CONSUMER INFORMATION PROCESSING DIFFERENCES IN RESPONSE TO FIRM
PRICING: REMEMBERING VERSUS REASONING

By
Chang Seob Yeo

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Business Administration—Doctor of Philosophy
2014

ABSTRACT
CONSUMER INFORMATION PROCESSING DIFFERENCES IN RESPONSE TO FIRM
PRICING: REMEMBERING VERSUS REASONING
By
Chang Seob Yeo
Previous literature has found evidence of firm’s motivation to encourage consumers’ stockpiling
behavior as well as benefits occurred from stockpiling. Stockpiling in this dissertation refers to
the propensity of consumers to increase their inventory level for future consumption of stock-up
products which are nonperishable and in a unit size that consumers frequently purchase. A sales
promotion (i.e., a price discount) is a fundamental condition for consumers’ stockpiling behavior.
Four experiments of this dissertation intend to make contributions to literature of sales
promotions, stockpiling, and price knowledge by answering (1) What types of price discounts
trigger stockpiling behavior? and (2) Would types of discounts operate differently for different
products (i.e., benefit changed products) and for different situations (i.e., an introduction of a
new product under an established brand name or under new brand name)? as well as additionally
investigating the direct effects depth and frequency of a discount on consumers’ price knowledge
and perceived similarity with overall shopping experiences. Across all of four experiments,
significant direct effects of discount depth and frequency on stockpiling; that interaction are
found. Such main effects are moderated by benefit changes, new product introduction, and new
brand introduction and also directionality among moderators vary. Lastly, evidence for proposed
mediators which are price recall error, price comparison error, and consumers’ perceived
similarity with overall shopping experiences is not found.

Copyright by
CHANG SEOB YEO
2014

ACKNOWLEDGEMENTS

Without the support, patience, and guidance of many people, this dissertation would not have been
completed
First and foremost, I would like to express my sincerest gratitude to Dr. Roger J. Calantone,
for his excellent guidance, unstinting support, and genuine concern. I would also like to thank my
committee members, Dr. R. Dale Wilson, Dr. Thomas J. Page, Dr. Clay M. Voorhees, and Dr.
Donald E. Conlon. Their academic support and encouragement are greatly appreciated.
Also I would like to thank Praneet Randhawa and Hannah S. Lee who are my special
friends. I could make it through all the challenges, thanks to them.
My lovely wife, Sujeong Lee, and my sweet daughter, Claire S. Yeo, you all are my
reasons for living and for walking forward. Thank you so much for your generous support and
love. My father and mother, Tae Yeol Yeo and Youngsook Kim, words would not be enough to
express my appreciation and love. Father, I believe you are watching all from heaven and really
hope you could be here with me. I send my special appreciation and recognition to my parents.
Also I thank my parents-in-law, Moonseok Lee and Jongsu Lee, for their patience as well as
emotional support.

iv

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................................ vii
LIST OF FIGURES ....................................................................................................................... ix
INTRODUCTION ...........................................................................................................................1
CONCEPTUAL BACKGROUND ..................................................................................................5
Consumer Stockpiling Behavior ..................................................................................................5
Consumer Price Search Behavior ................................................................................................7
Price Cognition and Representation of Price Information ...........................................................8
The Feeling of Familiarity .........................................................................................................10
STUDY 1 .......................................................................................................................................13
Hypothesis Development ...........................................................................................................13
The Impact of Discount Depth .............................................................................................13
The Impact of Discount Frequency......................................................................................15
The Mediation Effects of PRE, PCE, and Perceived Familiarity ........................................17
Research Design.........................................................................................................................19
Manipulation and Product Category ....................................................................................19
Procedure of Shopping Tasks ...............................................................................................21
Measures ..............................................................................................................................22
Data Collection and Samples ...............................................................................................23
Means of Analysis ................................................................................................................24
Results ........................................................................................................................................24
Manipulation Checks ...........................................................................................................25
Assessment of the Measures ................................................................................................27
Assessment of the Main Effects ...........................................................................................29
Assessment of the Mediation Effects ...................................................................................31
Discussion ..................................................................................................................................32
STUDY 2 .......................................................................................................................................34
Research Design.........................................................................................................................34
Results ........................................................................................................................................35
Manipulation Checks ...........................................................................................................36
Assessment of the Measures ................................................................................................38
Assessment of the Main Effects ...........................................................................................39
Assessment of the Mediation Effects ...................................................................................44
Discussion ..................................................................................................................................46
STUDY 3 .......................................................................................................................................47
Research Design.........................................................................................................................47
Results ........................................................................................................................................48

v

Manipulation Checks ...........................................................................................................48
Assessment of the Measures ................................................................................................48
Assessment of the Main Effects ...........................................................................................51
Assessment of the Mediation Effects ...................................................................................56
Discussion ..................................................................................................................................58
STUDY 4 .......................................................................................................................................60
Research Design.........................................................................................................................60
Results ........................................................................................................................................61
Manipulation Checks ...........................................................................................................61
Assessment of the Measures ................................................................................................63
Assessment of the Main Effects ...........................................................................................64
Assessment of the Mediation Effects ...................................................................................68
Discussion ..................................................................................................................................71
GENERAL DISCUSSION ............................................................................................................72
Managerial Implications ............................................................................................................74
Theoretical Implications ............................................................................................................75
APPENDICES ...............................................................................................................................78
Appendix A Complete Scenarios: Study 1 .................................................................................79
Appendix B Complete Scenarios: Study 1.................................................................................84
Appendix C Complete Scenarios: Study 1.................................................................................86
Appendix D Complete Scenarios: Study 1 ................................................................................88
REFERENCES ..............................................................................................................................90

vi

LIST OF TABLES

Table 1-1: Correlation Matrix ........................................................................................................26
Table 1-2: Measurement Model .....................................................................................................27
Table 1-3: Descriptive Statistics ....................................................................................................28
Table 1-4: Descriptive Information of Measures ...........................................................................29
Table 1-5: Multivariate Tests .........................................................................................................29
Table 1-6: Univariate Tests ............................................................................................................30
Table 1-7: Test of Mediation Effects ..............................................................................................32
Table 2-1: Correlation Matrix ........................................................................................................37
Table 2-2: Measurement Model .....................................................................................................38
Table 2-3: Descriptive Information of Measures ...........................................................................39
Table 2-4: Descriptive Statistics ....................................................................................................40
Table 2-5: Multivariate Tests .........................................................................................................41
Table 2-6: Between-Subjects Tests for an Interaction between DEP and FRQ .............................41
Table 2-7: Univariate Test ..............................................................................................................42
Table 2-8: Tests of Mediation Effects ............................................................................................44
Table 3-1: Correlation Matrix ........................................................................................................50
Table 3-2: Measurement Model .....................................................................................................51
Table 3-3: Descriptive Information of Measures ...........................................................................51
Table 3-4: Descriptive Statistics ....................................................................................................53
Table 3-5: Multivariate Tests .........................................................................................................54
Table 3-6: Univariate Tests ............................................................................................................54
Table 3-7: Between-Subjects Tests for an Interactions ..................................................................55

vii

Table 3-8: Tests of Moderated Mediation Effects ..........................................................................57
Table 4-1: Correlation Matrix ........................................................................................................62
Table 4-2: Measurement Model .....................................................................................................63
Table 4-3: Descriptive Information of Measures ...........................................................................64
Table 4-4: Descriptive Statistics ....................................................................................................65
Table 4-5: Multivariate Tests .........................................................................................................66
Table 4-6: Univariate Tests ............................................................................................................66
Table 4-7: Between-Subjects Tests for an Interactions ..................................................................68
Table 4-8: Tests of Moderated Mediation Effects ..........................................................................70
Table 5: Summary of Hypotheses Testing .....................................................................................72

viii

LIST OF FIGURES

Figure 1: Proposed Conceptual Model ..........................................................................................24
Figure 2: Mean Plots for Discount DEP and FRQ .........................................................................31
Figure 3: 2x2 Mean Plots for Discount DEP, Discount FRQ, and BEF ........................................43
Figure 4: 2x2 Mean Plots for Discount DEP, Discount FRQ, and NP ..........................................56
Figure 5: 2x2 Mean Plots for Discount DEP, Discount FRQ, and NB ..........................................69
Figure 6–1: Price Tags of Shopping Task #1: Low DEP and Low FRQ........................................79
Figure 6–2: Price Tags of Shopping Task #2: Low DEP and Low FRQ........................................80
Figure 6–3: Price Tags of Shopping Task #1: High DEP and High FRQ ......................................80
Figure 6–4: Price Tags of Shopping Task #2: High DEP and High FRQ ......................................81
Figure 6–5: Price Tags of Shopping Task #1: High DEP and Low FRQ .......................................81
Figure 6–6: Price Tags of Shopping Task #2: High DEP and Low FRQ .......................................82
Figure 6–7: Price Tags of Shopping Task #1: Low DEP and High FRQ .......................................82
Figure 6–8: Price Tags of Shopping Task #2: Low DEP and High FRQ .......................................83
Figure 7–1: Price Tags of Shopping Task #2: Low BEF, High DEP, and High/Low FRQ ............84
Figure 7–2: Price Tags of Shopping Task #2: Low BEF, Low DEP, and High/Low FRQ ............84
Figure 7–3: Price Tags of Shopping Task #2: High BEF, Low DEP, and High/Low FRQ ............85
Figure 7–4: Price Tags of Shopping Task #2: High BEF, High DEP, and High/Low FRQ ...........85
Figure 8–1: Price Tags of Shopping Task #2: NP, Low DEP, and High/Low FRQ .......................86
Figure 8–2: Price Tags of Shopping Task #2: NP, High DEP, and High/Low FRQ ......................87
Figure 9–1: Price Tags of Shopping Task #2: NB, High DEP, and High/Low FRQ .....................88
Figure 9–2: Price Tags of Shopping Task #2: NB, Low DEP, and High/Low FRQ ......................89

ix

INTRODUCTION
For marketing managers, how to design a sales promotion in a way that it motivates consumers to
purchase more than usual (i.e., stockpiling behavior) and more frequently is an important strategic
decision (e.g., Wansink 1994; Wansink, Kent, and Hoch 1998) along with purchase incidence and
brand choice of consumers. Managers would encourage consumers to stockpile, if doing so allows
them to passes inventory holding costs to consumers (e.g., Blattberg, Fppen, and Liebemian 1981).
The transmission of inventory holding costs reduces the level of firm’s inventory which may be
out-of date and to be shipped back. Moreover, while conventionally stockpiling was deemed as a
loss factor because it is commonly driven by a sales promotion which reduces profit margin as
well as because promotion induced stockpiling delays consumers’ subsequent purchases (i.e., postpromotion dip) in which they would buy the promoted brand at regular price (e.g., Ailawadi,
Gedenk, Lutzky, and Neslin 2007). However, Ailawadi, Gedenk, Lutzky, and Neslin (2007) finds
evidence of important benefits of stockpiling (e.g., preemptive brand switches and repeat
purchases). On consumers’ side, a trade-off is between the promotional savings for future
consumption and the added costs incurred by carrying extra inventory (Assuncao and Meyer 1993;
Helsen and Schmittlein 1992; Krishna 1992, 1994).
Stockpiling is to be defined as the inclination of consumers to increase their inventory
level by purchasing more items than usual at a purchase occasion and/or by advancing their
purchase time before the expected time of their subsequent purchase (e.g., Neslin, Henderson, and
Ouelch 1985; Macé and Neslin 2004; Blattberg and Neslin 1990). Stockpiling would occur for
nonperishable and stockable products. Thus, stock-up products are the focus of this dissertation
which refer to products that are nonperishable and in a unit size consumers frequently purchase
and consume (Litback, Calantone, and Warshaw 1985).

1

A sales promotion is a fundamental condition to trigger consumes’ stockpiling behavior
(Blattberg, Eppen, and Lieberman 1980; Neslin 2002). The propensity of consumers to stockpile
the promoted brand would be intensified when a sales promotion is infrequently provided, when
the promotional saving is greater than expected, and when consumers are more certain about the
regular as well as the deal price of the product (Kristiaan and Schmittlein 1992). In addition,
consumers who have higher income Stilley, Inman, and Wakefield (2010) and loyal Sun, Neslin,
and Srinivasan (2003) to the promoted brand are more likely to stockpile.
There have been various studies investigating consumers’ behavior on firm’s promotional
deals, but less attention was directed toward comprehending the underlying psychological process
of stockpiling behavior. One primary exception is Wansink, Kent, and Hoch (1998). They
examined an anchoring and adjustment judgment process as a psychological process through
which consumers go to make their purchase quantity decision. Also we have paucity in our
understanding of how the relationship between a discount and stockpiling would vary.
Thus, four experiments of this dissertation designs and implements various types of price
discounts and consumers’ price knowledge and their perceived overall similarity with shopping
experiences are introduced as mediators of the relationship between a price discount and
stockpiling. Furthermore, this dissertation investigates how proposed direct and indirect
relationships would vary when there is a change in product size as well as in market situation (i.e.,
new product introduction and new brand introduction). In general, this dissertation intends to
answer the following questions: (1) What types of price discounts trigger stockpiling behavior?
and (2) Would types of discounts operate differently for different products (i.e., benefit changed
products) and for different situations (i.e., an introduction of a new product under an established
brand name or under new brand name)? In addition to the main effects of a price discount on

2

consumers’ stockpiling behavior, its direct effects on price consumers’ price knowledge and
perceived similarity with overall shopping experiences are tested. A product category focused here
is carefully chosen. It should be nonperishable, stockable, regularly purchased, and commonly
discounted product.
This dissertation attempts to make the following contributions by answering those
research questions above. The importance of understanding how a sales promotion should be
designed to motivate consumers’ stockpiling behavior is addressed in a recent study examining
effects of stockpiling (Ailawadi, Gedenk, Lutzky, and Neslin 2007). Thus, first, this dissertation
attempts to make a contribution to the literature of sales promotion and stockpiling by
demonstrating how a sales promotion can be configured and scheduled to drive consumers’
stockpiling and how such relationships vary across product types as well as situations.
Attractiveness of a sales promotion is to be determined by how deep a discount is and how
frequently a discount has been offered over time. If a given sales promotion is suffice for attracting
consumers to stockpile, a firm will enable to capture benefits (e.g., reduced inventory costs,
increased repeat purchases, preempted switches to competitors’ brands) and, in turn, to achieve its
strategic goals. Also the attractiveness of the same price discount may vary when the product size
of the promoted brand is changed and when a new product is launched under the established brand
or the new brand. As a result, firms will understand how to better calibrate their pricing discounts
when trying to trigger stockpiling behavior for different products and for different situations.
The need to understand consumers’ process through which they make “how much”
decision pulled our attention (e.g., Wansink, Kent, and Hoch 1998). The second contribution is to
identify key mediating mechanisms which underlie the relationship between a price discount and
consumers’ stockpiling behavior. Proposed mediators are price knowledge and consumers’

3

perceived similarity between their shopping events over time. In addition, both absolute (i.e., price
recall) and relative (i.e., price comparison) price knowledge are measured and utilized in
hypotheses testing to capture more aspects of consumer price knowledge. That is, it is argued that
both measures are necessary to more fully capture and comprehend consumer price knowledge
which is a direct response to the necessity of multiple measures of price knowledge by Anderson
and Simester (2009, p. 153).
Overall, four experiments find that stockpiling behavior is more likely to occur when a
deeper discount is less frequently offered. Regarding to price knowledge, when a shallower
discount is more frequently provided, consumers show the propensity to more thoroughly search
price and their price knowledge is less error prone. Also among proposed moderators, benefit
change and new product introduction attenuate the main effects of discount depth and frequency
on stockpiling while new brand introduction amplified such main effects on stockpiling.

4

CONCEPTUAL BACKGROUND
Consumer Stockpiling Behavior
Consumers’ stockpiling behavior has been commonly found in various studies (Bucklin and Gupta
1992; Bucklin, Gupta, and Siddarth 1998; Chintagunta and Haldar 1998; Gupta 1988; Macé and
Neslin 2004; Van Heerde, Leeflang, and Wittink 2000, 2004). A fundamental drive consumers’
stockpiling behavior is a deal which allows consumers to have promotional savings (Blattberg,
Eppen, and Lieberman 1981; Neslin 2002) which induces them to buy more or sooner than usual
(Blattberg, Eppen, and Lieberman 1981; Neslin, Henderson, and Quelch 1985). In other words,
price and an offered sales promotion are important criteria consumers utilize to decide “what to
buy” as well as “how many to buy” (e.g., Tellis and Zufryden 1995). Retailers offer promotional
activities because promotion-induced stockpiling shifts inventory holding costs of retailers to
consumers (Blattberg, Eppen, and Lieberman (1981). Retailers’ inventory holding costs comes
from two sources that are the capital entangled in inventory as well as the space utilized to hold
inventory. Consumers’ costs generated by carrying a couple more products than usual would not
be as costly as retailers’. One of motivations for manufacturers to trigger consumers’ stockpiling
behavior is to boost nonusers’ trial of their brands (Blattberg, Eppen, and Lieberman 1981).
Ailawadi, Gedenk, Lutzky, and Neslin (2007) find that, after consumers leveraging a sales
promotion, the resultant extra inventory results in the increased consumers’ consumption of the
promoted brand (i.e., flexible consumption), the preempted future purchases of the promoted brand
(i.e., postpromotion dip) (e.g., Neslin, Powell, and Stone 1995), the preempted future purchases of
competitors’ brands (e.g., Lodish 1986), and the probable repeat purchase of the promoted brand.
Also they showed preempted switching to competitors’ brands and consumers’ repeated purchases
of the promoted brand enable manufacturers to compensate their financial loss from delayed
5

subsequent purchase in which consumes would have bought at a regular price. Consumers’
stockpiling behavior increases the sales of the promoted brand and such an effect was greater for
higher-share brands than for the smaller-share brands (Chan, Narasimhan, and Zhang 2008; Macé
and Neslin 2004).
For consumers, buying more is a trade-off between costs engendered by carrying more
items than usual and savings from a reduced price for future consumption. Promotional savings
encourage consumers’ stockpiling behavior for future consumption (Chintagunta 1993; Heilman,
Nakamoto, and Rao 2002; Nijs, Dekimpe, Steenkamp, and Hanssens 2001; Pauwels, Hanssens,
and Siddarth 2002). Overall stockpiling is a rational purchase behavior of consumers to minimize
their costs (Dellaert, Golounov, and Prabhu 2005). Consumers’ purchase quantity decision may be
based on the utility of a sales promotion in the long term as well as the short term (e.g., Litvack,
Calantone, and Warshaw 1985) or consumers simply cannot go by a deal (e.g., Meyer and
Assuncao 1990). Also consumers’ long-term exposure to sales promotion leads to “lie-in-wait”
which manifests consumers’ stockpiling behavior on a given purchase incidence and the
postponement of their subsequent purchase incidence till another promotion is available (Mela,
Jedidi, and Bowman 1998). That is, previously paid prices influence on consumers to develop their
future price expectation (or expectation on a sales promotion) (e.g., Gonul and Srinvasan 1996;
Jacobson and Obermiller 1990). In other words, consumers adapt their purchase quantity decision
to a price pattern over time (Meyer and Assuncao 1990). Krishna (1994) shows, when deals are
provided in regular pattern and consumers can predict a future deal, they purchase more. However,
the same study finds that irregularity in firm’s deal offering plan increased purchase quantity. The
latter may appear because consumers do not want to miss a chance to get promotional savings for
future consumption of the promoted product.

