Name-Letters and Birthday-Numbers: Implicit Egotism Effects in Pricing
Name-Letters and Birthday-Numbers: Implicit Egotism Effects in Pricing
Name-Letters and Birthday-Numbers: Implicit Egotism Effects in Pricing
net/publication/280311630
CITATIONS READS
14 1,568
2 authors:
Some of the authors of this publication are also working on these related projects:
Reframing the Discount with a Comparison to the Sale Price: Does It Make the Discount More Attractive? View project
All content following this page was uploaded by Keith Coulter on 24 July 2015.
Keith S. Coulter
Dhruv Grewal
The authors appreciate the feedback of Rajneesh Suri, Scott Motyka and Maastricht University
seminar participants.
Name-Letter and Birthday-Number
Implicit Egotism Effects in Pricing
Abstract
This research examines how the implicit egotism resulting from consumers’ positive self-
associations affects their evaluations of product prices. The effects can occur when the product’s
price and the consumer share either name letters (name-letter/price effect), or birthday numbers
(birthday-number/price effect).Through a series of studies, we demonstrate that the positive
affect linked to name-letters and birthday-numbers transfers directly to consumers’ price
predilections, and ultimately affects their purchase intentions. More specifically, consumers like
prices (e.g., “fifty-five dollars”) that contain digits beginning with the same first letter (e.g., “F”)
as their own name (e.g., “Fred” or “Mr. Frank”) more than prices that do not. Similarly, prices
containing cents digits (e.g., $49.15) that correspond to a consumer’s date of birth (e.g., April 26)
also enhance pricing liking and purchase intentions. Across groups of consumers, our findings
demonstrate that implicit egotism effects can result in greater purchase intentions for a higher
priced product compared to a lower-priced product.
personalization programs, in which product and price offerings are tailored toward the individual
consumer. In such programs, each customer is treated as “relevant” and unique, and that personal
term “implicit egotism” refers to the fact that people have positive associations about themselves,
and these associations can non-consciously spill over into their evaluations of objects that they
associate with the self. The “spill-over” can only occur, however, if implicit egotism cues are
provided that allows the consumer to make the self/object connection (Greenwald and Banaji
1995; Nuttin 1987). Two powerful implicit egotism cues that have not been examined within the
marketing literature are (1) the match between the first letter of a customer’s first (or last) name
and the first letters of the digits in a price-name, and (2) the match between a customer’s birthday
As an example of the first cue, we show that Fred (or Mr. Franklin) is more likely to
purchase a brand of coffee at a sale price of $8.55 than at a sale price of $8.66, whereas Sally (or
Ms. Smith) is more likely to purchase that same brand of coffee at a sale price of $8.66. For Fred
(or Mr. Franklin) and Sally (or Ms. Smith), the first letters of the cents digits (i.e., “fifty-five”
and “sixty-six,” respectively) directly correspond to the first letters (i.e., F and S) of their names.
Regarding the second cue, we demonstrate that Arthur (born on January 15), might be more
likely to purchase a hamburger if it is priced at $4.15 than if it is priced at $4.10, whereas Betty
(born on February 10) might be more likely to purchase that hamburger priced at $4.10.
investigated in the context of brand names (Brendl, Chattopadhyay, Pelham, and Carvallo 2005).
For example, research has shown that Tom (or Mr. Thomas) is more likely to buy a Toyota,
2
Linda (or Mrs. Lundy) is more likely to purchase a Lexus, and Henry (born on January 5) might
prefer “Five Guys” brand chilidogs over those of other restaurant chains. However, the practical
significance of these findings is limited in that marketers typically have little ability to alter
Thus, if these same implicit egotism effects were found to apply to (i.e., or could be evoked by)
Attribute-specific egotism effects have been hypothesized (Brendl et al 2005), but never
conclusively demonstrated. In this paper, we demonstrate such findings regarding the price
attribute.
We examine our expectations and hypotheses in the context of two experiments, two
quasi-experiments, and one field study. In our first study, we examine “T” and “E” name-letter
effects associated with two prices ($622 or $688) presented in verbal format. The name letters
correspond to participants’ last names. We find that name-letter/price matching enhances both
price liking and purchase intentions. In our second study, price stimuli ($8.55 or $8.66) are
presented as Arabic numerals in visual format, and we examine the effects associated with
matching the (last) name-letters “F” and “S.” We also examine the moderating role of price
rehearsal, which allows us to confirm process. Our third study expands the generalizability of our
findings by demonstrating several different (first) name-letter effects across a range of prices in a
field setting (i.e., involving retail automobile purchases). In this study, we are able to examine
In our fourth (experimental) study, we demonstrate that implicit egotism can also
manifest in the form of birthday-number/price matching effects, and that ego threat moderates
(i.e., enhances) these effects. Finally in a fifth (experimental) study, we examine birthday-
3
different process mediators, we are able to show that these two matching strategies impact
purchase intentions not only in terms of their effects on attribute liking, but also in terms of their
more general effects on ad evaluations. In Figure 1, we present our organizing framework, and
In summary, this research makes several important contributions. It is the first series of
studies to: 1) conclusively demonstrate that implicit egotism affects product attributes, 2)
investigate the price attribute in terms of both name-letter and birthday-number matching effects,
format (i.e., visual versus verbal) to numerical processing mode, 4) show that birthday-number
and name-letter effects can operate independently, but concurrently, within the same advertising
purchase intentions) relationship, and 6) demonstrate that the implicit ego-driven affect elicited
by either name-letter or birthday-number matching can transfer not only to the associated price
THEORETICAL DEVELOPMENT
As noted earlier, theories of implicit egotism suggest that people’s positive associations
about themselves spill over into their evaluations of objects associated with the self (Greenwald
and Banaji 1995; Pelham et al. 2002, Perham, Carvallo, and Jones 2005). Thus, people are
attracted to cities, states, careers, or brands that either consciously or non-consciously remind
them of themselves (Pelham et al. 2003, p.800). Because people prefer letters contained in their
4
own names more than other letters (e.g., Johnson 1986; Hoorens and Nuttin 1993; Kitayama and
Rarasawa 1997; Jones et al. 2002), and because they are reminded of themselves when they
encounter other people, objects, or events whose name begins with the same letter (or series of
letters) as their own names (e.g., Johnson 1986; Nuttin 1987; Jones et al. 2002), they tend to
express a greater liking of, or predisposition toward, those name-related objects. This positive
The implicit egotism that drives name letter effects has also been shown to create
preference for numbers (and objects associated with those numbers) that reflect one’s date-of-
birth (i.e., both day-of-the-month as well as year; Kitayama and Rarasawa 1997; Koole,
Dijksterhui and van Knoppernberg 2001; Jones et al. 2002; Jones et al. 2004). We refer to this as
the birthday-number effect. A representative review of findings in the literature regarding both
A limitation of some of these studies is that they involve archival data, which cannot
establish causality and increases the likelihood that the results are confounded with a number of
other variables (Gallucci 2003; Simonsohn 2011). In addition, review of Table 1 studies
demonstrates a need to establish these effects in more realistic experimental and field settings,
and to generalize to other factors that managers have greater control over (e.g., attributes such as
Presumably, personal names and dates carry positive affect because they serve to define a
person’s self-concept (Koole and Pelham 2003). Thus if a person and a brand share name letters,
1
With regard to preferences for things that do not have first and last names (e.g., cities, occupations), people’s first
and last names offer interchangeable predictors (Pelham et al. 2002).
5
one possibility is that the valence could transfer directly from the person to the brand (i.e.,
the Dual Mediation model of persuasion (MacKenzie, Lutz, and Belch 1986; Mitchell 1986).
Alternatively (and in concurrence with parallel constraint satisfaction models; Read, Vanman,
and Miller 1997), Brendl et al. (2005) suggest that valence transfer from the name letter to the
brand is attribute-specific, rather than global in nature. They argue that name letters directly
relevant to a current active need, which then drive their overall liking for the brand. The actual
name letters in the attribute may be consciously or non-consciously perceived, but the name-
letter linkage (and the affect transfer) likely occurs below the threshold of conscious awareness.
Importantly, despite Brendl et al.’s (2005) indirect transfer hypothesis, their studies
showed that the preferred brand shared name letters with the personal source, not the brand’s
attribute. As such, the authors could not definitively confirm their arguments (Brendl et al. 2005,
p. 412). If valence transfer does indeed occur through an attribute, we should expect implicit
egotism to be just as likely to influence purchase intentions if the personal source and the
attribute share first letters. Thus, Tom may be more likely to buy a Toyota - but he may also buy
The attribute that we investigate is price. Compared with other attributes, prices are
unique in that they can be expressed as a series of letters (e.g., “eight dollars and twenty-two
cents”) or numeric characters (e.g., $8.22). Thus, focusing on the price attribute enables us to
Name-Letter/Price Effects
Research in the field of numerical cognition suggests that people process numerical
6
information, including prices, using a dedicated cognitive subsystem (Ashcraft 1992; Dehaene
1997), and that numbers can be mentally represented in both auditory and visual formats.2
Numbers (e.g., 22) processed in the auditory (phonological) code are represented by a sequence
of word sounds (e.g., “twent-ē-tü”), whereas numbers processed in the visual code are
represented on a visuo-spatial medium based on their appearance as either Arabic numerals (e.g.,
“22”) or written words (e.g., “twenty-two”) (Coulter, Choi, and Monroe 2012).
