1954
Exploring Decision Rules for Sellers in Business-to-Consumer (B2C) Internet Auctions
In the tree-building process, CART requires
that the user select a computational method
for validating the tree. CART provides cross
validation in which the dataset is divided into N
sub-datasets. N-1 subsets are used as training
datasets, and the remaining dataset is used for
testing the model. To validate our trees, we use
10-fold cross-validation, a procedure in which
nine subsets are used as a training sample, and
one subset is used as a test sample (Steinberg &
Colla, 1997). In the10-fold cross-validation pro-
cess, the data are divided into approximately 10
equal subsets, where subsets are determined by
random sampling on the criterion variable, and
the tree-growing process is repeated 10 times.
RESULTS
Figure 1 and Figure 2 demonstrate the decision
trees of the two different data sets that have been
induced using CART analysis. Table 4 summarizes
the decision rules derived from the trees. In the
¿UVWGDWDVHWWKH'9'PRYLHWKHFODVVL¿FDWLRQ
tree analysis shows that if the initial bid price
LVJUHDWHUWKDQWKHQWKH¿QDOELGSULFHLV
DERYHWKHDYHUDJH¿QDOELGGLQJSULFH,Q RWKHU
words, when sellers set the initial bid price at a
level greater than $9.63, these sellers earn a price
premium (PP). This result (shown both at the top
of Figure 1 and also as Rule 1 in Table 4) shows
WKDWWKHLQLWLDOELGSULFHVLJQL¿FDQWO\LPSDFWVWKH

¿QDOELGSULFHD¿QGLQJFRQVLVWHQWZLWKHDUOLHU
regression-based Internet auction studies). The
remaining rules pertain to situations in which the
sellers’ initial bid price is less than $9.63. These
rules together show that shipping cost, positive
feedback, and the auction ending time are im-
portant determinants of price premiums. Rule 2
(see both Figure 1 and Table 4) shows that if the
shipping cost is less than or equal to $3.97, a price
premium is earned. Rules 3 through 5 indicate that,
in the situation of relatively-high shipping cost,
positive feedback and an ending time during the
PM hours of the weekday are predictors of price
premiums. In Rule 3, if the seller has less than
549.5 positive feedback ratings, the seller fails to
earn a price premium. Rule 4 and Rule 5 together
show that if the seller has more than 549.5 positive
feedback ratings, the ending time of the auction
plays an important role. If the auction does not
end during the PM hours of a weekday, the items
fail to earn a price premium.
Initial
Bid Price (IB)
n=404
Shipping
Cost (SC)
n=343
Positive
Feedback (PF)
n=228

Auction Ending
Time (WDPM)
n=68
PP
n=61
IB > $9.63
PP
n=115
IB <= $9.63
SC<=$3.97SC> $3.97
NPP
n=160
PF<=549.5 PF> 549.5
PP
n=41
NPP
n=27
NO YES
PP: Price Premium Group
NPP: Non Price Premium Group
IB: Initial Bid Price
SC: Shipping Cost
PF: Positive Feedback
WDPM: Weekday Afternoon Auction Ending Time
LEGEND
Figure 1. Decision tree for DVD movie
1955
Exploring Decision Rules for Sellers in Business-to-Consumer (B2C) Internet Auctions
7KH¿WVWDWLVWLFVIRUWKHPRGHODUHDVIROORZV
Resubstitution relative cost for the optimal tree is

0.428 and its complexity is 0.006. In addition, error
UDWHVIRUPLVFODVVL¿FDWLRQEDVHGRQWKHFULWHULRQ
variable are 0.24 for price premiums and 0.23 for
non-price premium auctions. In other words, 24%
Shipping
Cost (SC)
n=366
Initial Bid (IB)
n=219
Initial Bid (IB)
n=147
SC<=14.50 SC > 14.50
PP
n= 17
PP
n= 82
Initial Bid (IB)
n=65
NPP
n= 48
NPP
n = 196
PP
n= 23
IB<=0.88
IB > 0.88
IB<=107.55
IB > 107.55
IB<=182.50 IB > 182.50
PP: Price Premium Group

