

149
臺大管理論叢
第
28
卷第
1
期
increased in the condition of incongruence than that of congruence.
However, the magnitude of increase may be different for consumers with different
dominant inferences. In the condition of congruence, the quality and social signals
conveyed from the information of sales volume and breadth of appeal are apparent and
both indicate that the broad-appeal product is a better purchase choice. On the other hand,
in the condition of incongruence, the quality signal for consumers with quality inference is
still apparent and indicates that the narrow-appeal product is a better purchase choice.
However, as for consumers with social inference, the higher sales of the narrow-appeal
product and the broad-appeal product with lower sales both represent a majority’s choice
(implying two potential reference groups), which causes consumers’ purchase decision on
the narrow-appeal product to be less concentrated than for those with the quality
inference. Therefore, we conclude that the probability of purchasing a narrow-appeal
product significantly increases in the condition of incongruence compared to that of
congruence; yet the magnitude of increase is higher for consumers with the quality
inference than for those with the social inference.
H2: When a product’s breadth of appeal and sales volume are incongruent in
e-commerce, the probability of purchasing a narrow-appeal product
significantly increases in comparison to the condition of congruence; yet the
magnitude of increase is higher for consumers with the quality inference than
for those with the social inference.
3. Research Methodology
3.1 Experimental Design
The entire experiment was conducted in the context of e-commerce. The key task
assigned to the participants was to purchase three kinds of food products in a given
commercial website: cookies, drinks and chips. Participants were given two choices for
each kind of product, a broad-appeal and narrow-appeal product (e.g., chocolate cookie,
which is widely available vs. cinnamon cookie, which is an unusual flavor only to be
found in certain stores) and were required to choose one among the two alternatives.
Therefore, the dependent variable was consumers’ purchase decision, which reflected the
probability of purchasing a broad-appeal or narrow-appeal product. The independent
variable was the congruency of popularity information (congruence vs. incongruence). A
product’s sales volume and breadth of appeal are congruent when a broad-appeal product
has a higher sales volume than a narrow-appeal product; whereas, it is the condition of