Site Loader

The Impact of Corporate Governance on Earnings Quality of Public Shareholding Companies listed in Amman Stock Exchange: Empirical Study
??? ????? ??????? ??? ???? ??????? ?? ??????? ???????? ?????? ???????? ??????? ?? ??? ???? ??????: ????? ????????
Prepared By
Shaden Sami Abu Nadi
Student ID
401610036

Supervised By
Professor Mohammad Matar

This thesis is submitted as a requirement for Master’s degree in Accounting

Accounting and Finance Department
Faculty of Business
2018

Authorization
I, Shaden Sami Abu Nadi, herby grant Middle East University the right to use hard and soft copies of this thesis and/or distribute it worldwide, in whole or in part, and/or my abstract, in whole or in part, to libraries, organizations, scientific research institutions, and other entities requesting it.

Name: Shaden Sami Abu Nadi
Dated:
Signature:

Examination Committee’s Decision
This thesis of Shaden Sami Abu Nadi, which studied the Impact of Corporate Governance on Earnings Quality of Public Shareholding Companies listed in Amman Stock Exchange, has been defined accepted on ………../2018.

No. Name Title Signature
1 Prof. Mohammad Matar Supervisor and head of the committee.
2 Internal member.
3 External member.

Acknowledgment
To begin with, I would like to thank Allah for the strength and patience he had given me to finish this work. This study could not have been achieved without having faith that Allah is there to support and help me. May he bless everyone who was there for me during my studying period.
My words cannot describe how grateful I am to Professor Mohammad Matar for his endless support. He is a name that shines when accounting is mentioned, I cannot express how lucky I am having him to supervise my thesis. Without his continuous guidance, and vast knowledge, I would not be writing this. His recommendations, guidance, advocacy, expertise, patience and endless support has led me to finish this work. I am sure I would not have made it without him.
I must thank my family and friends for encouraging me in all of my pursuits and inspiring me to follow my dreams. I am especially grateful to my parents, who supported me emotionally and financially. Thank you to my mother, who had many sleepless nights supporting and encouraging me that I can make it. This accomplishment would not have been possible without them. Thank you!
Finally, many thanks go to the examination committee for devoting their time to review and discuss the material of this study.

Shaden Sami Abu Nadi

Dedication
This thesis is dedicated to my family, my precious father Sami, my main reason of support, my super mother Duha, and my lovely brothers Hamzeh and Mohammad. I can remember myself on the verge of breaking down under stress and pressure, and having them there to give me an emotional push that I can and will make it. Their faith and belief in me is what gave me the power and courage to make it to this point.
Finally, to my professors, friends and colleagues, thank you for being there for me along the way.
I really appreciate what every and each one of you have done to me.
Shaden Sami Abu Nadi

Table of Contents
Subject Page
Authorization B
Examination Committee’s Decision C
Acknowledgement D
Dedication E
First Chapter: The Study Background and Importance 1-7
Introduction 1-2
The Study Problem 2-3
The Study Objective 3
The Study Importance 4
Main Questions the Study Aims to Answer 5
Hypotheses of The Study 5-6
Frame of the Study 6
Limitations of the Study 6
Definitions of Main Terms Used 6-7
Second Chapter: Theoretical Framework and Previous Studies 8-36
Theoretical Framework 8-20
The Study Model 21-23
Previous Studies 24-34
What Differs the Current Study from Previous Studies? 35-36
Third Chapter: Methodology 37-40
Introduction 37
Methodology of the Study 37
Population of the Study 37
Sample of the Study 38
Sources of Information 38
Variables of the Study 39
Statistical Techniques Used 40
Fourth Chapter: Statistical Analysis and Hypothesis Testing 41-54
Fifth Chapter: Results and Recommendations 55-61
References 62-64
Appendices 65-89

List of Tables
Table No. Table Page
1-3 Table (1): Sample of the Study 38
2-4 Table (2): Results of Descriptive Analysis of the Independent Study Variables 42
3-4 Table (3): Results of Descriptive Analysis of Earnings Quality Variable 44
4-4 Table (4) The validity of the study data for statistical analysis 47
5-4 Table (5): Results of Correlation Matrix between the Study Variables 48
6-4 Table (6) Multiple Regression Test Results of the Study 51
7-5 Table (7): The Relationship between Corporate Governance and Earnings Quality on each Economic Sector. 58

List of Figures
Figure No. Content Page
1-2 Figure (1): TACC effect on Quality of Earnings. 22
2-3 Figure (2): The Study Model 39
3-4 Figure (3): Normal Distribution of Residuals in the Study Model 46

List of Appendices
Appendix No. Content Page
Appendix 1 Corporate Governance Data 66
Appendix 2 Earnings Quality Data 74
Appendix 3 Componenets of the Equation of Earnings Quality Data 79
Appendix 4 Outputs of Statistical Analysis 84

The Impact of Corporate Governance on Earnings Quality of Public Shareholding Companies listed in Amman Stock Exchange: Empirical Study
Prepared By:
Shaden Sami Abu Nadi
Supervised By:
Professor Mohammad Matar

Abstract
This study examined the effect of corporate governance practices on the quality of reported earnings in Amman stock exchange (ASE). It examined the effect of the four most important corporate governance variables of public shareholding companies listed in ASE; number of board of directors members, number of board of directors meetings, number of board of audit committee members, number of audit committee meetings. The study used the total accruals model by Richardson and applied it on a sample of 60 companies of the population which included 131 companies listed in ASE from the three economic sectors in Jordanian ASE for the period 2013-2016.
The researcher used descriptive statistical analysis: through measuring mean, maximum, minimum, and standard deviation, Smirnov and Shapiro Normality Tests, Multicollinearity through variance inflation factor (VIF) and tolerance, Pearson Correlation Matrix and Multiple regression to test the hypothesis of the studies.
The researcher found a significant impact of CG practices when considered as a whole unit on earnings quality in public shareholding companies listed in ASE. It also revealed that there is a significant impact of the number of board of directors’ members on earnings quality. The study also found no significant impact of the number of boards of directors’ meetings, the number of audit committee members and the number of the audit committee’s meetings on earnings quality.
Finally, the researcher recommends mandating the disclosure of corporate governance practices in all annual reports published on ASE, and conducting more studies taking in consideration the independence of the board of directors’ members and the independence and expertise of the audit committee members.
Key Words: corporate governance, earnings quality, Amman stock exchange, board of directors, audit committee.

??? ????? ??????? ??? ???? ??????? ?? ??????? ???????? ?????? ???????? ??????? ?? ??? ???? ??????: ????? ????????
?????:
???? ???? ??? ??????
?????:
??????? ??????? ???? ???