NPP: Non Price Premium Group
SC: Shipping Cost
IB: Initial Bid Price
LEGEND
Figure 2. Decision tree for MP3 player
Rule for Terminal Node
&ODVVL¿FDWLRQ7UHH$QDO\VLV'HSHQGHQW9DULDEOH )LQDO%LG3ULFH
Rule 1 IF Initial Bid Price (IB) > $9.63, THEN Price Premium Group (PP).
Rule 2
IF Initial Bid Price (IB) <= $9.63 AND Shipping Cost (SC) <= $3.97, THEN Price Premium
Group (PP).
Rule 3
IF Initial Bid Price (IB) <= $9.63 AND Shipping Cost (SC) > $3.97 AND Positive Feedback
(PF) <= 549.5, THEN Non Price Premium Group (NPP).
Rule 4
IF Initial Bid Price (IB) <= $9.63 AND Shipping Cost (SC) > $3.97 AND Positive Feedback
3)!$1'(QGLQJ7LPH:HHNGD\DIWHUQRRQ7+(13ULFH3UHPLXP*URXS33
Rule 5
IF Initial Bid Price (IB) <= $9.63 AND Shipping Cost (SC) > $3.97 AND Positive Feedback
(PF) > 549.5 AND Ending Time = Weekday afternoon, THEN Non Price Premium Group
(NPP).
5HJUHVVLRQ7UHH$QDO\VLV'HSHQGHQW9DULDEOH 1XPEHURI%LGV
Rule 1 IF IB <= $ 3.15, THEN Average Number of Bids (AVG-NB) = 9.29
Rule 2 IF IB > $3.15 AND SC > $3.9, THEN Average Number of Bids (AVG-NB) = 2.93.
Rule 3
IF IB > $3.15 AND IB <= $9.25 AND SC <= $3.9, THEN Average Number of Bids (AVG-
NB) = 6.28
Rule 4 IF IB > $9.25 AND SC <= $3.9, THEN Average Number of Bids (AVG-NB) = 2.73
Table 4. Decision rules based on CART analysis for DVD dataset (N=404)
1956

Exploring Decision Rules for Sellers in Business-to-Consumer (B2C) Internet Auctions
RIREVHUYDWLRQVDUHPLVFODVVL¿HGWR133DQG
RIREVHUYDWLRQVDUHPLVFODVVL¿HGWR33
In regression-tree analysis, the initial bid
price and shipping cost are the most important
SUHGLFWRUVMXVWDVLQFODVVL¿FDWLRQWUHHDQDO\VLV
(see the bottom half of Table 4 for regression-
tree analysis). Rule 1 shows that if the initial bid
price is less than or equal to $3.15, the average
number of bids is 9.29. If the initial bid price is
set above $3.15 and the shipping cost is greater
than $3.90, then the average number of bids is
2.93 (Rule 2). If, however, the initial bid price is
set above $3.15 but the shipping cost is less than
$3.90, then the average number of bids is 6.28
(Rule 3). Finally, if the initial bid price is greater
than $9.25 and the shipping cost is less than or
equal to $3.90, then the average number of bids
is 2.73. In regression analysis, CART does not
UHSRUWDPLVFODVVL¿FDWLRQUDWH
,QWKH03SOD\HUGDWDVHWWKH¿UVWUXOHVKRZV
that if the shipping cost (SC) is greater than
$14.50 and the initial bid price (IB) is less than or
equal to $182.50, then no price premium (NPP)
is earned (see Figure 2 and Table 5, Rule 1). If,
however, the shipping cost (SC) is less than or
equal to $14.50 and the initial bid price is greater
than $182.50, then a price premium (PP) is earned
(Rule 2). These results show that the bidders were
VLJQL¿FDQWO\LQÀXHQFHGE\VKLSSLQJFRVWDQGWKH

initial bid price. The remaining rules pertain to
Rule for Terminal Node
&ODVVL¿FDWLRQ7UHH$QDO\VLV'HSHQGHQW9DULDEOH )LQDO%LG3ULFH
Rule
1
IF SC > $14.50 AND IB < = $182.50, THEN NPP
Rule
2
IF SC > $14.50 AND IB > $182.50, THEN PP
Rule
3
IF SC <= $14.50 AND IB <= $0.88, THEN PP
Rule
4
IF SC <= $14.50 AND IB > $0.88 AND IB <= $107.55, THEN NPP
Rule
5
IF SC <= $14.50 AND IB > $107.55, THEN PP
5HJUHVVLRQ7UHH$QDO\VLV'HSHQGHQW9DULDEOH 1XPEHURI%LGV
Rule
1
IF IB > $135.40, THEN Average Number of Bids (AVG-NB) = 2.27
Rule
2
IF IB >= $85.00 AND IB < $135.40, THEN Average Number of Bids (AVG-NB) = 14.44
Rule
3
IF IB >= $7.50 AND IB < 85.40, THEN Average Number of Bids (AVG-NB) = 21.34
Rule
4

