144 r“ mac—an:

N, v
1,, “p? V
h"" . k "‘I
:n .. "(h
:v .

xii-Mn
. .3 n ,7} ‘33
’4' '4 .

”“‘L

1_ ‘lggl
‘h‘f
. »

3‘46.

1'

«‘93:
9.9

h;

«:3-
FEE-Eh?

W!

W? 3'
, - w
'1» ’51".
{f ‘

.‘ rt
Q.?‘,¢.
jf‘r ‘1 3':
‘vll‘v \
,

#
Ax km I ..' ”.1
m x

LI.

»
u

C

l ‘ . ‘ , (‘9
.".\.‘v x ‘ -. . 'M" .‘

W m ’. - ' ‘ w‘; '

c - 7.1.5 h. ' . "ff-’1 ' V 9 J
, ‘1 5~ '4». ' . A ,V .n '

”‘1“ a fly

. H 7‘.
fl“ JELH “1‘ ‘
‘fi , .I“.{\
't .4 LI

fl
1“

w- -.4:
éiaa Egg
QE‘FV"! :-

3?“ wr.
fl‘ Wm;

16"“

v
. ‘ I «
" . " ‘ r3-
: ' V »‘ ‘- ‘ y ’t“ 12 {‘"
- ' 7.. _ . ‘ Ir'- - ,u' l .‘n‘g ‘
{'2 v 'n \2 “'5“ '4. A - .‘ . ”m”, - .A. ‘.“.n't.;‘;\‘.fiq;;wf§ ,,
, 4 "I . ’1 ' r ' »
1““ .1 1'! ‘ ‘ .‘ . “$413353 4
“g.
:-

‘“ iv
\ fim.‘
“$23‘--¢ ‘- -

I
“E!”
'q

. g“ g; m. .
35331?
5.

1;

.1
a
N!" 'u’n" 5:133th

l
‘7'

«*4

. L.
.3 ' ‘
Hi3; rfi‘y
1‘ -,.5 .

a;

.
.
45¢

*i‘ . Q
iii-fig?
:‘r‘

{A
Mi»,
.._ A 1qu
:1".
.rr '3‘“! 4

War.” 1

I)

.' .p,. Q;
55% “'Mfi?‘ ”
‘1 1945‘”

851‘. '3’33" ‘
\1;§é.g‘: (#2:

fi‘

,.
a ,7 “q.

1.; ..
#5,} y’. ‘M 3

\‘1

3‘

1:5} '-

, r“

1".”
r, 0'

 

WESIS

liiiiii\iii®iflij§i\fliifiili\l

, U I

 

 

 

This is to certify that the

dissertation entitled

THE INFLUENCE OF SCARCITY MESSAGES ON BUYING IMPULSE:
AN EXPERIMENTAL STUDY OF TV HOME SHOPPING PROGRAMS

presented by

Hairong Li

has been accepted towards fulfillment
of the requirements for

Ph.D. degree in Mass Media

 

 

 

 

61 ajor profegor V

Date March 28, 1995

MS U i: an Affirmative Action/Equal Opportunity Institution 0-12771

fi—“W fl <_

 

,____v‘ -" «W 4

. .- ~qfu * ‘

 

 

LIBRARY
Michigan State
University

 

 

 

PLACE ll RETURN BOX to remove thle checkout from your record.
TO AVOID FINES return on or More date due.

DATE DUE DATE DUE DATE DUE

    
      

 

FEB o 5 39337

W ’
V (3‘1 (‘2 a?!

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU Ie An Affirmative AottorVEquel Opportunity lnetttuton
Want

THE INFLUENCE OF SCARCITY MESSAGES ON BUYING IMPULSE:
AN EXPERIMENTAL STUDY OF TV HOME SHOPPING PROGRAMS

By

Hairong Li

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Mass Media Ph.D. Program
College of Communication Arts and Sciences

1995

ABSTRACT
THE INFLUENCE OF SCARCITY MESSAGES ON BUYING IMPULSE
AN EXPERIMENTAL STUDY OF TV HOME SHOPPING PROGRAMS
By
Hairong Li

This study examines the influence of scarcity messages on the impulse buying of
TV home shopping program viewers. Consistent with previous studies, this study
defined the impulse buy as a purchase that is made by a TV home shopping program
viewer who has had no intention to buy the product prior to tuning in to the shopping
channel. Additionally, this study identified two new dimensions of the impulse buying:
the response time and the strength of the impulse buying. Two types of scarcity
messages (claims of limited quantity of an item and indications of a diminishing amount
of time available for the viewer to order an item) are proposed to have a positive
influence on each dimension of the impulse buying. The viewers' materialistic values

also are postulated to affect their impulse buying in TV home shopping positively.

To test these research hypotheses, an experiment was conducted between May 24
and June 15, 1994, which consisted of 48 one-hour sessions for five conditions. A total
of 150 students participated in the experiment as subjects. The subjects answered
questions in a computerized instrument while they were viewing the experimental TV
home shopping programs. The computerized instrument recorded the subject's intention
to buy on impulse, the response time, and the strength of the buying impulse, among

other variables.

The study found that the levels of buying impulse vary across products and
gender, that the viewers tend to make their purchase decisions considerably before a
product segment ends, and that it takes much less time for the viewers to decide not to
buy a product than to decide to buy a product. The results also indicated that the most

buying impulses are accompanied by a strong or very strong intention to do so.

The most important finding of this study is that the factors of interest have
different impacts on different dimensions of a buying impulse. Specifically, a buying
impulse can be triggered by the viewer's positive product evaluation and materialistic
values. When a buying impulse occurs, it is subject to the impact of scarcity messages in
terms of the response time and strength of the buying impulse. Scarcity messages

themselves do not usually trigger a buying impulse.

To Ying, Kevin and Alan

iv

ACKNOWLEDGMENTS

I would like to thank many special people in my life at Michigan State University.
I give heartfelt thanks to Dr. Bradley Greenberg for his unusual support, understanding,
and guidance. I am lucky to have him as a mentor and friend. I thank Professor E
Lincoln James, who initiated my research interest in direct marketing. I am also indebted
to Professors Bonnie Reece and Bruce Vanden Bergh, who are excellent sources of

inspirations in consumer behavior research.

I wish to extend appreciation to Drs. Thomas Baldwin, Gorden Miracle, Keith
Adler, Nora Rifon, Cornelius Pratt, and Tuen-Yu Lau, who always gave me advice. I am
also grateful to a number of my colleagues, including Regan Kania, Jeff Brand, Kartik
Pashupati, Louisa Ha, and Rick Busselle. I thank Ken and Alice Jones for their

friendship. Finally, I thank my wife Ying Jiang for her love.

TABLE OF CONTENTS

LIST OF TABLES ......................................................................................................... viii
LIST OF FIGURES ........................................................................................................ x
CHAPTER
I. INTRODUCTION .............................................................................................. 1
TV Home Shopping ............................................................................................ 1
Evolution of the TV Home Shopping Business .................................................. 3
Consumer Behavior in TV Home Shopping ....................................................... 6
Purpose of Study ................................................................................................. 7
Endnotes .............................................................................................................. 9
II. REVIEW OF THEORY AND RESEARCH HYPOTHESES ............................ 10
Who is the TV Home Shopper ............................................................................ 10
Impulse Buying ................................................................................................... lO
Impulse Buying in TV Home Shopping ............................................................. 13
Impulse Buying Compared to Other Buying Behaviors ..................................... 15
Factors Contributing to Impulse Buying ............................................................. 18
Measurement of Impulse Buying ........................................................................ 21
Scarcity Messages ............................................................................................... 22
Theories of Scarcity Messages ............................................................................ 23
Conditions for Scarcity Messages to Function ................................................... 27
Materialism ......................................................................................................... 27
Materialism Over the Life Span .......................................................................... 30
Acquisition and Materialism ............................................................................... 31
Research Propositions and Hypotheses ............................................................... 32
Endnotes .............................................................................................................. 36
III. METHODOLOGY .............................................................................................. 37
Research Design .................................................................................................. 37
Stimulus Materials .............................................................................................. 40
Subjects ............................................................................................................... 47
Instruments .......................................................................................................... 48
Experimental Procedures and Data Collection ................................................... 51
Preliminary Data Analysis .................................................................................. 53
Endnotes .............................................................................................................. 62
IV. RESULTS ........................................................................................................... 63
Buying Impulse in General: Descriptive Findings .............................................. 63
Testing the Hypotheses ....................................................................................... 67
Beyond Hypothesis Testing ................................................................................ 77
Endnotes .............................................................................................................. 81

vi

V. SUMMARY AND DISCUSSION ...................................................................... 82

Impulse Buying in TV Home Shopping ............................................................. 82

Scarcity Messages and Impulse Buying .............................................................. 84

Materialism and Impulse Buying ........................................................................ 85

Product Evaluation and Believability of Scarcity Messages .............................. 86

Theoretical Implications ..................................................................................... 87

Limitations .......................................................................................................... 89

Future Research ................................................................................................... 90

Conclusions ......................................................................................................... 93
APPENDIX 1

The Script of a Focus Group Discussion on TV Home Shopping ...................... 94
APPENDIX 2

QVC Programs by Frequency of Run ................................................................. 111
APPENDIX 3

Viewership of TV Home Shopping Programs .................................................... 116
APPENDIX 4

The Computerized Questionnaire ....................................................................... 117
APPENDIX 5

The Paper-and-Pencil Questionnaire .................................................................. 122
BIBLIOGRAPHY ........................................................................................................... 132

vii

LIST OF TABLES

TABLE

1-1.TV Home Shopping Program Viewers: 1989-1992 ................................................ 5
3-1. Top Ten QVC Programs .......................................................................................... 41
3-2. Four Selected Products ............................................................................................ 42
3-3. Quantities For High and Low Scarcity: Pretest Results .......................................... 44
3-4. Scarcity Messages for the Five Programs ................................................................ 45
3-5. Lengths of the Programs and Scarcity Messages .................................................... 46
3-6. Subject Characteristics ............................................................................................. 53
3-7. Product Evaluation Indices ...................................................................................... 55
38. Items and Factor Loading of Materialism: Initial Results ....................................... 57
3-9. Items and Factor Loadings of Materialism: Final Results ....................................... 59
3-10. Materialism Scores by Level ................................................................................. 60
3-11. Characteristics of the Subjects in Five Conditions ................................................ 61
4-1. Buying Impulse for Each Product ............................................................................ 63
4-2. Multiple Purchases of the Sample ........................................................................... 64
48. Response time for Two Different Decisions ........................................................... 65
4-4. Strength of the Intention to Buy a Product .............................................................. 66
45. Intention to Buy in Scarcity and Non-Scarcity Conditions ..................................... 68
4-6. Response Time in Scarcity and Non-Scarcity Conditions ....................................... 68
4—7. Strength of Buying Impulse in Scarcity and Non-Scarcity Conditions ................... 69
48. Buying Impulse in High and Low Scarcity Conditions ........................................... 71
4-9. Impact of Materialism on Buying Impulse .............................................................. 72
410. Response Time by Materialism ............................................................................. 73

viii

411. Strength of Buying Impulse by Materialism ......................................................... 74
4-12. Buying Impulse by Scarcity and Materialism: Rolodex Electronic Planner

and Skewdriver Pro Kit .......................................................................................... 75
4— 13. Buying Impulse by Scarcity and Materialism: Nutritional Monitor ...................... 76
4-14. Buying Impulse by Scarcity and Materialism: Electronic Pedometer ................... 77
4-15. Buying Impulse by Product Evaluation ................................................................. 78
4-16. Believability of Two Types of Scarcity Messages ................................................ 79
4-17. Buying Impulse of Scarcity Believers and Non-Believers .................................... 8O

ix

LIST OF FIGURES

FIGURE
2-1. The Position of Need in Shopping Processes .......................................................... 18
3-1. The Strength of Buying Impulse by Scarcity Conditions ........................................ 39

CHAPTER 1: INTRODUCTION

TV home shopping is a distinctive combination of advertising, promotion, and
merchandising; it is a marketing enterprise that is still in its infancy. TV home shopping
and parallel interactive retailing formats will bloom in the next decade as the expanding
information superhighway reaches a majority of homes in the United States (Elmer-
DeWitt, 1993). Industry experts anticipate that TV home shopping will turn into a sort of
"video mall” where shoppers will browse through channels as through individual stores.
They will be able to ask for information and advice, order goods, and pay for them — all
without leaving the comfort of their homes (Zinn, George, Shoretz, Yang, & Forest,
1993) . The technological innovations on the horizon, the commitment and investments
of media conglomerates, and public acceptance are bringing this ultimate vision closer to
reality (Elmer-DeWitt, 1993). A recent annual survey of 1,000 US. adults by
Advertising Age found that 31.4% said they were aware of the term "interactive media,"
compared to 19.1% in the last year’s annual survey, and that 19.5% had used home
shopping by TV (Fawcatt, 1994).

TV Home Shopping
TV home shopping is an electronic form of direct marketing. It consists of TV

programs that are dedicated exclusively to selling products, an idea that originated from
marrying America’s twin passions of shopping and TV watching (Robichaux & Roberts,
1993) . Home shopping programs are designed to present "live” demonstration of
products in an entertaining and informative way. Transmitted via cable, broadcasting
television, and satellites, TV home shopping programs of two major suppliers, QVC
Network and Home Shopping Network, reach two-thirds of the U. S. households with

television, 24 hours a day and seven days a week (Zinn, et al., 1993).

A characteristic of TV home shopping is interactivity, which allows viewers to
talk with program hosts (also called product describers) and salespeople by telephone
while watching the program. Although shopping programs are carried on the current one-
way television system, they are actually used in a two-way mode in that viewers can use
the telephone to inquire about products and make their orders directly from their home.
The Direct Marketing Association defines direct marketing as "an interactive system of
marketing which uses one or more advertising media to effect a measurable response
and/or transaction at any location" (Stone, 1989) . Obviously, TV home shopping is more
interactive than other conventional direct marketing vehicles, such as newspapers,
magazines, and direct mail in that the two-way communication between shopping

program viewers and merchandisers can be simultaneous.

Another salient characteristic of TV home shopping is that the communication
between viewers and merchandisers is a consumer-initiated process. Unlike marketer-
initiated direct marketing efforts such as telemarketing or direct mail, shopping programs
provide customers the ability to tune in at their discretion at any time of day. Thus, TV
home shopping is less intrusive than other forms of direct response advertising like
telemarketing and direct mail. Intrusion means that marketing messages or materials are
delivered to consumers without their prior consent or request so that consumers feel
uncomfortable with the practice of marketers. In an age when concerns for privacy and
intrusion are mounting (Schultz, 1992), public acceptance of less intrusive media vehicles

like home shopping programs will have a high potential for growth.

TV home shopping programs are transient by nature, as in the case with other
broadcasting media. In normal viewing situations, viewers have no control over the flow
of programs. As products are introduced or demonstrated one by one at a relatively even

pace, viewers may face some time pressure when they find something interesting in

shopping programs, especially if they require more time than the program allows to think
about their possible decisions on interesting products. Further, time pressure may be very
high when perceived risks of various kinds1 are involved in viewers’ purchase decisions.
In an effort to reduce shoppers’ perceived risks, TV home shopping services offer lenient
return policies. For instance, QVC provides its customers with a 30-day unconditional
money-back guarantee. As a result, return rates are higher in TV home shopping than in
traditional retailing. In 1993, the average retum rate was 20 percent for the Home
Shopping Network; and returns on some merchandise categories run as much as 40

percent (Reilly, 1993).

These characteristics make TV home shopping a unique retailing environment
where the consumer’s decision-making processes may differ greatly from those in the

conventional retail stores. This issue will be elaborated in the following sections.

Evolution of the TV Home Shppping Business
TV home shopping originated in 1977 with a radio auction show in Clearwater,

Florida It was the predecessor of the Home Shopping Network (HSN). The show made
the leap to television in 1982 on a local system, and was named the Home Shopping
Club, a name the network still uses for its shopping programs. In 1985, the company
began televising its shopping programs nationally (OECD, 1992). The QVC Network,
another major TV home shopping service, was established in 1986. By 1988, more than
30 other home shopping channels existed. After a consolidation in the early 19908, HSN
and QVC became two TV home shopping giants (Ferraro, 1993). In 1992, the QVC
service reached 42.9 million households, with an average net sales of $24.85 per
household. Home Shopping network’s coverage was 57.9 million households, with an
average net sales of $18.96 per household. The combined net sales of HSN and QVC
were $2.2 billion in 1992 and $2.3 billions in 19932.

TV home shopping has had the image of a discount retail outlet since its
inception. For years, marketers have categorized TV home shoppers as junk-jewelry-
buying, home-bound women from blue-collar households (Reilly, 1993). This image is
partially attributable to the positioning strategy of a leading home shopping service, the
Home Shopping Network. This network has positioned itself as a middle level retail
outlet, as Garry Hogan, president and CEO of HSN, acknowledged in an Advertising Age

interview (Donaton, 1993):

You can think of us as being very close to a Wal-Mart in terms of the kind of
customer that we serve, the kind of product that we offer and, I hope, the kind of
service that we provide. If you think of QVC and the first retailer you think about
is Saks in terms of what they want to be, I’d like you to think of Wal-Mart when
you think of HSN. I like that broad middle market. (p. 12)

The image of TV home shopping is undergoing change due to the QVC network’s
positioning strategy. QVC intends to offer hi gher-quality merchandise, and use
celebrities to present its products. It also has a soft-selling approach. The new channel,
named Q2, is a cross between a lifestyle magazine and a specialty store. It targets a more
diverse and upscale audience. As companies like cataloger Spiegel and retailer R. H.
Macy & Co. plan to enter this business, the gentrification of TV home shopping is likely
to continue (Reilly, 1993).

The TV home shopping business has had a relatively stable customer base. As
presented in Table 1, 13.5 percent of adults reported that they watched TV home
shopping programs in 1989. This number was unchanged in 1990, but increased to 15.6
percent in 1991. In 1992, TV home shopping program viewership dropped to 12.4

percent. However, the viewers who purchased at least one item in the last three months

increased to 5.5 percent in 1992, from 3.7 percent in 1991. Female viewers watched and

shopped more often than male viewers in this period of time.

Table 1-1. TV Home Shopping Program Viewers: 1989-1992
(96)

 

Watch Bought in Last 3 Months

 

1989 1990 1991 1992 1989 1990 1991 1992

 

Total U.S. Adults 13.5 13.5 15.6 12.4 3.9 3.3 3.7 5.5
Men 11.7 11.8 12.8 10.2 3.6 2.8 2.8 5.1
Women 15.2 15. 1 18.3 14.4 4.3 3.7 4.5 6.0

 

 

Source: Study of Media and Markets, Simmons Market Research Bureau, 1989-1992.

The next wave in TV home shopping development is likely to be spawned by
technological advances in telecommunication fields. A merger of video, telephones, and
computers will make television an entirely two-way, or interactive, medium as
distinguished from the early version of two-way television such as the QUBE system in
Columbus, Ohio. These mergers will be a natural result of several technological
breakthroughs, including the ability to translate all audio and video communications into
digital information; new methods of storing digitized data and compressing them to allow
for travel through existing telephone and cable lines; fiber-optic wiring that provides a
virtually limitless transmission pipeline; and new switching techniques and other
innovations that will make it possible to bring all this to neighborhoods without the
necessity of rewiring every home. These are the fundamentals of what is referred to as

the ”information superhighway." (Elmer-DeWitt, 1993)

Innovations such as these are likely to contribute to the proliferation of home
shopping in at least two ways. First, an anticipated 500 TV channels will be available in
several years and most of the channels will offer services, home shopping, and other
types of information programming. Second, a real two-way television system will make
it possible for program viewers to communicate with merchandisers via a remote control
device. Instead of talking only with program hosts and salespeople, viewers will have
prompt access to huge amounts of service and product information stored on computer
servers. Shoppers are likely to become equipped with almost perfect knowledge of
services and products, an ideal situation that economists have dreamed of for several
decades (Scherer & Ross, 1990).

Consumer Behavior in TV Home Shopping
TV home shopping changes certain facets of the retailing industry by creating a

new shopping environment. Unlike conventional retail stores, home shopping programs
are characteristic of an interactive communication system between consumers and
salespeople. Also, because of the transient nature of TV shopping programs, viewers
may face some time pressure while watching certain programs and making purchase
decisions. These characteristics of TV home shopping call for a theory of consumer
behavior from a different perspective. A few researchers have already begun exploring
consumer behavior in this special marketing environment (Grant, Guthrie, & Ball-
Rokeach, 1991; James & Cunningham, 1987) , and more research is expected to

investigate this new phenomenon of marketing communication.

One interesting issue in TV home shopping is impulse buying. Impulse buying is
likely to account for a majority of daily transactions in TV home shopping settings
because of the nature of this type of electronic retailing, as discussed in previous sections.

This speculation was supported in a focus group discussion on TV home shopping (see

Appendix 1). Additionally, even though a monthly program guide that lists all shows
hour by hour is distributed to shoppers along with their order shipments, program viewers
are not likely to know what specific brands will be available prior to tuning in to a
shopping channel. Thus, viewers will have to evaluate products as they are displayed.
This requires that purchase decisions be made in a relatively short period of time.

On the other hand, shopping program producers tend to use different promotion
techniques to stimulate and facilitate the viewer’s spontaneous intent to buy. One set of
promotion techniques is what can be refened to as "messages of scarcity" — claims of
limited guantigz of an item, and indications of diminishing amount of time left for
viewers to order an item. These techniques are used to generate a sense of urgency
within viewers which may trigger a buying impulse. The effects of scarcity messages on
individuals’ valuation, liking, and desirability of a commodity have been investigated for
four decades (Lynn, 1992). However, no studies have been found that address the effect
of scarcity messages on impulse buying behavior, especially in TV home shopping
settings. Thus, an exploration of the impact of this TV programming strategy may

contribute to the theory of consumer behavior in an interactive media environment.

mm
This study examines the influence of scarcity messages on impulse bujg'ng
behavior of TV home shopping progm viewers. Two types of scarcity messages —

claims of limited quantity of an item and indications of a diminishing amount of time left
for viewers to order an item — will be examined in an experimental study. With a
factorial design, three kinds of influences of scarcity messages on impulse buying will be
investigated. They are the influence of scarcity messages versus an absence of scarcity
messages; the influence of high scarcity messages versus low scarcity messages; and the

influence of two types of scarcity messages — limited quantity versus limited time

available for action. The mediating effect of individual differences in materialistic values

also is explored.

Previous studies of the constructs of interest and research propositions and
hypotheses are developed in Chapter 2. Chapter 3 describes the experimental design,
stimulus materials, and data collection and analysis . Findings of the experiment are
presented in Chapter 4 and a summary and discussion of the implications and limitations

of the study in Chapter 5.

Endnotes

1 Murphy and Enis (1986) identified five kinds of risks consumer perceive in their
decision making. They were financial, psychological, physical, functional, and social

risks.

2 Sales numbers were from the 1992 and 1993 annual reports of the QVC Network and
the Home Shopping Network, respectively.

CHAPTER 2: REVIEW OF THEORY AND RESEARCH HYPOTHESES

This study investigates the influence of scarcity messages on impulse buying of TV
home shopping program viewers, as mediated by their materialistic values. This chapter
reviews previous studies and develops research propositions from which a set of
hypotheses is derived. Studies on impulse buying, scarcity messages, and materialism will
be discussed in separate sections. Research propositions and hypotheses are then
presented.

W139 is the TV Home Shmr
In a pioneering study of TV home shoppers, James and Cunningham (1987)

examined the motivational, attitudinal, psychographic, and socio-environmental
characteristics of TV home shoppers. A survey questionnaire was mailed to a sample of
2,250 known direct marketing TV shoppers nationwide and 956 questionnaire were
returned, of which 905 were usable. The discriminant function coefficients and values of
the group means of TV shoppers and non-shoppers were used to interpret the results. The
findings show that when compared to non-shoppers, individuals who engaged in TV
shopping were more likely to be guided by convenience and a greater need for affiliation.
They tended to be females, younger, and highly socially isolated They also were more
likely to watch TV between midnight and 6 am. and to have a positive attitude toward
credit. Further, TV shoppers were less likely to be black and less likely to have a non-

traditional marital arrangement

Impulse Bufl'ng

Impulse buying is one type of consumer behavior that has attracted marketing
researchers for more than four decades. Clover (1950) studied impulse sales in a unique

market situation when 154 retail stores in three west Texas towns were closed for two days

10

11

in five weeks due to a shortage of gas. It was assumed that the more important impulse
purchases were to a given store, the less likely the store would be to make up its sales after
it re-opened. The sales data in these stores showed tlmt in the week with the first one-day
closing, sales declined 18% percent from the previous week or the "base week". In the
week after the first closing, sales recovered upward to a level of just 1% below the base
week. In the following week with the second one-day closing, sales declined, this time by
14% compared to the base week. The succeeding week sales rose to a level of 4% below
the base week. Clover concluded that such a pattern indicated impulse buying definitely

influenced sales in retail stores.

West (1951) conducted 5,300 personal interviews in six Canadian cities with
shoppers who purchased a total of 15,500 items. He found that 37 percent of all purchases
were impulse purchases — items that shoppers decided to buy while in the stores.
(Bellenger, Robertson, & Hirschman, 1978) surveyed 1,600 customers of department
stores and found that 39 percent of the purchases were impulse buys. In a summary of
previous studies on impulse buying, Cobb and Hoyer (1986) concluded that there appears
to be a high level of impulse purchasing in today’s marketplace, and that this type of
purchase behavior is not confined to any one single type of product or outlet.