??????
???? ??? ??????? ????? ??????? ????? ??????? ??? ???? ??????? ??????? ???????? ?????? ???????? ??????? ?? ????? ????. ???? ??????? ?????? ??? ????? ??????? ????? ?????? ????? ??????? ?????? ??????? ??? ? ??? ????? ???? ??????? ???? ???????? ???? ??????? ???? ????? ???? ???????? ???? ???????? ???? ???????.
??????? ??????? ????? ????????? ????? ?? ??? ?????????? ?????? ??? ???? ?? 60 ???? ?? ????? ??????? ???? ???? 131 ????? ????? ?? ????? ???? ?? ???????? ?????????? ??????? ?? ??????? ???????? ?????? 2013-2016.
??????? ??????? ??????? ???????? ??????: ?? ???? ???? ??????? ??????? ? ???? ?????? ? ???? ?????? ? ? ???????? ????????. ???? ???????? ???? ???? ???????? ?? ??? ???????.
???? ??????? ?? ???? ??? ????? ?????? ??? ??????? ??????? ??? ?? ???? ??? ????? ????? ??? ???? ???????. ???? ??? ??? ?????????, ?? ??? ???, ???? ??????? ?? ???? ??? ????? ?????? ??? ??? ????? ???? ??????? ????? ???????, ????? ???? ????????? ??????? ??????, ??? ??? ???????? ???? ???????, ??? ????? ???? ???????, ???? ???????? ???? ??????? ??? ??? ?????? ???? ??? ???? ???????.
??????? ? ???? ??????? ?????? ??????? ?? ????? ?? ??????? ????? ??????? ?? ???? ???????? ??????? ???????? ?? ????? ???? ? ?????? ???? ?? ???????? ?????? ??? ????? ????????, ?? ????? ?? ???????? ????????? ????? ???? ??????? ?????????? ????? ????? ???? ???????.
??????? ?????????: ????? ??????? ? ???? ??????? ? ????? ???? ? ???? ??????? ? ???? ???????.
Chapter One
The Study Background and Importance
Introduction
Shareholders trust in the board of directors, started declining after many incidents of corporate misconducts happened worldwide, starting with the Enron scandal, losing 78 billion dollars in the stock market, and followed with multiple other fraud cases on an individual level like Bernard Madoff, and on a multinational corporation level like WorldCom, Tyco, Lehman Brothers, that ended with the shareholders loosing high percentages of their investments. Such cases arose concerns over the management of corporations, thus, it resulted in vast interest in corporate governance practices.
Corporate governance (CG) became the concern of the twentieth century were all multinational organizations aimed towards having a clear organizational structure in order to enhance both financial and managerial performance of the firms. Good (CG) gives investors and stakeholders the assurance that their money and interest are in safe hands. It increases the competitive advantage of the firm and gives it a stronger and more attractive position in the market. It boosts and enriches the concept of transparency in the financial data presented by the entity at the end of each reporting period, which increases the credibility and reliability of the firm. The attention for it in Jordan started growing lately, after many countries in the MENA area started trading on a global level, becoming a part of the World Trading Organizations (WTO’s). Committees’ and academic individuals started rising awareness, writing more papers, and assessing the current laws that could relate to it.
Transparency and credibility are both major factors representing the quality of earnings. Most companies’ main goal is to make money, yet, the firm’s earnings information is a key metric of its financial performance. Faithfully represented financial data implies adequate, precise and reliable earnings reported in the capital market and indicates the true value of the firm. Earnings quality is considered a strong basis for evaluation of a firm by current and potential investors. It helps them make right decisions about the risk associated with their investments. If that data is not accessible by the investors, they tend to have a red flag alert of potential risk, which will eventually affect the value of the firm as a whole. (Sarun, 2016)
The list of fundamental and enhancing qualitative characteristics mentioned in the conceptual framework of accounting, including relevance, faithful representation, comparability, verifiability, timeliness, and understandability, are in my point of view, characteristics that must reflect on any financial data published by the firm, to reveal good quality of earnings.
It has been said that, in order to implement these characteristics and have high quality earnings, good corporate governance practices shall apply, to control the processes of producing and reporting financial and non-financial data.
The Study Problem
Poor CG practices contributed to the creation of the agency problem. It results from the separation between ownership and control. As managers have more inside information than the financial providers, managers might use that information to maximize their wealth, putting on the side the interest of current investors (Chi-keung Man, 2013), which had led to many fraud and manipulation scenarios, not only in Jordan, but also worldwide, thus increased the fear of investment for many potential investors concerned of losing their wealth to bad selfish management. This study examines whether companies applying CG practices differ from others in their financial results. It debates if CG practices achieves the characteristics of transparency and reliability in order to achieve a good quality of earnings.
The main problem of this study is to assess the impact of corporate governance on the quality of earnings in public shareholding companies listed on Amman Stock Exchange (ASE), and to imply whether there is a direct effect of the number of board of directors members, the number of audit committee members, number of board of director’s meetings as well as the number of the audit committee meetings on the quality of reported earnings.
The Study Objective
This study aims to:
Identify the meaning and scope of corporate governance and earnings quality.
Identify principles and practices of corporate governance applied in Jordanian public shareholding companies.
Assessing the advantages of good governance, and the disadvantages of bad corporate governance.
Discuss the determinants of earnings quality.
Assess the factors affecting the quality of reported earnings.
Assessing the impact of corporate governance on the earnings quality of Jordanian public shareholding companies.
Develop a model to show the impact of corporate governance on the earnings quality.

The Study Importance
Aiming to correct the misconduct of applying corporate governance in Jordanian public shareholding companies, this study is hoped to enhance the current papers written in the area, and be a point of reference to academics, firms, and government officials.
The researcher hopes that the results will help in reinforcing efforts to increase the quality of earnings for companies listed in the Jordanian market, and helping to serve all related parties stockholders and other stakeholders, such as the ministry of labor, security deposits center, income and sales tax department and the central bank, who are interested in the credibility of the financial data published by Jordanian public shareholding companies in their annual reports, as well as help all these parties take the right economic decisions based on the quality of income rather than the quantity of profits generated.
The study will examine results in the period 2013-2016, evaluating a sample of companies from all sectors.
Being in the era of financial indignities, and having earnings management through changing accounting policies, estimates, bad debt write-offs and much more to be mentioned, leading to a sharp decrease in the quality of earnings, a spotlight on the management of the corporations is drawn to attention, and this study intends to make a clear image of it, in a sample of companies selected from Amman Stock Exchange, to check if any impact drawn from managerial corporate governance practices directly or indirectly affect earnings quality.

Main Questions the Study Aims to Answer
The study mainly aims to answer this main question:
Is there an impact of corporate governance on earnings quality of public shareholding companies listed in Amman Stock Exchange?
That question is divided into many subsequent ones:
Is there an impact of the number of board of directors members on earnings quality?
Is there an impact of the number of meetings of board of directors on earnings quality?
Is there an impact of the number of audit committee members on earnings quality?
Is there an impact of the number of meetings of the audit committee on earnings quality?
Hypotheses of The Study
The main hypothesis is:
H0: There’s no significant impact of corporate governance on the quality of earnings in public shareholding companies listed in ASE.
As an implication of that hypothesis, the following sub-hypotheses arise:
H01: There is no significant impact of the number of the board of directors members on earnings quality.
H02: There is no significant impact of the number of meetings held by the board of directors on earnings quality.
H03: There is no significant impact of the number of the audit committee members on earnings quality.
H04: There is no significant impact of the number of meetings held by the audit committee on earnings quality.
Frame of the Study
Time frame: it will cover the financial periods from 2013-2016.
Location: A sample of companies from all sectors; financial, services, and industrial, listed in ASE.
Limitations of the Study
One of the main limitations of the study, is the degree of the data disclosed by these companies on a public level. As well as their compliance with rules and regulations of reporting governance related information.
Definitions of Main Terms Used
Corporate governance: is the set of guidelines and principles applied to direct and control a corporation.
Earnings quality: is a scale showing the level to which reported earnings are actual compared with the accounting policies used, actual operating income and management orientations.
Earnings Management: involves the intentional manipulation of the financial reporting processing, by altering the reported accounting numbers to obtain a private gain.
Transparency: offering a true, concise, and credible financial reports.
Credibility: the degree to which financial data is true and trustworthy to all stakeholders.
Earnings: reported profit or loss resulting from regular business operations after tax and interest deductions, also known as net income or bottom line.
Earnings management: the misrepresentation of the actual earnings generated by the company for a specific financial period.
Accruals: are earned revenues and incurred expenses that have an overall impact on income, assets, accounts payables, account receivables and other balance sheet items.