IF IB < $7.50 AND PF > 2635.5, THEN Average Number of Bids (AVG-NB) = 25.74
Table 5. Decision rules based on CART analysis for MP3 player dataset (N=366)
1957
Exploring Decision Rules for Sellers in Business-to-Consumer (B2C) Internet Auctions
situations in which the shipping cost is less than
or equal to $14.50. Rule 3 shows that if the ship-
ping cost is less than or equal to $14.50, and the
s e l l e r s e t s t h e i n i t i a l b i d le s s t h a n o r e q u a l t o $ 0 . 8 8 ,
then a price premium was earned. In Rule 4, if
the shipping cost is less than or equal to $14.50
and if the seller set the initial bid between $0.88
and $107.55, then the seller failed to earn a price
premium. Rule 5 shows that if the shipping cost
is less than or equal to $14.50 and if the seller
set the initial bid price higher than $107.55, then
a price premium was earned. In this dataset,
resubstitution relative cost for the optimal tree
is 0.348 and its complexity is 0.008. In addition,
error rates for the NPP group and the PP group
are 0.06 and 0.27, respectively.
In regression-tree analysis, the initial bid price
and positive feedback ratings are the most impor-
tant predictors, a slightly different result than in
WKHFODVVL¿FDWLRQWUHHDQDO\VLV5XOHVKRZVWKDW
if the initial bid price is greater than $135.40, the
average number of bids is 2.27. If the initial bid
price is above $85.00 and below $135.40, then
the average number of bids is 14.44 (Rule 2). If
the initial bid price is between $7.50 and $85.00,
then the average number of bids is 21.34 (Rule

3). Finally, if the seller sets the initial bid price
lower than $7.50, positive feedback ratings plays
DQLPSRUWDQWUROH6SHFL¿FDOO\LIWKHLQLWLDOELG
price is lower than $7.50 and positive feedback
ratings are greater than 2635.5, then the average
number of bids is 25.74.
Based on the rules derived from CART analy-
sis, initial bid price, shipping cost, and positive
feedback appear as important variables to deter-
PLQHWKH¿QDOELGSULFHIRUERWKSURGXFWV
CONCLUDING REMARKS
Implications for Research
This study contributes to the literature in several
ways. This study, with its broad examination of
variables from earlier research, is a step towards
a comprehensive theoretical understanding of
B2C Internet auctions. Exploratory work to this
SRLQW KDV LGHQWL¿HG VLJQL¿FDQW YDULDEOHV DQG
here all variables have been considered concur-
UHQWO\WR¿QGWKHRQHVZLWKWKHJUHDWHVWUHODWLYH
import. Thus, this study may provide guidance
DV UHVHDUFKHUV EHJLQ WR GHYHORS GH¿QLWLYHOLVWV
of variables that impact the outcome of Internet
DXFWLRQV7KHFUHDWLRQRIVXFKGH¿QLWLYHOLVWVKDV
EHHQLGHQWL¿HGDVDQHFHVVDU\SUHOLPLQDU\VWHSWR
developing theory (Ba & Pavlou, 2002; Pavlou,
2002; Weick, 1995; Whetten, 1989).
0RUHVSHFL¿FDOO\ZHKDYHEHHQDEOHWRLGHQ-
tify three independent variables—shipping cost,
initial bid price, and reputation—that may play

a larger role in Internet B2C auctions than was
SUHYLRXVO\UHDOL]HG,QSDUWLFXODUWKHLGHQWL¿FDWLRQ
of shipping cost as a primary determinant of price
premiums has not been previously reported by
U H V H D U F K H U V  2 X U ¿ Q G L QJ V OH D G X V W R W K H F R Q FO X V L R Q 
t h a t s h i p p i n g c o s t i s t h e s i n g l e m o s t i m p o r t a n t f a c -
tor in earning price premiums in Internet auctions.
This is somewhat surprising in light of the fact
that so few researchers have considered shipping
cost as a part of their models. Positive feedback
was also found to be a critical factor in the ability
WRHDUQSULFHSUHPLXPV7KLV¿QGLQJLVLQOLQH
with recent studies that place seller reputation in
a place of great importance in Internet auctions
(Ba & Pavlou, 2002; Bruce, Haruvy, & Rao, 2004;
McDonald & Slawson, 2002; Ottaway, Bruneau,
& Evans, 2003). Initial bid price was found to be a
critical factor in our study, but has not been found
WREHVLJQL¿FDQWLQRWKHUV*LONHVRQ5H\QROGV
  :H D UJ XH W K D WL Q LW LD O ELGS U LF HL VV LJ Q L ¿F D QW 
because a low initial bid price attracts the greatest
number of possible bidders. Sellers desire to have
the largest possible number of bidders, because
having more bidders helps to ensure that an item
does not remain unpurchased. Because of the
inconclusive history of this variable, further ex-
amination is warranted. Finally, ending time has
EHHQSURSRVHGDVDVLJQL¿FDQWIDFWRULQSRSXODU
1958
Exploring Decision Rules for Sellers in Business-to-Consumer (B2C) Internet Auctions