Paramount importance was assigned to the definition of impulse buying on the
basis of unplanned 9229M (Weinberg & Gottwald 1982). As simply defined by
Bellenger et al. (1978), impulse purchases are purchases where the decision to buy is made
in the store. According to Engel & Blackwell (1982), impulse buying is a purchase made
by the shopper who has not recognized a need for the product or who does not before

entering the store intend to buy the product.

12

Various classifications of impulse buying have been made. Stern (1962) proposed
four types of unplanned purchases. A male imgLIse buy is a novelty or escape purchase
which breaks a normal buying pattern. mm buying occurs when the shopper
remembers prior experience with the product which lcads to the purchase. Suggestion
m buying occurs when a shopper sees a product for the first time and visualizes a
need for it. Planned impulse buying occurs when the shopper enters the store with the
intention to make certain purchases dependent upon the availability of price specials,
coupon offers, and the like.

Kollat and Willett (1967) used a matrix of intention and outcome categories to
identify nine possible purchase types; only one type was unplanned purchase — it was a
purchase by the shopper who did not recognize the existence of a need before entering the
store, or the need was latent until the shopper was exposed to a product in the store. More
recently, Cobb and Hoyer (1986) suggested that there were three types of shoppers:
planners — shoppers who had an intention to buy a specific brand before entering the
store; mar planners — shoppers who had an intention to buy a product in a certain
category but not a specific brand; and impulse purchasers — shoppers who had no
intention to buy either a product of certain category or a specific brand prior to the store

visit.

Some studies love addressed psychological aspects of impulse buying. Rook
( 1987) identified eight characteristics of impulse buying from a content analysis of his
respondents’ descriptions of their impulse buying experiences. According to his findings,
the buying impulse is a sudden urge to buy something immediately, which is likely to be
triggered by a visual confrontation with a product or some promotional stimuli. In certain
situations, however, buying impulse may arise internally without any apparent external

visual encouragement. Buying impulses sometimes take center stage quickly and become

13

intensely preoccupying. Buying impulse itself is stimulating and sometimes it is a source
of personal excitement. It discourages consideration of the behavior’s potential
consequences so that impulse buying is frequently accompanied by different feelings such
as indulgence and irrationality. These characteristics of impulse buying are summarized in
Rook’s definition:

Impulse buying occurs when a consumer experiences a sudden, often powerful and
persistent urge to buy something immediately. The impulse to buy is hedonically
complex and may stimulate emotional conflict Also, impulse buying is prone to
occur with diminished regard for its consequences. (p. 191)

Obviously, one difference between Rock’s definition and the definitions of impulse
buying as ”unplanned purchasing" is that, as Rook observed, his term of impulse buying
refers to a narrower range of phenomena. Some unplanned buying may have certain
psychological characteristics as Rook described, whereas other unplanned purchases may
lack these characteristics.

Impulse Bufl'ng in TV Home Shopm'ng
It is clcar that most previous studies of impulse buying have focused on unplanned

purchases for which decisions are made in the store (Bellenger et al., 1978; Engel &
Blackwell, 1982) and the factors that tend to trigger impulse purchases (Cox, 1964;
Phillips & Bradshaw, 1993; Prasad, 1975). Impulse purchases have been treated as a one-
dimensional phenomenon in these studies — whether the shoppers make their decisions in
the store or they have made their decisions before entering the store. When we think of the
buying impulse in a TV home shopping setting, it seems that two other dimensions are
worth investigating:

1) How quickly an intention to buy occurs in the viewing process, and
2) the strength of this intention.

14

In this study, how quickly an intention to buy an unplanned item occurs in a
viewing process is referred to as the resm time of the buya'pg impulse whereas how
strong an intention to buy is the strength of the buyipg impulse. Response time of buying
impulse can be defined as reaction time in psychological studies. Reaction time refers to
the time from the onset of a stimulus to the time the subject responds (James, Schneider, &
Hinds, 1992). It is a standard mcasure in cognitive and perceptual psychology. Reaction
time is often used as a dependent variable in studies that test how different stimulus
manipulations affect the speed of the subjects’ cognitive responses to these stimuli. It is
often measured in milliseconds by computerized instruments (Graves & Bradley, 1988;
Graves & Bradley, 1991; Mordkof f & Egeth, 1993; Tryon & Mulloy, 1993). In this
study, the response time of buying impulse is the time from the beginning of product
segment in a shopping program to the time the viewer indicates an intention to buy or not to
buy the product.

Response time is a key element in the TV home shopping business. Since products
are presented one after another by the program host, the shorter the average response time,
the more units the program can sell in a given period of time. Procter & Gamble Co.
Chairman-CEO Edwin L. Artzt was very impressed with this advantage of TV home
shopping. In his speech to the American Association of Advertising Agencies in May
1994, he (1994) said:

Barry Diller, at QVC, told me he can sell 20,000 pairs of earnings in 5 minutes on

his home shopping channel. Tlnt’s terrific for a company that sells impulse items.
(p. 24)

This magnitude of prompt response is a unique characteristic of TV home shopping
largely because of live presentations of products, the national coverage of shopping
channels, and the capacity for handling a huge volume of phone calls simultaneously.

15

Thus, response time is indeed an indispensable aspect of impulse buying in TV home

shopping settings.

The concept of strength of buying impulse is largely inspired by Rook’s (1987)
discussion of impulse buying. As previously reviewed, Rook noted that impulse buying
occurs when a consumer experiences a sudden, often powerful and persistent urge to buy
something immediately. He distinguished his definition of impulse buying from those that
equate all unplanned purchases to impulse buying. By his definition, only unplanned
purchases that are caused by a strong urge constitute impulse buying. Using the broad and
more operationalizable definition of impulse buying, this study treats any unplanned
purchases as impulse buying in TV home shopping situations. In addition, this study
attempts to examine the strength of that impulse buying. It is reasoned that if the viewers
decide to buy an unplanned product during TV home shopping programs, their intention to
buy will be strong instead of weak.

Impulse Bufl'ng Comm to Other Buying Behaviors

Impulse buying is distinct from other types of consumer behaviors, such as
reasoned action (Fishbein, 1980; Fishbein & Ajzen, 1975), planned behavior (Ajzen, 1985;
Ajzen, 1991), compulsive behavior (Faber & O’Guinn, 1989), and addictive behavior
(Scherhom, 1990).

The theory of reasoned action assumes that consumers consciously consider the
consequences of the alternative behaviors under consideration and choose one that leads to
the most desirable consequences. Behavioral intention is the single best predictor of their
actual action, and behavioral intention is influenced by attitude toward the behavior and
perceived subjective nonn about the behavior (Fishbein & Ajzen, 1975; Fishbein & Ajzen,
1980). Sheppard, Hartwick and Warshaw (1988) conducted two meta-analyses of studies

16

on the reasoned action model. They found strong overall evidence for the predictive utility
of the model, and although numerous instances were identified in which researchers
overstepped the boundary conditions initially proposed for the model, the predictive utility

remained strong across conditions.

Ajzen (1985, 1991) proposed a theory of plarm behavior by adding a new
variable, perceived behavior control, to the theory of reasoned action. According to the
theory of planned behavior, behavioral intention is determined not only by attitude toward
the behavior and the subjective norm, but also by perceived behavior control. By perceived
behavior control, Ajzen (1991) means the ability of the person to decide at will whether to
perform or not perform the behavior. Because most behaviors depend on opportunities or
resources (e.g., money, skill, cooperation of others), only to the extent that a person has
the necessary opportunities and resources and intends to perform the behavior, should the
person sucwed in doing so. Obviously, impulse buying is different from a reasoned action
or planned behavior in that impulse buying is by definition unplanned and it is ”prone to
occur with diminished regard for its consequences” (Rook, 1987).

Impulse buying also differs from compgsive or addictive buyg'ng behavior in the
degree of self-control, even though both types of behavior are similar in certain aspects.
According to Faber and O’Guinn (1989), compulsive buying is an abnormal form of
consumer behavior tint is typified by chronic buying episodes of a somewhat stereotyped
fashion in which the consumer feels unable to stop or significantly moderate the behavior.
Compulsive buying behavior is signified by agreement with such statements as ”Bought
something in order to make myself feel better,” "Went on a buying binge and wasn’t able to
stop," "Bought things even though I couldn’t afford them." Although compulsive buying
may produce some short-term positive feelings for the shopper, it ultimately is destructive

to normal life functioning and results in significant negative consequences.

17

Instead of using the term compulsive buying, Scherhom (1990) argued that the
term motive consumm'on was consistent with psychological literature, and would lead to
additional insights into consumer shopping behavior. He noted that a psychological
difference between the two concepts lies in the reason for the loss of control which both
addiction and compulsion have in common. Compulsion means that one feels pressed to
do and repeat something even against one’s will, whereas addiction is considered to be
driven by an irresistible internal urge one experiences. Addiction is a type of self-
estrangement. Scherhom (1990) used habitual drinking to illustrate this point. The result
of the behavior — being drunk, despised, and abandoned — is not the state the drinker
desires. As can be inferred from his articulated feelings, he would like to avoid such a
state. When he drinks again, he will get drunk. The self-estrangement prevents him from
realizing that he has a choice — not to drink. Addictive buying has a nature similar to that
of addictive drinking.

One difference between impulse buying and compulsive or addictive buying lies in
the individual’s motivation to engage in the purchase act. Whereas impulse buying is
characterized by a desire for specific items, research with compulsive buyers indicates that
many of them do not have a great deal of interest in items after they are purchased
(O’Guinn & Faber, 1989). Another difference is the severity of consequences of impulse
buying and compulsive or addictive buying. For most people, impulse buying does not
lead to consequences as severe and extreme as those resulting from compulsive buying,
such as severe debt, lowered self -esteem, and a deleterious impact on interpersonal

relationships (O’Guinn & Faber, 1989).

The differences among the reasoned or planned action, impulse buying, and
compulsive or addictive behavior may be better illustrated by introducing the concept need

18

recogrp'tiop. Kollat and Willett (1967) used need recognition to distinguish planned
purchase from impulse buying. In their words, ”Before entering the store the shopper does
notrecognizetheexistenceofaneedortheneedislatentuntilsheisinthestoreandhas
been exposed to its stimuli” (p. 21). Thus, it can be assumed that planned shopping starts
with recognized needs; the choice of a product is then evaluated based on those recognized
needs, and a decision to purchase is then made. Impulse shoppers, on the other hand, tend
to encounter products in the store, then attempt to recognize the needs that the product can
satisfy, and then make the buying decision. This process can be referred to as "a product-
need matching.” Still, for compulsive or addictive shoppers, such a matching process may
never take place. This type of shopper may just buy whatever they encounter and like,
without even considering needs the product may meet Or, their needs may be just the
buying itself. The position of recognized needs in different shopping processes is
diagrammed in Figure 2-1.

Figure 2-1. The Position of Need in Shopping Processes

Planned Bufl'ngz
Recognized Needs _’ Product Evaluation _> Buying Decision

Product Encounter _> Needs Recognized _> Buying Decision

Commlsive Bufl'ng:
Product Encounter + Buying Decision

Factors Contributing to Immlse prg'ng
Previous studies suggested that impulse buying may be triggered by three types of
factors: situational factors, product-related factors, and consumer factors. The following

sections review major studies of impulse buying for these three factors.

l9

Situational Factors like store types, location of product display and shelf space have
been found to affect impulse purchases. Prasad (1975) investigated unplanned buying in
two retail settings — department stores and discount stores. It was found that significant
differences existed between the discount and department stores in terms of the overall level
of unplanned buying. While the rate of incidence of unplanned buying was significantly
higher among the discount store shoppers, the average size of the unplanned purchases was
significantly higher among the department store shoppers. Cox (1964) tested the
hypotheses that food products sales were responsive to changes in shelf space, and that
impulse items were relatively more responsive than staples. He found, however, that sales
of staples did not increase as a result of a larger shelf space, and that only one of his three
experimental impulse items had higher unit sales in a larger shelf space. In a recent review
article, Phillips and Bradshaw (1993) emphasized that point of purchase is a major factor in
determining consumer purchasing behavior, especially impulse buying. They discussed
the impact of product display on shoppers’ selection of products by facilitating their visual
perception of the products, and concluded that an interaction between shoppers and
product display is required to "account for impulse or unplanned purchasing which is

shown to be a major factor in many areas of retailing” (p. 51).

Mum-related factors Previous studies have also explored the relationships
between impulse buying and product categories as well as price levels. Prasad (1975)
examined impulse purclmsea by product categories in department and discount stores. The
”Women and Girls’ Wear” category and the ”Children and Infants’ Wear” category had the
highest and the lowest unplanned buying rates, respectively, in the department stores. In
the discount stores, the "Men and Boys’ Wear" category and the "Toys“ category showed
the highest and the lowest rates, respectively. Bellenger et al. (1978) found that the

percentage of impulse purchases showed considerable variation by product line, with a low

20

of 27 percent in wornan’s lingerie and a high of 62 percent in costume jewelry. Deshpande
and Krishnan (1979) found that impulse purchases were associated with low cost items,
buttheyappearedtobeunrelatedtoincomeortheusageofcreditcards.

Consgner arsonal factors Several studies have attempted to identify the influence
of consumer characteristics on impulse buying. Kollat and Willett (1967) measured a
number of demographic, personality, and shopping behavior variables. They found that
impulse purchases occurred more so with a greater number of product purchases, and in a
related sense, with larger grocery bills and on major shopping trips, instead of fill-in

shopping trips.

Cobb and Hoyer (1986) examined five sets of impulse purchase predictors:
decision task variables, shopping lifestyles, general shopping behavior, personality
characteristics, and demographics. They found that, compared to planners (shoppers who
decided to buy a specific brand before entering the store) and partial planners (shoppers
who decided to buy a product in a certain category but not a specific brand before entering
the store), impulse shoppers of bathroom tissue were more likely to pick up the first brand
spotted in the store and they were people who shopped earlier in the week. These
researchers also found that impulse shoppers of coffee were the least likely to have an
emotional attachment to the selected brand, to shop for specials, and to carry a shopping
list, and that these shoppers tended to examine the fewest packages of coffee before a

purchase was made.

Lee (1990) explored the influence of mall shopper type (recreational vs. economic)
and mood (positive vs. negative) on impulse buying behavior. In his study, shopper type
was found to be more influential than mood on impulse buy behavior. Recreational
shoppers who entered the mall in a positive mood and economic shoppers who entered the

21

mall in a negative mood, revealed relatively higher impulse purchasing rates. In comparing
exiting mood states, among the shoppers who entered the mall in a positive mood, only
recreational shoppers who exited in a positive mood showed higher impulse buying rates
than those who exited in a negative mood.

Measurement of Impulse Buying
Impulse buying behavior is commonly measured by two approaches. One is what

can be referred to as the "entrance-inquiry and exit-count” method. Shoppers are asked
what they plan to purchase at the time they enter a store and, if they have a shopping list, it
will be copied by the interviewer. Then, these shoppers’ actual purchases are recorded at
the checkout. The purchases that are not mentioned or listed by the shoppers when
entering the store are considered to be impulse purchases (Kollat & Willett, 1967). Some
researchers are critical of this method. Pollay (1968) argued that by forcing the respondent
to recite his or her intentions, the pretest questioning tends to commit that respondent to
fulfill those intentions. Pollay also believed that, if shoppers expect to have their groceries
inventoried, they are more likely to change their purchase behavior. Thus, higher rates of
status purchases such as specialty foods might result for shoppers who anticipate

inventorying; these possibilities may bias responses.

Another method used by Cobb and Hoyer (1986) can be referred to as the "in-store
observation and interview" approach. In this approach, trained interviewers are stationed
in supermarket aisles to observe and question shoppers. Recorded during the observation
phase are such variables as number of packages examined, use of a shopping list, and
amount of time taken. Immediately after a brand choice, the shopper is asked to participate
in a brief in-store interview. Upon completion of the interview, the shopper is asked to
take an additional survey home to fill out and mail back. This method seems to be better
than the first approach in that it avoids possible biases that may be caused by entrance-

22

inquiries. However, it is able to unobtrusively record only one purchase per shopper and
data collection is less efficient. Moreover, this method categorizes shoppers into either
planners, partial planners, or impulse shoppers depending solely on their responses to two
questions: 1) "Did you plan to purchase the (imert product category name) before entering
the store today?" followed by, 2) ”Did you plan to purchase the (insert brand name) before
entering the store today?” It seems to be less objective than the entrance-inquiry and exit-

count method.

Scarcig Mwes

Scarcity messages are claims that an item is of limited availability. A scarcity claim
is an attempt to induce individuals to obtain an item while it is still available by informing
them that the item has a limited availability (Brock, 1968; Lynn, 1991). Scarcity messages
can be expressed in two formats — claims of limited quantity of an item and indications of
a diminishing amount of time left for individuals to obtain an item. Claims of limited
quantity are commonly used in promotions in conventional retail stores, whereas TV home

shopping settings facilitate the use of diminishing time left for viewers to order an item.

Kelman (1953) was one of the first researchers to study messages of scarcity. He
told classes of school children that if they were willing to write an essay favoring jungle
stories, then either an entire class (non-scarce condition), or only five members of the class
(scarce condition) would receive a movie pass to see a particular movie. The results
showed that the school children who received the message stating that the movie passes
were scarce wanted to see the movie more than those who received the message stating that

the movie passes were available to everyone.

Frornkin (1971) explored the effect of scarcity messages on the perceived monetary
value of an item. Scarcity was manipulated by telling one group that he had 1,000 pairs of

23

nylons and could distribute 50in their area (scarce condition). The second group was told
that he had 100,000 pairs and could distribute 1,000 in that area (non-scarce condition).

The results showed that the subjects who received the message stating that the nylons were
scarce were willing to pay $2.71 per pair versus $2.27 for the non-scarce group. Fromkin

assumed that the difference in the two prices represented a difference in value.

Knishinsky (1982) had beef salespeople telephone supermarkets and provided
them with one of two messages to potential customers. One group of customers received a
standard sales presentation that attempted to get them to place their beef order (non-scarce
condition). The second group received the standard sales pitch, but was also informed that
the supply of beef was likely to be in short supply in the upcoming months (scarce
condition). The results of the experiment showed that customers who were told that the
beef was likely to be in short supply ordered over two times more beef than customers who
received the standard sales pitch.

Theories of Scarcigg Messages
Different theories have been developed to explain the effects of scarcity messages.

One is commodig th_epry (Brock, 1968). Its principal claim is that "any commodity will be
valued to the extent that it is unavailable“ (Brock, 1968) . There are three basic concepts in
this theory: commodity, value, and availability. Commodity may be anything that can be
possessed, is useful to its possessor, and is transferable from one person to another. Value
refers to a commodity’s ”potency for affecting attitudes and behavior" (Brock 1968,
p.246). Since commodities have a positive utility, any enhancement of a commodity’s
value will increase its perceived utility and will make the commodity more desirable.
Availability refers to the perceived difficulty of obtaining the commodity (Bonolo &
Brock, 1992; Brock, 1968). Therefore, availability can be limits on the supply or the
number of suppliers of the commodity; costs of acquiring, keeping, or providing a

24

commodity; restrictions limiting possession of a commodity; and delays in providing a
commodity. According to Brock (1968), when a commodity is perceived as unavailable,
that commodity is valued to a greater extent and the motivation to obtain it is enhanced.
Scarcity is only one manner in which a commodity may be perceived as unavailable
(Bozzolo & Brock, 1992; Brock, 1968).

Lynn (1991) conducted a meta-analysis of 38 studies testing commodity theory
between 1968 and 1989. He found that although highly reliable, the scarcity effects were
fairly small, with a mean (correlation) effect size of .12. The effects ranged from -.541 to

.43, indicating signifieant heterogeneous effects across those studies.

Commodity theory did not originally specify the mechanism underlying scarcity
effects on value. However, Brock (1968) suggested that people may desire scarce
commodities more than comparable available commodities because the possession of scarce
commodities conveys feelings of personal distinctiveness or uniqueness. This speculation
was elaborated in the need-for-unigueness Orgy (Fromkin, 1968; Snyder & Fromkin,
1980). According to this theory, people are motivated to maintain a sense of specialness as
they define themselves on various important factors related to self and others. Because
commodities are an important source for defining one’s sense of self in Western culture, it
is reasoned that scarce products provide a vehicle for establishing one’s specialness when
the need for uniqueness is activated. To ensure their sense of uniqueness, people attempt

to acquire things not available to everyone.

In a classic study, Fromkin (1970) posited that people should be very desirous of
participating in a rare or scarce activity if they were feeling highly similar to other people
(i.e., their need for uniqueness is activated). After completing a 90-item questionnaire

about themselves, students were given computer-generated feedback showing that they

25

were either slightly or highly similar to 10,000 other students who had also taken the
questionnaire. Next, the research participants were given descriptions of the experimental
environments. Because of room availability constraints, one environment was described as
being available at all times (plentiful or non-scarce condition), whereas another
environment was available only one hour per week (scarce condition). Overall, results
revealed that people were more desirous of the scarce as compared to the plentiful
experience. However, the scarcity main effect was qualified by the predicted interaction of
scarcity by need for uniqueness. Under the slight-similarity feedback condition, there were
no differences in the ratings of the plentiful versus scarce experience. For the high-
similarity feedback condition, however, the research participants rated the scarce experience
as more desirable than the plentiful one, wanted to spend more time in the scarce condition,
and were less willing to give up an opportunity to participate in the scarce as compared to
the plentiful experience. To verify the reliability of this theory, Lynn (1991) conducted a
meta-analysis of 11 studies that included the necessary scarcity and need for uniqueness

variables. The anticipated interaction was obtained and reliable across studies (2 = 3.59,

p< .001).

Psychological reactance theory (Brehm, 1966; Brehm & Brehm, 1981; Wicklund,
1974) also has been used to explain why people react the way they do to scarcity messages
(Worchel, Lee, & Adewole, 1975). Reactance theory focuses on an individual’s reaction
to threatened or lost freedoms. It maintains that people have a strong desire to be free to
believe or behave as they wish. When people perceive that these frwdoms are threatened
with elimination, they experience a psychological discomfort described as reactance.
Brehms’ argument is that this discomfort tends to cause a person to react by restoring his or
her perceived lost or threatened freedoms (Brehm & Brehm, 1981). Psychological
reactance theory seems to be a reasonable explanation for the effect of scarcity messages.

In light of this theory, messages of scarcity suggest to people that their freedom to acquire

26

certain items is constrained due to the limited availability of these items. Any restriction on
the freedom should increase the perceived attractiveness of the item. A person’s desire for
a scarce item will be increased when he or she is threatened by its potential loss. To avoid

the loss of freedom, a person will attempt to acquire the scarce item.

In contrast to these psychological perspectives, naive economic meories (Lynn,
1992) assume that the effects of scarcity may be at least partially attributable to people’s
naive economic theories of scarcity. It is possible that scarcity’s enhancement of
desirability is mediated by the assumption that scarce things are more expensive. Then,
assumed expensiveness may increase the desirability either by leading people to attribute
higher quality to the commodity or by increasing the perceived status of scarce
commodities. These causal relationships have been partially supported by empirical studies
(Lynn, 1992).

Other studies on the influence of scarcity messages on perceived expensiveness
were less conclusive. According to naive economic theories, scarcity’s enhancement of
desirability is mediated by assumed expensiveness. Thus, it is logical that scarcity’s effects
should be weakened or eliminated by precluding people from assuming that the scarce
commodity costs more than the available one. If people know the price of a commodity,
scarcity should not result in more desirability. Lynn (1989) studied the desirability of a
white wine by manipulating the availability of the wine and price information. He found
that scarcity reliably enhanced both perceived expensiveness and the desirability of the wine
only when subjects did not know how much it cost. However, this finding was
inconsistent with an earlier study. In a study by Syzibillo (1973) female subjects were
presented with information about pant suits, with three levels of price (no price
information, low price, and high price) and three levels of scarcity (no scarcity information,

low scarcity, and high scarcity). Subjects then rated the suits on a number of dimensions.

27

An analysis of subjects’ desirability ratings indicated a scarcity main effect but not an
interaction of scarcity by price information. Thus, scarcity enhanced desirability in that

study even when subjects knew how much the product cost.

WM};

Studies have revealed that scarcity messages increase the attraction to whatever is
perceived as scarce (Lynn, 1992), raise the perceived value of the item (Fromkin, 1970),
and increase desire for the item (Brock & Brannon, 1992). However, the impact of
scarcity messages is not unconditional; scarcity messages function only under certain
circumstances. First, a scarcity message must be believable before it can affect a person’s
attitude (Yun, 1992). It is clear that if people do not believe a scarcity message, they will

not consider the item scarce. Thus, their attitude toward the item will not change.

Second, people must perceive that a scarce item has worth before the scarcity
message will affect their attitude toward that item. Yun (1992) used lottery tickets in his
scarcity experiment Scarcity messages on the lottery tickets were not found to affect
subjects’ attitude. (he explanation was that subjects used in the study did not perceive the

lottery tickets to have worth.

Materialism

This study also investigates the mediating effect of materialism on buying impulse
in scarcity conditions. Materialism is a useful construct in exploring the effect of scarcity
messages on impulse buying behavior because it is the consumption orientation that has a
fundamental impact on a person’s lifestyles and daily consumption

Although researchers may disagree on the date and place of modern consumption’s

emergence, they agree tlmt it has achieved a place in both industrial and post-industrial life

28

(Belk, 1984; Belk, 1985; Richins & Dawson, 1992). Theoretical notions about
materialism have benefited from decades of historical, political, anthropological, and
sociological explorations (see Richins (1990) for a review of origins of materialism). In
contrast to scholars of other disciplines who have tended to treat materialism as an
aggregate phenomenon, marketing researchers are interested in individual differences in
materialism or materialistic values. A series of attempts have been made to develop valid
and reliable measures of materialism that are consistent with theoretical notions of
materialism (Belk, 1984; Belk, 1985; Campbell, 1969; Mochins & Churchill, 1978;
Richins & Dawson, 1990; Richins & Dawson, 1992; Yamauchi & Templer, 1982).

