An empirical investigation of the effects of moderating and mediating variables in business research: Insights from an auditing report

Namazi, Mohammad; Namazi, Navid Reza Article An empirical investigation of the effects of moderating and mediating variables in business research: Insights from an auditing report Contemporary Economics Provided in Cooperation with: VIZJA University, Warsaw Suggested Citation: Namazi, Mohammad; Namazi, Navid Reza (2017) : An empirical investigation of the effects of moderating and mediating variables in business research: Insights from an auditing report, Contemporary Economics, ISSN 2300-8814, University of Finance and Management in Warsaw, Faculty of Management and Finance, Warsaw, Vol. 11, Iss. 4, pp. 459-470, https://doi.org/10.5709/ce.1897-9254.256 This Version is available at: https://hdl.handle.net/10419/195504 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ www.ce.vizja.pl 459 This work is licensed under a Creative Commons Attribution 4.0 International License. This study examines the empirical impacts of moderating (MO) and mediating (ME) variables in business research, within the context of auditing reports. size, complexity of the operation, and risk of the firm were selected as the independent variable, moderating variable, and mediating variable, respectively. The selection was attempted over 15 years of research (1455 year-firm data) for the firms listed on the Tehran Stock Exchange (TSE). The following techniques were employed for testing the hypothesis: Pearson correlation, Levin, Lin, and Chu test, Hausman unit root, and multiple regression. The results revealed that there is a significant relationship between the size of the firm and the type of the auditing report. Moreover, a significant relationship between the size of the company and the type of auditing report is shown in the inventory value of the moderating variables. Furthermore, the same has been displayed in return on the assets and institutional ownership of the mediating variables. However, inventory and institutional ownership demonstrate asignificant statistical relationship when moderating and mediating variables are considered simultaneously. The implication of this study is to demonstrate the moderating and mediating variables as significant impediments in the type of auditing report. This would change the design, theory, and implications of this research. 1. Introduction An auditing report (AR) is the most potent information that is assimilated by the auditor. It reveals the final results of investigation of the auditor and the auditing processes concerning various activities of the firms. Such reports are important as they are considered to be multi-faceted mechanisms for reducing conflicts of interests among various stakeholdersinclud- ing managers, stockholders, creditors, government, and investors. Hence, AR particularly posits a very significant role for the firms and auditors (Paracini, Malsch, & Paillé, 2014; Diaz, 2016). The auditing literature reveals that the complexity of the operations (CO) and risks (R) of the firms An empirical investigation of the effects of moderating and mediating variables in business research: Insights from anauditing report ABSTRACT M41, M49 KEY WORDS: JEL Classification: Business research, Auditing reports, Size of the firm, Complexity of operations, Firm risk 1 Shiraz University, College of Economics, Management & Social Science, Department of Accounting, Iran Correspondence concerning this article should be addressed to: Mohammad Namazi, Iran, Shiraz, Eram Paradis, College of Economics, Management & Social Science, Department of Accounting, postcode 71946 85111. Tel: +98 71 36460520, Fax: +98 71 36460520. E-mail: [email protected] Mohammad Namazi1, Navid Reza Namazi1 Primary submission: 21.03.2017 | Final acceptance: 26.07.2017 460 Mohammad Namazi, Navid Reza Namazi 10.5709/ce.1897-9254.256DOI: CONTEMPORARY ECONOMICS Vol. 11 Issue 4 459-4702017 are among the most influential factors, which could lead to significant uncertainty for the auditor. This could, in turn, affect the type of issuing AR (Milch & Laumann, 2016, Liu, 2015). The size (S) of the firm is also a potent variable affecting the issuance of AR and R factors (Youn, Hua, & Lee, 2015). The impact of the CO, R, and S of the firm are studied individually in the auditing literature (Cohen, Krishnamoorthy, & Wright, 2007; Bruynseels & Willekens, 2012; Sun & Cui, 2014, Rodríguez et al, 2014, Youn et al., 2015). However, simultaneous effects of the S, CO, and R factors of the firms on the type of the AR have not yet been studied. Furthermore, the simultaneous effects of various elements of the CO and R factors of the firms, as moderating (MO) and mediating (ME) variable, on the type of the AR, has not yet been explored. Hence, one major aim of this study is to respond to the following inquiries: 1) What is the effect of CO, as a moderating variable, on the type of the AR report, when the S is considered? 