86 / 372
九尾數定價效果在不同評估模式中的侷限：分別、聯合與依序評估
86
category, each with different fictional brand names (S and K) and ending prices (nine- and
zero-ending one dollar higher). The prices of the S-branded nail clippers, battery, and
baseball cap were manipulated to have nine-ending prices with lower left-digit values ($1.99,
$2.99, $6.99), whereas the prices of the K-branded nail clippers, battery, and baseball cap
were manipulated to have zero-ending prices with higher left-digit values ($2.00, $3.00,
$7.00). A budget was given for each product category ($47 for the nail clippers; $43 for the
batteries; $228 for the baseball caps) for both brands/ending prices.
The budget figures were manipulated so that diverse quantitative results would be
generated for each product to prevent inadequate estimations (e.g., the same estimate for all
three products). The participants were randomly divided into four subgroups, each of which
was randomly assigned to one of four conditions: SE of nine-ending prices, SE of zero-
ending prices, JE of both ending prices, and SQE of both ending prices. In the SE condition,
the participants only saw either a nine-ending or zero-ending price, but not both. In the JE
condition, they saw both prices side by side at the same time. In the SQE condition, they saw
products of the same category from two different brands independently shown back to back.
After receiving the questionnaires, the participants were asked to calculate the maximum
number of products at the offered price that could be purchased within a budget for each
product category.
3.2.2 Dependent Variable
Participants answered questions regarding pricing perception by responding to the
statement, “How many $_______ priced products can you buy with a $______ budget?”
with integer numbers based on their calculation. To determine whether the drop-off
(rounding down) heuristic among quantitative estimates made by the participants was
consistent with the nine-ending effect, each estimate was allocated to one of three categories,
as proposed by Bizer and Schindler (2005). The three categories were accurate estimation,
drop-off error, and non-drop-off error. For example, quantitative estimates of 23 or 24 were
categorized as accurate estimations, since participants could purchase 23.6 nail clippers
priced at $1.99 within a budget of $47.
Quantitative estimations influenced by the nine-ending effect that resulted in the price
ending being ignored or minimized were categorized as drop-off errors. In the example
given, if participants perceived the price of the nail clippers as $1.00, then their perception
was anchored by the leftmost digit and they ignored, or dropped off, the other digits leading
to a quantity estimation of 47. Quantitative estimations between 24 (accurate) and 47 (price
ending drop-off) were all classified as drop-off errors. Estimates that could not be classified