6

Furthermore previous literature unveiled consumers’ heterogeneous characteristics in
their reactions to a sales promotion. Stockpiling behavior is more likely to occur for consumers
with higher income than those who with lower income (Stilley, Inman, and Wakefield 2010). The
sensitivity to future savings is higher for consumers with tighter budget. (Hoch et al. 1995).
Consumers who are loyal to the promoted brand may more recognize the utility in their tradeoff
between the incurred inventory holding costs and the increased future consumption with
promotional savings than non-loyal consumers do (Krishna 1994; Sun, Neslin, and Srinivasan
2003). In contrast, a sales promotion does not motivate brand switchers to stockpile at all (Chan,
Narasimhan, and Zhang 2008). They also found that incurred promotional savings by stockpiling
the promoted brand on future consumption are more significant for heavy users than for light
users.
Consumer Price Search Behavior
Various studies argue that a large proportion of consumers’ price search is limited and also only
a small fraction of alterative products constitute consumers’ purchase options (Lapersonne,
Laurent, and LeGoff 1995; Moorthy, Ratchford, and Talukdar 1997; Ratchford and Srinivasan
1993). In the similar vein, Dickson and Sawyer (1990, p. 47) find that “slightly more than half
(57%) of the shoppers claimed to have checked the price of the chosen item and fewer than one
in four (21.6%) reported checking the price of an alternative brand … only 13.9% said they
checked prices to remember until next time and 12.8% checked the price to compare prices
between supermarkets.” In addition, Murthi and Srinivasan (1999) show that inclusion of
consumers’ partial evaluation behavior across alternatives increased the predictive power of a
model compared with other models without it. Vanhuele and Dreze (2002) argue that consumers
may actively search price information or incidentally notice price information.
7

By leveraging consumers’ limited price search behavior, which leads to limited price
knowledge, firms can charge higher price. However, marketers should be aware that even a
moderate level of knowledge can induce consumers to actively pursue available information (i.e.,
the concave relationship between knowledge and the extent of consumer price search; Moorthy,
Ratchford, and Talukdar 1997). Cherry pickers show no significant differences from those who
do not conduct within-store search (Vanhuele and Dreze 2002). Overall, the amount of past price
knowledge and the mental calculation of gains and losses from additional search may explain
consumers’ limited search behavior (Carlson and McAfee 1983; Dahlby and West 1986). In the
next two sections, I will briefly discuss representation and judgment of a price as well as the
feeling of familiarity.
Price Cognition and Representation of Price Information
Consumers’ price perception is an indispensable factor in their purchase decision making (e.g.,
Monroe 2003). During searching and processing price information, consumers judge a focal
price by comparing it to the latitude of their internal reference price as well as the external
reference prices (i.e., prices of brands in a product category). Such a process of price search and
judgment is subjective and relativistic (e.g., DelVecchio, Krishnan, and Smith 2007; Janiszewski
and Lichtenstein 1999; Kalyanaram and Little 1994). Price cognition, by definition, is the
cognitive processes consumers undergo when they make price magnitude judgments, and price
cognition consists of rule-based processes and instinctive-associative processes (i.e., heuristics)
(Thomas and Morwtiz 2009). Consumer responses to price information are also determined by
online arithmetic computation and by stored information in their memories (i.e., associative
knowledge structure) (e.g., Ashcraft 1995). When consumers are repetitively exposed to a similar
numeric, their responses tend to get faster and they perceive similar questions as easier the next
8

time. Price cognition consists of multiple psychological tasks that occur simultaneously or
sequentially in one’s mind: the retrieval of price information stored from previous events, the
pursuit of price information from a current event, and the arithmetic comparison (i.e., subjective
calculation) of prices. Consumes who fail to retrieve previously stored price information do not
have accessible information in their memories, or are not motivated to retrieve it likely compare
a target price and prices of alternatives at the point of purchase.
The retrieval of price information is influenced by how it was encoded and stored in
memory. During a previous shopping event, consumers might encode price information as
symbolic representation or analog representation (see Markman 1999). With analog
representation, a numeric value serves as a subjective magnitude on a mental numerical
continuum, similar to representations of other psychophysical stimuli (e.g., light, size) (Thomas
and Morwitz 2009). Symbolic representation is conceptually similar to a visual Arabic code of a
triple-code model. Under a triple-code model (Dehaene 1992; Dehaene and Akhavein 1995),
numeric values are coded in memory as three different forms—namely, auditory verbal code,
visual Arabic code, and analogue magnitude code. For example, if a price is encoded as auditory
verbal code, $7.15 will be represented and stored as “seven dollars and fifteen cents” in one’s
mind. If visual Arabic coding is used, one processes and stores the price as “$7.15.” Finally, with
analogue magnitude code, the price is captured as an approximate magnitude located between
“$7 and $8 and closer to $7” on one’s subjective numeric continuum.
In line with Thomas and Morwitz’s (2009) discussion, a processed numeric symbol (e.g.,
$7.15) can be converted into an analog magnitude code (e.g., between $7.00 and $7.50) quite
automatically and simultaneously at the encoding stage (Vanhuele and Dreze 2002). This coded
price can be compared not only with the latitude of consumers’ internal reference price, which

9

can be retrieved from long-term memory, but also with external reference prices at the point of
purchase. Such comparisons stem from online calculations of numeric symbols and/or holistic
comparisons of subjective numeric locations on a numeric continuum in one’s mind.
The Feeling of Familiarity

When consumers confront a shopping event that is similar to a previously experienced event,
they unconsciously or consciously compare them at both the item level (e.g., product, price) and
the holistic context level (e.g., overall shopping experience). This section presents a literature
review of the psychological processes in play for retrieval or reconstruction of a previously
experienced event. In general, recognition includes two types of processes: feeling of familiarity
and recollection. In line with the traditional dual-process perspective of recognition memory, the
feeling of familiarity does not require intentionality, consciousness, or consumption of conscious
resources; in contrast, recollection requires people to be effortful in search and retrieval of
information in memory and to pay attention to the process. Moreover, an unconscious feeling of
familiarity triggers people’s deliberate effort in recollection (Leboe and Whittlesea 2002).
Overall, the feeling of familiarity is an automatic process that is not consciously controlled, while
recollection is intentional and controllable (e.g., Atkinson and Juola 1973; Jacoby 1991, 1994,
1998; Mandler 1979, 1980, 1991).
Overall, prior research agrees that the feeling of familiarity derives from an unconscious
inferential process (e.g., Jacoby and Dallas 1981; Jacoby and Whitehouse 1989; Lindsay and
Kelley 1996; Rajaram 1993; Whittlesea 1993). In addition, the feeling of familiarity with an item
is strengthened by the experience with the same item on a previous occasion (e.g., Jacoby and
Dallas 1981). However, studies provide evidence that recollection can also be an unconscious
inferential process (Bartlett 1932; Loftus, Miller, and Burns 1978; Loftus and Palmer 1974;
10

Roediger and McDermott 1995, 1996). Thus, recollection can be considered the mental
reconstruction of a prior event based on the information that comes to mind. The accuracy of the
reconstruction is determined by the extent to which the person encoded details of a prior event. If
such objective and detailed information is not available, recollection can also emerge through an
inferential process that assesses a piece of the information that comes to mind with respect to its
clarity, completeness, fluency, and relationship to other event-related information. By definition,
then, recollection is subject to errors (i.e., illusions of recall; Leboe and Whittlesea 2002).
As discussed previously, although the dual process treats the feeling of familiarity as a heuristicsbased reconstruction and recollection as a controlled retrieval, it is argued that the differentiation
of the two processes as automatic versus controlled merely depicts the extensiveness of each
process (Leboe and Whittlesea 2002). Leboe and Whittlesea (2002, p. 823) state that a primary
difference in the two processes is “the subjective conviction that current processing reflects an
influence of the past.” Thus, a significant factor influencing consumers’ subjective recognition
experience is their unconscious assessment of information appearing to their minds in terms of its
quality and consistency. In addition recollection is more likely to require conscious resources than
the feeling of familiarity, and the feeling of familiarity may occur with the intention to recall (e.g.,
Leboe and Whittlesea 2002).
In conclusion, rational assumption on consumers’ decision making process is history.
Price knowledge developed over time and shopping behavior (e.g., price search) of consumers
influences their purchase quantity decision when they face a sales promotion. This dissertation
intends to extent our knowledge by designing and implementing four experiments which
investigate how the relationships between discount types and stockpiling behavior vary across
different product types and situations. Also such experiments examine if price knowledge and

11

perceived similarity of shopping experiences are mediating mechanisms of the relationships
between a discount and stockpiling.

12

STUDY 1
Study 1 is to establish the baseline effects of a sales promotion on stockpiling behavior. One of
primary effects of interest is to investigate how discount depth and frequency influence
stockpiling behavior. Another main goal of Study 1 is to introduce initial mediating mechanisms.
That is, it is proposed that consumer perceived similarity of the overall shopping experience,
price recall error, and price comparison error are mediators of the relationships between a sales
promotion and stockpiling behavior. In addition, if the effects of discount depth and frequency on
price recall error (PRE) and price comparison error (PCE), and perceived similarity are tested.
Hypothesis Development
The Impact of Discount Depth. The assumption of consumers’ perfect rationality in their purchase
decision making is unrealistic (e.g., Dickson and Sawyer 1990). In certain cases, due to limited
price knowledge, consumers may more rely on prices at the moment rather than ones developed
from past prices they paid in order to determine purchase, brand choice, and quantity. Provided
that consumers’ preferred brand is promoted, they may exert more efforts to process a current
discounted price, to compare it to an expected price (i.e., internal reference price), and to update
their internal price accordingly (i.e., contrast effect). Such intraitem price search behavior would
result in lower PRE (e.g., Vanhuele and Dreze 2002). In addition, they may compare the promoted
price to prices of alternative brands in order to evaluate promoted brand’s price position in a
category and support their decision. However, it is also possible that the attractiveness of a sales
promotion available for their preferred brand demotivates such a rational price search behavior,
that is to say higher PRE and PCE. In other words, a discount itself may make both inter- and intraitem price search unnecessary tasks for consumers (e.g., Dickson and Sawyer 1990). The negative
effect of discount depth on consumers’ price knowledge accuracy is to be intensified when a
13

discount rate gets deeper.
Of the two types of familiarity which are objective familiarity and perceived familiarity,
this study builds on self-assessed (perceived) familiarity (e.g., Lichtenstein and Fishhoff 1977),
which, by definition, is subject to systematic biases and heuristics in consumer decision making
(Park and Lessig 1981). The underlying process for the feeling of familiarity is an inferential and
constructive process through which consumers undergo (e.g., Marcel 1983). Thus, consumers in
the current shopping context consciously or unconsciously integrate various aspects from this
context and evaluate the degree of similarity to such aspects from the previous shopping
experience. Fundamentally, consumers’ recognition of a discount for their preferred brand would
interfere their familiarity judgment. The feeling of dissimilarity will be stronger with an increasing
discount depth.
When a price discount is provided on a stock-up product, consumers would take
advantage of it by stockpiling items for future consumption and by delaying the time of their
subsequent purchase (e.g., Litvack, Calantone, Warshaw 1985). In other words, consumers will
be willing to increase their short-term costs in order to reduce long-term costs. Overall the
positive relationship between discount depth and purchase quantity is supported (Bell, Chiang,
and Padmannabhan 1999).
H1 :

The depth of a price discount has a direct, positive effect on price recall error
(H1a) and price comparison error (H1b) to the extent that deeper discounts are
associated with higher error.

H2 :

The depth of a price discount has a direct, negative effect on perceived similarity
of the overall shopping experience.

H3 :

The depth of a price discount has a direct, positive effect on consumer stockpiling.

14

The Impact of Discount Frequency. If there is a consistency regarding prices paid over time,
consumers’ expectation of future price would be more accurate, price knowledge built on past
prices paid over time. At the same time, their feeling of confidence with own price knowledge
would increase as well. For a given product category, if consumers have a short purchase cycle
(i.e., their purchase frequency in a fixed time period), a price stored in their memory from a past
purchase occasion will be more accurate and up-to-date. Also their purchase behavior would be a
more habitual or routine-based task. This is because price sought, processed and stored in their
previous purchase occasion is easy to access due to a small time gap between purchase occasions.
The feeling of familiarity depends on predictiveness, uncertainty, the match between
expected information and current information, and the perceived surprise in the current shopping
event (Whittlesea and Williams 2001). Notice of any changes in the current shopping occasion
would serve as perceived surprise which precedes consumers’ judgment of the feeling of
familiarity. While shopping, various stimuli can evoke consumers’ feeling of familiarity with their
shopping experience (e.g., Koriat and Levy-Sadot 2001; Koriat and Lieblich 1977; Marcel 1983).
Frequent exposure to the same (or very similar) stimuli over time positively influences consumers’
expectations and predictiveness but negatively influences their uncertainty in the current shopping
occasion. When consumers face with an unexpected or unusual stimulus (e.g., an unusually deep
discount), they might disassociate the current shopping experience with the previous one, resulting
in unmet expectation and increased uncertainty about the current occasion (Whittlesea and
Williams 2001). Increasing frequency of exposure to a particular stimulus increases the feeling of
familiarity with that particular stimulus (e.g., Nhouyvanisvong and Reder 1998; Reder and Ritter
1992; Reder and Schunn 1996; Scunn et al. 1997). If consumers frequently purchase their preferred
brand with a discount over time, their perceived frequency of discount availability will increase

15

(Krishna 1991). In turn, consumers’ feeling of familiarity regarding their shopping experience will
increase.
Consumers’ knowledge about the pattern of sales promotions is highlighted as an
important drive of stockpiling (e.g., Krishna 1994). If consumers adapt to increasing sales
promotion frequency, the effect of a sales promotion would wear-out over time and/or consumers
may not see a discounted price as attractive as before (e.g., Foekens, Leeflang, and Wittink 1999;
Kopalle, Mela, and Marsh 1999). Thus consumers will wait for the next sales promotion for
future purchase and, on a given purchase occasion, stock up the promoted brand to the necessary
inventory level where consumers can consume till a subsequent deal is provided. The lying-inwait heuristic results from consumes’ certainty and expectation about when the next sales
promotion would be available (Krishna 1994). Based on behavioral learning perspective (e.g.,
Blattberg and Neslin 1990), it could be posited that the increasing opportunity for consumers to
have a promotional benefit allows them to learn to stockpile the promoted item consistently
(Ailawadi, Harlam, Cesar, and Trounce 2006). However, consumers may not more tend to
stockpile the promoted brand with increasing promotion frequency (e.g., Ailawadi, Harlam,
Cesar, and Trounce 2006). That is, consumers’ risk averse propensity combined with their higher
uncertainty about when the next sales promotion would be available may encourage them to
stockpile the promoted brand. Products with higher regular price and a random discount chance
are more likely to be stockpiled by consumers when they are promoted (e.g., Meyer and
Assuncao 1990). Consumers may feel an obligation to make a purchase, to stock up the
promoted item, and to carry additional items to a subsequent purchase occasion in order to
minimize future costs (e.g., Meyer and Assuncao 1990; Krishna 1994). This would be intensified
when their preferred brand is discounted.

16

H4 :

The frequency of a price discount has a direct, negative effect on price recall error
(H4a) and price comparison error (H4b) to the extent that less frequent discounts
are associated with less error.

H5 :

The frequency of a price discount has a direct, positive effect on perceived
similarity of the overall shopping experience.

H6 :

The frequency of a price discount has a direct, negative effect on consumer
stockpiling.

The Mediation Effects of PRE, PCE, and Perceived Familiarity of Overall Shopping Experience.
For stock-up products, consumers’ purchase quantity would be more determined by price they
know than one they remember (e.g., Monroe 2003; Monroe and Lee 1999). Remembering prices
require consumers to consume conscious resources in order to retrieve or recollect specific price
points, but knowing prices is more associated with a sense of familiarity (or the feeling of
familiarity) which does not require the consumption of conscious resources (e.g., Gardiner 1988;
Rajaram 1993; Tulving 1985). In other words, while knowing is an instinctive and fast task,
recollecting specific prices (i.e., past prices of the promoted brand) necessitate (1) the symbolic
representation of prices, (2) the deliberative and slow cognitive process, and (3) consumption of
conscious resources.
In addition, consumers’ perceived familiarity with shopping experience is suggested as
another factor which mediates the relationship between types of discount and stockpiling behaviors.
Park ad Lessig (1981) finds a positive monotonic relationship between the feeling of familiarity
and the confidence with a decision made. Thus, increasing chances to purchase consumers’
preferred brand at a promoted price may increase their perceived familiarity across purchase
occasions and, in turn, increasing perceived confidence with their choice and quantity will

17

motivate consumers’ stockpiling behavior. This is also associated with the concept of behavioral
learning (e.g., Blattberg and Neslin 1990). However, it should be also realized that the unexpected
surprise (e.g., unexpected deeper discount) may interfere consumers’ feeling of familiarity (i.e.,
contrast effect).
Purchase decision making and price magnitude judgments of consumes with a long
purchase cycle may be determined more by inferences to their perceived similarity to an item itself
and/or a context (i.e., price, product, and overall shopping experience). That is, because these
consumers’ price knowledge gets less accurate and more uncertain over time and because they
have some degree of expectations of future price, they will infer previously paid prices and
compare them with the current prices. Such an inference and reconstruction process is based on
the perceived similarity of various aspects such as current price, brand assortment, shopping
purpose, and so on. In turn, stockpiling behavior can be influenced by consumers’ perceived
familiarity with shopping experiences.
If consumers remember things about the previous shopping event (e.g., brand choice,
paid price, shelf location) that match the current event (i.e., affirmation of definite expectations;
Whittlesea and Williams 2001) and if such expectations are violated with a surprise (e.g.,
unusually deep discount, product size reduction) (i.e., violation of expectation; Whittlesea and
Williams 2001). If consumers attribute the produced information to the previous shopping event
and the feeling of familiarity is strong, they might not actively search but rather depend on price
information in the current shopping event (i.e., higher probability of price recall error and price
comparison error).

18

H7 :

The effects of the depth of a discount on stockpiling are mediated by price recall
error (H7a), price comparison error (H7b), and perceived similarity of the
experience (H7c).

H8 :

The effects of the frequency of a discount on stockpiling are mediated by price
recall error (H8a), price comparison error (H8b), and perceived similarity of the
experience (H8c).

Research Design

Study1 employs 2 (high vs. low discount depth) x 2 (high vs. low discount frequency) betweensubject design that serves as the base for other three experiments. Total 495 participants were
randomly assigned to one of four conditions. The experiment of Study1 consists of two
consecutive shopping tasks (i.e., computer simulated shopping events). Each task includes
written information (i.e., shopping scenario) and price tags of purchase options. Each price tag
displays regular price, discounted price, dollar amount to be saved from a given discount, and
unit price and package size. Following sections discuss (1) manipulations of discount depth and
frequency, (2) experiment procedure, (3) measures, (4) sample/data collection, and (5) means of
data analysis before discussing results and findings of Study1.
Manipulation and Product Category. The primary interest of my dissertation is how price
discounts determined by depth and frequency influence consumer stockpiling behavior and what
is an underlying psychological process of such relationships, in particular for low-involvement
products. Thus, a product category used in this experiment should be one that consumers
commonly and/or regularly shop for (e.g., consumer packaged goods). Also the formation of
internal reference price standards is more likely to arise with the presence of repurchase intention

19

in the near future (Mazumdarm, Raj, and Sinha 2005). Previous research investigating effects of
sales promotion and consumer stockpiling behavior classified products into two groups. The one
category was identified by consumer stockpiling and flexible consumption behaviors (e.g., salted
snacks). Consumers for the other category only showed stockpiling behavior (e.g., coffee) (Bell,
Chiang, and Padmanabhan 1999; Bell, Iyer, and Padmanabhan 2002). Taking all together,
ground coffee was chosen as the focal product category for the current experiment and 12oz
package size was used.
Four between-subject conditions were determined by manipulations of high vs. low
discount depth as well as high vs. low discount frequency. Levels of manipulations were
determined based on pre-studies and previous literature. Any price discounts below are not to be
easily recognizable (Litvack, Calantone, and Warshaw 1985). Kalwani and Yim (1992) found a
significant difference in the level of expected price between high and low discount depth groups
(i.e., 20% vs. 30%; 10% vs.20%; and 30% vs. 40%). DelVecchio, Krishnan, and Smith (2007)
found that participants perceive 43% (vs. 13%), 40% (vs.15%), and 50% (vs. 10%) discounts
higher than their corresponded lower discount rates. For the product category with consumer
stockpiling behavior and flexible consumption behavior, the average discount rate provided was
36% and the other one where consumers showed stockpiling behavior only, the average discount
rate was 26% (Bell, Iyer, and Padmanabhan 2002). Study1 chose 10% and 40% as low and high
level discounts and they were checked through two pre-studies. Results indicated that
participants perceived 40% discount rate significantly higher than 10% discount. The regular
price for a focal product (i.e., participant’s preferred brand) was set at $7.15 in the current
experiment. Thus, 10% discounted price is $6.44 and 40% discounted price is $4.29.