In the marketing literature, an early definition of materialism was developed from
studies of child and adolescent socialization. Ward and Wackrnan (1971) defined
materialism as "an orientation which views material goods and money as important for
personal happiness and social progress" (p.422). This definition has also been used in
later studies (Brand & Greenberg, 1992; Churchill & Mochins, 1979; Mochins & Moore,
1982).

More systematic studies of materialism have been conducted by Belk (1983; 1984;
1985). He defines materialism as:
The importance a consumer attaches to worldly possessions. At the highest levels

of materialism, such possessions assmne a central place in a person’s life and are
believed to provide the greatest sources of satisfaction and dissatisfaction. (p.291)

Belk proposed that materialism consists of three traits: possessiveness, nongenerosity, and
envy. According to Belk (1983; 1984; 1985), possessiveness is the inclination and
tendency to retain control or ownership of one’s possessions, and nongenerosity refers to
an unwillingness to give possessions to or share possessions with others. As for envy,

Belk (1985) adopted Shoeck’s definition of envy as displeasure and ill will at the

29

superiority of another person in happiness, success, reputation, or the possession of
anything desirable. The empirical test (Belk, 1985) indicated that possessiveness,

nongenerosity, and envy were indeed subseales of overall materialism.

Richins and Dawson (1990; 1992) disagreed with Belk’s treatment of materialism
as a set of traits. These researchers observed that traits are personal characteristics that are
formed at the early age, thus they are relatively unchanging over the life span. Also,
underlying traits are generally impervious to environmental stimuli (Richins and Dawson
1990). The conception of materialism from the literature does not fit the accepted
characterization of traits. Therefore, Richins and Dawson (1992) treated materialism as a
value. They suggested that defining materialism as a value is consistent with the notion that
materialism reflects the importance a person places on possessions and their acquisition as a
necessary or desirable form of conduct to reach desired end states, including happiness.

Considering materialism as a value, they proposed three different components of
materialism: Acquisition centrality, acquisition as the pursuit of happiness, and possession-
defined success (Richins & Dawson 1990; 1992). Aguisition centralig refers to the extent
to which one places possession acquisition at the center of one’s life. Acquisition as the
pursuit of hapg’ness is the belief that possessions are essential to satisfaction and well-
being in life. Possession-defined success refers to the extent to which one uses
possessions as indicators of success and achievement in life both in judging themselves and
others. From these propositions, these researchers have developed and tested the measures
of materialism in a series of efforts, resulting in a scale of materialism. The three
subscales and the combined scale have acceptable reliability coefficients and are found to be
free from social desirability bias.

30

Materialism Over the Life Smp

Csikszentmihalyi and Rochberg-Halton (1981) suggested there may be important
differences in materialism that occur during the life span of the individual. They found that
when three generations of a family’s members were asked to name their favorite
possessions and explain the significance of the designated items, the possessions named
and the rationales offered differed systematically across generations. The youngest
(teenager) generation was most likely to name products like stereo equipment that allow
them to do things. The middle generation pointed to a variety of objects from furniture to
trophies that reminded them of accomplishments and shared experiences. The oldest
generation most revered photograph albums and other memorabilia.

Belk (1985) expected that overall materialism would be highest for the middle
generation. Adolescents should value activities more than things, and older persons should
be less egoistic. For similar reasons, the middle generation was also expected to score
highest on his three subscales (i.e., possessiveness, nongenerosity, and envy) of
materialism. Belk (1985) defined the youngest generations as those of 13 years or older,
unmarried, without children, and living with parents; the middle as those married, with
children living in the household, and without grand children; and the oldest generation as
grandparents. His findings showed, as hypothesized, that the middle generation tended to
score highest in both the overall scale and the subscales. However, these scores were not
significantly higher than those of the youngest generation on two (possessiveness and
envy) of the three subscales. Additionally, the magnitudes of difference were somewhat
small. Nevertheless, Belk found that materialism and specific materialistic traits were
weakest among the oldest generation.

31

Acquisition and Mdalism

As an orientation to consumption, materialism has had a fundamental impact on
how people value acquisition. Researchers assumed that high materialistic people value
acquisition and the means to acquire possessions more than those low in materialism, and
that they also value possessions more than other life goals and more than their relationships
with other people (Belk, 1985; Richins & Dawson, 1992). Richins and Dawson (1992)
further assumed that people who desire a lot of possessions will need more money to
acquire those possessions and thus are expected to report a higher desired level of income.
Their empirical tests indicated that respondents high in materialism felt that they needed
significantly more income than those low in materialism. They also found that respondents
high in materialism were more likely to value ”financial security" and less likely to value
"warm relationships with others," and that for respondents low in materialism, four values
were rated as more important than ”financial security." They were self-respect, warm
relationships, family security, and a sense of accomplishment Based on those findings,
Richins & Dawson (1992) concluded that materialists are more likely to value acquisition

and the means to acquire possessions.

In a recent master’s thesis, Yun (1992) developed his propositions on the
relationship between people’s material values and their reactions to scarcity messages. He
assumed that highly materialistic people attain more happiness in their life by acquiring
possessions than do less materialistic people. As such, messages of scarcity should have a
more pronounced effect on highly materialistic individuals because these messages present
a direct threat to their ability to pursue happiness in their life. Yun hypothesized that highly
materialistic people were more likely to like, be attracted to, and value a scarce product than
less materialistic people. However this hypothesis was not confirmed, which was

attributed to lower overall levels of materialism among student subjects.

32

Research Propmitioms and Hmheses

Previous studies have indicated that impulse buying is a typical consumer behavior
in today’s marketplace. It has been defined as purchases for which the decision is made in
the store. Impulse buying is distinct from reasoned or planned purchases and compulsive
or addictive buying. The definition of impulse buying is expended in this study to
characterize major aspects of this phenomenon in TV home shopping settings. In addition
to the core aspect of unplanned purchase, the response time and strength of the impulse
buying in TV home shopping are identified and proposed for investigation. The new
concept of impulse buying is necessary to examine consumer behavior in an interactive

shopping process.

Impulse buying has been found to be affected by situational factors such as types of
stores, location of product display, and shelf space in conventional stores. In TV home
shopping settings, the channel is the store. Thus, TV programming strategies may affect
the viewer’s impulse buying behavior, as do situational factors in ordinary stores. Scarcity
messages, the focus of this study, are one type of TV programming strategy that is
assumed to affect the impulse buying behavior of TV home shoppers.

Different theories have been developed to explain the impact of scarcity messages
on perceived values of scarce items and desire to acquire these items. This study uses the
psychological reactance theory (Brehm, 1966) to justify the relationship between scarcity
messages and impulse buying in TV home shopping. This theory reasons that any event
that makes it difficult for an individual to exercise a freedom constitutes a threat to that
freedom, and threats to freedom typically should produce reactance arousal. A direct

behavioral effect of reactance arousal is an action to restore the threatened freedom. In a

33

shopping setting, impulse buying should occur when a shopper seeks the immediate
satisfaction of getting an item that may become unavailable unless a prompt action is taken.
'16.“ on: ' ' NM 1 ‘1' LL. scar' Henrik L’Vea m'ye' J woe” V- hof 41‘
dimensions a bu in i lse of TV ho e sh i viewers rovided

mg mg mflges are perceivm to 99 true and that the scarce items have mived
value to yigwem. Specifically,

H1: A buying impulse is more likely to occur in a scarcity condition than in a
non-scarcity condition.

H2: Response time of a buying impulse is shorter in a scarcity condition than in
a non-scarcity condition.

H3: A buying impulse is stronger in a scarcity condition than in a non-scarcity
condition.

In addition to a comparison of scarcity and non-scarcity situations, this study also
investigated the differences in buying impulse in the high scarcity and low scarcity
situations. If scarcity messages have a positive influence on buying impulse, then high
scarcity should have a stronger impact on buying impulse than low scarcity . This, the
following hypothesis is proposed:

H4: A buying impulse is more likely to occur in a high scarcity condition than in
a low scarcity condition.

As discussed, there are two types of scarcity messages in TV home shopping
settings — claims of limited quantity of an item and indications of diminishing time
available for action. Although no literature exists to suggest which type of scarcity
messages is more likely to affect impulse buying, this study anticipates that scarcity

messages of time have a stronger influence on buying impulse than scarcity messages of

34

quantity. This proposition is based on a fundamental characteristic of the U.S. economy.
Due to the U.S. huge capacities for mass production, shortage of goods supply has hardly
been an economic fact except for very special events in history, like the sugar crisis of 1974
('Sugar,‘ 1974) and the oil crisis of 1979 ("Energy Togethemess,” 1979). Consumers
have rarely experienced any shortage in the supply of daily goods. On the other hand,
consumers have numerous opportunities for special promotion offers in specified short

periods of time. Consumers have been trained by marketing and advertising practices to
respond to time-related scarcity messages. Therefore, it is r;ea_so_n_g that time scarcity

messages have a stronger imm on bum'ng impulse than guantig scarcig messages.
Specifically,

H5: A buying impulse is more likely to occur in the high scarcity of time
condition than in the high scarcity of quantity condition.

Materialism is an individual’s orientation to the acquisition and consumption of
goods. Studies have found that materialistic people are more likely to value acquisitions
than less materialistic people, and that materialists are more likely to seek happiness from

acquiring possessions than from other activities (Richins & Dawson, 1992). Because

materialism makes Eple more subject to material influence, it is mausible that more

materialistic viewers of TV home shopping Ems are more likely to buy on immlse
than viewers low in materialism. The influence of materialism on each of the three

dimensions of buying impulse is hypothesized:

H6: A buying impulse is more likely to occur in highly materialistic viewers,
somewhat likely to occur in moderately materialistic viewers, and less likely
in least materialistic viewers.

H7: Response time is shortest in highly materialistic viewers, somewhat short in
moderately materialistic viewers, and longest in least materialistic viewers.

35

H8: A buying impulse is strongest in highly materialistic viewers, somewhat
strong in moderately materialistic viewers, and weak in least materialistic
vrewers.

If materialism makes people more subject to material influence, it can be further
anticipated that materialistic people are more likely to be influenced by scarcity messages.
Therefore, it is plausible that materialism and scarcity messages interact to maximize the
buying impulse in viewers who are highly materialistic and in a high scarcity condition.

Thus,

H9. A buying impulse is more likely to occur in highly materialistic viewers who
are in a high scarcity condition than in less materialistic viewers who are in a
low scarcity condition.

In sum, nine research hypotheses are proposed in the study. Among them, five
hypotheses are on the impact of scarcity messages on buying impulse; three on the
influence of materialism on buying impulse; and the last one on an interaction of scarcity

messages and materialism on buying impulse.

36

Endnotes

1 Five studies in Lynn's (Lynn, 1991) meta-analysis produced significant reversals of
predicted scarcity effects, i.e., scarcity's enhancement of value may sometimes be

overwhelmed by concerns about being or appearing selfish, and thus, it reduces people's
desire for the scarce items. This effect is referred to as "reversals of scarcity effects”

(p.47).

CHAPTER 3: METHODOLOGY

This chapter describes the methods used to conduct the experiment, collect and
prepare data, and test the hypotheses outlined in Chapter 2. The hypotheses seek to assess
the influence of scarcity messages on the buying impulse of TV home shopping program
viewers, as mediated by the viewers’ materialistic values. Testing these hypotheses
requires a factorial design in which the effects of two types of scarcity messages (time and
quantity) on buying impulse can be explored simultaneously. The following sections
describes the research design, stimulus materials, subjects, instruments and experimental

procedures, as well as preliminary analyses of data.

Research Desigp
1. Dependent Variable

Buying impulse is the dependent variable in this study. To be consistent with
previous studies of impulse buying in traditional store environments (Cobb & Hoyer,
1986; Lee, 1990; Phillips & Bradshaw, 1993; Rook, 1987), this study defines impulse
purchase in a TV home shopping setting as a purchase that is made by a TV home shopping
program viewer who has not had the intention to buy a product prior to tuning in to the

shopping channel.

Whether a prior intention exists is the difference between an impulse buy and a
planned purchase. The existence of a prior intention can be determined by asking the
shopper if he or she has such an intention before the shopper enters the store (Kollat &
Willett, 1967). The drawbacks of this approach are that the pretest questioning tends to
commit the respondents to fulfill their intentions or it may result in status purchases if the
respondent anticipates inventorying at the end of shopping. Both possibilities may bias
responses (Pollay, 1968)

37

38

Lack of a prior intention can be assumed if the purchase is a new product. Because
the shopper has been unaware of a product before entering the store, purchase of this
product can be assumed to be an impulse buy. This study takes this approach. It uses all
new products in the experiment Thus, an intention to buy any of the products can be
considered an impulse buy.

As discussed in Chapter 2, two new dimensions of buying impulse are proposed in
this study. In addition to whether the viewer wants to buy an unplanned item, the response
time and the strength of an intention also are measured. The three dimensions — intention
to buy an unplanned item, response time of an intention, and the strength of an intention —
are all dependent measures. Measurement of the three dimensions is described in the

instruments section of this chapter.

2. Independent variables

Independent variables are scarcity manipulations and materialism. Two types of
scarcity messages — limited quantity and limited time available for action — were
examined in this study. For each type of scarcity message, two levels of scarcity (either
high or low) were implemented in the experiment. Specifically, the quantity of an item was
either very limited (high scarcity of quantity) or somewhat limited (low scarcity of
quantity). In the same vein, the time available to order an item was either very short (high
scarcity of time) or somewhat short (low scarcity of time). The quantification of scarcity
messages for each condition is discussed in the stimulus section of this chapter. This 2 by
2 factorial design is diagrammed in Figure 1.

This figure shows the magnitude of one dimension of buying impulse (strength of

an intention) in each of the four conditions. In addition to these four scarcity conditions, a

39

control condition also is implemented in the study, whereby no message is given regarding
quantity or time available. This research design makes two comparisons possible. First, a
comparison of buying impulse can be made between scarcity and non-scarcity conditions
(Cells 1, 2, 3, and 4 vs. Cell 5). Second, buying impulse can be compared for high
scarcity and low scarcity conditions (Cells 1 and 3 vs. Cells 2 and 4 for quantity scarcity
and Cells 1 and 2 vs. Cells 3 and 4 for time scarcity). An analysis of variance is

appropriate for this design.

Figure 3-1. The Strength of Buying Impulse by Scarcity Conditions

 

 

 

 

 

Scarcigt of Quantity
High Low None
High Strong Moderate
(Cell 1) (Cell 2)
Scarcity of Time Low Moderate Weak
(Cell 3) (Cell 4)
Weakest
N
one (Cell 5)

 

 

 

Control

4o

Materialism of TV home shopping viewers, another independent variable in the
study, was categorized at three levels: low, medium, and high. Discussion of materialism

measurement is in the materialism scale section of this chapter.

Stimulus Materials

TV home shopping programs from the QVC Network were used in this study. The
reasons for using the QVC instead of the Home Shopping Club programs as the pool of
experimental materials are threefold. First, QVC is commonly perceived as a higher quality
shopping channel because of its soft-selling approach, frequent use of celebrities to present
products and its quality merchandise. QVC programs are more appropriate considering use
of college students as subjects of the study. Second, the QVC Network provides its
customers with a monthly program guide that outlines daily shopping programs by hour.
Such a program guide was not available for the Home Shopping Club channel. The QVC
program guide greatly facilitated the selection and preparation of stimulus materials.
Finally, only the QVC shopping channel is available 24 hours a day on the cable system in
the area where the author lives and works. The local availability of the QVC channel made

it possible for the author to observe its programs on a regular basis.

1. Program Selection

A content analysis of the QVC programming shows that a variety of shopping
programs is available on this shopping channel (for content analysis results, see Appendix
2). In a four-week period from July 26 to August 22, 1993, the QVC channel aired 175
programs, totaling 672 hours. The length of individual programs was from 1 hour, a
standard format, to 24 hours for "Frfth Klondike Gold Rush." The ten most frequent
programs, presented in Table 3-1, were all one-hour shows. Topping the list, "The

Jewelry Showcase” was shown 44 times in this four-week period.

41

Table 3-1. Top Ten QVC Programs

 

 

Length
Program Name (in hours) Frequency
1. The Jewelry Showcase l 44
2. Fun & Leisure 1 36
3. Superbargains 1 31
4. Make Life Easier 1 27
5. Around the House 1 26
6. Now You’re Cooking 1 23
7. The QVC Sampler 1 l7
8. Fashion Coordinates 1 12
9. The Gold Hour 1 10
10. The Ring Showcase 1 9

 

Source: The QVC Program Guide, July 26 - August 22, 1993.

2. Product Selection

"The QVC Sampler," ranking seventh in this tabulation, was selected as the pool of
experimental programs primarily because this program typically sells new products. Eight
QVC Sampler shows were randomly selected from the program guide between March 28
and April 24, 1994 and videotaped from the QVC channel. The eight shows presented 39
different products, with their prices ranging from $13.75 for a 1994 NCAA Final Four T-
Shirt to $1,990.00 for an IBM 486/25 PS1 computer with SVGA monitor.

After a review of the prices of these products, items that cost more than $50 were
excluded from the list because bi g-tag products were less likely to be impulse items and
thus less relevant for this study. This resulted in a list of 29 products as the pool for
identifying products to be used in the experiment.

42

To ensure that experimental products were of general appeal to the subjects, a
pretest was conducted among a sample of students at Michigan State University. Students
were asked to read a list of the 29 products and short descriptions of these products, and to
pick five products that they thought had the greatest value for themselves, considering each
product’s cost and usefulness. Four products were given highest multiple picks and thus
were selected as the experimental products (see Table 3-2).

Table 3-2. Four Selected Products

 

 

Product Price Sample Male Female
1. Rolodex Electronic Planner $16.43 20 13 7
2. Electronic Pedometer $28.60 18 8 10
3. Nutritional Monitor $38.14 16 4 12
4. Skewdriver Pro Kit $45.00 16 13 3

 

Note: The pretest sample consisted of 40 students, with equal numbers of males and
females.

These four products were not gender-free items. Male students preferred the
Rolodex Electronic Planner and Skewdriver Pro Kit whereas female students preferred the
Electronic Pedometer and Nutritional Monitor. The Rolodex Electronic Planner and
Skewdriver Pro Kit were presented by a male host and the Electronic Pedometer and
Nutritional Monitor by a female host The lengths of the program segments for these four

products ranged from 4 minutes 50 seconds to 7 minutes 15 seconds (see Table 3-5 for the

length of each program).

3. Quantification of Scarcity Messages
As discussed in the previous sections, both types (time and quantity) of scarcity
messages were categorized as either high or low. The quantification of the two levels of

scarcity of tim_e was determined by the QVC programming practice. The quantification of

43

the two levels of scarcity of gmttty was determined from a pretest of MSU students who
were asked how many units of a product in TV home shopping program would be

perceived as high in scarcity and low in scarcity.

S_c_atcim of Time. A three-month observation of the QVC programs has indicated
that a count-down timer is frequently used in the QVC shopping programs. When the time
runs out, a different product is usually introduced. For those who are interested in the
previous product, this practice may generate a sense of urgency although viewers can still
order the previous item as long as they know the name or the ordering number of that item.
The length of a count-down timer in the QVC programs ranges from one minute to four
nrinutes. It is assumed that the shorter the time available for action, the stronger the
urgency to act This study adopted the QVC practice, using a one-minute timer in the high

scarcity condition and a four-minute timer in the low scarcity condition.

Scarcity of Quantig. The observation of the QVC programs has revealed that the
quantity of an item available varies greatly across products. For certain collectibles like
baseball bats signed by a sports celebrity or certain artistic works, the inventory was
usually within a few hundred. For other products like kitchenware and jewelry, their
availability was almost unlimited. The QVC program hosts usually mention the quantity of
an item at the beginning of the show if the inventory of that item is limited. The number of
orders for an item is sometimes shown on the screen to indicate the popularity of an item
among viewers. The empirical quantification of the two levels of scarcity of quantity

cannot be determined by any QVC programming practice.

To determine the specific quantity for each of the two levels — high and low
scarcity, a pretest was conducted among seven MSU students. Students were presented a

situational question:

Suppose that you are watching a TV home shopping program, and you see some
products being sold on the show. What quantity of a product will make you think it
is very scarce? Somewhat scarce? By very scarce, we mean that the inventory of a
product is so small that only a small fraction of the program audience is able to buy
one. By somewhat scarce, we mean that not everybody who wants one can get
one.

Then, they were asked to read the features and prices of the four products and write a

number for each scarcity level for each product. The pretest results are in Table 3-3.

Table 3-3. Quantities For High and Low Scarcity. Pretest Results

 

Product/Condition Minimum Maximum Mean SD Low-High Ratio

 

Rolodex Electronic

Planner
High Scarcity 4 100 39 34.2 3.22
Low Scarcity 8 400 129 134.9

Nutritional Monitor
High Scarcity 8 75 26 23.7 3.48
Low Scarcity 10 300 96 110.4

Skewdriver Pro Kit
High Scarcity 6 50 22 16.1 3.78
Low Scarcity 8 200 80 67.6

Electronic Pedometer
High Scarcity 5 75 32 23.2 3.28
Low Scarcity 10 300 103 100.5

 

It was found that students had difficulty in determining the quantity for each
scarcity level because they lacked knowledge of the TV home shopping business and its
sales in the United Statesl. Thus, the numbers they provided seemed to be too small and
unrealistic. However, the pretest was still instructive because a consistent ”low-high ratio"

was found across four products. A low-high ratio was the quotient of the quantity for the

45

low scarcity condition divided by the quantity for the high scarcity condition. The ratios
for the four products were between three and four. That is, a low scarcity item was 3-4
times more available than a high scarcity product In light of these findings, this study
decided to use 500 units and 2,000 units as the quantities for the high scarcity and low

scarcity conditions, respectively.

4. Making the Experimental Programs

A control program was first produwd in the studio of the Department of
Telecommunication at Michigan State University. It consisted of the four products but
contained no scarcity messages. The program began with a QVC logo and then started
presenting the four products one after another, with the same QVC logo between two
product segments. The four products were presented in the following order: Rolodex
Electronic Planner, Nutritional Monitor, Skewdriver Pro Kit, and Electronic Pedometer.

The control program was 26 minutes 5 seconds long.

Four experimental programs were then duplicated from the control program and
superimposed with different scarcity messages of each type. The specific amount of each
type of scarcity message is in Table 3-4.

Table 3-4. Scarcity Messages for the Five Programs

 

 

Time Quantity
Condition 1 1 minute 500 units
Condition 2 1 minute 2,000 units
Condition 3 4 minutes 500 units
Condition 4 4 minutes 2,000 units

Control None None

 

46

Different scarcity conditions were created by superimposing different scarcity
messages in the programs. The subjects in different conditions viewed the programs that
were identical in audio but different in video. Specifically, a phrase "500 Units Available"
or ”2000 Units Available" was superimposed in the program 20 seconds after each product
segment began. The phrase was shown on the left-lower part of the screen, just below the
product information window. The phrase was on the screen for 20 seconds and then
disappeared A one-minute timer or a four-minute timer was superimposed in the program
at such point of time that a product segment will end when the time is out The timer was
located on the right-bottom comer of the screen. The lengths of the programs and scarcity
messages are in Table 3-5.

Table 3-5. Lengths of the Programs and Scarcity Messages
(Minutes : Seconds)

 

Conditions 1 & 2 Conditions 3 & 4

 

Beginning Ending Length Beginning Ending Length

 

Rolodex Planner 0.00 6:40 6:40 000 6:40 6:40
Quantity Message 020 0:40 020 020 0:40 0:20
Time Message 5:40 6:40 1:00 2:40 6:40 4:00

Nutritional Monitor 7:00 11:50 4:50 7:00 11:50 4:50
Quantity Message 7:20 7:40 0:20 7:20 7:40 0.20
Time Message 10:50 11:50 1:00 7:50 11:50 4:00

Skewdriver Kit 12: 10 1925 7: 15 12:10 1925 7: 15
Quantity Message 12:30 12:50 0:20 12:30 12:50 0:20
Time Message 18:25 1925 1:00 15:25 1925 4:00

Electronic Pedometer 1945 26: 05 6:20 1945 26:05 6:20
Quantity Message 20: 05 20.25 0:20 20:05 20.25 0:20

Time Message

25:05

26: 05

1:00

22:05

26: 05

4:00

 

47

The length of the four experimental programs was identical to that of the control
program because they were duplicates of the control program before scarcity messages
were added. Quantity messages for each product appeared at the same time. The length of
the count-down timer was one minute for Conditions 1 and 2, and four minutes for

Conditions 3 and 4. Timers began one minute or four minutes before the end of each

product segment

fleets

The demographics of TV home shopping audiences, reported annually in the
Simmons Market Research Bureau’s Study of Media and Markets, had been reviewed
before decisions were made on who would be the subjects of this experiment. According
to SMRB, 15.6% of the adult population, or 28.5 million people, reported that they
watched TV home shopping programs in 1991, while 72.7% said they did not and 11.7%
did not know. Of the viewers, 6.7 millions, or one in four, said that they had bought one
or more items from TV home shopping programs in the last three months (see the SMRB
statistics in Appendix 3). Similar patterns were found in the 1989 and 1990 SMRB
reports. As for viewing and buying behaviors, they were almost flat among most age,
education and income groups, even though females were more likely to view and buy from
shopping programs than males. Given the exigencies of experimental studies, college

students were the experimental subjects.