literature (Ribeiro, 2004; Witt, 2005) but, to our
knowledge, has not been conclusively linked to
auction outcome until this study found a weekday
DIWHUQRRQHQGWLPHWREHDVLJQL¿FDQWSUHGLFWRU
of the ability to earn a price premium.
We a l s o n o t e t h a t t h e C A RT a n a l y s i s p e r f o r me d
here produces results that are similar to the vast
majority of existing empirical work on Internet
auctions. The majority of previous work has used
some form of regression analysis to reach its con-
clusions; CART corroborates these results using
a different methodology. CART has been used in
e-commerce research (Kim, Lee, Shaw, Chang,
& Nelson, 2001), but has not, to our knowledge,
EHHQDSSOLHGWRWKHVSHFL¿FWRSLFRI,QWHUQHWDXF-
tions. One of the strengths of CART is that it is a
non-parametric technique, which means that no
assumptions are made regarding the distributions
of the predictor variables. Normal, non-normal,
skewed, categorical, and ordinal variables can be
included in CART analyses. These conditions may
be present in the datasets collected by Internet
auction researchers, thus making CART a poten-
tially useful analytic tool. For these reasons, the
application of CART in Internet auction research
is a new methodological contribution.
Implications for Practice
Internet Auctions have become a popular sales
and marketing channel for businesses seeking to
HQKDQFHSUR¿WV2IIHULQJDXFWLRQVHUYLFHVVHOOLQJ

by auction, and adopting the appropriate auction-
pricing policy may increase the attractiveness of a
Web site, reduce the inventory cost of slow-selling
products, reduce transaction costs, and provide
valuable insight into customer preferences. In
some cases, retailers have realized revenue gains
of 50% or more on excess inventory sold through
Internet auctions rather than through liquidators
(Gentry, 2003).
)RUWKHEHQH¿WRISUDFWLWLRQHUVZHKDYHGHP-
onstrated the use of a relatively new tool, CART
Analysis, which can be used to investigate auc-
tions. Businesses can use this tool, available in
a number of data mining software packages, to
glean insights from their data about how to most
effectively list items for auction. Useful outputs
of this analysis technique are the decision tree,
which gives a fairly intuitive visual representation
RIWKHFULWLFDOYDULDEOHVZKLFKDUHLGHQWL¿HGDQG
the table of decision rules, which demonstrates the
priority of the critical factors. CART can provide
complex, exact models that include decision rules
for all variables considered in a given analysis.
Here, we have attempted to strike a balance
between simplicity and detail by demonstrating
a simpler decision tree that indicates only the
most important variables. Outputs are simple to
interpret (particularly for individuals with little
statistical training in regression techniques), pro-
vide precise results, and suggest logical, sequential

decisions to practitioners. Also, we note that CART
analysis is essentially a data-driven method; it is
a tool that can be useful even when the analyst
KDVOLWWOHH[SHULHQFHLQVHOOLQJWKHVSHFL¿FSURGXFW
for which he or she may be developing decision
rules. Finally, as noted in the previous section,
CART does not require that the data be normally
distributed, thus making it a viable option in
situations where techniques with more stringent
assumptions will not work.
0RUH SUDFWLFDOO\ EDVHG XSRQ RXU ¿QGLQJV
merchants desiring to utilize Internet auctions
should consider competing with other sellers on
the bases of shipping cost, reputation, initial bid
price, and auction ending time. We urge caution,
KRZHYHULQDSSO\LQJWKHVH¿QGLQJVWRSURGXFWV
that differ from the ones used in our study. The
VSHFL¿FGHFLVLRQUXOHVJHQHUDWHGE\RXUDQDO\VLV
may not be widely generalizable to other products
(or even to other movie titles or other MP3 play-
ers). Investigations of other types of products may
generate different decision rules. We encourage
sellers to investigate each of the variables listed
HDUOLHU LQ 7DEOH  WR ¿QG WKH IDFWRUV WKDW DUH
most important for the particular product being
auctioned. Nevertheless, the growing corpus
1959
Exploring Decision Rules for Sellers in Business-to-Consumer (B2C) Internet Auctions
of literature on Internet auctions indicates that
shipping cost, reputation, initial bid price, and