20

Furthermore, an alternative brand option is a hypothetical brand, Café Value. Its price was set at
$4.78 that is between $6.44 and $4.29.
One’s perception on discount frequency over time was more accurate when the
regularity was employed than when not (Krishna 1991). In the same research, discount frequency
was manipulated by providing three discounts for low discount frequency vs. six discounts for
high discount frequency over a 12-week period regardless of regularity. Here discount frequency
was manipulated by a combination of written information given in each shopping scenario and
the number of discounts available across two shopping tasks. Participants in low discount
frequency group had only one discount which was available at the second shopping and those
who in high discount frequency group had total four discounts sequentially (i.e., two discounts
from the written information and two discounts provided at all of two shopping tasks).
Procedure of Shopping Tasks. In order to complete the experiment, participants conducted two
consecutive shopping tasks. For each task, they were required to answer which brand they chose
to purchase and how many items they decided to purchase. Prior to these main shopping tasks,
participants answered an array of questions. In particular, three questions were used to complete
shopping scenarios. The first question was about participant’s preferred brand name of ground
coffee that one purchases most often. Such brand name (i.e., the focal brand) was presented in
shopping tasks as one of choice options and also was manipulated in terms of discount depth and
frequency. The second question was about one’s current product inventory level at home. The
current level of product inventory would have a significant effect on one’s purchase quantity
decision. Thus, of the two shopping tasks, the current inventory information was embedded in
the first shopping scenario and one could defer the first purchase. The last question was about the
level of product inventory where one usually decides to purchase to refill one’s inventory. This

21

information was piped in the second shopping scenario. In the second shopping task, one cannot
defer one’s purchase. In addition, by asking one to imagine s/he is shopping a ground coffee in
retail store where they often shop for grocery products, the potential influence from the store
location on one’s decision making was to be controlled. As the last step, questions for predicting
variables and covariates were asked and demographic information were collected. Appendices
(i.e., Appendix A – D) present the complete scenarios of the experiment.
Measures. This section only discusses sources of measurement items. The means of assessment
of the measures and evidences of reliabilities and validities are discussed in later sections.
Stockpiling. This is measured by purchase quantity in the second shopping task. If a participant
chose to purchase one of alternative options, his or her purchase quantity is recoded to zero.
Price recall error (PRE) represents represents consumer’s absolute price knowledge and
is adapted from the measure Zeithaml (1982) used to examine consumers’ cognitive response to
price information. It is operationalized as the absolute difference between actual price and
recalled price. Unlike PRE, price comparison error (PCE) reflects consumer’s relative price
knowledge and is adopted from Zeithaml (1982). It is operationalized as the number of incorrect
rankings made by participants. Similarity which is participants’ perceived similarity between
shopping tasks is measured with regard to overall shopping experience. Participants used a
seven-point scale (1 = “very different,” 7= “very similar”) to respond to the following statement:
“Please indicate how similar your current shopping trip you just made and your previous
shopping trip is, in terms of the following….” Category knowledge, participants’ product
category knowledge, is operationalized by adapting two measures of subjective knowledge from
Brucks (1985). To form a single scale, the average score of two measures is used. To measure
deal proneness, six measurement items are adopted from Lichtenstein, Ridgway, and Netemeyer

22

(1993). According to Murthi and Rao (2012), deal proneness is one of the reasons for consumer
price unawareness. All six items are measured on a seven-point Likert scale, and an average
score is used for hypotheses testing. Certainty of recalled price is measured by single
measurement item which is adapted from the measure of consumers’ affective (i.e., attitude)
response to price information in Zeithaml’s (1982) conceptual model. It asks a participant to
answer the degree of certainty about one’s recalled price. For purchaser variable, before a
participant conducts two shopping tasks, one was asked to identify oneself as a buyer or a nonbuyer. Lastly, as demographic variables, household size and household income are collected as
covariates. In particular, household size is operationalized as the number of people in a
household
Data Collection and Samples. Data were collected through Amazon Mechanical Turk (MTurk),
which is based on Internet crowdsourcing and is a popular platform to find participants. Various
researchers have used MTurk as a source to select participants from the general population and
have noted its usefulness and reliability (e.g., Buhrmester, Kwang, and Gosling 2011; Goodman,
Cryder, and Cheema 2013; Paolacci, Chandler, and Ipeirotis 2010; Rand 2011). In their research
noting differences between participants from MTurk and those from other sources (e.g.,
community members, students), Goodman, Cryder, and Cheema (2013) suggests that researchers
use screening/filtering measures and also consider demographic properness of participants for
their studies. Thus, for this study, data were collected in such a way that only those who met two
preset conditions were able to participate in one of experiment conditions: (1) incentive approval
ratio (i.e., >89%) and (2) country in which they reside (i.e., United States). Some additional
screening questions were asked prior to experimental tasks. Last, in exchange for one’s

23

participation, a monetary compensation was granted. For the current experiment, cell sizes are in
the range of 88 to 139. Total sample size is 495.
Means of Analysis. Overall, data analysis was a three-stage process. First, reliabilities and
validities of measures were assessed. Second, main effects of discount depth and frequency on
purchase quantity, PRE, PCE, and similarity were tested along with potential interaction effects.
Finally, mediation effects of similarity, PRE, and PCE were examined with counting covariates
into the proposed model in Figure 1.
Figure 1: Proposed Conceptual Model

Price Recall Error
Discount
Depth

Price Comparison
Error

Stockpiling

Discount
Frequency
Perceived Similarity:
Overall Shopping

Benefit Change
New Product Introduction
New Brand Introduction
The first is the assessment of the measures. For single item measures, the existence of
overlapping confidence intervals was checked, That is, a variable should not have correlations
which overlapping 1 with other separate variables. For two multi-item measures, category
knowledge and deal proneness, a Confirmatory Factor Analysis (CFA) was used. To check
reliabilities and validities of the measures, Cronbach’s alpha, average variance extracted (AVE),

24

and composite reliability (CR) were calculated (Fornell and Larcker 1981). To form a single
scale for those two multi-item measures, the average score in indicators is used for hypotheses
testing.
Secondly, Multivariate Analysis of Variance (MANOVA) was used to test Hypotheses 1
– 6 associated with main effects of discount depth and frequency on PRE, PCE, and similarity.
The primary focus is on between-subject tests and comparisons. First of all, overall Wilk’s
Lambdas were checked as evidence of multivariate test. If the multivariate tests shows statistical
significance, univariate tests for each mediator and purchase quantity were followed. Here
Bonferroni adjustments were made.
Lastly, PROCESS macro in SPSS (Hayes and Preacher 2014) was used to test mediation
effects, H7 – 8. In particular, for Study 1, Model 4 template was applied. This approach allowed
a direct effect of single predicting variable on a dependent variable and its multiple indirect
effects (i.e., multiple mediators) to be examined simultaneously. The magnitude and significance
of the indirect effects were assessed by looking at upper and lower bounds for the confidence
intervals surrounding each indirect effect.
Results

Manipulation Checks. Indicated average discount rates between high discount depth group
(33.81%) and low discount depth group (16.97) were significantly different (F1,493=142.718,
p<.01). The difference between average perceptions on a given discount rate between two groups
was also significant (F1,493=124.044, p<.01). Between high and low discount frequency groups,
their perceived number of discounts was significantly different (1.11 vs. 1.88) (F1,493=138.239,
p<.01). Frequency perception was significantly different between two groups (F1,493=274.418,
p<.01).
25

Table 1-1: Correlation Matrix
Knowledge Deal

Quantity

PRE

PCE

Certainty

Similarity Household Household
Size
Income

Knowledge
Deal

.796
-.012

Lower 95% CI
Upper 95% CI

-.100
.076

Quantity

.046

-.028

Lower 95% CI
Upper 95% CI

-.042
.134

-.116
.060

.097*

-.135**

-.040

Lower 95% CI
Upper 95% CI

.009
.184

-.221
-.047

-.128
.048

PCE

.014

-.090*

-.027

.032

Lower 95% CI
Upper 95% CI

-.074
.102

-.177
-.002

-.115
.061

-.056
.120

Certainty

.070

.047

.142**

Lower 95% CI
Upper 95% CI

-.018
.157

-.041
.135

.055
.227

-.322
-.156

-.093
.083

Similarity

.035

.060

-.026

-.023

-.050

.119**

Lower 95% CI
Upper 95% CI

-.053
.123

-.028
.147

-.114
.062

-.111
.065

-.138
.038

.031
.205

Household
Size

.066

-.033

.090*

.023

-.026

.087

.019

Lower 95% CI
Upper 95% CI

-.022
.153

-.121
.055

.002
.177

-.065
.111

-.114
.062

-.001
.174

-.069
.107

Household
Income

.016

-.154**

.140**

-.031

-.020

.076

.046

.288**

Lower 95% CI
Upper 95% CI

-.073
.104

-.239
-.067

.053
.225

-.119
.057

-.108
.068

-.012
.163

-.042
.134

.205
.367

PRE

.0002
.536
.
.
.

-.241** -.005

.
.
.

.

**. Correlation is significant at the 0.01 level (2-tailed).
*. Correlation is significant at the 0.05 level (2-tailed).
Notes: Correlations are in the lower triangle. Shared variances are in the upper triangle. AVE is on the diagonal.

26

Assessment of the Measures. As shown in Table1-1, single item measures did not show any
correlations overlapping 1 for separate variables. Table1-2 presents a measurement model
estimated by CFA. This model composed of two reflective multi-item latent constructs (i.e.,
category knowledge and deal proneness). Fit indices indicated a good-fit between the proposed
model and the data (CFI=.976; SRMR=.039; RMSEA=.069) (Hu and Bentler, 1999). Evidence of
reliabilities and validities was found on category knowledge (AVE=.7964, CR=.887, Cronbach’s
alpha=.868) and on deal proneness (AVE=.5369, CR=.87, Cronbach’s alpha=.888). AVE was
greater than the shared variance (Table1-1). In addition, Table1-4 has the descriptive information
of measures.
Table 1-2: Measurement Model
λ
Category Knowledge (CR=.887; AVE=.7964)
 Rate your overall knowledge of “ground coffee”, as compared to the
average consumer.
 Please check the scale that best describes your familiarity with
“ground coffee”.
Deal Proneness (CR=.87; AVE=.5369)
 If a product is on sale, that can be a reason for me to buy it.
 When I buy a brand that is on sale, I feel that I am getting a good
deal.
 I have favorite brand, but most of time I buy the brand that is on
sale.
 One should try to buy the brand that is on sale.
 I am more likely to buy brands that are on sale.
 Compared to most people, I am more likely to buy brands that are on
sale.
Goodness-of-Fit Indexes
CFI=.976; SRMR=.039; RMSEA=.069

27

.89
.89

.55
.61
.70
.73
.87
.87

Table 1-3: Descriptive Statistics
DEP
Quantity Low
High
PRE
Low
High
PCE
Low
High
Similarity Low
High
FRQ
Quantity Low
High
PRE
Low
High
PCE
Low
High
Similarity Low
High
DEP
Quantity Low
High
PRE

Low
High

PCE

Low
High

Similarity Low
High

Mean
1.548
2.127
44.638
56.728
1.243
0.903
5.106
5.123

SD
0.079
0.071
6.778
6.078
0.06
0.054
0.091
0.082

95% CI
1.392 1.704
1.987 2.266
31.321 57.955
44.785 68.67
1.125 1.361
0.797 1.009
4.927 5.285
4.962 5.284

2.007
0.07
1.869 2.145
1.668
0.08
1.511 1.824
52.354 6.023 40.519 64.189
49.012 6.827 35.599 62.425
1.045 0.053
0.94
1.15
1.101
0.06
0.982
1.22
4.914 0.081 4.755 5.073
5.315 0.092 5.135 5.495
FRQ
Mean
SD
95% CI
Low
1.619 0.099 1.425 1.812
High
1.477 0.124 1.234 1.721
Low
2.396
0.1
2.198 2.593
High
1.858
0.1
1.661 2.055
Low
44.64
8.44 28.057 61.223
High 44.636 10.607 23.795 65.478
Low 60.067 8.596 43.178 76.957
High 53.388 8.596 36.499 70.278
Low
1.201 0.075 1.055 1.348
High
1.284 0.094 1.099 1.469
Low
0.888 0.076 0.738 1.038
High
0.918 0.076 0.768 1.068
Low
5.007 0.114 4.784
5.23
High
5.205 0.143 4.924 5.485
Low
4.821 0.116 4.594 5.048
High
5.425 0.116 5.198 5.653

28

Table 1-4: Descriptive Information of Measures
N Mini Max
Mean
SD
Quantity
495 1.00
6.00
1.869 1.209
PRE
495 0.00 570.00 51.184 99.417
PCE
495 0.00
2.00 1.0545 .895
Similarity
495
1
7
5.11
1.355
Certainty
495
1
7
4.21
1.513
Household Size
495
1
5
2.63
1.280
Household Income 492
1
14
6.30
3.631
Knowledge
495 1.00
7.00
4.211 1.459
Deal Proneness
495 1.00
7.00
4.740 1.134
Valid N
492

Assessment of the Main Effects. The main effect of discount depth on mediators and the main
predicted variable was significant (Wilks’λ=.905, p<.01). Also that of discount frequency
confirmed its significance (Wilks’λ=.957, p<.01. Given the fact that overall multivariate tests of
discount depth and frequency are significant (Table1-5), univariate tests of each mediator and a
predicted variable (quantity) are discussed.
Table 1-5: Multivariate Tests
Wilks’ λ
F4,488
Sig. Partial Eta Squared Powerc
b
Intercept
.050
2305.699 .000
.950
1.00
b
DEP
.905
12.872
.000
.095
1.00
FRQ
.957
5.479b
.000
.043
.976
b
DEP*FRQ
.987
1.605
.172
.013
.496
a. Design: Intercept + DEP + FRQ + DEP * FRQ
b. Exact statistics
c. Computed using alpha = .05

Table1-6 and Figure2 present the significance of each test and directions of relationships.
(also see Table1-3). Discount depth showed its positive association with purchase quantity
(F1,491=29.642, p<.01). In contrast, discount frequency revealed its negative relationship with
purchase quantity (F1,491=10.188, p<.01). Thus H3 and H6 are supported. In terms of PCE, an
increase in discount depth decreased PCE (F1,491=17.743, p<.01). This may tell that when

29

consumers’ preferred brand is more deeply discounted than the price they paid (i.e., internal price
reference) in their previous shopping, they might be more likely to compare the price to those of
other brands in a product category. The relationship between discount frequency and similarity
was significant and positive (F1,491=10.716, p<.01).
Table 1-6: Univariate Tests
DEP
Quantity

Contrast
Error
PRE
Contrast
Error
PCE
Contrast
Error
Similarity Contrast
Error
FRQ
Quantity

Contrast
Error
PRE
Contrast
Error
PCE
Contrast
Error
Similarity Contrast
Error

MS
F1,491 Sig. Partial Eta Squared Powera
40.032
29.642 .000
.057
1.00
1.350
17459.574 1.763 .185
.004
.263
9901.512
13.792
17.743 .000
.035
.988
.777
.036
.020 .888
.000
.052
1.792
MS
13.759
1.350
1333.887
9901.512
.378
.777
19.202
1.792

F1,491 Sig. Partial Eta Squared Powera
10.188 .002
.020
.890
.135

.714

.000

.066

.486

.486

.001

.107

10.716 .001

.021

.904

30

Figure 2: Mean Plots for Discount DEP and FRQ
Quantity

PRE

Similarity

PCE

Assessment of the Mediation Effects. Table1-7 presents results of mediation tests for PRE, PCE,
and similarity. Discount depth showed its positive and significant direct effect on quantity
(t=5.068, p<.01). Discount frequency indicated its negative and significant direct effect on
quantity (t=-2.889, p<.01). However, all of proposed mediators did not have significant indirect
effects. 95% confidence intervals (CI) of each indirect effect when it was tested with discount
depth are (1) PRE: -0.058 ~ 0.027, (2) PCE: -0.048 ~ 0.030, and (3) similarity: -0.022 ~ 0.015.
95% CIs of indirect effects which were tested with discount frequency are (1) PRE: -.007
~ .014), (2) PCE: -.021 ~ .008), and (3) similarity (-0.022 ~ 0.015).

31

Table 1-7: Test of Mediation Effects
DIRECT AND INDIRECT EFFECTS
Direct effect of DEP on Quantity
Effect
SE
t
p
LLCI
.549
.108
5.068
.000
.336
Indirect effect of DEP on Quantity
Effect
Boot SE BootLLCI BootULCI
TOTAL
-.013
.023
-.058
.027
PRE
-.005
.008
-.030
.004
PCE
-.008
.020
-.048
.030
Similarity
-.0003
.008
-.022
.015
DIRECT AND INDIRECT EFFECTS
Direct effect of FRQ on Quantity
Effect
SE
t
p
LLCI
-.314
.109
-2.889
.004
-.528
Indirect effect of FRQ on Quantity
Effect
Boot SE BootLLCI BootULCI
TOTAL
-.013
.020
-.061
.021
PRE
.001
.005
-.007
.014
PCE
-.001
.006
-.021
.008
Similarity
-.013
.018
-.057
.017

ULCI
.762

ULCI
-.101

Discussion

Study1 examined direct relationships between types of discounts determined by its depth and
frequency and consumer stockpiling behavior (i.e., purchase quantity) as well as indirect
relationships through three mediators (i.e., PRE, PCE, similarity). Table 5 summarizes results of
the hypotheses testing.
In general, primary findings of Study1 are two folds. First, if a firm provides relatively
deeper discount comparing to previously offered discount, consumers may purchase more. As
hypothesized, if a firm make a discount available frequently and regularly, consumers may
decrease their purchase quantity. Second, with regard to customer price search behavior and price

32

knowledge, when customers recognize an available high depth discount, they seems to be more
motivated to search and compare prices at the category level (i.e., interitem search) than usual.
Based on the findings of Study1, we need to extend our understanding to how consumer
stockpiling behavior may vary when another pricing strategic aspect is employed. Thus, Study2
add benefit change into the model of Study1. Benefit change is to be manipulated by changing
product size. It is a pervasive way of changing price in order for firms to stabilize their profit
margin when there is an increase in price of raw materials. Thus, Study2 introduces benefit change
as a moderator and examines how established relationships between types of price discounts and
purchase quantity would be varied by benefit change.