Students who were taking undergraduate-level courses in the College of
Communication Arts and Sciences at Michigan State University in Summer 1994 were
selected as the pool of experimental subjects. It also was decided there would be a
minimum of 30 subjects in each of the five conditions. Students were recruited in their

classrooms by the author. Approximately 200 students volunteered for this experiment

I ents

Data were collected with two separate instruments in the experiment Both were
self administered questionnaires: one was a paper questionnaire and the other was a
questionnaire administered on personal computers. The contents of these instruments and

the variables they measured are described below.

1. Buying Impulse
Buying impulse is the dependent variable in the study. It was operationalized as the
viewer’s spontaneous intention to buy a product during a TV home shopping program.

Three dimensions of buying impulse were measured:

1. Whether the viewer has a spontaneous intention to buy a product while
viewing a shopping program;

2. How quickly the intention occurs in a viewing process; and

3. How strong the intention is.

These variables were measured by a computerized questionnaire executed in
CASESZ, a computer-assisted telephone interview program. In this study, this program
was used as a self administered instrtunent. Each subject sat in front of a personal
computer, and, while viewing a product being sold in the program, they were asked to
answer this question shown on the computer screen:

As soon as you decide you want to buy this product, type " 1" (for YES) and press

the "Enter” key. When you decide you do not want to buy it, type "5" (for NO)
and press the ”Enter” key.

This question was asked for all four products. If the answer was yes, then the
subject was asked how strong his or her intention was to buy the product. The response

categories were: very strong, somewhat strong, somewhat weak, very weak, and don’t

49

know. In addition, the computerized questionnaire was designed so that the time the
subject keyed in an answer was automatically recorded in increments of hundredths of
seconds. This resulted in a measure of the ”reaction time?” of the subjects for each product

(the script of the computerized questionnaire is in Appendix 4).

2. Believability of Scarcity Messages

Subjects were asked how believable they thought the information about time limits
and quantity available were. The response categories were: very believable, believable,
somewhat believable, not believable at all, and don’t know.

3. Evaluation of the Products

1 Subjects also were asked to evaluate each of the four products in five aspects on a
seven-point scale. The five aspects were: good/bad, useful/useless, good quality/poor
quality, a bargain/costs too much, and like/dislike. Response categories ranged from 1 to
7, with 4 as the neutral point

4. Materialism

The materialism scale developed by Richins and Dawson (1992) was used in this
study because it was found to be of relatively high reliability and validity, and of virtually
no contamination from social desirability bias. This scale consists of 18 statements,
measuring three factors of the construct — acquisition centrality, acquisition as the pursuit
of happiness, and possession-defined success (Richins & Dawson, 1992 ). Seven
statements measure the guisition centrality: I usually buy only the things I need; I try to
keep my life simple, as far as possessions are concerned; The things I own aren’t all that
important to me; I enjoy spending money on things that aren’t practical; Buying things
gives me a lot of pleasure; I like a lot of luxury in my life; and I put less emphasis on

material things than most people I know. Five statements measure the wuisition as the

50

W: I have all the things I really need to enjoy life; My life would be better
if I owned certain things I don’t have; I wouldn’t be any happier if I owned nicer things;
I’d be happier if I could afford to buy more things; and It sometimes bothers me quite a bit
that I can’t afford to buy all the things I’d like. Six statements measure the asession-
defined success: I admire people who own expensive homes, cars, and clothes; Some of
the most important achievements in life include acquiring material possessions; I don’t
place much emphasis on the amount of material objects people own as a sign of success;
The things I own say a lot about how well I’m doing in life; I like to own things that
impress people; and I don’t pay much attention to the material objects other people own
The response categories for each of the statements were: strongly agree, agree, neutral,

disagree, and strongly disagree.

5. Background Information

Questions were asked about the subjects’ media behavior: how many hours of TV
they watch on a typical weekday and a typical weekend (response categories were: none,
less than 1 hour, 1-2 hours, 34 hours, 5 or more hours, and don’t know); how many
magazines and newspapers they read regularly (none, one, two, three, four or more, and
don’t know); and whether they had ever watched any TV home shopping channels before
(yes or no). Questions also were asked about their purchase history: how many times they
made purchases by mail in the past six months (none, one time, two times, three times,
four or more times) and whether they had ever ordered anything from TV home shopping
programs (yes or no).

Demographic questions asked of the subjects were their gender, their age as of their
last birthday, their major, their monthly spending after paying tuition, rent, utilities and
food (under $99, $100-$199, $200-$299, $300-$399, $400-$499, $500 or more, and
don’t know), and the total yearly income of their parent or parents (under $9,999; $10,000-

51

$19,999; $20,000-$29,999; $30,000-$39,999; $40,000-$49,999; $50,000 or more; and

don’t know).

Exw’ mental Procedures and Data Collection
The experiment was conducted in the Information Technology and Services Lab of

the Department of Telecommunication at Michigan State University. Forty-eight one-hour
experimental sessions were held in 15 days between May 24 and June 15, 1994. Exactly
150 students participated in the experiment as subjects. Among the 48 sessions, four
sessions had eight subjects per session. For most of the sessions, the number of subjects

per session ranged from 2 to 5. There was only one subject in 12 sessions.

1. Experimental Procedures

The CASES application software and the computerized questionnaire were installed
on the computer server and accessible to eight personal computers in the Lab. Each subject
signed a consent form before he or she was assigned an ID for the experiment A session

began with an introduction to the experiment by the author.

In this experiment, you will be watching a TV home shopping program. The
program presents four products. It begins with a QVC logo and then follows the ,
first product. When the first product is over there is another QVC logo, followed by
the second product. This pattern continues for the third and fourth product The
program lasts about 26 minutes. While you are watching the program, you answer
a couple of questions for each product on your computer.

The subjects then were asked to activate a training questionnaire on the computer.
Under the author’s guidance, they went through all the questions they would answer in the
formal experiment After all subjects learned how to handle the computerized
questionnaire, the author started a shopping program videotape on a 3/4 inch tape deck and

52

the program was shown on a 27-inch television set The subjects pressed a key on the
keyboard to activate the formal questionnaire when they heard the very first word from the
program host. The first question was shown on the computer screen and an experimental

session began.

After the shopping program ended and they answered all the questions, the subjects
were given another paper questionnaire to fill out The entire session took approximately

45 nrinutes. The subjects were debriefed after they turned in the second questionnaire.

2. Randomization

The subjects were randomly shown one of the five different shopping programs
(four experimental programs and one control program). The order in which the five
programs were shown was determined initially with random numbers. After each of the
five programs was shown once, the program that was viewed by fewest subjects was
shown next. Thus, the number of subjects for each condition accumulated approximately

evenly.

3. Subject Characteristics

Among the 150 students who participated in the study, 55% were females. The age
of the sample concentrated on 20 (19%), 21 (29%) and 22 (23%). Advertising majors
accounted for 32% of the subjects; other majors in the College of Communications Arts and
Sciences (communication, journalism, speech, and telecommunication) 43%; business

majors 5% and all other majors 19%. A profile of the subjects is in Table 3-6.

53

Table 3-6. Subject Characteristics

 

Frequency Sample Percentage

 

 

9!;an
Nhle 67 45%
Female 83 55
Ass
18 - 19 12 8
20 28 19
21 44 29
22 34 23
23 and Over 29 19
Missing 3 2
Major
Advertising 48 32
Com/Joumalism/Speechfl‘elecom 64 43
Business 8 5
Other 29 19
Missing 1 1
Preliminary Data Analysis

Data automatically recorded in the computerized instrument were outputted using
the CASES program when the experiment was completed. Data from the paper
questionnaires were keypunched. The two data sets were merged through SPSS for
Windows (version 6.0) and evaluated prior to analysis. As indicated by Tabachnick and
Fidell (1983), the first step is to inspect out-of-range values, plausible means and
dispersions, and variation for accuracy of input Frequencies for all variables were run to
check for keypunch errors. Variables with suspicious distributions were checked for

accuracy of input

The next step is to examine size and distribution of missing data. If only a few
units of data are missing from a large data set, the problem will not be serious and almost
all procedures for handling them will yield similar results (Tabachnick & Fidell, 1983).
Although there are no rules regarding how much missing data can be tolerated for a given
size of sample, it was decided that for this study of 150 subjects, variables with 5% or less
data missing should not present serious problems for data analysis. Using this criterion,

no variables presented serious problems.

The third step is to identify and deal with outliers. Outliers are cases with such
extreme values on one variable that they unduly affect the average value or the variability of
the scores. Because all questions in both instruments of this study were close-ended there

were no outliers.

1. Data Transformation

Measures of the subject’s response time for each of the four products were
automatically recorded in intervals of a hundredth of a second. These variables were
converted into the number of minutes with two decimal places. Such a conversion made
the measure more intuitive, and easier for comparisons with the lengths of the experimental

programs and scarcity messages.

2. Product Evaluation Indices

The subjects were asked to evaluate each of the four products in five aspects on a
seven-point scale. The five aspects are: good/bad, useless/useful, good quality/poor
quality, costs too much/a bargain, and like/dislike. The direction of two pairs of the
adjective phrases (useless/useful and costs too much/a bargain) was deliberately reversed to
avoid response set bias. The score for each aspect was so coded that the higher the score,

55

the more positive the evaluation. The scores for all five aspects were summed into a
product evaluation index. This study categorized a product evaluation index at three levels:
negative evaluation (scored 0 to 17.5), neutral evaluation (17.6 to 22.5), and positive

evaluation (22.6 to 35)4. The number of subjects at each level is in Table 3-7.

Table 3-7. Product Evaluation Indices

 

 

Scale
Evaluation Reliability
Product Negative Neutral Positive Missing a
1. Rolodex Electronic Planner 39 36 70 5 .85
2. Electronic Pedometer 57 53 36 4 .83
3. Nutritional Monitor 41 44 60 5 .81
4. Skewdriver Pro Kit 77 37 32 4 .86

 

3. Materialism Measurement

This study used the materialism scale developed by Richins and Damson (1992) to
measure the materialistic values of the subjects. This scale consists of 18 statements,
measuring three factors of materialism. The validity, reliability and the factorial structure of
the scale have been tested in the study by Richins and Dawson (1992). To verify the
validity and reliability of the scale, a series of confirmatory factor analyses was conducted
in this study.

Confirmatory factor analyses (CFA) allow the researchers to specify expected
dimensions and determine how well the given items fit the theoretical measurement models
(Hunter & Gerbing, 1982). Several statistical packages are available to conduct
confinnatory factor analyses. Among them are PACAGE (Hunter & Lim, 1987), EQS

56

(Bentler, 1989), and LISREL (Jereskog & Scrbom, 1989)5. This study used LISREL 7.2
in SPSS for Windows.

LISREL begins a CFA with a priori specification of the factorial structure of a
measurement model (Jbreskog & Scrbom, 1989). It reproduces a series of parameters from
the covariance matrix of the observed variables by the method of maximum likelihood5.
These parameters can be used to assess a measurement model in terms of both global and
local fit to the data. Key parameters of the global-fit test are chi-square and its probability
for a given degrees of freedom. If a measurement model fits the data, the chi-square
should be relatively small, indicating that the covariance of observed variables are not
significantly different from the theoretical variance of these variables given the specified
factorial structure. T-values can be used to assess the significance of factor loading of
individual items on specified factors or latent variables. Parameters whose t-values are
larger than two in magnitude are normally judged to be different from zero (J'oreskog &
Scrbom, 1989).

LISREL also produces a table of modification indices from which noisy items can
be easily identified when the model does not fit the data Deleting a noisy item from the
model or setting the offending item free — assuming that item also measures some factor
other than the one it is supposed to measure — usually, but not always, decreases the
overall chi-square and leads to the improvement of the model’s fit After a noisy item is
dealt with, the CFA procedures should be repeated.

Table 38 presents the initial results of a confirmatory factor analysis of materialism
variables. Although the factor loadings of most items are relatively strong and only two are
smaller than .30, the chi-square was quite large, indicating that the global fit of the model

was not good. Some revision of the model is thus necessary.

Table 38. Items and Factor Loading of Materialism: Initial Results

 

 

Item Factor I Factor 11 Factor III
Factor I: Aguisition Centrality
I like a lot of luxury in my life. .706
Buying things give me a lot of pleasure. .650
I try to keep my life simple, as far as possessions
are concerned. .564
I usually buy things I need. .515
I enjoy spending money on things that aren’t .431
practical.
I put less emphasis on material things that most
people I know. .320
The things I own aren’t all that important to me. .250
Factor II: Amuisition as the Pursuit of Happjness
My life would be better if I owned certain things I
don’t have. .650
I’d be happier if I could afford to buy more things. .632
It sometimes bothers me quite a bit that I can’t
afford to buy all the things I’d like. .610
I wouldn’t be any happier if I owned nicer things. .395
I have all things I really need to enjoy life .274
Factor III: Possession-Defined Success
I like to own things that impress people. .925
I admire people who own expensive homes, cars, .698
and clothes.
I don’t place much emphasis on the amount of .657
material objects people own as a sign of success.
The things I own say a lot about how well I’m .621
doing in my life.
Some of the most important achievements in life .577
include acquiring material possessions
I don’t pay much attention to the material objects
other people own. .486

 

Note: Chi-Square (132 d.f.) = 261.52 (p = .000).

58

The table of modification indices was used to identify offending items in the model.
Six items with the largest residuals or weak factor loadings were detected and dropped in
the sequential CFA runs and eventually, the analysis resulted in a 12-item measurement
model of materialism that fits the data (see Table 3-9). The chi-square of the new model
was 58.57, with the probability of .217. The 12 items accounted for 95.5 of the total
variance of the model. As for the local fit of individual items, the t-value associated with
each item ranged from 4.1 to 10.5, indicating that all factor loadings were significantly
different from zero (Jereskog & Sorbom, 1989). The correlations were .61 between
acquisition centrality and acquisition as the pursuit of happiness, .87 between acquisition
centrality and possession-defined success, and .60 between acquisition as the pursuit of

happiness and possession-defined success.

The reliability alpha was .63 for the subscale of acquisition centrality, .61 for the
acquisition as the pursuit of happiness, and .78 for the possession-defined success. A
combination of the three subscales resulted in a more reliable measure of materialism (alpha
= .83) Therefore, this study used the overall scale as the measure of materialism.

59

Table 3-9. Items and Factor Loadings of Materialism: final Results

 

Item Factorl FactorII FactorIII Alpha

 

Factor I: Aguisition Centralig .63
I like a lot of luxury in my life. .767

Buying things give me a lot of pleasure. .627
I try to keep my life simple, as far as

possessions are concerned. "‘ .472
I put less emphasis on material things that

most people I know.* .312

Factor 11: Aguisition as the Pursuit of

Happiness .61
I’d be happier if I could afford to buy more .737

things.
My life would be better if I owned certain .626

things I don’t have.
I wouldn’t be any happier if I owned nicer .450

things)“

Factor III: Possession-Defined Success .78
I like to own things that impress people. .938
I admire people who own expensive homes,

cars, and clothes. .712
I don’t place much emphasis on the amount

of material objects people own as a sign .654

of success
The things I own say a lot about how well

I’m doing in my life. .598
I don’t pay much attention to the material

objects other people own.* .498

Correlatiops between Factors
Factor 11 .61

Factor III .87 .60
Overall Scale .83

 

Note: An asterisk indicates reverse scored items.

Chi-Square (51 d.f.) = 58.57 (p = .217); GFI Index = .94; AGFI Index = .91;

Each item was so scored that the higher the score, the higher materialism it

represents. Scores of all 12 items were then summed to make the scale score. Because

60

each item was measured on a seven-point scale, the possible sum of the 12 items should
range from 12 to 84. However, the actual sum ranged from 18 to 56, with a mean of
35.85 and standard deviation of 7.22. As stated in the preceding section, materialism is
eategorized into three levels: low, medium, and high. The cutoff points for categorization
were decided by the percentile ranking method (McClave & Dietrich 1985). The
materialism scores associated with the commutative 33.3% and 66.6% of the valid
frequency were used as the cut for low level and high level, respectively. The numbers of

subjects in each level and the ranges of subscale scores are in Table 3-10.

Table 3-10. Materialism Scores by Level

 

 

 

Materialism Score
Level Subjects Minimum Maximum Mean SD
Low 52 18 33 28. 1 4.0
Medimn 48 34 39 36.4 1 .8
High 47 40 56 43.9 3.3

 

The numbers of subjects were relatively even across three levels and the mean
differences among the three levels were approximately 8 points. Standard deviations
between the low level and high level were very similar whereas the standard deviation of

the meditun level was relatively small.

4. Randomization Check
As described in the preceding section, the subjects were randomly shown one of the
four experimental programs and one control program. To examine the outcome of this

procedure, comparisons were made across five conditions on the subjects’ media usage,

61

purchase history, monthly spending, income of parent or parents and gender. The results

are shown in Table 3-11.

Table 3-11. Characteristics of the Subjects in Five Conditions

 

 

F/xZ
Cond. 1 Cond. 2 Cond. 3 Cond. 4 Control

Watch TV on Weekday 3.33 3.48 3.42 3.59 3.47 .319
Watch TV on Weekend 3.26 3.28 3.42 3.32 3.23 .1450
Read magazines 2.76 2.70 2.61 2.66 2.67 .0578
Read Newspapers 2.10 2.17 2.23 2.21 2.24 .1372
Purchases by Mail 2.23 1.73 2.23 1.93 1.80 1.1329
Orders from TV Programs 13.8% 10.0% 6.5% 17.2% 10.0% 2.011
Watch Shopping Programs 79.3% 73.3% 80.6% 79.3% 73.3% .86491
Gender: females 48.3% 63.3% 48.4% 62.1% 56.7% 2.516
Monthly Spending 2.93 2.63 2.26 2.92 2.25 1.273
Income of Parent(s) 5.04 5.08 5.14 5.08 5.21 .0653

 

Note. Response categories of the ten variables in the table are in the instruments section of

this chapter.

No significant differences across all five conditions exist for the ten variables. It

can be assumed that the randomization was successful and that the subjects in the five

conditions were not significantly different from each other. Therefore, differences in the

subjects’ buying impulse are attributable to the manipulations — scarcity messages of

different levels.

62

Endnotes

1 In 1993, the QVC shopping programs reached 45.8 million households; the Home
Shopping Club programs reached 62.5 million households, according to the 1993 annual
reports of these two companies.

2 CASES (Version 3.5, July 1992), the Computer-Assisted Survey Execution System, is
developed by the Computer-assisted Survey Methods Program (CSM) at the University of
California at Berkeley.

3 Reaction time is a standard measure in cognitive and perceptual psychology. It is
commonly defined as the time from the onset of a stimulus to the time the subject responds
(James, et al., 1992).

4 A sum of all five 7-point subscales resulted in an index score ranged from 5 to 35.
Instead of using 20 as the neutral point, this study used 17.6 to 22.5 as the range of neutral
evaluation for the index.

5 Statistical methods underlying confirmatory factors analysis are different for these three
packages. PACKAGE uses the least square method whereas LISREL uses the maximum
likelihood method. EQS gives users options for the least square, maximum likelihood, and
other several methods.

6 For a brief discussion of the maximum likelihood method, see Pedhazur (1982) in page
638.

CHAPTER 4: RESULTS

This chapter presents three types of results. It begins with a description of the
characteristics of buying impulse of the experimental subjects. Because this study is one of
the initial attempts to investigate the buying impulse in TV home shopping settings, it is
constructive to conduct some descriptive armlyses of this phenomenon. Next, the main
body of this chapter tests our hypotheses. finally, this chapter presents some exploratory
analyses beyond hypothesis testing as to the causality of buying impulse in TV home
shopping settings.

Bum'ng Immtse in General: Descrim've Findings
This section describes the characteristics of buying impulse for the entire sample.

As discussed in preceding sections, three dimensions of buying impulse are examined in
this study: 1) whether the viewers intend to buy a product while viewing the program; 2)
how quickly an intention occurs in the viewing process; and 3) how strong the intention is
to buy a product The percentages who experienced the impulse to buy each product are in
Table 4-1. Of the 150 subjects, 32% (48 subjects) intended to buy a Rolodex Electronic
Planner whereas only 19% (29 subjects) wanted to buy a Nutritional Monitor.

Table 4-1. Buying Impulse for Each Product

 

 

Product n Percent
1. Rolodex Electronic Planner 48 32%
2. Nutritional Monitor 29 19
3. Skewdriver Pro Kit 35 23
4. Electronic Pedometer 35 23

 

Note: The sample consisted of 150 students, of whom 83 were female.

A number of subjects had a buying impulse for more than one product, as shown in
Table 4—2. Numbers on diagonal were numbers of subjects who want to buy that particular
item whereas numbers off diagonal were numbers of subjects who also want to buy other
items. For instance, among 48 subjects who intended to buy a Rolodex Electronic Planner,
15 also intended to buy a Nutritional Monitor; 10 also intended to buy a Skewdriver Pro
Kit; and 17 also intended to buy an Electronic Pedometer. In the same fashion, of 29
people who intended to buy a Nutritional Monitor, 15 also intended to buy a Rolodex
Electronic Planner while 6 also intended to buy a Skewdriver Pro Kit, etc. Multiple
purchases existed for the two other products. Of the entire sample, 40% intended to buy
none of the products, 33% intended to buy one product, 18% intended to buy two
products, and 9% intended to buy three or four products.

Table 4-2. Multiple Purchases of the Sample
(I!)

 

 

1 2 3 4
1. Rolodex Electronic Planner 48
2. Nutritional Monitor 15 29
3. Skewdriver Pro Kit 10 6 35
4. Electronic Pedometer 17 14 10 35

 

The second dimension of buying impulse is response time -— how quickly an
intention to buy occurs in a TV home shopping viewing process. In the experiment, the
time the subjects keyed in whether or not they wanted to buy a product was automatically
recorded in the computerized instrument Response time for each product was exactly the
time from the beginning of a product segment to the time the subject indicated a product
purchase intention. Table 43 presents the findings as to how quickly the subjects made

their decisions.

65

Table 4-3. Response time for Two Different Decisions
(Minutes : Seconds)

 

 

 

Who Decided Who Decided
to Buy Not to Buy
Segment
Product Length Time n Time n t

l. Rolodex Electronic Planner 6:40 3:31 48 2:29 102 3.02"
2. Nutritional Monitor 4:50 3: 18 29 2: 16 121 3.72“"
3. Skewdriver Pro Kit 7: 15 2:46 35 2:07 115 2.09"

4. Electronic Pedometer 6:20 2:37 35 2:01 1 15 2. 1 1*

 

*p < .05; “p < .01; "*p < .001.

There are three interesting findings on the response time of buying impulse. First,

the subjects had made their decisions considerably before a product segment ended. For

instance, the Rolodex Electronic Planner segment lasted 6 minutes and 40 seconds and the

subjects made their decisions in the first 2.5 to 3.5 minutes. The subjects did not need to

view the entire segment to make their purchase decisions.

Second, it took significantly more time to decide to buy a product than to decide _rr_et©_

m buy a product For example, it took the subjects an average of 3 minutes 31 seconds to

decide to buy a Rolodex Electronic Planner whereas it took the subjects only 2 minutes 29

seconds to decide not to buy the same product This pattern was consistent across all four

products.

Finally, as Table 4-3 indicates, the subjects tended to accelerate their decisions for

subsequent products. That is, they could make quicker and quicker decisions as the

viewing process continued. To test the significance of these differences as to how quickly

66

a decision was made, t-tests were conducted for each pair of the products. Substantial
differences existed for products that were not adjacent For instance, it took less time for
the subjectstomakeadecision forthethirdproductthanthefirstone,and tomakea
decision for the fourth product than the second one. This finding suggests there is a

learning curve in the experiment

The third dimension of buying impulse is how strong an intention is to buy a
product. The subjects who indicated that they had an intention to buy a product were asked
the question: "How strong is your intention to buy the product?" The subjects who had no
intention to buy a product skipped this question. Table 4-4 presents the findings on this

question for all four products.

Table 44. Strength of the Intention to Buy a Product
(96)

 

Very Somewhat Somewhat Very

 

Product n Strong Strong Weak Weak DK
1. Rolodex Electronic Planner 48 15% 48% 31% 6% 0%
2. Nutritional Monitor 29 21 55 24 0 0
3. Skewdriver Pro Kit 35 18 44 35 3 3
4. Electronic Pedometer 35 17 51 14 11 6

 

Among the subjects who wanted to buy a product, 44% to 55% had a "somewhat
strong” intention. A sum of ”very strong” and ”somewhat strong” ranged from 62% (for
the Skewdriver Pro Kit) to 76% (for the Nutritional Monitor). This finding suggests that if
only strong or very strong intention to buy an unplanned product is categorized as buying
impulse, as Rook (1987) proposed, then buying impulse so defined will still account for a
large proportion of purchases in TV home shopping settings.

67

Eating the Hymtheses
1. Scarcity Messages and Buying Impulse

The first three hypotheses address the issue as to whether scarcity affects the
buying impulse in TV home shopping settings. The components of these hypotheses are
three dimensions of a buying impulse (an intention to buy an unplanned item, response
time, and strength of an intention) and two scarcity conditions (the scarcity condition
versus non-scarcity condition). Tests were conducted to examine the influence of each type
of scarcity message — time and quantity — for all four products in the two scarcity

conditions.

Hypothesis 1 maintains that a buying impulse is more likely to occur in a scarcity
condition than in a non-scarcity condition. To test the hypothesis, the four scarcity
conditions (high scarcity of time, low scarcity of time, high scarcity of quantity, and low
scarcity of quantity) were combined into the scarcity condition and compared with the
control condition (non-scarcity condition). Crosstabulations were made of the occurrences
of buying impulse in these two conditions. The findings in Table 45 indicate that no chi-

squares were significant. Hypothesis 1 was not confirmed.