auction ending time may be important in the sale
of many other types of items as well (e.g., Brint,
2003; Gilkeson & Reynolds, 2003; McDonald &
6ODZVRQ6WDQGL¿UG
Limitations
2QHOLPLWDWLRQRIWKLVVWXG\LVWKHXVHRID³JUHHG\´
FODVVL¿FDWLRQDOJRULWKP&ODVVL¿FDWLRQWUHHVXVH
ZKDWLVNQRZQDVD³JUHHG\´DOJRULWKPWRGHWHU-
mine splits in the dataset (Harrison, 1997). Some
KDYHFULWLFL]HGWKHXVHRIFODVVL¿FDWLRQWUHHVEH-
FDXVHRIWKHXVHRI³JUHHG\´DOJRULWKPVLQVWHDG
arguing that splits should be made based upon
two or more levels at once (TwoCrows, 1999).
(VVHQWLDOO\WKH³JUHHG\´DOJRULWKPH[HFXWHVLWV
task without considering the impact that any split
m a y h a ve o n s u b s e q u e n t s p l i t s ( Tw o C r o w s , 199 9).
Other criticisms come from researchers who have
H[SUHVVHGDGHVLUHIRUFODVVL¿FDWLRQV\VWHPVWKDW
make multivariate rather than univariate splits
(TwoCrows, 1999). These debates are outside the
scope of this study. However, it is worth noting
WKDWLQVSLWHRIWKHVHWKHRUHWLFDOLVVXHVFODVVL¿FD-
tion trees continue to be widely used and trusted
in data mining applications by researchers and
software developers.
Another limitation of our study is that we have
analyzed data on only two products. While the
fact that we have only examined a DVD movie
and an MP3 player does not diminish the fact
that CART analysis has been demonstrated as a

useful technique in Internet auction research and
practice, it is a limitation from the standpoint of
LGHQWLI\LQJVLJQL¿FDQWLQGHSHQGHQWYDULDEOHV$
broader selection of products and different types
of analysis might generate more broadly-appli-
cable guidelines for Internet auction retailers.
For instance, both DVDs and MP3 players are
small, easily transportable, internationally-used
products. It is conceivable that different products
would generate decision trees with different criti-
F D O I D F W R U V  7 K X V  Z K L O H R X UV S H F L ¿ F G H F L V LR Q U XO H V 
may not be widely applicable, the technique that
we have demonstrated is.
Similarly, we have collected data from only
one Internet auctioneer. While eBay has a larger
international presence than any other Internet
auctioneer, we remind readers that caution should
EHXVHGZKHQDSSO\LQJWKH¿QGLQJVKHUHWRDXF-
tions conducted in different contexts.
A different type of data analysis might also
SUHVHQW DGGLWLRQDO XVHIXO ¿QGLQJV 2WKHU DS-
proaches for analyzing Internet auction datasets
include, but are not limited to, binary logistic
UHJUHVVLRQZLWKWKHDXFWLRQ¶V¿QDOFORVLQJSULFH
as a binary dependent variable, for instance) and
multiple regression (perhaps with number of bids
as a dependent variable). Using the same data from
which the decision trees were generated, both
binary logistic regression and multiple regression
LGHQWLI\SUHGLFWRUVYHU\VLPLODUWRWKRVHLGHQWL¿HG

in CART analysis (interested readers may consult
Appendix B for these results). These analyses
LGHQWLI\VLJQL¿FDQWLQGHSHQGHQWYDULDEOHVEXW
do not generate sequential decision rules. Thus,
it is likely that such techniques may be of greater
interest to researchers than to practitioners. One
¿QDOSRLQWZLWKUHJDUGWRDOWHUQDWHGDWDDQDO\VLV
techniques deserves mention. The reader will
recall from earlier discussion that regression
techniques have more stringent assumptions than
CART, and thus regression may not be appropriate
for analysis of all datasets.
Future Research
One opportunity for future research is the inves-
tigation of auctions of different types of products.
DVDs and MP3 players represent only a small
fraction of the myriad items that are auctioned on
the Internet. It is conceivable that auctions of dif-
ferent types of items may yield different decision
rules. An examination of how the type of product
impacts auction outcome may be a fruitful area
of inquiry for researchers.
1960
Exploring Decision Rules for Sellers in Business-to-Consumer (B2C) Internet Auctions
Future work will also need to delve into the
motivations for bidder, seller, and auctioneer
behavior. This is necessary in order to more
FRPSOHWHO\H[SODLQWKHVLJ QL ¿FDQFHRI¿QGLQJVLQ
Internet auction studies. Some researchers have
examined the roles that bidder experience (Ward

& Clark, 2002; Wilcox, 2000), bidder strategy
(Bapna, Goes, & Gupta, 2003; Easley & Tenorio,
2004), bidder acceptance of technology (Stafford
& Stern, 2002), and bidder motivation (Cameron
*DOORZD\6WDQGL¿UG5RHORIV'XUKDP
2004) play in Internet auctions. As this study has
assisted in the development of sellers’ decision
rules and strategies for Internet auctions, we be-
lieve future research should continue to examine
bidders’ decision rules and strategies as well.
)XWXUHZRUNFRXOGLQWHJUDWH¿QGLQJVIURPWKHVH
WZRVWUHDPVRI,QWHUQHWDXFWLRQUHVHDUFK¿WWLQJ
the seller model of behavior to the bidder model.
The ultimate goal should be the development of
a comprehensive model of buyer and seller char-
acteristics, motivation, and behavior.
REFERENCES
Agarwal, R., Arning, A., Bollinger, T., Mehta, M.,
Shcafer, J., & Srikant, R. (1996). The quest data
mining system. Paper presented at the 2nd Inter-
national Conference on Knowledge Discovery in
Databases and Data Mining, Portland, OR.
Ba, S., & Pavlou, P. A. (2002). Evidence of the
effect of trust building technology in electronic
markets: Price premiums and buyer behavior. MIS
Quarterly, 26(3), 243-269.
Ba, S., Whinston, A. B., & Zhang, H. (2003).
Building trust in online auction markets through
an economic incentive mechanism. Decision Sup-
port Systems, 35(3), 273-286.

