o) (13.1 /o) (20.9 /o) (22.7 /0) Signiﬁcance
Overnight Meana
stays 4.41, 4.3 4.4 4.3 4.4 4.3 F = 0.482
None/ few p=0.749
(6%) 4.5 3.8 7.4 4.8 8.4
About half
(%) 12.0 9.8 7.4 5.6 7.7
Most/ all
(%) 83.5 86.4 85.2 89.6 83.9
Out-of-state Mean
trips 3; 3.4 3.4 3.2 3.3 3.3 F ___ 1.494
NW” few 20 1 24 o 29 2 27 7 26 8 —
(%) ' ‘ ' ‘ - p—0.202
About half 30.5 26.8 28.6 28.9 27.2
(%)
Most/ all
(%) 49.0 49.2 42.2 43.4 46.0
Out-of-country Mean
trips Lé 1.8 1.7 1.5 1.5 1.4 F = 12950
NW” few 83.5 89.5 92.0 93.6 96.0 p<0-001
(%)
About half
(%) 10.3 8.1 6.1 4.4 2.5
Most/ all
(%) 6.2 2.4 1.9 2.0 1.5

 

a Mean values measured on the basis of a S-point likert scale (1=none, 3=about half, 5=all).

b Mean score for overnight stays in the ﬁve segments.

126

 

Table 37. Behavior Proﬁle toward the Performing Arts and Pleasure Trips of the
Inclination to Travel Out of State and/or Out of Country to Attend Live
Performances Five Sub-Segments

 

Behavior proﬁle

 

Segment 1

a, 66'

symphonies, Jazz concerts,”

9, ‘6 99 ‘6

Mainly attended “musicals,
“opera” and “rock concerts.”

plays,

Spent an average of $47.00 for purchasing tickets, with $91.00 as the
average highest price paid, and $792.00 as the average total spending on
tickets for the live performances.

The largest average total spending on tickets, compared to other segments.

,9 6‘

Mostly used “newspapers, websites of performing art organizations or
centers,” “acquaintances, friends, family or relatives,” and “websites
specialized for online ticketing” as the information sources.

Always or usually used the lntemet to research and purchase tickets for
performing arts but relatively less intemet users than other segments.

,9 £6

“Historical/heritage attractions, sites and districts, museums,” and
“performing arts” were the most important attractions and activities in
selecting pleasure trip destinations.

The highest mean scores for out-of-state trips and out-of-country trips,
among the ﬁve segments.

 

Segment 2

 

 

,9 66'

” “ ” “ Jazz concerts,

9, 6‘

Largely attended “musicals, plays, rock

concerts” and “country concerts.”

symphonies,

Spent an average of $50.00 for purchasing tickets, with $92.00 as the
average highest price paid, and $772.00 as the average total spending on
tickets for the performing arts.

The highest average amounts for the highest price paid for tickets among all
of the segments.

Chieﬂy used “websites of performing art organizations or centers,”
“newspapers,” “websites specialized for online ticketing,” and
“acquaintances, friends, family or relatives,” as information sources.

,9 6‘

“Historical/heritage attractions, sites and districts, performing arts,” and
“gardens, zoos and aquariums” were the most important attractions and
activities when selecting destinations for pleasure trips.

The highest mean score for overnight stays, among the ﬁve segments.

 

127

 

Table 37. (cont’d)

 

Behavior proﬁle

 

Segment 3 o

,9 6‘ ,9 (6 ’9 ‘6

Mostly attended “musicals, rock concerts, symphonies,” and

“jazz concerts.”

plays,

,9 66

The highest attendance of “ballet,” “ethnic dance,
concerts,” and “hip hop concerts.”

new and experimental

Spent an average of $46.00 for purchasing tickets, with $87.00 as the
average highest price paid, and $749.00 the average total spending on tickets
purchases for the performing arts.

The lowest amount for the average ticket purchases.

Mostly used “websites of performing art organizations or centers,”
“newspapers,” “websites specialized for online ticketing,” and
“acquaintances, friends, family or relatives” as information sources.

Always or usually used the lntemet to research and purchase tickets for the
performing arts.

9, ‘6

“Historical/heritage attractions, sites and districts, gardens, zoos and
aquariums,” “museums” and “perfonning arts,” were the most important
attractions and activities for pleasure trip destinations.

The lowest mean scores for overnight stays and out-of-state trips among the
ﬁve segments.

 

Segment 4 o

 

 

9, 6‘ 99 6‘

Frequently attended live performances of “musicals, rock

concerts, “symphonies,” and “jazz concerts.”

plays,

Spent an average of $49.00 for purchasing tickets, with $83.00 as the
average highest price paid, and $733.00 as the average total spending on
tickets for the live performances.

Chieﬂy used “websites of performing art organizations or centers,”
“newspapers,” “websites specialized for online ticketing, acquaintances,
friends, family or relatives,” and “radio” as information sources.

,9 ‘6

The highest lntemet users, and always or usually used the lntemet to
research and purchase tickets for the performing arts.

9, ‘6

“Historical/heritage attractions, sites and districts, gardens, zoos, and
aquariums,” and “performing arts” were the most important attractions and
activities in selecting pleasure trip destinations.

Most or all of pleasure trips involved overnight stays, and more than half
were related to out-of-state trips.

 

128

 

 

Table 37. (cont’d)

 

Behavior proﬁle

 

,9 6‘

Segment 5 0 Frequently attended “musicals,” “plays,” “rock concerts,
“jazz concerts.”

symphonies” and

o The lowest attendance of performing arts, among the ﬁve segments.

0 Spent an average of $48.00 for purchasing tickets, with $81.00 as the
average highest price paid, and $712.00 as the average total spending on
tickets for the live performances.

o The least average highest price and the least average total amount for tickets
purchases, compared with other segments.

0 Primarily used “websites of performing art organizations or centers,”
“websites specialized for online ticketing,” “newspapers” and
“acquaintances, friends, family or relatives” as information sources.

 

0 Always or usually used the lntemet to research and purchase tickets for the
performing arts.

0 “Gardens, zoos and aquariums,” “historical/heritage attractions, sites and
districts,” and “performing arts” were the most important attractions and
activities when selecting destinations for their pleasure trips.

0 Most or all pleasure trips involved overnight stays, and more than half were
related to out-of-state trips.

0 Showed the lowest mean score for out-of-country trips among the ﬁve
segments.

 

 

 

 

Summary of Results
A summary of market relevant proﬁles and marketing mix strategies for the ten
heavy consumer groups, the propensity to donate to arts organizations six sub-segments,
and the inclination to travel out of state and/or out of country to attend live performances

ﬁve sub-segments is presented in Table 38, Table 39 and Table 40 respectively.

129

 

.Eoto

.33.». 05.5 38.0.0 08:003. -
.8058 sauna—5...: 0.50 0.5 A6... 3 $500.0.“ 05:5 5.. “03.3.0050 80.303
.05 3.5... 88:00 5 25.03.5090 0.5 wﬁgoztoq .5 80.303 $0..me 82.30302 -

 

 

 

 

 

 

 

Ea «Each. 30: wad—000.50 0000.58.50 - 0.00.0.0 .5 wﬁvcoau. .89 .
0.000.030 .0 0820.5“. 9 0000.00 25.589:— - 05 0.. 3mm .50 63 .5 00.5 80.0.0 08...»... 0 .vmm m. 0.00.0.0 5.. 20.. 00.5 08003.. - a .0 vm
8.9.0.5. winBEb 0>.800ww< - @5va 8.0050 0.08 can
@6ch $000.50 ﬁsh AﬁommV 8.5—.58.? .3ch $3... $0.53 0.00.88 00 00:00:03.. -
.8808 80.830 - 08.08 06 2:85 -
A83. 02.»... .5 00.30 0w0=o0 80»-.. -
.2808 05.5 88.6: 6855.. -
.8058 5.855.... 050 0.5 @6th 03.00.00}..Em..\m.0=0..¢\80=5£0:.000
05:75-59 8500 =2.—2.3 .05 362.. 0.00500 .5 m:0.8~.:0w5 t0 5.55.00“. .5 80.303 63an 000530302 -
@000. 9 02085000 .05 :08 .0006 3.35250 - .88 wﬁccotoa 05 8 300.0: .5 5.20% .82
80.803 .05 002.8030: co 3.0.0.0200 0:00... - 06 0.. $3 .05 .23 .5 00.5 000.0: 08:»... 0 .03 m. 0.00.0.0 5.. 0.0.. 00.5 0w80>< - 36 m.
.mO—QSOU ho.“ m—NOU ”WGv—UMQ .Ao\o — NV Wat—00:00 v—OOH U5
5 38.0.0 05500.0 0.00.0.0 5.00% 025k. - Aﬁcmmv 3.5050 05% .3ch 8.558.? .36va 9.0... .36me 0.00.0.2: 00 00:00:00.3. -
.883 06.52 -
.38... 20 800.. 40.9. -
.8058 5.005085 0.50 0.3 36$. 9500.0: 05.5 5.. 03.5.0050 80.303
0.0.05. 2.3 8 $600.85 aim—5.8.0”. - .05 362.. 000500 .5 05.80.5090 :0 wEEotﬂ. .5 80.303 .15va 000950302 -
. m 0: m n 0 3 0000.0: .5 950.5% .89
8 Ba 8388...: a 050.028 88.085 - 65 0 820 Ba .30 .8 8% 0.8.60 :20... a .3 0 08.6: s. 28 8E 60863.. - 2 mm
0.3... 0.00.85 .5 5.50.0» 5.50.0 am... 8&0 - 8% .3603 8.555% .36va 0.8:. .363. 0.00.02: .0 00.50.2030 500555 -
.383 202 a 08 8.86 6585 -
.368. 00;»... 5 00030 0w0=00 50.4 -
ex...
0005.
000.008
.8085. 5:. wiﬁfaﬁ .0385“. 05.05 000.003. .5 05m 0.002

 

 

8qu 00:50:00 .900: 05 .5. 8.930% 0...). $500.82 .05 850.5 508.0% 000.00.). .mm 030...

130

. 1.. .tr....\ .1." 2:2... s.

 

88.0.8.0 5.. 8008.0 -
8:00:00 0.00: 0:0

.88. 8:8 0200 0302.... -
.8058 50085.8. 050 08 @000. 8.80.0.0 0:..:0 5.. 0088.008 80.303

0:0 .82.. 88:00 .5 80.808090 :0 8.85.0.0: .5 0.8.303 .080... 82.08302 -
88.0.. :0 8.0008 .80.

 

 

 

 

.2802: in... .o 8080... 0008 0 0032.. - 05 m. m .8 0:0 .000 .o 800 .800 08:0... .08 m. 0300 a. 0.8 80.. 00802 - mm .0
88.303 0:0 8000830: :0 88.80.60 300". - .3... .. 0.38 .9850 0:0 0x0. .. 0:00:00
5. .82. 3.0% .3... m. 88:8 .02 .88. 2802: .800. 3.0... .0 85053. -
A0300. 00.082 -
.88. so as» 8-2 -
A0800. 0:..:0 88.0.. 80:08.. -
.8088 5.88.0.8. 050
08 0x. . D 8.3.8.000...8.805580008800000 0:0 @003 8.80.00 0:..:0 5..
88.0.8-0 5.. 80085 - . -
80:08:50-0 0:0 00:08.5 303 w:0.:m - 0088.008 88.303 @000. 80800 5 30.88 :0w5 .8 38.80.00: .5 80.303
. 8 000 .5 88.0.. :0 8.0008 n. m 0
3223-0 028:. 9 0.3 2050 05 0.00 - 82 20 a 88 Ba .08 .o 80.. .300 0.20... .08 0800 a. 0.8 800 000.03. -
8808.8 8.80.880 0>.80:ww< - Aﬂomm. 8.02.08»... 0:0 Ackvm.
8.0050 000.0 .083. 8:00:00 0.00: .088. 9.0.: 6.80... 0.00.38 8 00:00:82 -
.3000. 80.00.” - 80.0w» .5 0800:. -
.82.. 05:52.80 05330 -
.8058 5.88.5.8. 050 08 0x03. 8>.8_0.3.88.3:058008800000
0:0 @003 80800 .5 80.888090 :0 8.85.50 .0 88.303 AXE”. 80008302 -
.5008 ...0..053 .
800.0: :0 w80008
0:0 82303 8000830: :0 8.0.8300 300... - . . . .
.30.: 0:0 0.00808 80.6.00 0w0 :0 0:00“. - .80. 05 8 one» 0:0 00% .. 0 00.5 80.0.. .83... 2% 88.0.. 5.. 0.00 00.5 0w0._0>< - 0.0 nm
. .mmOOOhn— .Agh—v WP—DUGOO Owns—HOBOM Uﬂd
0800.590 00.55 0. >03 0.08.0 0:0 >00”... - exam. 8:00:00 000.. A803 8.02.088 .3000. 880.: A083. 0.00.38 8 00:00:82 -
.300... 00.3... 5 00800 0w0=00 80..-.» -
.88. 52° 3 .50. mm -
0x...
088
80.88
308.8 38 3.80.88 8.80.0: 0....05 80.8.). .5 08m 0002

 

 

 

 

 

.0008. .8 0.000

131

 

 

800.0080
0. :0803 0.8.0 0w0500:0 88 8:0>m. -

.8008 00-053

A0080. 0:..:0 88.0.. 88.0.50 -
80058 5.88.0.8 0.50 0.8 0.0.8. 8.8.0..
0:..:0 5. 000:0.008 08.8.03 0:0 0.0.8. 0028.000...08.000002883880008
A0000. 80800 5 80.80.0080 :0 8.85.50 .0 08.0003 A0050. 8000830 Z -
88.0.. :0 w:.0:08

 

 

 

 

 

 

000 0.0.0; 80000030.. 00 000.026.. 0.000.. - 0.0. 2.. 00 88 000 .000 .0 800 0.0.. 02.0... 0 .000 .0 800 .80.. 0000:. - 0.0 00

.:0803 6.0.0.. 8:00 0:0 0.0.0.. 8:00:00 0.00.. 0000..