Importantly, although research has shown that the different codes prompt processing in
different areas of the brain (Dehaene et al. 1993; Dehaene and Cohen 1995), research has also
demonstrated that the visual, analog, and auditory codes are neurologically connected with one
another - that is, there are paths linking the different representations within and across each of
the brain hemispheres (Noel 2001; Pinel et al. 1999). This means that although a visual pricing
stimulus (e.g., $22) will lead to a price being represented in the corresponding visual code (e.g.,
$22), and a verbal pricing stimulus will lead to a price being represented in the corresponding
auditory code (e.g.,“twent-ē-tü”), either form of encoding can also lead to a price being
2009; Logie, Gillhooly, & Wynn 1994; Coulter and Coulter 2010; Coulter, Choi, and Monroe
2012).
demonstrated that the verbal format associated with a spoken word and the visual format
associated with a written word are closely linked (see Berninger et al. 2006 for a discussion).
Further, written language skills are learned through recognition of sounds as words comprised of
letters (Gillam and Johnston 1992). Thus if a price stimulus is presented and encoded verbally
2
According to Dehaene’s “Triple Code” model, numbers can also be processed and encoded as analog magnitude
representations, which are judgments of relative size (e.g., “large” or “small”) arrayed along a left-to-right-oriented
mental number line (Dehaene, Bossini, & Giraux 1993).
7
(e.g., as “twent-ē-tü”), we expect that a written word representation (e.g., “twenty-two,” rather
than the numeric character representation “22”) will also be evoked. The consumer will then be
able to make the non-conscious connection between his or her own name letters (e.g., “Ted” or
Ms. “Thomas”) and the letters contained in the price (e.g., “Twenty-Two”). We refer to this as
resulting in greater liking for the price and ultimately greater purchase intentions. Importantly,
the first letter of the consumer’s name and the first letter of the price need not represent an exact
phonetic match (e.g., the “E” in “Edith” and the “E” in “Eight” represent different sounds).
Rather it is sufficient that the verbal representation elicit that visual letter in the mind of the
consumer. We expect:
H1 : If prices are presented verbally, consumers will demonstrate greater price liking and
purchase intentions if the price does (vs. does not) share name first-letter with the
consumer.
If a price is not presented verbally, but rather is presented visually as a series of numeric
characters (e.g., $30), visual encoding of those characters is expected to occur. But research has
shown that visual/auditory encoding as a series of words or word-sounds may also occur,
depending on the needs of the task (Dehaene 1992; Gallistel and Gelman 1992; Cohen and
Dehaene 1996; Dehaene et al. 1998; Coulter, Choi, and Monroe 2012). For example, when it is
necessary to write out numeric characters (e.g., $8.55) on a check, visual as well as verbal word
representations (e.g., “eight dollars and fifty-five cents”) will also be encoded (Coulter, Choi,
Research has also shown that auditory (together with visual and analog) representations
are encoded during the mental rehearsal of (visually presented) Arabic numerals (e.g., $4) in a
simulated comparative shopping environment (Vanhuele, Laurent, and Dreze 2006; Coulter and
8
Coulter 2010). Here, rehearsal of the price information equates to verbal presentation in the
previous hypothesis. When consumers mentally “hear” the prices, both auditory representations
as well as visual word-names should be evoked (e.g., “4” evokes “fö(ə)r” and “four”). That
evocation should allow the consumer to make the connection between the first letter in his or her
own name, and the first letter in the price. Therefore, when the price is presented visually as
numeric characters, we would expect the name-letter/price effect to occur when the price is
rehearsed, but not to occur when not rehearsed. This implies a three-way (rehearsal x price x
H2 : There will be a three-way interaction of price, name-letter, and rehearsal on price liking
and purchase intentions. More specifically, when consumers mentally rehearse the price,
we expect a name-letter/price effect. When they do not mentally rehearse the price, we
do not expect a name-letter/price effect.
Birthday-Number/Price Effects
Although far less research has been carried out with regard to birthday-numbers,
evidence also indicates that people are positively biased toward other persons, objects, or events
that share numbers corresponding to their birthdays (Perham, Mirenberg, and Jones 2002; Jones
et al. 2004). Thus, for example, Patsy might be attracted to Gregg because they share the same
actual birthday (Pelham et al. 2003), or because his street address corresponds to her birthday
As noted earlier, birthday number effects are hypothesized to result from the same
implicit egotism that drives name-letter preference (Jones et al. 2004). Just as encoding in
spoken/written word format may be necessary in order for name-letter effects to manifest, we
expect that visual encoding as Arabic numerals will be particularly salient for manifesting
birthday-number effects. That is, if a price stimulus is presented visually (i.e., in the form of
numeric characters), a consumer should make the connection between his or her birthday-
9
numbers (e.g., the 30th of April) and the numbers in the price stimulus (e.g., $30). We refer to
this as the “Birthday-Number/Price Effect.” The affect that is transferred should manifest as
liking of the price attribute and increased brand purchase intentions. Thus we posit:
H3 : If prices are presented visually (as numeric characters), consumers will demonstrate
greater price liking and purchase intentions if the price does (vs. does not) share birthday
numbers with the consumer.
(Greenwald and Banaji 1995; Nuttin 1987), then it stands to reason that when a person’s positive
sense of self is compromised, he/she might attempt to overcome that “threat” by showing
exaggerated name-letter preference. In support of this argument, several findings show that when
respondents have recently written about a personal flaw, they overcome the implied ego
challenge by expressing even greater predilection toward people, objects, or events that reflect
their name-letter preference (Jones et al. 2002). We expect that the same moderating effects
should apply toward a product attribute. More specifically, when consumers’ egos are
threatened, an ordinal interaction effect should occur such that the hypothesized birthday-
H4: The greater price liking and purchase intentions that consumers demonstrate under
birthday-number/price matching conditions will be increased when consumers’ implicit
egotism is threatened.
Assuming the appropriate presentation context, we expect that both name-letter and
birthday-number effects will manifest in the case of single-digit prices, or multiple-digit prices
when all digits match the consumer’s name-letter or birthday-number. In the case of multiple-
digit prices when all digits do not match (e.g., Edward and “Eight-thirty-four”), manifestation of
10
Regarding the first two factors, there is no a priori theoretical basis for determining what
expectation is that a majority of matches would generate the effect. Regarding the third factor,
prices are distinctive in that each numeric character represents some unique unit of value, and
comma (e.g., $8,197) or as cents digits (e.g., $81.97).3 For example, the three numeric characters
are expressed as “red.” What this means is that although it is likely that affect could transfer from
“Robert” to “red” (i.e., name-letter matches first letter of attribute) but not from “Dan” to “red”
(i.e., name letter matches last letter of attribute), in the case of the price attribute it may be
equally likely that affect could transfer from “Fred” to “$8.25” (i.e., name-letter matches first
letter of third numeric character) as from “Fred to “$5.28” (name-letter matches first letter of
Research in numerical cognition has demonstrated both left-digit (Thomas and Morwitz
2009) as well as right-digit (Coulter and Coulter 2007) effects. Therefore, one might expect that
matching on left or right digits could create more pronounced implicit egotism effects than
matching on “interior” digits. Further, due to economic considerations, managers may have
greater latitude in varying right digits in general, and cents digits in particular, than they would in
varying left-most dollar digits. Thus in both our experimental and quasi-experimental studies,
choice of digit manipulations is based primarily on prior research, and practical rather than
theoretical considerations.
3
Exceptions to this rule would involve such numbers as 10, 11, 12, and 13-19.
11
STUDY 1
The base 10 numerical system consists of ten integers, each with a unique word name
(e.g., 1 = “one”; 2 = “two”). The numbers 11 (“eleven”) and 12 (“twelve”) also represent unique
word names. Beyond the number 12, however, numerical word names are essentially variations
of single integer names (e.g., “thirteen” = “three plus ten”). Therefore, we expect that price-
related name-letter effects are limited to the following letters: “O” (one), “T” (two, three, ten,
twelve), “F” (four, five), “S” (six, seven), “E” (eight, eleven), and “N” (nine). For the purposes
of our first study, we randomly chose to examine the letters “T” and “E.” We examine other
Study 1 involved a 2 (last name letter: T/E) 2 (price number: $622/$688) between-
radio commercials for a fictitious sporting goods retailer. In each commercial, the announcer
mentioned a special price for a fictitious brand of bicycle, and the brand name (Mizuki Pro) did
not begin with either of the target letters. The price was either $622 (i.e., “six twenty-two,”
consistent with the name-letter “T”) or $688 (i.e., “six eighty-eight,” consistent with the name-
letter “E”). We chose to manipulate the tens and units digits (rather than hundreds, which
remained constant at 6), because: a) prior research has indicated that the auditory manipulation
of these digits affects price preference and choice (e.g., Coulter, Choi, and Monroe 2012), b)
manipulating two digits rather than one was expected to create a stronger effect, and c)
The participants (N = 120) were students, faculty, and staff chosen from phone
directories and e-mail lists at several major Eastern U.S. universities, based on alphabetical
12
listings of their last names (only participants who did not also have target letters as the first letter
in their first names were included in this sample). Participants were contacted by e-mail and
asked to click on a link that allowed them to take part in our study. After they clicked on the link,
the scenario explained that a new sporting goods retailer would soon be moving to the area, and
we wanted to examine whether it could be price competitive. Each person then heard a radio
commercial with one of the two (randomly determined) prices. Next, they completed a brief
questionnaire to assess price liking and purchase intentions (Appendix A), using items consistent
with previous research (e.g., Dodds, Monroe and Grewal 1991; Suri and Monroe 2003). The
questionnaire also contained relevant demographic questions. Our expectation was that
participants whose last names began with “E” (“T”) and who were exposed to the $688 ($622)
price would express greater price liking and purchase intentions than those whose last name
began with “E” (“T”) and who were exposed to the $622 ($688) price.