2) What is the effect of the R of the firm as a mediating variable on the type of the AR, when the S of the firm is considered? 3) What is the simultaneous effect of the S of the firm as moderating and mediating variables, on the type of AR? The structure of the study is as follows: Section 2 provides a theoretical basis, previous studies, and related hypotheses. Sections 3 and 4 present the research methodsincluding significant hypotheses, models, statistical techniques employed, and the findings, respectively. Consequently, the conclusion, discussion, limitations, and suggestions are presented in Sections 5 and 6. 2. Theoretical background, prior studies, and hypotheses An AR is the end result of the auditing process. Based on the agency theory (Jensen & Meckling, 1976; Namazi, 1985; 2013) it is used as a mechanism for reducing agency conflict between the managers and other stakeholders, (specifically, shareholders). The type of AR peculiarly affects the interests of the managers, stockholders, and other stakeholders by decreasing the informational asymmetry between the agent and principles. Moreover, it accurately discloses the degree of reality of the firm with the help of the judgement of the independent auditor (Taffler, Lu, & Kausar, 2004). However, the type of auditing report, which is issued by the independent auditor, depends on various variables. The most notable variables are as follows: Size of the Firm (S)- When the S of the firm is small, the probability of issuing a qualified or adverse/disclaimer report by the independent auditor increases. This is because small firms usually do not maintain required legal documents appropriately. They do not maintain suitable accounting systems, and lack wellinformed accounting personnel. Therefore, these small firms do not maintain appropriate accounting books, records, and financial statements consistent with the prescribed accounting, tax, and commercial standards and principles. In addition, the amount of their materiality level is low, due to their small sizes. These factors would increase the risks of the firm and auditing to the extent that the auditor may issue the qualified or adverse/disclaimer report in accordance to the Auditing Standards No. 42 and 51. Earnhart and Leonard (2013) studied the role of the organizational structure and S in the auditing process. They concluded that corporate governance and S posit a significant relationship with the issuance of AR. Tsipouridou and Spathis (2014) also, by considering the significant factors affecting the type of AR, concluded that a significant relationship is predominant between S and R with the type of AR. Recently, Wang and Dou (2015) found that S is an influential factor affecting the type of AR by investigating the relationship between S and AR. Corbella, Florio, Gotti and Mastrolia (2015), investigated the factors affecting the quality of AR, and concluded that the S of the client’s company is an influential factor affecting the type of AR. Heliodoro, Carreira and Lopes (2016) also arrived at a conclusion that there is a significant relationship between S and the qualified audit opinions. The previous studies unambiguously indicated that a significant effect exists between the S of the firm and the type of AR. Complexity of the Operations (CO): Another potent factor which affects the AR type is the CO of the firm (Blandón & Bosch, 2013). CO could reduce the www.ce.vizja.pl 461 An empirical investigation of the effects of moderating and mediating variables in business research: Insights from an auditing report This work is licensed under a Creative Commons Attribution 4.0 International License. efficiency and effectiveness of the firm’s systems and would provide a basis for the emergence of the risk for auditors. System effectiveness and CO are considered as apart of the firm’s risks, which ultimately affect AR type (Kardes, Cavusgil, & Cavusgil, 2013). To study the effect of CO, Hylas and Ashton (1982) tested 281 errors, which were required to be adjusted in the financial statements, and these were reported in 158 AR. The result revealed that the most infringed and distorted errors were related to accounts receivables and inventory. Xie, Cai and Ye (2010) concluded that CO is a significant factor affecting the type of AR. Ho and Kang (2013) identified the following as the components of CO: accounts