Chapter Two
Theoretical Framework and Previous Studies
This chapter aims to clarify the basic concepts associated with corporate governance and its effect on the quality of earnings presented by firms at the end of each reporting period. It will also go through a few previous studies revolving around the same.
First: Theoretical Framework
Corporate Governance
Introduction
Good corporate governance (CG) helps to secure investor confidence, enhance access to capital markets, promote growth and strengthen economies. This makes CG necessary, beneficial and useful for all sectors and types of companies whether they are multinationals, governmental, domestic firms, small businesses or family-owned operations.
Although CG frameworks differ from country to country based on the legal, regulatory and institutional environment, they all aim to identify the rights, responsibilities, and expected outcomes of owners and managers.
History and Development of Corporate Governance
The interest in corporate governance started immediately after World War II, the U.S. economy bloomed and large corporations started from scratch, especially in the industrial sector. Amongst the widespread prosperity, the phrase “corporate governance” was not in use, and corporate governance was the least important issue to taken into consideration. All they had applied was that managers led, and directors and shareholders followed. In the 1970’s, the discovery of widespread illegal payments by U.S. corporations to foreign officials drew the S.E.C further into the corporate governance realm. The extensive corporate bribery, assured that the managers only enjoyed the position and reflected the failure of the system of corporate accountability. With the government overseeing all that, managers main concern was then the prospect of unpreventable government supervision and control. It was not long until their concerns turned to reality and a majority started asking that corporate governance should be reformed through the enactment of federal laws returning the board to its historical role as internal auditor of the corporation responsible for constraining management from violations of law and breaches of trust. (Cheffins, 2012).
After that, many of laws and regulations calling for the implementation of CG practices arose. In 1999, the OECD published its Principles of Corporate Governance, the first international code of good corporate governance approved by governments. These principles are not prescriptive or binding, they are more advisory whereas each country could apply them the way that suits them the most.
The principles went through multiple developments and ended up calling for a stronger role for shareholders in a number of important areas, including executive remuneration and the hiring process of board members. They ask companies to make sure they have mechanisms to address possible conflicts of interest, to recognize and protect the rights of stakeholder’s, and implement open door policy to hear the voices of those who foresee any misconduct. They also identify the responsibilities of auditors to shareholders and the important role the institutional investors could play in monitoring company performance and in discussing their concerns to the boards. They can challenge or support the board through voting at the general meetings of shareholders and they are well placed to take their concerns directly to the board and to propose a course of action.
In the past few decades, the importance of corporate governance was drawn after the Enron wreckage. Freed power in the hand of the CEO that led to unethical behavior within the Enron organization that continued to come to light long after its downfall. Overall, corporate governance in Enron was weak in almost all aspects. Thus, the board of directors is composed of a number of people who lacks moral character. Also, they are often willing to engage themselves in fraudulent activity. This was the genuine root of the company’s corporate governance failure, and the reason behind the Sarbanes-Oxley act. (Dibra, 2016)
The OECD was a point of encouragement for the Middle Eastern countries in the application and practice of corporate governance practices. United Arab Emirates founded The Institute of Corporate Governance (Hawkamah) as a joint effort between Dubai International Financial Center, OECD and Arab Banks Union. (Matar, 2009, P. 459-476)
In Jordan, corporate governance was an ambiguous concept, until in 2003, where Amman’s Chamber of commerce started an individual department called The Companies Control Department, that observes and assures that corporations are implementing the right applicable CG practices, under the aim of increasing the competitive advantage of the national economy. (Jordanian Corporate Governance Guide, P: 04). That guide included the basic guidelines of corporate governance disclosure all companies are supposed to follow. The code has five parts; it enlists the board of directors’ roles and responsibilities, the control environment, transparency and disclosure, rights of shareholders and stakeholders.
Board of directors should be competent to achieve their aligned responsibilities, and represent the organizations products and services to their target audiences. They should have the expertise in management, finance, business, national and international markets and applicable laws and regulations. They should be independent and their judgements should be free of bias to benefit the organization they represent. The code requires that the size of the board should be an odd number, and have a minimum of three, and maximum of thirteen members, based on the size of the firm they present.
When it comes to the number of meetings the board holds, the principle states that the BOD should meet at least once a quarter to discuss organization-related-issues.
The BOD should establish an audit committee to review the financial statements, ensure accountability, review internal audit department, guarantee the adherence to accounting policies and recommend an internal audit. The committee shall be composed of a minimum three independent members, elected by the BOD, and shall meet at least once quarterly to assess the financial performance, and managerial decisions associated with the organization.
The application of these guidelines was voluntary until a new announcement was publicized making it obligatory starting 2017.
Definitions of Corporate Governance
CG is a very broad concept, it is viewed and interpreted differently by each part of the organization and the community. Stakeholders view it as the guidelines that protects their rights, while management views it as their code of business conduct. This clearly results that corporate governance does not have a shared definition in literature and it is not easy to describe. Sternberg (1998) defines it: “Corporate Governance describes ways of ensuring that corporate actions, assets and agents are directed at achieving the corporate objectives established by the corporation’s shareholders”. .It is also defined as “the system by which business corporations are directed and controlled” (Rankin, 2012, P:188). The OECD Principles of Corporate Governance (2004) provides one of the most used broad definitions since it serves as a reference for all OECD member countries: “Corporate Governance defines a set of relationships between a company’s management, its board, its shareholders and other stakeholders. Corporate Governance also provides the structure through which the objectives (i.e. strategy) of the company are set, and the means of obtaining those objectives and monitoring performance are determined”.
Corporate Governance attracted a good deal of public interest because of its apparent importance for the economic health of corporations and society in general. The way companies are governed, their goals and the objectives they pursue, the rights they are entrusted for, the responsibilities they recognize, and the distribution of the value they create became highly significant, not only for their directors and shareholders, but also for the wider community. (Fiori, 2008)
The Importance of Corporate Governance
Good corporate governance allows the corporation to work smoothly due to the existence of a clear level of accountability and communication amongst the organization, as well as people understanding what their roles and responsibilities are. Accountability, being one of the most important characteristics in accounting practitioners, will clarify all involved parties’ rights and responsibilities, through enhancing professional ethics.
To properly understand and exploit corporate governance, it is required to understand and apply its most important ideologies. The general principles of all forms of corporate governance are generally related to the shareholders, board of directors, and stakeholders. In addition to that, corporate governance places a solid emphasis on the behavior of the corporation and the data it discloses publicly.
The most basic corporate governance principles include: (OECD, 2011)
Keep the Interest of Stakeholders in Mind
Treating Shareholders Equally
Identifying the Roles of the Board of Directors
Ethical Behavior
Transparency
Accounatbility
Disclosure
A good CG has many advantages, starting from a better competitive advantage, which means that investors will prefer investing in the company applying CG, other than investing in others in the same industry. It will also gain a better reputation, attract qualified reputable employees, reduce perceived risk to investors which can reduce the cost of capital as well as facilitate economic growth. (Rankin, 2012)
CG is now moving towards a broader view, taking risk management and sustainability in consideration. The concern of proper corporate accountability is related to the need of an efficient risk management and internal control systems. Uncertainty regarding effect of voluntary corporate governance guidelines and risk management drew the attention towards the alleged relationship (Kwamboka, 2010).
The impact of CG elements on the different dimensions of corporate sustainability reporting is also an unexplored area, especially that sustainability is a point of concern for corporations nowadays. It’s been argued that various corporate activities have a direct or indirect impact on the external environment, and that is why corporations should be held accountable to a wider audience than simply its shareholders. (Mahmoud, et al, 2018).
And after the Global Financial Crisis (GFC) the awareness increased and managements of companies where obliged to start two new committees, one for sustainability, and the other for risk management, and they asked the audit committee to put risk management on the top of its’ priorities. (Rankin, 2012, P.200)
CG is also debated to affect the quality of reported earnings. Earnings quality is defined as “the presence of earnings, which reflect current performance, are useful for predicting future performance and correctly discount intrinsic firm value” (Black, 1998)
Quality of Earnings
Quality of earnings reflects in the usefulness of financial statements to both internal and external users including investors, creditors, managers and all other stakeholders contracting with the corporation. It includes much more than understatement or overstatement of net income. Good earnings quality, which is said to improve capital market efficiency, is represented when fundamental characteristics, reliability and relevance, appear. Reliability means that the data is verifiable, free from error or bias and truly represents the actual numbers. Relevance refers to timeliness, and helpfulness to predict and forecast the future outcomes, good earnings quality means it is repeatable over a series of reporting periods. It also refers to the stability of income statement components, maintaining capital and realization risk of assets. Other factors must apply in order to assure good quality of earnings including persistence, accruals quality, predictability, value relevance, timeliness, smoothness, and conservatism. Bad earnings, or poor earnings lack those fundamental qualities, and are deceiving for decision makers. These can be detected when there is an aggressive use of accounting rules, high inflation, and a known gain for sale of assets.
Investors like to see high-quality earnings, since these results tend to be repeated in future periods and provide more cash flows for investors, as well as reduce their fear of keeping or expanding their investments. Thus, entities that have high-quality earnings are also more likely to have higher share prices, and a better competitive advantage.
Earnings quality, in a way or another, is associated with good corporate governance practices. Managements of firms have the control of the way accounting policies are chosen, reports are presented, data is disclosed, and can assess the degree to which that data is relevant, reliable and faithfully represented to decision makers and investors. In order to observe earnings quality, organizations should provide regular detailed reports regarding sources of earnings, and prediction of any changes in the future trends of these sources. They could also implement a conservative accounting approach, in order to correctly recognize their revenues and expenses. (Katsuo, 2008) To simplify it, managements follow CG guidelines and overview the sources of its’ earnings, how they’re presented and to what extent are they accurate or manipulated.
A very important question in relation to the assessment of earnings quality is to determine the factors influencing the quality of financial reports, these consist of the business model and operating environment of the firm, the management’s financial reporting decisions related to policies and estimates, the quality of the standards, the credibility and independence of the auditors, and governance activities. Other factors affecting the quality of earnings could be the size of the firm, diversity of cash flows and revenues sources, and length of the operating cycle.
Quality of earnings has many determinants, the main important are:
Firm Characteristics: Many studies provided evidence that firm characteristics are associated with quality of earnings. These traits include firm’s choice of accounting principles, properties of its earnings such as persistence and volatility, and accruals. Firm performance could indicate any possible scenarios of earnings management and manipulation. The company’s debt indicates the degree of their financial leverage, and if their management is funding it’s earnings by borrowing and increasing its liability. Such action could reduce the quality of earnings, where there is substantial evidence that debt levels are associated with various measures of earnings quality (Malmquist, 1990).
Financial reporting: is the disclosure of financial information to meet the needs of various stakeholders about the financial performance of a firm. Accounting methods, principles and estimates, financial statement classification and disclosures affect the quality of earnings based on the amount of data shared with the public, and whether the managements have chosen the right practices or not.
Governance: internal controls including characteristics of the Board of Directors (BOD), internal control procedures, and managerial share ownership has an impact on earnings quality. Internal control mechanisms are viewed as monitors of the financial reporting system that constrain a manager’s opportunity or ability to manage earnings, while managerial share ownership and managerial compensation are generally predicted to affect earnings quality because they provide incentives for earnings management. It is debated that stronger internal control mechanisms lead to less manipulation and higher quality of earnings. When it comes to the BOD, some studies have shown that independent boards, and higher audit committee quality are associated with less earnings management (Abbott et al., 2004).
Managerial ownership also has an effect on decisions taken regarding accounting policies, methods and estimates chosen. Finally, internal control mechanisms should be assessed to reveal the impact on the various proxies for earnings quality.
Auditors: researchers hypothesize that auditors are a determinant of earnings quality because of their ability to detect a material misstatement and to adjust for or report it (DeAngelo, 1981). Auditors are able to detect errors based on his effort and effectiveness, and has the freedom whether to report or correct errors based on factors such as litigation risk, reputation costs, and independence. Auditor’s firm size and fees could affect the results, which in result affect the quality of reported earnings as well.
Capital market incentives: it hypothesized that the cost/benefit trade-offs of accounting choices change during periods when a firm raises capital, and thus its earnings quality, may differ when a firm is raising capital. One-time accounting choices can have long-term consequences, including a diminished reputation for reliable reporting, which in turn may negatively affect equity valuation due to decreased reporting credibility, creating an alarming alert to investors and stakeholders.
In order to keep the firms attractiveness, managements making decisions that affect reported earnings must be careful. They need to assess objectives related to their compensation, debt contract provisions, or incentives that could cause fluctuations in share prices.
Quality of earnings could mean the degree to which management’s choices of accounting estimates can affect reported income (Weil, 2009). These estimates could be in vague areas, that do not have a certain way to measure and needs judgement to quantify, which leaves stakeholders with a concern that the opportunity of manipulation is there, which causes users to think earnings numbers have a low quality. Earnings management has a lot in common with earnings quality. Earning management occurs when managers use their personal judgement in financial reporting, altering the actual results that could mislead some shareholders, stakeholders and potential investors about the companies actual economic performance, and affect their decisions based on that data (Healy, Wahlen, 1998). Accounting for business operations requires judgement and estimates, as we cannot measure revenue without estimating the prices customers will pay, and forecasting the expected sales, the bad debt, credit terms, refunds and related costs of selling. In these cases, managements’ has to choose a number, for example the percentage of uncollectable amounts, based on actual results of previous years. Anyhow, other economic and political factors affect them, and thus these percentages are not constant. The wider the range of reasonable estimates, the more management’s choice will influence net income, and that could fall under earnings management. Other management decisions such as timing of transactions, inventory valuation methods, depreciation methods, fall under their sole judgement, in which they could use to manipulate earnings, and thus affect the quality of the reported earnings negatively. Earnings’ management has many techniques and shapes, like “cookie jar reserve” technique, this technique deals with future estimates of future obligations to be paid as a result of transactions happening in the current period. In this case managements’ do not have a reasonable answer, and could use this technique to overestimate these expenses, and when these expenses are accrued, the difference between the actual and estimated are put in the “cookie jar” to be used later and boost the earnings of a not-really-good performance periods. These results are misconceiving and does not represent the actual financial performance, which on the other hand, lower the quality of reported earnings. “Big Bath write-offs” is another practice used to manipulate earnings. This method is used to write-off bad debt, write-down assets or change an operating segment. In these cases, a loss is recorded against earnings, which will lower the net income, the quality of reported earnings and cause a drop in the share prices. A “Big Bet on the Future” that happens at acquisitions could open a door for manipulation, the companies could manipulate R&D (Research and Development) costs by matching them to the earnings occurring from that acquisition, which will increase the earnings for future periods, causing a false increase in reported earnings.
Another method to manage earnings is “Throw out” a problem child, selling a losing subsidiarys’ equity will cause a gain in the reported earnings of that period. Many other methods like introducing new accounting standards, write off of long term operating assets, leaseback, early retirement of debt and use of derivatives are also used to manage the reported earnings (Rehman,2013).
To continue on the previous, we can actually say that earnings management is closely associated with manipulating accruals. An accountant makes adjustments for revenue earned but not recorded in the accounts, and expenses that incurred and not recorded in the accounts. These accruals are added through an adjusting entry and are reported on the financial statements.
Accruals could be both discretionary, which are non-obligatory expenses that are not realized but recorded in the accounts books like bonuses of higher management. Managements have full control over these and arise from normal business transactions. They lead to an increase in earnings and are based on the changes of the firms’ performance. Non-discretionary accruals are mandatory expenses to be accrued in the future, and are recorded in the accounts books, like utilities and rent expenses. Such expenses are usually budgeted for based on past actual expenses accrued. Discretionary accruals are the point of debate as they are solely based on management’s choices and flexibility, which can be easily tampered with, causing a decrease of the quality in reported earnings.
It is said that highly managed earnings have lower quality. Yet, not having earnings management, does not in fact guarantee high quality of earnings. Sometimes earnings management, or “income smoothing: could help firms in keeping it’s good position in the financial markets in expected financial distresses.