0. .00000 .0508 5.3 0.00.0 0:0 0.00.002 - 0.808 0.8000 .0.\..NN. 0082.08 .0003. 8.0.0 600$. 0.00.008 .0 00:00:05.. -

A0003. 000.0 w:.>.. 0:850 -

0.080. 80. 5 00.800 0w0..00 800-0 -

00.0. 20 000.. m0 000 0000. 0.0 e000 8 00.5 -

.008. 0000.. -

.8:0>0 .00 0%.: 8.00803 .0000». 0:..:0 88.0.. 88.050 -
0:0 800.5080. .000. 8 .0080 5.3 88.0.. 80058 5.88.5.8 0.50 08 0.0.0... 0038.83.80.800008808800000

00.05.08 00 .000 2.0.00.0 008000 0030... - 0:0 000.. 05.80.. 00.0.0 .0. 000.0020 0.0.0003 .0000. 000000302 -

8008.08.05? 88.0.. :0 w:.0:08 .05. .
0:0 08.8.0 3 .80008 30: :0 8.0.80 >00 0000... - 8... m. 0300 0:0 . . 5 .0 00.00 08.0.. 800w... 0 .03 m. 88.0.. 5. 0.00 00.00 0w000>< - 0. . Nm

88.0.. 8008.0 0:0 08080.0 5.08 00.5.0 - .000.... 8:00:00 000.0 0:0

.0200; 000 .. 8.000020 008. 0.0808 000. 00$. 00.0 .0000. 0.00000. .0 0000002.... -

0. .00000 8.008 5.3 0.00.0 0:0 800.00.). - A0080. 0.2.800: 0:. :. w:.0.m0._ 0.0000 03... -

.000... 00.0 .0 0.0% 8 -

.008. 0.000.. -

00.
0.8..»
3088 0.8 8.08.88 8.8050 0F. 0 .— 0 08.8.). “WWW“ 0002

 

 

80:00. .wm 0.00...

132

 

.3528 x09. 28
mam—n £862: go .8528 not? a 0285 -

.3..va 8:5 $83: 822.5 -
.8058 gang—85 88 Ba 3&8 ware—oz 2.2.8 8m cod—«Beam 8:303

Ea 335 £088 .5 8233530 5 was—Stun Mo 83303 A$N8 EaaamBoZ -
98.0: so 335%

 

 

 

 

 

 

 

32:38 5352 525 Ba 25 - Es 2: a 8mm Ea .5 ho 8% 3°: .8ng a 62 mo 8% E2. omens =< - on 2

as???» 8 £3.35 - .35 €350 ﬁg 2a 2%:

.Sﬁowou how—o =8 83:8 65 ﬂagm - 06:8 3:38 Aﬁoomv mtoocoo x92 Agowv 2862: 6x33 $83 3 359.82 -

@2355 wave—88-0 o>_mwo._ww< - AXES v3.52 -

.3..th Eon—$0380 253......— -

.333 2o .5: 3.3 -

éxbwv 2&5 88.0: 829:5 -
.8858 cougar—£5 28 Ba 3&3 wﬁuoxo: 05:8 he tau—302$ 8533

98 3..th ﬂaunmko: Aﬂoat «.353 8 gong—EMS ta wsgotoq mo 8:303 -
.3828 x08 can .383: cc 9:2on

.933 £332: mo 5:023 35> a 025.5 - .82 05 3 ma Ea .msm ho SE “8.0: $232 a .3» mo 8E “83: owﬁo>< - E _ mm

.wﬁﬂcgum .82:th :55 c5 Baum - .Qotv 8.60:8 S35 3.6:

.mﬁmEggb co £3235 - mﬂcoaaﬁxm AXES 330:8 x02 .3..an mam—n AXES £832: 3 35282 -

.Agmmv $2 8 3&3 owe—Bo 323. -

3.3 so as» $3 -

Aﬂoat oEEum -

3.9
98%
8x38
338% 5:. maze—HE 3338 2:95 uuﬁa—z mo 35 252

 

 

5:83 .mm 2.3

133

I"! -Iu“ Ill-IIqunlau.“ Iii.“ :Qnutllﬂh iFhlmle\~Fi& hi‘x‘th-E .AVMx. IJNBN~N.~.

.83 wﬁgotoa 2: mo E08338 >23: 5 98% “SEE ho $8895;

 

.393. 2:5 8 maze—BE aim—8:23— -
3:303 93 ﬂog—Ems»: co wﬁmﬁozva mace..— -

AXES 05:8 38.0: 08:23 -
.8958 :ouaEhomﬁ ban—ta 8a AXES wcuoxo: 05:8 he uni—38% 8:303
28 afoot £350 .8 acegﬁwho 5 wash—eta uo 8:303 Afamt “rummage Z ..
.996: :o wsvcoam .89

 

 

 

 

 

 

 

 

uoﬁowou 5.80 :3 8.9.8 85 aco>m - 2: m_ mpg v5 .3» no 3E. .86: 3.233 a .3» mm 383: he 23 8E owﬁo>< - «ANNA»: N

.momcoﬁﬁzm Ea mam—n £832: .3..va 8.8250 *6.— EB @ch o m
bzasa AME? cage—om 35> a 0255 - 350:8 NNE. AXRG moEosag AXES man—a Ackowv £832.: “a 352.22 -
ex... a 3:32 -
.333 3.3: 8 25.85 we 0:805 22345: 3:52 -
3&3 2o 3 as» 3.3 -
.3..wa 3:5 383: 08:8?— -

.8828 =2§E£E 28 2a 3&3 833.8}::auuucthouaaaEmzvoa
5205.36.83 c5 AXES E053 3 m=o§_§w3 ta wisp—atom mo 8:303 Angv ﬁning—""302 -
. .883: so 3:52: .88 :3:
can 8:303 ”£35330: so was???“ «soon. - o m. 5 . 0 3:9 u u. 8.3. a . m. o o .5 an aura own—3 a . _
83235? 93 983 $3238 5 . nth v on: . a x .2 .n 3% .AMVMV ”tow”? xoor 23 Aim“ 9 m
o O

.0525 .350 canoe—om 35> a 0235 88:8 NNE. A33; 880595? AX.me $3:— Ao\ov$ £362: “a ougaoz< -
AXLE 8&8 owe—Bo .393. 55 8.3: -
3&3 2° 3 as» 3.3 -

§V

08%
“SEE
$886 58 mate—BE .535qu oan 8x32 .«o 35 Eon—Mom
macoawom

-nsm 5m mcouﬁimwuo 3.2 9 Bacon 8 bmmconoi 05 Sm momwoﬁbm 52 wane—.82 98 81.35 Ego—om you—.82 .9. 28¢

134

.mﬁa wﬁnnotom 05 mo £083.50 ~93... 5 82m 8%.: mo 09350th a

 

 

 

 

 

 

 

.358 25:0 $86: 08:25 -
.8958 cog—585 E8895 2m 3&3 9.38.0: 0525 he @336on 8:303
28 AXES ﬂog—mamas: Afavwv E028 S maoumﬂcawk. ta wﬁgoton mo 8:303 -
.3338 N53. 93 33:8 *5.— .883: :o 9%:on .88 .
33 $235 mo :23»? 35> a ”as; - as a SS Ba .23 ho 8% 33 “8ng a ._3 2 as? .8 23 8% ome”; - «3 .MQ v
.mgommho: .8 gouache.— m gum - Aﬂoa: 93:8 335 @0va N h
.m_:u>u~-o so mania—m -
£3onme wauoxaﬁé o>_mmo.&m< - moEoiE? éxamv 3.80:8 v.02 ARES 933 6x63 £81,388 352.82 -
.393: 33%: 1 8on mo 2:85 233.3: .352 -
£8: 2.. 3% $-90. -
.333 25:5“. -
drama 25.8 38.0: 085:5 -
@858 :oamﬁouﬁ 5.5.3 0.3 3&8 wage—o: 25:0 Bu uoﬁﬁoam 8:395
£5383.— .moO. 8 6.5% .23 383: can AXE» €8.50 8 mcounnmswmho ta wEE—otoa mo 3:303 .3..me ﬂoaaqmioz -
wSE 68 a .5..on own can u So - .386: .8 wﬁnconm .39 «£63
. .n 8 no m m x u. E . . . M
$3.02 “5.8.0.6 2a mica 823. 332.. - 2: g 35 Ba 3» mo 8% Eu: :2»; a 5 a mean he 23 8% ”we”: - m o
.e.\&: 350:8 8:3 Ea 3:8
350:8 Nana .318 moEonag Aﬂoat 933 6x63 £362: “a 35282 -
§§ 2238.. 2.. a 328. oaoa 9.: -
£2: 53 a as» 2 -
Qt
22?.
8x38
3898 5.: mate—.88 32:88 0:35 “8.82 .8 35 Eoﬁwom

 

 

558v .3. ”Bag

135

at: m5=:0.t00 05 .8 50:50:00 >80: 5 08cm .00—.85 .8 88:00:00 0

 

 

 

 

 

 

 

Aggy 8:8 $050: 08:05: -
80500 8:05.805 8.3:: 0::
3&8 83:50:}_80u::0Eh00§50:000 :8 0.003 50500 8 0:585:88

$00.0: 5:000:00 0:305 - :0 w5::0.t00 .8 00:30? 6.0th w::00_0: 05.8 .80 :8:500% 8:303 - .
0:00:00 .9580 :8 0:00:00 0.00: 900.0: :0 w5::0% .83 «6.0: o
as: gassehwahwshw 8%“ Human”: - 2: 0 a: 2: .80 :0 8E 30: :20: a .30 33:: a: :8 8:: 00822 - N m

0588.8 wmwﬁwawEb 02.00?” - AXE: 0:28:50 :8 0.00 5 50:5

3580 .380 0:00:00 0.00: 6:003 98.: 6.003 0.0505: 8 005::0:< -

Aﬂoat 80.83 :5: m8— .8 0:805 20:00:2— _0:::< -

0.08: 20 an: 2 30.5 -

.0080 2:8 39:: 08:05.. -
000.500 888005 8.3.: 0.5 $08 50000030: :5 @003 M50008: 0530

0:00:00 ~€580 :5 0:00:00 0.00: 8.: 25.5009. 005303 $.0ch 88:00 8 08:05:88 5 w5E.8.t00 .8 85303 -

was: .2058:— 00 88080. 85> a 0:305 - 300.0: :0 w5::000 .80: GN:
055039 85.5 20 0 :5 Ba .5 :o 8E .30: :20: a .20 0 03.0: a: 28 8E 00202 - a 3 m

0000— 0: 00855000 :8 :05 805: m5:5500 - .?\..m _V 850.85% :5W 3%: 0:55

$558380 :0 58:05m - .3580 4.x. _ Q 3.80:8 0.00: AXES 0.050 6.003 5050::— 5 008::0:< -

850:3» w5u00t050 0>mmm0hww< - .385 0.53.3 800.05”: -

Aﬂoat 000.03 I 80.03 ‘8 0:805 20:00:05 _0:::< -

.008: 20 as» $2 -

0x:
08%
80:05
30:30 55 w58¥85 .5508: 0:88 .00—:02 .8 05m 8033

 

 

3.503 .0m 030,0

136

.33 ms Eaten 05 mo 8083.50 ban: 5 0.3% H338 mo 0383.5;

 

.3..va 03:8 38.0: 320:5 -
.AgotmooSom =28§8E :38 2a wﬁwuxou 2E5 E wag—£03m 8:303

 

 

 

..mo_=ona:$m 23 c5 axﬁwv Maegan—mac: AXES £028 .5 mcoumnmcawuo ta mascara mo 8:303 ..
933 .2852: Mo .8328 not? a 0235 - ﬂow—oz co wﬁncoam .89 36.8
.253th Bo: he Evian. a gum - 05 ﬂ N55 98 £3 mo 8% “8.0: $23: a 63 m_ ﬂow—or you 23 3E owﬂo>< - a N6 m

.wﬁmmtozuab 35> 0255 - éxﬁmv mtoocoo x02 and @005

83085. 33858-0 o>_mmu._ww< - mtoocoo mug. 65va weigh—Eb“ Aﬂoat $83 .313 £375.: a ouﬁvcoﬁx -

.Ao\om.owv 22.8.5: 05 E wﬁEmﬂ 20:. 8 28.—an 03,—. -

$.an $2 .5 8&8 03:8 so: -

§§ 52° 3 e3» 3.? -
.8958 533:85 Batons: v.8 3&8 8>=£9<>=gmbucoﬁhoocﬁﬁascoa

.983 .3832: EB 3&3 ﬁasco 8 m=o_§_§w8 ta wﬁcuotoa .«o 8:303 exams Eunanguz -
@320 5.: mo cocoa—ow 35> a 025.5 .386: no 9.65% .89

8:3? 93 exams? so we???“ mac“— - 2: a 85 2a _a me 8E 5%: ~8ng a a a aﬂoﬁe uwahwmwamwwgm - “aw: _
95.5.59. .82 «a .555 .23 38.0: 3:8 N52 6. 70:8 9F? A: V93 a .? vmwwwsﬁ a om ham - _ e

wEEnEoo 8 :25 88,—on owe—own 025.5 - mt . \oov . .— \omw _ \ia _ . a cat 3<

Ao\ommv 20:032. 05 5 wEEmE 89:3 93; -

.308: 8&8 owo=oo 30%.? 55 any: -

Axes 3% s as» 3.? -

AX;
98%
“8.8:.
$883 55 maniac. .358; oan “3.32 mo 8% EoEwom

 

 

 

 

 

mﬁcoﬁwombsm 3E mooﬁgctum o>3 ccou<

8 bassoongéso 8:28 BSmuHAESO 322,—. 3 soumczoﬁ 05 SM momwoﬁbm £2 wave—32 can 858m Ego—om #8134 .3 033‘

137

at: w£:::0.::0: 0:: :0 £053.80 .900: 5 03:0 8:35 :0 0w:::00:0m .