Hypothesis 2 asserts that the response time of a buying impulse is shorter in the
scarcity condition than in the non-scarcity condition. The t-test results of the response time
in these two conditions, presented in Table 46, indicate that the response time in the
scarcity condition was shorter than in the non-scarcity condition and this pattern was
consistent across all four products. However, only the difference in response time for the
Nutritional Monitor was significant (t = 3.36; p < .01). A mean z-scorel for all four
products was calculated and analyzed in a t-test. The results (t = 1.72 and p = .089) are in
Table 4-6. These findings suggest that scarcity messages do affect the response time but
this effect was relatively weak in this study.

Table 4—5. Intention to Buy in Scarcity and Non-Scarcity Conditions

 

Product Scarcity

Non-Scarcity

 

X2
1. Rolodex Electronic Planner 31% 37% .375
2. Nutritional Monitor 21 13 .866
3. Skewdriver Pro Kit 23 27 .233
4. Electronic Pedometer 24 20 .233

 

Table 46. Response Time in Scarcity and Non-Scarcity Conditions

(Minutes: Seconds)

 

 

Product Scarcity Non-Scarcity t
1. Rolodex Electronic Planner 3:16 4:21 1.52
2. Nutritional Monitor 3:11 4:04 3.36"
3. Skewdriver Pro Kit 2:39 3:11 .88
4. Electronic Pedometer 2:33 2:58 .74
Mean Z-Score -.0396 .3438 1.72

 

**p< .01

Hypothesis 3 declares that a buying impulse is stronger in the scarcity condition

than in the non-scarcity condition. T-tests were conducted to compare the strength of

buying impulse in the scarcity and non-scarcity conditions. The results are in Table 4-7.

69

Q

Table 4-7. Strength of Buying Impulse in Scarcity and Non-Scarcity Conditions

(1 = very weak; 2 = somewhat weak; 3 = somewhat strong; 4 = very strong)

 

 

 

Product Scarcity Non-Scarcity t
1. Rolodex Electronic Planner 2.8 2.5 .77
2. Nutritional Monitor 3.1 2.3 2.46"
3. Skewdriver Pro Kit 2.8 2.6 .57
4. Electronic Pedometer 2.8 2.8 .14
Mean Z-Score -.0331 -.3229 1.32
*p < .05

The findings in Table 4-7 show that the buying impulse in the scarcity condition
was stronger than in the non-scarcity condition for three of the four products. For these
three products, the buying impulse was ”somewhat strong" in the scarcity condition
whereas it was between "somewhat weak” and "somewhat strong” in the comparison
condition. However, only the strength of buying impulse for the Nutritiorml Monitor was
significantly different between the scarcity and non-scarcity conditions, and the overall
effect of scarcity messages was relatively weak, as indicated by the mean z-score ( t = 1.32
and p = .189) in Table 47. Given these findings, it is prudent to conclude that Hypothesis
3 is partly confirmed in this study.

2. High Scarcity and Low Scarcity

Hypothesis 4 asserts that a buying impulse is more likely to occur in a high scarcity
condition than in a low scarcity condition. This hypothesis is tested in both quantity
scarcity and time scarcity conditions. Hypothesis 5 maintains that a buying impulse is

more likely to occur in the high scarcity of tim_e condition than in the high scarcity of

70

quantity condition. With a factorial design, these two hypotheses can be tested in one 2-
way analysis of variance. Table 48 presents the ANOVA results.

An inspection of the cell means in Table 48 shows there were no significant
segment interactions between the scarcity of quantity and scarcity of time on the buying
impulse for any product Thus, we can further examine the main effects of both types of
scarcity to test Hypothesis 4. No significant differences were found in the buying impulse
between the high and low scarcity of gua_n_tig( conditions. However, it seems that the high
scarcity of time had an impact on the buying impulse. The impact of high scarcity of time
on impulse buying was consistent across both high scarcity of quantity and low scarcity of
quantity conditions for seven of eight possible comparisons. An exception existed only in
the low scarcity of quantity condition for the Skewdriver Pro kit where the two means (.24
and .23) were virtually identical. Nevertheless, these differences were significant only for
one product, the Nutritional Monitor. For this product, high scarcity of time had a
significant impact on the buying impulse. An analysis of variance of the mean z-score of
all four products (see Table 4-8) suggests that high scarcity of time had a significant overall
impact on the buying impulse for all four products. Thus, it can be concluded that
Hypotheses 4 was confirmed in the time scarcity condition but not in the quantity scarcity
condition. Given the findings that high scarcity of time had a significant impact on the
buying impulse whereas high scarcity of quantity had no significant impact on any product,
it is concluded that Hypothesis 5 also was confirmed.

Table 48. Buying Impulse in High and Low Scarcity Conditions

71

(O = not intend to buy; 1 = intend to buy)

 

 

 

 

 

 

Scarcity of Quantity
Scarcity
Product of Time Low High Mean E
1. Rolodex Electronic Low .28 .19 .23 Main Effects 1.593
Planner
High .33 .43 .38 Quantity .011
Time 3.180
Mean .31 .31 Interaction 1.172
Explained 1.452
2. Nutritional Monitor Low .14 .13 .13 Main Effects 3535*
High .40 .17 .28 Quantity 2.791
Time 4.163*
Mean .27 . 15 Interaction 2.394
Explained 3.155*
3. Skewdriver Pro Kit Low .24 .16 .20 Main Effects .255
High .23 .27 .25 Quantity .091
Time .413
Mean .24 .21 Interaction .539
Explained .350
4. Electronic Pedometer Low .24 . 13 .18 Main Effects 2.363
High .37 .23 .30 Quantity 2.484
Time 2.163
Mean .31 . 18 Interaction 1.581
Explained 1.581
Mean Z-Score Low -.05 -.21 -.13 Main Effects 4480*
High .22 .06 . 14 Quantity 2.244
Time 6585*
Mean .09 -.08 Interaction .002
Explained 2.987*

 

 

*p < .05.

72

3. Scarcity Messages and Materialism

Hypothesis 6 states that a buying impulse is more likely to occur among highly
materialistic viewers, somewhat likely to occur in moderately materialistic viewers, and less
likely in least materialistic viewers. One-way analysis of variance of the buying impulse
across three levels of materialism tests this hypothesis. Table 49 presents the results.

Table 49. Impact of Materialism on Buying Impulse
(0 = Not intend to buy; 1 = intend to buy)

 

 

Materialism
Product Low Medium High F
1. Rolodex Electronic .38 .27 .30 .811
Planner

2. Nutritional Monitor ,10b .15b .3421 5.524"
3. Skewdriver Pro Kit .15 .23 .34 2.414
4. Electronic Pedometer . 15 .21 .36 3185*

Mean Z-Score -.12b -.07b 22a 5.082"

 

Note. Different superscripts denote significant mean differences.
*p < .05; **p < .01.

Table 49 shows that a linear and significant impact of materialism on the buying
impulse existed for two of the four products, the Nutritional Monitor and Electronic
Pedometer. The impact of materialism also was linear for the Skewdriver Pro Kit but not
significant. A Scheffe test was conducted to examine the pairwise differences between
three levels of materialism for the Nutritional Monitor and Electronic Pedometer. The
findings are that for the Nutritional Monitor, high materialistic subjects were more likely to
buy on impulse than moderately materialistic and low materialistic subjects, and that there

was no significant difference between moderately materialistic subjects and low

73

materialistic subjects. No pairwise differences were found for the Electronic Pedometer.
An analysis of variance of the mean z-score for the four products (see Table 49) showed
that high materialism had a stronger impact on the buying impulse than low and moderate
materialism but again, no significant difference existed between low and moderately
materialistic viewers. Therefore, it is concluded that Hypothesis 6 is only partially
supported by the data.

Hypothesis 7 claims that the response time is shortest in highly materialistic
viewers, somewhat short in moderately materialistic viewers, and longest in least
materialistic viewers. One-way analyses of variance were conducted to test this
hypothesis. The results are in Table 4— 10.

Table 4-10. Response Time by Materialism
(Minutes : Seconds)

 

 

Materialism
Product Low Medium High F
l. Rolodex Electronic Planner 4:21 2:37 2:59 3514*
2. Nutritional Monitor 3:29 3:27 3:07 .326
3. Skewdriver Pro Kit 2:36 2:41 2:56 .164
4. Electronic Pedometer 2.31 2:23 2:49 .438
Mean Z-Score .2814 -.0772 -.0961 1.65

 

*p < .05.

The findings in Table 4— 10 shows that materialism has a significant impact on
buying impulse only for one product, the Rolodex Electronic Planner. However, the

shortest response time was for the moderately materialistic viewers instead of the high

74

materialistic viewers as proposed. A Scheffe pairwise comparison failed to identify any
significant differences between high materialism and moderate materialism or between
moderate materialism and low materialism. The mean z-score shows no significant
difference in response time across three materialism levels (F = 1.65 and p = .199) Thus,
Hypothesis 7 was not confirmed.

Hypothesis 8 maintains that a buying impulse is strongest in highly materialistic
viewers, somewhat strong in moderately materialistic viewers, and weak in least
materialistic viewers. The strength of buying impulse was compared across three levels of
materialism through one-way analyses of variance. The findings, presented in Table 411,
indicate that materialism had no significant impact on the strength of buying impulse for all
product Hypothesis 8 was not confirmed.

Table 411. Strength of Buying Impulse by Materialism

(1 = very weak, 2 = somewhat weak, 3 = somewhat strong, 4 = very strong.)

 

 

Materialism

Product Low Medium High F

1. Rolodex Electronic 2.7 2.8 2.7 .085
Planner

2. Nutritional Monitor 3.2 3.0 2.9 .280

3. Skewdriver Pro Kit 3.0 3.0 2.5 2.084

4. Electronic Pedometer 2.7 2.9 2.8 .132

Mean Z-Score -.0311 -.0182 -.1884 .373

 

This study assumes that materialism adds to the size of effect of scarcity messages
on the buying impulse of the TV home shopping program viewers. Hypothesis 9

75

nraintains that a buying impulse is more likely to occur in highly materialistic viewers who
are in a high scarcity condition than in less materialistic viewers who are in a low scarcity
condition. To test this hypothesis, materialism was categorized at two levels (high and
low). A three-way analysis of variance was conducted for all four products. No
significant main effects and interactions were found for the Rolodex Electronic Planner and
Skewdriver Pro kit; their means of impulse buying are reported in Table 412. Tables 413
and 414 show the three-way ANOVA results for two other products — the Nutritional

Monitor and Electronic Pedometer, respectively.

Table 412. Buying Impulse by Scarcity and Materialism: Rolodex Electronic Planner and
Skewdriver Pro Kit

(0 = Not intend to buy; 1 = intend to buy)

 

 

Low Materialism High Materialism
Low High Low High
Scarcity of Time Scarcity of Scarcity of Scarcity of Scarcity of

Quantity Quantity Quantity Quantity

 

Rolodex Electronic Planner

Low .23 (n=13) .29 (14) .27 (15) .12 (17)

High .30 (10) .62 (16) .35 (20) .23 (13)
Skewdriver Pro Kit

Low .31 .0 .20 .29

High .10 .19 .30 .38

 

The most striking finding in Table 412 was there was a significant interaction
between the scarcity of quantity and materialism. An examination of the cell means of these
two variables suggests that a buying impulse for the Nutritional Monitor was most likely to
occur among high materialistic subjects in the high scarcity time but low scarcity of quantity

condition. This finding was replicated for the Electronic Pedometer (see Table 413).

76

These findings did not support Hypothesis 9 which states that a buying impulse is more

likely to occur in high materialistic viewers in the high scarcity condition.

Table 413. Buying Impulse by Scarcity and Materialism: Nutritional Monitor
(0 = Not intend to buy; 1 = intend to buy)

 

 

 

 

Low Materialism High Materialism
Low High Low High
Scarcity of Time Scarcity of Scarcity of Scarcity of Scarcity of
Quantity Quantity Quantity Quantity
Low .0 (n: 13) .21 (14) .27 (15) .06 (17)
High .10 (10) .25 (16) .55 (20) .08 (13)
E.
Main Effects 3.174*
Quantity 2.463
Time 4384*
Materialism 1.848
2-Way Interactions 5136"
Quantity and Time 1.514
Quantity and Materialism 13.199***
Time and Materialism .302
3-way Interaction 493

Explained 3:632"

 

Table 414. Buying Impulse by Scarcity and Materialism: Electronic Pedometer
(0 = Not intend to buy; 1 = intend to buy)

 

 

Low Materialism (11) High Materialism (n)
Low High Low High
Scarcity of Time Scarcity of Scarcity of Scarcity of Scarcity of

Quantity Quantity Quantity Quantity

 

 

 

Low .15 (13) .21 (14) .33 (15) .06 (17)
High .20 (10) .38 (16) .45 (20) .08 (13)
E
Main Effects 1.627
Quantity 2.545
Time 2.180
Materialism .032
2-Way Interactions 2.584
Quantity and Time .000
Quantity and Materialism 7692"
Time and Materialism .066
3-way Interaction .454
Explained 1.870
e o thesis Testin

This study basically takes a behavioral approach. One assumption underlying this
approach is that all behavior is controlled by environmental factors (Peter & Olson, 1990) .
Thus, this study focuses on the influence of scarcity messages on the buying impulse of
TV home shopping viewers. Although this study included the viewer’s materialism as an
independent variable, the fact that many hypotheses were not confirmed suggests a need to
include more cognitive factors in the causal analysis of the buying impulse in TV home
shopping settings. This study measured some other cognitive variables (product

evaluation, believability of scarcity messages, etc.) that can be used in a further exploration

78

of the causality of the buying impulse. This section presents the results of some
exploratory analyses that were not proposed in the research hypotheses.

1. Product Evaluation and Buying Impulse

As noted in Chapter 3, this study asked the subjects to evaluate each of the four
products after they had viewed the shopping program. Evaluations were made on five
aspects of a product and the scores were summed into a product evaluation index. The
index was further categorized at three levels: negative, neutral and positive (see Chapter 3).
An exploratory analysis of the relationships of these indices and the buying impulse
revealed some significant findings which are presented in Table 415.

Table 415. Buying Impulse by Product Evaluation
( 0 = Not intend to buy; 1 = intend to buy)

 

 

Evaluation
Product Negative Neutral Positive E
l. Rolodex Electronic Harmer .051) .06b 59a 3 5078*"
2. Nutritional Monitor ,00b .1 1b .47a 22,455":
3. Skewdriver Pro Kit 06“ .27b 59%! 23.205***
4. Electronic Pedometer .02b .09b .583 38.837***

 

Note. Different superscripts denote significant mean differences
***p < .001.

These findings indicate that the positive product evaluation had a significant
influence on the buying impulse of the subjects. That is, the subjects whose evaluation of a
product was positive are more likely to buy that product on impulse. One-way analyses of

variance also were conducted to examine potential differences in two other dimensions of

79

buying impulse — the response time and strength of buying impulse — at different levels
of product evaluation. No significant effect of product evaluation was found on these two
dimensions of the buying impulse for any product. Further, a three-way ANOVA was
conducted but no significant interactions were found among the scarcity of time, scarcity of
quantity, and product evaluation. The main effect of product evaluation was significant on
the buying impulse for all four products.

2. Believability of Scarcity Messages

This study measured how believable scarcity messages in TV home shopping
program were to the subjects who viewed the experimental shopping programs. Table 4
16 presents the results of the believability of scarcity of quantity and scarcity of time

 

 

messages.
Table 416. Believability of Two Types of Scarcity Messages
(n = 120)
Very Somewhat Not Don’t
Product Believable Believable Believable Believable Know
Scarcity of Time 2% 11% 28% 53% 6%
Scarcity of Quantity 2 13 38 40 8

 

The findings show that 41% of the subjects believed the messages of scarcity of
time, and 52% believed the messages of scarcity of quantity. It was suggested in previous
studies of scarcity messages (Y un, 1992) that scarcity messages have no impact on
people’s perception or behavior if they do not trust these messages. Thus, it is reasonable
to examine the impact of the believability of scarcity messages on the buying impulse of the

subjects in this study.

80

An index of believability was created by combining the believability of time and
believability of quantity. The score of the index ranged from 2 (neither time nor quantity
scarcity messages were believable) to 8 (both types of scarcity messages were very
believable). The index was categorized further at two levels: scarcity believers (those with
an index score of 8) and scarcity non-believers (those with an index score from 2 to 7). T-
tests were conducted to compare the buying impulse means between these two groups of

subjects. The results are in Table 417.

Table 417. Buying Impulse of Scarcity Believers and Non-Believers
(0 = Not intend to buy; 1 = Intend to buy.)

 

 

Believers Non-Believers t
Product (n=38) (n=71)
1. Rolodex Electronic Harmer . 18 .41 2.41*
2. Nutritional Monitor .23 .21 . 18
3. Skewdriver Pro Kit .24 . 15 .99
4. Electronic Pedometer .33 .1 l 2.90"

 

Note. For this believer index, 11 subjects of 120 were missing.
*p < .05; **p < .01.

There was a significant difference in buying impulse between the subjects who
believed scarcity messages and those who did not for the Electronic Pedometer. This
difference is also significant for the Rolodex Electronic Harmer but the direction of the
impact was opposite. Three-way analyses of variance were conducted to examine the
effects of scarcity of time, scarcity of quantity, and believability of scarcity messages. No
significant interactions among the three variables were found for any product A significant
main effect of believability of scarcity messages existed on buying impulse for the same
two products, the Rolodex Electronic Harmer and Electronic Pedometer. Again, the impact
of believability on impulse buying for the Rolodex Electronic Harmer is opposite to that for
the Electronic Pedometer.

81

Endnotes

1 The z-score transformation standardizes variables to the same scale, producing new
variables with a mean of 0 and a standard deviation of l (SPSS, Inc. 1990). Because the
lengths of four product segments were different in this study, it was necessary to
standardize the response time and strength variables before they were aggregated.
Therefore, z-scores for all four experimental products were calculated and an average of

these scores was used in the analysis.

CHAPTER 5: SUMMARY AND DISCUSSION

This chapter summarizes the results presented in the preceding chapter. First, it
reviews the findings of a descriptive analysis of impulse buying in TV home shopping
settings. Next, this chapter elaborates the results of hypothesis testing. Then, it discusses
the findings of an exploratory analysis of the factors that were not specified in the
hypotheses but had an impact on impulse buying. Finally, this chapter discusses

limitations of the study and suggestions for further research.

Impulse Bum’ng in TV Home Showing
Previous studies have defined impulse buying as an unplanned purchase for which

the decision to buy is made in the store. To be consistent with previous research, this
study defines impulse buying in TV home shopping as a purchase that is made by a TV
home shopping program viewer who has had no intention to buy the product prior to
tuning in to the shopping channel. Different from previous studies, however, this study
identifies two new dimensions of impulse buying in TV home shopping settings: the
response time and the strength of the buying impulse. Response time refers to how quickly
an intention to buy an unplanned product occurs in the viewing process. The strength of
the buying impulse refers to how strong the intention to buy is. These three dimensions
have been investigated in this study and the findings are summarized in the following

sections.

A moderate level of impulse buying was found to exist in this experiment, from
19% for the Nutritional Monitor to 32% for Rolodex Electronic Planner. These findings
suggest the levels of the imwlse bum'ng in TV home showing vm among products.
These percentages, however, may not be generalizable to normal TV home shoppers

because of differences between an experiment and the home viewing settings.

82

The measures of the response time indicated that the viewers of TV home shopping
programs tend to make their purchase decisions considerably before a product segment
ends. They do not need to view an entire product segment to decide whether to buy or not
to buy. This is probably because the products in this experiment were low involvement
products and the information processing in the decisions for these products was relatively
simple. Another explanation may be that subjects concentrated on viewing of the products
in the experiment so they could process the product information in a more efficient way

than viewers in normal home settings.

There is a substantial difference between decisions to buy and decisions not to buy
in terms of how quickly these decisions are made in TV home shopping. It takes less time
for the viewers to decide not to buy a product than to decide to buy a product Probably
because certain risks are involved in purchase decisions, decisions to buy require more time
to evaluate the involved risks. This finding may be generalizable to purchase decisions in
normal home settings. The measures of the response time also reveal that the viewers tend
to make quicker and quicker decisions as they continue viewing the shopping program.
This fact suggests there is a learning process in the experiment The implication of this
finding is that more experienced TV home shoppers may be able to make their decisions
(either to buy or not to buy) more quickly than TV home shopping novices.

Most buying impulses are accompanied by a strong or very strong intention to do
so. This finding supports Rook’s (1987) definition of buying impulse that impulse buying
occurs when a consumer experiences a sudden, often powerful and persistent urge to buy

something immediately. The strength of buying impulse is indeed an important factor.

Scarcim Messagm and Impulse Bum’ng
This study proposed five research hypotheses on the influence of scarcity messages

on impulse buying, of which one was not confirmed and four were partially confirmed.
Partially confirmed means that the hypotheses were supported with one or two of the four

experimental products. This section summarizes the testing results of these hypotheses.

The data did not support the hypothesis that a buying impulse is more likely to
occur in a scarcity condition than in a non-scarcity condition. That is, there was no
significant difference in the buying impulse of TV home shopping viewers between the
scarcity and non-scarcity condition. The study found, however, a consistent pattern across
all four products: response time was shorter in the scarcity condition than in the non-
scarcity condition although the difference was significant for only one product and an
overall impact was relatively weak. This study also showed that the buying impulse was
stronger in the scarcity condition than in the non-scarcity condition for three of the four

products although the difference was significant for only one product.

It can be concluded from these findings that scarcigt messages do not influence the

finnence of a bum'ng imMse itself; however, they do imgt the resm time and

strength of the buying impulse after it has been triggered by other factors (e. g., product
evaluation, which is discussed in the following section of this chapter).

It is obvious that the influence of scarcity messages on the response time and
strength of the buying impulse was relatively weak in this experiment This may be
attributable to the methods of this study. The subjects in the experiment knew that the
shopping program they were watching was not broadcast live and that they could not buy
any item from it even though they might want to. The viewers’ realization of this fact

might have had a subtle impact on their response to scarcity messages.

85

This study distinguished two types of scarcity messages in TV home shopping
programs - scarcity of time and scarcity of quantity. It examined the relative influences of
these two types of scarcity messages. The findings indicated that high scarcity of time had
a significant impact on the buying impulse whereas high scarcity of quantity had no
significant impact Given the fact that no significant difference in the occurrences of
buying impulse was found between the scarcity and non-scarcity condition, it is prudent to
take these findings with caution. This study suggests that further studies be conducted

before conclusions are made on the impact of the high scarcity of time on impulse buying.

Materialism and lmflse Bum'ng
Three hypotheses were proposed to address the influence of materialism on the

buying impulse of TV home shoppers, and one hypothesis to specify an interaction of
scarcity messages and materialism on the buying impulse. Among these four hypotheses
only one was partially confirmed.

The study partially confirmed the hypotheses that a buying impulse is most likely to
occur among highly materialistic viewers, somewhat likely to occur in moderately
materialistic viewers, and least likely to occur in least materialistic viewers. That is, a
significant difference in buying impulse exists between high materialism and low or
moderate materialism but not between moderate and low materialism. Given the fact that a
consistent pattern of materialism impact existed for three of the four products, I_t_t_s_
con_cluded that materialism does imgt (m the bum'ng impulse of TV home shwwrs. As
for two other dimensions of the buying impulse, the study failed to find significant
differences in the response time and the strength of the buying impulse among the viewers
of different materialistic levels. This study also did not support the hypothesis that scarcity
messages and materialism interact to impact on the buying impulse.

The impact of materialism on buying impulse was not as strong as anticipated.
There are two possible explanations. First, because the subjects of this experiment were
college students, they might be homogeneous in materialistic values. If that is the case,
then it cannot be expected that levels of materialism will have a substantial impact on the
buying impulse. A review of Table 3-10 and the related discussion in Chapter 3 shows that
the scores of materialism for the sample ranged from 18 to 56, with a mean of 35.85 and
standard deviation of 7.22. Theses statistics suggest that the subjects are instead quite

heterogeneous in materialism.

Another explanation is that materialism has a relatively weak impact on the
acquisition of less expensive items. As discussed in Chapter 2, previous studies have
found that highly materialistic people value acquisition and the means to acquire
possessions more than those low in materialism. They also value possessions more than
other life goals. However, no studies exist to answer the question whether materialism has
an impact on the acquisition of everything or an impact on acquisition of only expensive or
valuable items. Because materialism had a weak impact on the buying impulse in this study
and all experimental products were priced less than $50, it can be speculated that

materialism may have less impact on acquisition of less expensive items.

Product Evaluation and Believabilig of Scarcigt Mflges
Although the impact of product evaluation was not hypothesized in this study, the

subjects’ evaluations of the four products were measured in the experiment An
exploratory analysis indicated that product evaluation has a significant impact on the buying
impulse. The viewers who evaluated a product positively were more likely to buy that
product on impulse. However, no significant impact of product evaluation was found on

the other two dimensions of buying impulse for any product That is, the response time

87

and strength of the buying impulse were not subject to the influence of the product

evaluation. It can be concluded from these findings that a buja'ng impulse for a product can

 

The believability of searcity messages had an impact on the buying impulse for only
one product. The difference was also significant for another product but the direction of
the impact was opposite. Given the size and pattern of the impact of believability of
searcity messages, this study concludes that the believability of searcity messages has no
substantial impact on the buying impulse.

Image; Imflieations

This study has several theoretieal implications. It reveals that scarcity messages
tend to generate a sense of urgency which is indieated by a shorter response time in the
searcity condition. This effect has not been investigated in previous studies, probably
beeause a sense of urgency in retail stores may not be so conspicuous as in TV home
shopping settings and it also eannot be precisely measured with a paper-and-pencil
instrument. In interactive media, a sense of urgency may be an important mediating

variable in the consumer’s decision making.