33

STUDY 2
The primary goals of Study 2 are two folds. The first one is to replicate the baseline effects noted
in Study 1 and the second one is to introduce moderating effects of benefit change. In other words,
Study 2 investigates, when product size is changed for consumers’ preferred brand, how the
established relationships in Study 1 change. Because reducing product size means less product
benefit for consumers, consumers may adjust their purchase behavior accordingly.
If a benefit change is recognized, it would make consumers necessary to update their
knowledge because it overall influences the value calculation. Under concepts of behavioral
learning (e.g., Blattberg and Neslin 1990) and feeling of familiarity (e.g., Whittlesea and
Williams 2001), because of consumers’ learned benefits over time through promotional savings,
even though there is a change in product size (i.e., reduced product benefit), consumers may not
reevaluate product value. Even if consumers notice a change, if a given discount is unusual and
deeper, they are less likely to update their price knowledge and more likely to stockpile the
promoted item to maximize benefits. Updating their knowledge by reevaluating product value
may occur under a condition in which a change in stimulus is distinctive enough for consumers
to be encouraged to do so. However, as a recent article addresses, it seldom appears (The New
York Times 2011). This may because firms make a change under the level of just-noticeabledifference of consumers and/or because consumers limited information search behavior.
H9 :

A benefit change moderates the effects of depth/frequency to the extent that the
effects of both discounts are amplified when there is a benefit change.

Research Design

34

The experiment of Study2 employs 2 (high vs. low discount depth) x 2 (high vs. low
discount frequency) x 2 (high vs. low benefit/ between-subject design. A primary
difference of Study2 from Study1 is the addition of benefit change (i.e., change in product
size) into the base model tested in Study1. Study2 retains manipulations of discount depth
and frequency from Study1.
Benefit (i.e., product size) was operationalized by changing price information in
price tag. After a 9% reduction in product size (i.e., low benefit group), unit price increased
from 60 cents per oz. to 65 cents per oz. and product size decreased from 12 oz. to 11 oz.
For both numerical information, the left digits were kept the same. However, with a 33%
reduction (i.e., high benefit group), product size decreased from 12 oz. to 8 oz. The twodigit number decreased to a single digit; the single digit number, 8, also was greater than
each digit of the base size (i.e., 1 and 2). Product unit price increased from 60 cents to 85
cents and involved a change in the left digit. Of the two shopping tasks, benefit
manipulation is introduced in the second shopping task. Study2 kept other experimental
aspects exactly same with those of Study2 including procedure, scenarios, data collection
platform, and all of the measures. Lastly, for Study2, cell sizes are in the range of 72 to 174.
Total sample size is 625.
In terms of means of data analysis, hypotheses testing was conducted by following the
same process of Study1 with the addition of the moderating effect of benefit change. So, in order
to test the moderated mediation, a new model template in PROCESS macro was used (i.e., model
8 for moderated mediation).
Results

35

Manipulation Checks. Indicated average discount rates between high discount group (32.83%) and
low discount depth group (18.34%) were significantly different (F1,623=146.769, p<.01). The
difference between average perceptions on a given discount rate between two groups were also
significant (F1,623=134.782, p<.01). Indicated number of discounts showed a significant difference
between low discount frequency group (1.08) and high discount frequency group (1.96)
(F1,623=225.710, p<.01). Also participants’ frequency perception between two groups was
significantly different (F1,623=418.388, p<.01). Checking the manipulation of benefit change was
two step approach. First, whether participants recognized benefit change or not was asked. Second,
those who recognized a given benefit change were led to questions associated with their objective
and subjective perceptions on the given level of manipulation. Of the participants, 22.2% (i.e., 139
of 625) recognized the change. High benefit group answered there was a decrease in product size
by 3.45oz and low benefit group indicated a decrease by 2.38oz (F1, 137 = 8.178, p<.01). In addition,
with regard to participants’ perception on a given amount of benefit change, there was a significant
difference between two groups (F1,137=2/.618, p<.01).

36

Table 2-1: Correlation Matrix
Knowledge Deal
Knowledge
Deal

.837
-.127**

Lower 95% CI
Upper 95% CI

-.203
-.049

Quantity

PRE

PCE

Certainty

Similarity Household Household
Size
Income

.016
.462

.122**

-.037

Lower 95% CI
Upper 95% CI

.044
.199

-.115
.042

PRE

.026

-.160**

.024

Lower 95% CI
Upper 95% CI

-.053
.104

-.235
-.083

-.055
.102

PCE

.046

-.024

-.039

.072

Lower 95% CI
Upper 95% CI

-.033
.124

-.102
.055

-.117
.040

-.006
.150

Certainty

.032

.058

.102*

-.175**

.031

Lower 95% CI
Upper 95% CI

-.047
.110

-.021
.136

.024
.179

-.250
-.098

-.048
.109

.086*

.064

-.050

-.007

.000

.017

Lower 95% CI
Upper 95% CI

.008
.163

-.015
.142

-.128
.029

-.085
.071

-.078
.078

-.062
.095

Household
Size

-.048

.024

.027

-.027

-.041

.052

.025

Lower 95% CI
Upper 95% CI

-.126
.031

-.055
.102

-.052
.105

-.105
.052

-.119
.038

-.027
.130

-.054
.103

Household
Income

-.012

-.118**

.032

-.041

-.063

.023

-.043

.272**

Lower 95% CI
Upper 95% CI

-.090
.066

-.195
-.040

-.047
.110

-.119
.038

-.141
.016

-.056
.101

-.121
.036

.198
.343

Quantity

Similarity

.
.
.
.
.
.

.

**. Correlation is significant at the 0.01 level (2-tailed).
*. Correlation is significant at the 0.05 level (2-tailed).
Notes: Correlations are in the lower triangle. Shared variances are in the upper triangle. AVE is on the diagonal.

37

Assessment of the Measures. Table 2-1 indicates there are no signle measure which has a
correlation overlapping 1 with other variables. Table 2-2 presents a measurement model estimated
by CFA. It shows a good fit between the model and the data (CFI=.974; SRMR=.036;
RMSEA=.075) (Hu and Bentler, 1999). Acceptable level of reliability and validity were found for
both category knowledge (AVE=.837, CR=.911, Cronbach’s alpha=.901) and deal proneness
(AVE=.462, CR=.835, Cronbach’s alpha=.880). In addition, AVE was greater than its shared
variances. Table 2-3 shows descriptive information of measures.
Table 2-2: Measurement Model
λ
Category Knowledge (CR=.911; AVE=.837)
 Rate your overall knowledge of “ground coffee”, as compared to the
average consumer.
 Please check the scale that best describes your familiarity with
“ground coffee”.
Deal Proneness (CR=.835; AVE=.462)
 If a product is on sale, that can be a reason for me to buy it.
 When I buy a brand that is on sale, I feel that I am getting a good
deal.
 I have favorite brand, but most of time I buy the brand that is on
sale.
 One should try to buy the brand that is on sale.
 I am more likely to buy brands that are on sale.
 Compared to most people, I am more likely to buy brands that are on
sale.
Goodness-of-Fit Indexes
CFI=.974; SRMR=.036; RMSEA=.075

38

1.0
.82

.63
.65
.73
.73
.87
.85

Table 2-3: Descriptive Information of Measures
N Mini Max
Mean
SD
Quantity
625 1.00
6.00
1.810
1.113
PRE
625 .00 570.00 41.4906 80.985
PCE
625 .00
2.00
1.013
.904
Similarity
625
1
7
5.16
1.379
Certainty
625
1
7
4.20
1.542
Household Size
625
1
5
2.59
1.196
Household Income 622
1
14
6.09
3.417
Knowledge
625 1.00
7.00
4.295
1.513
Deal Proneness
625 1.00
7.00
4.755
1.124
Valid N
622

Assessment of the Main Effects. Multivariate tests revealed significant main effects of benefit
change (Wilks’λ=.984, p < .001, discount depth (Wilks’λ=.954, p < .001), and discount depth
(Wilks’λ=.910, p < .001). In addition, a significant interaction between discount depth and
frequency was found (Wilks’λ=.982, p < .005). Table 2-4 presents means, standard deviations, and
sample sizes for all four pricing groups. Given the fact that overall multivariate tests of discount
depth, discount frequency, and an interaction between discount depth and frequency were
significant (Table2-5), univariate tests of such each mediator and a predicted variable as well as
between-subjects tests of an interaction effect on those variables are discussed.
Table 2-6, 2-7 and Figure3 present the significance of each test and directions of
relationships (also see Table 2-4). An increase in benefit change caused a decrease in PCE
(F1,617=8.191, p<.01) which would indicate that those who recognize any changes in product size
will deliberately search and compare price information across products (i.e., interitem search) at
the product category level. Discount depth was positively associated with quantity (F1,617=28.893,
p<.01). Discount frequency showed a significant negative relationship with quantity
(F1,617=30.934, p<.01) and a significant positive relationship with similarity (F1,617=28.125, p<.01).

39

Table 2-4: Descriptive Statistics
SD
95% CI
BEF Mean
Quan Low 1.847 0.059
1.73; 1.964
High 1.788 0.062 1.667; 1.909
PRE Low 44.307 4.533 35.406; 53.208
High 38.423 4.693 29.206; 47.64
PCE Low 0.907 0.05
0.809; 1.006
High 1.114 0.052 1.012; 1.216
Sim
Low 5.199 0.075 5.052; 5.347
High 5.126 0.078 4.973; 5.278
Quan DEP FRQ
Low Low

BEF Mean
SD
Low 1.681 0.112
High 1.807 0.117
High Low 1.471 0.116
High 1.474 0.122
High Low Low 2.487 0.122
High 2.247 0.125
High Low 1.75 0.126
High 1.623 0.128
PRE Low Low Low 49.495 8.516
High 42.735 8.917
High Low 36.659 8.812
High 33.421 9.319
High Low Low 40.382 9.319
High 38.753 9.509
High Low 50.694 9.574
High 38.783 9.78
*Quan=purchase quantity; Sim=similarity

SD
95% CI
SD
95% CI
DEP Mean
FRQ Mean
Low 1.608 0.058 1.494; 1.723
Low 2.055 0.06 1.939; 2.172
High 2.027 0.063
1.904; 2.15
High 1.579 0.062
1.459; 1.7
Low 40.577 4.448 31.842; 49.312 Low 42.841 4.537 33.932; 51.75
High 42.153 4.774 32.779; 51.527 High 39.889 4.689 30.68; 49.098
Low 1.071 0.049 0.974; 1.168
Low 1.038 0.05 0.939; 1.137
High 0.95 0.053 0.847; 1.054 High 0.983 0.052 0.881; 1.085
Low 5.258 0.074 5.113; 5.402
Low 4.876 0.075 4.728; 5.023
High 5.067 0.079 4.912; 5.222 High 5.449 0.078 5.296; 5.601

95% CI
Sim DEP FRQ BEF
1.462 1.901
Low Low Low
1.577 2.037
High
1.244 1.698
High Low
1.234 1.714
High
2.247 2.727
High Low Low
2.002 2.492
High
1.503 1.997
High Low
1.371 1.875
High
32.77 66.219 PCE Low Low Low
25.223 60.247
High
19.354 53.964
High Low
15.12 51.722
High
22.081 58.683
High Low Low
20.08 57.427
High
31.892 69.497
High Low
19.576 57.99
High

40

Mean
5.253
4.904
5.506
5.368
4.566
4.781
5.472
5.449
1.055
1.145
0.953
1.132
0.816
1.137
0.806
1.043

SD
0.141
0.148
0.146
0.154
0.154
0.158
0.159
0.162
0.094
0.099
0.098
0.103
0.103
0.105
0.106
0.108

95% CI
4.976 5.53
4.614 5.194
5.219 5.793
5.065 5.672
4.263 4.869
4.471 5.09
5.161 5.784
5.131 5.767
0.87 1.24
0.951 1.339
0.761 1.145
0.929 1.334
0.613 1.019
0.93 1.344
0.597 1.014
0.831 1.256

Table 2-5: Multivariate Tests
Wilks’ λ
F4,488
Sig. Partial Eta Squared
b
Intercept
.050
2918.666 0.000
.950
BEF
.984
2.552b
.038
.016
b
DEP
.954
7.393
.000
.046
FRQ
.910
15.194b
.000
.090
BEF*DEP
.991
1.394b
.235
.009
b
BEF*FRQ
1.000
.018
.999
.000
DEP*FRQ
.982
2.871b
.022
.018
b
BEF*DEP*FRQ
.996
.542
.705
.004
a. Design: Intercept + BEF + DEP + FRQ + BEF * DEP + BEF * FRQ
+ DEP * FRQ + BEF * DEP * FRQ
b. Exact statistic
c. Computed using alpha = .05

Powerc
1.000
.722
.997
1.000
.436
.054
.779
.182

In addition, there was a significant interaction effect between discount depth and
frequency on quantity (F1,617=5.678, p<.05). The positive relationship between discount depth and
quantity was significantly attenuated by an increase in discount frequency, while an increase in
discount depth strengthened the negative relationship between discount frequency and quantity.
Another significant effect between discount depth and frequency was found for similarity
(F1,617=3.929, p<.05). That is, the positive relationship between discount frequency and similarity
was significantly strengthened by an increase in discount depth.
Table 2-6: Between-Subjects Tests for an Interaction between DEP and FRQ
DEP*FRQ
Quantity
Error
PRE
Error
PCE
Error
Similarity
Error

MS
6.451
1.136
10229.903
6600.284
.001
.810
7.116
1.811

F1,617 Sig. Partial Eta Squared Powera
5.678 .017
.009
.662
1.550 .214

.003

.237

.002

.969

.000

.050

3.929 .048

.006

.508

41

Table 2-7: Univariate Test
BEF
Quantity

Contrast
Error
PRE
Contrast
Error
PCE
Contrast
Error
Similarity Contrast
Error
DEP
Quantity

Contrast
Error
PRE
Contrast
Error
PCE
Contrast
Error
Similarity Contrast
Error
FRQ
Quantity

Contrast
Error
PRE
Contrast
Error
PCE
Contrast
Error
Similarity Contrast
Error

MS
.549
1.136
5368.378
6600.284
6.634
.810
.840
1.811

F1,617
.483

Sig. Partial Eta Squared Powera
.487
.001
.107

.813

.367

.001

.147

8.191

.004

.013

.815

.464

.496

.001

.104

MS
F1,617 Sig. Partial Eta Squared Powera
27.147 23.893 .000
.037
.998
1.136
384.935
.058 .809
.000
.057
6600.284
2.253
2.782 .096
.004
.384
.810
5.635
3.111 .078
.005
.421
1.811
MS
F1,617 Sig. Partial Eta Squared Powera
35.147 30.934 .000
.048
1.000
1.136
1350.971 .205 .651
.000
.074
6600.284
.464
.573 .450
.001
.118
.810
50.942 28.125 .000
.044
1.000
1.811

42

Figure 3: 2x2 Mean Plots for Discount DEP, Discount FRQ, and BEF
Quantity

PRE

PCE

Similarity

43

Assessment of the Mediation Effects. Table 2-8 presents results of moderated mediation tests. The
positive and significant effect of discount depth on quantity was attenuated by an increase in
benefit change. That is its effect was reduced from .529 (t=4.292, p<.01) to .253 (t=2.001, p<.05).
The negative association between discount frequency and purchase quantity also was weakened
by an increasing level of benefit change. The strength of such relationship changed from -.514 (t=4.181, p<.01) to -.443 (t=-.3.508, p<.01).
Table 2-8: Tests of Mediation Effects
DIRECT AND INDIRECT EFFECTS
Conditional Direct effects of DEP on Quantity at values of the moderator (BEF)
BEF
Effect
SE
t
p
LLCI
Low
.529
.123
4.292
.000
.287
High
.253
.127
2.001
.046
.005
Conditional Indirect effect of DEP on Quantity at values of the moderator (BEF)
Effect
Boot SE BootLLCI BootULCI
.002
.007
-.006
.033
PRE at Low BEF
.002
.007
-.005
.029
PRE at High BEF
.008
.012
-.010
.042
PCE at Low BEF
.002
.007
-.006
.026
PCE at High BEF
Similarity at Low
.014
.014
-.006
.052
BEF
Similarity at High
.0014
.0077
-.0089
.0276
BEF
Indirect effect of highest order interaction
Effect
Boot SE BootLLCI BootULCI
-.001
.010
-.023
.017
PRE
-.006
.013
-.046
.008
PCE
-.012
.015
-.060
.006
Similarity
INDEX OF MODERATED MEDIATION
Index
Boot SE BootLLCI BootULCI
-.001
.010
-.023
.017
PRE
-.006
.013
-.046
.008
PCE
-.012
.015
-.060
.006
Similarity

44

ULCI
.772
.502

Table 2-8: (cont’d)
DIRECT AND INDIRECT EFFECTS
Conditional Direct effects of FRQ on Quantity at values of the moderator (BEF)
BEF
Effect
SE
t
p
LLCI
-.514
.123
-4.181
.000
-.756
Low
-.443
.126
-3.508
.001
-.692
High
Conditional Indirect effect of FRQ on Quantity at values of the moderator (BEF)
Effect
Boot SE BootLLCI BootULCI
.001
.008
-.009
.028
PRE at Low BEF
-.004
.008
-.035
.004
PRE at High BEF
.004
.009
-.006
.034
PCE at Low BEF
.003
.009
-.007
.046
PCE at High BEF
Similarity at Low
-.008
.018
-.049
.024
BEF
Similarity at High
-.008
.019
-.051
.028
BEF
Indirect effect of highest order interaction
Effect
Boot SE BootLLCI BootULCI
-.005
.012
-.052
.006
PRE
-.001
.012
-.038
.015
PCE
-.001
.009
-.029
.012
Similarity
INDEX OF MODERATED MEDIATION
Index
Boot SE BootLLCI BootULCI
-.005
.012
-.052
.006
PRE
-.001
.012
-.038
.015
PCE
-.001
.009
-.029
.012
Similarity

ULCI
-.273
-.195

However, regardless of the level of benefit change, all of indirect effects were not
significant. 95% CI of index of moderated mediation of the model with discount depth for each
mediator is (1) PRE: -0.023 ~ 0.017, (2) PCE: -0.046 ~ 0.008, and (3) similarity: -0.060 ~ 0.006.
95% CI of index moderated mediation of the model with discount frequency for each mediator is
(1) PCE: -0.052 ~ 0.006, (2) PRE: -0.038 ~ 0.015, and (3) similarity: -0.029 ~ 0.012.

45

Discussion

Table 5 shows the results of Study 2 hypotheses testing. First, consumers who recognize changed
benefit in their preferred product may conduct thorough price search at the category level. Second,
direct effects of discount depth and frequency revealed the same directional effects with those from
Study1. Third, an interaction between discount depth and frequency was found for purchase
quantity. That is, if high depth discount is frequently and regularly provided, its positive impact on
purchase quantity will be reduced. Furthermore, a firm should expect larger sales reduction when
high discount is provided than when low discount is provided, provided that a firm is planning to
increase discount frequency over time.
Study1 and 2 examined how different types of price discount influence stockpiling
behavior and how such effects may vary when product benefit is changed. However, little is known
about whether those effects would vary with an introduction of new product. Thus, Study3 will
investigate the relationships between types of discount and purchase quantity under the condition
in which a new product is launched.

46

STUDY 3
The primary goal of Study 3 is to replicate main and mediating effects tested in Study 1 and to
introduce the second moderating effect. A major difference from Study 2 is, in Study 3, a new
product which is under the established brand (i.e., consumers’ preferred brand) is added in the
experiment and serves as a moderator. Unlike a change in product size in Study 2, an
introduction of new product under consumers’ preferred brand is relatively easier to notice. In
other words, a launched new product would suffice to catch consumers’ attention at a category
level. If it serves as a wake-up call for consumers, they would be motivated to be more rational
on their purchase decision making. However, consumers may not be motivated to update their
price knowledge, if consumers have high satisfaction with the established product, the new
product requires them to change their behavior to use it, and they are risk averse (i.e., high
possibility to have cognitive dissonance). Thus, in general, the presence of new product may
make consumers more committed to the established product and more strive to leverage benefits
of a sales promotion by stockpiling.
H10 :

A new product moderates the effects of depth/frequency to the extent that the
effects of both discounts are amplified when a new product under the same brand
with an established product is introduced.