Results
letter groups could introduce confounds if the groups vary on some other, untested variable
conditions in terms of demographics (i.e., age, gender, marital status, educational, nationality,
and income). In addition, we questioned participants regarding the nature of our experiment;
none revealed an understanding of its purpose or any awareness of the name-letter/price linkage.
interaction (F(1,116) = 17.19, p < .001) (Figure 2a). Neither main effect was significant. When
participants’ last names began with “E,” the group exposed to the higher $688 price indicated
significantly greater price liking than the group exposed to the $622 price (M$688=4.42, M$622=
13
3.29; F(1, 116) = 9.18, p < .01). Similarly, when participants’ last names began with “T,” the group
exposed to the $622 price indicated significantly greater price-liking than the group exposed to
the $688 price (M$622=4.43, M$688= 3.38; F(1, 116) = 8.04 p < .01). Thus, H1 is supported.
Purchase intention. ANOVA results again revealed the expected two-way (name letter
price-ending) interaction (F(1,116) = 22.52, p < .001) (Figure 2b). Neither main effect was
significant. ANCOVA’s were also run including all of the aforementioned demographic
variables as covariates, and none were significant at p>.10. Follow-up contrasts indicated that
when participants’ last name began with “E,” those exposed to the higher $688 price were
significantly more likely to purchase than were those exposed to the $622 price (M$688=4.91,
M$622= 3.88; F(1,116) = 8.19, p < .01). Thus, H1 is again supported. This result also demonstrates
that name-letter/price matching can overcome economic value in driving purchase intentions.
When participants’ last name began with “T,” those exposed to the $622 price were
significantly more likely to purchase than were those exposed to the $688 price (M$622=5.13,
M$688= 3.75; F(1, 116) = 14.83, p < .001). Of course, one might expect greater purchase intention at
$622 than at $688 due to economic considerations, but not to the extent that a 9% decrease in
Mediation Analysis. As a final step in our analysis, we tested whether price liking
mediates the effects of the (name-letter x price-name) interaction on purchase intentions. For the
test of mediation, we used the bootstrap PROCESS macro method suggested by Hayes (2012).
PROCESS Model 8 estimates the conditional (i.e., moderated by name-letter) indirect effects of
a causal variable (i.e., price) on an outcome variable (i.e., purchase intentions) through a
proposed mediator (i.e., price liking), as well as the conditional direct effect of that causal
14
variable on the outcome variable. Here [(a1i +a3iW) bi] represents the indirect effects of the
independent variable on the dependent variable through the mediator, and (c1’+c3’W) represents
the conditional direct effect of the independent variable on the dependent variable.
Our results showed that both the overall liking [F(3,116)= 5.75, p=.0011] and purchase
intentions [F(4,115)= 8.01, p<.001] “outcome” models were significant. The significant (price x
name-letter) interaction (t= 3.56, p<.001) implies that the indirect effect of price on purchase
intentions through price liking is moderated by name-letter. The normal theory tests for the
conditional indirect effects indicated positive non-standardized coefficients at both values of the
moderator (1.89 and 2.06), and the bootstrap results showed that the bias corrected and
accelerated 95% confidence intervals (CI) excluding zero were .84(lower)/2.93 (upper) and
both values of the name-letter moderator [coefficient = -1.13, t= -3.13, p<.005, and coefficient =
.76, t= 2.08, p<.05]. Lower/upper confidence intervals for the direct effects were -1.85/-.41 and
.037/1.48, respectively. Therefore, price liking partially mediates the (price x name-letter)
Discussion
Study 1 supports our expectations regarding name-letter effects resulting from verbal
price presentation. Purchase intentions were greater for the higher priced brand in the “E” name-
letter conditions, which suggests that the positive affect transfer resulting from participants’
name-letter/price associations (rather than the economic value of the prices) drove both price
liking and brand purchase intentions. Interestingly, our partial mediation results indicate that
although matching name-letter to price triggered implicit egotism valence transfer, price was not
the only construct to benefit from that transfer. Rather (as we will address in a subsequent study),
15
affect could also have been transferred to either other pertinent attributes (e.g., manufacturing
quality) or to the ad itself, which also impacts purchase intentions. In the next quasi-experiment,
we present prices in written numeral format to determine if these same effects (but involving
different letters) will again become manifest. We also examine a third (moderating) variable –
price rehearsal.
STUDY 2
Study 2 involved a 2 (last name letter: F/S) 2 (price number: $8.55/$8.66) 2 (price
Name letter was the fixed factor; assignment of participants to the price number and price
rehearsal conditions was randomly determined. One of the two price combinations appeared in a
print ad for a fictitious brand of coffee (see Appendix B). The brand name (LIVE coffee) did not
begin with either of the target letters, and the price was not included in any lines of text.
In the rehearsal conditions, participants were asked to “read to yourself, without speaking
out loud” any sale prices they encountered, and to rehearse the prices internally to keep them in
memory (i.e., because they would later be asked to recall those prices). This rehearsal
manipulation is consistent with previous studies that have based their design on in-store price
comparisons (e.g., Vanhuele et al. 2006). Because products are not always located contiguously,
consumers comparing prices must retain the prices associated with particular products or brands
in short-term working memory. This paired association learning is facilitated by internal price
recitation. The manipulation instructions ensured that participants mentally “heard” the prices,
evoked the verbal/visual price-word representations, and therefore made the hypothesized name-
As in Study 1, participants (N = 130) included students, faculty, and staff chosen from
telephone directories and e-mail lists at several major Eastern U.S. universities (i.e., based on
alphabetical listings of their last names). After clicking on an online link, participants viewed the
print ad containing one of the two prices; they subsequently completed a brief questionnaire with
items otherwise identical to Study 1, but modified to match the product. We expected that if
participants whose last names began with “F” (“S”) mentally rehearsed the price-names, those
who saw the $8.55 ($8.66) price would express greater price liking and purchase intentions (H2)
than those who were exposed to the $8.66 ($8.55) price. We expected that these effects would be
Results
differences across name-letter groups in terms of age, gender, marital status, educational level,
nationality, or income. In addition, in response to questions about the nature of the experiment,
none of the participants revealed an understanding of the purpose of the study or an awareness of
price-ending rehearsal) interaction (F(1,122) = 23.64, p < .001). Further as expected, in the
(F(1,122) = 45.33, p < .001). Follow-up contrasts revealed that when participants’ last names began
with “F” and they rehearsed the prices, those who viewed the $8.55 price indicated significantly
greater price liking than those who viewed the $8.66 price (M$8.55=5.33, M$8.66= 3.40; F(1, 122) =
24.75, p < .001). Conversely, when participants’ last name began with “S” and they rehearsed the
17
prices, those who viewed the $8.66 price indicated significantly greater price liking than those
who viewed the $8.55 price (M$8.66=5.15, M$8.55= 3.35; F(1, 122) = 20.70, p < .001). As expected,
the two-way interaction was non-significant in the non-rehearsal condition (F(1,122) = 0.02, n.s.),
and there were no significant differences across “F” (M8.55 = 3.78, M8.66 = 3.85) or “S” (M8.55 =
rehearsal) interaction (F(1,122) = 11.86, p < .01) (Figure 3). As expected and again in support of
interaction (F(1,122) = 22.39, p< .001) (see Figure 3a). Follow-up contrasts revealed that when
participants’ last name began with “S” and they rehearsed the prices, those who viewed the
higher $8.66 price were significantly more likely to purchase than those who viewed the $8.55
Further, when participants’ last name began with “F” and they rehearsed the prices, those
who viewed the $8.55 price were significantly more likely to purchase the brand than those who
viewed the $8.66 price (M$8.55=5.63, M$8.66= 3.95; F(1, 122) = 11.10, p = .001). Although one
would expect greater purchase intentions at $8.55 than at $8.66 due to economic value, one
would not expect that a 1% decrease in price could result in a 43% increase in purchase
intentions. Finally, in the non-rehearsal condition, as expected we did not find a significant two-
way (name-letter price-ending) interaction (F(1,122) = 0.02, n.s.), and there was no significant
difference in means across the four groups (F: M8.55 = 4.10 vs. M8.66 = 4.10; S: M8.55 = 4.20 vs.
however this time employing PROCESS Model 12 (i.e., due to the presence of two moderators;
Hayes 2012). As expected, results revealed that price liking partially mediates the (price x name-
letter) purchase intentions relationship, but only in the price rehearsal conditions.4
Discussion
These results support our expectations regarding name-letter effects as they relate to
prices presented visually as Arabic numerals. Name-letter preferences only became manifest
when affect was transferred due to the mental rehearsal of price names. We argue that this
rehearsal non-consciously evoked personal name-letter associations, and that is was implicit
egotism, rather than the economic value of the prices, that drove both price liking and brand
purchase intentions. When only the numeric characters were processed and encoded (i.e., as
Arabic numerals in the non-rehearsal conditions), name-letter effects did not emerge. Our
findings of partial mediation again imply that affect was transferred not only to the triggering
attribute (i.e., price), but to other attributes (e.g., “time well spent” – see Appendix B), to the ad
itself, or directly to the Live Coffee brand. We examine some of these other possible mediators
in a subsequent study.
