230 / 372
社會壓力、承諾與稅務員反功能行為：實驗研究
230
National Taxation Bureau (national taxation) and the Local Tax Bureau (local taxation) of
Kaohsiung City. A total of 117 tax officials participated in this experiment. Of the
participants in the experiment, 78% were female, 62% were aged between 21-40, 51% had
no more than 10 years of service, and 91% reported to have a college education. In order to
avoid errors in the responses provided by the experimental participants who had personally
experienced case-based stress when dealing with business tax collection, we attempted to
exclude tax officials who were responsible for business tax collection from the experiment.
However, since the experiment was conducted during non-office hours, it was not easy to
control site conditions and exclude business tax officials completely. The final participants
still included 14 business tax officials. The participants were randomly divided into two
experimental groups (one with superior pressure plus outside pressure and the other with
peer pressure plus outside pressure) and one control group (with only outside pressure). To
clarify sources of error and obtain more accurate results (Snedecor and Cochran, 1980), we
performed an analysis of covariance. Since the tax officials’ work experience was associated
with the major variables of this study (i.e., commitment, audit quality, and turnover
intention) (Bluedorn, 1982; Meyer and Allen, 1984; Meyer et al., 1990; Cotton and Tuttle,
1986; Bonner and Pennington, 1991; Bonner and Walker, 1994; Ni et al., 2009), we
considered work experience a covariance.
The results of the experimental study confirmed the negative effects of social pressures
on the behaviors of tax officials, with the impact of superior pressure being especially
notable. Superior pressure would cause a reduction in audit quality and an increase in
turnover intentions. In addition, the effect of peer pressure on audit quality was less than that
of superior pressure, but its influence on turnover intentions was still significant. As to the
moderating effects of organizational commitment and professional commitment, we found
that neither of them moderated the relationship between social pressure and audit quality.
This finding was consistent with the results of Lord and DeZoort (2001). We also found that
organizational commitment did not moderate the relationship between social pressure and
turnover intentions. In other words, tax officials with high organizational commitment were
not necessarily able to resist social pressures, and thus organizational commitment is
ineffective in reducing their dysfunctional behaviors. However, tax officials with only
outside pressure (i.e., the control group) were the exceptions. In this group, the presence of
high professional commitment resulted in lower turnover intentions than that of low
professional commitment.
The managerial implications and suggestions of this study are as follows. First of all,