Contrary to previous studies (Brock, 1968; Fromkin, 1968; Knishinsky, 1982;
Lynn, 1991), scarcity messages in this study did not make people more desirous of the
scarce items. In light of the theories of searcity messages reviewed in Chapter 2, this
unexpected result may be attributable to the W in the study. According to the
need-for-uniqueness theory (Fromkin 1968; Snyder & Fromkin 1980), scarce goods are
more attractive to people who want themselves to be special. In this study, however, the
experimental products (Rolodex Planner, Nutritional Monitor, Skewdriver Pro Kit, and

88

Electronic Pedometer) might not be the right type of products to satisfy the people’s need
for uniqueness. Therefore, although they were made to seem searce in the experiment,

these products were not more attractive to the subjects.

From the commodity theory (Bozzolo & Brock 1992; Brock 1968) perspective,
when a commodity is perceived as unavailable, it will be valwd to a greater extent and the
desire to obtain it will be enhanced. Although this theory does not specify the types of
products to which it applies, it seems tint the effect of unavailability or scarcity is more
likely to occur on daily necessities. For instance, the 1974 sugar crisis ('Sugar," 1974)
and the 1979 oil crisis ('Energy: Togetherness," 1979) did generate a nationwide rush for
the scarce products. However, the products in the present study were obviously different.
They were not people’s daily staples. That might be the reason scarcity messages did not

generate the subjects’ spontaneous intention to buy the products.

In sum, this study raises the question whether the need-for-uniqueness theory and
commodity theory are appropriate for all types of products when they are used to explain
the effect of scarcity messages. The fact that searcity messages failed to make a difference
in buying impulse suggested that the impact of scarcity messages may be mguct-mific
and the existing theories of searcity may need to be tested among different types of

products such as necessities and luxury goods.

This study also sheds light on the cause of a buying impulse of TV home shoppers.
It revealed that the positive evaluation of a product was highly correlated with a buying
impulse. It ean be assumed that the positive evaluation of a product is a determinant of an
impulse buy, instead of vice versa. It can be further assumed that both planned purchases
and impulse buying are the same in terms of the existence of a product evaluation in the

decision making. As discussed in Chapter 2, a key difference between the two types of

89

consumer behaviors is whether nfl recogm'tion exists before the viewer tunes in to a TV
home shopping channel. These assumptions could be tested by asking TV home shopping
program viewers who have made a purchase to answer two questions: 1) whether they
decided to buy the brand or a brand in that product category before tuning in to the
shopping channel, and 2) whether they recognize a need for the brand or a brand in tlut
product category before seeing the product. It is very likely that impulse buyers do not
recognize a need for the product until they see it in the program.

Qmitatign_s

This study had difficulty in quantifying high and low scarcity of quantity messages.
Although a l to 4 ratio was found in the pretest and used as the basis for deciding the
amounts of items in the low and high searcity conditions, the author was uncertain whether
these numbers (500 for high searcity condition and 2,000 for low scarcity condition) would
generate the anticipated sense of searcity. This uncertainty was largely because both the
respondents of the pretest and the subjects of the experiment lacked knowledge about the
TV home shopping business in terms of its penetration in the U.S. households and the
volume of its yearly sales. Thus, the impact of searcity messages in this study may not be
generalizable to other products.

Another limitation of the study is that the subjects were not evenly distributed in
each session for each condition. For each session, the subjects were always assigned to
the condition that had fewest subjects by then. Thus, there were eight subjects in two
sessions whereas there was only one subject in several sessions. It is unclear whether
subjects would have behaved differently among experimental sessions with different
number of subjects. It would be ideal if there were an equal number of subjects in each

session and an equal number of sessions for each of the five conditions.

90

Two instruments (one was computerized and the other was paper-and-pencil
questionnaire; see Appendices 4 and 5) were used to collect the data for this study. The
paper-and-pencil questionnaire adopted two seales from previous studies — a scale of
shopping orientations and a seale of impulsiveness. This study did not pretest these two
scales. Data analysis of these scales failed to identify the proposed factor structures. Thus,
the data from these seales were not used in this study. The questionnaire also asked the
subjects whether they noticed the scarcity messages after they had viewed the shopping
program. Because the questions could not distinguish the subjects who did not notice
searcity messages from the subjects who could not recall their notice of scarcity messages,
data from these questions could not be used reliably. These problems might have been
avoided if a pretest of the questions had been conducted.

Future Research

Consumer behavior in interactive media such as TV home shopping is a promising
research direction. New theories of consumer behavior will be in demand as the
information superhighway reaches more and more households. We need a better

understanding of consumer behavior in a completely new environment.

This study has left several important issues in TV home shopping unanswered. A
preliminary analysis of the TV home shopping programs indieates that a variety of
W have been used to affect program viewers and to promote the sales.
Typical tactics are uses of different prices, celebrities, reference groups and models in the
programs. Measures are taken to reduce the viewers’ perceived risks in their purchases
(unconditional 30-day return policy) and to facilitate the viewers’ orders (installment pay
plans). It is unclear what impacts these tactics may have on impulse buying in TV home

shopping. Further studies can investigate the impacts of these tactics on impulse buying.

91

For instance, celebrities are frequent presenters in the QVC shopping programs
which suggests that using celebrities is an efficient way to generate viewership and promote
sales. However, no studies have been found to address why celebrity presenters make TV
home shoppers more likely to watch shopping programs and to buy products. From the
uses and gratifications perspective, a parasocial relationship may exist between the celebrity
and viewers which mediates the viewers’ responses to the celebrity-hosted programs. As
Horton and Wohl (1956) suggested, a bond of intimacy is developed with media
personalities through shared experiences existing only through viewing of the personality
or persona over time. Levy (1979) observed that viewers who found parasocial interaction
gratifying increase their exposure to television news for more contact with the news
personalities. Robin and McHugh (1987) noted that parasocial interaction, as a form of
intimacy, might be seen as an antecedent to future media use. In light of this theory, future
studies can explore what gratifieations the viewer seeks through viewing the celebrity-

hosted shopping programs and buying from these programs.

Use of models to demonstrate products is another characteristic of TV home
shopping programs. Social learning theory (Bandura 1977) can be used to explain the
effect of this promotion strategy. This theory identifies four processes that intervene in
modeling: attentional, retentional, production and motivational processes. This theory may
be able to explain why TV home shopping program viewers recognize new needs and
imagine new uses by watching model—demonstrated products, especially new products.

This study included a set of viewer-related factors such as materialism, product
evaluation, and believability of scarcity messages. This study also measured shopping
orientations and impulsiveness but they were not used in the analysis. More reliable
measures of these constructs need to be developed in future studies to delineate the impact

of cognitive factors on impulse purchase.

92

In addition to programming factors and consumer-related cognitive factors,
product-related factors such as prices and types of products also can be research topics.
Beeause the present study used all low-priced items ($50 or less), issues such as low or
high involvement and perceived risks were not addressed. Psychological pricing is
frequently used in TV home shopping programs whose impact on buying impulse can be
studied across different types of products (necessities vs. luxuries or low involvement

goods vs. high involvement goods, etc.).

Studies can be conducted to compare the relative impact of these three types of
factors on impulse buying and planned purchases in TV home shopping settings. Different
purchase patterns may exist in TV home shopping: some viewers may flip though TV
channels and be attracted by some merchandise in a shopping channel which results in a
impulse buy; whereas other viewers may tune in to a shopping channel with a clear
intention to buy some products but actually buy something else at the end. Constructs such
as product evaluation, need recognition, and scarcity messages may play different roles in
different purchase patterns. Theories need to be developed to explain potential differences
which may be useful to understand consumer behavior in other interactive media where a

mix of both planned and impulse behavior is more common.

Ideal studies in interactive media can be conducted using "real” data — the
information about transactions or usage that are recorded in the systems. An advantage of
interactive media is that the systems store automatic feedback information. For instance,
the QVC Network has records of purchases for each customer by time in its database. The
Prodigy or Ameriea Online can record the usage of each program or service in their
systems. If such data are available, they can be analyzed with a content analysis of the
programs or services that parallel the data in time. Such an approach will no doubt produce

more accurate and reliable findings.

Conclusions
The most important finding of this study is that different factors have different

impacts on the dimensions of a buying impulse of TV home shoppers. Specifically, a
buying impulse can be triggered by the viewer’s positive product evaluation and
materialistic values, and the response time and strength of the buying impulse can be
enforced by scarcity messages. Scarcity messages themselves do not usually trigger a
buying impulse.

In a broader sense, a better understanding of impulse buying in TV home shopping
settings requires a combination of both cognitive and behavioral approaches. A cognitive
approach assumes that humans are autonomous, independent centers of action for whom
predominant controlling variables are cognitive factors. A behavioral approach assumes
that all behavior is controlled by environmental factors and cognitive factors are merely
mediators (Peter & Olson, 1990). What this study demonstrated is that neither approach
can individually solve the puzzle of impulse buying in interactive media. We need a skillful

combination of both.

APPENDICES

APPENDIX 1
The Script of a Focus Group Discussion on TV Home Shopping ...................... 94

APPENDIX 2
QVC Programs by Frequency of Run ................................................................. 111

APPENDIX 3
Viewership of TV Home Shopping Programs .................................................... 116

APPENDIX 4
The Computerized Questionnaire ....................................................................... 117

APPENDIX 5
The Paper-and-Pencil Questionnaire .................................................................. 122

APPENDIX 1: The Script of a Focus Group Discussion on TV Home Shopping

Hairong Li's Note: To get some common-sense ideas on TV home shopping, a focus
group was held in Dr. Bradley S. Greenberg's ofiice on November I I, 1993. Three
advertising majors and a graduate student of psychology participated in the discussion.
Among them were three females and one male. Regan Kania moderated the 45-minute
discussion. An interviewer of the Institute for Public Policy and Social Research at
Michigan State University transcribed the entire session. Hairong Li reviewed the script
and underlined the sentences he considered more relevant.

(M = Moderator, P = Participants, H = Hairong Li)

M: The first question on our list is what personal viewing or shopping experiences have
you had with the home shopping network or with QVC?

Pl: Ummm.
M: Go ahead, you can say it.
P1: Well, I've never bought anything from the home shopping, but I watch, like, well if I

am flipping through the channels and there is nothing on, well, I'll, you know, watch it
just for a moment or two, especially if they have, like a celebrity on, like sometimes. L111

watch it when they have Jgp Rivers op, or has everyone heard of Joan Rivers?
M: Urn hmm.

P1: Other than that I kind of like making fun of it.

M: Do you?

P1: Yeah

M: Yea, so you probably would not buy things from it, or you might.
Pl: No, l 'nkI wou d 't tit.

P2: I've bouth things from it.

M: Have you?

P2: Yep.

P3: Yea, me too, for example, this necklace and this ring.
M: Uh huh.

P4: And I have a ring.

M: How often do you watch?

95

P2: At least, I probably sit down, like, not steady, but about three times a week. I watch it
for at [mt tep minutes or so.

P3: Especially if it is really late at night.
P2: At night.
P3: And there is nothing else on.

P2: And you'll sit there picking out Christmas gifts and my mom watches it too. So at
home she'll sit there. That's her life. "This is great; this is what wanted.”

M: What period of time do you watch for normally?

P2: Like?

M: Like how long would you say you watch?

P?: Probably like...

M: On an average basis.

P2: At the most probably ten minutes.

M: OK, so not for very long?

P2: While I'm waitin or the next ro to come on or...
M: OK.

P2: Something decent to come on or...

M: So you flip to it during commercials and go back to it during other programs?

P2: Well, I've said I've watched it too, like if I'm doing dishes or something like that. I

just, if there is nothing I want to see on TV. I'll turn it on and see if they've got anyflng

good. I mean I wouldn't for the clothes and stuff, but some of the jewelg and stuff, being
that I got something there and my mom has tons of stuff from QVC. We never have any

trouble, you know, with it turning our hands green or anything. Seems to do fine for me.
M: Umm, hmm, how about you? how often do you watch and how long?

P3: Well, probably like, Oh, I would say once a week or once every two weeks. If I watch

it, I usually watch it for the whole time. Like if they have jewelry on I usually sit and

watch the whole section as long as it's not over a half hour. If it gets over a half hour then
I et bo .

M: OK.

P3: It takes them awhile to get to the next item. If you sit there and watch the same item
over and over and you want to see more. So, I think they sit on one item too long, but I'll

sit up to a half hog to watch it.

M: How about you, Shawnti?
P4: I'm really not that familiar with Home Shopping Network.
M: OK.

P4: I don't have cable so I don't, and I don't watch TV very much. I'm just familiar with it
dmo er cause she watched it all e time (M: Uh huh) and not myself.

M: OK, we have a sample tape to show just for a couple of minutes, so Hairong, why
don't you start that.

H: OK.
M: We're just going to show a couple of minutes of the home shopping channel.

H: I also have a QVC annual report, and a Home Shopping Network annual report if you
want to...

(While viewing the tape)

P4: Oh, same thing as? These aren't real obviously.
H: They're not real, I just want, they're not real.
P2: Almost all their products aren't real.

P3: They need to clarify that, too.

P4: Um Hmm.

P3: On there they should. I think it's just like a lot of people think they're getting real stuff
and...

(Tape viewing over)

M: We just saw one ad for a ring and one ad for earrings. Diamond looking earrings. OK.
Well, what was your, what do you think about that?

P4: 1 just think it would be so boring to watch. It's not for me. It's not enteM’ning. It's not
something. I would have more fun going out and shopping. I think, I like to shop and I

don't don't ' there would be an 'n am interested in. I understand alot of

We like it.

M: What kind of people do you think would like it?

P4: Mm, the m of mph that don't have the time to shop or don't like to shop.
P2: or not as mobile, elderly or...

M: Do you think?

P2: Or even pppple that dislike crowds, like at Christmas time I do not want to deal with
going into a mall, waiting in line for a half an hour, or, you know, those m of mle

9‘7

esmially one that can't, or dop't drive or you know stuff like that. And they don't have
easy access to up and go shopping.

M: Do you agree or disagree with that?
P3: Yea, I agree with it.
P4: I thought so too.

P1: Yea, I still think. I still think of my grandmother because she is the type of person
that if you go into a store and tell her you know you really need this. This is a product
that you really need. You know, she is kind of has disgable income. She is maybe a
little naive so she'll believe that. I think that maybe older pe_gple that do have the extra

income who are kind of naive and buy into that mrsuasion.

M: All right, good. How do you think people buy? Do you think they plan the purchases
that they make on the Home Shopping Channel or do you think they buy on impulse? For
example, how do you define impulse, Hairong?

H: Yea, basically, consumer behavior studies categorize behavior into three types. One is
planned action. There are a lot of models and theories. You want to buy a car or you want
to buy a computer so you have criteria you want, and you look over a lot of brands then
you check the price and features. Then, eventually, you make your purchase decision so
that is called planned. You know a specific category and you buy it. Another one is
impulse. For example, before you step in the retail store you do not have an idea, you
look inside, and you see some cookies and say, ”Oh, that's nice, that's a new brand, why
don't] taste this." And then you grab them and put them in the shopping cart. Sol would
call that impulse. You do not have any plan to buy either the product category or the
brand. You don't have any idea about that particular brand. Just when you see it, you like
it and you buy it.

M: So what type of purchases do you think people make here. Do you think they're
planned or impulse?

P4: Definitely impulse.
P1: Impulse.

P2: I think there is a lot more impulse because you never know when you turn that on. I
mean you never know, I mean they may tell you they have a segment coming up but you
might sit down and say, "OK, I need a ring. " So you migl_rt sit down and watch it. There
might be a little planning to it. For the most like on Hi the channel and are l' e

“Wow, look at that, I really want that," so you call and...

M: OK.

H: I gave you that program guide. I'm not sure if any of you, any of you, received some
before. I called them and they send me some. Sometime when your order something they
send you a program guide. So you can see different product...

M: Different product categories are offered at different times of the day.

. I know even if ot this I babl wouldn't even look at it. I would not even
look at it. I would like, well, I should probably stay home between, you know, 4 p.m.,

because at 4 pm. this is conring on.

H: I did a couple of analyses of program types. They have some programs that are run a
couple times a day. This one is for several days in a month period. So, this is jewelry
showcase, just one hour program. They run forty-four times during the thirty days. So

apbably someone, you know, who wants the prom guide eand if you like jewelgor
yop like clothes or dolls and then you Qp set yog; VCR and tag, but I don't think that rs
a fig momrtion of viewers.

P1: When you say that they run the same program over and over, are you saying like the
exact same tape? They just put it in there forty-four times or do they show the same
merchandise and like present it. Do you know what I am saying?

H: Yea, I know. The program name is the same but the content is different. Most of them
are broadcast live. I'm going to go there and take a look at their studios. But I have not
made the trip. Their headquarters... I read in an article something like this — they have a
producer for every program that sits behind the studio. You know when the phone, let me
say this, they have a very huge room for the telephone operators. See this one, how many
they have. Over a thousand calls coming in. So, how many orders are placed at the same
time the producer can watch the number go up so when he sees the curve go up and then
it reaches a purchase peek, and then when it goes down, the producer would say "next."
So, it is live. Most of them are live.

 

M: What do you think are some factors that contribute to the purchases that are the result
of impulse on the home Shopping Channel? What would make someone want to make an
impulse purchase?

P1: I think just the whole concep_t of it. it gleans itself really well to impulse eause like
they have to phone up, you know, and just push a couple of buttons. I think they really
push that. You know "three easy payments." They try to make it sound really well, you

know and things like that. Just the whole and just the fact that you can't really plan what's
go jpgto go on, as someone already said, it's just kind of what rs on, and rs what's there...

M: It forces you to buy an impulse image?
P1: Right.
M: What do you think contributes to the impulse purchase?

P3: Spine things that I gunk that would contribute to it, and that they don't do veg much

of , is they should put up the return mlicies more. So that mpple could, if they buy it and
they are not satisfied with it, then they can return it. Because a lot of times they don't

even talk about that really. But...

M: Is there any thing else that they do on the program now that would cause people to
order? Can you think of any other things they do now?

P2: The pricing. The way they at the prices up there. It is here is the original price, and
this is our price and then You'll notice a couple of times, sometimes, when they put the

Mud up, they will cut the price like four times. They will be like well we can sell it to

you for this. I have sat there and I have seen like four times over numbers and I will sit

there and I will be dying. It will be at like tweng five dollars when it started at like a

hundred and all of the sudden, you are like "Well, that's got to be like the best deal in
town!" So you know it's like any one who will just pick up the phone. Especially, if that's

99

not something you look for everyday. You know if you are out shopping every day
looking for, you know, diamond earrings or whatever then maybe you would know that
well they got these at Meijers or whatever for $14. 99 or whatever. These are going to be

products you don't look for everyday.

M: What other things do you see in the program that might cause people to buy on
impulse?

P4: Well, I was also going to say the price. My grandma would keep a list to keep track

of all the savings that she would have for each product and she would add it up and she

was so happy that she had saved about $2,000 buying these products. I mean she would

33y things that she would never use. She bought my dad a tool set and my uncle a power
'11 and...

M: Hm, so the dollar here was pretty important?
P4: Yea.
M: For the Money.

P2: People think "oh, wow, I am saving." Even gough, they may not be overall. You
think you did because you don't know.

P1: Also, I think like the type of products they have. I think most of their things aren't
really essentials. I think like something you need you see something and go "Oh, I like
that" or

P2: To better yourself. "Well, if I get that it's going to look really good on me."

Pl: Yes, I think that kind of and also the way they talk. The whole persuasion and the
way they talk is trying to, you know like I said the format is trying to. You know, M

8% out saying, "No, 1 don't want that," and then there are good things to Ey and get you
to buy it. And t_ry and get you to, get that impulse for you.

M: Um, hmm is there anything else that motivates people to buy on impulse? Can you
think of any other factors?

P4: You don't have anything else to do.
M: OK. So maybe if the audience is bored.

P2: When you're sitting there watching it and you see somebody else with it. And that
little counter says the number of pe_ople who are bufl'ng this right now. And they keep
track for you and let you know "237 calls have come in and there's only 40 left and po_w_
m‘d better get on that phone now and call us now." And you think, look at all those
people that are buying this product and there are only forty left and that means only about
280 of these products. Nobody else in the neighborhood is going to lflve one so I m
going to get this and I'll be different. You know not like earrings or something like that
but some of their other products, like their sweaters or whatever. That's different from

n an one in the nei borhood rs oin to have and that the numbers are u and

yo t_g; like "Oh, I just have to do it now," you know?

P3: Or when they have like satisfied customers call and say that they like their products
or whatever. I think some of the people are hesitant about buying over the phone or

100

whatever because they don't know what they are going to get.
M: Yea, we watched an example of a caller talking from her home.
P3: Yea.

H: Most of the people, they are happy when they talk, right? I found very few customers
that are sad, sad customers that called. If they call they may cut them out.

M: Yea, if they call they would cut them out.
H: I've never heard a complaint over the phone.

M: Is there anything else that you remember noticing that would trigger an impulse
buying decision?

P4: I think that good lighting techniques that can make the diamonds spafle.
P2: Really, sparkle.
M: Any other factors at all that you can remember?

H: For the lighting, they use a different lens, you know, when they show a particular item
you see the lens and the reflections and that type of thing. You look at the real thing and
you say oh now that is..

M: Are there any other persuasive factors that they used to try and get you to buy or ?

P4: They showfi the women with the earrings on and she was an attractive model. You
would be able to think I would look that good in those earrings.

M: OK. Any other factors?

Pl: Well maybe like in, I mean, I am just thinking like in the example, he is trying to
show you just how you can use it and getting it. Usually they g to give you examples of
different ways you can use the prguct or things like that.

M: Mm hmm.

P2: Plus making it seem like a very practieal normal gift that someone will use. Where he
said his mom gave him that little thing, that clip, that grasshopper thing, or whatever, @

kind of sau'ng "Well. why don't buy those things for somebody. This is much better,
more practical apd somebody will really like iL," and you know, they'll say, "come on,
you have got to buy this produc "

M: So, the announcer is pretty persuasive.

 

P2 and P3: Yea.

P2: But he also tried to be kind of , he also tried to be kind of funny sometimes, so he
didn't look as pushy like he was trying to be real persuasive (M: OK) but you know.

M: Anything else you can remember? (Pause) OK Do you remember a little clock down
in the corner and what did that represent, do you remember?

101

P2: Is that like the time that you have left or time that they had been calling the item?

M: Do you think that would influence anybody to purchase, to try and get the call in
before the clock expires?

P3 and P4: yea

M: OK All right.

P1: I always wondered though if you can call like, you know, two seconds afterwards.
M: Yea, I'm not sure. Do you know?

H: I don't know. Do you think that is believable?

P2: I don't know. I never called them afterwards.

H: Is that true?

P3: I heard that when they say they are going down and when they say they only have
seven products or whptever t_hey really have a lot more than that left. And that the reason

that they do th t rs to get you to call. And then after they have so many left over or
whatever, they will start it again like three weeks later. They'll start that product thing

again.
H: That is a kind of deception, right?

 

M: Possibly.

P4. See I would use the clock. I_______f it said two minutes and twenty seconds if I didn't like

the product thev were showing, then I could flip the channel and I could, you know, flipit
back rn two minutes.

M: So you would use it to decide when to come back?

 

 

P4: Yea.

P3: Well, if a mrson is sitting there and they are thinking about bufl'ng it and there is
only a minute left and they are like "Oh, I better hm up"

M: Do you think that is really another form of pressure to get them to buy it even if they
are not sure?

P2: Yea. Cause you're like, "well if I don't get it within the next minute then any chance
of my having it is going to be gone." At that ppint that is the only thing you are thinking
of. You are not thinking of, well, it might come back on or I might find it in the stores.

M: So, are there differences within people on who's more prone to buy something on a
short time limit like that than who isn't. Like, what type of person is going to buy at
impulse? I asked who you thought would watch the home shopping network now, and
who do you think would buy at impulse? What type of person?

P2: I think somebody, like more middle income. I mean I just can not picture my dad like

102

sitting down and buying something like this he would be like yea right. (P4: yea) I mean
he is the type that would be like at Lord and Taylor's, you know, and at the expensive
jewelers and where price is no matter. You know, they want the best and the higher price
they pay the better they feel. Where my mom, she's really low income and when she buys
something off there she feels satisfied, like she really got a bargain here, and it was better

for her. Sol would say like more low to like medium, like the lower Ell pf the medinm
men. People maybe with, I know it's not real nice to say, but pe_pple with a little less
intelligence, they arpn't out to look mom and chgpk Qn'ngs out.

M: OK.
P2: Not necessarily into college or maybe just two years of college or something like that.

M: OK. Who do you think, Shawnti, would buy at impulse more than others?

: le who e sad. Peo le w feel satisfied tocheer the u r some 'n .
I don't know how long it takes to get mg Mnct pnce you order it but some apple feel

better from just buying something.
M: OK, do you have any input?

P1: Like, I think somebody that is reluctant or hesitating and isn't quite sure and they
push them over the edge and say, "Well, I'll just go out and get it or something." This is
going to sound really sexist but I kind of think, somebody, a male or maybe somebody
with a more male personality style would be like "no, I'm not going to buy it." You know
they made their decision. Where maybe somebody more emotional or more of a feminine
style who is kind of more hesitated and is kind of more likely to buy on impulse and is
hesitating like "OK, I got to have it," you know instead of you know just saying no.

M: Do you agree or disagree with that?

P3: I disagree with that. I think that it would be more someone like "I want it now." They

don't hesitate. They see something. They want it and they've got to have it now. They
don't think about it. They're just like "Yep I am going to buy it." That's what I mean. If I
see something that I really have to have I'll get it.

M: Well, you mentioned low income people rrright be more likely to buy it. Are they
equal as likely to afford what they are buying or do you think they buy beyond their
means?