receivables, inventory, return on assets, number of departments of foreign sales, merger, consolidation activities, and the ratio of the market value to book value. Their findings showed that the designated CO variables are influential in the operations of the firms. Moreover, Davies and Mackenzie (2014) inferred that the levels and volume of the activities exhibit a significant relationship with CO. Recently, Milch and Laumann (2016) deduced that CO would affect the risk associated with the firms. Previous studies indicated that CO exerts a significant effect on the relationship between S and AR. Hence, the first hypothesis is provided as follows: Hypothesis 1: Complexity of the operations of the firm, as a moderating variable, will affect the relationship between the size of the firm and the type of auditing report. Risk (R): According to the Auditing Standard No. 42, the duty of the auditor is to assess the probability and significant effects of the risk (R). The R which cannot be ordinarily identified by substantive tests must also be identified by the auditor. The importance of the evaluation of R by the auditor has also been emphasized in other auditing standards (for instance, No. 107). The higher the degree of R, the higher will be the amount of uncertainty and going concern of the firm. Therefore, R factor can create asituation, which will result in the issuance of the nonunqualified report by the auditor (Taffler et al., 2004). The effect of the internal control and inherent risks was evaluated by Messier and Austen (2000). They showed that the CO and R factors maintain asignificant relationship with the evaluation and issuance of the type of AR. Geiger, Raghunandan and Rama (2006) also derived at the conclusion that the uncertainty factor in “going concern”, which is a part of R factor, posits the most significant influence on the judgment formation of the auditor in issuing AR. The role of corporate governance on the evaluation of the auditing R and auditor’s planning was also studied by Cohen et al. (2007). Their findings showed that the board of directors would have a significant relation with identifying the risk of the firm. Bruynseels and Willekens (2012) studied the type of AR in more risky American firms. They discovered that the R of the firm and going concern are extremely important in issuing the type of AR. Tsipouridou and Spathis (2014) also concluded that among the financial ratios of the clients such as profitability and S, activity uncertainty, and R, exhibit a significant relationship with the type of AR. Furthermore, Sun and Cui (2014) considered corporate governance variables and found that shares of the non-executive board members, duality of director’s task, and institutional shareholders are regarded as the most important factors of the firm’s R. Recently, Habib and Jiang (2015) investigated the R factors of China’s firms including corporate governance, independence of the board of directors, and financial reporting. The result of their study indicated that the firm’s R factors are influential impediments for financial reporting. Chen, He, Ma and Derrald (2016), also by considering the paradigms of the going concern in the process of issuing AR, arrived at a conclusion that R variables are the major factors affecting the type of AR. The previous research indicated that the following are considered as significant R variables of the firms: quick ratio, return on assets, ratio of long-term liabilities and earnings before extraordinary items, share of non-executive board members, duality of director’s task, and institutional shareholders. These R variables are considered to affect the issuance of the type of AR. Furthermore the mediating role in the relationship between S and AR is posited by the R of the firm. Thus, a firm which would experience a high R would probably not receive an unqualified AR. This is because the firm’s R will eventually limit the scope of the investigation and deviate it from the accounting and auditing 462 Mohammad Namazi, Navid Reza Namazi 10.5709/ce.1897-9254.256DOI: CONTEMPORARY ECONOMICS Vol. 11 Issue 4 459-4702017 standards. These factors, in effect, influence the type of AR issued by the auditor. Hence, the following hypotheses are provided: Hypothesis 2: The risk of the firm, as a mediating factor, affects the relationship between the size of the firm and the type of auditing report. Hypothesis 3: The complexity of the operations and the risk of the firm posit a mediated-moderated effect on the relationship between the size of firm and the type of auditing report. 