 

 

 

 

 

.8050: :0.:::E:0.::_ £2: 0:: £63 £33030: :5 @0th $500.0: 05—:0
:_ wﬁnzﬂoam 8:303 6x3: 0:02:00 :0 m:0_::~:::m:0 t: w£:::0.::0: :0 00:30? -
900—0: :0 w£::0%
8:30; Ea eawhmwﬁmmwmwmwﬁmﬁmu .59 2:: 8 NE: Ba .3: :o 8:: :33: :30: a .3 :o 8E :83: ”020% - «$3 :0.
.m0_w0::::m 3:80.380 0>_mm0:ww< - 6x6: 3:00:00 $.35: AR.» 3 o n
03:23:»: AXES 3:00:00 :00: Aﬁmov 98.: .303: 283:8 :0 005::0::< -
.3..th 000.35 .. oocdmm :0 08005 20:00.8: _::::< -
.393 20 :8: m3». -
.§§ 930: 03:05: -
00050.: cots—E85 >0:— 0:: £02.: wag—0:: 0:=:0 E 33302:: 00:30;
:5: exp: 5.39.30: .333 m:0::00 :0 22:32:30 :0 wﬁgota :0 00:80? -
8:303 :5: 0:093:30: :0 3:03:33 0:00“: .. .3003: :0 2:59. 8.3:
05:52:... 00:5 022:: - .39 2:: a 3: Ba .3: :o 8E .90: 52:0: a .3: :o 8:: :80: 00802 - a . v
.:0:00: 0:: 00:38:: wESxEEb 0>_m00:wm< .. :be 3:00:00 am :5 £53 v o
00:06:: :0w:=0> 0: .0250 3:: «0::0> - 8:233: AXES 3:00:00 :00: AXES 93.: 6x38 £862: :: 005::0::< -
.308: 30.35 - c863 .:0 0800:: 20:84.0: :0:::< -
$.30 2o :8: mm 55 $0.: -
.8058 :0_:::::0.::_ SEE: 0:: 30:8 33:20:}::3:h::0£)00:8£m=30
:5: 302.: :33030: .3th 0:3:00 :0 0:03—35:90 t: wagotoa :0 8:303 -
.m0_:0:a:im :5 30:0: :0 9:53
99:: £632.:— .:0 :0_:00_0m :0_:a> : 0:305 - .80: 0:: m: 35 :5: .53 .:0 00:: 30.0: :00:w_: a 6:: .:0 00:: 8:0: 0w0:0>< - «2.2: m
.5:0::..:0.::03 A0510 3:00:00 533:: @ommV Q:
:5: 8:303 .m:0mn%30: :0 w::0_::0>:: 0:00“— - 03:23:»: ﬁx.» 9 3:00:00 :00: 6x02: 93.: .338 28:05:: :: 005::09: -
.308: 0:0. :0 00:wo: 0w0=00 80%-: -
3&3 52° 5 $0: 3.5. -
ax:
0:..:0
8:38
30:98 0::: $300.88 350:0: 01.35 30.82 :0 0N_m ::0_::wom

 

 

 

 

 

6.38: .8 03a:

138

Based on the results of three Exhaustive CHAID analyses and statistical

comparisons of the segments, the results of the testing of three hypotheses presented in

Table 41.

Table 41. Results of the Testing of Three Hypotheses

 

 

 

 

 

 

 

Heavy consumers of performing arts can be further
divided into substantial, measurable, accessible and
actionable market segments, based on the inclination
to travel out-of-state and/or out-of-country to attend
live performances.

 

 

to travel sub-
groups are
mutually
exclusive.

 

Hypothesis Result
Hypothesis 1 Ten heavy Accepted
Persons who attend live performances (eg., play, folk consumer

dances, musicals and rock concerts) can be classiﬁed groups of the

into substantial, measurable, accessible and actionable performing arts

market segments based on the number of tickets are mutually

purchased to the performing arts. exclusive.

Hypothesis 2 Six propensity Accepted
Heavy consumers of performing arts can be further to donate sub-

divided into substantial, measurable, accessible and groups are

actionable market segments, based on the propensity mutually

to donate to arts, heritage or cultural organizations. exclusive.

Hypothesis 3 Five inclination Accepted

 

139

 

CHAPTER 5

DISCUSSION AND CONCLUSION

Overview

The market segmentation process is a critical component of positioning
performing arts organizations whereby marketers identify distinct consumer segments
that best maximize value to performing arts organizations. To understand the market of
performing arts consumers, this study has three main purposes: 1) to explore and identify
the important predictors which distinguish performing arts consumer groups; 2) to
segment and proﬁle the propensity to donate to arts, heritage or cultural organizations of
heavy consumers; and 3) to segment and proﬁle the inclination to travel out of state
or/and out of country to attend live performances among heavy consumers. Three
decision trees were developed using a criterion-based predictive segmentation method,
Chi-squared Automatic Interaction Detection (CHAID) (Wedel & Kamakura, 2000;
Magidson, 1994; Kass, 1980).

Through the process of three segmentations, the overall sample or selected study
sample were divided into mutually exclusive homogeneous subgroups. Each subgroup
has a different probability of target outcome based on a combination of three types of
predictors — socioeconomics, household information and arts-related experiences.
Exhaustive CHAID was selected over the original CHAID algorithm because it has a
better chance of ﬁnding optimal association between predictors and a dependent variable.
The target category of each criterion in the decision trees was: 1) heavy consumers
(purchased 10 or more tickets to live performances); 2) donors (donated money to arts

organizations); and 3) cultural tourists (attended live performances in a different state or

140

country). ‘Donation activity’ and ‘travel out-of-state and/or out-of-country to attend live
performances’ were used as independent variables in the ﬁrst CHAID analysis and as
dependent variables in the second and third CHAID analyses. The three CHAID analyses
were considered to be distinct.

Data were collected through an online survey of E-club members of the Wharton
Center for Performing Arts at Michigan State University who had purchased a ticket at
least once from the Wharton Center and were on its email list. The E-club newsletter
announcing the online survey and providing the link to the survey was sent to a total of
39,011 valid email addresses and approximately 13,503 members opened the newsletter.
Overall, the survey lasted for 12 days in January/February 2007, and a total of 4,744 E-
club members responded to the survey, for a response rate of 35.1%.

Most valuable consumers usually represent a large percentage of its revenue in a
business (Jackson & Wang, 1994). Since the heavy consumers represented approximately
30% of the performing arts market and accounted for over 50% of actual spending on
tickets, this study focused on them. The results identiﬁed important predictors of
performing arts consumers, based on the number of tickets purchased, and distinguished
ten-heavy consumer segments. There were viable sub-segments from two different
decision tree models for ‘the propensity to donate arts, heritage or cultural organizations’
and ‘the inclination to travel out-of-state and/or out-of-country to attend live
performances’ among heavy consumers. Chi-square tests and one-way ANOVAS were
conducted to examine statistically signiﬁcant differences among derived segments with

regard to their characteristics and behaviors toward the performing arts and pleasure trips.

141

This chapter summarizes the signiﬁcant ﬁndings, provides implications to
managers and marketers of performing arts organizations, discusses limitations of the

study, and presents recommendations for future study.

Summary of Findings
Heavy Consumers of the Performing Arts (objective I)

The sample of 4,138 respondents who were divided into two categories by light
consumers (purchased 1-9 tickets, 69.4%) and heavy consumers (purchased 10 or more
tickets, 30.6%) - was utilized to identify signiﬁcant predictors of the heavy and light
performing arts consumers. The decision tree resulting from the Exhaustive CHAID
analysis identiﬁed signiﬁcant predictors of heavy and light consumers consisting of 6
levels and containing 34 nodes. Donation activity was the most important predictor of
heavy consumers; attendance at live performances in a different state or country, income,
age, membership status, gender, purchase of art products, employment, children’s status,
volunteer status and marital status were other signiﬁcant predictors.

This ﬁnding supports previous empirical studies which indicated socioeconomic
characteristics were important predictors of consumers’ behavior patterns toward the
performing arts (Willis & Snowball, 2009; Favaro & Frateschi, 2007; Werck &
Heyndels, 2007; Levy-Garboua & Montmarquette, 2002). Twenty mutually exclusive
consumer groups in the performing arts were identiﬁed according to their index scores
that showed greater and less than 100% which represented the proportion of the heavy

consumers in the overall sample. Since heavy consumers were a target category, ten

142

consumer groups were identiﬁed as heavy attendance segments which were more likely
to purchase 10 or more tickets to live performances a year.

The best three groups of heavy consumers were: 1) attendees at the performing
arts, who were 45 years or older, donated money to arts organizations, had membership
in arts organizations, and attended live performances in a different state or country
(72.3% heavy consumers); 2) attendees, who had an annual household income of
$150,000 or higher and donated money to arts organizations but did not attend live
performances in a different state or country (55.4% heavy consumers); 3) attendees at the
performing arts, who were 45 years or older, donated money to arts organizations, and
attended live performances in a different state or country, but did .not have membership in
arts organizations (54.4% heavy consumers).

The least likely group to be heavy consumers of the performing arts was female
with an annual household income of less than $50,000, with no donation activity, no
volunteer in arts organizations, and no art products purchase. The ﬁndings of this study
distinguish not only heavy and light consumers of the performing arts but also predict the
heavy attendance segments more likely to purchase tickets to the performing arts.
Donation Activity to Arts Organizations (objective 2)

The second CHAID analysis identiﬁed the donate activity to arts organizations of
heavy consumers. A total of 1,267 heavy consumers of the performing arts who divided
into two categories - donors (61.3%) and non-donors (38.7%) - were selected for this
CHAID analysis. The decision tree resulting from the Exhaustive CHAID analysis,
consisting of 3 levels and containing 9 nodes, identiﬁed the propensity to donate six sub-

segments, including three donor groups and three non-donor groups.

143

Membership status in arts organizations was the most signiﬁcant predictor of
donation activity, and volunteer status, age and purchase of art products were important
predictors. Compared with the overall sample, segments 1, 2 and 3 had greater index
scores than 100% that indicated these respondents were more likely to be donors; while
segments 4, 5 and 6 had lower index scores than 100% indicating that these respondents
were less likely to be donors. Consistently with the previous research and regardless of
the segment, a majority of the respondents in the propensity to donate were female,
Caucasian or White, had a 4-year college degree or higher, an annual household income
of $50,000 or higher, and two persons or more residing in their households (Kastenholz et
al., 2005; Colbert, 2003; McCarthy et al., 2001; Notzke, 1999; Bergonzi & Smith, 1996;
Silberberg, 1995).

Segment 1, consisting of 16.4% of the heavy consumers, was the best segment of
donors (93.3% donors) among the six segments. These respondents, who all had a
membership and volunteered in arts organizations, were mostly female, 45 years or older
and had completed a bachelor’s degree or higher. Almost 60% had an annual household
income of $100,000 or more.

Segment 2, consisting of 16.2% of the heavy consumers, was a donor group
(87.3% donors), who all had memberships, but did not volunteer, in arts organizations.
They were mainly female and 45 years or older and had completed a bachelor’s degree or
higher. More than a quarter had an annual household income of $150,000 or higher and
over half had two persons residing in the household.

Segment 3, consisting of 20.7% of the heavy consumers, was also a donor group

(62.6% donors), who were all 55 years or older but did not have a membership in arts

144

organizations. A majority had completed at least a bachelor’s degree, had an annual
household income of between $50,000 and $149,999, and. had two persons residing in the
household.

Segment 4, consisting of 23.5% of the heavy consumers, was less likely to be a
donor groups (52.3% donors), who were all 35-54 years old, purchased art products but
did not have a membership in arts organizations. A majority were female, had completed
at least a bachelor’s degree or higher, and had an annual household income of $50,000 or
higher. More than half were married/partnered with children living at home.

Segment 5, consisting of 12.6% of the heavy consumers, was less likely to be a
donor group (30.3% donors), who were~ all 35-54 years old, did not purchase arts
products and did not have a membership in arts organizations. They were mainly female,
had an annual household income of between $50,000 and $149,999, were employed ﬁrll-
time, had three persons or more residing in the household, and were married/partnered
with children living at home.

Segment 6, consisting of 10.6% of the heavy consumers, was the least likely to be
a donor group (26.1% donors), who were under 35 years old and did not have a
membership in arts organizations. A majority were female, had an annual household
income of less than $100,000, and were single or married without children. The
proportion of gay or lesbian was higher here than in other segments.

Differences of the Propensity to Donate to Arts Organizations Sub-Segments
(objective 3)

Objective three of this study was to investigate signiﬁcant differences among the
propensity to donate sub-segments with regards to their characteristics and behaviors

including type of attendance at live performances, spending on tickets, information

145

sources, online searches and purchases, the importance of cultural attractions and
activities in selecting pleasure trip destinations and pleasure trip behaviors.

Regardless of the segment, musicals and plays were the most popular live
performances. Segment 1, the most likely to be donor group, showed higher attendance of
most types of live performances than other segments, including ‘plays’, ‘symphonies’,
‘jazz concerts’, ‘operas’, ‘folk/ethnic concerts’ and ‘ballet’. Segment 6, mostly young and
single people, on the other hand, attended ‘rock concerts’, ‘new and experimental
concerts’, ‘hip-hop concerts’ and ‘world concerts’ more than other segments.

Segments 3 and 4 paid the greatest average price for a ticket to a live
performance; segment 2 paid the greatest average highest price for a ticket; and segment
3 had the greatest average total spending on tickets; while segment 6 paid the lowest
average, and the lowest average highest price for a ticket, as well as had the lowest
average total spending on tickets to the live performances. Segment 3 comprised of older
respondents (55 or older) who were married/partnered with children no longer living at
home. They may have had less time constraint, thus more chances to attend live
performances than other segments. Segments 2 and 4 included more respondents who had
an annual household income of $150,000 or higher than other segments, while segment 6
included younger respondents (less than 35) who had a lower annual household income
than others. Therefore, amount spent on tickets was considerably related to the income
levels.

With regards to information sources, there was no difference among the six
segments in the traditional information sources including ‘TV’, ‘magazines’ and ‘radio’;

however, ‘newspapers’ were more effective information sources for segments 1, 2 and 3.

146

Websites of performing art organizations or centers or specialized in online ticketing
were more useful for segments 4, 5, and 6 than others. Furthermore, segments 1, 4 and 6
obtained information on the performing arts from ‘acquaintances, ﬁ‘iends, family or
relatives’, and segments 1 and 2 used leaﬂets/brochures and art-related publications as
information sources more than other segments.

Since the sample of this study was collected from the E-club members at the
Wharton Center for Performing Arts, almost all of the six segments used Internet for
research and purchases. Segments 4, 5 and 6 researched and purchased tickets to the
performing arts through web sites more than other segments due to their younger ages
(less than 55 years old). In addition, segments 1 and 4 did more online research for arts or
antiques and collections than other segments.

When selecting pleasure tn'p destinations, segment 1 indicated the highest mean
scores of the importance of cultural attractions and activities; while segment 5 showed
the lowest mean scores of the importance of all the attractions and activities. Segments l,
2, 3 and 4 indicated that “historical/heritage attractions, sites and districts” were the most
important of attractions and activities, while segment 5 and 6 showed that “gardens, zoos
and aquariums” were most important destinations. Segments 1, 3 and 6 indicated
“performing arts” as the second most important attractions and activities in selecting
pleasure trip destinations.

As for the pleasure trip behavior, regardless of the segment, most or all of their
pleasure trips was involved overnight stays and more than half took trips out-of-state.
Interestingly, segment 6 took more out-of-state trips, while segment 1 took more out-of-

country trips than other segments.

147

Attendance at Live Performances in a Different State or Country (objective 4)

The third CHAID analysis was used to predict cultural tourists who were more
likely to attend live performances in a different state or country among the heavy
consumers of the performing arts. A total of 1,255 respondents who were heavy
consumers and attended live performances in another state or country other than they
resided during the last 12 months were selected for this analysis. The decision tree
resulting from this Exhaustive CHAID analysis identiﬁed ﬁve segments, including two
cultural tourist groups and three non-cultural tourist groups. The most signiﬁcant
predictor was membership status in arts organizations; education level was the most
important predictor for members; while purchase of arts products was the most important
predictor for non-members; and donation activity was best predictor of non-members
who did not purchase art products. Segments l and 2 had higher index scores, while
segments 3, 4 and 5 had lower index scores, compared to the overall sample (38.2%
cultural tourists in the heavy sample). Therefore, those in segments 1 and 2 were more
likely to attend live performances in a different state or country while respondents in
segment 3, 4 and 5 were less likely to be cultural tourists.