P2: I think they buy beyond, but it's something that, you know, when you walk into a
store it's just that there are sales. People and you look at these price tags. You see really, I
think that it more hits on there when ou see one rice or item and it sa 8 2.00 and
y_qu just look at it, you go, "oh, my goodness." Then you have to write out this check n'gnt
m, and you know the whole time your like. "No, I really can't do this." Plus, you
constantly have sales people standing there and everyone just kind of nagging at you and
you just don't know. So, you just turn around and you are like "I am out of here." Wl_r_jle if

you're watching television and you make a misgge nobody knows you made a mistake.
You can send it back, and they are never going to see your face or anmng like that.

It's not necessarily low income people that can never afford this are just calling and trying
to get rip them off but people that do maybe once a month or something just for
something a little more than they can afford, or to get something for themselves to make
them feel better. But they don't feel like dealing with the store. And stuff like that. I mean

103

I think he's right that somebody that's a little more, there are probably less men that do
but I think there are probably men that do.

Pl: Yea, I am not saying that there aren't.

P2: I was saying they are the same way there probably are more women. You hear a lot
more women voices calling in (H: Yea) and talking to them than you do male voices. But
males are just like "I don't want to talk on the air. Just leave me alone." And women like
their games with friends and they will talk on the phone for hours. That's their job. You
know they hold onto that phone. It's something more Quainted with women.

M: Shawnti, what do you think about people that call in and make the orders? What
characteristics do you think they have? I know we talked a little bit about it, but?

P4: I think it is really easy to not even realize that your mm’ng for the product. You're just
m f vowredit card, you don't have to think about that money gone, you know

out of you and where if you are writing a check. It is a lot easier. It's a lot more guilt
free.

 

M: And that's the result of buying through the tilt television or telephone as opposed to
buying in person and with cash.

P4: right.

P1: I just thought of another, like, persuasive. Just how, I was thinking how, even though

it is through television, I think it is veg pprsonal, like the arson is talking right at you.

flfnere are not a lot of distractions there. If you go shopping in a store there is a lot of
distractions. The sales person, while you will not have a conversation with them, you will

kind of tell them what you want. And they will help you pick it out or something, but you
know that is pretty personal and a lot of time people would like, you know a host or like.
People will watch when, you know, when this one sells and the hostess is on, you know.
Do you know what I am saying?

M: What are you saying?
P1: find of like; me 1539mm“. thi e ad 'd on the hone she 1' their

ities or somethin .l ' it is ve rso

M: Do you think they would trust more the personality on the TV screen or something in
a live shopping environment? What do you think?

P4: 1 think definitely in a live shopping environment.

M: You would trust that person more?

Pl: Yea,. I think of all the times when I was in the store, I didn't want to buy something
because I believed the person was conning me. I ended up being dissatisfied. But I would
be persuaded to buy with someone there talking to me than on the television.

M: Do you agree?

P2: I think of it because I mean working in a retail store. There are people that come in

and its almost like they think you are working on a commission. You know that might
even be the first question out of their mouth. Are you working on a commission? Almost

 

104

where on TV they look at this and their TV mph have to be honest. Where I could be
working on commission and I could lie flat out to you that this is the best product in the

world, and it will fall apart in your bag on the way home, for real. They'll think well she's
just out to make her buck. They don't know who's making the dollar there. You know a
lot of people I think are thrown by that. Because I've had it asked many a times, "Do you
work on commission?" and if I do they are walking out of the store.

M: Who do you think is more believable?

P1: See, I think the total opposite of her because I think, you said you think because they
£132 TV mph they will be more trustful. I think because they are TV mh they will be
much much more distrustful. And like I know, or] am sure these people are working on
commission just like the whole, you know, TV thing is, like the whole Hollywood thing
is to be fake and be like a fake facade thing.

 

P2: But for older people, I think older 0 e have this sort of well if it's on television
they can't be lying to me.

Pl: It gives it credibility.

P2: Because somebody must be put on this station, and if they are lying somebody can be
sued real bad. You know so there has got to be some type of credibility. Where if you are
standing in a store you also have a judgment to make by sitting there looking at it. You
know what I am saying? You are standing there and you are looking at it, and you can
hold the shirt. If it falls apart in your hand, obviously, you can hand it back. Whereas if
it's over the TV and there is somebody lying to you directly, there is going to be, you
know, the National Inquirer or whoever it is that is going to be running after you, and
advertising group they are not going run any ads and not have a consequence. If you are
sending something you know that is going to break in the mail on the way there.

M: What do you think?

P4: Well I really couldn't trust either one, but just seeing these people when I watched it

yesterday for five minutes. This women had on the ugliest jacket and the guy was telling
her it is so beautiful on you. You can see them tell mople that these look beautiful when
it really doesn't in my opinion. I think the difference is just in the stores you get to make
the decision you get to tg on. And you get to see what the quality of it and so that is how

I would base my decision not on what someone else would tell me. You know any sales
person would probably tell me it looks good. They are both distrustful.

 

M: Well how comfortable do you think people are buying over the phone as opposed to
live then? You said you can touch the item in the store and you can't try it on telephone
ordering from the TV ad. You can't so. ] don't know how comfortable do you think
people are buying over the phone a product they can't touch?

P1: I think, like, when a person came on or like when it first started, I thought now "I am
never going to trust this at all." like I am never going to send them the check because I'm
never even going to get the product. But I think] know I'll get the product and everything

and I'll trust the product and everything. The only thing I would not like about is how it is

gging to look on TV versus in mrson. I don't think they are going to lie to me. But if ]
want to return it for some reamson it is going to be so much more of a hassle to do it. So]

think it's not that I don't trust them. It's just you know you're never sure at the same time
and it's more of a hassle. I would rather go to the store and buy it.

 

105

M: What do you think?

P4: I think it depends on the person. People think of shopping as a social kind of thing.
you know, they go with their friends. So If you want to go out and do something like that,
it all depends if you are homebody or. My mrception of the moph that shop for clothes
are more ppople who stay home and watch TV all day and really have no kind of life,
really, I mean honestly that's how I feel gd they are really flsive nnd just kind of let
things hapmn.

M: OK, What do you think? Who is more trustworthy? A live person a TV or what kind
of people order over the phone as opposed to live?

 

P3: What type of people?

M: Yea, Like she said social people like to go out shopping. It's more of a social. Do you
agree with that?

P3: I don't know. That's hard. I'm not sure. I think that you see a lot of stuff on TV like
the thigh master and all that stuff. It just seems like the biggest rip off. All that stuff that
they have that they are trying to get you to buy. Some of it, I just sit there and laugh. So
you sit and think, and in your mind you just associate it with those kinds of shows. S__o
you might think that only stupid @ple would buy through the TV flse obviously there
are all these other rip off things on TV. I mean I have really think they are rip off things
so some people might associate that with them. Although the things that I have, or my
mom, has bought me] have liked and been happy with it.

 

M: Well how do you think QVC compares to other f ull-length advertisements for
products, the thirty minute advertisement or any other commercial product that's
advertisement? How do you think the shopping channel compares?

P2: I think people. I do not watcn when somebody says here is a thim minute
advertisement on a channel. I turn it off immediately. I have only watched one and that

was with Cher and that cosmetic thing, her hair care thing, because she has that great hair
dressing lady or whntever. Bnt even t_h;a_t I won't even watch all the way througl_r, but I

think that catches you because it is Cher.
M: Celebrities help this type of commercial?

P2: Right. I just turn those off as soon as they say commercial. I think that kind of turns
somebody away. Whereas the Home Shopping or the QVC, you know, what you are
getting into when you turn that channel. You know your looking products and you know
that is what you want to buy. When you have to sit and listen to somebody describe a
product for thirty minutes you just turn it off.

M: What do you think? How does QVC compare to other commercials?

Pl: I think QVC has n little bit more legit. People view it a little more legitimately. But]

like the infomercials. I think they are funnier or something. Like just when you said it, I
was thinking about stuff the insanity women or whatever. Like I will watch it just

bggnuse I think she is funny. Some of the informercials are more like gadgets. Like I am
thinking of like they have a pasta machine. Like] think that's kind of neat. Just, you
know? Just more like, you know, gadgets.

M: So what it has more entertainment value more commercial for you?

106

P1: Yea, maybe for me.
M: So that would encourage you to watch the full length commercial over the QVC?

P1: Yea, but ] think that I would. That] would watch it for the same way I would watch
QVC. You know like I would watch to fill up time while I am waiting for something else
you know. You know ten rrrinutes or something like that.

M: So just for short periods of time?
Pl: Yea,

M: What about you? How do you compare the two?

P4: I think QVC is probably faster M. It shows more products in a shorter amount of
time. More mmonabh. I don't think I've ever watched an informercial just for me. Maybe

about five minutes. But I don't think you can call them when they get on the air. Or
maybe some people like to hear their voice on the air or something. I don't know.

P2: That is just like the participation thing I mean I am. Cause a lot of those half an hour
informercial or whatever, you call this 1-900 number, call this 1-800 number, just like
you do with them. But they make it seem like they are to talk to you. Where these people
are like just call and order because it makes your hair look pretty and we've told you
about our product for a half an hour.

M: So this is more a two way communication?

P2: They make it seem like there is more you personal stuff involved here because you're
buying a product of your dreams. Whereas it's not as though you are watching a Kmart's
commercial or just a commercial for any product you know, Ajax or whatever, you know
this is something more; this is you purchase.

P: And the callers are there waiting to hear...

M: Right. How do you think the two compare?
P3: 1 think that the Home Shopping Network is n lot more legitimate than the

informercials. But I think that having the informercials on, I mean usually the times I
would be watching the Home Shopping Club. I think it kind of drags it on a little bit.

People tend to associate one with the other and that drags the legitimacy of the Home
Shopping Network down.

M: Do you have a comment?

Pl: Yea, I was just going to say that I think meir audiences are probably a lot different
because just the time of day, like you can watch the Home Shopping Network, you know

in the afternoon and stuff like that. You know the older people or just people with
families you know, whereas people with families and the informercials are usually made
an'gm, and maybe for older or if you have a family, you know, it's more for people that
are up and want to watch TV and there is nothing else on, you know so I kind of think

they have a different audience.
M: All right. Do you have any other questions, Hairong?

107

H: Yea, I have another question. I was thinking of an experiment. What kind of product
do you think would appeal to students?

P4: What segment?
M: What type of product would students like to watch?

P4: Jewelg, Mt's what I always watch.

P2: Even like collegiate products, you know, stuff like you call and can order n swgapghirt
and they can say we can out your school on it you know or whatever, they can say we
have these different schools or you know.

M: Do they have that at all?
P: Do they do it?
M: Do they have that type of product?

P2: I've never seen it on their channels but I mean I've never. Maybe it is something they
can aim at.

M: But I think, we're not asking what QVC should show. We are asking what they should
show that they already show. Am I right? Is that your question?

H: Yea.
M: He wants to know what they can show so that it can be an interesting product.
P2: Oh, I see. I don't know. Maybe they do have sweatshirts. I don't know if I've.

P3: They have this one like when] was watching it and they had like I've seen their
clothes on before. And] thought they were really ugly but this one time they had like
evening outfits on, evening wear, evening dresses. A lot of them was really pretty stuff

gd it was really inexgnsive and even like the next day I had seen some of the outfits
like they had seen on TV in Victoria's Secret. I mean I would be interested in clothes that

were more contemm rather than they seem to have a lot of clothes grandmas wear
and stuff like that.

Pl: I'm thinking of, like I'm, I'm not sure but I think one time, they were selling like a
m Broom album. You know some music or something like that I gness.

H: CD?

Pl: Yea, something like that. And the reason I am thinking music is because it would
,4pr to both men and women. Plus it is something mpph in our age group buy a log,
and it would have to be something kind of middle of the road, like you know, nothing that

would offend apple but nothing that mle would think, this really is really boring gm
w, you know?

M: How long of a period of time do you think you would want to view if you were the
experiment? Before you got to bored?

108

P4: The time they do know is fine.

M: That's probably about five minutes.

H: Fivg minutes.

P2: my»! about Irve minutes because...

M: That's the normal time you watch anyway...

P2: 1th would Q something that students do. That would probably be as long as they
womg want to watch. I mean it is something that we do between.

H: OK.

P1: Yea, and I think mple are mm famjlinr; wimi rt,w wiom you uknow you don't nnvgto
explain to them wflt rt 1 y,ou know?

M: OK.

H: You know another question is, that bothers me, that I have to have a devise or a
mechanism to measure the sudden urge to buy. I'm not sure the way how we could
measure. For example, if I want to do an experiment in the classroom or in an office,
that's different from home environments. When you watch TV at home and you want to
buy, you buy. How do we measure this, for example, if we do this rn the classroom?

P2: How many people would buy it?

H: For example, if you really want to buy this, how do you express your intention? What
can...

M: You know, you are saying like a thermometer measures temperature.
H: Yea.

M: How can he measure how likely you are to buy something.

P1: Blood Pressure (laughter).

M: Pupil Dilation.

H: That part I don't know. If you have any suggestions...

P3: Even if you use those scantron sheets you know and you don't have to say your name

or anything like that. Because I know 1 f] was in a group with, you know, fifg other

mph rn a room. I sure wouldn't be the first one shooting up my hand saying "ym, I'd
buy that product." A lot of people, the reason they watch that rs because they are trying

to, not necessarily hide it from somebody else. They might like it, but who knows what
everyone else thinks. You know one would ever know and go through like five products
or something like that. As far as likelihood and three being right in the middle, I might
buy it depending... You know or say either no I wouldn't buy it, or yes I would.

H: So, a measurement. That's a good point I'm thinking about. I would like to ask
everyone about products and ask them "would you buy this?"

109

P1: I think people. I was just thinking that I think mph would be pretty honest if you
just ask them "would you buy this?" I think most people would just say "yes," but I think

just like an experiment, I think it would be pretty hard. Cause you couldn't just have a
thousand subjects. And maybe, one would just say they would buy it cause that is kind of
the whole concept. They work on that. They reach a million people and maybe only a
thousand would buy it. But selling a thousand products...You know what I am saying?
M: Yep, I know what you are saying.

Pl: You've thought about it?

M: A little bit (laughter).

P3: You have to differentiate between like] would when saw those earrings. Now I
would buy those cause they are simple and they are something I like. But] wouldn't buy
it like right now. I probably, I don't know, I would probably.

M: Hope it came out again sometime?

P3: Yea, maybe.

P2: Either that or that's a product you know you can go into the store and get.

P4: Yea that's true.

M: And get a better deal?

P4: It is also easier to say that you would buy that rather than just picking up the phone
and really getting that product.

M: Right. So a questionnaire. We don't know would a questionnaire be an accurate
response?

P4: Sure I would buy those earrings but I mean] wouldn't.

P2: if there was the phone there? would I call, etc.?

P4: I don't know if I like them enough to call.

P2: I mean you would have to be specific and say, "All right, you have a phone next to
you right now, and a credit card." And are you on it, on the phone talking to these people
when they are giving them the number or are you just saying "oh yeah, I'd buy it, that's a
nice product." It's tough.

M: OK. Look through this then and tell them what you want to know.

H: I would like each of you to evaluate these programming tactics. They are promotion
technologies. Promotion would signal impulse or whatever, if we drink that's likely to
generate or trigger impulse purchases, you write a five next to the item. If you think it is
not very important or less likely to generate an impulse, you can just assign a zero there
or a one.

M: Do you guys have pens in here or any?

110

P4: I have one.

P2: Yea.

M: Would you like one?

H: I should mention that a lot of items is like this...
P2: So if we feel that it will?

H: Yea, if you feels you will.

M: Five is very likely and one is not.

P2: To cause impulse buying?

H: Yea, in terms of...

P1: How come there's only a hundred, oh...

P2: So we just use zero and five and that's it or should we use everything in between.
M: Use everything in between.

H: Five is the most likely and...

M: Zero is the least likely...

H: Zero is the least likely to generate an impulse purchase in your opinion.

APPENDIX 2: QVC Programs by Frequency of Run
(July 26-August 22, 1993)

Program Name otal Hours

Showcase 44

 

(Continued)

111

112

(Continued)

 

(continued)

113

(Continued)

VCI'S

et Annr
vers
rn a etrme
ng rt wr

 

(Continued)

114

(Continued)

 

(Continued)

115

(Continued)

tan'y yes

rctonan trments
te
rant

VCS

TOTAL HOURS (24 hours X 7 days X 4 weeks) 300

 

Source: QVC Program Guide, QVC Network, July 26-August 22, 1993.

APPENDIX 3: Viewership of TV Home Shopping Programs
(96)

 

Who Bought
Who Watch in Last 3 Months

 

1991 1990 1989 1991 1990 1989

 

Total U.S. Adults 15.6 13.5 13.5 3.7 3.3 3.9
Men 12.8 11.8 11.7 2.8 2.8 3.6
Women 18.3 15. 1 15.2 4.5 3.7 4.3
18-24 18.3 14.8 15.6 4. 1 3. 1 4.6
25-34 16.8 15.2 15.1 3.3 3.6 4.2
35-44 15.8 13.1 14.1 3.3 3.6 3.5
45-54 15.9 14.4 15.0 4.6 2.9 4.6
55-64 15.1 14.0 11.2 5.0 3.8 3.4
65 and over 11.4 9.2 9.3 2.6 2.5 3.3
Graduated College 13.2 10.6 11.0 3.4 2.6 3.6
Attended College 15.6 12.9 14.6 3.1 3.0 4.0
Graduated High School 17.3 15.5 14.4 4.3 3.7 3.9
Did not Graduate High School 15.0 13.1 13.3 3.4 3.3 4.2
$40,000 or More 14.8 13.2 13.3 3.9 3.6 4.0
$30,000 - $39,999 16.5 14.3 14.0 4.2 3.4 4.1
$20,000 - $29,999 17.1 14.5 13.9 4.0 2.9 4.3
$10,000 - $19,999 15.6 13.0 13.6 2.8 3.1 3.7
Under $10,000 15.7 12.6 12.9 3.2 2.7 3.2

 

Source: Study of Media and Markets, Simmons Market Research Bureau, 1989, 1990,
and 1991.

116

>zrid<
>id<

>date<
>time<

>tml<
>tm2<
>tm3<
>tm4<
>tm5<
>if<

>u1<

>u2<

>u3<

>u4<

APPENDIX 4: The Computerized Questionnaire

[allow 4] [10¢ 0/1] [inputloc 1/1]

[allow 4] [10¢ 1/1] [inputloc 1/1] [caseid]
[allow 6] [setdate date]

[allow 4] [settime time]

[allow 6]
[allow 6]
[allow 6]
[allow 6]
[allow 6]
[if id le <600>][goto u4][endif]

In this short training session, you will go though the
questions you will answer in the formal experiment.

To answer a question, you type in a number for the option
you select and press the "Enter" key.

Are you ready to start the training session?

TYPE <1> rum mss "unwan- KEY To sum. [goto u3]
ups <5> turn mass "ENTER" as! T0 QUIT rm: SESSION.

.88--)

Are you sure that you want to quit?

<1> YES, I WANT To QUIT. [goto endl]
<5) No, I WANT TO CONTINUE.

TYPE A NUMBER AND PRESS TEE ”ENTER" KEY.

33...)

As mentioned earlier, you will watch a TV shopping
program in this experiment. When you hear the very
first word from.the program host, type "1" and press
the "Enter" key to start.

TYPE <1> AND PRESS TEE “ENTER“ KEY TO START.

[goto pl]
an---)

Let's start the formal experiment.

Remember: When you hear the first word from the

program host, type "1" and press the "Enter" key

to get your very first question.

TYPE <1> AND PRESS THE ”ENTER“ KEY IMMEDIATELY
WHEN YOU HEAR THE FIRST WORD FROM THE HOST.

.338-)

117

>p1<

>p1a<

>wla<

>w1b<

118

[start timer]
Now, you are watching Rolodex 3R Electronic Paper
Pocket Planner. Relax and enjoy watching.

As soon as you decide you want to buy this product,
type "1" and press "Enter" key.

When you decide you do not want to buy it,
type "5" and press the "Enter" key.

<1> YES, I WANT TO BUY IT. [goto pla]
<5> No, I DON'T WANT TO BUY IT. [goto wlb]

TYPE A NUMBER AND PRESS THE “ENTER" KEY.

--.-I)

[record timer in tml]
How strong is your intention to buy?

<1> VERY STRONG

<2> SOMEWBAT STRONG

<3> SOMEWEAT WEAR

<4> VERY WEAR

<8> DON'T KNOW

TYPE A NUMBER AND PRESS THE "ENTER" REY.

3----)
That's all questions for Rolodex 3R Planner.
Please continue watching. There is a QVC logo after this

product. When you see the QVC logo, type "2" and press
the "Enter" key to start the second session.

TYPE <2) AND PRESS TEE ”ENTER" KEY TO START THE SECOND SESSION.

3.3:.) [ gate p2 ]

[record timer in tml]
That's the only question for Rolodex 3K Planner.

Please continue watching. There is a QVC logo after this
product. When you see the QVC logo, type "2" and press
the "Enter" key to start the second session.

TYPE <2) AND PRESS TEE ”ENTER” KEY TO START THE SECOND SESSION.

.3...)

>p2<

>p2a<

>w2 a<

>w2b<

119

Now, you are watching the second product, NutriSmart
Nutritional Dietary Monitor.

As soon as you decide you want to buy this product,
type "1" and press "Enter" key.

When you decide you do not want to buy it,
type "5" and press the "Enter" key.

<1> YES, I WANT TO BUY IT [goto p2a]
<5> No, I DON'T wrun- ro BUY IT. [goto w2b]

TYPE A NUMBER AND PRESS THE ”ENTER” KEY

------)

[record timer in tm2]
How strong is your intention to buy?

<1> VERY STRONG

<2) SONEWHAT STRONG

<3> SOMEWEAT WEAR

<4> VERY WEAR

<8> DON'T KNOW

TYPE A NUMBER AND PRESS TEE "ENTER" REY.

----S)

That's all questions for NutriSmart Nutritional
Dietary Monitor.

Please continue watching. There is a QVC logo after this
product. When you see the QVC logo, type "3" and press

the "Enter" key to start the next session.

TYPE <3> AND PRESS THE "ENTER" KEY TO START THE THIRD SESSION.
3....) [ goto p3 ]

[record timer in tm2]

That's the only question for NutriSmart Nutritional
Dietary Monitor.

Please continue watching. There is a QVC logo after this
product. When you see the QVC logo, type "3" and press
the "Enter" key to start the next session.

TYPE <3> AND PRESS THE ENTER KEY TO START THE THIRD SESSION.

-3---)

>p3<

>p3a<

>w3a<

>w3b<

120

Here is the third product, Skewdriver Pro Kit.

As soon as you decide you want to buy this product,
type "1" and press the "Enter" key.

When you decide you do not want to buy it,
type "5" and press the "Enter" key.

<1> YES, I WANT TO BUY IT. [goto p3a]
<5> No, I DON'T WANT TO BUY IT. [goto w3b]

TYPE A NUMBER AND PRESS THE ”ENTER" KEY.

-...->

[record timer in tm3]
Bow strong is your intention to buy?

<1> VERY STRONG
<2> SOMEWHAT STRONG
<3> SOMEWHAT WEAK
<4) VERY WEAK

<8> DON'T KNOW

TYPE A NUMBER AND PRESS THE "ENTER" KEY.

--===>

That's all questions for Skewdriver Pro Kit.

Please continue watching. There is a QVC logo after this
product. When you see the QVC logo, type "4" and press

the "Enter" key to start the last session.

TYPE <4) AND PRESS THE “ENTER" KEY
TO START THE FOURTH AND LAST SESSION.

.....> [goto 941

[record timer in tm3]
That's the only question for Skewdriver Pro Kit.

Please continue watching. There is a QVC logo after this
product. When you see the QVC logo, type "4" and press
the "Enter" key to start the last session.

TYPE <4) AND PRESS THE ”ENTER” KEY
TO START THE FOURTH AND LAST SESSION.

3....)

121

>p4< Now, you are watching the last product in the show,
Electronic Pedometer and Fitness Monitor.

As soon as you decide you want to buy this product,
type "1" and press "Enter" key.

When you decide you do not want to buy it,
type "5" and press the "Enter" key.

<1> YES, I WANT TO BUY IT. [goto p4a]
<5> No, I DON'T WANT TO BUY IT. [goto end2]

TYPE A NUMBER AND PRESS THE "ENTER" KEY.

83....)

>p4a< [record timer in tm4]
Bow strong is your intention to buy?

<1> VERY STRONG

<2> SOMEWHAT STRONG

<3> SOMEWBAT WEAR

<4> VERY WEAR

<8> DON'T KNOW

TYPE A NUMBER AND PRESS THE "ENTER" KEY.

----->

>end1< Thank you! Please take a break. When the program is
over, there is a questionnaire for you to fill out.

TYPE <9> AND PRESS THE "ENTER" KEY TO END THIS PROGRAM.
-----> [goto fnl]

>end2< [record timer in tm4]
Thank you] Please have a break. When the program is
over, there is a questionnaire for you to fill out.
TYPE <9> AND PRESS THE "ENTER" KEY TO END THIS PROGRAM.

8....)

>fn1< [stop timer ] [record timer in.tm5] [complete]

APPENDIX 5: The Paper-and-Pencil Questionnaire

Thank you for your participation in the experiment. We would like you to fill out this
questionnaire to complete the study. Please ask should you have any questions.

Section A

1.

When you were viewing the first product, Rolodex 3K Electronic Paper Pocket
Planner, did you notice a count-down timer on the screen?

