Page 149 - 33-3

NTU Management Review Vol. 33 No. 3 Dec. 2023

dissatisfied with the existing IT are likely to switch to other ITs (Bhattacherjee et al.,
2012). However, considering the effect of switching costs, when consumers compare the
benefits of switching or not switching, they sometimes lose their switching intention. This
is because the switching costs may exceed the consumer’s expectations about new IT. IT

switching is considered a risky behavior and the results are often unpredictable. Therefore,
consumers who are afraid of taking risks and are cost sensitive may be reluctant to switch
to new IT even though they are slightly dissatisfied with the existing IT. Based on the

argument above, if a health app is considered an IT, in the context of this study, switching
cost may be expected to weaken the positive impact of dissatisfaction on the intention to
switch to a new health app. Thus, we propose the following hypothesis:
Hypothesis 6 (H6): Procedural switching cost has a negative moderating effect on
dissatisfaction with the current health app.

4. Research Methodology

4.1 Questionnaire Development

In order to test our research model, we conduct a survey including items for each
of the constructs. We adopt existing validated scales from the well-established and
reliable research instruments and turn into a three-part questionnaire. The first part is a
detailed description of the purpose of this study to ensure that respondents have used

health applications (apps). The second part includes the items for each of the constructs
measured by a seven-point Likert scale. The last part is the nominal scale, collecting the
basic information of the respondents and the control variables of the research model. The
Appendix shows the questionnaire of second part.

4.1.1 Item Development
To ensure the validity of the research content, we rewrite and develop each structured
survey item from previous research to fit the health app research environment, as shown in
the Appendix. We measure each item with a seven-point Likert scale, with 1 being “very

disagree” and 7 being “very agree”. Since both research conducting and data collection are
in Taiwan, we translate the original questionnaire from English into Chinese. To ensure the
quality of the translation, we review the Chinese and English versions and conduct a pilot
study with faculties, doctoral students, and graduate students. The feedback from the pilot

141