An alternate explanation for our findings is that it was merely the alliterative sounds of
the prices themselves that had an effect. Alliteration does not require a repetition of letters but
rather of sounds, and prior research has demonstrated that alliteration may be linked not only to
brand name recall but to positive attitudes as well (Lowrey, Shrum, and Dubitsky 2003; Leigh
1994; Reece, Vanden Bergh and Li 1994; McQuarrie and Mick 1992, 1993). In each of our first
two studies we manipulated two digits rather than one – thus alliteration was already present in
our prices in both target and non-target conditions. That presence seemingly rules out alliteration
4
ANCOVA with purchase intentions as the dependent measure and the demographic variables as covariates
revealed that none of the covariates were significant at p>.10. See Web Appendix for mediation analysis details.
19
name-letter (“F” vs. “Ph”) – but not phoneme.5 Results (i.e., an effect in the case of “F” but not
“Ph”) confirmed our expectations that alliteration is not driving our results.
The external generalizability of the findings from Studies 1 and 2 may be limited,
however, in that we investigated only two sets of price endings and two specific name letters in
each quasi-experiment. Furthermore, our sample populations were restricted, and we were unable
to randomly assign participants to our blocking variable (name-letter group). Thus to address
these threats to external validity, in our next Study we examine the name-letter/price matching
effect using actual customer purchase data from a retail business establishment. We also examine
which price digits may be the most likely to impact name letter effects.
(FIELD) STUDY 3
An automobile dealership with three retail business locations in two Eastern U.S. states
agreed to provide the data for our field study. Each dealership sold approximately 60 vehicles per
month, of which roughly 25% (15 per dealership; total 45) were non-financed. We did not have
access to specific automobile make or model purchase information, but we learned the final
purchase prices (i.e., which included taxes and fees). We decided to examine cash (i.e., non-
dealer-financed) purchases only, because final payments (minus any prior deposits) typically
relied on cashier’s checks, for which the purchase prices appeared on the second line of the
check in written-word format. Assuming buyers read the amount they were paying, we could be
5
In the follow-up study, a 2 (first name-letter: F/Ph) 2 (price: $8.55/ $8.22) between-subjects quasi-experimental
design was employed (22 participants/cell). All stimuli and procedures were identical to Study 2. ANOVA revealed
a significant name-letter price-ending interaction (F(1,84) = 13.83, p< .001): (“F”: M$8.55=4.89, M$8.22= 3.89; F(1,84)
= 12.55, p= .001) and (“Ph”: M$8.22=4.30, M$8.55= 3.82; F(1,84) = 2.95, n.s.). Results rule out alliteration as an
alternate explanation and support the name-letter/price effect. More details are provided in the Web-Appendix.
20
reasonably certain that visual as well as verbal price-word representations had been processed
only. Of the approximately 45 non-financed automobile sales per month, approximately 17%
were sold to persons whose first names began with T, F, S, E, N, or O (7–8 vehicles/month).
Over a three-year period, 275 purchases thus matched our name-letter criteria. After the
purchases had been finalized, customers also completed a very brief customer satisfaction
survey, and we employed the results from the final question, “How would you rate your liking of
the deal you received at (dealership)?” (1–10, “liked very little/liked very much”), as our
dependent measure. There was no survey question that specifically addressed purchase prices;
however, price liking and deal liking should be highly correlated. Approximately 67% of
customers completed this survey question, resulting in a final usable sample group of N = 183.
All of the purchase prices were between $10,000 and $99,000. Thus each price had a total
of seven digits, including the two that represented cents. For each customer’s name-letter
purchase price, we recorded the number of times that the corresponding price-names appeared in
that seven-digit price. For example, if Fred paid $25, 471.46 for his vehicle (including taxes and
fees), three of the digits in that price (i.e., the 5, the first 4, and second 4) represent price names
that match Fred’s name-letter. If Samantha paid $17,429.46 for her new vehicle, two digits from
In general, price names representing multiple digits (i.e., “ten,” “eleven,” and “twelve”)
could correspond to any two consecutive digits except the thousands, hundreds, units, or second
cents digit. For example, the price name “ten” appears in the price $25,410.67, but not in the
price $25,104.67. No digit follows the second cents digit, so it could not represent a 10, 11, or
21
12. Also in the case of any “ten,” “eleven,” or “twelve,” we recorded only the first of the two
corresponding digits as a match. Thus, for example, when examining “T” name-letter purchases
match (“twenty”), as would the tens (“twelve”). But the unit digits would not be recorded as a
Finally, the price-name “one” could be used to represent the thousands digit (e.g.,
$21,555.55), the hundreds digit (e.g., $25,155.55), the units digit (e.g., $25,551.55), or the
second cents digit (e.g., $25,555.51), but not the ten-thousands, tens, or first cents digits (see
Table 2). In sum, we evaluated the prices according to their verbal/written-word representations,
such that not all price names could represent all seven digits. However, all seven digits could be
represented by a minimum of nine price names (i.e., the blank column spaces in Table 2). Once
the price names corresponding to customer name letters had been recorded, we examined
As noted earlier, for each of the seven digits, we recorded whether the customer’s name
matched a price name. We then summed across digits to create our independent variable (labeled
“total-match”). When deal liking was regressed on the total-match variable, the model was
Because not all price names could represent all seven digits in our target prices, we next
examined each of the digits individually in a multiple regression analysis (all variance inflation
factors were less than 1.1). Matching on four of the seven digits affected deal liking, but
matching on the other three did not. Specifically, matching on the ten-thousands (F(1,181) = 7.47, p
22
= .007), thousands (F(1,181) = 8.41, p<.01), first cents (F(1,181) = 14.29, p< .001), and second cents
(F(1,181) = 6.34, p = .01) digits influenced consumers’ feelings about the deal. There was no effect
of matching customers’ name letters with price names for the hundreds (F(1,181) = .04, n.s.), tens
(F(1,181) = .62, n.s.), or units (F(1,181) = .59, n.s.) digits. Thus, both left-digit (primacy) and right-
digit (recency) effects appeared to occur. Other than contributing to the total, matching the
Finally, we examined name letters individually. When deal liking was regressed on the
total-match variable, the model was significant for the name letters “O” (F(1,17) = 4.40, p = .05),
“T” (F(1,24) = 9.31, p <.01), “F” (F(1,24) = 11.51, p <.005), “S” (F(1,67) = 5.43, p<.05), and “E”
(F(1,16) = 6.18, p <.05). Notably, the model was significant for the name letter “O,” despite (a) the
few data points (i.e., 17), (b) its match with only one price name (i.e., “one”), and (c) the
potential for “one” to represent only four of the seven digits (i.e., thousands, hundreds, units, and
second cents) in our seven-digit prices (see Table 2). Therefore, this finding lends strong support
to our theory.
Although the results were in the predicted direction, our model was not significant in the
case of the name-letter “N” (F(1,23) = 1.86, n.s.). In many ways, this result was not surprising,
because a wealth of studies have demonstrated the positive effects of 9-ending prices on brand
preference and purchase intentions (see Anderson and Simester 2003a, 2003b; Schindler 1991).
Thus, all customers (not just those with matching name letters) might have been predisposed
toward prices with nines appearing in the thousands, units, or second-cents places. Perhaps such
Finally, we also analyzed cases where the buyer’s name had zero matches with the price
(M=7.72, n=39), 1 match with the price (M=7.92, n=74), or 2 or more matches with the price
23
(M=8.21, n=70). The results demonstrated that deal liking was directionally stronger (than zero
matches) for 1 match (F(1,180) = 1.02, p=.31), but was significantly higher for 2 or more matches
F(1,180) = 12.24, p < .001). The results also supported the difference between 1 and 2 or more
matches (F(1,180) = 9.97, p < .01). Thus, this analysis seems to suggest that in the case of multiple
digit prices, two or more digit matches (or 29% of total digits) are required in order for name-
Discussion
As posited in our first two hypotheses, consumers will exhibit name-letter preferences if
they form the word representations of price-names that elicit name-letter associations. In Study
1, we argue that these representations were encoded due to the verbal presentation of prices in
radio commercials. In Study 2, we argue that these representations were encoded due to the
mental rehearsal of price-names that accompanied the visual presentation of Arabic numerals
(e.g., $8.55). Finally in our (field) Study 3, we argue that these representations were encoded due
to the transcription of price-names that appeared on customers’ checks (Coulter, Choi, and
Monroe 2012). But can implicit egotism drive price liking and brand purchase intentions if prices
As noted earlier, implicit egotism has also been shown to impact preference effects
associated with birthday numbers (Nuttin 1985; Pelham et al., 2003). Also as noted earlier,
whereas the verbal code (and accompanying visual encoding as price words) may be necessary in
order for price-related name-letter effects to manifest, we expect that the numeric-character form
of visual encoding will be particularly salient for birthday numbers. Therefore to further
generalize the findings observed in our earlier studies to that of a purely “visual processing/
visual encoding” environment, we next examine implicit egotism effects as they apply to
24
examine the process behind our hypothesized effects (Jones et al. 2002).