1. No (go to next question)
2. Yes > If yes, do you remember how many nrinutes
the timer started from?

 

1. 1minute
2. 2 minutes
3. 3 minutes
4. 4 minutes
8. Don't remember

When you were viewing Rolodex 3K Electronic Paper Pocket Planner, did you
notice a message about how many of that product were available?

1. No (go to next question)
2. Yes > If yes, do you remember the quantity?

 

1. 500 units available
2. 1,000 units available
3. 2,000 units available
4. 3,000 units available
8. Don't remember

When you were viewing the second product, NutriSmart Nutritional Dietary
Monitor, did you notice a count-down timer on the screen?

1. No (go to next question)
2. Yes > If yes, do you remember how many nrinutes
the timer started from?

 

1. 1 minute
2. 2 minutes
3. 3 minutes
4. 4 minutes
8. Don't remember

122

123

When you were viewing NutriSmart Nutritional Dietary Monitor, did you notice a
message about how many of that product were available?

1. No (go to next question)
2. Yes > If yes, do you remember the quantity?

 

l. 500 units available
2. 1,000 units available
3. 2,000 units available
4. 3,000 units available
8. Don't remember

When you were viewing the third product, Skewdriver Pro Kit, did you notice a
count-down timer on the screen?

1. No (go to next question)
2. Yes > If yes, do you remember how many rrrinutes
the timer started from?

 

1. 1 minute
2. 2 minutes
3. 3 minutes
4. 4 minutes
8. Don't remember

When you were viewing Skewdriver Pro Kit, did you notice a message about how
many of that product were available?

1. No (go to next question)
2. Yes > If yes, do you remember the quantity?

 

1. 500 units available
2. 1,000 units available
3. 2,000 units available
4. 3,000 units available
8. Don't remember

When you were viewing the last product, Electronic Pedometer and Fitness
Monitor, did you notice a count-down timer on the screen?

1. No (go to next question)
2. Yes > If yes, do you remember how many minutes
the timer started from?

 

1. 1minute
2. 2 minutes
3. 3 minutes
4. 4 minutes
8. Don't remember

124

 

8. When you were viewing Electronic Pedometer and Fitness Monitor, did you notice
a message about how many of that product were available?
1. No (go to next question)
2. Yes > If yes, do you remember the quantity?
1. 500 units available
2. 1,000 units available
3. 2,000 units available
4. 3,000 units available
8. Don't remember
9. TV home shopping programs often show a count-down timer indicating the viewer
has a limited amount of time to order some merchandise. Generally, how
believable do you think that information is?
1. Very believable
2. Believable
3. Somewhat believable
4. Not believable at all
8. Don't know
10. TV home shopping programs often show a limited quantity of some merchandise
available for viewers to order. Generally, how believable do you think that
information is?
1. Very believable
2. Believable
3. Somewhat believable
4. Not believable at all
8. Don't know
Sectim B

We would like to know how you evaluate the four products in the show. Please make
your evaluation by putting an "x" on the line under a number. If you think that the
product is very good, put an "x" on the line under 1; if you think that it is very bad, put a
"x" on the line under 7. If you think that it is just neutral, put a "x" on the line under 4;
put a "x" any where between the neutral point and an extreme if your evaluation is
moderate in either direction.

1.

How do you evaluate the first product, Rolodex 3K Paper Pocket Planner?

 

 

 

 

 

 

1 2 3 4 5 6 7
Good : : : : : : Bad
Useless : : : : : : Useful
Good quality : : : : : : Poor Quality
Costs too much : : : : : : A Bargain

 

 

Like : : : : : : Dislike

125

How do you evaluate the second product, Nutritional Dietary Monitor?

1 2 3 4 5 6 7
Good . . . . . .

 

 

Useless

 

 

Good quality

 

 

Costs too much

 

Like

 

 

How do you evaluate the third product, Skewdriver Pro Kit?

1 2 3 4 5 6 7
Good

 

 

Useless

 

 

 

 

Good quality
Costs too much

 

 

Like

 

 

How do you evaluate the last product, Electronic Pedometer?

1 2 3 4 5 6 7
Good

 

 

Useless

 

 

Good quality

 

 

Costs too much

 

 

Like

 

 

Bad

Useful

Poor Quality
A Bargain
Dislike

Bad

Useful

Poor Quality
A Bargain
Dislike

Bad

Useful

Poor Quality
A Bargain
Dislike

126

Section C

People have different values and attitudes. Please read the following statements and
circle a number to indicate how you agree or disagree with each of them.

Strongly Strongly
Agm Agm Neutral Disam Disagm
1. I admire people who own expensive
homes, cars, and clothes. ............................. 5 4 3 2 1
2. I have all the things I really need to
enjoy life. .................................................... 5 4 3 2 1
3. I usually buy only the things I need. ........... 5 4 3 2 1
4. I try to keep my life simple, as far as
possessions are concerned. .......................... 5 4 3 2 1
5. My life would be better if I owned
certain things I don't have. ........................... 5 4 3 2 1

6. Some of the most important achievements
in life include acquiring material
possessions. ................................................. 5 4 3 2 1

7. I don't place much emphasis on the
amount of material objects people own

as a sign of success. .................................... 5 4 3 2 1
8. I wouldn't be any happier if I owned nicer

things. .......................................................... 5 4 3 2 1
9. The things I own aren't all that important

to me. ........................................................... 5 4 3 2 l
10. I enjoy spending money on things that

aren't practical. ............................................. 5 4 3 2 1
11. I'd be happier if I could afford to buy more

things. .......................................................... 5 4 3 2 1
12. The things I own say a lot about how well

I'm doing in life. .......................................... 5 4 3 2 1
13. Buying things gives me a lot of pleasure. 5 4 3 2 l
14. I like to own things that impress people. ..... 5 4 3 2 1

15. It sometimes bothers me quite a bit that
I can't afford to buy all the things I'd like. 5 4 3 2 1

(continued)

127

(continued) Strongly Strongly
Am Am 116m MM

16. I don't pay much attention to the material

objects other people own. ............................ 5 4 3 2 1
17. I like a lot of luxury in my life. .................... 5 4 3 2 1
18. I put less emphasis on material things than

most people I know. .................................... 5 4 3 2 1
Section D

The following are some statements about people's shopping orientations. Please indicate
your habits by circling a number for each statement.

Strongly Strongly
Am Am _Neutml _LDisa ee 2182212
1. I learn about products through catalogs
to save time. ................................................. 5 4 3 2 1
2. I spend a lot of time browsing through
stores. ........................................................... 5 4 3 2 1
3. I appreciate a catalog which is attractively
designed. ...................................................... 5 4 3 2 1
4. The availability of high quality
merchandise is very important to me. .......... 5 4 3 2 1
5. I make purchases through catalogs to
save time. ..................................................... 5 4 3 2 1
6. I enjoy traveling to stores and
shopping centers. ......................................... 5 4 3 2 1
7. I enjoy receiving catalogs in the mail. ......... 5 4 3 2 1
8. The availability of a wide selection of
merchandise is very important to me. .......... 5 4 3 2 1
9. I enjoy making purchases through catalogs. 5 4 3 2 1

10. I enjoy browsing through catalogs
with a friend. ................................................. 5 4 3 2 1

11. I enjoy browsing through stores even
though I may not make a purchase. .............. 5 4 3 2 1

(continued)

128

(continued) Strongly Strongly
Agm Agm Neutral Disam Disagme;

12. The availability of retailer brands is

very important to me. ................................... 5 4 3 2 1
13. I enjoy learning about products by

paging through catalogs. .............................. 5 4 3 2 1
14. The availability of unique merchandise

is very important to me. ............................... 5 4 3 2 1
15. I enjoy learning about products by going

to stores. ....................................................... 5 4 3 2 1
16. I enjoy looking at catalogs even though

I may not make a purchase. ......................... 5 4 3 2 1
Section E

People differ in the ways they act and think in different situations. We would like to
know the ways in which you act and think. Please read each statement and circle a
number on the right side of the page.

Never Sometimes Often Alwa s

l. I plan tasks carefully. ......................................... 1 2 3 4
2. I do things without thinking. .............................. 1 2 3 4
3. I make-up my mind quickly. .............................. 1 2 3 4
4. I am happy-go-lucky. ......................................... 1 2 3 4
5. I don't "pay attention." ....................................... 1 2 3 4
6. I plan trips well ahead of time. .......................... 1 2 3 4
7. I am self -controlled. ........................................... 1 2 3 4
8. I concentrate easily. ............................................ 1 2 3 4
9. I save regularly. .................................................. 1 2 3 4
10. I "squirm" at plays and lectures. ......................... 1 2 3 4
11. I am a careful thinker. ......................................... 1 2 3 4

(Continued)

129

(Continued)
PM! Sometimes _OLte_n M

12. I plan for job security. ........................................ 1 2 3 4
13. I say things without thinking. ............................. 1 2 3 4
14. I like to think about complex problems. ............. 1 2 3 4
15. I change jobs. ...................................................... 1 2 3 4
16. I act "on impulse." .............................................. 1 2 3 4
17. I get easily bored when solving thought

problems. ............................................................ 1 2 3 4
18. I act on the spur of the moment. ......................... 1 2 3 4
19. I am a steady thinker. ......................................... 1 2 3 4
20. I change residences. ............................................ 1 2 3 4
21. I buy things on impulse. ..................................... 1 2 3 4
22. I can only think about one problem at a time. 1 2 3 4
23. I change hobbies. ................................................ 1 2 3 4
24. I spend or charge more than I earn. .................... 1 2 3 4
25. I often have extraneous thoughts when thinking. 1 2 3 4
26. I am more interested in the present than

the future. ............................................................ 1 2 3 4
27. I am restless at the theater or lectures. ................ 1 2 3 4
28. I am future oriented. ........................................... 1 2 3 4
Section F

The last section includes some questions about your media usage and shopping
experiences, as well as a few questions about yourself.

1. On a typical weekday (Monday through Friday) how many hours of TV do you
watch?

None

Less than 1 hour
1-2 hours

3-4 hours

5 or more hours
Don't know

@MPP’N!‘

130

On a typical weekend day (Saturday or Sunday) how many hours of TV do you
watch?

”MPP’N!‘

None

less than 1 hour
1-2 hours

3-4 hours

5 or more hours
Don't know

How many magazines do you read regularly?

@MPP’N!‘

None

1 magazine

2 magazines

3 magazines

4 or more magazines
Don't know

How many newspapers do you read regularly?

@MPPN!‘

None

1 newspaper

2 newspapers

3 newspapers

4 or more newspapers
Don't know

How many times did you make purchases by mail in the past six months?

1
2
3
4. 3 times
5
8

None
1 time
2 times

4 or more times
Don't know

Have you ever ordered anything from TV home shopping programs?
1. Yes

2.
8.

No
Don't know

Have you ever watched any TV home shopping channels before?

1. Yes

2.
8.

No
Don't know

131

8. What is your gender?

1. Male
2. Female

9. What is your age as of your last birthday?

years old.

10. How much money do you have to spend each month after you pay for tuition,
rent, utilities and food?

Under $99

$ 100-$ 199
$200-$299
$300-$399
$400-$499
$500 or more
Don't know

WQMPP’NF

11. Which category best approximates the total yearly income of your parent(s)?

Under $9,999
$10,000-$ 19,999
$20,000-$29,999
330,000-339,999
$40.000-$49.999
$50,000 or more
Don't know

909$“?pr

12. What is your major?

 

That's all. Thank you so much for your time and cooperation!

BIBLIOGRAPHIES

BIBLIOGRAPHY

Ajzen, I. (1985). From Intentions to Actions: A Theory of Planned Behavior. In J.

Kuhland & J. Beckman (Eds), Action-Control: Erom Cogm'tions to Behavior (pp. 11-
39). Heidelberg: Springer.

Ajzen, I. (1991). The Theory of Planned Behavior. Organizational Behavior and Human
Decision Processes 5_0(2), 179-211.

Artzt, E. L. (1994, May 23). P&G’s Artzt: TV Advertising in Danger. Advertising Age, p.
24, 40, 42.

Belk, R. W. (1983). Wordly Possessions: Issues and Criticisms. In R. P. Bagozzi & A.
M. Tybout (Eds), Advances in Consumer Research (pp. 514—519). Ann Arbor, MI:
The Association for Consumer Research.

Belk, R. W. (1984). Three Scales to Measure Constructs Related to Materialism:
Reliability, Validity, and Relationships to Measures of Happiness. In T. C. Kinnear
(Ed.), Advances in Consumer Research (pp. 291-297). Provo, UT: The Association of
Consumer Research.

Belk, R W. (1985). Materialism: Trait Aspects of Living in the Material World. Journal of
Consumer Researcn, _1;(Dec.), 265-280.

Bellenger, D. N., Robertson, D. H., & Hirschman, E. C. (1978). Impulse Buying Varies
by Product. Jorunal of Advertising Research, 1_8_(December), 15-18.

Bentler, P. M. (1989). EQS: Structural Equations Progmm Manual. Los Angeles: BMDP
Statistical Software.

Bozzolo, A. M., & Brock, T. C. (1992). Unavailability Effects on Message Processing: A
Theoretical Analysis and an Empirical Test. Basic and Applied Social Psychology,
fi( 1), 93-101.

Brand, J. E., & Greenberg, B. S. (1992). commercials in the Classroom: The Impact of
Channel One Advertising. Manuscript Michigan State University.

Brehm, J. W. (1966). A Theog of Psychological Reacmnce. New York: Academic.

Brehm, S. S., & Brehm, J. W. (1981). Psychplogjcal Reactance: A Theog of Freedom
and Control. New York: Academic.

Brock, T. C. (1968). Implications of Commodity Theory for Value Change. In A. G.
Greenwald, T. C. Brock, & T. M. Ostrom (Eds), Psychological Foundations of
Attitudes (pp. 243-275). New York: Academic.

Brock, T. C., & Brannon, L. A. (1992). Liberalization of Commodity Theory. Basic and
Applied Social Psycholpgy, 13(1), 135-144.

Campbell, D. T. (1969). Materialism. In J. P. Robinson & P. R. Shaver (Eds), Measures
of Social Attitudes (pp. 651-652). Ann Arbor, MI: Institute for Social Research.

132

133

Churchill, G. A., & Mochins, G. T. (1979). Television and Interpersonal Influences on
Adolescent Consumer learning. Journal of Marketing Research, gJune), 23-35.

Clover, V. T. (1950). Relative Importance of Impulse Buying in Retail Stores. Journal of
Marketing, QJUIY). 66-70.

Cobb, C. J., & Hoyer, W. D. (1986). Planned Versus Impulse Purchase Behavior.
Journal of Retailing, 62(4), 384-409.

Cox, K. (1964). The Responsiveness of Food Sales to Shelf Space Changes in
Supermarkets. Journal 01 Marketing Res_ea_[ch, I (May). 63-67.

Csikszentrrrihalyi, M., & Rochberg-Halton, E. (1981). The Meanings of Thingp: Domestic
Symbols and the Self. Cambridge: Cambridge University Press:

Deshpande, R., & Krishnan, S. (1979). Consumer Impulse Purchase and Credit Car
Usage: An Empirical Examination Using the Log linear Model. In J. C. Olson (Ed.),
Advgces in Consumer Emh (pp. 792-795). Ann Arbor, MI: Association for
Consumer Research.

Donaton, S. (1993, July 12). Pleased to be TV’s Wal-Mart. Advertising Age, p. 12.

Elmer-DeWitt, P. (1993, April 12). Take a Trip Into the Future on the Electronic
Superhighway. Time, p. 50-58.

Engel, J., & Blackwell, R. (1982). Consumer Behavior. Chicago: Dryden Press.

Faber, R. J., & O’Guinn, T. C. (1989). Classifying Compulsive Consumers: Advances in
the Development of a Diagnostic Tool. In T. K. Srull (Ed), Advances in Consumer
Research (pp. 738-744). Provo, UT: Association for Consumer Research.

Fawcatt, A. W. (1994, October 3). Interactive Awareness Growing. Advertising Age, p.
8.

Ferraro, C. (1993, July 13). QVC Buying Rival Home Shopping Network Investor
Business Daily, p. 1.

Fishbein, M. (1980). A Theory of Reasoned Action: Some Applications and Implications.

In H. Home & M. Page (Eds), Nebraska Symmium on Motivation (pp. 65-116).
Lincoln, NB: University of Nebraska Press.

Fishbein, M., & Ajzen, I. (1975). Belief Attitude Intention and Behavior". An
Introduction to Theog and Research. Reading, MA: Addision-Wesley.

 

Fishbein, M., & Ajzen, I. (1980). Predicting and Understanding Consumer Behavior:
Attitude-Behavior Correspondence. In I. Ajzen & M. Fishbein (Eds), Understanding

Attitudes and Predicting Social Behavior (pp. 148-172). Prentice Hall.

Fromkin, H. L. (1968). Affective and Valuational Conmuences of Self-Perceived
Uniqueness Deprivation. Unpublished dissertation, Ohio State University, Columbus.

Fromkin, H. L (1970). Effects of Experimentally Aroused Feelings of Undistinctiveness
Upon Valuation of Searce and Novel Experiences. Journal of Personality and Social
Psychology (16), 521-529.

134

Fromkin, H. L. (1971). A Commodity Theory Analysis of Consumer Preferences for

Scarce Products. In man of the 79th Am Convention of the American
Psychological Association (6), 521-654.

Goldman, P., Deth T. M., & Hubbard, H. (1979). Energy: Togetherness.
Newsweek, 22(April 16), 44-46.

Grant, A. E., Guthrie, K. K, & Ball-Rokeach, S. J. (1991). Television Shopping: A
Media System Dependency Perspective. Commnm'cation Research, _1_8_(6), 773-798.

Graves, R. E., & Bradley, R. (1988). More on Millisecond Timing and Tachistoscope
Applications for the IBM PC. Behavior Research Methods, Instruments, & Commters,
E4), 408-412.

Graves, R. E, & Bradley, R. (1991). Millisecond Timing in the IBM PC/XT/AT and
PS/2: A Review of the Options and Corrections for the Graves and Bradley Algorithm.

Behavior Research Methods, Instruments, & Computers, £3), 377-379.

Hunter, J. E., & Gerbing, D. W. (1982). Unidimensional Measurement, Second Order
Factor Analysis, and Causal Models. Research in Orgm'zational Behavior (pp. 267-
320). JAI Press, Inc.

Hunter, J. E., & Lim, T. S. (1987). PACKAGE Unpublished manuscript. Department of
Communication, Michigan State University.

James, J. S., Schneider, W., & Hinds, K. (1992). Mel Lab: Expgriments in Perception,

Cogm'tion, Social Enychology mm Human Factors. Pittsburgh, PA: Psychology
Software Tools.

James, L. E., & Cunningham, I. C. M. (1987). A Profile of Direct Marketing Television
Shoppers. Joum of Dimt Marketing. 1(4), 12-23.

JUreskog, K. G., & Scrbom, D. (1989). LISREL 7: A Guide to the Progmm and
Applications (2nd. ed.). Chicago: SPSS Inc.

Kelman, H. C. (1953). Attitude Change as a Function of Response Restriction. Human
Relations(6), 185-214.

Knishinsky, A. (1982). The Effects of Scarcig of Material and Exclusivity of Information

on Industrial Buyer Perceived Risk in Provoking a Purchase Decision. Unpublished
dissertation, Arizona State University.

Kollat, D. T., & Willett, R. P. (1967). Customer Impulse Purchasing Behavior. Journal of
Marketing Research, 4(Feb.), 21-31.

Lee, D. W. (1990). The Influence of Shomr Tm and Mood on Impulse Purchasing
Behavior. Unpublished dissertation, Michigan State University.

Lynn, M. (1989). Scarcity Effects on Value: Mediated by Assumed Expensiveness?
Journal of Economic Psychology, IQ, 257-274.

Lynn, M. (1991). Scarcity Effects on Value: A Quantitative Review of the Commodity
Theory Literature. Psychology and Marketing, 8, 67-78.

135

Lynn, M. (1992). Scarcity’s Enhancement of Desirability. The Role of Naive Economic
Theories. Basic and Applied Social Psychology, ma), 67-78.

McCIave, J. T., & Frank H. Dietrich, I. (1985). Statistics (3rd ed.). San Francisco, CA:
Dellen Publishing Company.

Mochins, G. P., & Churchill, G. A. (1978). Consumer Socialization: A Theoretical and
Empirical Analysis. Journal of Marketing Research, 15(4), 544-609.

Mochins, G. P., & Moore, R. L. (1982). A longitudinal Study of Television Advertising
Effects. Journal of Consumer Research, 2(Dec.), 279-286.

Mordkoff, J. T., & Egeth, H. E. (1993). Response Time and Accuracy Revisited:
Converging Support for the Interactive Race Model. Journal of Exmrimental

Psychology: Human Perception and Performanc_e, 1_9(5), 981-991.

Murphy, P. E., & Enis, B. M. (1986). Classifying Products Strategically. Journal of
Marketing, flluly). 24-42.

O’Guinn, T. C., & Faber, T. (1989). Compulsive Buying: A Phenomenological
Exploration. Journal of Consumer Research, 16(September), 147-156.

OECD (1992). New Home Shopping Technologies. Organization for Econonric
Cooperation and Development, France.

Pedhazur, E. J. (1982). Multiple Remion in Behavioral Research: Explanation and
Prediction (2nd ed.). New York, NY: Holt, Rinehart and Winston.

Peter, J. P., & Olson, J. C. (1990). Consumer Behavior and Marketing Strategy (2nd
ed.). Homewood, IL: Irwin.

Phillips, H., & Bradshaw, R. (1993). How Customers Actually Shop: Customer
Interaction with the Point of Sale. Journal of the Market Research Socieg, 35(1), 51-
62.

Pollay, R. W. (1968). Customer Impulse Purchasing Behavior. A Reexamination. Journal
of Marketing Researgh, _5_(Aug.), 323-5.

Prasad, V. K. (1975). Unplanned Buying in Two Retail Settings. Journal of Retailing,
5_1(Fall), 3-12.

Reilly, P. M. (1993, November 16). TV Shopping Hooks High-Toned Viewers. The Wall
Street Journal p. B3.

 

Richins, M. L, & Dawson, S. (1990). Measuring Material Values: A Preliminary Report
of Scale Development. In M. E Goldberg, G. Gorn, & R. W. Pollay (Eds),
Advances in Consumer Research (pp. 169-175). Provo, UT: The Association for
Consumer Research.

Richins, M. L., & Dawson, S. (1992). A Consumer Values Orientation for Materialism
and Its Measurement: Scale Development and Validation. Journal of Consumer
Research 3(3), 303-316.

 

136

Robichaux, M., & Roberts, J. L. (1993, August 12). Home Shopping’s Roy Speer
Resigns Chairman’s Post. ijhg Wfll Street hm, p. B6.

Rook, D. W. (1987). The Buying Impulse. Journal of Consumer Research,
15(8eptember), 189-199.

Scherer, F. M., & Ross, D. (1990). Industrial Market Structure and Economic
Performance (3rd ed.). Boston, MA: Houghton Mifflin.

Scherhom, G. (1990). The Addictive Trait in Buying Behavior. Journal of Consumer
Policy, 13(1), 33-51.

Schultz, D. E (1992). From the Editor: Who Will Fulfill the Promise? Journal of Direct
Marketing, 6(1), 4.

Sheppard, B. H., Hartwick, J., & Warshaw, P. R. (1988). The Theory of Reasoned
Action: A meta-Analysis of Past Research with Recommendations for Modifications
and Future Research. Journal of Consumer Research 1_5(December), 325-343.

Snyder, C. R., & Fromkin, H. L. (1980). Uniqueness: The Human Pursuit of Difference.
New York: Plenum.

Stern, H. (1962). The Significance of Impulse Buying Today. Journal of Marketing
2_6_(April), 59-62.

Stone, B. (1989). Successful Direct Marketing Methods (4th ed.). Lincolnwood, IL: NTC
Business Books.

Sugar: A Story of Mismanagement.(1974). Business Week, _1_2(November 16).

Syzibillo, G. J. (1973). The Effects of Price and Scarcig on the Valuation of Fashions by
Fashion ijnion leaders and Non-Opinion Leaders. Unpublished dissertation, Purdue

University.

Tabachnick, B. G., & Fidel], L. S. (1983). Using Multivariate Statistics. New York:
Harper & Row.

Tryon, W. W., & Mulloy, J. M. (1993). Further Validation of Cornputer-Assessed
Response Time to Emotionally Evocative Stimuli. Journal of Personality Assessment,
@112), 231-236.

Ward, S., & Wackrnan, D. (1971). Family and Media Influences on Adolescent Learning.
American Behavioral Scientist, HMS-427).

Weinberg, P., & Gottwald, W. (1982). Impulsive Consumer Buying as a Result of
Emotions. Journal of Business Research(March), 43-57.

West, J. C. (1951). Results of Two Years of Study into Impulse Buying. Journal of
Marketing, fiJanuary), 362-363.

Wicklund, R. A. (1974). Freedom and Reactance. Hillsdale, NJ: Lawrence Erlbarun
Associates, Inc.

137

Worchel, S., lee, J., & Adewole, A. (1975). Effects of Supply and Demand on Ratings of
Object Value. Journal of Egrsonalig m SocinI Echology, 3;, 906-914.

Yamauchi, K T., & Templer, D. I. (1982). The Development of a Money Attitude Scale.
Journal of Personality Assessment, 4_6(Oct.), 522-528.

Yun, K A. (1992). The Effect of Scarcig Mess_ages and Materialistic Attitudes on lilting,
Attraction, and Value. Unpublished master’s thesis, Kansas State University.

Zinn, L., George, G. D., Shoretz, R., Yang, D. J., & Forest, S. A. (1993, July 26).
Retailing Will Never be the Same. Business Week p. 54-57, 59-60.

 

"Illillllllllillllllllf