3. Research methodology This study is a quantitative applied research that utilizes the quasi-design approach (Smith, 2014). The population of this study encompasses all listed firms in the Tehran Stock Exchange (TSE) market, Iran’s largest stock exchange. It is a member of the World Federation of Exchanges and a founding member of the Federation of Euro-Asian Stock Exchanges. The period of this study is 15 years (2001-2015). However, firms were selected on the basis of a systematic approach according to the following criteria: 1) The financial period of the firms is at the end of the year. 2) Financial institutions, banks, and holding companies are not among the firms selected. 3) The required financial information of the companies is the information available during the period of the study. 4) Firms were active during the period of the study. Given the preceding conditions, 97 firms (1,455 yearfirm) were identified for this study. The required data was obtained through the Sahra and Tadbir Pardas’ software. Furthermore, the information relating to institutional ownership and market value was collected from the firm’s financial statements and board of directors’ reports. SPSS was employed for testing the hypotheses. Eviews, panel data, and multiple regression analysis were also employed for hypothesis testing. 3.1. Research variables and design In this study, the type of (AR) was chosen as the dependent variable. This variable was also expended by Earnhart, and Leonard (2013) and Rodríguez, Eldridge, Roldán, Leal-Millán and Ortega-Gutiérrez (2014). Furthermore, the (R) of the firm was extracted as the dependent variable in order to reflect the nature of the mediating variable in Hypotheses 2 and 3. The (S) of the firm was selected as an independent variable. This variable has also been used in Earnhart, and Leonard (2013) and Rodríguez et al. (2014), among others. The (CO) was used as the MO variable based on the present literature in order to: 1) extend the theory and design the relationship between the S of the company and the type of AR to a more realistic and complex situation, 2) determine “when” and under “what condition” the direct relationship between S and AR variables would hold significant, and 3) be consistent with the definition of the moderating variable (See, Baron & Kenny, 1986; Wilken, Jacob, Prime, 2013; Namazi & Namazi, 2016). Generally, as the CO is increased, the amount of uncertainty and disagreement of the auditor would also increase. Therefore, the type of AR could be affected. In this study, Figure 1 illustrates the schematic relation of the MO variable. In Figure 1, three paths (A, B and C) are shown to consequently determine the effect of S, MO, and their interactions on the AR type. The significant effect of the MO variable is particularly analyzed by concentrating on path C –the interactive effect of MO and S variables. Following Hylas and Ashton (1982), Ho and Kang (2013), Davies and Mackenzie (2014), accounts receivables, inventory, return on assets and the ratio of market value to book value were expended in order to study the effect of the MO variable. These variables were selected because they are the measures of complexity of the firm and are, in fact, effective in issuance of the AR. Hypothesis 2 was provided in order to extend the design and theory of the relationship between S and AR. Based on the literature, the R of the firm was considered as the ME variable as it was hypothesized that the relationship between S and AR can be studied by a chain of causal relations. This relation considers R as an ME factor between S and AR. Moreover, the firm would actually maintain two roles: 1) The dependent variable with respect to S, and 2) the independent variable in regard to AR variable. In essence, the introduction of ME could explain “how” and “why” a correlation exists between S and AR (See, Baron & Kenny, 1986, Namazi & Namazi, 2016). The R of the firm has also been used by Messier and Austen www.ce.vizja.pl 463 An empirical investigation of the effects of moderating and mediating variables in business research: Insights from an auditing report This work is licensed under a Creative Commons Attribution 4.0 International License. Figure 1. The effect of the moderating variable (complexity of the operations) Step 1 A B Step 2 C Interaction effect Figure 1: The effect of the moderating variable (complexity of the operations)  Independent variable (X) (Size of the firm) MO variables (Complexity variables) (Y) Dependent variable ( Auditin g Re p orts ) (X*MO) Figure 3. The simultaneous effects of the MO and ME (The mediation-moderation) effect A A B C D E Figure 3: The simultaneous effects of the MO and ME (The mediation-moderation) effect  Independent variable (X) (Size of the firm) MO variables ( Com p lexit