160 /304
產品受歡迎資訊與網路購物：以消費者解讀為干擾變數
160
opinion leaders to share the functional specifications of the product either through a store
event or through blogs. In addition, e-businesses may emphasize the high sales volume of
the narrow-appeal product to stimulate consumers’ inference of high quality. With the
right kind of inference in mind and the right strategies aligned, the effect of popularity
information may be maximized and consumers’ purchase intentions may be affected as
intended.
5.5 Limitations
This research contains some limitations that warrant future research. First, the
methodology chosen for this study is a laboratory experiment, which is often criticized on
its external validity. Thus, caution needs to be taken when generalizing findings. Future
research could re-validate the findings in a real shopping website or through a field study.
Second, although previous studies have reported that product type may impact consumers’
purchase behavior, we do not treat the product type as a contingent factor in this research
because it is not within the scope of this research. Based on the recommendations
provided by Lynch (1982), conducting an experiment with homogenous product type
would allow researchers to focus their observations on the main causal relationship.
Therefore, we intentionally kept the product type in the experiment homogenous to food-
related products only. In order to minimize the impact of product type, all the selected
products are low-cost and could be frequently consumed by different levels of consumers.
Nevertheless, we still did not eliminate the possibilities that product type in some
conditions may become a contingent factor that influences the dependent variable or its
homogeneity may affect the internal validity of the experiment. Due to those conditions,
additional studies expanding to different product categories would be helpful.
Third, the selections between two breadths of appeal are geared toward the taste of
the Taiwanese/Chinese audiences, which may be different from audiences in other
countries. For example, a cinnamon-flavored cookie is considered as a narrow-appeal
product in the Taiwan market whereas it could be considered as a broad-appeal product in
America because of its wide availability. Future research could consider selecting broad-
appeal and narrow-appeal products that are more generalized to avoid the limitation of
localization. Forth, although the sales volume we assigned to each pair of product is based
on the principles used in the prior research (Tucker and Zhang, 2011), the volume (e.g.,
64) is relatively small in many conditions in e-commerce. Future research could consider
exploring different levels of sales volume with different products and examining whether