STUDY 4
We first created an online survey that included an ad for a pasta dinner promotion at a
local (fictitious) restaurant (see Appendix C). In the advertised price, the dollars digits remained
constant at $39 (greater than any possible day-of-the-month birthday number), whereas the cents
digits varied. We included the phrase “includes 15% gratuity and all taxes” to disguise the
potentially “unusual” price-endings. Final sale prices for items purchased online typically
include taxes and shipping, which increases variability and reduces the probability of observing
the more typical round numbers. Two hundred five students from a major Eastern U.S. university
After they first clicked on the survey link, participants were told that the purpose of the
study was to examine the relationship between consumers’ “time frame horizon” (i.e., as
the following page they were asked a series of questions regarding historical dates (e.g., date
J.F.K. was slain; Valentine’s Day). Included among those questions was an item pertaining to
their birthdate. The purpose of the other items was to disguise our interest in birthday numbers.
In the “threat” condition, participants were informed that their score on the historical
items was 10% below the mean of other survey respondents. In the “non-threat” condition,
substandard test performance relative to one’s peers will serve as an ego threat (e.g., Greenberg
and Pyszczynski 1985: Heatherton and Vohs, 2000; Vohs and Heatherton 2001, 2004).
The birthday-number matching manipulation was operationalized by using the day of the
month entered as the cents digits of the price in the stimulus ad. For example, if a participant
indicated that his or her birthdate was April 16, the price appearing in the stimulus advertisement
would appear as $39.16. For the other two conditions, participants viewed a price where five
cents was either added to or subtracted from their day of birth. Thus, in the previous example, the
price in the stimulus advertisement would appear as either $39.11 or $39.21. Whether
containing cents digits that were five cents above or below that number was randomly
determined. The exception to this randomization occurred in the case of birthday numbers less
than 5; to avoid negative numbers or change the dollar amount, only the 5-cents added condition
was employed. Our expectation was that those participants who viewed their birthday number as
part of the price would express greater price-liking and purchase intentions. Our stimulus ad
appeared on the third page of the questionnaire. After viewing the stimulus, participants
Results
All possible birthday numbers (i.e., 1–31) were represented in our sample, with
frequencies ranging from 2 (i.e., 10) to 9 (i.e., 6, 15, 17). For purposes of analysis, we created a
(F(2,199)= 2.86, p= .05). The differences between match/no-match means in both the threat (MM=
26
5.19 vs. MNMA=3.39 and MNMB=3.59; F(1,199) = 51.16, p < .01)6 and non-threat conditions (MM=
4.58 vs. MNMA= 3.72 and MNMB= 3.74; F(1,199) = 9.94, p < .001) were significant, supporting H3.
However the match/no match difference was significantly greater in the former (threat)
condition than in the latter (non-threat) condition (F(1,199) = 5.53, p < .05), demonstrating an
ordinal interaction and confirming underlying process. Thus H4 is supported. Within each threat
condition, there was no difference between the two non-match conditions. As in previous studies,
price liking results mirrored those of purchase intentions, and mediation analysis indicated
partial mediation.7
Discussion
Importantly, by including a manipulation of personal “threat,” we confirm the process behind our
results (Jones et al. 2002). In our final study, we investigate birthday-number/price and name-
these relationships, we are able to determine whether the affect-transfer triggered by implicit
egotism in our matching conditions impacts either non-relevant attributes, or reactions to the ad
itself. For example, if the price-ending is matched to respondents’ birthday numbers, does that
impact price liking only, or can it also affect the liking of other product attributes, evaluation of
the advertisement, or some other variable that also impacts purchase intentions? In Study 5 we
STUDY 5
In Study 5 we replicate Study 4, but a) only examine the non-threat condition, b) employ
6
M= match condition; NMA = no-match, cents-added condition; NMB = no match, cents-subtracted condition
7
See Web Appendix for price liking and mediation analysis results.
27
only the 5-cents added manipulation for the non-birthday-matching prices, and c) include a (first)
name-letter/brand matching variable as one of our manipulations. Thus our study involves a 2
The stimulus used in our experiment was modified, such that the price and “brand-
cuisine” attribute information was prominently presented in the upper left portion of the ad (see
Appendix D). We chose 26 Italian restaurant names (e.g., “Eduardo’s) and 26 cuisine adjectives
(e.g., “Excellent”), such that the first name letter of the restaurant always matched the first letter
of the adjective. Thus, in our name-letter matching conditions, Ethan might see an ad for
“Eduardo’s Excellent” cuisine, whereas in our non-matching conditions Ethan might see an ad
The cover story was altered such that participants were told that the purpose of the study
was to examine not only the relationship between consumers’ “time frame horizon” (i.e., as
indicated by responses to historical “date-related” questions) and online purchase patterns, but
also the relationship between consumers’ “recognition of popular names” (i.e., as indicated by
responses to “name-letter” questions) and online purchase patterns. Further, on the following
page they were asked a series of questions regarding both historical dates and names of popular
figures (e.g., first name of the Queen of England). Included among those questions were items
pertaining to the first letter of their first name as well as their birthdate. Finally questionnaire
items also included an item assessing food liking and two items measuring ad liking (Appendix
8
Specific name-letters and birthday numbers were pooled in the match/no-match conditions. Thus, it is possible that
any given name-letter (e.g., “z”) or birthday-number (e.g., “26”) could appear in the match condition, but not in the
non-match condition (or vice-versa). Ensuring equal and all letters/numbers in both conditions would require a (26 x
2 x 31 x 2) quasi-experimental design.
28
A). All other procedures and measures were identical to Study 4. One hundred fifty-seven
Results
All possible birthday numbers were represented in our sample, with frequencies ranging
from 2 (i.e., 4, 6, 12, 18, 27) to 9 (i.e., 10, 14, 20, 26). All possible name-letters except Q and X
were also represented in our sample, with frequencies ranging from 2 (i.e., I, N, O, W, Y) to 15
(i.e., J). For purposes of analysis, we created a birthday-number/cents-digit matching versus non-
number/price matching effect [MMatch= 3.84 vs. MNon-Match=3.16; F(1,153)= 9.87, p<.005].
matching. Thus H3 is again supported. Consistent with previous research (Brendl et al., 2005),
the name-letter/brand matching effect was also significant [F(1,153)= 12.53, p=.001]. Mean
purchase intentions in the name-letter/brand matching condition (M= 3.88) were significantly
greater than mean purchase intentions in the non-matching condition. The interaction between
birthday-number/price match and name-letter/brand match was not significant (F(1,153)= .50, n.s.].
This result suggests that the price matching effects we have demonstrated are robust, and can
Price Liking, Product Liking, and Ad Evaluations. Consistent with Study 5, for price
liking there was a significant birthday-number/cents-digit matching effect [F(1,153)= 6.88, p=.01].
Mean price liking in the matching condition (M= 3.32) was significantly greater than in the non-
29
matching, cents-added condition (M = 2.71). The name-letter matching variable was not
significant (p=.501).
name-letter/brand matching effect [F(1,153)= 17.45, p<.001]. Food liking in the name-letter
matching condition (M= 5.14) was significantly greater than in the non-matching condition (M=
4.24). In this instance the birthday-number matching effect was not significant.
matching effect [F(1,153)= 11.42, p=.001]. Ad evaluations in the birthday matching condition (M =
4.76) were significantly greater than ad evaluations in the non-matching condition (M = 4.13),
and ad evaluations in the name-letter/brand matching condition (M= 4.79) were significantly
Mediation Analysis. Due to the significant ad evaluation and product attribute evaluation
effects, as a final step in our analysis we investigated whether either of these variables might
mediators were not expected to be linked in series, we utilized PROCESS Model 4 (Hayes
2012). In regard to the first relationship, results were similar to previous studies - that is, price
purchase intentions [total effect (c)= .67, t= 3.05, p<.005, LLCI= .24, ULCI= 1.11; indirect
effect (a x b)= .25, Sobel Z= 2.37, p=.02, LLCI= .15, ULCI= .74]. Interestingly, however, ad
evaluations were also found to be a significant mediator [indirect effect (a x b) =.12, Sobel Z=
1.99, p=.05, LLCI= .03, ULCI= .31]. The mediating effect of product-attribute (food) liking was
30
not significant (p=.44). Because the direct effect (c’) of birthday-matching on purchase intentions
was also non-significant (p=.17), these results indicate that ad and price liking fully mediate the
With regard to the second (name-letter) relationship, findings were as expected given
previous results reported in the literature (Brendl et al., 2005) - that is, product-attribute (food)
significant mediator, but the mediating effect of price liking was not significant. As in our
Discussion
Results of Study 5 indicate that both name-letter and birthday-number sources of implicit
egotism can drive purchase intentions simultaneously. In this study we assume that these are
separate and distinct processes – that is, birthday-numbers affect the price attribute, whereas
name-letters affect the product attribute. As demonstrated in earlier studies, name-letters could
also impact the price attribute,10 but only if prices are rehearsed (i.e., which is not the case here).