The Study Model:
To examine the impact of corporate governance practices on earnings quality, the researcher used total accruals model developed by Richardson. This model measures the total accruals that consists of three main variables: The change in working capital, the change in non-current operating assets and the change in net financial assets.
The model is represented by this mathematical formula:
Total Accruals Model: (Richardson et al., 2003)
TACC = ?WC + ?NCO + ?FIN, Where:
?WC = (?Current Assets – ?Cash & Short Term Investment) – (?Current Liability – ?Short Term Debt)
?NCO = (?Total Assets – ?Current Assets– ?Investment and Advances) – (?Total Liabilities – ?Current Liabilities – ?Long-Term Debt)
?FIN = (?Short Term Investment + ?Long Term Investment) – (?long Term Debt + ?Short Term Debt + ?Preferred Stock)
Where
TACC: absolute vale of total accruals.
?WC: net change in working capital.
?NCO: net change in non-operating assets – net change in non-operating liabilities.
?FIN: net change in financial assets – net change in financial liabilities.
This model was applied on data extracted from Amman Stock Exchange, to test the quality of earnings in companies applying, and not applying corporate governance practices.
It helped the researcher to reach that earnings quality is measured as below:
EQ= ?0 + ?1*NBD + ?2* NMBD + ?3* NAC + ?4* NMAC + ¥, Where:
EQ: Earnings Quality and the measure of total accruals, as defined is the probability of having a sustainable current income in the future (Richardson et al., 2003)
NBD: Number of the board of directors members.
NMBD: Number of meetings held by the board of directors.
NAC: Number of the audit committee members.
NMAC: Number of meetings held by the audit committee.
The relationship between total accruals and quality of earnings is of a negative correlation, the higher the accruals are, the higher earnings management is, which as a result decreases the quality of earnings.

Figure (1): TACC effect on Quality of Earnings.
Total accruals (TACC) present both discretionary and non-discretionary accruals, which includes accounting principles and estimates that are affected by managements judgement, and are easily manipulated based on their own interest. This clearly shows a negative relationship between (TACC) and earnings quality, and thus, to determine if there is earnings’ quality for each company in the sample, the researcher had calculated the arithmetic mean of (TACC) for all companies in the sample to be a benchmark.
That benchmark was taken to state that if (TACC) for a certain company is higher than the mean, then there is earnings quality in that company.