The best segment of cultural tourists is segment 1, consisting of 19.8% of the
heavy consumers, who had higher than a bachelor’s degree and had membership to arts
organizations (53.2% cultural tourists). A majority were female, married/partnered and
45 years or older. More than half had two persons residing in their household.

Segment 2, consisting of 23.5% of the heavy consumers, was more likely to be
cultural tourists who purchased art products and donated money but did not have

membership in arts organizations (42.4% cultural tourists). These respondents were

148

mainly female, married, between 35 and 64 years old, and had completed at least a
bachelor’s degree or higher. More than 30% had an annual household income of
$150,000 or higher.

Segment 3, consisting of 13.1% of the heavy consumers, was less likely to be
cultural tourists who had a bachelor’s degree or less and had membership to arts
organizations (37.0% cultural tourists). They had the highest proportion of females and
an annual household income of less than $50,000.

Segment 4, consisting of 20.9% of the heavy consumers, was also less likely to be
cultural tourists who purchased art products but did not have membership and did not
donate money to arts organizations (31.7% cultural tourists). A majority were female,
less than 65 years old, had an annual household income of between $50,000 and
$149999, and were employed full-time. More than one-ﬁfth was single without children.

Segment 5, consisting of 22.7% of the heavy consumers, was the least likely to be
cultural tourists who did not have membership to arts organizations and did not purchase
art products (27.7% for cultural tourists). A majority were female and less than 55 years
old but this segment contained the highest proportion of males among the segments.
More than 40% were married/partnered with children living at home.

Differences of the Inclination to Travel Out of State or/and Out of Country to Attend
Live Performances S ub-Segments (objective 5)

Signiﬁcant differences among the inclination to travel out of state or/and out of
country to attend live performances sub-segments were found in their characteristics and
behaviors including the type of attendance at live performances, spending on tickets,
information sources, online searches and purchases, the importance of cultural attractions

and activities in selecting pleasure trip destinations, and pleasure trip behaviors.

149

Regardless of the type of attendance, musicals and plays were the most popular
live performances among the ﬁve segments. Segment 1, the best cultural tourist group,
attended more types of live performances than did other segments, including ‘plays’,
‘symphonies’, ‘jazz concerts’, ‘opera’, ‘folk/ethnic concert’, ‘blue concerts’, ‘modern
dance’, ‘world concerts’ and ‘folk dance’, while segment 3 attended ‘ballet’ and ‘rock
concerts’ more than others did.

Segments 2 paid the greatest average price for a ticket and the greatest average
highest price for a ticket to a live performance; while segment 1 had the greatest average
total spending on tickets during the last 12 months. Segments 2 included more
respondents who had an annual household income of $150,000 or higher than other
segments, while segment 1 included more respondents with higher education (a
bachelor’s degree or higher). Therefore, amount of spending on tickets was considerably
related to their income and education levels.

There were no statistically signiﬁcant differences where information sources used
were ‘TV’, ‘magazines’, ‘radio’ and ‘acquaintances, friends, family or relatives.’
‘Newspapers’ were more effective information sources for segment 1, while ‘websites of
performing art organizations or centers’ were most popular for segment 2. Furthermore,
segments 4 and 5 used ‘websites specializing in online ticketing’ more often than other
segments while segment 1 used art-related publications and segment 3 used from
leaﬂets/brochures more than others to learn about live theater, dances, performances, and
music concerts.

Regardless of the segment, almost all of respondents in the ﬁve segments used

lntemet for research and purchase of tickets to the performing arts. In particular,

150

segments 2 and 4 researched arts and tickets to the performing arts through online the
most while segment 3 researched antiques and collections more than did other segments.
On the other hand, segment 4 purchased arts and tickets to the performing arts through
online more than did other segments.

When selecting pleasure trip destinations, segment 1 indicated the highest mean
scores of the importance of cultural attractions and activities in selecting pleasure trip
destinations; while segment 5 showed the lowest mean scores of the importance of most
destinations. Segments 1 and 2 indicated that ‘historical/heritage attractions, sites and
districts’, ‘museums’ and ‘performing arts’ were the most important of attractions and
activities, while segment 3, 4 and 5 showed that ‘historical/heritage attractions, sites and
districts’, ‘gardens, zoos and aquariums’ and ‘performing arts’ were the most important
attractions or activities. ‘Agricultural attractions and events’ were considered less
important than other attractions and activities to all of the ﬁve segments.

For pleasure trip behavior, regardless of segment, most or all of their pleasure
trips were involved overnight stays and more than half took out-of-state trips.
Furthermore, segment 1 took more out-of-country trips than other segments while

segment 5 took the fewest trips out-of-country.

Implications
For effective marketing communication, performing arts organizations need to
ﬁrst deﬁne target segments and identify their media which the target segments use the
most (Kotler, 2005). This study explored signiﬁcantly important predictors among

performing arts consumers and identiﬁed characteristics of heavy attendance consumers

151

using CHAID segmentation. In addition, based on their characteristics and art-related
experiences, two segmentations of performing arts heavy consumers into propensity to
donate sub-segments, and cultural tourist sub-segments resulted in mutually exclusive six
and ﬁve sub-segments in each analysis. The ﬁndings of this study should be of interest
not only to the Wharton Center for Performing Arts, but also to performing arts
organizations in general. The characteristics of performing arts consumers and the
speciﬁc segments could be used as a primary strategic resource for performing arts
marketers and managers. They can identify and penetrate their target markets and build
effective marketing plans and strategies on program development, service delivery,
pricing, product advertising and sales promotion (Bernstein, 2007; Kotler, 2005).
Heavy Consumers of the Performing Arts

This study identiﬁed important predictors and the characteristics of heavy
attendance segments that can be easily identiﬁed by performing arts organizations and
can provide useful insights for marketers and managers to communicate with their target
segments. The CHAID analysis allowed for the examination of complex interactions and
interrelationships between performing arts consumers’ characteristics and art-related
experiences and heavy- and light-purchase consumers. This allowed for an investigation
of the characteristics of the top and bottom presenting subgroups and enhancing
understanding of the predictors associated with each consumer group of the performing
arts. Heavy consumers who have attended live performances often tend to attend and
purchase more performances in the future (Bernstein, 2007). The ﬁndings demonstrated
how different art-experience variables interact with socioeconomic and household

variables, predicting distinct heavy consumer groups of the performing arts. Previous

152

studies have been identiﬁed important predictors of attendees in the performing arts but
did not clearly demonstrated interactions or interrelationships of predictors and heavy and
light consumer groups. Therefore, this information contributes to understanding the
relationships between predictors as well as the characteristics of consumer groups of the
performing arts.

Donation activity was the most important predictor of heavy attendance segments
of the performing arts. Overall, donors, who attended live performances in a different
state or country, were 45 years or older and had membership in arts organizations were
the most closely associated with best characteristics the heavy consumer group. Using
combination of predictors, marketers and managers in performing arts organizations
could easily predict the characteristics of heavy consumers and of purchase propensity of
their target segments.

The ﬁndings provide clear deﬁnitions of heavy and light consumer groups.
Females have been found usually referred as heavier performing arts attendees, ticket
purchasers, and donors than males (Kastenholz et al., 2005; Borgonovi, 2004; Colbert,
2003; McCarthy et al., 2001; Notzke, 1999; Andreasen & Belk, 1980). The current study,
however, showed one light consumer group that was characterized by females whose
annual household income was less than $50,000, with no donation activity, no
volunteering, and no purchase of art products. Furthermore, previous research has shown
that childhood or current art education was associated with performing arts attendance
(Borgonovi, 2004; Richards, 1996; Urry, 1990) but these factors did not distinguish
between heavy- and light-purchase consumers in the present study where clusters of

predictors are more useful than one or two predictors alone.

153

Propensity to Donate to Arts Organizations Sub-Segments

The ﬁndings of segmentation provide insights into the propensity to donate of
heavy consumers and provide the primary managerial implications to marketers and
managers of performing arts organizations. To penetrate performing arts donation market,
managers should be able to identify characteristics of each segment as well as its
likelihood of donating money to performing arts organizations. It is crucial in performing
arts organizations to understand donation activity, which is most strongly associated with
performing arts heavy consumers.

Researchers have noted that performing arts sponsorship is higher than other arts
sponsorship (Colbert et al., 2005). This study explained important predictors of
propensity to donate and showed how combinations of these predictors were segmented
into performing arts donor and non-donor groups, and how these segments can be linked
to other proﬁling variables. This CHAID segmentation enables the diversity of individual
heavy consumers to be reduced to a manageable set of clearly deﬁnable predictors:
membership in arts organizations, volunteer status, age and purchase of art products.
Since previous research (Brown, 2007a; Harvey, 1990) had not focused on these core
segments of heavy-purchasers and donors at the same time, this study contributes to the
current understanding of donation segments of heavy consumers of the performing arts.

This study provides not only socioeconomic proﬁle of the donation segments but
also detailed behavior proﬁles regarding attendance, spending, information source, online
searching and pleasure trips of the six segments. This information will help performing
arts marketers and managers extend their knowledge of these market segments, enabling

organizations to provide appropriate programs and services and deliver their message or

154

promotions efﬁciently to each segment. For example, plays and musicals were popular in
all segments, but not all segments showed the same preferences for other live
performances.

Information about performing arts organizations and their programs is directly
associated with purchase behaviors (Murray, 1991). Providing useful information in
appropriate ways for valid target segments will be crucial. For example, newspapers were
the most effective information source for segments 1, 2 and 3, which were more likely to
be donors, while websites specialized for online ticketing were a better information
source to segments 3, 4 and 5, which were more likely to be non-donors.

In this study, all of the respondents in the segments were E-club members of the
Wharton Center and high intemet users for researching or purchasing art- and culture-
related products. They usually used the lntemet to research and purchase tickets for the
performing arts through websites specialized for online ticketing or websites of
performing art organizations or center. Therefore, performing arts organizations should
consider signiﬁcant cost savings and effective ways from high-technology marketing.
Electronic media can be used better communicate with their consumers in visual and
audio effects (Bernstein, 2007).

Inclination to Travel Out of State and/or Out of Country to Attend Live Performances
Sub-Segments

This study adds useful information about heavy consumers of the performing arts
who are more likely to attend live performances in a different state or country and
become cultural tourists for performing arts. Previous tourism studies show that cultural
tourists are signiﬁcant segments which are repeated visitors, spend more money and stay

longer than other tourists (Nichols, 2006a; Dolnicar, 2002). In present study supports

155

previous research that cultural tourist segments for the performing arts spend an average
of highest price for a ticket and on higher total spending on tickets than non-cultural
tourist segments. Consequently, cultural tourists are more proﬁtable segments and worthy
of focus.

Understanding heavy consumers of the performing arts and their travel behaviors
related to performing arts is a vital to the marketing managers of performing arts
organizations in various tourism destinations, for example Santa Fe, New Mexico
(Kimmel, 1995). While cultural arts attractions are an important economic resource for
community-building, as well as for attracting more tourists. The results extend beyond
previous studies of cultural tourists for performing arts (Kim et al., 2007; Kastenholz et
al., 2005; Notzke, 1999; Silberberg, 1995) by segmenting heavy consumers of the
performing arts.

Two cultural tourist segments more likely to attend live performances in a
different state or country were identiﬁed, and their characteristics and behaviors were
analyzed. Performing arts organizations and destination marketers interested in attracting
and retaining cultural tourists for the performing arts should focus on people who had
more than a bachelor’s degree and memberships in arts organizations, or who do not have
such memberships but purchase art products and donate money to arts organizations.

Cultural tourist segments who had a higher proportion of two-person households
took more trips out-of-state and/or out-of-country than non-cultural tourist segments. The
results imply that the composition of household is related to not only attend the
performing arts, but also long distance pleasure trip behaviors. This supports tourism

literatures which show'people with young children plan fewer international trips than

156

people with no children. Performing arts organizations in tourist destinations should
position themselves for those target arts consumers and consider their values.

In addition, these segments indicated that historical/heritage attractions, sites and
districts; performing arts; and museums were the most important attractions in selecting
pleasure trip destinations. By, comprehending the needs, preferences and behaviors of
cultural tourists, performing arts organizations can focus on them and redeﬁne the market
for their present products or create a new marketplace according to their values.

Compared to cultural tourist segments, non-cultural tourist segments among the
heavy consumers may have greater lo yalty to performing arts organizations in their
community. These segments had a higher proportion of young people; they usually
obtained information from the websites of performing arts organizations or centers.
Performing arts organizations can use this information in their marketing strategies to
sustain a strong relationship with, and to expand commitment from non-cultural tourist
segments.

These accurate and applicable results can aid marketers and managers in
identifying differences among performing arts consumers, and help them as their tailor
marketing strategies for their target segments and positioning new programs and services.
Not only is it important to know what variables related to ticket purchases, but also who
is more likely to be a heavy consumer, a donor and a cultural tourist for the performing

arts.

157

Limitation of the Study

Despite of the contributions to the broader theoretical and practical understanding
of the characteristics and behavior of consumers of the performing arts, this study has
several limitations.

First, generalizability is a common limitation of social science research. This
study used the E-club members of the Wharton Center Performing arts at Michigan State
University (Wharton Center) as a study population. These respondents tended to be more
familiar with programs and more involved with the organizations, thus these results may
be over-represented (Brown, 2007a). Therefore, generalizability of the ﬁndings may be
limited to apply to other performing arts organizations that are of different size and in
different geographic region (Trochim, 2001). In addition, survey of other members of the
Wharton Center, who were not E-club members, may show different results from the
ﬁndings of this study because long-term subscribers at the Wharton Center programs are
senior persons who may purchase 10 or more tickets to live performances per year but do
not use lntemet.

Second, a common limitation of CHAID applies to this study. The segments of
each CHAID analysis have deﬁned based on a single criterion variable because it is
difﬁcult to obtain a result of a single common segmentation if multiple criteria were used
(Magidson & Vermunt, 2005). In addition, using one criterion variable resulted in a set of
segments in each study. However, using alternative criterion variables or different study
populations with identical design might not produce the same set of segments. The

categories of a predictor may merge and split differently, depending on the criterion

158

variable. Therefore, the results of this study may not be generalized to other population or
a similar criterion variable (Chen, 2003b).