y variables ) (X*MO) (Y) Dependent variable (Auditing Reports) ME variables (Risk) Figure 2. Path diagram of the mediating variable (risk) D Dʹ Figure 2: Path diagram of the mediating variable (risk)  Independent variable (X) (Size of the fir m ) (Y) Dependent variable (Auditing Reports) Independent variable (X) (Y) Dependent variable ME variables ( Risk ) 464 Mohammad Namazi, Navid Reza Namazi 10.5709/ce.1897-9254.256DOI: CONTEMPORARY ECONOMICS Vol. 11 Issue 4 459-4702017 (2000), Cohen et al. (2007), Bruynseels and Willekens (2012) and Sun and Cui (2014) in their studies. Figure 3 illustrates the simultaneous effects of the MO and ME (The mediation-moderation effect). It shows that the relationship between S and AR is not direct; rather it is based on the combination of the MO and ME variables. In this situation, the design and theory of the relationship between S and AR are extended to situations, in which the ME variable is affected by the determined interaction effect of the MO variable on the independent variable (S*MO). The ME variable, in turn, affects the dependent variable. 3.2. Models The following equations were considered for each hypothesis, based on the Baron and Kenny (1986), Muller, Judd and Yzerbyt (2005) and Ro (2012). First Hypothesis: Y= α10 + α11 X + α12 MO + α13 (X*MO) + ε1 (1) Hence, OPIN = α10 + α11 LNTA + {α12(1) REC + α12(2) INV+ + α12(3) ATURN+ α12(4) MB} + {(α13(1) LNTA*REC + + α13(2) LNTA*INV + α13(3) LNTA*ATURN + + α13(4) LNTA*MB) + ε1 For the second hypothesis, the following three multiple regressions were formed to reflect the nature of the mediating variables: Second Hypothesis: Y= α20 + α21 X + ε2 (2) ME= α30 + α31 X + ε3 (3) Y= α40 + α41 X + α42 ME+ ε4 (4) Hence, OPIN= α20 + α21 LNTA + ε2 (QUICK + ROA + LEV + PROFIT/LOSS + BI + + CEOCHR + INST) = α30 + α31 LNTA + ε3 OPIN = α40 + α41 LNTA + α42 (QUICK + ROA + LEV + + PROFIT/LOSS + BI + CEOCHR + INST) + ε4 Third Hypothesis: Y= α50 + α51 X + α52 MO + α53 (X*MO) + ε5 (5) ME= α60 + α61 X + α62 MO + α63 (X*MO) + ε6 (6) Y= α70 + α71X + α72 MO + α73 (X*MO) + α74 ME + + α75 (ME*MO) + ε7 (7) Equation 5 is exactly the same as Equation 1. The right side expression of Equation 6 is exactly the same as that of the Equation 5. Hence, OPIN = α50 + α51 LNTA + {α52(1) REC + α52(2) INV + + α52(3) ATURN+ α52(4) MB} + {(α53(1) LNTA*REC + + α53(2) LNTA*INV + α53(3) LNTA*ATURN + + α53(4) LNTA*MB) + ε1 (QUICK + ROA + LEV + PROFIT/LOSS + BI + + CEOCHR + INST) = α60 + α61 LNTA + + {α62(1) REC + α62(2) INV + α62(3) ATURN + α62(4) MB} + + {(α63(1) LNTA*REC + α63(2) LNTA*INV + + α63(3) LNTA*ATURN + α63(4) LNTA*MB) + ε1 OPIN= α70 + α71 LNTA + α72 INV + α73 ATURN + + α74 (LNTA*INV) + α75 (LNTA*ATURN) + α76 ROA + + α77 INST + α78 (ROA*INV) + α79 (ROA*ATURN) + + α80 (INST*INV) + α81 (INST*ATURN) Variables: OPIN: This stands for the audit report. This variable was considered as a dichotomous variable. Thus, if the firm received an unqualified report, the number 1, was assigned. In other cases,0 was assigned. LNTA: This stands for the natural logarithm of the total assets. REC: This refers to the percentages of the accounts receivables to total assets. INV: This signifies the percentages of the inventory to total assets. www.ce.vizja.pl 465 An empirical investigation of the effects of moderating and mediating variables in business research: Insights from an auditing report This work is licensed under a Creative Commons Attribution 4.0 International License. ATURN: This stands for asset turnover determined as total sales divided by total assets MB: This denotes the ratio of the market value to the book value of the firm. QUICK: Quick ratio = current assets - inventories and prepaid expenses ÷ current liabilities. ROA: Return on assets = income before extraordi- nary item ÷ total assets. LEV: It signifies the ratio of long-term liability = longterm liability ÷ total assets. PROFIT/LOSS: This is the income before the extra-or- dinary item. This variable was considered as dichotomous. If this item was less than 0, the number 1 was assigned; otherwise, 0 was assigned. BI: It denotes for the percentage of board independence. CEOCHR: This signifies the duality of the director’s duty. If the CEO is also the chairman, the number 1 was considered; otherwise, 0 was considered. INST: This stands for the number of shares held by institutional investor’s ÷ total shares of the firm. 4. Hypothesis testing and findings Table 1 shows the descriptive statistics related to the dependent, independent, moderating and mediating variables. This table indicates that 84% of the TSE companies received non-unqualified reports during the period of the study. The average logarithm of the total assets of the TSE companies is equal to 3.21 million Rials. Furthermore, among the complexity variables, the market to book value ratio exhibits the highest average, median, and standard deviation. This could be associated with the inflation effect related to the sales of the companies and/ or the historical cost computation of the total assets, which is calculated in this ratio. 