Bapna, R., Goes, P., & Gupta, A. (2000). A theo-
retical and empirical investigation of multi-item
on-line auctions. Information Technology and
Management, 1(1-2), 1-23.
Bapna, R., Goes, P., & Gupta, A. (2001). Insights
and analyses of online auctions. Association for
Computing Machinery. Communications of the
ACM, 44(11), 42-50.
Bapna, R., Goes, P., & Gupta, A. (2003). Analysis
and design of business-to-consumer online auc-
tions. Management Science, 49(1), 85-102.
Bapna, R., Goes, P., Gupta, A., & Karuga, G.
(2002). Optimal design of the online auction
channel: Analytical, empirical, and computational
insights. Decision Sciences, 33(4), 557-578.
Berry, M., & Linoff, G. (1997). Data mining
techniques for marketing, sales, and customer
support. New York: John Wiley & Sons.
Breiman, L., Friedman, J., Olshen, R., & Stone,
C. (1984). &ODVVL¿FDWLRQ DQG UHJUHVVLRQ WUHHV.
Belmont, CA: Wadsworth International Group.
Brint, A. T. (2003). Investigating buyer and
seller strategies in online auctions. The Journal
of the Operational Research Society, 54(11),
1177-1188.
Bruce, N., Haruvy, E., & Rao, R. (2004). Seller rat-
ing, price, and default in online auctions. Journal
of Interactive Marketing, 18(4), 37-51.
Budish, E., & Takeyama, L. N. (2001). Buy prices
in online auctions: Irrationality on the Internet?

Economics Letters, 72(3), 325-334.
Cameron, D., & Galloway, A. (2005). Consumer
motivations and concerns in online auctions. In-
ternational Journal of Consumer Studies, 29(3),
181-193.
Chong, B., & Wong, M. (2005). Crafting an ef-
fective customer retention strategy: A review of
halo effect on customer satisfaction in online
auctions. International Journal of Management
and Enterprise Development, 2(1), 12-26.
1961
Exploring Decision Rules for Sellers in Business-to-Consumer (B2C) Internet Auctions
Crockett, R. O. (1999). Going, going richer.
Business Week (3659), EB16.
Dholakia, U. M. (2005a). Concept discovery,
process explanation, and theory deepening in e-
marketing research: The case of online auctions.
Marketing Theory, 5(1), 117-124.
Dholakia, U. M. (2005b). The usefulness of bid-
ders’ reputation ratings to sellers in online auc-
tions. Journal of Interactive Marketing, 19(1),
31-41.
Dholakia, U. M., & Soltysinski, K. (2001). Coveted
or overlooked? The psychology of bidding for
comparable listings in digital auctions. Marketing
Letters, 12(3), 225-237.
Ding, M., Elishaberg, J., Huber, J., & Saini, R.
(2005). Emotional bidders - An analytical and
experimental examination of consumers’ behavior
in a Priceline-like reverse auction. Management

Science, 51(3), 352-365.
Easley, R. F., & Tenorio, R. (2004). Jump bidding
strategies in Internet auctions. Management Sci-
ence, 50(10), 1407-1420.
eBay (2006). eBay Inc. announces fourth quarter
DQG IXOO \HDU  ¿QDQFLDO UHVXOWV 5HWULHYHG
April 3, 2006, from www.investor.ebay.com/news/
Q105/ebay0420-2932ld.pdf
Engelbrecht-Wiggans, R. (1980). Auctions and
bidding models: A survey. Management Science,
26(2), 119-143.
Geng, X., Stinchcombe, M., & Whinston, A.
B. (2001). Radically new product introduction
using on-line auctions. International Journal of
Electronic Commerce, 5(3), 169-189.
Gentry, C. R. (2003). Recycled returns. Chain
Store Age, 79(1), 72-73.
Gilkeson, J. H., & Reynolds, K. (2003). Determi-
nants of Internet auction success and closing price:
An exploratory study. Psychology & Marketing,
20(6), 537-566.
Gregg, D., & Walczak, S. (2003). E-commerce
auction agents and online-auction dynamics.
Electronic Markets, 13(3), 242-251.
Grow, B. (2002). Excess inventory? EBay to the
rescue. Business Week, 8.
Harrison, J. G. (1997). Enhancements to the data
mining process. Unpublished Ph.D., Stanford
University, CA.
Hu, X., Lin, Z., Whinston, A. B., & Zhang, H.