Third, potential predictors used in the CHAID analyses were 15 variables from
three categories. Since heavy and light consumers are one of the most common criteria in
market segmentation, other signiﬁcant predictors, not included in this study, may exist to
identify heavy- and light-purchase consumers in the population. This also could yield
different results.

Fourth, a common limitation of online survey also applies in this study. Due to
multiple email addresses for the same person, multiple survey completions by some
participants, and invalid email addresses, the online survey might be caused sampling
issues (Wright, 2005; Andrews et al., 2003; Couper, 2000). A systematic bias could have
occurred between respondents and non-respondents because younger cohorts are more
likely to complete online surveys than older cohorts in general (Wright, 2005). In
addition, since the data were self-reported, there was no guarantee that participants

provided accurate information on the survey.

Future Research
This present study focused on heavy consumers rather light consumers of the
performing arts. Future research should examine in detail the light consumers and
identify their needs, values and preferences. To distinguish heavy and light consumers,
this study conveniently called all of those who purchased one to nine tickets to live
performances during the last 12 months as a light consumer. However, light consumers of

the performing arts including moderate- and light-purchasers in this study identiﬁed

159

several mutually exclusive homogeneous groups and whose characteristics were
signiﬁcant different from each other. Some of these groups have more potential to grow
their spending on tickets since heavy consumers have already purchased a number of
tickets and are strongly linked to performing arts organizations, obtaining additional
revenues from them is perhaps unlikely (Liu, 2007). Further study will help performing
arts organizations cater to moderate- and light-purchasers, and to provide and implement
effective market strategies of programs, promotions, prices,-advertisement. In addition,
student segments were not highlighted in this study because few participated, and most of
those who did small number of tickets during the last 12 months. However, students
represent a potentially large segment of the Wharton Center. Future segmentation study
must focus on students and identify their needs, wants, lifestyle and values to attract their
interest and attendance.

Further research should also include target consumers who do not use intemet but
heavy consumers of the performing arts such as subscribers, group buyers, box office
buyers. If those are no signiﬁcant differences between E-club and non-E-club members,
present results can be generalized to all of the heavy consumers at the Wharton Center. If
there are signiﬁcant differences between two groups, same CHAID segmentations should
- be need for performing arts consumers who do not use intemet at the Wharton Center,
thus identifying their segments and behaviors.

Different patterns of performing arts consumptions are strongly correlated with
socioeconomic and regional characteristics (Willis & Snowball, 2009). Moreover,
donation activity of performing arts consumers differ from the size, type and venue of

performing arts organizations (Frank & Geppert, 2004). These results may not be

160

generalized, thus examining to other performing arts organizations and their consumers is
recommended. Future research, using the same methods in this study, might compare
performing arts consumers in different regions in US. as well as different nations

Not only socioeconomic and household characteristics and art-related experience
variables, but also motivation, preferences, satisfactions are signiﬁcantly associated with
attendance at performing arts (Hume, 2008; Wiggins, 2004). In further studies, all of
external, internal and socioeconomic variables should be used as potential predictors to
examine behaviors of performing arts consumers. In addition, because segmentation
methodology has created multiple views of the same market (Wyner 1995), future studies
should conduct not only CHAID segmentations, but also diverse segmentation schemes
which have various strengths and could be better applied for different purposes and point
of the performing arts consumer market (Hsu & Kang, 2007).

In conclusion, an important marketing principle, the so-called 80-20 rule,
indicates that 20% of the customers, who are usually referred to as heavy consumers
create 80% of the revenue (Koch, 1999). It is important to understand who performing
arts consumers are and how to identify heavy consumers in the market. There is a small
number of heavy consumers who contribute revenue disproportionately higher than light
consumers even though the numbers (80/20) may differ between performing arts market
and the general retailing market. A segmentation study of diverse demand is essential for
the purpose of developing strategies for effective and efﬁcient marketing plans in the
performing arts market.

This study identiﬁes heavy and light consumers of the performing arts by

applying the CHAID algorithm to the segmentation of the performing arts market. To

161

examine detail information about heavy consumer market, two different sets of CHAID
segmentation were then conducted to identify heavy consumers and donation market and
cultural tourist market of the performing arts. The results of the Exhaustive CHAID
analyses provided non-binary decision trees where combinations of predictors identiﬁed
mutually exclusive market segments (Wedel & Kamakura, 2000; Magidson, 1994; Kass,
1980). Each of the segments explained how likely they are to engage in the target
behavior. The CHAID analyses showed the most appropriate technique for selecting the
more meaningful and important segmentation variables as an intermediate step for market
segmentation (Hsu & Kang, 2007; Galguera et al., 2006; Chung, Oh, Kim & Han, 2004).
Therefore, this study provides crucial information about performing arts
consumers, as well as developing a useful approach that could permit performing arts
organizations to identify valuable consumers effectively. In addition, the identiﬁed target
segments and their proﬁles will be essential in building effective communication and

promotion strategies according to their marketing purposes.

162

APPENDICES

163

Appendix A. The Online Survey Questionnaire

 

 

Your Attendance to Performances
During the last 12 months have you attended any live theater performances (e.g.,
plays, musicals or operas), live dance performances (e. g., ballet, jazz or modem)
or any live music concerts?
0 Yes 0 No
How many live performances have you attended during the last 12 months?

01 02 03 04-5 06-9 OlOormore

What type of live performances have you attended during the last 12 months?
Check all that apply.

Cl Play El Folk dance 1:] New and experimental concert
El Musical Cl Folk and ethnic concert [:1 Rock concert

El Opera [:1 Jazz concert Cl Hip hop concert

Cl Ballet El Blues concert El World concert

D Modern dance El Symphony D Other

Cl Ethnic dance E] Country concert

Were any of these live performances in a different state or country other than
where your permanent residence is located?

0 Yes 0 No

Were any of these live performances fee (did not require the purchase of a ticket
or subscription)?

0 Yes 0 No

What proportion of the live performances you attended during the last 12 months
was free (did not require the purchase of a ticker or subscription)?

0 Few (less than half) of the performances were free
0 About half of the performances were free
0 Most (more than half) of the performances were free

 

 

[ Previous Page ] [ Next Page J

 

 

 

164

 

 

 

 

Your Attendance to Performances
Have you purchased tickets to live performances during the last 12 months?

0 Yes 0 No

How many tickets to live performances did you purchase (for yourself or members
of your household) during the last 12 months?

01 02 03 04-5 06-9 OlOormore

What was the average price that you paid for a ticket to a live performance during
the last 12 months? Please, round to the nearest dollar.

$- .00

What was the highest price that you paid for a ticket to a live performance during
the last 12 months? Please, round to the nearest dollar.

s- .00

About how much in total did you pay for live performance tickets during the last
12 months?

s— .00

 

 

[Previous Page il l Next Page ]

 

 

 

165

 

 

Your Participation in Performing Arts

During the last 3 years, have you been engaged in any performing arts, such as
theater, dance, or music?

0 Yes 0 No

Check the performing arts in which you have been engaged during the last 3 years.
Please check all that apply.

1:] Playing a musical instrument
E1 Playing in band or orchestra
El Solo singing

C] Group or choir singing

El Dance

Cl Acting

1:] Comedy, magic or mime

Cl Story telling and readings

C] Other performing arts

During the last 3 years, were you professionally engaged in any performing arts
(e. g., theater, dance, music)?

0 Yes 0 No

 

 

[ Previous Page i [ Next Page ]

 

 

 

166

 

 

 

Your Participation in Visual Arts

During the last 3 years, have you been engaged in any visual arts, such as drawing,
painting, printmaking, writing, textiles, photography, and/or sculpture?

0 Yes 0 No

Check the visual arts in which you have been engaged during the last 3 years.
Please check all that apply.

[:1 Drawing

Cl Painting

El Printmaking

El Sculpture

[:1 Textile arts

Cl Photography

[:1 Crafts

Cl Folk or traditional art
13 Media arts

Cl Writing

[:1 Other visual art

During the last 3 years, were you professionally engaged in any visual arts (e. g.,
drawing, craﬁs, writing)?

0 Yes 0 No

 

 

(Previous Page i F Next Page ]

 

 

 

167

 

 

 

Your Participation in Art and Culture Related Disciplines

During the last 3 years, have you been engaged in any art-related or cultural
disciplines such as architecture; landscape architecture; fashion, graphic, interior or
industrial design; historic preservation or restoration; genealogy; or archaeology?

0 Yes 0 No

Check the arts and cultural-related disciplines in which you have been engaged
during the last 3 years. Please check all that apply.

D Architecture

[:1 Landscape architecture

[:1 Fashion design

Cl Graphic design

Cl Interior design

Cl Industrial design

[:1 Historic preservation/restoration
Cl Genealogy

D Archaeology

During the last 12 months, were you professionally engaged in any art-related or
cultural disciplines (e.g., architecture, design, genealogy)?

 

0 Yes 0 No

 

 

[ Previous Page ] [ Next Page ]

 

 

168

 

 

 

Professional Artists
Do you consider yourself to be a professional artist?
0 Yes 0 No
Have you received formal education/training in your artistic discipline?

0 Yes 0 No

Are you currently a member of an arts-related guild, union or
association/organization?

0 Yes 0 No

How much did you earn (before costs) during the last 12 months from your artistic
activities?

0 No income 0 $6,000 - 9,999 O$60,000 - 99,999
0 Less than $2,000 0 $10,000 - 19,999 O$100,000 - 199,999
O$2,000 - 3,999 O$20,000 - 39,999 O$200,000 or more

O$4,000 - 5,999 O$40,000 - 59,999

 

 

[Previous Page ] [ Next Page ]

 

169

 

 

 

Purchases of Art Products

Have you purchased any visual arts in including drawing, paintings, writings, prints,
graphic designs, crafts or photographs during the last 12 months?

0 Yes 0 No

What types of visual art have you purchased during the last 12 months?
Check all that apply.

CI Paintings

El Books

[:1 Photographs (not including hiring a photographer, e.g., wedding)

Cl Drawings

[:1 Prints

El Sculptures

E1 Crafts (e. g., glass, ﬁber, metal, ceramics, leather, wood, plastic and mixed
media)

U Folk or traditional art

D Other

Approximately how much did you spend for visual art during the last 12 months?
Round to the nearest dollar. Don’t type symbols in the box; only numbers.

 

$[ I .00

 

What was the highest price you paid for any one piece of the visual art you
purchased? Round to the nearest dollar. Don’t type symbols in the box; only numbers.

 

sl 1 .00

In what ways did you purchase visual art during the last 12 months? Check all that
apply.

CI Art Fair

Cl Auction

CI Art Gallery/Bookstore

E] Museum Store

El Catalog

D Web/intemet

El Direct from an artist (no intermediary)
El Commissioned a special artistic work
El Other

 

 

ﬁrevious Page] [ Next Page ]

 

 

 

170

 

 

 

Purchases of Cultural Products

Have you purchased or rented any books, videotapes, DVDs, CDs, records, tapes,
or music downloads during the last 12 months?

0 Yes 0 No

Please indicate how much you spent on the following items during the last 12
months. Please round to the nearest dollar. Do not type commas or symbols in
the box; only numbers.

1) Books $[:1
2) Videotapes, DVDs $ 1:]
3) CDs, records, tapes, music downloads $l:|

 

 

[ Previous Page 1 [ Next Page ]

 

 

 

171

 

 

 

 

Employment of Artists and Performers

Have you hired any performing artists including musicians/bands, actors, comedians,
mimes/performance artists during the last 12 months?

0 Yes 0 No

What types of performing artists have you hired during the last 12 months? Please,
check all that apply.

Cl Solo musicians

El Musical bands/groups

El Solo singers

Cl Singing groups

[:1 Actors or theater companies

[:1 Comedians, magicians or mime artists
Cl Performance artists

Cl Dancers or dance companies

Cl DJs

Did you hire them for a business event/purpose or personal reason/event?

0 Business
0 Personal
0 Both business and personal

Approximately how much did you spend in fees for their services/performances
during the last 12 months? Round to the nearest dollar. Do not type commas or
symbols in the box; only numbers.

$C:l .00

 

 

[ Previous Page ] [ Next Page ]

 

 

 

 

172

 

 

Have you hired any architects or designers during the last 12 months?

0 Yes 0 No

Have you hired any historic preservation/restoration, archaeological, or genealogical
services during the last 12 months?

0 Yes 0 No

What types of heritage related services have you hired during the last 12 months?
Please, check all that apply.

[:1 Historic preservation/restoration services
Cl Archaeological services

Cl Genealogical services

Cl Home/building restoration services

 

 

[ Previous Page ] [ Next Page ]

 

 

 

173

 

 

 

Tourism

How important are the following attractions and activities when you select
destinations for your pleasure trips.

     

i 01

     

lzxtrcmclx' Somcuhut ;, ., Not , .,
Important Important ll]?112‘:,'\tl‘l‘”1
Museums 0 O O O
ltural
gfrs/festivals/events O O O 0
Performing arts 0 O O O
Placest bu local
arts/cra s y 0 O O O
Histo 'cal/herita e
attracfions, sites gand O O O O O
rstrrcts
Gardens, zoos,
aquariums
A ric ltural
a racfions and events 0 O O O O
Purchasin roducts
Fro or grgcessed O O O O O
oca y
c 'tecture and
milgings O O
ustoms and wa s of
Ci:ving y 0 0
Libraries, liter
ve ts and ary O O O O O

OO SIOI'CS

Have you taken any pleasure trips to destinations more than 50 miles from your home
during the last 12 months?

0 Yes 0 No

 

 

[ Previous Page ] [ Next Page ]

 

 

 

174

 

 

 

Your Pleasure Trips During the Last 12 Months

How many pleasure trips to destinations more than 50 miles from your home did you
take during the last 12 months?

01 02-3 04-5 06-9 0 lOormore
Taking in mind your pleasure trips(s) during the last 12 months to destinations more

than 50 miles from your home:
What proportion of these trips involved an overnight stay away from home?

 

0 None

0 Few (less than half but at least 1)
0 About half

0 Most (more than half but not all)
0 All

What proportion of these pleasure trips were to a destination within the state where
you reside?

0 None

0 Few (less than half but at least 1)
0 About half

0 Most (more than half but not all)
0 All

What proportion of these pleasure trips were to another country?

0 None

0 Few (less than half but at least 1)
0 About half

0 Most (more than half but not all)
0 All

 

175

 

 

 

Taking in mind your pleasure trip(s) during the last 12 months to destinations more
than 50 miles from your home:
How often did you participate in the following arts and culture related activities?

.-:\I\\1_\.\ I s111le St ILIUIll

 

Attended ?0 theatrical performance,
dance pero rmance or concert

Visited a museum
Attended a festival

Visited an historical site, attraction
or commum ty

Visited a zogbaquarium, botanical
garden or ar retum

00000
00000
00000
000000

Visited a library

Taking in mind your pleasure trip(s) during the last 12 months to destinations mpg
than 50 miles from your home:

What proportion of these trips had the primary or only purpose to participate in the
following arts and culture related activities?