The Reason Behind Choosing Richardson Model:
This model can be used with all sectors, unlike other models that can only be applicable on one sector.
It examines total accruals (discretionary and non-discretionary), while other models only examines only discretionary accruals.
It is considered the newest model as it was developed in 2003.

Second: Previous Studies
Here are some previous studies that the researcher found them most related to this study. They are presented according to its publishing date from the oldest to newest.
(Dechow, Dichev, 2001) titled:
“The Quality of Accruals and Earnings: The Role of Accrual Estimation Errors”
This study suggested a new way of how which accruals could affected earnings quality. It assumed that the quality of earnings decreases because of estimation errors. They have developed a working capital model to assess their assumption, and found that accruals quality are negatively related to total accruals, length of operating cycle, standard deviation of sales, cash flows and earnings. They had also found a positive relationship between accruals quality and earnings persistence.
(Hwang, Lin, 2010) titled:
“Audit Quality, Corporate Governance, and Earnings Management: A Meta-Analysis”
The purpose of this study was to use meta-analysis techniques to assess the outcomes of existing studies on the causes of earnings management. The main focus was on the impact of the effectiveness of corporate governance and audit quality. The study’s findings had shown that the independence of the board of directors and its expertise, the audit committee’s independence, its size, expertise, and the number of meetings have a negative relationship with earnings management. It was also found that the audit committee’s share ownership has a positive relationship with earnings management.

(Islam et al, 2011) titled:
“Is Modified Jones Model Effective in Detecting Earnings Management? Evidence from A Developing Economy”
The study examined the effectiveness of Modified Jones Model in detecting earnings management among all public companies that are listed in the Dhaka Stock Exchange (DSE), for the period covering between 1985 – 2005. The researchers employed the modified Jones model to detect earning management in context of Bangladesh capital market and found out that it was not very effective, and attempted to extend the modified Jones model by adding factors like revenue, depreciation expenses, retirement benefit expenses, asset disposal gains/losses and found it to be very successful.

(Ismail, 2011) titled:
“Earnings Quality, Family Influence and Corporate Governance: Empirical Evidence from Malaysia”
This study had taken place in Malaysia. It aimed to examine the association between family firms and earnings quality in Malaysia, to examine the association between corporate governance and earnings quality in Malaysia and to assess whether the existence of family firms mitigates the relationship between corporate governance and earnings quality. The results of the study shown that firms with significant family ownership and control have much higher quality of earnings compared to public firms. The results support the notion that family firms have special features that substitute for the monitoring role of governance over the firm’s activities. The higher earnings quality of family firms could be a result of the family firm’s selflessness and long-term orientation that reduces the motivations for earnings manipulation. The results could also be due to the fact that family firms have lower agency costs and greater expertise relating to the firm’s operations.

(Hollimans, 2011) titled:
“Earnings Quality: Accounting Conservatism in Public and Private Firms in the Netherlands”
This study aimed to test the difference of accounting conservatism policies used in public and private firms in Netherlands, where financial conservatism is debated to increase the quality of earnings. It also tested the difference of the companies’ characteristics and the degree to which they implement IFRS and how that affects accounting conservatism.
The quality of earnings was measured using Ball and Shivakumar model. The results indicated that Private companies are less conservative than the public companies due tohaving internal communication represented by their financial reports, before and after applying IFRS.

(Abu Ali et al, 2011) titled:
“The Impact of Earnings Quality on the Cost of Equity based on the International Financial Reporting Standards”
This study aimed to identify the characteristics of financial earnings and their quality on the cost of capital, through the cost of equity for the public shareholding services and industrial companies listed in ASE. The study assumed that the quality of earnings leads to getting economic gains through decreasing the information risk which leads to decreasing the cost of capital. A sample composed of (57) industrial companies and (27) services companies was tested through examining the qualitative characteristics of accounting information for the period covering (1993-1007).
The study found that there is an impact of earnings quality on the cost of equity, and that there is a relationship between the qualitative characteristics of accounting information on the cost of equity. It also found that accounting factors are more attributing to decreasing the cost of capital than the financial factors.

(Bistrova et al 2012) titled:
“Quality of Corporate Governance System and Quality of Reported Earnings: Evidence from CEE Companies”
This study had taken place in Croatia, Czech Republic, Estonia, Hungary, Latvia, Lithuania,
Romania, Poland, Slovakia, and Slovenia. The data was extracted from annual reports published on their stock exchanges, and the analysis was for the period of three years (2007-2010). It aimed to look into the quality of the corporate governance and the quality of earnings of the CEE companies and clarify how the relationship between these two determinants affects the sustainable shareholder value. Earnings quality was assessed based on the widely-accepted measures to detect creative accounting practices: level of accruals and comparison of net income level to operating cash flow. The most important findings of the study were that there is less risk to face financial results manipulation if the high quality corporate governance applies, meaning that its board and management team act according to the best practice, information on the company is transparent and publicly available, and that well-managed companies are the best in improving quality of the reported earnings during the study’s timeframe. The analysis of the overall earnings management in CEE region implied that the market is favorable to investors whereas manipulation is low, based on the accruals analysis and comparison of net income to operating cash flow.

(Bonetti et al 2012) titled:
“The Influence of Country- and Firm-Level Governance on Financial Reporting Quality: Revisiting the Evidence”
This paper aimed to revisit the joint effect of country-level legal enforcement and firm-level governance on the quality of financial reporting. The sample was a group of companies implementing IFRS. The main results implied that firms operating in weak legal enforcement countries enhance their earnings quality by having strong efficient CG practices. This finding suggested that in strong enforcement countries, CG is a voluntary process. The main finding was that IFRS adoption by itself does not much affect earnings quality and that any such effect is conditional upon firm- and country-level governance.

(Hamdan et al, 2012)
“The Factors Affecting the Quality of Earnings: Evidence from Jordanian Industrial Companies”
This study aimed to check the quality of earnings for public shareholding industrial companies listed in ASE, as well as testing the factors that affect the level of quality such as: accounting discretion, debt contracts, company size, return on investment, quality of the audit and audit committees. The sample of 50 companies covering the period 2004-2009 was examined. The top results indicated high quality of earnings in the industrial sector, as well as a positive relationship between the company’s size, audit quality and debt contracts with the quality of earnings.

(Meeampol et al, 2013) titled:
“The Relationship between Corporate Governance and Earnings Quality: Case Study of Listed Companies in the Stock Exchange of Thailand (SET)”
The objective of this study was to investigate the practices of corporate governance of listed companies in SET in the period 2006-2008, and assed the quality of their earnings. The most important results this study had reached that there is a relationship between IOD score and standard deviation of net income. It also found that if there is a standard deviation for firms listed in (SET) has IOD scores at 5 percent significance level, then the relationship between governance and earnings quality is positive and the firm has high quality CG, lower earnings management, and higher quality of earnings.

(Basilico, 2013) titled:
“The Quality of Earnings, Governance and Future Stock Returns in Europe. An Empirical Study.”
This study aimed to investigate the practice of inappropriate devious behavior of managers in compiling financial reports and its impact on investors through the links between the quality of earnings, corporate governance and future stocks returns. It reviewed the impact IFRS on the quality of earnings to determine in which European countries it is possible to exploit the accruals mispricing to build outperforming stocks portfolios. The study assessed the relationship between manipulating accruals and industry affiliation across different European countries and studied the importance of corporate governance characteristics to add value to the quality of earnings.
The results were that earnings management decreased, but the accruals mispricing is still present is some European countries. The accruals misrepresentation is not present in all industries within the European sample studied. It was also found that CG quality matters and is linked to higher quality and higher future stock returns in the Netherlands.

(Zgarni et al, 2014) titled:
“Do the Characteristics of Board of Directors Constrain Real Earnings Management in Emerging Markets? – Evidence from the Tunisian Context”
This study aimed to test the effect of the board of director’s independence, size and number of annual meetings on lowering the levels of earnings management. A sample of (29) non-financial companies listed in the Tunisian financial market was tested using Roychowdhury model.
The results had shown that the independence of the board decreases the percentage of earnings management, and that the implementation of the financial instruments law led to decreasing earnings management. It had also shown a negative relationship between the number of the board’s annuals meetings and the percentage of earnings management.