Thus, in both instances, affect is transferred due to visual associations involving alpha-numeric
A second implication of this study is that the implicit ego-driven affect elicited by either
name-letter or birthday-number matching can transfer not only to the associated price or product-
related attribute, but also to other aspects of the ad (i.e., in this case, ad evaluations). This
conclusion is in agreement with the partial mediation results of our earlier studies, and has
important managerial relevance. Specifically it demonstrates that marketers could benefit from
9
See Web Appendix for complete statistics related to name-letter mediation results.
10
Such a scenario could be envisioned, for example, if a respondent named “Fred” was also born on the 15 th (and
was included in our birthday-matching condition).
31
matching name-letter or birthday-number to price, even if price is not the primary determinant
attribute in driving choice. In such instances effects through price liking might be mitigated,
Third, in our earlier follow-up to Study 2 we find that alliteration is not driving our name-
letter results (see Footnote 13 and Web Appendix). Here, however, all participants are exposed
letter effects still occur seems to once again rule out the alliterative process as an alternate
explanation.
The results of our five studies support our expectations regarding name-letter and
participants demonstrated greater price liking and purchase intentions if the price shared name-
letters with the participant than if it did not. The name-letter preferences manifested when price
names were either orally presented or internally rehearsed (i.e., which allowed participants to
mentally “hear” the prices). This suggests that auditory encoding of sounds - and the
accompanying visual encoding of those sounds as words - was necessary in order to form name-
letter associations and elicit affect transfer. It was that positive affect (and not the economic
value of the prices) that drove both price liking and brand purchase intentions. When only the
numeric characters were processed and encoded (i.e., in the non-rehearsal conditions of Study 2),
or when the auditory representations were not linked to the specific name-letters (i.e., in the non-
purchase prices corresponding to the consumer name letters O, T, F, S, E, and N. Unlike our first
32
two studies, this study used first names rather than last names, thereby increasing the external
validity and generalizability of our findings. Although we cannot be certain that purchase prices
were verbally spoken, the fact that they were transcribed in written word format on check
payments was sufficient to elicit the name letter associations. Deal liking was predicted by name
letter matching, not only in terms of total matches, but also in terms of four of the seven
individual price digits. With one exception (N), deal liking was also predicted by name-letter
matching for each of the individual name letters. Therefore, the preponderance of evidence
matching birthday-number to price, rather than name-letter to price. Consistent with earlier
findings, our Study 4 mediation analysis confirms that the positive affect associated with
birthday-numbers can directly influence consumers’ liking of prices, which can subsequently
drive brand purchase intentions. In our series of studies, we show that this valence transfer can
occur when the personal source and the price share name-letters or birthday-numbers, rather than
just the personal source and the brand. This finding is particularly important, because it greatly
Study 5. In addition, we show that the implicit ego-driven affect that is elicited by such matching
can transfer not only to the associated price or product-related attribute (i.e., manifesting as
attribute “liking”), but to other aspects of the ad as well (i.e., in this case, ad evaluations). This
means, for example, that even if price is not the most important attribute driving purchase,
matching names or birthday-numbers to price could have potentially beneficial effects because
In our research, we focused on price promotions for goods and services that were very
expensive (e.g., thousand-dollar automobiles in our Field study), expensive (e.g., hundred-dollar
bicycles in Study 1), less expensive (thirty-nine dollar meals in Studies 4 and 5) and inexpensive
(eight-dollar coffee in Study 2). Thus, the research demonstrates fairly consistent findings across
a variety of both expensive and inexpensive goods and services, using a variety of research
strength of the price matching effect on price-liking and purchase intentions, b) the overall
significance of the effects, and c) the robustness of our results. Regarding (a), we found nine
effects on price liking with an average weighted r=.39, and eight effects on purchase intentions
with an average weighted r=.41. Regarding (b), we tested the significance of the overall
relationships using Rosenthal and Rosnow’s (2008) p-value combination technique. The results
indicate that the overall relationships were significant at p < .001 for both variables. Finally
regarding (c), we used Rosenthal and Rosnow’s (2008) file drawer technique to determine that it
would take over 400 (300) null studies to reduce the significance of our price liking (purchase
intention) results to the .05 level, suggesting that the results are highly robust.
and Rosnow’s (1982) Binomial Effect Size Display (BESD). Given that the effect size was r
=.41 for purchase intentions (explaining over 16% of the variance), we examined a hypothetical
scenario of 200 consumers (of which 100 see a price that creates a name or birthday price match,
and 100 see a price that does not create such a match). For the price match group, approximately
70 are predicted to have had higher purchase intentions, whereas out of the non-match group,
only 30. Thus, when a price match exists, the propensity to buy is shown to have increased by a
34
factor of 2.33. This finding dramatically illustrates the practical significance of our results.
With the advent of database marketing, the marketer’s ability to quickly and efficiently
tailor product offers to meet the individual customer’s specific needs has greatly increased.
Similarly, the use of online and mobile vehicles - as well as location-based services - has enabled
firms to become much more successful in tailoring marketing communications, and in presenting
customers with unique offers. The ability to deliver customized messages and customized prices
that appeal to both the visual and auditory senses has grown exponentially in recent years with
the proliferation of computers, smart phones and tablets. A key element to the success of these
Our results highlight the importance of implicit egotism in increasing that personal
relevance. Both name-letters and birthday numbers carry symbolic meaning; by “matching”
these letters and numbers to prices (or other pertinent attributes), customer-brand associations are
formed. Attribute associations trigger the affect transfer that can ultimately strengthen both
interest in, and preference for, a given product (Keller 1993). Thus brand managers should
carefully design their brands as well as product and promotional materials to capitalize on the
utilization of such cues. Managers should also be aware of language and cultural variations that
could either increase or decrease the opportunities for name-letter matching (Hoorens and Nuttin
price, particularly in the case of “big ticket” items where some degree of price negotiation is
customary, and where face-to-face interaction facilitates the sharing of personal information. For
35
example, an automobile salesperson negotiating a price on a new car with a potential buyer
might be more likely to close the deal if he or she incorporates name-letters (which will be
known to the seller) or birthday-numbers (which may or may not be known to the seller) into the
final selling price. According to our field research, the first (left-most) and last (right-most) digits
of the price appear to be particularly important. Presumably (i.e., due to their relatively minor
impact on economic value), the salesperson should have a great degree of latitude in matching
retail settings, where computer-generated prices can be quickly and easily altered with the click
of a mouse. For example, once customers have provided personal “check-out” information, the
final price can be matched with their name or date-of-birth (i.e., again in the case of cents digits,
with only a very minor impact on economic value) by varying the shipping and handling costs
charged. In the case of “return” customers, the initial sale prices to which the customer is
campaign sponsored by a favorite restaurant chain could receive deal offers with prices
name-letter matching, the likelihood of effects manifesting may be increased by: a) presenting
the prices verbally, b) persuading customers to “hear” the prices by encouraging them to “read
over” the information before confirming the sale, or c) embedding the prices in a line of text.
For any of these price customization strategies, privacy concerns regarding the provision
and use of personal information can arise. One might also question the ethical implications of
arbitrarily manipulating shipping and handling costs. It is not the purpose of this article to delve
36
into the moral or ethical implications of such price manipulation strategies—merely to discuss
their implications. Further research and scholarly discourse should address these relevant
concerns.
Four of the six possible name letters associated with price (T, E, F, and S) were utilized
in our quasi-experiments; we recommend that future research replicate our field study findings in
an experimental setting using O’s and N’s. Also, in this paper we do not investigate the
consumers will demonstrate birthday number preferences if the brand attribute does (vs. does
not) share birthday numbers with the consumer, and if the consumer does (vs. does not) form
visual representations of the price numbers. Investigation of such a hypothesis would require that
participants first hear their birthday number (i.e., perhaps embedded within an audio ad), then
write down that number before evaluating the product offer. Future research might also examine
With the exception of Study 5, we addressed only one of a possible myriad number of
product attributes (i.e., price). Further studies might attempt to replicate our results by addressing
investigated the day the customer was born (1–31), but not the month (1–12) or year (e.g., 1965).
Whether our findings are consistent across these latter two categories awaits further study.
The field study results provided a number of interesting insights. The results demonstrate
support for both left-digit (primacy) and right-digit (recency) effects. In our experiments and
to examine the relative effects of matching left-digits vs. right digits. For example, a price of
37
$44.77 would be verbally encoded as “Forty-Four dollars and Seventy-Seven Cents.” Such a
price would represent a right digit match for Samantha, but a left digit match for Frank.
The growth in online and mobile media provides a plethora of opportunities to further
explore these concepts in greater detail (see Grewal et al. 2011). It is highly likely that media
matching effects. For example, name-letter/price effects are likely to be stronger for ads viewed
on tablets where consumers are used to processing auditory information in addition to visual
information.
Finally, personal relevance (and implicit egotism) can also be explored in the context of
customized direct-marketing advertising messages that include the customer’s name and provide
them with willing-to-pay (WTP) pricing options (Wertenbroch and Skiera 2002). It is likely that
customers who indicate a price that matches their name-letter or birthday-number should feel
REFERENCES
Anderson, Eric and Duncan Simester (2003), "Mind Your Pricing Cues" Harvard Business
Review, 81(9), 96-103.