4.1. Assumption analysis In this study, Jarque-Bera test was first employed in order to examine the normality of the data. However, Levin, Lin and Chu test was conducted in order to perform the reliability and stationary position of the research variables. The result revealed that the data for each test was normal (p-value = 0.00) and reliable (p-value = 0.00) at a significance level of 5%. As per the result, Hausman test was extricated to investigate the collinearity of the equations of the models. The results revealed that the model could be estimated based on the fixed approach. (The statistical printout will not be presented here because of space limitation). After appropriate estimation, Pearson correlation method was carried out in order to Statistical indicator Variable Central indexes Index of Distribution Statistical indicator Variable Central indexes Index of Distribution Median Average Standard deviation Median Average Standard deviation OPIN 0.841242 0.789103 0.428844 LNTA 3.210170 2.162516 5.370082 ATURN 0.671690 1.000000 0.469851 ROA 0.038751 0.000000 0.193106 REC 0.039828 0.000000 0.195660 INV 0.841242 0.789103 0.428844 QUICK 0.655749 0.600000 0.176924 LEV 63.04841 73.30000 31.38912 MB 5.773185 5.683248 0.690706 LOSS 0.240614 0.237581 0.173225 BI 0.771804 0.718241 0.438342 CEOCHR 14.11531 12.57199 13.77684 INST 0.657850 0.653242 0.263460 Table 1. Descriptive statistics of variables 466 Mohammad Namazi, Navid Reza Namazi 10.5709/ce.1897-9254.256DOI: CONTEMPORARY ECONOMICS Vol. 11 Issue 4 459-4702017 examine the association relationships. In all of the analysis, the significance level was set at 5%. Table 2 shows the result of the Pearson correlation. Consequently, for testing the hypothesis of the study, only the significant correlation coefficients were entered into the model. 4.2. Results of the hypotheses Hierarchical multiple regression analysis was performed for the first hypothesis. Table 3 shows that the relationship between AR and S (P=0.01), and MO variables for INV (P= 0.009), and ATURN (P=0.000) are significant, and the amount of R2 is 0.018. Table 4 shows that only INV (P=0.000) and LNTA*INV (P=0.002) are significant among the MO variables. This will unambiguously show the significance of the INV among the MO variables. Hence, H1 is confirmed and the R2 of the model changes to 0.026. In pursuing the second hypothesis, the statistical relationship between significant designated variables was tested through multiple regressions analysis based on the Equations 2 and 4, and MANOVA for Equations 3. The results of Equation 2, revealed that the relationship between S (P = 0.005) and AR is significant. The result for MANOVA, using Wilk’s Lambda, was also significant (P = 0.000). The regression analysis of the model 3, showed that both ROA (P = 0.000) and INST (P = 0.000) as ME variables, are significant. Moreover, the R2 of the model is increased to 0.056 (see table 5). Since the absolute coefficient of LNTA in model 4 (α41 = 0.012) is smaller than its coefficient in model 2 (α21 = 0.050), the mediation relationship exists. Hence, H2 was supported. For the third hypothesis, Equation 5 is used to assess moderation of the overall treatment effect of Equation 1. Equation 6 allows the treatment effect on the mediator in Equation 3 to be moderated. Equation 7 is a moderated version of Equation 4, in which both mediator and the residual effect are controlled and allowed to be moderated. In the third hypothesis, the simultaneous effect of the S and CO (INV and ATURN) and ME (ROA and INST) was considered. The result (of Equation 7) is shown in Table 6. Table 6 shows that the relationship between LNTA and AR is significant (P =.003). INV Correlations LNTA ATURN ROA REC INV QUICK LEV MB LOSS BI CEOCHR INST OPIN Pearson Correlation -.085* -.071* -.141* -.030 .072* -.037 .019 -.025 .047 .045 .023 -.215* Sig. (2-tailed) .001 .007 .000 .252 .006 .166 .460 .376 .076 .124 .409 .000 Table 2. The result of Pearson correlation * Significant Kind of variable Variables Coefficient Standard deviation T statistic Sig (R2)(Adjusted R2) Independent variable LNTA -0.050 0.019 -2.552 0.010* 0.018 0.016MO variables INV 0.265 0.102 2.595 0.009* ATURN -0.108 0.028 -3.757 0.000* C1.004 0.128 7.806 0.000* Table 3. First regression results of model 1-Hypothesis 1 * Significant