(Al Areeni et al, 2014) titled:
“The Modified Jones and Yoon Models in Detecting Earnings Management in Palestine
Exchange (PEX)”
This study aimed to investigate the effectiveness of the Modified Jones (1995) and Yoon et al., (2006) model in Palestine Exchange, and how well these models are efficient in detecting earnings management practiced by Palestinian listed companies in the PEX from the period of 2006 – 2011. The population of the study included all listed companies in the PEX in all sectors such as banking, industry, insurance, investments and services.
The results indicated that the Yoon et al., (2006) model is better than the Modified Jones (1995) model in detecting earnings management practiced by targeted companies in PEX.

(Matar et al, 2014) titled:
“Earnings Management Methods and their Impact on the Reliability of the Published Financial Statements of Jordanian Public Shareholding Companies”
The study’s objective was to reveal the extent to which managements manipulate and manage earnings, and how that could reflect on the credibility of its published financial statements. The researchers implemented a descriptive analytical methodology and found that any earnings management techniques have a direct or indirect effect on the trustworthiness of the financial reports published. It has also reached that companies with fraudulent managements are considered the most significant, while those using income smoothing are considered the least significant, when it comes to the reliability of the data reported.
(Hashmi et al, 2016) titled:
“Impact of Corporate Governance Measures on Earnings Quality: Evidence from Pakistan”
The purpose of this study was to examine the impact of corporate governance on earnings quality on a sample taken from Karachi Stock Exchange of 100 index including 70 non-financial listed firms. Whereas, earning quality of the firm has been addressed by earnings management and discretionary accruals used to measure the level of earnings management. The Study’s main results had shown a negative impact of audit quality and board size on earnings management, while the relationship between firm size and earnings management is positively significant.

(Abdul-Hamid et al, 2016) titled:
“The Impact of Disclosure Quality on Corporate Governance and Earnings Management: Evidence from Companies in Indonesia”
This research studied the quality of disclosure and its relationship with CG and earnings management. The study’s sample was 175 listed manufacturing firms on Indonesia Stocks Exchange (IDX) during period 2009-2013. The analysis methodology used was Moderated Regression Analysis (MRA). The results indicated that there is a significant effect on disclosure quality on the relationship between corporate governance practices such as number of members in the board of directors, their compensations, and earnings management. The results indicated that disclosure quality and good corporate governance can reduce earnings management and manipulation. This study also found insignificant moderation effect of disclosure quality on the relationship between independence of the audit committee and earnings management variables.

(Machdar et al, 2017) titled:
“The Effects of Earnings Quality, Conservatism, and Real Earnings Management on the Company’s Performance and Information Asymmetry as a Moderating Variable”
This study researched the deviation of information and its’ effect on the strength or weakness of earnings quality. It was conducted in Indonesia and Singapore and covered the period from 2004-2013. The main findings of the research included that that impact of earnings quality on the company’s performance is varying. Accruals quality can significantly affect the company’s performance, while income smoothing does not affect the company’s performance. Information asymmetry weakens the effect of earnings management and accruals quality on stock returns. On the other hand, it strengthens the real earnings management influence on company performance.

(Al-Attar et al, 2017) titled:
“The Effect of Earnings Quality on the Predictability of Accruals and Cash Flow Models In Forecasting Future Cash Flows”
This study answered two main questions, the first was, “is the superiority of aggregate earnings over aggregate cash flows affected by the level of earnings quality?” In addition, the second was, “Is the superiority of aggregate earnings over the main components of earnings (i.e. operating cash flows, accounts receivable, inventory, accounts payable, and depreciation) affected by the level of earnings quality?”. The Study took place in Jordan and examined a sample of Industrial and services firms. The main findings of the study were that the quality of earnings affect the predictability of cash flow and earnings, as well as providing evidence of using all accounting data reported in the financial statements to enhance the predictability of future cash flows.

The Difference between Previous Studies and this Study
Corporate governance became a hot research topic in this era, and caught the attention of many researchers to study its practices and effects on many things including sustainability, stock prices, market value, earnings management, risk management, organizational structure and foremost, the quality of earnings.
The researcher aimed to determine the impact of CG practices on the quality of earnings for public shareholding companies listed in ASE. This study comes to follow those efforts and aims to differ from the previous by studying a sample from all industries of public shareholding companies listed in ASE. It will study the CG factors that might affect quality of earnings such as number of the board of directors’ members, and the number of their meetings. It will also assess number of the audit committee members, and the number of their meetings. The period the study will cover a more recent period than previous studies, covering the period from 2013-2016, and covering a wider span than previous studies. The purposive sample will have a bigger oversight on the market as a whole, which will make the implementation and retesting of this study easier for future studies, and more accurate to base results and recommendations upon.
What differs the study from previous studies is mainly:
It is a modern study, covering the period from 2013-2016.
The sample of the study was inclusive of all sectors: industrial, financial and services.
It used the most important variables of corporate governance: the set-up of the Board of directors and the set-up of the audit committee.
The researcher used the Total Accruals model by Richardson, which is more modern compared with other models used in other previous studies such as Jones Model or Jones Modified Model.

Chapter Three
Methodology
Introduction
This chapter will present the methodology, the population, the sample of the study, its’ variables, and the statistical analysis techniques which were used to test it, as well as the resources of the information gathered.
Methodology of the Study
This study used a descriptive analytical methodology based on an empirical study using actual financial data published on ASE for the period 2013-2016, as it is considered the most efficient way to get the most accurate results possible regarding the impact of CG on quality of earnings in public shareholding companies listed in Amman Stock Exchange.
Population of the Study
This study’s population is the public shareholding companies listed in ASE from all sectors: financial, services and industrial. As reported during the period 2013-2016, the number of the public shareholding companies listed are 131 companies in the three economic sectors.

Sample of the Study
The study will examine a sample composed of 60 public shareholding companies, selected from a population of 131 companies in ASE based on the data disclosed by these firms, from the ASE database. The study will cover their financials reported in the period 2013-2016. The sample was selected from ASE because the amount of data disclosed by most of companies was not available on ASE, neither was given when the companies were contacted or interviewed.
Table (1): Sample of the Study
Industry No. of Companies % from Population % from Sample
Financial 13 76% 22%
Industrial 37 55% 61%
Services 10 22% 17%
60 100%

Sources of Information
Three sources of information used in the study were:
Primary source of information; which was the annual reports and financial data of the sample disclosed on ASE website.
Secondary source of information: previous literature, books, magazines, journals, articles and any other sources of information that might be relevant.
The researcher also contacted many of the factories in the industrial sector, while only one provided the required information.
Variables of the Study

Figure (2): The Study Model
EQ= ?0 + ?1*NBD + ?2* NMBD + ?3* NAC + ?4* NMAC + ¥
Designed by the researcher.

Statistical Techniques Used:
The study will use the data collected from the annual reports of the sample in order to test the hypotheses through five methods:
Descriptive statistical analysis: through measuring mean, maximum, minimum, and standard deviation.
Smirnov and Shapiro Normality Tests.
Multicollinearity through variance inflation factor (VIF) and tolerance.
Pearson Correlation Matrix.
Multiple regression.

Chapter Four
Statistical Analysis and Hypotheses Testing
Introduction
The data collected from the sample companies were analyzed in order to obtain the results of the study. This chapter contains four main sections. The first section includes descriptive analysis of the study data to describe the sample depending on the descriptive statistical measures, the second section contains the verification of the study data and its validity for the statistical analysis, and to ensure that they are compatible with the normal distribution hypotheses in addition to Multicollinearity Test, the third section examined the correlation between the variables of the study, and the final section examined the hypotheses of this study depending on multiple regression models.
The Descriptive Statistics of the Study and its Variables
In the analysis of the descriptive data of the independent and dependent variables of this study, the researcher examined the published financial statements of the companies included in the study sample which are (60) companies and published in the annexes. These data have been divided into two parts:
The First Part:
After data has been collected associated with Corporate Governance which represented by (number of the board of directors members (NBD), number of meetings held by the board of directors (NMBD), number of the audit committee members (NAC), and number of meetings held by the audit committee (NMAC)) from the financial reports of the sample companies presented in appendix 1, the descriptive analysis of the independent variables was measured as follows:
Table (2): Results of descriptive analysis of the independent study variables
Measurements – variables NBD NMBD NAC NMAC
Maximum 19 19 7 20
Minimum 4 0 0 0
Mean 9 7 3 4
Std. Deviation 2.739 2.454 1.483 3.148
Kurtosis -0.029 2.425 1.256 1.461
Skewness 0.539 0.334 -0.193 0.744

