85 / 372
臺大管理論叢
第
26
卷第
2
期
85
3.1.4 Discussion
The results from Study 1 support H1, which postulates that the difference in magnitude
perception between nine-ending and zero-ending prices is greater in the SE condition than in
the JE condition. The results also support Hsee (1996) proposal that stimulus options (nine-
ending and zero-ending prices) become relatively easier to evaluate when presented
simultaneously (jointly) rather than individually (separately). Compared to the SE condition,
there was very little difference in the mean magnitude perception between nine-ending and
zero-ending prices one level higher (e.g., $1.99 vs. $2.00) in the JE condition, indicating that
the nine-ending pricing effect was weakened in the joint condition.
3.2 Study 2
Study 2 was conducted to further test H2 and H3, which pertain to the differences in the
nine-ending pricing effect among the separate, sequential, and joint evaluation conditions. In
contrast to Study 1, which focused on the psychological interpretation of price perception,
Study 2 utilized the perception of practical purchases of nine-ending and zero-ending priced
items as manipulated by Bizer and Schindler (2005) to simulate the product purchasing
instead of a price-perceived judgment. This manipulation may not only improve the
generalization of this article but also be closely related to the real world by examining
whether the difference in quantitative estimations between nine-ending and zero-ending
priced items is greatest in the SE condition, least in the JE condition, and midway in the SQE
condition (H2 and H3).
3.2.1 Method
Study participants were a random sample of consumers (n = 158) recruited from a large
hypermarket. The sample consisted of 78 males and 80 females with a mean age of 31.4
(over two-thirds were between 25 and 55 years). Over half the respondents reported being
married, and over half completed an associate, bachelor’s, or graduate degree. Three-fourths
were employed full time (including self-employed, with < 5% unemployed and looking for
work). The sample was predominantly middle class (almost two-thirds reported an annual
household income between $40,000 and $180,000, 4% earned > $180,000, 16% reported
earning < $50,000, and 10% preferred to not respond to the income question).
This study utilized a one-factor, four-level (separate nine ending, separate zero ending,
joint nine ending and zero ending, and sequential nine ending and zero ending) design. The
participants were asked to make quantitative estimations of three products (nail clipper,
battery, and baseball cap) within a given budget. There were two items within each product