)111
I11OII'1 -'-

 

   

Attendedr? theatrical performance,
dance pe ormance or concert

Visited a museum
Attended a festival

Visited an historical site, attraction or
communrt y

Visited a 20%, aquarium, botanical
garden orar oretum

000000
00000
00000
00000
000000

Visited a library

 

 

[Previous Page I I Next Page I

 

 

 

176

 

 

 

Cultural Membership

Are you afﬁliated (e.g., member, ofﬁcer, and employee) with a national, state or
local membership-based organization that supports, sponsors, or implements arts,
heritage or cultural activities?

This could include artist organizations that fundraise and sponsor and cultural
activities and facilities (e.g., concerts, art shows, heritage preservation, and
community libraries).

0 Yes 0 No

What is the primary mission and purposes of the organization(s) you are a
member of?

0 Supports and encourages visual arts and artists

0 Supports and encourages performing arts and artists

0 Supports and encourages libraries

0 Supports and encourages arts education

0 Supports and encourages public art

0 Supports and encourages heritage preservation and/or restoration
0 Supports public awareness and cultural development

0 Supports ﬁlm and ﬁlm makers

O Other

 

 

I Previous Page I I Next Page I

 

 

177

 

 

 

Donations to Arts and Cultural Causes

During the last 12 months, have you donated any money to an arts, heritage or
cultural organization (or cause) including but not limited to organizations that you
were/are a member?

0 Yes 0 No

Please check the one box that best describes the amount that you donated during the
last 12 months to arts, cultural or heritage organizations or causes.

0 less than $100 O$550-$699 O $2,000-$4,999
0 $100-$199 O$700-$849 O $5,000-$9,999
0 $200-$299 O$850-$999 O $10,000-$19,999
0 $300-$399 O$1,000-$1,499 0 $20,000 or more
0 $400-$549 O$1,500-$l,999

What did your donations support? Check all that apply.

CI Local arts, cultural or heritage causes, organizations or programs

CI State arts, cultural or heritage causes, organizations or programs

CI National arts, cultural or heritage causes, organizations or programs

CI International arts, cultural or heritage causes, organizations or programs

 

 

I Previous Page ] f Next Page I

 

 

 

178

 

 

 

Volunteering to Arts and Cultural Organizations

During the last 12 months, have you volunteered any time to an arts, heritage or
cultural organization (or cause) including but not limited to organizations that you
were/are a member?

0 Yes 0 No

With which causes, organizations or programs did you volunteer? Check all that
apply.

CI Local arts, cultural or heritage causes, organizations or programs

E] State arts, cultural or heritage causes, organizations or programs

El National arts, cultural or heritage causes, organizations or programs

CI International arts, cultural or heritage causes, organizations or programs

 

 

[Previous Page ] I Next Page ]

 

 

 

179

 

 

 

Your Sources of Information

How oﬁen do you use the following information sources to learn about live theater,
dance performances, and music concerts?

leuulI} Seldom
O O O

 

 

 

TV 0
Magazine 0 O O 0
Newspapers 0 O O 0
Radio 0 O O O
W bs'tes s ecialized fo online
tic etIng p r O O O 0
Websites of erforrnin art
organization? or centergs O O O O
A intances, friends, famil or
re 333% y 0 O O O
Art-related publications 0 O O O
Leaﬂets/Brochures O O O 0
Do you use the intemet to research and/or purchase products?
0 Yes 0 No
I Previous Page I I Next Page ]

 

 

 

180

 

 

 

Use of Internet

How often do you use the intemet to research the following art and culture related
products?

   

O O O 0

Art (e.g., paintings, photos)

Tickets for rforrnin arts e. .,
concerts) pe g ( g 0 O ' O O
Antiques and collections 0 O O 0

How often do you use the intemet to purchase the following art and culture related
products?

0 O O 0

Art (e.g., paintings, photos)

 

 

Tickets for erforrnin arts e. .,

concerts) p g ( g 0 O O O

Antiques and collections 0 O O O
I Previous Page I I Next Page ]

 

 

 

181

 

 

 

Your Information
Please, indicate your gender:
0 Male 0 Female
How old are you?

0Under18 018-24 025-34 035-44 045-54 055-64
0650rolder

What is your sexual orientation?

0 Straight/Heterosexual
0 Gay/Lesbian
0 Other

0 Prefer not to answer
What is the highest level of formal education you have achieved?

0 Less than high school

0 Completed some high school

0 Completed high school

0 Some college

0 2-year college degree

04-year college degree

0 Completed some graduate courses

0 Master’s degree (MA, MS, MBA, MFA, MPA, etc.)

0 Doctoral degree (PhD, EdD, DVM, MD, DO, DD, etc.)

Are you or your spouse afﬁliated with Michigan State University (MSU)? If so,
what is the nature of your afﬁliation? (Please check all that apply).

CI No

El Yes, employed as faculty at MSU
[I Yes, employed as staff at MSU
D Yes, alumni of MSU

El Yes, a current student at MSU

El Yes, an MSU donor

D Yes, retired from MSU

 

182

 

 

 

 

Which is the country of your permanent residence?

0 United States (U.S.)
0 Canada

0 Mexico

0 Other

Please indicate your email address.
(Email address will only be used to notify the winner of the $100 Wharton Center
gift certiﬁcate and the dinner for two at the Kellogg Center’s State Room.)

L I

 

What is your zipcode/postal code of your permanent residence?

I I

 

 

 

I Previous Page I I Next Page J

 

 

 

183

 

 

 

 

Your Family Information
How many persons currently reside in your household, including yourself?

0 I live alone 0 4 persons
0 2 persons 0 5 persons
0 3 persons 0 6 persons or more

How would you describe your current family status?

0 Single without children

0 Single with children living at home

0 Single with children no longer living at home

0 Married/Partnered without children

0 Married/Partnered with children living at home

0 Married/Partnered with children no longer living at home

How many children are living at home with you?

0 1 child 0 4 children
0 2 children 0 5 children
0 3 children 0 6 children or more

Which are the age categories of children currently living in your household?
Check all that apply.

El Less than 6 yours old
[:1 6 - 12 years old
I313 -17 years old

[I 18 years old or more

 

 

IPrevious Page I I Next Page I

 

 

 

184

 

 

 

Art and Culture Related Experiences of Your Child/Children

During the last 12 months, have you taken any children residing in your household
to arts or cultural performances or exhibits?

0 Yes 0 No

To how many arts or cultural performances or exhibits have you taken your children
during the last 12 months?

0 l performance/exhibit

0 2-4 performances/exhibits

0 5-9 performances/exhibits

0 10 or more performances/exhibits

During the last 12 months, did any of the children residing in your household
participate in any elementary, middle or high school student performances or
exhibits (e.g., school play, choir performance)?

0 Yes 0 No

During the last 12 months, have any children residing in your household taken any
type of visual (e.g., painting or drawing), performing (e.g., dance, singing, or
musical instrument), or literary arts or crafts classes or lessons?

0 Yes 0 No

What types of classes did children living in your household take during the last 12
months? Check all that apply.

CI Visual arts

CI History/Appraisal

[:1 Performing arts

CI Music (vocal/instruments)

CI Literary arts

CI Applied arts (e.g., architecture, design, and genealogy and archaeology)
CI Crafts

In total, how much did you pay for arts education classes for your children during
the last 12 months?

0 Less than $100
0 $100 - $199
0 $200 - $499
0 $500 - $999
0 $1,000 - $1,999
0 $2,000 - $4,999
0 $5,000 or more

 

185

 

 

 

Your Art-Related Experience

When you were a child (up to 18 yrs old) how oﬁen did your parents or other adults
take you to arts or cultural performances or events?

0 Never 0 Rarely 0 Occasionally 0 Frequently 0 Very frequently

When you were a child did you take any type of visual (e. g., painting and drawing),
performing (e. g., dance, singing, and musical instruments) or literary arts or crafts
lessons or classes?

0 Yes 0 No

What type of classes did you take when you were a child?

El Visual arts

CI Performing arts

CI Music (vocal/instruments)
1] Literary arts

CI Crafts

During the last 12 months have you taken any type of visual (e.g., painting and
drawing), performing (e. g., dance, singing, and musical instruments) or literary
arts or crafts classes or lessons?

0 Yes 0 No

What type of art or cultural related classes did you take during the last 12 months?

CI Visual arts

El History/Appraisal

[:1 Performing arts

El Music (vocal/instrumental)

CI Literary arts

D Applied arts (e. g., architecture, design, genealogy and archaeology)

In total, how much did you pay for your arts education classes during the last 12
months?

0 Less than $100
0 $100 - $199
0 $200 - $499
0 $500 - $999
0 $1,000 - $1,999
0 $2,000 - $4,999
0 $5,000 or more

I Previous Page I I Submit I

 

 

 

 

186

 

 

 

Appendix B. Descriptive Statistics of Performing Arts Respondents

Appendix Bl. Socioeconomic Characteristics of the Respondents

 

 

Frequency %
Gender 4,685
Male 1,245 26.6
Female 3,440 73.4
Age 4,722
Under 25 358 7.6
25 — 35 674 14.3
35 —- 44 952 20.2
45 — 54 1,391 29.4
55 — 64 1,077 22.8
65 or older 270 5.7
Sexual orientation 4,703
Straight/Heterosexual 4,2 54 90.5
Gay/Lesbian 1 16 2.5
Other 2 5 0.5
Prefer not to respond 303 6.5
Education 4,729
Less than or completed high school 193 4.1
Some or 2-year college degree 1,433 30.4
4-year college degree 1,237 26.1
Completed some graduate courses 535 1 1.3
Master's degree 983 20.3
Doctoral degree 343 7_3
Employment status a 5,158
Full time employed 3,045 64.4
Part-time employed 678 14.3
Homemaker (care’s for family/house) 367 7.8
Unemployed 56 l .2
Retired 603 12.8
Student 409 3.7

 

a Multiple responses.

187

Appendix B1. (cont’d)

 

 

Frequency %
Racial type 4,655
Caucasian or White 4,439 95.4
African American or Black 30 1.7
Asian or Paciﬁc Islander 65 1.4
Mixed 56 1.2
Native American or Aleutian Eskimo 15 0.3
Hispanic or Latino origin or descent 4,606
Hispanic or Latino 88 1.9
Non Hispanic or Latino 4,513 98.1
Family annual Income in 2006 4,232
Less than $30,000 360 8.6
$30,000 - $49,999 579 13.7
$50,000 - $74,999 910 21.5
$75,000 - $99,999 332 20.8
$100,000 - $ 149,999 986 23.2
$150,000 - $199,999 290 6.9
$200,000 or more 225 5.3
Country of permanent residence 4,743
United States (U.S.) 4713 99.5
Canada [0 0.2
Other 15 0.3

 

188

Appendix B2. Household Information for the Online Survey Respondents

 

 

Frequency %
Number of people in household 4,718
1 live alone 669 14.2
2 persons 1,965 41.6
3 persons 811 17.2
4 persons 857 18.2
5 or more persons 416 8.8
Family status 4,744
Single without children 865 18.2
Single with children living at home 243 5.2
Single with children no longer living at home 239 5.1
Married/partnered without children 619 13.0
Married/partnered with children living at home 1,661 35.0
Married/partnered with children no longer living at
home 1,] 17 23.5
Number of children living at home a 1,878
1 child 736 39.2
2 children 739 42.0
3 children 268 14.3
4 children 65 3.5
5 or more children 20 1.0
Age categories of children living at home a b 2,574
Less than 6 years old 504 26.8
6 — 12 years old 709 37.7
13 — 17 years old 749 39.8
18 years old or more 612 32.5

 

a This only includes respondents who have children living at home (40.2%), either single (5.2%) or
married/partnered (35.0%).

Multiple responses.

189

Appendix B3. Consumption Pattern of Live Performances during the Last 12 Months

 

 

Frequency %
Attendance at live performances in the last 12 months 4,744
Attended a live performance(s) 4,434 93.5
Did not attend a live performance 310 6.5
Number of live performances attended 4,257
1 performance 598 14.0
2 performances 347 19.9
3 performances 626 14.7
4-5 performances 932 21.8
6-9 performances 729 17.1
10 performances or more 535 12.5
Types of performances attended a 4,434
Play 2,437 55.0
Musical 3,493 78.8
Opera 319 7.2
Ballet 392 8.8
Modern dance 256 5.8
Ethnic dance 152 3.4
Folk dance 35 1.9
Folk and ethnic concert 442 10.0
Jazz concert 673 15.2
Blues concert 419 9.5
Symphony 763 l 7.2
Country concert 471 10.6
New and experimental concert 1 16 2.6
Rock concert 1,059 23.9
Hip hop concert 69 1.6
World concert 129 2.9
Other 1,067 24.1
Attended live performances in a state or country other than 4 415
resrdence ’
Yes 1,210 27.4
No 3,205 72.6
Attended free live performances 4,434
Yes 858 19.4
No 3,575 80.6

 

a Multiple Responses.

190

 

Appendix B4. Formal Arts Education of Respondents

 

 

Frequency Percentage

Attended performances during childhood 4,727

Never 310 17.1

Rarely 1,474 31.2

Occasionally 1,549 32.8

Frequently 691 14.6

Very frequently 200 4.3
Took art classes during childhood 4,744

Yes 3,515 74.1

No 1,229 25.9
Type of art classes taken during childhood 3 b 3509

Visual arts 340 23.9

Performing arts 1,502 42.8

Music (vocal/instruments) 2,935 83.6

Literary arts 380 10.8

Crafts 1,241 35 .4

 

a This only includes respondents who had taken art classes during their childhood.
b Multiple responses.

191

 

Appendix B5. Current Art Education of the Respondents

 

 

Frequency %

Took art classes during the last 12 months 4,744
Yes 742 '5-6
No 4,002 84.4

Type of art classes taken in the last 12 months a b 728
Visual arts 318 43.7
History/appraisal 2 1 2.9
Performing arts 183 25.2
Music (vocal/instruments) 219 30.1
Literary arts 48 6.6
Applied arts (e.g., architecture, design, and genealogy) 1 '9 16.4

Total cost of art classes taken in the last 12 months a 723
Less than $100 283 38.9
$100-$199 160 22.0
$200 - $499 151 20.7
$500 - $999 80 11.0
$1,000 - $1,999 31 4.3
$2,000 or more 23 3.1

 

a This only includes respondents who have taken art classes during the last 12 months .

b Multiple responses.