Table (2) presents the results of descriptive analysis of the independent study variables represented by (number of the board of directors members (NBD), number of meetings held by the board of directors (NMBD), number of the audit committee members (NAC) and number of meetings held by the audit committee (NMAC)), it is noted from the table that the arithmetic mean of number of members of the board of directors is (9) which means that the average number of members of the board of directors in the study sample companies is nine members. The largest number of members of the board of directors in the sample companies was (19) members, where it was in the Arab Potash Company in 2016, while the lowest number of members of the board of directors was (4) members, where it was in a number of companies such as Sheba Metal Casting Company in 2014 and Jordanian Dairy Company in 2014 and 2015. The value of the standard deviation of the number of directors reached (2.739), where this value indicates the dispersion of the study sample from its arithmetic mean. The table above also shows the arithmetic mean number of meetings held by the board of directors which is (7), it indicates that the average of number of meetings held by the board of directors in the sample companies is seven during the year. The largest number of meetings was held by the members of the board of directors in the sample companies was (19) meetings, where they were held at the Commercial Bank of Jordan in 2015 and the Jordanian Free Markets Company in 2014, while the lowest number of meetings of the board of directors was (0) meeting, where it was in Jordan Pipes Manufacturing company. As for the number of the audit committee members, its arithmetic mean in the study sample companies was (3) members, the largest number of the audit committee members in the study sample companies was (7) members, while the lowest number of the audit committee members was (0) member, where it was in more than one company such as National Steel Industry Company and National Aluminum Industrial Company, the standard deviation reached (1.483) which indicates the extent of dispersion of the study sample data.
The table above shows the arithmetic mean of the number of meetings held by the audit committee was (4), which indicates that the average number of meetings held by the audit committee in the study sample companies. The largest number of meetings was held by the audit committees in the sample companies was (20) meetings where it was in the Jordanian Capital Bank in 2014,, while the lowest number of meetings was held by the audit committees was (0) meeting such as the Jordanian Chemical Industries company, and the National Poultry.
It should be noted that all the sample companies comply with the Corporate Governance Law in terms of the number of members of the Board of Directors and the number of meetings, while it was found that many companies do not comply with the standard in terms of the number of members of the Audit Committee and the number of their meetings.
Also it is noted that the indicators of Skewness and Kurtosis of all the variables are within the appropriate range. According to Kamiya et al., 2014, the decision rule is that the value of Skewness should be between (-1.3) and (+1.3), and the value of Kurtosis should be between (-3.75) and (3.75) in order for the data to be suitable for normal distribution conditions. According to each of the values in the above table, it is clear that they meet the requirements of normal distribution.

The Second Part:
After collecting the financial statements of the dependent variable which is Earnings Quality depending on (Ricardson et al., 2003), from the financial reports of the sample companies presented in annex 1, the descriptive analysis of the earnings management method was measured as follows:
Table (3): Results of descriptive analysis of earnings quality variable
Years- Measurements Minimum Maximum Mean Std. Deviation
2014 0.0024 0.9061 0.1445 0.1622
2015 0.0001 0.7702 0.1404 0.1493
2016 0.0007 1.0866 0.1279 0.1846
All Years 0.0001 1.0866 0.1376 0.1653
Skewness : 1.081 Kurtosis : 3.293

Table (3) shows the results of the descriptive test of the dependent study variable represented by Earnings Quality and measured by the size of the total accruals. When the total accruals increase, this means that the earnings quality decrease, and vice versa. According to the above table, it is noted that the highest average of earnings quality was in 2016, where the total accruals reached (0.1279), while the lowest average of earnings quality was in 2014, where the total accruals reached (0.1445). Also, as for the arithmetic mean of the total accruals of the sample companies in all the years which reached (0.1376), it is noted that it became lower than the average of total accruals in 2014 and 2015, and higher than the average of total accruals in 2016, which indicates that the earnings quality decreased in 2014 and 2015 from the average of the study years, and the earnings quality increased in 2016 from the average of the study years. Moreover, it is noted that the highest value of earnings quality was in 2015, where is was in the National Oil and Electricity Production from Oil Shale Company, its total accruals in that year reached (0.0001), while the lowest value of earnings quality was in 2016, where it was in Comprehensive Multiple Projects Company, its total accruals in that year reached (1.0866).
Also it is noted from table number (3) that the indicators of Skewness and Kurtosis of all the variables are within the appropriate range. According to Kamiya et al., 2014, the decision rule is that the value of Skewness should be between (-1.3) and (+1.3), and the value of Kurtosis should be between (-3.75) and (3.75) in order for the data to be suitable for normal distribution conditions. According to each of the values in the above table, it is clear that they meet the requirements of normal distribution.
The Verification of the Validity of Vata for the Statistical Analysis
It is important to ensure that the study data meet the conditions that confirm that it is suitable for the hypothesis analysis tests by conducting the normal distribution test of the Standardized Residual. The linear regression hypotheses confirm the need to distribute the residuals naturally (Babbie et al., 2018), where it is important to follow the residuals of natural distribution. In the case of non-compliance with the normal distribution conditions, a non- parametric statistical analysis should be conducted for data that is characterized by abnormal distribution. The results of the verification tests of the study data will be presented as follows: testing the normal distribution of the residuals and linear interference tests:

Normal Distribution Test

Figure (3): Normal distribution of residuals in the study model
Figure (3) presents the results of normal distribution verification of the residuals using the Kolmogorov-Smirnov test. According to this test, the decision rule indicates that the residuals follow normal distribution if the P-value is greater than (0.05) (Hayduket al., 2007), in contrast, the distribution is considered abnormal for the residuals. According to figure 1, the residuals follow the normal distribution, where the P-value of this test reached (0.072), therefore, it is possible to use the parametric tests of data related to finding the impact of Corporate Governance on the Earnings Quality in Public Shareholding Companies which represent the study sample.

Linear Interference Test
Multicollinearity Test was used to test the validity of study data for statistical analysis, where the researchers Luo et al., (2007) defined Multicollinearity Test as a phenomenon that occurs when there is a high correlation between two or more independent variables in the multiple regression models. This leads to negative effects on analysis by eliminating the ability to clarify the results and conclusions of this study, which in turn will affect the generalization and accuracy of the study model. In order to ensure that there is no problem in linear interference, variance inflation factor (VIF) and tolerance coefficient will be used, as shown in table (4) as follows:
Table (4) The validity of the study data for statistical analysis
Multicollinearity
VIF Variable’s Tolerance
1.290 Number of the board of directors members (NBD) 0.775
1.396 Number of meetings held by the board of directors (NMBD) 0.716
1.520 Number of the audit committee members (NAC) 0.658
2.165 Number of meetings held by the audit committee (NMAC) 0.462

Variance inflation factor (VIF) is used to determine interference between variables and the general rule of VIF. There is inflation according to this indicator when the value is greater than 10, thus, the regression coefficients are poorly determined due to increased inflation between independent variables (Schreiber & Jackson, 2017). According to Table (4), all values were less than 10 in relation to the inflation coefficient.
As for tolerance coefficient, it is another test that used to detect the self-correlation problem. It can be determined that there is a problem of self-correlation if the tolerance factor value is less than (0.20). As for table (4), all values were greater than (0.20) in relation to the inflation coefficient. Based on the hypotheses of the two previous indicators, it is clear that all study variables exceeded these two indicators, which means that there is no problem of linear interference and self-correlation in the study model.
Correlation Matrix for Study Variables
Pearson correlation matrix was adopted to find the correlation between independent variables and between each of them with the dependent variable as follows:
Table (5): Results of correlation matrix between the study variables
Correlation
Prob EQ NBD NMBD NAC NMAC
EQ 1
—–
NBD -0.216** 1
0.004 —–
NMBD -0.043 0.128 1
0.568 0.086 —–
NAC -0.151* 0.362** 0.123 1
0.000 0.000 0.099 —–
NMAC -0.165* 0.445** 0.496** 0.548** 1
0.027 0.000 0.000 0.000 —–