Research Design
The experiment of Study3 employs 2 (high vs. low discount depth) x 2 (high vs. low
discount frequency) x 2 (with vs. without new product) between-subject design. A major
difference from previous two studies is an introduction of new product. Study3 retains
manipulations of discount depth and frequency from Study1.

47

For Study 3, a new product was added into the second shopping task as an
alternative purchase option. This new product is under the same preferred brand name
provided by a participant. Its difference from the established product under the same
brand is a product benefit (i.e., brew ready packet). The price of new product is set at $7.44
which is slightly higher than the regular price $7.15.
In terms of means of data analysis, hypotheses testing is conducted by following the
exact same process of Study 2 with the addition of the new product introduction as a
moderator. That is, in order to test the moderated mediation, the model8 template in
PROCESS macro was used. Lastly, the cell sizes of Study3 are in the range of 49 to 139 and
total sample is 720.
Results

Manipulation Checks. Indicated average discount rates between high discount group (33.74%) and
low discount depth group (16.58%) were significantly different (F1,718=243.973, p<.01). The
difference between average perceptions on a given discount rate between two groups were also
significant (F1,718=203.541, p<.01). Indicated number of discounts showed a significant difference
between low discount frequency group (1.11) and high discount frequency group (1.89)
(F1,718=215.608, p<.01). Also participants’ frequency perception between two groups was
significantly different (F1,718=381.091, p<.01).
Assessment of the Measures. Table 3-1 indicates there are no signle measure which has a
correlation overlapping 1 with other variables. Table 3-2 presents a measurement model estimated
by CFA. It shows a good fit between the model and the data (CFI=.981; SRMR=.034;
RMSEA=.067) (Hu and Bentler, 1999). Acceptable level of reliability and validity were found for

48

both category knowledge (AVE=.831, CR=.907, Cronbach’s alpha=.896) and deal proneness
(AVE=.440, CR=.822, Cronbach’s alpha=.867). In addition, AVE was greater than its shared
variances. Table 3-3 shows descriptive information of measures.

49

Table 3-1: Correlation Matrix
Knowledge Deal
Knowledge
Deal
Lower 95% CI
Upper 95% CI

Quantity
Lower 95% CI
Upper 95% CI

PRE
Lower 95% CI
Upper 95% CI

PCE
Lower 95% CI
Upper 95% CI

Certainty
Lower 95% CI
Upper 95% CI

Similarity
Lower 95% CI
Upper 95% CI

Household Size
Lower 95% CI
Upper 95% CI

Household
Income
Lower 95% CI
Upper 95% CI

.831
-.008

Quantity PRE

PCE

Certainty Similarity Household
Size

Household
Income

.000
.440

-0.081
0.065

.091*

-.013

0.018
0.163

-0.086
0.060

.

.068

-.136**

-.012

-0.005
0.140

-0.207
-0.064

-0.085
0.061

.049

-.111**

-.008

.061

-0.024
0.122

-0.183
-0.038

-0.081
0.065

-0.012
0.133

.118**

.058

.119**

0.045
0.189

-0.015
0.131

0.046
0.190

-0.301
-0.163

-0.075
0.071

.
.

-.233** -.002

.

.063

.025

-.036

-.049

-.043

.118**

-0.010
0.135

-0.048
0.098

-0.109
0.037

-0.122
0.024

-0.116
0.030

0.045
0.189

.

.032

-.043

.061

.023

-.043

.061

.054

-0.041
0.105

-0.116
0.030

-0.012
0.133

-0.050
0.096

-0.116
0.030

-0.012
0.133

-0.019
0.127

.015

-.174**

.141**

-.007

-.005

.063

.062

.258**

-0.058
0.088

-0.244
-0.102

0.069
0.212

-0.080
0.066

-0.078
0.068

-0.010
0.135

-0.011
0.134

0.188
0.325

.

**. Correlation is significant at the 0.01 level (2-tailed).
*. Correlation is significant at the 0.05 level (2-tailed).
Notes: Correlations are in the lower triangle. Shared variances are in the upper triangle. AVE is on the diagonal.

50

.

Table 3-2: Measurement Model
λ
Category Knowledge (CR=.907; AVE=.831)
 Rate your overall knowledge of “ground coffee”, as compared to the
average consumer.
 Please check the scale that best describes your familiarity with
“ground coffee”.
Deal Proneness (CR=.822; AVE=.440)
 If a product is on sale, that can be a reason for me to buy it.
 When I buy a brand that is on sale, I feel that I am getting a good
deal.
 I have favorite brand, but most of time I buy the brand that is on
sale.
 One should try to buy the brand that is on sale.
 I am more likely to buy brands that are on sale.
 Compared to most people, I am more likely to buy brands that are on
sale.

.81
1.0

.56
.62
.72
.73
.86
.84

Goodness-of-Fit Indexes
CFI=.981; SRMR=.034; RMSEA=.062

Table 3-3: Descriptive Information of Measures
N Mini Max
Mean
SD
Quantity
720 1.00
6.00
1.804 1.153
PRE
720 0.00 570.00 50.211 97.584
PCE
720 0.00
3.00
1.106 0.965
Similarity
720
1
7
5.063 1.356
Certainty
720
1
7
4.286 1.520
Household Size
720
1
5
2.653 1.259
Household Income 716
1
14
6.307 3.547
Knowledge
720 1.00
7.00
4.189 1.470
Deal Proneness
720 1.00
7.00
4.768 1.127
Valid N
716

Assessment of the Main Effects. Multivariate tests revealed significant main effects of new product
(Wilks’λ=.986, p < .05, discount depth (Wilks’λ=.929, p < .001), and discount depth
(Wilks’λ=.969, p < .01). In addition, two significant interactions between new product and
discount depth (Wilks’λ=.894, p < .01) as well as new product and discount frequency
51

(Wilks’λ=.976, p < .01) were found. Table3-4 presents means, standard deviations, and sample
sizes for all six pricing groups. Given the fact that overall multivariate tests of new product,
discount depth, discount frequency, and two interactions were significant (Table 3-5), univariate
tests of such each mediator and a predicted variable as well as between-subjects tests of interaction
effects on those variables are discussed. Table 3-6 and Figure 4 present the significance of each
test and directions of relationships (also see Table3-7).
The introduction of new product was negatively associated with purchase quantity
(F1,712=4.076, p<.05) and also positively with PCE (F1,712=4.256, p<.05). Discount depth showed
its significant and positive relationships with quantity (F1,712=30.514, p<.01) and PCE
(F1,712=19.402, p<.01). Discount frequency was show to have a negative relationship with quantity
(F1,712=9.381, p<.05) and with PCE (F1,712=8.508, p<.01), and a positive relationship with
similarity (F1,712=4.686, p<.05).
Two significant interactions revealed that PCE is influenced by interactions between new
product introduction and discount depth (F1,712=82.005, p<.01) as well as new product introduction
and frequency (F1,712=13.593, p<.01). With new product introduction, an increase in discount
depth increased PCE. When a low discount was available, PCE was higher without new product
introduction than with new production introduction. However, with an available high discount,
participants made more PCE when new product was introduced. In terms of the relationship
between discount frequency and PCE, when a discount was provided in low frequency,
introduction of new product made participants to make more PCE. However, with high discount
frequency, the less amount of PCE was made by participants without new product introduction
than with new product introduction.

52

Table 3-4: Descriptive Statistics
SD
95% CI
NP Mean
Quan w/o 1.837 0.051 1.738; 1.937
w/
1.51; 1.802
1.656 0.074
PRE w/o 50.683 4.47 41.906; 59.46
w/ 48.838 6.54 35.997; 61.678
PCE w/o 1.073 0.041 0.992; 1.154
w/
1.224 0.06 1.105; 1.342
Sim w/o 5.115 0.061 4.994; 5.235
w/
4.966 0.09 4.789; 5.143

SD
95% CI
SD
95% CI
DEP Mean
FRQ Mean
Low 1.498 0.064 1.372; 1.625
Low 1.884 0.061 1.764; 2.005
High 1.995 0.063 1.872; 2.118 High 1.609 0.066 1.480; 1.738
Low 42.853 5.661 31.739; 53.967 Low 48.853 5.406 38.239; 59.467
High 56.667 5.542 45.786; 67.548 High 50.668 5.791 39.299; 62.037
Low 0.987 0.052
0.885; 1.09
Low 1.255 0.050 1.157; 1.353
High 1.309 0.051 1.209; 1.409 High 1.042 0.053 0.937; 1.147
Low 5.113 0.078
4.96; 5.266
Low 4.922 0.074 4.776; 5.068
High 4.967 0.076 4.818; 5.117 High 5.158 0.080 5.002; 5.315

Quan

NP DEP FRQ Mean
SD
95% CI
Sim NP DEP FRQ Mean SD
95% CI
w/o Low Low 1.619 0.094 1.434 1.803
w/o Low 5.007 0.114 4.783 5.231 5.007
1.477
0.118
1.245
1.709
High
5.205 0.143 4.923 5.486 5.205
2.396
0.096
2.207
2.584
High Low
High 4.821 0.116 4.593 5.049 4.821
1.858
0.096
1.670
2.046
High
5.425 0.116 5.197 5.653 5.425
1.491
0.150
1.197
1.784
w/ Low Low
w/ Low 5.055 0.181 4.699 5.41 5.055
1.407
0.144
1.123
1.69
High
5.186 0.175 4.843 5.53 5.186
2.032
0.141
1.756
2.309
High Low
High 4.806 0.171 4.471 5.142 4.806
1.694
0.158
1.383
2.005
High
4.816 0.192 4.439 5.193 4.816
PRE w/o Low Low 44.640 8.289 28.367 60.914 PCE w/o Low Low 1.201 0.076 1.051 1.352
High 44.636 10.417 24.184 65.089
High 1.284 0.096 1.095 1.473
High Low 60.067 8.442 43.493 76.641
High Low 0.888 0.078 0.735 1.041
High 53.388 8.442 36.814 69.962
High 0.918 0.078 0.765 1.071
w/ Low Low 44.509 13.177 18.639 70.379
w/ Low Low 1.091 0.122 0.852 1.329
High 37.627 12.722 12.649 62.605
High 0.373 0.117 0.143 0.603
High Low 46.194 12.411 21.827 70.56
High Low 1.839 0.114 1.614 2.063
High 67.02 13.96 39.612 94.429
High 1.592 0.129 1.339 1.845
*Quan=purchase quantity; Sim=similarity; w/o=without; w/=with

53

Table 3-5: Multivariate Tests
Wilks’ λ
F4,7/9
Sig.
Partial Eta Squared
b
Intercept
.058
2878.312
0.000
.942
NP
.986
2.573b
.037
.014
b
DEP
.929
13.477
.000
.071
FRQ
.969
5.585b
.000
.031
NP*DEP
.894
21.082b
.000
.106
b
NP*FRQ
.976
4.294
.002
.024
DEP*FRQ
.992
1.498b
.201
.008
b
NP*DEP*FRQ
.992
1.401
.232
.008
a. Design: Intercept + NP + DEP + FRQ + NP * DEP + NP * FRQ
+ DEP * FRQ + NP * DEP * FRQ
b. Exact statistic
c. Computed using alpha = .05

Powerc
1.000
.727
1.000
.979
1.000
.930
.467
.439

Table 3-6: Univariate Tests
NP
Quantity

Contrast
Error
PRE
Contrast
Error
PCE
Contrast
Error
Similarity Contrast
Error
DEP
Quantity

Contrast
Error
PRE
Contrast
Error
PCE
Contrast
Error
Similarity Contrast
Error

MS
5.011
1.230
518.202
9549.762
3.456
.812
3.358
1.807

F1,712
4.076

Sig. Partial Eta Squared Powera
.044
.006
.522

.054

.816

.000

.056

4.256

.039

.006

.540

1.858

.173

.003

.275

MS
F1,712 Sig. Partial Eta Squared Powera
37.517
30.514 .000
.041
1.000
1.230
29036.801 3.041 .082
.004
.414
9549.762
15.757
19.402 .000
.027
.993
.812
3.240
1.793 .181
.003
.267
1.807

54

Table 3-6: (cont’d)
FRQ
Quantity

Contrast
Error
PRE
Contrast
Error
PCE
Contrast
Error
Similarity Contrast
Error

MS
11.534
1.230
501.511
9549.762
6.910
.812
8.468
1.807

F1,712 Sig. Partial Eta Squared Powera
9.381 .002
.013
.864
.053

.819

.000

.056

8.508 .004

.012

.830

4.686 .031

.007

.580

Table 3-7: Between-Subjects Tests for an Interactions
NP*DEP
MS
F1,712 Sig. Partial Eta Squared Powera
Quantity
1.031
.839 .360
.001
.150
Error
875.416
PRE
452.667
.047 .828
.000
.055
Error
6799430.794
PCE
66.599
82.005 .000
.103
1.000
Error
578.241
Similarity
4.052
2.242 .135
.003
.321
Error
1286.576
NP*FRQ
MS
F1,712 Sig. Partial Eta Squared Powera
Quantity
.624
.508 .476
.001
.110
Error
875.416
PRE
4046.649
.424 .515
.001
.100
Error
6799430.794
PCE
11.039
13.593 .000
.019
.957
Error
578.241
Similarity
4.143
2.293 .130
.003
.327
Error
1286.576

55

Figure 4: 2x2 Mean Plots for Discount DEP, Discount FRQ, and NP
Quantity

PRE

PCE

Similarity

Assessment of the Mediation Effects. Table 3-8 presents results of moderated mediation tests. The
introduction of new product decreased the positive direct effect of discount depth on quantity. That

56

is, its effect was reduced from .527 (t=5.186, p<.01) to .433 (t=2.785, p<.01). While the negative
association between discount frequency and purchase quantity, -.303 (t=-2.950, p<.01) without
new product introduction was significant, new product introduction made the relationship
insignificant as well as weaker, -.187 (t=-1.217, p>.10).
With regard to moderated mediation effects, regardless of the presence of new product, all
of indirect effects were not significant. 95% CI of index moderated mediation of the model with
discount depth for each mediator is (1) PRE: -.008 ~ .023, (2) PCE: -.133 ~ .107, and (3) similarity:
-.002 ~ .069. 95% CI of index moderated mediation of the model with discount frequency for each
mediator is (1) PRE: -.0096 ~ .0188, (2) PCE: -.0426 ~ .0702, and (3) similarity: -.0044 ~ .0614.
Table 3-8: Tests of Moderated Mediation Effects
DIRECT AND INDIRECT EFFECTS
Conditional Direct effects of DEP on Quantity at values of the moderator (NP)
NP
Effect
SE
t
p
LLCI
Without
.527
.102
5.186
.000
.328
With
.433
.156
2.785
.006
.128
Conditional Indirect effect of DEP on Quantity at values of the moderator (NP)
Effect
Boot SE BootLLCI BootULCI
-.001
.006
-.019
.008
PRE at w/o NP
-.001
.006
-.019
.007
PRE at w/ NP
.004
.015
-.026
.036
PCE at w/o BEF
-.013
.046
-.099
.083
PCE at w/ BEF
Similarity at w/o
-.001
.008
-.017
.015
NP
.018
.015
-.001
.061
Similarity at w/ NP
Indirect effect of highest order interaction
Effect
Boot SE
PRE
PCE
Similarity

.001
-.017
.019

BootLLCI BootULCI

.006
.061
.016

-.008
-.133
-.002

57

.023
.107
.069

ULCI
.727
.738

Table 3-8: (cont’d)
INDEX OF MODERATED MEDIATION
Index
Boot SE BootLLCI BootULCI
PRE
PCE
Similarity

.001
.006
-.008
.023
-.017
.061
-.133
.107
.019
.016
-.002
.069
DIRECT AND INDIRECT EFFECTS
Conditional Direct effects of FRQ on Quantity at values of the moderator (NP)
NP
Effect
SE
t
p
LLCI
-.303
.103
-2.950
.003
-.504
Without
-.187
.153
-1.217
.224
-.488
With
Conditional Indirect effect of FRQ on Quantity at values of the moderator (NP)
Effect
Boot SE BootLLCI BootULCI
.000
.003
-.010
.005
PRE at w/o NP
.000
.005
-.008
.014
PRE at w/ NP
.000
.004
-.010
.008
PCE at w/o BEF
.010
.026
-.040
.066
PCE at w/ BEF
Similarity at w/o
-.020
.015
-.063
.002
NP
-.008
.012
-0.049
0.006
Similarity at w/ NP
Indirect effect of highest order interaction
Effect
Boot SE BootLLCI BootULCI
PRE
PCE
Similarity

PRE
PCE
Similarity

ULCI
-.101
0.114

.000
.006
-.010
.019
.010
.027
-.043
.070
.012
.015
-.004
.061
INDEX OF MODERATED MEDIATION
Index
Boot SE BootLLCI BootULCI
.0004
.0061
-.0096
.0188
.0101
.0266
-.0426
.0702
.0122
.0154
-.0044
.0614

Discussion

Study3 examined and compared proposed direct and indirect relationships introduced from Study
across two situations, without new product introduction vs. with new product introduction. Table
5 is the summary of results. First, when a firm introduces new product under the same brand, an

58

increase in discount depth would make consumers less motivated to search and compare prices at
the category level opposed to its negative relationship without it. Second, it would be expected for
firm’s new product introduction to lead to narrower price search and knowledge at the category
level, and to change the positive association between discount frequency and PCE into strongly
negative association.
Tests of moderated mediation found no significant indirect effects of discount depth or
frequency on purchase quantity regardless of the presence of new product. Across two situations,
discount depth showed its significant and positive impact on purchase quantity. However, new
product introduction made the negative influence of discount frequency on purchase quantity
insignificant despite its same directional influence.
Study3 examined how the relationships established in Study1 and 2 would vary across
two different situations (i.e., new product introduction). Then would new competitor’s entrance
(i.e., new brand) would make changes in such relationships? Thus, Study4 is to examine how new
brand introduction would influence direct effects of discount on purchase quantity.

59

STUDY 4
The objectives of Study 4 is similar with ones of Study 3. The only difference from Study 3 is that
this current study proposes the introduction of a new product under new brand name as a moderator.
Unlike a new product under the same preferred brand name, new brand would be more challenged
to catch consumers. Unless consumers perceive it as a better solution, it will be ignored.
Furthermore it may encourage consumers more strive to leverage benefits of a sales promotion by
stockpiling and make a given promotion to their preferred brand more attractive.
H11 :

A new brand moderates the effects of depth/frequency to the extent that the effects
of both discounts are amplified when a new brand is introduced.

Research Design
The experiment of Study4 employs 2 (high vs. low discount depth) x 2 (high vs. low
discount frequency) x 2 (with vs. without new brand) between-subject design. Unlike
Study3, Study4 introduces a ground coffee product under new brand. Study4 retails
manipulations of discount depth and frequency from Study1.
For Study4, a new brand is introduced as an alternative purchase option in the
second shopping task. Like Café Vale, this new brand is hypothetical, Venezia Espresso. The
price of new brand is set at $7.44 which is the same regular price of new product in Study3.
In terms of means of data analysis, hypotheses testing is conducted by following the
exact same process of Study2 with the addition of new brand introduction as the moderating
effect. That is, in order to test the moderated mediation, the template of model8 in PROCESS
macro was used. Lastly, cell sizes of Study4 are in the range of 52 to 139 and total sample is 713.