192

Appendix B6. Art Product Purchase Experience of the Respondents

 

 

Frequency %
Visual arts purchases during the last 12 months 4,744
Purchased visual arts 2,967 62.5
Did not purchase visual arts 1,777 37.5
Types of visual arts purchased a b 2,960
Paintings 1,231 41.6
Books 1,786 60.3
Photographs 1,098 37 .1
Drawings 322 10.9
Prints 1,286 43.4
Sculptures 323 I 1-1
Crafts 1,821 61.5
Folk or traditional art 470 15.9
Other types of visual art 231 7.8
Channels where visual arts were purchased a b 2,948
Art fair 1.783 60.5
Auction 334 l 1.3
Art gallery / bookstore 1,61 1 54.5
Museum store 745 25.3
Catalog 539 18.3
Web / intemet 333 23-3
Direct from an artist (no intermediary) 876 29.7
Commissioned a special artistic work 97 3. 3
Other ways 476 16.1

 

a This only includes respondents who purchased visual arts during the previous 12 months.
b Multiple responses.

193

Appendix B7. Cultural Membership Status of the Respondents

 

 

Frequency %
Afﬁliated with arts, heritage, or culturally related organizations 4,744
Afﬁliated 1,032 21.8
Not afﬁliated 3,712 78.2
Primary mission/purpose of the organization(s) for which
respondents were a member8| 1’030
Supports and encourages visual arts and artists 1 13 1 1.0
Supports and encourages performing arts and artists 324 31.5
Supports and encourages libraries 94 9.1
Supports and encourages arts education 61 5.9
Supports and encourages public art 31 3.0
Supports and encourages heritage preservation and/or
restoration 1 52 [4'8
Supports public awareness and cultural development 1 10 10.7
Supports ﬁlm and ﬁlmmakers 20 1.9
Other 125 12.1

 

a This only includes respondents who are afﬁliated with heritage or cultural organizations.

Appendix B8. Volunteering Status to Arts and Cultural Organizations of the Respondents

 

 

Frequency %
Volunteered to arts, heritage or cultural organizations 4,744
Did volunteer 344 17.8
Did not volunteer 3,900 82.2
Types of causes, organizations or programs volunteer service
was - a 842
provrded
Local arts, cultural or heritage causes, organizations or 93 6
programs 788 ‘
State arts, cultural or heritage causes, organizations or 12 9
programs 109 '
National arts, cultural or heritage causes, organizations or 4 8
programs 40 '
lntemational arts, cultural or heritage causes, organizations 3 6
or programs 30 '

 

a This only includes respondents who volunteered their time to an arts, heritage or cultural organization or

cause in the last 12 months.
b .
Multrple responses.

194

 

 

Appendix B9. Donation Activity to Arts and Cultural Organizations and Causes of the

 

 

Respondents
Frequency %
Donations during the previous 12 months to arts, heritage or
cultural organizations 4,744
Did make a donation 1,830 38.6
Did not make a donation 2,914 61.4
Amount donated during the previous 12 months a 1,771
Less than $100 561 31.7
$100 - $199 499 28.2
$200 - $399 316 17.8
$400 - $999 225 12.7
$1,000 or more 170 9.6
Types of causes supported a b 1,302
Local arts, cultural or heritage causes, organizations or
programs 1’527 84'7
State arts, cultural or heritage causes, organizations or
4 51 25.0
programs
National arts, cultural or heritage causes, organizations or 18 3
programs 330 '
lntemational arts, cultural or heritage causes, organizations 6 1
or programs I 10 '

 

 

a This only includes respondents who made a donation to an arts, heritage or cultural organization or cause
during the last 12 months.

Multiple responses.

195

 

Appendix C. Comparison of Heavy and Light Consumer of the Performing Arts

Appendix C 1. Comparison of Socioeconomics, Household lnforrnation and Art-related
Experiences between Heavy and Light Consumers of the Performing Arts

 

2

 

 

Heavy Light X p-value
(30.6%) (69.4%)
% %
Gender 7.610 0.006M
Male 30.3 26.1
Female 69.7 73.9
Age 143.111 o.ooo***
Under 35 12.1 25.9
35 - 44 17-6 21.5
45 - 54 33.5 27.9
55 - 64 28.1 20.3
65 or older 8.7 4.4
Education 87.415 0000*"
Completed high school or less 2.6 40
Some or 2-year college degree 21.5 32.7
4-year college degree 26.2 26.7
Completed some graduate courses 13.1 1 1_()
Master’s degree 25.8 19, 5
Doctoral degree 10.7 6. 1
Income 149.333 0.000***
Less than $50,000 13.1 24.5
$50,000 - $99,999 38.9 43.7
$100,000 - $149,999 27.2 22.6
$ 150,000 or more 208 9.2
Employment 7.651 0006* *
Full-time employment 59.7 64.2
Other 40.3 3 5.3
**, *** indicates signiﬁcance levels at 0.01 and 0.001, respectively.

196

Appendix C 1. (cont’d)

 

 

 

Heavy Light )8 p-value
(30.6%) (69.4%)
% %

Number of people in household 7.205 0.125
1 live alone 13.3 14.6
2 persons 44.5 40. 5
3 persons 15.6 17.9
4 persons 17.9 18.5
5 persons or more 8.6 8. 5

Marital status 13.353 0000*»:4:
Single 24.5 30_1
Married 75.5 69.9

Children’s status 22.800 0.000***
No Children 26.8 33.1
Children living at home 40.3 402
Children no longer living at home 32.8 26.7

Attended live performances in a different state or country 96-942 0000*"
Yes 38.2 232
N0 61.8 76.8

Art education during childhood 0.591 0.442
Yes 75.2 74. 1
No 24.8 259

Recent art education 15.159 0000* **
Yes 19.6 14.7
No 80.4 85.3

**, *** indicates signiﬁcance levels at 0.01 and 0.001, respectively.

197

 

 

Appendix C1. (cont’d)

 

 

Heavy Light X2 p-value
(30.6%) (69.4%)
% %
Purchase art products 33 .421 0.0001: :1: :1:
Yes 70.3 61.0
No 29.7 39_0
Cultural membership 96.474 0000* :1: :1:
Yes 32.6 13.7
No 67.4 81.3
Volunteer status 54.452 0.0004: :1: :1:
Yes 25.5 15.3
No 74.5 84.2
Donation activity 269.666 0_000* :1: :1:
Yes 61.2 3 3.9
N" 33-3 66.1

 

**, *** indicates signiﬁcance levels at 0.01 and 0.001, respectively.

198

BIBLIOGRAPHY

199

BIBLIOGRAPHY

Americans for the Arts (2007). Arts & economic prosperity 111. Retrieved January 2009,
from http://www.pinellasarts.org/pdf/economic_pros__cty.pdf.

Andreasen, A. R., & Belk, R. W. (1980). Predictors of attendance at the performing arts.
Journal of Consumer Research, 7(2), 112-120.

Andrews, D., Nonnecke, B., & Preece, .1. (2003). Electronic survey methodology: a case
study in reaching hard-to-involve intemet users. International Journal of Human-
Computer Interaction, 16(2), 185-210.

Baumol, W. J ., & Bowen, W. G. (1966). Performing arts: the economic dilemma. a study
0 problems common to theatre, opera, music and dance. New York: Twentieth
Century Fund.

Bakken, S. K. (2005). Use of chi-squared automatic interaction detector in the prediction
of vocational rehabilitation outcomes among veterans with substance use
disorders. Unpublished doctoral dissertation, Univeristy of Wisconsin-Madison.

Bagdadli, S., & Arrigoni, L. (2005). Strategic positioning of the Venice Biennial:
analysing the market for periodic contemporary art exhibitions. International
Journal of Arts Management, 7(3), 734-752.

Bergonzi, L., & Smith, J. (1996). Effects of arts education on participation in the arts.
Santa Ana, CA: National Endowment for the Arts.

Bernstein, J. S. (2007). Arts marketing insights: the dynamics of building and retaining
performing arts audiences. San Francisco, CA: John Wiley & Sons, Inc.

Best, R. J. (2000). Market-based management: strategies for growing customer value and
proﬁtability (2nd ed.). Upper Saddle River, NJ: Prentice-Hall.

Biggs, D., De Ville, B., & Suen, E. (1991). A method of choosing multiway partitions for
classiﬁcation and decision trees. Journal of Applied Statistics, 18(1), 49-62.

Bojanic, D. C. (2007). Customer proﬁle of the “carryout” segment for restaurants.
International Journal of Contemporary Hospitality Management, 19(1), 21 -3 1.

Borgonovi, F. (2004). Performing arts attendance: an economic approach. Applied
Economics, 36(17), 1871-1886.

Bouder-Pailler, D. (1999). A model for measuring the goals of theatre attendance.
International Journal of Arts Management, 1(2), 4-15.

200

Bowen, J. T. (1998). Marketing segmentation in hospitality research: no longer a
sequential process. International Journal of Contemporary Hospitality
Management, 10(7), 289-296.

Broderick, A., & Mueller, R. (1999). A theoretical and empirical exegesis of the
consumer involvement construct: the psychology of the food shopper. Journal of
Marketing Theory and Practice, 7(4), 97-108.

Brown, A. S. (2007a). A segmentation model for performing arts ticket buyers. Retrieved
March 2009, from
http://www.wolfbrown.com/mups_down1oads/MUP_Ticket Buyer_Segmentation

Reportpdf.

Brown, A. S. (2007b). A segmentation model for donors to 12 university presenting
programs. Retrieved March 2009, from
http://www.wolfbrown.com/mups_downloads/MUP_Donor_Segmentation_Repor

t.pdf.

Byrd, E. T. (2003). An analysis of variables that Inﬂuence stakeholder participation and
support for sustainable tourism development in rural North Carolina.
Unpublished Ph.D. Dissertation, North Carolina State University.

Caldwell, M., & Woodside, A. G. (2003). The role of cultural capital in performing arts
patronage. International Journal of Arts Management, 5(3), 34-51.

Chen, J. S. (2003a). Developing a travel segmentation methodology: a criterion-based
approach. Journal of Hospitality & Tourism Research, 27(3), 310-327.

Chen, J. S. (2003b). Market segmentation by tourists’ sentiments. Annals of Tourism
Research, 30(1), 178-193.

Chung, K. Y., Oh, S. Y., Kim, S. S., & Han, S. Y. (2004). Three representative market
segmentation methodologies for hotel guest room customers. Tourism
Management, 25(4), 429-441.

Clopton, S. W., Stoddard, J. E., & Dave, D. (2006). Event preferences among arts
patrons: implications for market segmentation and arts management. International

Journal of Arts Management, 9(1), 48-77.

Colbert, F. (1997). Changes in marketing environment and their impact on cultural policy.
Journal of Arts Management, Law, and Society, 27(3), 177-186.

Colbert, F. (2003). Entrepreneurship and leadership in marketing the arts. International
Journal of Arts Management, 6(1), 30-39.

201

Colbert, F ., d' Astous, A., & Parmentier, M. (2005). Consumer perception of private
versus public sponsorship of the arts. International Journal of Arts Management,
8(1), 48-59.

Conway, T., & Whitelock, J. (2007). Relationship marketing in the subsidized arts: the
key to a strategic marketing focus? European Journal of Marketing, 41(1), 199-
222.

Couper, M. P. (2000). Web surveys. Public opinion quarterly, 64(4), 464-494.

Croes, H. (2008). Accommodation portfolio and market differentiation: the case of
Aruba. Tourism Review, 56(2), 185-197.

Cuadrado, M., & Molla, A. (2000). Grouping performing arts consumers according to
attendance goals. International Journal of Arts Management, 2(3), 54-60.

d' Astous, A., Legoux, R., & Colbert, F. (2004). Consumer perceptions of promotional
offers in the performing arts: an experimental approach. Canadian Journal of
Administrative, 21(3), 242-254.

DiMaggio, P., & Mukhtar, T. (2004). Arts participation as cultural capital in the United
States, 1982—2002: Signs of decline? Poetics, 32(2), 169-194.

Dolnicar, S. (2002). A review of data-driven market segmentation in tourism. Journal of
Travel & Tourism Marketing, 12(1), 1-22.

F avaro, D., & Frateschi, C. (2007). A discrete choice model of consumption of cultural
goods: the case of music. Journal of Cultural Economics, 31(3), 205-234.

Frank, B., & Geppert, K. (2004). Are small recipients overlooked by sponsors? an
empirical note. Journal of Cultural Economics, 28(2), 143-156.

Gainer, B. (1993). The impact of gender on marketing the arts. Paper presented at the
Second lntemational Conference on Arts Management, Paris.

Galguera, L., Luna, D., & Méndez, M. (2006). Predictive segmentation in action: using
CHAID to segment loyalty card holders. International Journal of Market
Research, 48(4), 459-479.

Gan, L. L., Maysami, R. C., & Koh, H. C. (2008). Singapore credit cardholders:
ownership, usage patterns, and perceptions. Journal of Services Marketing, 22(4),
267-279.

Goldsmith, R. E., Flynn, L. R., & Bonn, M. (1994). An empirical study of heavy users of
travel agencies. Journal of Travel Research, 33(1), 38-43.

202

Goldsmith, R. E., & Litvin, S. W. (1999). Heavy users of travel agents: a segmentation
analysis of vacation travelers. Journal of Travel Research, 38(2), 127-133.

Gomez, M., & Benito, N. R. (2008). Manufacturer's characteristics that determine the

choice of producing store brands. European Journal of Marketing, 42(1), 154-
177.

Guiltinan, J. P., & Paul, GW. (1991). Marketing management. strategies and programs
(4th ed.). New York: McGraw-Hill.

Gursoy, D., & Chen, J. S. (2000). Competitive analysis of cross cultural information
search behaviour. Tourism Management, 21(6), 583-590.

Hager, M. A., & Pollak, T. H. (2002). The capacity of performing arts presenting
organizations. Washington, DC: the Association of Performing Arts Presenters,
and the Urban Institute.

Harvey, J. W. (1990). Beneﬁt segmentation for fund raisers. Journal of the Academy of
Marketing Science, 18(1), 77-86.

Hair, I. R, Anderson, R. E., Tatham, R. L., & Black, W. C. (1998). Multivariate data
analysis (5‘'1 ed.). Upper Saddle River, NJ: Prentice Hall.

Hoare, R. (2004). Using CHAID for classification problems. The New Zealand Statistical
Association Conference, Wellington, New Zealand.

Hsu, C. H., & Kang, S. K. (2007). CHAID-based segmentation: international visitors' trip
characteristics and perceptions. Journal of Travel Research, 46(2), 207-216.

Hudson, S. (2000). The segmentation of potential tourists: constraint differences between
men and women. Journal of Travel Research, 38(4), 363-368.