(2004). Hope or hype: On the viability of escrow
services as trusted third parties in online auction
environments. Information Systems Research,
15(3), 236-250.
Kagel, J. H. (1995). A survey of experimental
research. In J. H. Kagel & A. E. Roth (Eds.),
The handbook of experimental economics (pp.
501-586). Princeton, NJ: Princeton University
Press.
Kannan, P. K., & Kopalle, P. (2001). Dynamic pric-
ing on the Internet: Importance and implications
for consumer behavior. International Journal of
Electronic Commerce, 5(3), 63-83.
Kim, J. W., Lee, B. H., Shaw, M. J., Chang, H. -L.,
& Nelson, M. (2001). Application of decision-tree
induction techniques to personalized advertise-
ments on Internet storefronts. International Jour-
nal of Electronic Commerce, 5(3), 45-62.
. OH L Q  %   / HI ÀH U.  %    7 K H U R OH R I P D U N H W 
forces in assuring contractual performance. The
Journal of Political Economy, 89(4), 615-641.
Klemperer, P. (1999). Auction theory: A guide
to the literature. Journal of Economic Surveys,
13(3), 227-286.
Klemperer, P. (Ed.). (2000). The economic theory
of auctions. Cheltenham, UK: Edward Elgar.
Krishna, V. (2002). Auction theory. San Diego,
CA: Academic Press.
1962
Exploring Decision Rules for Sellers in Business-to-Consumer (B2C) Internet Auctions

Lhose, G. L., Biolsi, K., Walker, N., & Reuter, H. H.
$FODVVL¿FDWLRQRIYLVXDOUHSUHVHQWDWLRQV
Communications of the ACM, 37(12), 36-49.
List, J. A., & Lucking-Reiley, D. (2000). Demand
reduction in multiunit auctions: Evidence from
D VSRUWVFDUG ¿HOG H[SHULPHQW The American
Economic Review, 90(4), 961-982.
/X F N L Q J 5 H L OH \'   8V L Q J ¿H OG H [ S H U L P H Q W V 
to test equivalence between auction formats:
Magic on the Internet. The American Economic
Review, 89(5), 1063-1081.
Lucking-Reiley, D. (2000a). Auctions on the Inter-
net: What’s being auctioned, and how? The Jour-
nal of Industrial Economics, 48(3), 227-253.
Lucking-Reiley, D. (2000b). Vickrey auctions in
practice: From nineteenth-century philately to
WZHQW\¿UVWFHQWXU\ HFRPPHUFHThe Journal
of Economic Perspectives, 14(3), 183-193.
McAfee, R. P., & McMillan, J. (1987). Auctions
and bidding. Journal of Economic Literature,
25(2), 699-739.
McDonald, C. G., & Slawson, V. C. (2002). Repu-
tation in an Internet auction market. Economic
Inquiry, 40(4), 633-650.
Mehta, K. (2002). Understanding revenue driv-
ers of electronic auctions. Unpublished Ph.D.,
University of Illinois at Chicago.
Mehta, M., Agarwal, R., & Rissanen, J. (1996).
6/,4$IDVWVFDODEOHFODVVL¿HUIRUGDWDPLQ-
ing. Paper presented at the Fifth International

Conference on Extending Database Technology,
Avignon, France.
Milgrom, P. (1985). The economics of competi-
tive bidding: A selective survey. In L. Hurwicz,
D. Schmeidler, and H. Sonnenschein (Ed.), So-
cial goals and social organization. Cambridge:
Cambridge University Press.
Milgrom, P. (1986). Auction theory. In T. E. Bewley
(Ed.), Advances in economic theory. Cambridge:
Cambridge University Press.
Milgrom, P. R., & Weber, R. J. (1982). A theory of
auctions and competitive bidding. Econometrica,
50(5), 1089-1122.
Neo, B. S. (1992). The implementation of an
electronic market for pig trading in Singapore.
Journal of Strategic Management Systems, 1(5),
278-288.
Oh, W. (2002). C2C versus B2C: A comparison
of the winner’s curse in two types of electronic
auctions. International Journal of Electronic
Commerce, 6(4), 115-138.
Ottaway, T. A., Bruneau, C. L., & Evans, G. E.
(2003). The impact of auction item image and
b u y e r / s e l l e r f e e d b a c k r a t i n g o n e l e c t r o n i c a u c t i o n s .
The Journal of Computer Information Systems,
43(3), 56-60.
Pavlou, P. A. (2002). Trustworthiness as a source of
competitive advantage in online auction markets.
Academy of Management Proceedings, A1-A7.
Porter, R., & Shoham, Y. (2004). On cheating in

