THE EFFECT OF SELF-EFFICACY AND OBEDIENCE PRESSURE ON AUDIT JUDGMENT BY USING TASK COMPLEXITY AND MORAL REASONING AS THE MODERATING VARIABLE

Ensuring that the implementation of good governance activities was necessary for functional supervision in the government that can be carried out by internal and external auditors. This study aims to provide a review of the determinant factors in influencing the audit judgment of the government's external auditor. As many as 58 auditors from the State Audit Board of the Republic of Indonesia Representative of Southeast Sulawesi Province were sampled in this study. Data collection using a questionnaire instrument. And data were analyzed using Moderated Regression Analysis (MRA). The results of this study showed that (1) self-efficacy has a significant effect on audit judgment, (2) the obedience pressure has a significant effect on audit judgment, (3) the task complexity moderates the relationship of self-efficacy on audit judgment, positive-signed regression coefficients that indicate that the task complexity strengthens the relationship of self-efficacy on audit judgment, (4) the task complexity moderates the relationship of obedience pressure on audit judgment. Positive-signed regression coefficients that indicate that the task complexity strengthens the relationship of obedience pressure on audit judgment, (5) Moral reasoning does not moderate the relationship of self-efficacy to audit judgment, and (6) Moral reasoning moderates the relationship of obedience pressure on audit judgment. The regression coefficient is negative, indicating that the existence of moral reasoning weakens the relationship of obedience pressure on audit judgment. Originality for this paper shows as: conduct a research about task complexity and moral reasoning of auditor external in governmental sector in producing the audit judgment especially in Southeast Sulawesi Province in Indonesia, and this research retest the research result from Jackman and Oldham (1980), Cohen et al. (2001), Jamilah et al. (2007), Iskandar and Sanusi (2011); Yuliani (2012); Sanusi et al. (2015); and Nursaid (2016). Based on these, researcher use model combination and quantitative approach.