Anderson, Eric and Duncan Simester (2003), "Effects of $9 Price Endings on Retail Sales:
Evidence from Field Experiments," Quantitative Marketing and Economics, 1 (March),
93-110.
Anseel, F. and W. Duyck (2008), “Unconscious Applicants: A Systematic Test of the Name-
Letter Effect,” Psychological Science, 19 (10), 1059-1061.
Ashcraft, Mark H. (1992), “Cognitive Arithmetic: A Review of Data and Theory,” Cognition, 44
(1-2), 75 106.
Berninger, Vigrinia W., Robert D. Abbott, Janine Jones, Beverly J. Wolf, Laura Gould, Marci
Anderson-Youngstrom, Shirley Shimada and Ken Apel (2006), “Early Development of
language by Hand: Composing Reading, Listening, and Speaking Connections,”
Developmental Neuropsychology, 29 (1), 61-92.
Best, Roger J. (2009), Market-Based Management, 5th Edition. Pearson/Prentice Hall: Upper
Saddle River, New Jersey.
Brendl, C. Miguel, Amitava Chattopadhyay, Brett W. Pelham and Mauricio Carvallo (2005),
“Name Letter Branding: Valence Transfers When Product Specific Needs Are Active,”
Journal of Consumer Research, 32 (3), 405-415.
Cohen, Laurent and Stanislas Dehaene (1996), “Cerebral Networks for Number Processing:
Evidence from a Case of Posterior Callosal Lesion,” NeuroCase, 2 (3), 155 174.
Coulter, Keith S. and Robin A. Coulter, (2010) “Number Sounds and Price Perceptions: The
Effects of Phonetic Symbolism on Price Discount and Value Perceptions,” Journal of
Consumer Research, 37 (August), 315-328.
Coulter, Keith S., Pilsik Choi, and Kent B. Monroe (2012), “Comma N’ Cents in Pricing: The
Effects of Auditory Representation Encoding on Price Magnitude Perceptions,” Journal
of Consumer Psychology, 22 (3), 395-407.
Dehaene, Stanislas (1997), The Number Sense, Oxford: Oxford University Press.
Dehaene, Stanislas, Serge Bossini and Pascal Giraux (1993), “The Mental Representation of
Parity and Number Magnitude,” Journal of Experimental Psychology: General, 122 (3),
371 396.
39
Dehaene, Stanislas and Laurent Cohen (1995), “Towards an Anatomical and Functional Model
of Number Processing,” Mathematical Cognition, 1 (1), 83 120.
Dodds, William B., Kent B. Monroe and Dhruv Grewal (1991), "Effects of Price, Brand, and
Store Information on Buyers' Product Evaluations," Journal of Marketing Research, 28
(August), 307-319.
Feys, Jos (1995), “Mere Ownership: Affective Self-Bias or Evaluative Conditioning?” European
Journal of Social Psychology, 25 (5), 559-575.
Gallistel, Charles R. and Rochel Gelman (1992), “Preverbal and Verbal Counting and
Computation,” Cognition, 44 (1-2), 43 74.
Gallucci, Marcello (2003), “I Sell Seashells by the Seashore and My Name is Jack: Comment on
Pelham, Mirenberg, and Jones (2002)” Journal of Personality and Social Psychology, 85
(5), 789-799.
Gillam, Ronald B. and Judith R. Johnston (1992), “Spoken and Written Language Relationships
in Language/Learning – Impaired and Normally Achieving School-Age Children,”
Journal of Speech and Hearing Research, 35 (December), 1303-1315.
Greenwald, Anthony G. and Mahzarin R. Banaji (1995), “Implicit Social Cognition: Attitudes,
Self-Esteem, and Stereotypes,” Psychological Review, 102 (1), 4-27.
Grewal, Dhruv, Kusum Ailawadi, Dinesh Gauri, Kevin Hall, Praveen Kopalle and Jane
Robertson (2011), “Innovations in Retail Pricing and Promotions,” Journal of Retailing,
87 (S1), S43-S52.
Heatherton, Todd F. and Kathleen D. Vohs (2000), “Interpersonal evaluations following threats
to self: Role of self-esteem,” Journal of Personality and Social Psychology, 78 (4),
725-736.
Hoorens, Vera and Jozef M. Nuttin (1993), “Overevaluation of Own Attributes: Mere Ownership
or Subjective Frequency?” Social Cognition, 11 (1), 117-200.
40
Hoorens, Vera and Elka Todorova (1988), “The Name Letter Effect: Attachment to Self or
Primacy of Own Name Writing?” European Journal of Social Psychology, 18 (4), 365-
368.
Hoorens, Vera, Jozef M. Nuttin, Ildiko Erdelyi-Herman and Ubolwanna Pavakanun (1990),
“Mastery Pleasure Versus Mere Ownership: A Quasi-Experimental Cross-Cultural and
Cross-Alphabetical Test for the Name Letter Effect,” European Journal of Social
Psychology, 20 (3), 181-205.
Hodson, Gordon and James M. Olson (2005), “Testing the Generality of the Name Letter Effect:
Name Initials and Everyday Attitudes,” Personality and Social Psychology Bulletin, 31
(8), 1099-1111.
Jain, Shailendra Pratap and Maheswaran, Durairaj (2000), “Motivated Reasoning: A Depth-of
Processing Perspective,” Journal of Consumer Research, 26 (4), 358-371.
Jones, John T., Brett W. Pelham, Mauricio Carvallo and Matthew C. Mirenberg (2004), “How
Do I Love Thee? Let Me Count the J’s: Implicit Egotism and Interpersonal Attraction,”
Journal of Personality and Social Psychology, 87 (5), 665-683.
Jones, John T., Brett W. Pelham, Matthew C. Mirenberg and John J. Hetts (2002), “Name Letter
Preferences Are Not Merely Mere Exposure: Implicit Egotism as Self-Regulation,”
Journal of Experimental Social Psychology, 38 (March), 170-177.
Johnson, Mitzi M. S. (1986), “The Initial Letter Effect: Ego-Attachment or Mere Exposure?”
Unpublished doctoral dissertation, Ohio State University.
Keller, Kevin Lane (1993), “Conceptualizing, Measuring, and Managing Customer-Based Brand
Equity,” Journal of Marketing, 57 (1), 1-22.
Kim, John, Chris T. Allen and Frank R. Kardes (1996), “An Investigation of the Mediational
Mechanisms Underlying Attitudinal Conditioning,” Journal of Marketing Research, 33
(August), 318-328.
Kitayama, Shinobu and Mayumi Rarasawa, M. (1997), “Implicit Self-Esteem in Japan: Name
Letters and Birthday Numbers,” Personality and Social Psychology Bulletin, 23 (7), 736-
742.
Knewtson, Heather S. and Richard W. Sias (2010), “Why Susie Owns Starbucks: The Name
Letter Effect in Security Selection,” Journal of Business Research, 63 (December), 1324-
1336.
Komori, Megumi and Koji Murata (2008), “Implicit Egotism in Japan: Preference for First
and Family Name Initials,” Hitotsubashi Journal of Social Studies, 40(2), 101-109.
Koole, Sander, Ap Kijksterhuis and Ad van Knippenberg (2001), “What’s in a Name: Implicit
41
Self-Esteem and the Automatic Self,” Journal of Personality and Social Psychology, 80
(4), 669-685.
Koole, Sander and Brett W. Pelham (2003) “On the Nature of Implicit Self-Esteem: The Case of
the Name Letter Effect,” in Motivated Social Perception, Steven Spencer, Steven Fein,
Mark P. Zanna, and James Olson (Eds.), Lawrence Erlbaum: Mahway, N.J., 93-116.
Leigh, James H. “The Use of Figures of Speech in Print Ad Headlines,” Journal of Advertising,
23 (2), 17-26.
Lowrey, Tina M., L.J. Shrum and Tony Dubitsky (2003), “The Relation Between Brand Name
Linguistic Characteristics and Brand-Name Memory,” Journal of Advertising, 32 (3), 7-
17.
MacKenzie, Scott B., Richard J. Lutz and George E. Belch (1986), “The Role of Attitude
Toward the Ad as a Mediator of Advertising Effectiveness: A Test of Competing
Explanations,” Journal of Marketing Research, 23 (December), 130-143.
McQuarrie, Edward F. and David Glen Mick (1992), “On Resonance: A Critical Pluralistic
Inquiry into Advertising Rhetoric,” Journal of Consumer Research, 19 (September), 180-
197.
McQuarrie, Edward F. and David Glen Mick (1993), “Special Topic Session: The New
Advertising Rhetoric,” in Advances in Consumer Research, Vol. 20 (1), Leigh McAlister
and Michael Rothschild, eds., Provo, UT: Association for Consumer Research, 308.
Melnyk, Valentyna, Klein, Kristine and Volckner, Franziska (2012), “The Double Edged Sword
of Foreign Brand Names for Companies from Emerging Countries,” Journal of
Marketing, 76 (6), 21-37.
Mitchell, Andrew (1986), “The Effect of Verbal and Visual Components of Advertisements on
Brand Attitudes and Attitude Toward the Advertisement,” Journal of Consumer
Research, 13 (June), 12-24.