Hume, M. (2008). Developing a conceptual model for repurchase intention in the
performing arts: the roles of emotion, core service and service delivery.
International Journal of Arts Management, 10(2), 40-55.

Hume, M., & Mort, G. S. (2008). Satisfaction in performing arts: the role of value?
European Journal of Marketing, 42(3), 311-326.

Hume, M., Mort, G. S., Liesch, P. W., & Winzar, H. (2006). Understanding service
experience in non-proﬁt performing arts: implications for operations and service

management. Journal of Operations Management, 24(4), 304-324.

Hunt, SD. (2002). Foundations of marketing theory: toward a general theory of
marketing, ME. Sharpe, Armonk, NY.

203

Huntington, C. (2007). Reevaluating segmentation practices and public policy in classical
performing arts marketing: a macro approach. Journal of Arts Management, Law
and Society, 37(2), 127-141.

Jackson, R., & Wang, P. (1994). Strategic database marketing, NTC Books,
Lincolnwood, IL.

Jain, K., & Srinivasan, N. (1990). An empirical assessment of multiple
operationalizations of involvement. Advances in Consumer Research, 17(1), 594-
602.

Kass, G. V. (1980). An exploratory technique for investigating large quantities of
categorical data. Journal of the Royal Statistical Society. Applied Statistics, 29(2),
119-127.

Kastenholz, E., Carneiro, M. J., & Eusebio, C. (2005). The impact of socio-demographics
on tourist behavior analyzing segments of cultural tourists visiting Coimbra.
Paper presented at the lntemational Conference Theoretical Advances in Tourism
Economics.

Kim, H., Cheng, C., & O’ Leary, J. T. (2007). Understanding participation patterns and
rends in tourism cultural attractions. Tourism Management, 28(6), 1366-1371.

Kimmel, J. R. (1995). Art and tourism in Santa Fe, New Mexico. Journal of Travel
Research, 33(28), 28-30.

Knobbe, A. A. (2003). Marketing efforts of performing arts venues. Unpublished Master
Dissertation, Michigan State University.

Koch, R. (1999). The 80/20 principle: the secret of achieving more with less: Doubleday.

Kotler, P. (1972). A generic concept of marketing. The Journal of Marketing, 36(2), 46-
54.

Kotler, P. (1979). Strategies, for introduction marketing into nonproﬁt organizations.
Journal of Marketing, 43(1), 37-44.

Kotler, P. (1991). Marketing management: analysis, planning, implementation and
control (7 ed.). Englewood Cliffs, NJ : Prentice-Hall.

Kotler, P. (2005). According to Kotler: the world 's foremost authority on marketing
answers your questions. New York: American Management Association.

Kotler, P., & Armstrong, G. (1987). Marketing an introduction. Englewood Cliffs, NJ:
Prentice-Hall.

204

Kotler, P., & Armstrong, G. (2006). Principles of marketing (1 1th ed.). Upper Saddle
River, NJ: Pearson Prentice Hall.

Kotler, P., & Scheff, J. (1997). Standing room only: Strategies for marketing the
performing arts: Harvard Business School Pr.

Laurent, G., & Kapferer, J. N. (1985). Measuring consumer involvement proﬁles.
Journal of Marketing Research, 22(1), 41-53.

Legohérel, P., & Wong, K. (2006). Market segmentation in the tourism industry and

consumers' spending: what about direct expenditures. Journal of Travel &
Tourism Marketing, 20(2), 15-29.

Levy, M., & Weitz, B. A. (1992). Retailing Management. Chicago, IL: Irwin Press.
Levy-Garboua, L., & Montmarquette, C. (2002). The demand for the arts: Cirano.

Litvin, S. W. (2000). Revisiting the heavy-user segment for vacation travel marketing.
Journal of Vacation Marketing, 6(4), 346-356.

Liu, Y. (2007). The long-tenn impact of loyalty programs on consumer purchase
behavior and loyalty. Journal of Marketing, 71(4), 19-35.

Lu, W. (2005). The analysis of bases of market segmentation. Chinese Business Review,
4(10), 71-72.

Magidson, J. (1994). The CHAID approach to segmentation modelling: chi-squared
automatic interaction detection. in R. P. Bagozzi (ed.), In advanced methods of
marketing research (pp. 118-159). Cambridge, MA: Blackwell.

Magidson, J., & Vermunt, J. K. (2005). An extension of the CHAID tree-based
segmentation algorithm to multiple dependent variables. In Weihs C, Gaul W
(eds), Classiﬁcation: The Ubiquitous Challenge. Springer: Heidelberg.

Martorella, R. (1999). Government and corporate ideologies in support of the arts. In R.
Moulin (ed.), Sociologie de 1’ art. Paris: L' Harmattan.

Mazanec, J. A. (1984). How to detect travel market segments: a clustering approach.
Journal of Travel Research, 23(1), 17-21.

Mazanec, J. A. (1992). Classifying tourists into market segments:A neural network
approach. Journal of Travel & Tourism Marketing, 1(1), 39-59.

McArthur, S., & Balasubramanian, R. (2003). Beyond demographic segmentation: A

behavioural approach to performing arts marketing. Paper presented at the
Massey University seminar presentation.

205

McCarthy, B. (2006). Building and sustaining trust in networks: lessons from the
performing arts. Irish Marketing Review. Retrieved January 2009, from
http://ﬁndarticles.com/p/artic1es/mi_qa5500/is_20060 1 /ai_n2 1 403 75 1/?tag=conte
nt;coll.

McCarthy, K. F., Brooks, A., Lowell, J., & Zakaras, L. (2001). The performing arts in a
new era. Santa Monica: RAND Corp.

McCarthy, K. F ., & J innett, K. (2001). A new framework for building participation in the
arts. Santa Monica, CA: RAND.

Mejon, J. C., F ransi, E. C., & Johansson, A. T. (2004). Marketing management in cultural
organizations : a case study of catalan museums. International Journal of Arts
Management, 6(2), 11-22.

Mencarelli, R., & Pulh, M. (2006). Positioning the supply of live performance: innovative
managerial practices relating to the interaction of spectator, performance and
venue. International Journal of Arts Management, 8(3), 19-29.

Mitchell, A. (1984). Nine American lifestyles: Values and Societal Change. Futurist,
18(4), 4-14.

Molera, L., & Albaladejo, I. P. (2007). Proﬁling segments of tourists in rural areas of
south-eastem Spain. Tourism Management, 28(3), 757-767.

Montgomery, S. S., & Robinson, M. D. (2005). Take me out to the opera: are sports and
arts complements? Evidence from the performing arts research coalition data.
Paper presented to the 8‘h lntemational Conference on Arts and Cultural
Management, HEC Montreal.

Muller, T., & Cleaver, M. (2000). Targeting the CANZUS baby boomer explorer and
adventurer segments. Journal of Vacation Marketing, 6(2), 154-169.

Murray, K. B. (1991). A test of services marketing theory: consumer information
acquisition activities. Journal of Marketing, 55(1), 10-25.

Myers, J. H. (1996). Segmentation and positioning for strategic marketing decisions.
Alibris, NV: Cengage.

Nichols, B. (2003). Demographic characteristics of arts attendance, 2002, NEA Research
Division. Note #82, National Endowment for the Arts. Washington DC.

Nichols, B. (2005). Arts and leisure activities: evidence from the 2002 survey of public

participation in the arts. Note #89, National Endowment for the Arts. Washington
DC.

206

Nichols, B. (2006a). Consumer spending on performing arts: outlays ﬂat for 2005; non-
spectator categories show growth. Research Note #91, National Endowment for
the Arts. Washington DC.

Nichols, B. (2006b). Performing arts companies in 2002. Research Note #93, National
Endowment for the Arts. Washington DC.

Nichols, B. (2006c). Volunteers with arts or cultural organizations: a 2005 proﬁle.
Research Note #95, National Endowment for the Arts. Washington DC.

Nielsen, R. P., McQueen, C., & Nielsen, A. B. (1974). Performing arts audience
segments. Journal of the Academy of Marketing Science, 2(4), 602-609.

Notzke. (1999). Indigenous tourism development in the arctic. Annals of Tourism
Research, 26(1), 55-76.

Ouellet, J.-F., Savard, M.-A., & Colbert, F. (2008). The personality of performing arts
venues: developing a measurement scale. International Journal of Arts
Management, 10(3), 49-59.

Palakurthi, R. R., & Parks, S. J. (2000). The effect of selected socio-demographic factors
on lodging demand in the USA. International Journal of Contemporary
Hospitality Management, 12(2), 135-142.

Park, M., Yang, X., Lee, B., Jang, H., & Stokowski, P. (2002). Segmenting casino
gamblers by involvement proﬁles: a Colorado example. Tourism Management,
23(1), 55-65.

Park, D.-B., & Yoon, Y.-S. (2009). Segmentation by motivation in rural tourism: a
Korean case study. Tourism Management, 30(1), 99-108.

Perreault, W. D. (1980). Model-free approach for analysis of complex contingency data
in survey research. Journal of Marketing Research, 1(4), 503-515.

Pink, B., & Matthews, M. (2003). A measure of culture: cultural experiences and cultural
spending in New Zealand. Published report by Statistics New Zealand and
Ministry for Culture and Heritage. Retrieved March 2009, from
http://www.stats.govt.n2/reports/analytical-reports/measure-of-culture.aspx.

Prentice, R., & Andersen, V. (2003). Festival as creative destination. Annals of Tourism
Research, 30(1), 7-30.

Reid, R. D., & Bojanic, D. C. (2005). Hospitality marketing management (4 ed.).
Hoboken, New Jersey: Wiley &Sons.

207

Rentschler, R., Radboume, j., Carr, R., & Rickard, J. (2002). Relationship marketing,
audience retention and performing arts organizations viability. International
Journal of Nonproﬁt and Voluntary Sector Marketing, 7(2), 1 18-130.

Reynolds, T. J ., & Olson, J. C. (2001). Understanding Consumer Decision Making.
Mahwah, New Jersey: Lawrence Erlbaum Associates.

Rich, J. D. (2001). The market in marketing culture and the arts. in Colbert, F. (Ed.), (2
ed., pp. 54-57). Montreal: Presses HEC.

Richards, G. (1996). Cultural tourism in context. in Cultural Tourism in Europe:
Wallingford: CABI.

Roan, N. A., Jones, J ., & Roan, W. W. (2006). Association of performing arts presenters,
conversations with the ﬁeld report. Retrieved
March 2009, from http://www.artspresenters.org/about/convﬂdreport.pdf.

Robbins, J. E., & Robbins, S. S. (1981). Museum marketing: identiﬁcation of high,
moderate, and low attendee segments. Academy of Marketing Science Journal,
9(1), 66-75

Royce, A. P. (2004). Anthropology of the performing arts: artistry, virtuosity, and
interpretation in a cross-cultural perspective. Singer, Milton: Walnut Creek: Alta
Mira Press.

Ruﬁn, R. (2007). Sales growth of Spanish tourist ﬁrms: some implications of Gibrat’s
Law on marketing management. Tourism Management, 28(3), 788-805.

Scheff, J ., & Kotler, P. (1996). Crisis in the arts: The marketing response. California
Management Review, 39(1), 28-52.

Silberberg, T. (1995). Cultural tourism and business opportunities for museums and
heritage sites. Tourism Management, 16(5), 361-365.

Smith, W. (1956). Product differentiation and market segmentation as alternative
marketing strategies. Journal of Marketing, 21 (3), 3-8.

Snedecor, G. W., & Cochran, w. 'G. (1989). Statistical Methods (8th ed.): Iowa State
University Press.

SPSS (2001). Answer Tree 3.0 user’s guide. Chicago, IL: Author.

Stepchenkova, S., & Morrison, A. M. (2006). The destination image of Russia: from the
online induced perspective. Tourism Management, 27(5), 943-956.

208

Trochim, W. (2001). The Research Mehtods Knowledge Base (2nd ed.). Cincinnati, OH:
Atomic Dog Publishing.

Turrini, A. (2006). Measuring audience addiction to arts: the case of an Italian theatre.
International Journal of Arts Management, 8(3), 43-53.

Urry, J. (1990). The tourist gaze: leisure and travel in contemporary societies: Sage
Publications (CA).

US. Department of Commerce, Bureau of Economic Analysis and US. Census Bureau
(2008). Arts, entertainment, and recreation services (Naics 71) — Estimated year-
to-year percent change in revenue for taxable employer ﬁrms: 1 999 through 2006.
Retrieved March 2009, from
http://www.census.gov/svsd/www/services/sas/sas data/71/2006 NAICS71.pdf.

Van Middelkoop, M., Borgers, A., & Timmerrnans, H. (2003). Inducing heuristic

principles of tourist choice of travel mode: a rule-based approach. Journal of
Travel Research, 42(1), 75-83.

Venugopal, V., & Baets, V. W. (1994). Neural networks and statistical techniques in

marketing research: a conceptual comparison. Marketing Intelligence and
Planning, 12(7), 30-38.

Vermunt, J. K., & Magidson, J. (2003). Latent class models for classiﬁcation.
Computational Statistics and Data Analysis, 41(3-4), 531-537.

Wedel, M., & Kamakura, W. A. (2000). Market segmentation: conceptual and
methodological foundations: Kluwer Academic Pub.

Werck, K., & Heyndels, B. (2007). Programmatic choices and the demand for theatre: the
case of Flemish theatres. Journal of Cultural Economics, 31(1), 25-41.

Wharton Center for Performing Arts (2008). Michigan State University Wharton Center
for Performing Arts. 2007-2008 Annual Report.

Wiggins, J. (2004). Motivation, ability and opportunity to participate: a
reconceptualization of the RAND Model of audience development. International
Journal of Arts Management, 7(1), 22-33.

Willis, K. G., & Snowball, J. D. (2009). Investigating how the attributes of live theatre
productions inﬂuence consumption choices using conjoint analysis: the example
of the national arts festival, South Africa. Journal of Cultural Economics, 33(3),
167-183.

Worsely, A., & Lea, E. (2008). Consumer concerns about food and health. British Food
Journal, 110(11),1106-1118

209

Wright, K. B. (2005). Researching intemet-based populations: advantages and
disadvantages of online survey research, online questionnaire authoring software
packages, and web survey services. Journal of Computer-Mediated
Communication, 10(3). Retrieved September 2009, from
http://icmcindianggdu/vol10/issue3/wmpthtml.

Wyner, G. A. (1995). Segmentation analysis, then and now. Marketing Research: A
Magazine of Management and Application, 7(1), 40-41.

Yu, L., & Wang, L. (2007). Mining customer preferences with ANN-DT. Wireless

communication, Networking and Mobile Computing, 2007. WiCom 2007.
lntemational Conference on, 5484-5487.

210

   

IIIIIIIIIIIIIIIIIIIIIIIIIIEs