60

Results

Manipulation Checks. Indicated average discount rates between high discount group (33.18%) and
low discount depth group (17.14%) were significantly different (F1,711=186.818, p<.01). The
difference between average perceptions on a given discount rate between two groups were also
significant (F1,711=158.006, p<.01). Indicated number of discounts showed a significant difference
between low discount frequency group (1.11) and high discount frequency group (1.94)
(F1,711=230.943, p<.01). Also participants’ frequency perception between two groups was
significantly different (F1,711=444.794, p<.01).

61

Table 4-1: Correlation Matrix
Knowledge Deal
Knowledge
Deal
Lower 95% CI
Upper 95% CI

Quantity
Lower 95% CI
Upper 95% CI

PRE
Lower 95% CI
Upper 95% CI

PCE
Lower 95% CI
Upper 95% CI

Certainty
Lower 95% CI
Upper 95% CI

Similarity
Lower 95% CI
Upper 95% CI

Household Size
Lower 95% CI
Upper 95% CI

Household
Income
Lower 95% CI
Upper 95% CI

.805
-.034

Quantity PRE

PCE

Certainty Similarity Household
Size

Household
Income

.002
.461

-.107
.040

.111**

-0.036

.038
.183

-.109
.038

.132**

-.150**

-.058

.059
.203

-.221
-.077

-.131
.015

.000

-.94*

.031

.109**

-.073
.073

-.166
-.021

-.043
.104

.036
.181

.043

.032

.89*

-.031
.116

-.042
.105

.016
.161

-.325
-.188

-.104
.043

.029

.045

-.067

-.008

-.04

.127**

-.045
.102

-.029
.118

-.140
.006

-.081
.065

-.113
.034

.054
.199

.081*

.001

.087*

.016

-.032

.072

.039

.008
.154

-.072
.074

.014
.159

-.057
.089

-.105
.042

-.001
.145

-.035
.112

.036

-.121**

.105**

-.023

-.018

.100**

0.056

.278**

-.038
.109

-.193
-.048

.032
.177

-.096
.051

-.091
.055

.027
.172

-.017
.129

.209
.344

.
.
.

-.258** -.031

.
.
.

**. Correlation is significant at the 0.01 level (2-tailed).
*. Correlation is significant at the 0.05 level (2-tailed).
Notes: Correlations are in the lower triangle. Shared variances are in the upper triangle. AVE is on the diagonal.

62

.

Assessment of the Measures. Table 4-1 indicates there are no signle measure which has a
correlation overlapping 1 with other variables. Table 4-2 presents a measurement model estimated
by CFA. It shows a good fit between the model and the data (CFI=.978; SRMR=.035;
RMSEA=.068) (Hu and Bentler, 1999). Acceptable level of reliability and validity were found for
both category knowledge (AVE=.805, CR=.891, Cronbach’s alpha=.889) and deal proneness
(AVE=.461, CR=.834, Cronbach’s alpha=.879). In addition, AVE was greater than its shared
variances. Table 4-3 shows descriptive information of measures.
Table 4-2: Measurement Model
λ
Category Knowledge (CR=.891; AVE=.805)
 Rate your overall knowledge of “ground coffee”, as compared to the
average consumer.
 Please check the scale that best describes your familiarity with
“ground coffee”.
Deal Proneness (CR=.834; AVE=.461)
 If a product is on sale, that can be a reason for me to buy it.
 When I buy a brand that is on sale, I feel that I am getting a good
deal.
 I have favorite brand, but most of time I buy the brand that is on
sale.
 One should try to buy the brand that is on sale.
 I am more likely to buy brands that are on sale.
 Compared to most people, I am more likely to buy brands that are on
sale.
Goodness-of-Fit Indexes
CFI=.978; SRMR=.035; RMSEA=.068

63

.85
.94

0.59
0.63
0.73
0.75
0.88
0.86

Table 4-3: Descriptive Information of Measures
N Mini Max
Mean
SD
Quantity
713 1.00
6.00
1.889
1.236
PRE
713 0.00 570.00 53.993 102.355
PCE
713 0.00
3.00
1.097
.964
Similarity
713
1
7
5.11
1.383
Certainty
713
1
7
4.19
1.521
Household Size
713
1
5
2.66
1.263
Household Income 709
1
14
6.21
3.547
Knowledge
713 1.00
7.00
4.223
1.445
Deal Proneness
713 1.00
7.00
4.732
1.169
Valid N
709

Assessment of the Main Effects. Multivariate tests revealed significant main effects of discount
depth (Wilks’λ=.898, p <.01), and discount frequency (Wilks’λ=.944, p <.01). In addition, two
significant interactions between new brand and discount depth (Wilks’λ=.905, p <.01) as well as
discount depth and discount frequency (Wilks’λ=.975, p <.01) were found. Table 4-4 presents
means, standard deviations, and sample sizes for all six pricing groups. Given the fact that overall
multivariate tests of new product, discount depth, discount frequency, and two interactions were
significant (Table4-5), univariate tests of such each mediator and a predicted variable as well as
between-subjects tests of interaction effects on those variables are discussed. Table 4-6, 4-7 and
Figure5 present the significance of each test and directions of relationships.
Discount depth had positive relationships with quantity (F1,705=48.590, p<.01), PRE
(F1,705=11.346, p<.01) and PCE (F1,705=12.225, p<.01), and a negative relationship with similarity
(F1,705=6.837, p<.01). Discount frequency was found to be negatively associated with quantity
(F1,705=22.271, p<.01) but positively associated with similarity (F1,705=19.851, p<.01).

64

Table 4-4: Descriptive Statistics
SD
95% CI
SD
95% CI
SD
95% CI
NB Mean
DEP Mean
FRQ Mean
Quan w/o 1.837 .054
1.732; 1.943 Low 1.544 .070
1.407; 1.682
Low 2.107 .067
1.976; 2.238
w/ 1.923 .080
1.766; 2.076 High 2.216 .066
2.086; 2.345 High 1.653 .069
1.517; 1.789
PRE w/o 50.683 4.654 41.546; 58.820 Low 41.077 6.053 29.192; 52.962 Low 52.772 5.765 41.454; 64.090
w/ 59.490 6.895 45.953; 73.027 High 69.096 5.705 57.895; 80.297 High 57.401 5.997 45.627; 69.174
PCE w/o 1.073 .042
.990; 1.156
Low .993
.055
.885; 1.101
Low 1.140 .052
1.037; 1.243
w/ 1.177 .063
1.054; 1.300 High 1.257 .052
1.155; 1.359 High 1.110 .055
1.003; 1.217
Sim w/o 5.115 .062
4.994; 5.235 Low 5.267 .080
5.110; 5.424
Low 4.878 .076
4.728; 5.028
w/ 5.132 .091
4.953; 2.311 High 4.979 .076
4.831; 5.128 High 5.369 .079
5.213; 5.524
Quan NB DEP FRQ Mean
SD
95% CI
Sim NB DEP FRQ Mean SD
95% CI
.100
1.423
1.815
w/o Low Low 1.619
w/o Low 5.007 .114 4.783 5.231 5.007
.125
1.231
1.724
5.205 .144 4.923 5.486 5.205
High 1.477
.102
2.196
2.595
High Low 2.396
High 4.821 .116 4.593 5.049 4.821
.102
1.659
2.058
5.425 .116 5.197 5.654 5.425
High 1.858
.162
1.418
2.053
w/ Low Low 1.736
w/ Low 5.434 .185 5.071 5.797 5.434
1.346
.163
1.026
1.667
5.423 .187 5.057 5.790 5.423
High
.157
2.370
2.987
High Low 2.679
High 4.250 .180 3.897 4.603 4.250
.156
1.624
2.236
5.421 .178 5.071 5.771 5.421
High 1.930
PRE w/o Low Low 44.640 8.628 27.700 61.581 PCE w/o Low 1.201 .078 1.047 1.355 1.201
1.284 .099 1.090 1.478 1.284
High 44.636 10.844 23.346 65.927
High Low 60.067 8.788 42.814 77.321
High .888 .080 .731 1.045 .888
.918 .080 .761 1.075 .918
High 53.388 8.788 36.134 70.642
w/ Low Low 39.434 13.973 12.000 66.868
w/ Low .755 .127 .505 1.004 .755
.731 .128 .479 .983 .731
High 35.596 14.107 7.899 63.293
High Low 66.946 13.594 40.257 93.636
High 1.714 .124 1.472 1.957 1.714
1.509 .123 1.268 1.749 1.509
High 95.982 13.474 69.528 122.437
*Quan=purchase quantity; Sim=similarity; w/o=without; w/=with

65

Table 4-5: Multivariate Tests
Wilks’ λ
F4,702
Sig. Partial Eta Squared Powerc
Intercept
.058
2827.127b 0.000
.942
1.000
NB
.995
.928b
.447
.005
.296
b
DEP
.898
20.001
.000
.102
1.000
FRQ
.944
10.406b
.000
.056
1.000
NB*DEP
.905
18.320b
.000
.095
1.000
b
NB*FRQ
.994
1.053
.379
.006
.334
DEP*FRQ
.975
4.416b
.002
.025
.937
b
NB*DEP*FRQ
.993
1.208
.306
.007
.381
a. Design: Intercept + NB + DEP + FRQ + NB * DEP + NB * FRQ
+ DEP * FRQ + NB * DEP * FRQ
b. Exact statistic
c. Computed using alpha = .05

Table 4-6: Univariate Tests
NB
Quantity

Contrast
Error
PRE
Contrast
Error
PCE
Contrast
Error
Similarity Contrast
Error
DEP
Quantity

Contrast
Error
PRE
Contrast
Error
PCE
Contrast
Error
Similarity Contrast
Error

MS
1.085
1.386
1.085
1.386
1.085
1.386
1.085
1.386

F1,705
.783

Sig. Partial Eta Squared Powera
.377
.001
.143

.783

.377

.001

.143

.783

.377

.001

.143

.783

.377

.001

.143

MS
F1,705 Sig. Partial Eta Squared Powera
67.344
48.590 .000
.064
1.000
1.386
117414.320 11.346 .001
.016
.920
10348.437
10.463
12.225 .001
.017
.937
.856
12.393
6.837 .009
.010
.743
1.813

66

Table 4-6: (cont’d)
FRQ
Quantity

Contrast
Error
PRE
Contrast
Error
PCE
Contrast
Error
Similarity Contrast
Error

MS
F1,705 Sig. Partial Eta Squared Powera
30.866
22.271 .000
.031
.997
1.386
3204.364
.310 .578
.000
.086
10348.437
.128
.149 .699
.000
.067
.856
35.982
19.851 .000
.027
.994
1.813

In addition, two interactions effects were found. First, interaction effects between new
brand and discount depth were found for PCE (F1,705=63.808, p<.01) and similarity (F1,705=7.682,
p<.01). With new brand introduction, an increase in discount depth increased PCE. When a low
discount was available, PCE was higher without the presence of new brand than with it. However,
with a high discount, participants made more PCE when new brand was introduced. In terms of
similarity, without new brand, similarity was not significantly difference between high and low
depth groups, but after, new brand was introduced, similarity decreased drastically with an increase
of discount depth. Second, interaction effects between discount depth and discount frequency were
found for similarity (F1,705=23.603, p<.01). The level of discount depth strengthened the
relationship between discount frequency and similarity and the level discount frequency attenuated
the relationship between discount depth and similarity.

67

Table 4-7: Between-Subjects Tests for an Interactions
NB*DEP
Quantity
Error
PRE
Error
PCE
Error
Similarity
Error
DEP*FRQ
Quantity
Error
PRE
Error
PCE
Error
Similarity
Error

MS
F1,705 Sig. Partial Eta Squared Powera
1.270
.917 .339
.001
.159
1.386
37952.434 3.667 .056
.005
.481
10348.437
54.612
63.808 .000
.083
1.000
.856
13.924
7.682 .006
.011
.790
1.813
MS
F1,705
5.327
3.844
1.386
6415.691
.620
10348.437
.513
.600
.856
23.603
13.022
1.813

Sig. Partial Eta Squared Powera
.050
.005
.499
.431

.001

.123

.439

.001

.121

.000

.018

.950

Assessment of the Mediation Effects. Table 4-8 presents results of moderated mediation tests. The
introduction of new brand increased the positive direct effect of discount depth on quantity. The
effect changed from .573 (t=5.265, p<.01) to .728 (t=4.331, p<.01). Also the negative association
between discount frequency and purchase quantity was strengthened by new brand introduction.
The effect increased from -.308 (t=-2.798, p<.01) to -.607 (t=-3.674, p<.01).
With regard to moderated mediation effects, regardless of the presence of new product, all
of indirect effects were not significant. 95% CI of index moderated mediation of the model with
discount depth for each mediator is (1) PRE: -.083 ~ .007, (2) PCE: -.082 ~ .165, and (3) similarity:
-.00 ~ .110. 95% CI of index moderated mediation of the model with discount frequency for each
mediator is (1) PRE: -.054 ~ .005, (2) PCE: -.049 ~ .007, and (3) similarity: -.066 ~ .011.

68

Figure 5: 2x2 Mean Plots for Discount DEP, Discount FRQ, and NB
Quantity

PRE

PCE

Similarity

69

Table 4-8: Tests of Moderated Mediation Effects
DIRECT AND INDIRECT EFFECTS
Conditional Direct effects of DEP on Quantity at values of the moderator (NP)
NB
Effect
SE
t
p
LLCI
Without
.573
.109
5.265
.000
.359
With
.728
.168
4.331
.000
.398
Conditional Indirect effect of DEP on Quantity at values of the moderator (NP)
Effect
Boot SE BootLLCI BootULCI
PRE at w/o NB
-.015
.012
-.047
.002
PRE at w/ NB
-.033
.021
-.091
-.005
PCE at w/o BEF
-.011
.018
-.052
.022
PCE at w/ BEF
.028
.045
-.058
.122
Similarity at w/o
NB
-.001
.009
-.019
.018
Similarity at w/ NB
.034
.026
.000
.110
Indirect effect of highest order interaction
Effect
Boot SE BootLLCI BootULCI
-.018
.020
-.083
.007
PRE
.039
.061
-.082
.165
PCE
.035
.027
-.001
.110
Similarity

PRE
PCE
Similarity

ULCI
.786
1.058

INDEX OF MODERATED MEDIATION
Index
Boot SE BootLLCI BootULCI
-.018
.020
-.083
.007
.039
.061
-.082
.165
.035
.027
-.001
.110

DIRECT AND INDIRECT EFFECTS
Conditional Direct effects of FRQ on Quantity at values of the moderator (NP)
NB
Effect
SE
t
p
LLCI
Without
-.308
.110
-2.798
.005
-.523
With
-.607
.165
-3.674
.000
-.931
Conditional Indirect effect of FRQ on Quantity at values of the moderator (NP)
Effect
Boot SE BootLLCI BootULCI
PRE at w/o NB
.002
.007
-.009
.024
PRE at w/ NB
-.009
.011
-.047
.004
PCE at w/o BEF
.001
.005
-.006
.016
PCE at w/ BEF
-.005
.010
-.039
.006
Similarity at w/o
NB
-.022
.018
-.069
.005
Similarity at w/ NB
-.028
.025
-.096
.005

70

ULCI
-.092
-.283

Table 4-8: (cont’d)
Indirect effect of highest order interaction
Effect
Boot SE

BootLLCI BootULCI

PRE
PCE
Similarity

-.011
.014
-.054
.005
-.006
.012
-.049
.007
-.007
.015
-.066
.011
INDEX OF MODERATED MEDIATION
Index
Boot SE BootLLCI BootULCI

PRE
PCE
Similarity

-.011
-.006
-.007

.014
.012
.015

-.054
-.049
-.066

.005
.007
.011

Discussion

Study4 examined and compared proposed direct and indirect relationships introduced from Study1
across two situations, without new brand introduction vs. with new brand introduction. Table 5
summarizes the results of hypotheses testing.
A significant interaction effect between new brand and discount depth on PCE suggests
inherent complexity in consumer price interitem search behavior. In addition, with newly launched
brand product, consumers may purchase more with the presence of a high discount. However, at
the same time, negative influence of discount frequency was strengthened.

71

GENERAL DISCUSSION

Across all of four experiments, a positive association between discount depth and stockpiling and
a negative association between discount frequency and stockpiling are found. Regarding to the
effect of discount depth on purchase quantity, consumers are not likely to miss the chance to
purchase more items for future consumption. It could be posited that, for stock-up products (i.e.,
nonperishable and frequently purchased consumed products), if there is an increasing opportunity
to purchase consumers’ preferred brand at a discounted price, they strive to minimize costs by
purchasing the normal quantity on each purchase occasion which does not increase their inventory
level above the normal level.
Table 5: Summary of Hypotheses Testing
Hypothesis
Study1
Study2
Study3
Study4
DEP  PRE(H1a)
+
n/s
n/s
n/s
Supported
DEP  PCE(H1b)
+ Non supported
n/s
Supported
Supported
DEP  SIM(H2)
–
n/s
n/s
n/s
n/s
DEP  Q(H3)
+
Supported
Supported
Supported
Supported
FRQ  PRE(H4a)
–
n/s
n/s
n/s
n/s
FRQ  PCE(H4b)
–
n/s
n/s
Supported
n/s
FRQ  SIM(H5)
+
Supported
Supported
Supported
Supported
FRQ  Q(H6)
–
Supported
Supported
Supported
Supported
DEP  PRE  Q(H7a)
n/s
DEP  PCE  Q(H7b)
n/s
DEP  SIM  Q(H7c)
n/s
FRQ  PRE  Q(H8a)
n/s
FRQ  PCE  Q(H8b)
n/s
FRQ  SIM  Q(H8c)
n/s
a
b
BEF moderator(H9)
+
n/a
Not support
n/a
n/a
NP moderator(H10)a
+
n/a
n/a
Not supportb
n/a
a
NB moderator(H11)
+
n/a
n/a
n/a
Supported
DEP = discount depth; FRQ = discount frequency; BEF = benefit change; NP = new product;
NB = new brand; PRE = price recall error; PCE = price comparison error; SIM = perceived
similarity; Q = purchase quantity; n/a = not applicable; n/s = not significant
a
Moderating effects are found significant only for the main effects of discount depth (frequency)
on quantity.
b
”Not support” means the result is significant but directionally opposite.

