Quality assurance audit research – Term Paper

6.3    Interpreting hypothesis test results

The hypothesized structural model was tested with its results presented in Table 4. As revealed by results, the chi-square statistic for the model is significant (X2 =309.35, df =107, p<0.01). Additionally, other indices for testing the fit of the model lie within the range of the recommended values. CFI and IFI both stand at 0.90. GFI was found to be 0.89. Lastly, RMSEA and standardized RMR values are 0.077 and 0.07 respectively. From these values, it is clear that the model is acceptable.

Furthermore, 11% of client identification variation is explained by the exogenous variables. Also, the exogenous variables explain 16% of the variance in client acquiescence scores. Two more tests were performed to ascertain the importance of client identification. The first model shows that client identification is affected by client importance. Consequently, client identification affects client acquiescence.  A sensitivity test was performed to find out if client acquiescence was directly affected by client importance. It is expected that client importance will affect client acquiescence directly if client importance only concentrated on the economic dependence. From the test, t=30 and this implies insignificance of client importance on client acquiescence. Rather, client acquiescence is affected by client concern through the effects on client identification by the importance of the client. This association holds even when client importance effect on client importance is removed.

Table 4

General Summary of Fit and Explained Structural Model Variances

          Tests                                                                                                                          Result

X²                                                                                                                                                                                     309.35

df                                                                                                                         107

p-value                                                                                                               <0.001

Indices of Fit      

CFI                                                                                                                      0.90

IFI                                                                                                                       0.90

GFI                                                                                                                      0.89

AGFI                                                                                                                   0.83

Analyzing Residuals

Standardized RMR                                                                                          0.07

RMSEA                                                                                                              0.078

Dependent Variables R²

Client Identification                                                                                         0.12

Acquiescence of Client                                                                    0.17

The second test was in comparison of the hypothesized model, presented in Figure 1, with another model in which client acquiescence is directly affected by all exogenous variables. To be more specific, this competing model removes client identification’s mediating effect. The fit of the competing model (X2 =364.54, df =105, CFI =0.87, and RMSEA = 0.094) is inferior when compared to the hypothesized model’s fit. In this model, client acquiescence is related to only three factors in a significant way. One of them is professional identification. Auditor experience and identification of clients are also included which proves the superiority of the hypothesized model. Moreover, standardized parameter estimates (except one) in all the hypothesized relationships have significant values. The structural model’s significant paths are presented in Figure 2.

The first hypothesis of the study examines the various antecedents of client identification. Results shown in Figure and Table 5 show that three factors increase necessary identification as suggested. The Social Identity Theory suggests these factors.

These findings complement earlier results that prove the existence of identification of auditors with clients. Variables in the first three hypotheses all have a significant association with higher auditors; client identification. These results thus support h1a, H1b and H1c.

;

Table 5

Results for Structural Model and Coefficients for the Hypotheses; Model

Hypothesis Explanatory Variable;;;;; Explained ;;;;; Standardized; t-values;;;;;;;;;;;;;;;;;;;;;; Conclusion

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; Variable;;;;; Coefficients

HIa;;;;;;; Years client has been;;;;;;;;;; ;;; Client;;;;;;;;;;;;;;;;;;;;; ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

;;;;;;;;;;; ; Audited by auditor;; ;;;;;;;;;; Identification; ;;;;;;; 0.17;;;;;;;;;;;;;;;;;;; ;;; 2.55;;;;;;;;;;;; Significant at p ; 0.01

H1b;;;; ; Client Importance;;; Client Identification;;;;; 0.16;;;;;;;;;;;; ;;; 2.10;;;;;;;;;;;; Significant at p ; 0.05

H1c;;;; ; Client Image;;;;;;;;;;;; Client Identification;;;;; 0.15;;;;;;;;;;;; ;;; 1.93;;;;;;;;;;;; Significant at p ; 0.05