sealed bid auctions. Decision Support Systems,
(39), 41-54.
Quinlan, J. R. (1990). Induction of decision trees.
I n J . W. C h a v l i k , W. J u d e & T. G . D i e t t e r i c h ( E d s .),
Readings in machine learning (pp. 57-69). San
Mateo, CA: Morgan Kauffman.
Ribeiro, J. (2004, March 2). Researcher compiles
auction data for online merchants. Retrieved June 7,
2005, from />agementtopics/ebusiness/story/0,10801,90668,00.
html
Rothkopf, M. H. (1969). A model of rational
competitive bidding. Management Science, 15(7),
362-374.
1963
Exploring Decision Rules for Sellers in Business-to-Consumer (B2C) Internet Auctions
Rothkopf, M. H., & Harstad, R. M. (1994).
Modeling competitive bidding: A critical essay.
Management Science, 40(3), 364-384.
Segev, A., Beam, C., & Shanthikumar, J. G.
(2001). Optimal design of Internet-based auc-
tions. Information Technology and Management,
2(2), 121-163.
Seth, A., & Thomas, H. (1994). Theories of the
¿UP,PSOLFDWLRQVIRUVWUDWHJ\UHVHDUFKJournal
of Management Studies, 31, 165-191.
Shapiro, C. (1983). Premiums for high quality
products as returns to reputations. The Quarterly
Journal of Economics, 98(4), 659-680.
Sinha, A. R., & Greenleaf, E. A. (2000). The im-
pact of discrete bidding and bidder aggressiveness

on sellers’ strategies in open English auctions:
Reserves and covert shilling. Marketing Science,
(19), 244-266.
Sorensen, E. H., Miller, K. L., & Ooi, C. K. (2000).
The decision tree approach to stock selection.
Journal of Portfolio Management, 27(1), 42-52.
Stafford, M. R., & Stern, B. (2002). Consumer
bidding behavior on Internet auction sites. In-
ternational Journal of Electronic Commerce,
7(1), 135-150.
6WDQGL¿UG665HSXWDWLRQDQGHFRP-
merce: EBay auctions and the asymmetrical
impact of positive and negative ratings. Journal
of Management, 27(3), 279-295.
6WDQGL¿UG665RHORIV05'XUKDP<
(2004). The impact of eBay’s buy-it-now func-
tion on bidder behavior. International Journal of
Electronic Commerce, 9(2), 167-176.
Stark, R., & Rothkopf, M. H. (1979). Competitive
bidding: A comprehensive bibliography. Opera-
tions Research, 27(2), 364-391.
Steinberg, D., & Colla, P. (1997). CART: Inter-
face and documentation. San Diego, CA: Salford
Systems.
Subramaniam, R., Mittal, V., & Inman, J. J. (2004).
Revenue equivalence in online auctions: The
case of Yahoo and eBay. In P. T. L. Popkowski
Leszczyc (Ed.), Advances in consumer research
(Vol. 31, pp. 1-2).
Tsai, L. H., & Koehler, G. J. (1993). The accuracy

of concepts learned from induction. Decision
Support Systems, 10(2), 161-172.
TwoCrows (1999). Introduction to data mining
and knowledge discovery. Potomac, MD: Two
Crows Corporation.
van Heck, E., & van Damme, E. (1997). New
HQWUDQWVDQGWKHUROHRI,77KHWHOHÀRZHUDXF-
tions in the Netherlands. Paper presented at the
30th HICSS, Hawaii.
Vickrey, W. (1961). Counterspeculation, auc-
tions, and competitive sealed tenders. Journal of
Finance, 16(1), 8-37.
Vogelstein, F., Boyle, M., Lewis, P., & Kirkpat-
rick, D. (2004). 10 tech trends to bet on. Fortune,
149(4), 75-81.
Ward, S. G., & Clark, J. M. (2002). Bidding be-
havior in on-line auctions: An examination of the
eBay Pokemon card market. International Journal
of Electronic Commerce, 6(4), 139-155.
Weick, K. (1995). What theory is not, theoriz-
ing is. Administrative Science Quarterly, 40(3),
385-390.
Whetten, D. (1989). What constitutes a theoretical
contribution. Academy of Management Review,
14(4), 490-495.
Wilcox, R. T. (2000). Experts and amateurs: The
role of experience in Internet auctions. Marketing
Letters, 11
(4), 363-374.
Wilson, R. B. (1969). Communications to the edi-

tor: Competitive bidding with disparate informa-
tion. Management Science, 15(7), 446-449.