72

Those relationships are negatively influenced by a moderating effect of a benefit change
(i.e., a reduction in product size) (Study 2) and of an introduction of new product under the same
preferred brand name (Study3). In contrast, an introduction of new brand (Study 4) amplified the
relationships. A greater reduction in product benefit weakened the positive relationship between
discount depth and stockpiling behavior and the negative relationship between discount frequency
and stockpiling. It may be posited that, when a given negative change in product size is recognized
and is greater, consumers are likely to have less confidence with their purchase quantity decision
making which may come from greater expected cognitive dissonance regarding to purchase
quantity and lower level satisfaction with regard to the current shopping experience. New product
introduction shows the same effects with benefit change and it may increase consumers’ cognitive
dissonance on a forgone new alternative. However, with an introduction of new brand, a greater
discount frequency more decreases purchase quantity than without the introduction. The new brand
introduction strengthened the positive association of discount depth and purchase quantity. The
presence of new brand might increase consumers’ anticipated regret on their current purchase
quantity which will be based on discount depth and frequency.
With regard to consumers’ price search behavior and price knowledge, Study 1 shows a
provided deeper discount on their preferred brand more motivate their interitem price search and
have more accurate price knowledge at the category level than a shallower discount. In Study 2,
benefit change shows its negative association with PCE. It can be posited that benefit change
encouraged consumers to search and compare prices at the category level under one important
condition of consumers’ recognition of a given change occurred in their preferred brand.
However, in Study 2 and 3 when a new option is introduced which can be under the same

73

preferred brand or new brand, consumers are more error-prone regarding to their price
knowledge.
Managerial Implications
Similar with the findings of Helsen and Schmittlein (1992) and Assuncao and Meyer (1993), the
increasing probability to purchase a product at a discounted price increases negatively influence
the effect of discount regarding on stockpiling behavior. Overall stockpiling is positively related
with discount depth and negatively with discount frequency. The negative direct effect of discount
frequency on stockpiling can be attenuated by offering a deeper discount. Promotion induced
stockpiling can be weakened by increasing discount frequency. These results indicate that
developing a plan for sales promotions (i.e., price discounts) is inherently complex firm activity.
A well-crafted discount plan will allow a firm to achieve its marketing objectives (e.g., market
share) by increasing the sales promotion efficiency as well as a tactical tool to compete. For
example, a firm may be want to put non stockpiling-inducing discount rate earlier and offer a
discount rated in which it triggers consumers’ stockpiling followed by the previous level discount
offers. In other words, by managing tactical pricing practices, firms could have a better control on
timing of stockpiling and more accurate forecasts for future sales as well as profits.
At first blush, discount depth and frequency seems to be sufficient elements of firm’s sales
promotion action plan. However, this dissertation unveiled the greater complexity by proposing
three moderators. According to Rao (1984), they are elements which all are closely related to firm
pricing. A change in product size is one of common pricing practices used by a firm in order to
keep the specific level of its profit margin. If the change made is beyond consumers’ threshold of
just noticeable difference, its established discount plan should adjusted accordingly. Furthermore,
unlike situations where product size was changed and a new product is introduced, a new brand

74

introduction amplified established direct effects of discount depth and frequency. Industries where
consumers are price sensitive and not loyal to a specific brand, how a firm plans its short- and
long-term sales promotions would determine its market share and sales revenue.
This study also sheds some light on capturing how consumers develop price knowledge.
Specifically, the study argues that, rather than a dichotomous investigation of price knowledge
(i.e., short-term and long-term), firms need to understand how consumers’ price search and
knowledge change over time and across pricing practices by measuring price recall and price
comparison on a regular basis to better understand price knowledge evolution. Because
understanding consumers’ price knowledge as it evolves over time is a significant criterion for
firms’ pricing decision making (Vanhuele and Dreze 2002) in terms of the effectiveness of using
price cues (e.g., Anderson, Cho, Harlam, and Simester 2008; Anderson and Simester 1998) and
also managers generally overestimate the proportion of consumers who are conscious of and
accurately aware of price (e.g., Urbany, Dickson, and Key 1990; Urbany, Dickson, and Sawyer
2000). Results here show that consumers’ price search behavior and, in turn, price knowledge are
influenced by types of a sales promotion available at the moment of purchase. Thus if managers
better understand how consumers’ price knowledge evolve by regularly collecting consumers’
price recall error and price comparison error, they will be able to increase the effectiveness of
price cues.
Theoretical Implications
This dissertation extends our theoretical understanding about the effects of sales promotions and
consumers’ stockpiling behavior. It presents what types of a sales promotion have an influence on
consumers’ promotion induced stockpiling and what are mediating mechanisms for the
relationship between a sales promotion and stockpiling behavior (Study 1). By proposing three

75

moderators, it also examines how such relationships would vary (Study 2, 3, and 4). Additional
objectives of the current dissertation is that how consumes’ price search behavior and knowledge
is influenced by different types of a sales promotion. Lastly even though all of proposed mediating
mechanisms are found insignificant, this dissertation explicitly addresses theoretical and
managerial importance of our understanding of underlying psychological process.
Even though the current four experiments are designed to capture a long-term as well as a
short-term promotional effects because a carryover effect from past promotional exposures is
expected and influential on consumers’ current purchase behavior, there are clear limitations. Thus,
the first limitation appears because shopping tasks are computer simulated and simplified ones.
The current results of theory testing need to be tested by using scanner data and/or in a more
realistic retail environment. Second limitation is from the assumption that there is little competitive
reaction from competitors. By using participants’ own preferred brand as a focal brand and
hypothetical brands as alternative options and previous literature finds mixed results on
competitors’ responses (e.g., Ailawadi, Kopalle, and Neslin 2005), it is plausible that competitors
could react differently across types of sales promotion employed and their reactions could have
some influences on consumers’ stockpiling.
There are some important routes to extend this work. First, how the effects of discount
depth and frequency on stockpiling vary according to brand and category traits. Second, although
this dissertation only focuses on a price discount and not more on the other options such as
coupons, rebates, etc., it might be worthwhile to investigate whether there are any synergy
effects when they are combined together. Also, a follow-up question could be “what would be an
optimal combination to trigger consumers’ stockpiling behavior?” Third, future empirical
research should examine competitors’ response to types of price discounts and the effects of

76

competitive reactions to stockpiling behavior. Fourth, future research could compare established
effects from four experiments to ones when various sizes of the brand are available and/or more
brand alternatives are available. Lastly, future research could bring other consumers’
characteristics into the proposed model to examine whether there are significant differences
regarding to the effects of a sales promotion on stockpiling.

77

APPENDICES

78

APPENDIX A COMPLETE SCENARIOS: STUDY 1
Introduction
Before proceeding to the next page, please carefully read this instruction, From now you will be
asked to purchase “ground coffee” for your household.
You need to decide (1) which Brand you will purchase and (2) how many packages of the brand
your will purchase. Note that you should choose only one brand and are allowed to buy multiple
packages of your brand choice. Lastly, please do this hypothetical shopping as you usually do in
a retail store you often go.
Low Depth and Low Frequency
Shopping #1
You just came to a grocery store where you usually purchase groceries. Today there are two
available “ground coffee” brands which are <Preferred brand name> (12 oz) and Café Value (12
oz). You know you have <Preferred brand name> of coffee at home now. Please decide which
brand and how many packages of the brand you will purchase.
Figure 6–1: Price Tags of Shopping Task #1: Low DEP and Low FRQ
Preferred brand name

Café Value

Shopping #2
YOUR NEXT SHOPPING TRIP TO THIS STORE:
In this shopping trip, "ground coffee" is on your shopping list, because <Preferred brand name>
is left at home. Today there are two available brands which are <Preferred brand name> (12 oz)
and Café Value (12 oz). You recall that you have never seen <Preferred brand name> on sale in
the past. Please decide which brand and how many packages of the brand you will purchase.

79

Figure 6–2: Price Tags of Shopping Task #2: Low DEP and Low FRQ
Preferred brand name

Café Value

High Depth and High Frequency
Shopping #1
You just came to a grocery store where you usually purchase groceries. Today there are two
available “ground coffee” brands which are <Preferred brand name> (12 oz) and Café Value (12
oz). You know you have <Preferred brand name> of coffee at home now. Please decide which
brand and how many packages of the brand you will purchase.
Figure 6–3: Price Tags of Shopping Task #1: High DEP and High FRQ
Preferred brand name

Café Value

80

Shopping #2
YOUR NEXT SHOPPING TRIP TO THIS STORE:
In this shopping trip, "ground coffee" is on your shopping list,
because ${q://QID2371/ChoiceGroup/SelectedChoices} is left at home. Today there are two
available brands which are <Preferred brand name> (12 oz) and Café Value (12 oz). You recall
that, in your last three shopping trips you made, <Preferred brand name> was on sale. Please
decide which brand and how many packages of the brand you will purchase.
Figure 6–4: Price Tags of Shopping Task #2: High DEP and High FRQ
Preferred brand name

Café Value

High Depth and Low Frequency
Shopping #1
You just came to a grocery store where you usually purchase groceries. Today there are two
available “ground coffee” brands which are <Preferred brand name> (12 oz) and Café Value (12
oz). You know you have <Preferred brand name> of coffee at home now. Please decide which
brand and how many packages of the brand you will purchase.
Figure 6–5: Price Tags of Shopping Task #1: High DEP and Low FRQ
Preferred brand name

Café Value

81

Shopping #2
YOUR NEXT SHOPPING TRIP TO THIS STORE:
In this shopping trip, "ground coffee" is on your shopping list, because <Preferred brand
name> is left at home. Today there are two available brands which
are ${q://QID2183/ChoiceTextEntryValue} (12 oz) and Café Value (12 oz). You recall that you
have never seen <Preferred brand name> on sale in the past. Please decide which brand and
how many packages of the brand you will purchase.
Figure 6–6: Price Tags of Shopping Task #2: High DEP and Low FRQ
Preferred brand name

Café Value

Low Depth and High Frequency
Shopping #1
You just came to a grocery store where you usually purchase groceries. Today there are two
available “ground coffee” brands which are <Preferred brand name> (12 oz) and Café Value (12
oz). You know you have <Preferred brand name> of coffee at home now. Please decide which
brand and how many packages of the brand you will purchase.
Figure 6–7: Price Tags of Shopping Task #1: Low DEP and High FRQ
Preferred brand name

Café Value

82

Shopping #2
YOUR NEXT SHOPPING TRIP TO THIS STORE:
In this shopping trip, "ground coffee" is on your shopping list, because <Preferred brand name>
is left at home. Today there are two available brands which are <Preferred brand name> (12 oz)
and Café Value (12 oz). You recall that, in your last three shopping trips you made, <Preferred
brand name> was on sale. Please decide which brand and how many packages of the brand you
will purchase.
Figure 6–8: Price Tags of Shopping Task #2: Low DEP and High FRQ
Preferred brand name

Café Value

83

APPENDIX B COMPLETE SCENARIOS: STUDY 2
Other than purchase options of Shopping #2, all other parts are the same with Study 1
Low Benefit, High Depth, and High/Low Frequency
Shopping #2
Figure 7–1: Price Tags of Shopping Task #2: Low BEF, High DEP, and High/Low FRQ
Preferred brand name

Café Value

Low Benefit, Low Depth, and High/Low Frequency
Shopping #2
Figure 7–2: Price Tags of Shopping Task #2: Low BEF, Low DEP, and High/Low FRQ
Preferred brand name

Café Value

84

High Benefit, Low Depth, and High/Low Frequency
Shopping #2
Figure 7–3: Price Tags of Shopping Task #2: High BEF, Low DEP, and High/Low FRQ
Preferred brand name

Café Value

High Benefit, High Depth, and High/Low Frequency
Shopping #2
Figure 7–4: Price Tags of Shopping Task #2: High BEF, High DEP, and High/Low FRQ
Preferred brand name

Café Value

85

APPENDIX C COMPLETE SCENARIOS: STUDY 3
Other than purchase options of Shopping #2, all other parts are the same with Study 1
New Product, Low Depth, and High/Low Frequency
Shopping #2
Figure 8–1: Price Tags of Shopping Task #2: NP, Low DEP, and High/Low FRQ
Preferred brand name

Preferred brand name:
NEW!! Ready to Brew Instant Pack

Café Value

86

New Product, High Depth, and High/Low Frequency
Shopping #2
Figure 8–2: Price Tags of Shopping Task #2: NP, High DEP, and High/Low FRQ
Preferred brand name

Preferred brand name:
NEW!! Ready to Brew Instant Pack

Café Value

87

APPENDIX D COMPLETE SCENARIOS: STUDY 4
Other than purchase options of Shopping #2, all other parts are the same with Study 1
New Brand, High Depth, and High/Low Frequency
Shopping #2
Figure 9–1: Price Tags of Shopping Task #2: NB, High DEP, and High/Low FRQ
Preferred brand name

Venezia Espresso

Café Value

88

New Brand, Low Depth, and High/Low Frequency
Shopping #2
Figure 9–2: Price Tags of Shopping Task #2: NB, Low DEP, and High/Low FRQ
Preferred brand name

Venezia Espresso

Café Value

89

REFERENCES

90

REFERENCES

Addelman, Sidney (1962) “Orthogonal Main Effects Plans for Asymmetrical Factorial
Experiments,” Technometrics, 4 (1), 21–46.
Ailawadi, Kusum L., Bari A. Harlam, Jacques Cesar, and David Trounce (2006), “Promotion
Profitability for a Retailer: The Role of Promotion, Brand, Category, and Store
Characteristics”, Journal of Marketing Research, 43 (November), 518–535.
______, Karen Gedenk, Christian Lutzky, and Scott A. Neslin (2007), “Decomposition of the Sales
Impact of Promotion-Induced Stockpiling”, Journal of Marketing Research, 44 (August),
450–467.
______, Praveen K. Kopalle, and Scott A. Neslin (2005), “Predicting Competitive Response to a
Major Policy Change: Combining Game-Theoretic and Empirical Analysis,” Marketing
Science, 24 (February), 12–24.
Anderson, Eric T. and Duncan I. Simester (1998), “The Role of Sale Signs,” Marketing Science,
17 (2), 121–142.
______ and ______ (2009), “Price Cues and Customer Price Knowledge,” in Handbook of Pricing
Research in Marketing, Vithala R. Rao, eds. Northampton, Edward Elgar, 150– 65.
______, Edward Ku Cho, Bari Harlam, and Duncan I. Simester (2008), How Does Consumers’
Knowledge of Prices Affect the Demand–Response to Price Changes and Price Cues,
Cambridge: Marketing Science Institute, 9–211.
Ashcraft, Mark H. (1995), “Cognitive Psychology and Simple Arithmetic: A Review and Summary
of New Directions,” Mathematical Cognition, 1 (1), 3–34.
Assuncao, Joao and Robert J. Meyer (1993), "The Rational Effect of Price Promotions on Sales
and Consumption," Management Science, 39 (5), 517–35.
Atkinson, R. C. and J. F. Juola (1973), “Factors Influencing Speed and Accuracy of Word
Recognition,” in Attention and Performance, Vol 4, Sylvan Kornblum, ed. New York,
Academic Press, 583–612.
Bartlett, Frederic C. (1932), Remembering: A Study in Experimental and Social Psychology.
Cambridge: Cambridge University Press.
Bell, David R., Ganesh Iyer, and V. Padmanabhan (2002), “Price Competition under Stockpiling
and Flexible Consumption,” Journal of Marketing Research, 39 (3), 292–303.
______, Jeongwen Chiang, and V. Padmanabhan (1999), “The Decomposition of Promotional

91

Response: An Empirical Generalization,” Marketing Science, 18 (4), 504–526.
Bennett, Peter D. and Robert M. Mandell (1969), “Prepurchase Information Seeking Behavior of
New Car Purchasers: The Learning Hypothesis,” Journal of Marketing Research, 6 (4),
430–433.
Blattberg, Robert C., Gary D. Eppen, and Joshua Liberman (1981), “A Theoretical and Empirical
Evaluation of Price Deals for Consumer Nondurables,” Journal of Marketing, 45(1), 116–
129.
______ and Scott A. Neslin (1990), Sales Promotion: Concepts, Methods, and Strategies.
Englewood Cliffs, NJ: Prentice Hall.
Brucks, Merrie (1985), “The Effects of Product Class Knowledge on Information Search Behavior,”
Journal of Consumer Research, 12 (1), 1–16.
Bucklin, Randolph E. and Sunil Gupta (1992), “Brand Choice, Purchase Incidence, and
Segmentation: An Integrated Modeling Approach,” Journal of Marketing Research, 29
(May), 201–215.
______, ______, and S. Siddarth (1998), “Determining Segmentation in Sales Response Across
Consumer Purchase Behaviors,” Journal of Marketing Research, 35 (May), 189–97.
Buhrmester, Michael, Tracy Kwang, and Samuel D. Gosling (2011), “Amazon’s Mechanical Turk:
A New Source of Cheap, yet High-Quality, Data?,” Perspectives on Psychological Science,
6 (1), 3–5.
Carlson, John H. and R. Preston McAfee (1983), “Discrete Equilibrium Price Dispersion,” Journal
of Political Economy, 91 (3), 480–493.
Chan, Tat, Chakravarthi Narasimhan, and Qin Zhang (2008), “Decomposing Promotional Effects
with a Dynamic Structural Model of Flexible Consumption,” Journal of Marketing
Research, 45 (August), 487–498.
Chintagunta, Pradeep K. (1993), “Investigating Purchase Inci- dence, Brand Choice, and Purchase
Quantity Decisions of Households,” Marketing Science, 12 (Spring), 184–208.
______ and Sudeep Haldar (1998), “Investigating Purchase Timing Behavior in Two Related
Product Categories,” Journal of Marketing Research, 35 (February), 43–53.
Clifford, Stephanie and Catherine Rampell (2011), Food Inflation Kept Hidden in Tinier Bags”
The New York Times, March 28.
Dahlby, Bev and Douglas S. West (1986), “Price Dispersion in An Automobile Insurance Market,”
Journal of Political Economy, 94 (2), 418–438.

92

Dehaene, Stanislas (1992), “Varieties of Numerical Abilities,” Cognition, 44 (1), 1–42.
______ and Rokny Akhavein (1995), “Attention, Automaticity, and Levels of Representation in
Number Processing,” Journal of Experimental Psychology, 21 (2), 314–326.
Dellaert, Benedict G.C., Vladislav Y. Golounov, and Jaideep Prabhu (2005), “The Impact of Price
Disclosure on Dynamic Shopping Decisions,” Marketing Letters, 16 (1), 37–52.
DelVecchio, Devon, H. Shanker Krishnan, and Daniel C. Smith (2007), “Cents or Percent? The
Effects of Promotion Framing on Price Expectations and Choice,” Journal of Marketing,
71 (3), 158–170.
Dickson, Peter R. and Alan G. Sawyer (1990), “The Price Knowledge and Search of Supermarket
Shoppers,” Journal of Marketing, 54 (3), 42–53.
Edwards, Allen L. (1948), “Note on the "Correction of Continuity" in Testing the Significance of
the Difference between Correlated Proportions,” Psychometrika, 13 (3), 185–187.
Erdem, Tülin, Michael P. Keane, and Baohong Sun (2008), “A Dynamic Model of Brand Choice
When Price and Advertising Signal Product Quality,” Marketing Science, 27 (6), 1111–
1125.
Foekens, Eijte W., Peter S.H. Leeflang, and Dick R. Wittink (1999), “Varying Parameter Models
to Accommodate Dynamic Promotion Effects,” Journal of Econometrics, 89 (1-2), 249–
68.
Fornell, Cleas and David F. Larcker (1981), “Evaluating structural equation models with
unobserved variables and measurement error”, Journal of Marketing Research, 18 (1), 39–
50.
Fox, Edwards J. and Stephen J. Hoch (2005), “Cheery-Picking,” Journal of Marketing, 69 (1), 46–
62.
Gardiner, John M. (1988), “Functional Aspects of Recollective Experience,” Memory and
Cognition, 16 (4), 309–313.
Gauri, Dinesh K., K. Sudhir, and Debabrata Talukdar (2007), “The Temporal and Spatial
Dimensions of Price Search: Insights From Matching Hou sehold Survey and Purchase
Data,” Journal of Marketing Research, 45 (2), 226–240.
Gonul, Fusun and Kannan Srinivasan (1996), “Estimating the Impact of Consumer Expectations
of Coupons on Purchase Behavior: A Dynamic Structural Model,” Marketing Science, 15
(3), 262-79.
Goodman, Joseph K., Cynthia E. Cryder, and Amar Gheema (2013), “Data Collection in a Flat
World: The Strengths and Weaknesses of Mechanical Turk Samples,” Journal of
Behavioral Decision Making, 26 (3), 21–224.