Monroe, Kent B. (2003), Pricing: Making Profitable Decisions (3rd edition), New York:
McGraw-Hill.
Nelson, Leif and Joseph P. Simmons (2007), “Moniker Maladies: When Names Sabotage
Success,” Psychological Science, 18 (12), 1106-1112.
Nuttin, Jozef M. (1987), “Affective Consequences of Mere Ownership: The Name Letter Effect
in Twelve European Languages, European Journal of Social Psychology, 17 (4), 381-
402.
Nuttin, Jozef M. (1985), “Narcissism Beyond Gestalt and Awareness: The Name Letter Effect,”
European Journal of Social Psychology, 15 (July-September), 353-361.
Pelham, Brett W., Mauricio R. Carvallo, Tracy DeHart and John T. Jones (2003), “Assessing the
Validity of Implicit Egotism: a Reply to Gallucci,” Journal of Personality and Social
Psychology, 85 (November), 800-807.
Pelham, Brett W., Matthew C. Mirenberg and John T. Jones (2002), “Why Susie Sells Seashells
by the Seashore: Implicit Egotism and Major Life Decisions,” Journal of Personality and
Social Psychology, 82 (April), 469-487.
Pelham, Brett W., Carvallo, Mauricio and Jones, John T. (2005), “Implicit Egotism,” Current
Directions in Psychological Science, 14 (2), 106–110.
Pinel, Philippe, Gurvan Le Clech, Pierre-Francois Van de Moortele, Lionel Naccache, Denis
LeBihan and Stanislas Dehaene (1999), “Event-Related fMRI Analysis of the Cerebral
Circuit for Number Comparison,” Neuroreport, 10 (7), 1473 1479.
Price, C.J., R. J. Wise and R. S. Frackowiak (1996), “Demonstrating the Implicit Processing of
Visually Presented Words and Pseudowords,” Cerebral Cortex, 6 (1), 63-70.
Read, Stephen J., Eric J. Vanham and Lynn C. Miller (1997), “Connectionism, Parallel
Constraint Satisfaction Processes, and Gestalt Principles: (Re)Introducing Cognitive
Dynamics to Social Psychology,” Personality and Social Psychology Review, 1 (1), 26-
53.
Reese, Bonnie B., Bruce Vanden Bergh and Hairong Li (1994), “What Makes a Slogan
Memorable and Who Remembers It,” Journal of Current Issues and Research in
Advertising, 16 (2), 41-57
Rosenthal, Robert and Ralph L. Rosnow (2008), Essentials of Behavioral Research: Methods
and Data Analysis (3rd ed.). Boston, MA: McGraw-Hill.
Schindler, Robert (1991), "Symbolic Meanings of a Price Ending," in Holman, R.H. and
Solomon, M.R. (Ed.), Advances in Consumer Research, Vol. 18, Association for
Consumer Research, Provo, UT, 794-801.
Simonsohn, Uri (2011), “Spurious? Name Similarity Effects (Implicit Egotism) in Marriage, Job,
and Moving Decisions,” Journal of Personality and Social Psychology, 101 (1), 1-24.
Thomas, Manoj and Vicki Morwitz (2009), “The Ease-of-Computation Effect: The Interplay of
Metacognitive Experiences and Naïve Theories in Judgments of Price Differences,”
Journal of Marketing Research, 46 (1), 81 91.
Vanhuele, Marc, Gilles Laurent and Xavier Dreze (2006), “Consumers’ Immediate Memory for
Prices,” Journal of Consumer Research, 33 (2), 163 72.
Vohs, Kathleen. D. and Todd F. Heatherton (2001), “Self-esteem and Threats to Self:
Implications for Self-Construals and Interpersonal Perceptions,” Journal of Personality
and Social Psychology, 81 (December), 1103-1118.
Vohs, Kathleen D. and Todd F. Heatherton (2004), “Ego Threat Elicits Different Social
Comparison Processes Among High and Low Self-esteem People: Implications for
Interpersonal Perceptions,” Social Cognition, 22 (February), 168–191.
Wertenbroch, Klaus and Bernd Skiera (2002), “Measuring Consumers’ Willingness to Pay at the
Point of Purchase,” Journal of Marketing Research, 39 (May), 228-241.
Zajonc, Robert B. (1980), “Feeling and Thinking: Preferences Need No Inferences,” American
Psychologist, 35 (2), 151-175.
Zajonc, Robert B. (1998), “Emotions,” in D. Gilbert, S.T. Fiske, and G. Lindzey (Eds.),
Handbook of Social Psychology, 4th Edition, 591-632. New York, NY: McGraw-Hill.
44
Studies with yocked sample Study 1: 38 flemish 6th graders; choice of letter from pairs or triads of name
Nuttin (1985) pairs. Study 2: 96 Flemish undergrads. Name letter. letters and non-name letters. Name letters were more often chosen.
Name letter (First and Last
Johnson (1986) One study. Study with 597 subjects initial letters). Attractiveness rating on a 9-point sclae. Name letters were more attractive.
Study run with 13 country 2047 undergrads studying in 12 Name letters were more often chosen.
samples and 12 language languages studiyng in numerous Name Letters (First name These results generalize across
Nuttin (1987) samples. universities in different countries. and/or Family name) choice of six most preferred letters countires.
Difference between Greater choice of name letter and
Study 1: 66 flemish high school the proportions of name-letters selected by attractiveness of name letter effect.Rules
students; Study 2: 136 flemish the person vs. other in the yoked couple out name letter frequency
Hoorens and Nuttin high school students; Study 3: (Study 1 & 2) . Ranking of alphabet from 1 overestimation as an alternative
(1993) Experimental studies 200 Flemish students Name letters to 26 in terms of attractiveness. explanation.
Name letter effects were found in both
Hoorens and Todorova Study with a sample of Belgian Name letters (Full, First, Choice of letters in both Cyrillic alphabet and languages and stronger in their normal
(1988) students. 100 Bulgarian undergrads. Family and Initials) Roman language. language --Cyrillic.
Jones, Pelham, Mirenberg Name Letter (First name initial Rated liking of 26 capital letters on a 9 point
and Hetts (2002) Study of name effects. 143 undergraduates and Last name initial). scale. Name letters were preferred.
Rated liking of 1-33 numbers on a 9-point
Study of birthday effects. 143 undergraduates Birthday scale. Birthday numbers were preferred.
11-point liking scale for a dating context
Jones, Pelham, Carvallo Study 6: 86 undergraduate men: Name Letter (First or Last (Study 6), and picking jersey that had
and Mirenberg (2004) Two experiments. Study 7: 22 undergraduates. name). number that was paired to name. Name letter effect were supported.
Study 5a: 52 undergraduate
women; Study 5b: 110 Birthday number effects (Data Experiments focused on 11-point liking scale
Two experiments. undergraduate women and Month match). for a dating context (Study 5a, Study 5b). Birthday effects were supported.
Archival
Proportion living in state with name Increased tendency to live in state that
Pelham, Mirenberg and correspondence compared to non-name name resembles. Also similar results in
Jones(2002) Archival date Name. correspondence. eight largest US & Canadian cities.
Archival data focused on marrying others
Jones, Pelham, Carvallo Name Letter (First or Last whose first or last name resembled their
and Mirenberg (2004) Four archival studies. name). name Name letter effect were supported.
Looked at a variety of performances: Name letter effect can also lead to
Nelson and Simmons Strikeouts (Study 1), GPA (Study 2), Tier negative performance highlighting
(2007) Three archival studies First name, of Law school attended (Study 4) unconscious processing.
Calculated probability of matches between
Belgian data base (582,0007 Name Effect (first three letters these letters and the employees company Name letter matches were greater than
Anseel and Duyck (2008) Archival study. employees). of last name) name. the expected probability.
45
Table 2
(Field) Study 3
Seven-Figure Digit Place/Price Name Correspondence
1 X X X
2
3
4
5
6
7
8
9
10 X X X X
11 X X X X
12 X X X X
Figure 2
Study 1 Results
a. Liking
$622 $688
4.42 4.43
3.29 3.38
b. Purchase Intention
$622 $688
4.91 5.13
3.88 3.75
Figure 3
Study 2 Results (Rehearsal Conditions)
a. Purchase Intentions
$8.66 $8.55
4
5.63 5.6
3
3.95 3.89
2
1
Last Name: F Last Name: S
b. Purchase Intentions
$8.66 $8.55
3
4.1 4.1 4.1 4.2
2
1
Last Name: F Last Name: S
49
APPENDIX A
Price Liking [Cronbach α = .93 (S1); .92 (S2); .74 (S2A); .94 (S4), .92(S5)]
How much did you like this price? [1-7 (dislike extremely/like extremely)]
What were your feelings about the listed sale price? [1-7 (dislike extremely/like extremely)]
Rate the degree that you liked this price [1-7 (not at all/very much)]
Purchase Intention [Cronbach α = .91 (S1); .89 (S2); .87 (S2A); .91 (S4), .88 (S5)]
What is the likelihood that you would buy the advertised brand if it were for sale in the local
area? [1-7 (definitely would not/definitely would)]
Rate the probability that you would buy the brand if it were for sale in the local area. [1-7
(definitely will not/definitely will)]
How likely do you feel you would be to buy the advertised item at the listed price?
[1-7 (very unlikely/very likely]