H2;;;;;; ; Client Identification Client Acquiescence;;;; 0.20;;;;;;;;;;;;; ;;; 3.08;;;;;;;;;;;; Significant at p ; 0.01

H3;;;;;; ; Professional Identification Client Acquiescence;;;; -0.12;; ;;; -1.85;;;;;;;;;;;;;;;;;;;;;; Significant at p ; 0.05

H4;;;;;; ;; Years client has been;;;;;;;; Client Acquiescence;;; -0.09;; ;;; -1.54;;;;;;;;;;;;;;;;;;;;;; Marginally significant at;;;;; been audited by firm ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; p=12

;H5a;;; ;;; Client Size;;;;;;;;;;;;; Client Acquiescence;;;; -0.05 ;;; -0.94;;;;;;;;;;;;;;;;;;;;;; Insignificant

;

H5b;;;; ;;; Auditor;s experience ;;;;;;;; Client Acquiescence;;;; -0.31 ;;; -5.21;;;;;;;;;;;;;;;;;;;;;; Significant at p ; 0.001

As it is now evident that auditor-client identification exists, ascertaining whether it compromises auditor objectivity would be important; H2 hypothesizes that client identification impairs auditor objectivity. This means that the extent of client identification is directly related to the acquiescence of the auditor. This hypothesis is supported by the results in Figure 2. Likelihood of auditors to end the conflict according to the wishes of the client are significantly and positively influenced by client identification (H2: t = 3.08 at p; 0.01).

Auditor work

Two factors that can improve auditor objectivity are examined here. First, how professionals identify with their work. According to H3, this increases their objectivity because those who possess greater levels have a lower likelihood of acquiescing to the treatment preferred by the client and, therefore, have a higher likelihood of ensuring that the unrecorded liabilities are included. The second figure reveals that the likelihood of auditors to resolve the conflict according to the wishes of the client is negatively influenced by professional identification (H3: t = -1.85 at p; 0.05).

Secondly,; this study suggests that tenure of the firm can affect objectivity. Firm Tenure refers to the years that a client is involved in auditing engagements with the particular the firm. Statistically, H4 is significant but only to a limited extent, as shown in Figure 2. These findings agree with other archival research works but are a contradiction to the efforts towards audit firm rotation. Overall, this study finds evidence of the association of firm tenure and greater auditor objectivity.

The last hypothesis examines two contextual factors that could affect auditor objectivity. These factors are client size and auditor’s firm tenure. A client’s size and acquiescence related to it are not significantly associated as it is evident in Table 5. When a client becomes of importance to some extent, client size has a minimal differential effect.

It can be concluded that auditors who have been with their firms for quite a long time, have a lower likelihood of acquiescing to client wishes and will most probably require that the missing liabilities be recorded. An earlier analysis of auditors’ rank had similar findings.

7. Conclusion and Discussions

In general it can be noted that the quality of audits is influenced a lot by client identification. Auditors and Auditing firms should hence aim at reducing the same while enhancing client commitment. A good way of reducing client identification is for auditors to approach audits from the perspectives of shareholders and other key players. Client identification essentially encourages bias. Previous studies suggest that strong, independent relations between auditors and clients characterized by fairness and respect go a long way towards improving the quality of the auditors and client satisfaction. My research takes a perspective of social incentives in the examination of the threat faced by auditor independence. The focus of that being that there is a better alternative in the form of Auditor-Client Commitment where the two entities work independently of each other. When there is independence between the Client’s party and the Auditor’s party there is a high likelihood for a good audit- because professional discourse takes precedence over personal relations. In the course of this study, more light is shed on previous research that mainly focused on the financial incentives of auditors. The Social Identity Theory has been very helpful in the establishment of a comprehensive model of the auditor-client state of being identified. Also, it was widely applied in coming up with hypothesized antecedents and in the examination of the consequences of client identification. From the results, it can be said that social incentives influence audit judgments.