93

Grewal, Dhruv, Kent B. Monroe, and R. Krishnan (1998), “The Effects of Price-Comparison
Advertising on Buyers’ Perceptions of Acquisition Value, Transaction Value, and
Behavioral Intentions,” Journal of Marketing, 62 (2), 46–59.
Hardie, Bruce G. S., Eric J. Johnson, and Peter S. Fader (1993), “Modeling of Loss Aversion
Reference Dependence Effects on Brand Choice,” Marketing Science, 12 (4), 378–394.
Hauser, John R. and Glen Urban (1986), “The Value Priority Hypotheses for Consumer Budget
Plan,” Journal of Consumer Research, 12 (4), 446–462.
______ and Steven M. Shugan (1983), “Defensive Marketing Strategies,” Marketing Science, 2
(4), 319–360.
Hayes, Andrew F. and Kristopher J. Preacher (2014), “Statistical Mediation Analysis with a
Multicategorical Independent Variable,” British Journal of Mathematical and Statistical
Psychology, 67 (3), 451 – 470.
Heilman, M. Carrie, Kent Nakamoto, and Ambar G. Rao (2002), “Pleasant Surprises: Consumer
Response to Unexpected In- Store Coupons,” Journal of Marketing Research, 34 (May),
242–52.
Helsen, Kristiaan and David C. Schmittlein (1992), “Some Characterizations of Stockpiling
Behavior Under Uncertainty,” Marketing Letters, 3 (1), 5 – 17.
Hoch, Stephen J., Byung-Do Kim, Alan L. Montgomery, and Peter E. Rossi (1995), “Determinants
of Store-Level Price Elasticity,” Journal of Marketing Research, 32 (February), 17-29.
Howard, John A. and Jagdish N. Sheth (1967), “Theory of Buyer Behavior,” in Changing
Marketing Systems: Corporate and Government Interfaces, Reed Moyer, ed. Chicago,
American Marketing Associations, 253–262.
Hu, Li-tze. and Peter M. Bentler (1999), “Cutoff Criteria for Fit Indexes in Covariance Structure
Analysis: Conventional Criteria versus New Alternatives,” Structural Equation Modeling,
6 (1), 1–55.
Kopalle, Praveen K., Carl F. Mela, and Lawrence Marsh (1999), "The Dynamic Effect of
Discounting on Sales: Empirical Analysis and Normative Pricing Implications," Marketing
Science, 18 (3), 317–32.
Jacoby, Jacob, Jerry C. Olson, and Rafael A. Haddock (1971), “Price, Brand Name, and Product
Composition Characteristics as Determinants of Perceived Quality,” Journal of Applied
Psychology, 55 (6), 570–579.
Jacobson, Robert and Carl Obermiller (1990), “The Formation of Expected Future Price: A
Reference Price for Forward-Looking Consumers,” Journal of Consumer Research, 16

94

(March), 420–432.
Jacoby, Larry L. (1991), “A Process Dissociation Framework Separating Automatic from
Intentional Uses of Memory,” Journal of Memory and Language, 30 (5), 513–541.
______ (1994), “Measuring Recollection: Strategic vs. Automatic Influences of Associative
context,” in Attention and Performance, Vol. 15, Carlo Umilta and Morris Moscovitch, eds.
Cambridge, Bradford, 661–679.
______ (1998), “Invariance in Automatic Influences of Memory: Towards a User’s Guide for The
Process-Dissociation Procedure,” Journal of Experimental Psychology Learning, Memory,
and Cognition, 24 (1), 3–26.
______ and Kevin Whitehouse (1989), “An Illusion of Memory: False Recognition Influenced by
Unconscious Perception,” Journal of Experimental Psychology: General, 118 (2), 126–
135.
______ and Dallas Mark (1981), “On the Relationship between Autobiographical Memory and
Perceptual Learning,” Journal of Experimental Psychology: General, 110 (3), 306–340.
Janiszewski, Chris and Donald R. Lichtenstein (1999), “A Range Theory Account of Price
Perception,” Journal of Consumer Research, 25 (4), 353–368.
Kalyanaram, Gurumurthy and John D. C. Little (1994), “An Empirical Analysis of Latitude of
Price Acceptance in Consumer Package Goods,” Journal of Consumer Research, 21 (3),
408–418.
______ and Russell S. Winer (1995), “Empirical Generalizations from Reference Price Research,”
Marketing Science, 14 (3, Pt.2), G161–G169.
Kalwani, Manohar U. and Chi Kin Yim (1992), “Consumer Price and Promotion Expectations: An
Experimental Study,” Journal of Marketing Research, 29 (1), 90–100.
Keane, Michael P. (1997), “Current Issues in Discrete Choice Modeling,” Marketing Letters, 8 (3),
307–333.
Koriat, Asher and Israel Lieblich (1977), “A Study of Memory Pointers,” Acta Psychologica, 41
(2-3), 151–164.
______ and Ravit Levy-Sadot (2001), “The Combined Contributions of the Cue-Familiarity and
Accessibility Heuristics to Feelings of Knowing,” Journal of Experimental Psychology:
Learning, Memory, and Cognition, 27 (1), 34–53.
Krishna, Aradhna (1991), “Effect of Dealing Patterns on Consumer Perceptions of Deal Frequency
and Willingness to Pay,” Journal of Marketing Research, 28 (4), 441–451.

95

______ (1992), "The Normative Impact of Consumer Price Expectations for Multiple Brands on
Consumer Purchase Behavior," Marketing Science, 11 (3), 266–86.
______ (1994), “The Impact of Dealing Patterns on Purchase Behavior” Marketing Science, 13
(4), 351–373.
Lapersonne, Eric, Gilles Laurent, and Jean-Jacques LeGoff (1995), “Consideration Sets of Size
One: An Empirical Investigation of Automobile Purchases,” International Journal of
Research in Marketing, 12 (1), 55–66.
Lastovicka, John L. (1979), “Questioning the Concept of Involvement Defined Product Classes,”
in NA - Advances in Consumer Research, Vol. 6, William L. Wilkie, ed. Ann Arbor.
Association for Consumer Research, 174–179.
Leboe, Jason P. and Bruce W. A. Whittlesea (2002), “The Inferential Basis of Familiarity and
Recall: Evidence for a Common Underlying Process,” Journal of Memory and Language,
46 (4), 804–829.
Lehmann, Donald R., William L. Moore, and Terry Elrod (1982), “The Development of Distinct
Choice Process Segments Over Time: A Stochastic Modeling Approach,” Journal of
Marketing, 46 (2), 48–59.
Lichtenstein, Donald R., Nancy M. Ridgway, and Richard G. Netemeyer (1993), “Price
Perceptions and Consumer Shopping Behavior: A Field Study,” Journal of Marketing
Research, 30 (2), 234–245.
Lichtenstein, Sarah and Baruch Fishhoff (1977), “Do Those Who Know More Also Know More
about How Much They Know?” Organizational Behavior and Human Performance, 20 (2),
159–183.
Lindsay, D. Stephen and Colleen M. Kelley (1996), “Creating Illusions of Familiarity in a Cued
Recall Remember/Know Paradigm,” Journal of Memory and Language, 35 (2), 197–211.
Litvack, David S., Roger J. Calantone, and Paul R. Warshaw (1985), “An Examination of ShortTerm Retail Grocery Price Effects,” Journal of Retailing, 61 (3), 9 – 25.
Loftus, Elizabeth F., David G. Miller, and Helen J. Burns (1978), “Semantic Integration of Verbal
Information into a Visual Memory,” Journal of Experimental Psychology: Human
Learning and Memory, 4 (1), 19–31.
______ and John C. Palmer (1974), “Reconstruction of Automobile Destruction: An Example of
the Interaction between Language and Memory,” Journal of Verbal Learning and VerbalBehavior, 13 (5), 585–589.
Louviere, Jordan J. (1988), Analyzing Decision Making: Metric Conjoint Analysis, Beverly Hills:
Sage.

96

Macé, Sandrine and Scott A. Neslin (2004), “The Determinants of Pre- and Postpromotion Dips
in Sales of Frequently Purchased Goods,” Journal of Marketing Research, 41 (August),
339–50.
Mandler, George (1979), “Organization and Repetition: Organizational Principles with Special
Reference to Rote Learning,” in Perspectives on Memory Research, L. G. Nilsson, ed. New
York, Academic Press, 293–327.
______ (1980), “Recognizing: The Judgment of Previous Occurrence,” Psychological Review, 87
(3), 252–271.
______ (1991), “Your Face Looks Familiar But I Can’t Remember Your Name: Your Name: A
Review of Dual Process Theory,” in Relating Theory and Data: Essays on Human Memory
in Honor of Bennett B. Murdock, William. E. Hockley and Stephan Lewandowsky, eds.
Hillsdale, Erlbaum, 207–225.
Lodish, Leonard M. (1986), The Advertising and Promotion Chal lenge: Vaguely Right or
Precisely Wrong. New York: Oxford University Press.
Marcel, Anthony J. (1983), “Conscious and Unconscious Perception: An Approach to the
Relations between Phenomenal Experience and Perceptual Processes,” Cognitive
Psychology, 15 (2), 238–300.
Markman, Arthur (1999), Knowledge Representation, Mahwah: Lawrence Erlbaum.
Mazumdar, Tridib, S.P. Raj, and Indrajit Sinha (2005), “Reference Price Research: Review and
Propositions,” Journal of Marketing, 69 (4), 84–102.
McNemar, Quinn (1947), “Note on the Sampling Error of the Difference between Correlated
Proportions or Percentages,” Psychometrika, 12 (2), 153–157.
Meyer, Robert J. and Joao Assuncao (1990), “The Optimality of Consumer Stockpiling Strategies,”
Marketing Science, 9 (1), 18–41.
Mela, Carl F., Kamel Jedidi, and Douglas Bowman (1998), “The Long-Term Impact of Promotions
on Consumer Stockpiling Behavior,” Journal of Marketing Research, 35 (2), 250–262.
Monroe, Kent B. (2003), Pricing: Making Profitable Decisions, New York: McGraw-Hill, Irwin.
______ and Angela Y. Lee (1999), “Remembering Versus Knowing: Issues in Buyers’ Processing
of Price Information,” Journal of the Academy of Marketing Science, 27 (2), 207–225.
Moorthy, Sridhar, Brian T. Ratchford, and Debabrata Talukdar (1997), “Consumer Information
Search Revisited: Theory and Empirical Analysis,” Journal of Consumer Research, 23 (4),
263–277.

97

Murthi, B.P.S. and K. Srinivasan (1999), “Consumers’ Extent of Evaluation in Brand Choice,” The
Journal of Business, 72 (2), 229–256.
______ and Ram C. Rao (2012), “Price Awareness and Consumers’ Use of Deals in Brand Choice,”
Journal of Retailing, 88 (1), 34–46.
Neslin, Scott A. (2002), Sales Promotion. Cambridge, MA: Marketing Science Institute.
______, Caroline Henderson, and John Quelch. (1985). “Consumer Promotions and the
Acceleration of Product Purchases," Marketing Science 4 (Spring), 147-165.
———, Stephen G. Powell, and Linda Schneider Stone (1995), “The Effects of Retailer and
Consumer Response on Optimal Manufacturer Advertising and Trade Promotion
Strategies,” Management Science, 41 (May), 749–66.
Nhouyvanisvong, Adisack and Lynne M. Reder (1998), “Rapid Feeling-of-Knowing: A Strategy
Selection Mechanism,” in Metacognition: Cognitive and Social Dimensions, Vincent Y.
Yzerbyt, Guy Lories, and Benoit Dardenne, eds. London, Sage, 35–52.
Nijs, Vincent R., Marnik G. Dekimpe, Jan-Benedict E.M. Steenkamp, and Dominque M. Hanssens
(2001), “The Category- Demand Effects of Price Promotions,” Marketing Science, 20 (1),
1–22.
Olson, Jerry Corrie (1977), “Price as an Informational Cue: Effects in Product Evaluation,” in
Consumer and Industrial Buying Behavior, Arch G. Woodside, Jagdish N. Sheth, and Peter
D. Bennet, eds. New York, North Holland Publishing Company, 267–286.
Paolacci, Gabriele, Jesse Chandler, and Panagiotis G. Ipeirotis (2010), “Running Experiments on
Amazon Mechanical Turk,” Judgment and Decision Making, 5 (5), 411–419.
Park, C. Whan and V. Parker Lessig (1981), “Familiarity and its Impact on Consumer Decision
Biases and Heuristics,” Journal of Consumer Research, 8 (2), 223–231.
______ and Frederick W. Winter (1979), “Product Quality Judgment: Information Processing
Approach,” Journal of the Market Research Society, 21 (3), 211–217.
Pauwels, Koen, Dominique M. Hanssens, and S. Siddarth (2002), “The Long-Term Effects of Price
Promotions on Category Incidence, Brand Choice, and Purchase Quantity,” Journal of
Marketing Research, 39 (November), 421–39.
Putler, Daniel S. (1992), “Incorporating Reference Price Effects into A Theory of Consumer
Choice,” Marketing Science, 11 (3), 287–309.
Rajaram, Suparna (1993), “Remembering and Knowing: Two Means of Access to the Personal
Past,” Memory & Cognition, 21 (1), 89–102.

98

Rand, David G. (2011), “The Promise of Mechanical Turk: How Online Labor Markets Can Help
Theorists Run Behavioral Experiments,” Journal of Theoretical Biology, 299 (21), 172–
179.
Rao, Vithala R. (1984), “Pricing Research in Marketing: The State of the Art,” Journal of Business,
57 (1), S39–S60.
Ratchford, Brian T. (2009), “Consumer Search and Pricing” in Handbook of Pricing Research in
Marketing, Vithala R. Rao, ed. Northampton, Edward Elgar, 91–107.
______ and Narasimhan Srinivasan (1993), “An Empirical Investigation of Returns to Search,”
Marketing Science, 12 (1), 73–87.
Reder, Lynne M. and Frank E. Ritter (1992), “What Determines Initial Feeling of Knowing?
Familiarity with Question Terms, not with the Answer,” Journal of Experimental
Psychology: Learning, Memory, and Cognition, 18 (3), 435–451.
______ and Christian D. Schunn (1996), “Metacognition Does Not Imply Awareness: Strategy
Choice Is Governed by Implicit Learning and Memory,” in Implicit Memory and
Metacognition, Lynne M. Reder, ed. Hillsdale, Erlbaum, 45–78.
Roediger, Henry L.H. L., III and Kathleen B. McDermott (1995), “Creating False Memories:
Remembering Words Not Presented in Lists,” Journal of Experimental Psychology:
Learning, Memory, and Cognition, 21 (4), 803–814.
______ and ______ (1996), “False Perceptions of False Memories,” Journal of Experimental
Psychology: Learning, Memory, and Cognition, 22 (3), 814–816.
Sawyer, Alan G. and Peter Dickson (1984), “Psychological Perspectives on Consumer Response
to Sales Promotion,” in Research on Sales Promotion: Collected Papers, Katherine Jocz,
ed. Cambridge, Marketing Science Institute.
Schunn, Christian D., Lynne M. Reder, Adisack Nhouyvanisvong, Daniel R. Richards, and Philip
J. Stroffolino (1997) “To Calculate or Not to Calculate: A Source Activation Confusion
Model of Problem Familiarity's Role in Strategy Selection,” Journal of Experimental
Psychology: Learning, Memory, and Cognition, 23 (1), 3–29.
Sivakumar, K and S. P. Raj (1997), “Quality Tier Competition: How Price Change Influences
Brand Choice and Category Choice,” Journal of Marketing, 61 (3), 71–83.
Stilley, Karen M., J. Jeffrey Inman, and Kirk L. Wakefield (2010), “Spending on the Fly: Mental
Budgets, Promotions, and Spending Behavior,” Journal of Marketing, 74 (May), 34–47.
Stokes, Raymond Curtis. (1985), “The Effect of Price Package Design, and Brand Familiarity on
Perceived Quality,” in Perceived Quality, Jacob Jacoby and Jerry Olson, eds. Lexington,

99

Lexington Press, 233–246.
Sun, Baohong, Scott A. Neslin, and Kannan Srinivasan (2003), “Measuring the Impact of
Promotions on Brand Switching When Consumers Are Forward Looking,” Journal of
Marketing Research, 40 (November), 389–405.
Thomas, Manoj and Vicki Morwitz (2005), “Penny Wise and Pound Foolish: The Left Digit Effect
in Price Cognition,” Journal of Consumer Research, 32 (1), 54–65.
______ and ______ (2009), “Heuristics in Numerical Cognition: Implications for Pricing,” in
Handbook of Pricing Research in Marketing, Vithala R. Rao, ed. Northampton, Edward
Elgar, 132–149.
Tellis, Gerard J. and Birger Wernerfelt (1987), “Competitive Price and Quality under Asymmetric
Information,” Marketing Science, 6 (3). 240–253.
______ and Fred S. Zufryden (1995), “Tackling the Retailer Decision Maze: Which Brands to
Discount, How Much, When, and Why?” Marketing Science, 14 (August), 271–99.
Tulving, Endel, (1985), “Memory and Consciousness,” Canadian Psychologist, 26 (1), 1–12.
Urbany, Joel E., Peter R. Dickson, and Alan G. Sawyer (2000), “Insights into Cross- and WithinStore Price Search: Retailer Estimates vs. Consumer Self-Reports.” Journal of Retailing,
76 (2), 243–258.
______, ______, and Rosemary Kalapurakal (1996), “Price Search in the Retail Grocery Market,”
Journal of Marketing, 60 (2), 91–104.
______, ______, and Rosemary Key (1990), “Actual and Perceived Consumer Vigilance in the
Retail Grocery Industry,” Marketing Letters, 2 (1), 15–25.
Van Heerde, Harald J., Peter S.H. Leeflang, and Dick R. Wittink (2000), “The Estimation of Preand Postpromotion Dips with Store-Level Scanner Data,” Journal of Marketing Research,
37 (August), 383–95.
———, ———, and ——— (2004), “Decomposing the Sales Promotion Bump with Store Data,”
Marketing Science, 23 (August), 317–34.
Vanhuele, Marc and Xavier Dreze (2002), “Measuring the Price Knowledge Shoppers Bring to the
Store,” Journal of Marketing, 66 (4), 72–85.
Wansink, Brian (1994), “Advertising’s Impact on Category Substitution,” Journal of Marketing
Research, 31 (November), 505–515.
———, Robert J. Kent, and Stephen J. Hoch (1998), “An Anchoring and Adjustment Model of
Purchase Quantity Decisions,” Journal of Marketing Research, 35 (February), 71–81.

100

Wakefield, Kirk L. and J. Jeffrey Inman (1993), “Who Are the Price Vigilantes? An Investigation
of Differentiating Characteristics Influencing Price Information Processing,” Journal of
Retailing, 69 (2), 216–233.
Wathieu, Luc and Marco Bertini (2007), “Price as a Stimulus to Think: The Case for Willful
Overpricing,” Marketing Science, 26 (1), 118–129.
Whittlesea, Bruce W. A. (1993), “Illusions of Familiarity,” Journal of Experimental Psychology:
Learning, Memory, and Cognition, 19 (6), 1235–1253.
______ (1997), “Production, Evaluation, and Preservation of Experiences: Constructive
Processing in Remembering and Performance Tasks,” in The Psychology of Learning and
Motivation: Advances in Research and Theory, Vol. 37, Douglas L. Medin, ed. San Diego,
Academic Press, 211–264.
______ and Jason P. Leboe (2000), “The Heuristic Basis of Remembering and Classification:
Fluency, Generation, and Resemblance,” Journal of Experimental Psychology: General,
129 (1), 84–106.
______ and Lisa D. Williams (2001), “The Discrepancy-Attribution Hypothesis: II. Expectation,
Uncertainty, Surprise, and Feelings of Familiarity,” Journal of Experimental Psychology:
Learning, Memory, and Cognition, 27 (1), 14–33.
Zeithaml, Valarie A. (1982), “Consumer Response to In-Store Price Information Environments,”
Journal of Consumer Research, 8 (4), 357–369.
______ (1988), “Consumer Perceptions of Price, Quality, and Value: A Means-End Model and
Synthesis of Evidence,” Journal of Marketing, 52 (3), 2–22.

101