According to statistical data in Table (5), it is clear that:
There is statistically significant relationships between earnings quality and each of the independent variables except for the number of meetings held by the board of directors, where the significance value and correlation coefficients were respectively: (R = -0.043, Sig = 0.568), at a significant level (0.05), which means that there is no significant correlation between the number of meetings held by the board of directors and earnings quality.
As for the relationship between the dependent variable and the remaining independent variables, it is clear that the strongest correlation coefficient is the coefficient which correlates between earnings quality and number of the board of directors members, where the correlation value reached (R = -0.216, Sig = 0.004) , at a significant level (0.01), which means that there is a statistically significant negative correlation, which means that total accruals decrease with the increase of number of the board of directors members, which is reflected positively on earnings quality, followed by the correlation value which reached (R = -0.165, Sig = 0.027), at a significant level (0.05), between earnings quality and number of meetings held by the audit committee, which indicates that there is a statistically significant negative correlation, which means that total accruals decrease with the increase of number of meetings held by the audit committee, which is reflected positively on earnings quality. While the lowest correlation coefficient was between earnings quality and number of the audit committee members, where the correlation value which reached (R= -0.151, Sig=0.000), at a significant level (0.05), indicating that there is a statistically significant negative correlation, which means that total accruals decrease with the increase of number of the audit committee members, which is reflected positively on earnings quality.
Also, there is some statistically significant relationships with different values of the correlation coefficient among the independent variables, the strongest correlation coefficient is the coefficient which correlates between number of the audit committee members and number of meetings held by the audit committee, where the correlation value which reached (R= 0.548, Sig= 0.000), at a significant level (0.01), which means that there is a very strong positive correlation with statistical significance, which means that the number of meetings held by the audit committee increases with the increase of the number of the audit committee members, while the lowest correlation coefficient was between number of the board of directors members and number of meetings held by the board of directors, where the correlation value which reached (R = 0.362, Sig = 0.000), at a significant level (0.01), which means that there is a very strong positive correlation with statistical significance, which means that number of meetings held by the board of directors increases with the increase of number of the board of directors members.
Moreover, the results of correlation test analysis indicated that there is no correlation higher than 80% among the independent study variables, which indicates that there is no high correlation problem among the independent study variables, and this finding is consistent with what was previously achieved through the linear interference test.

Test of Study Hypotheses
Ho1: There’s no significant impact of corporate governance on the earnings quality in public shareholding companies listed in ASE.
In order to determine the result of the main null hypothesis, the study used multiple regression analysis. The Sig F value was adopted to accept or reject the study model and to determine the extent of its suitability to represent the relationship between the independent variables and the dependent variable, where the decision rule indicates that the model is accepted when the Sig F value is less than 0.05. In order to determine the impact of each one of the independent variable separately on the dependent variable, the Sig T value was adopted in this study, where the decision rule indicates that there is an effect when the value of Sig T is less than (0.05) in order to accept the alternative hypothesis and reject the null hypothesis, and in order to indicate the accuracy of the explanation of the independent variables to the dependent variable, the adjusted R Square value was adopted.
The next section presents the results of multiple regression analysis with an explanation of these results for the study model, and then the results will be compared with previous literature and studies.

The Study Model
EQ= ?0 + ?1*NBD + ?2* NMBD + ?3* NAC + ?4* NMAC + ¥
The study model is designed to examine the impact of corporate governance practices on the earnings quality of the sample companies listed on the Amman Stock Exchange, table (6) shows the results of the multiple regression test of the study model as follows:
Table (6) Multiple regression test results of the study

Variable Prob. (T-statistic) T-Statistic Coefficient ? Std. Error

Constant 0.000 4.383 ————– 0.249 0.057
NBD 0.046 -2.013 -0.168 -0.010 0.005
NMBD 0.816 0.233 0.020 0.001 0.006
NAC 0.552 -0.597 -0.054 -0.006 0.010
NMAC 0.517 -0.649 -0.070 -0.004 0.006

Prob(F-statistic) 0.041
S.E. of regression 0.162
F-statistic 2.553
Adjusted R-square 0.034
R 0.235
R-squared 0.055

Table (7) shows the results of the multiple regression tests of the independent study variables represented by corporate governance with (number of the board of directors members, number of meetings held by the board of directors, number of the audit committee members, and number of meetings held by the audit committee) and their impact on the dependent variable (earnings quality) for the Public Shareholding sample companies. It is noted from the table that the value of F calculated reached (2.553), at a significant level (0.05), indicating that the first proposed model of the study has very little suitable and has the explanatory power. The results of regression analysis showed that the value of Sig F reached (0.041), which is less than the significance of the test which is (0.05). Therefore, the main null hypothesis should be rejected and the alternative hypothesis should be accepted, which means that there’s a significant impact of corporate governance on the earnings quality in public shareholding companies listed in ASE.
The results indicated that the adjusted R2 value reached (0.034), which means that only about 3.4% of the fluctuations in the earnings quality of the sample companies can be explained by the changes that occur in one part of the dimensions of the corporate governance represented by number of the board of director’s members. It is noted that the value of the Adjusted R2 should be between 0-1 and, if its value is greater than 30%, it is possible to construct a mathematical equation to predict the earnings quality through corporate governance. The significant decrease in the adjusted R2 value indicates that there are other factors outside the interrelationship between these two variables may mutually affect each other. Therefore, the researcher will not be able to formulate the multiple linear regression equation in mathematical form to indicate this prediction.
In order to determine the impact of each variable of corporate governance on earnings quality, the multiple regression test results were adopted as follows:
Ho1-1: There is no significant impact of the number of the board of directors members on earnings quality.
It is noted from table (6) that the value of Sig T which was less than (0.05) reached (0.046), and according to the decision rule that indicates to reject the null hypothesis and accept the alternative hypothesis if the Sig T is less than (5%), and thus the number of the board of directors members impact on earnings quality in companies. Accordingly, it was found that there is an impact of the number of the board of director’s members on earnings quality. As for the coefficient value which reached (-0.168), there is a negative impact of the number of the boards of directors members on total accruals, which is reflected positively on earnings quality, which indicates that the number of the board of directors members is the only dimension that impact on earnings quality between the corporate governance dimensions of the study.
Ho1-2: There is no significant impact of the number of meetings held by the board of directors on earnings quality.
It is noted from the results of the analysis that Sig T value which was greater than (0.05) reached (0.816), and according to the decision rule that indicates to accept the null hypothesis if Sig T value is greater than (5%), and thus the number of meetings held by the board of directors has no statistically significant impact on earnings quality. In addition, it has a weak positive impact on total accruals, which is reflected negatively on earnings quality, where the value of the coefficient reached (0.020), at significance level greater than 5%.
Ho1-3: There is no significant impact of the number of the audit committee members on earnings quality.
It is noted from the results of the analysis that Sig T value which was greater than (0.05) reached (0.552), and according to the decision rule that indicates to accept the null hypothesis if Sig T value is greater than (5%), and thus the number of the audit committee members has no statistically significant impact on earnings quality. Thus, it has a weak negative impact on total accruals, which is reflected positively on earnings quality, where the value of the coefficient reached (-0.054), at significance level greater than 5%.
Ho1-4: There is no significant impact of the number of meetings held by the audit committee on earnings quality.
It is noted from the results of the analysis that Sig T value which was greater than (0.05) reached (0.517), and according to the decision rule that indicates to accept the null hypothesis if Sig T value is greater than (5%), and thus the number of meetings held by the audit committee has no statistically significant impact on earnings quality. In addition, it has a weak negative impact on total accruals, which is reflected positively on earnings quality, where the value of the coefficient reached (-0.070), at significance level greater than 5%.
On the other hand, the gradient regression analysis shows that although the test revealed that there is a statistically significant effect between independent variables each with the dependent variable, the relative weight of these variables varies in the formation of the new regression equation.
The largest weight in this equation is the number of the board of directors members. While the lowest weight is the number of the audit committee members. Therefore, taking into account the value of Constant (0.249), the linear gradient equation, which represents the model of predicting the earnings quality for the Jordanian public shareholding companies, was formulated. If only four independent variables were adopted as a function of this prediction:
EQ= 0.249 – 0.010*NBD + 0.001* NMBD – 0.006* NAC – 0.004* NMAC + ¥, Where:
EQ: Earnings Quality and the measure of total accruals.
NBD: Number of the board of directors members.
NMBD: Number of meetings held by the board of directors.
NAC: Number of the audit committee members.
NMAC: Number of meetings held by the audit committee.
¥: Margin of error.
Chapter Five
Results and Recommendations
Results
The study reached many results; the most important of which are:-
There is a significant impact of corporate governance practices when considered together as a unit on earnings quality in public shareholding companies listed on ASE. According to the researcher, this is due to the important role of corporate governance mechanisms which are mainly concerned with the protection and guarantee of the rights of shareholders and all stakeholders involved in the company’s business. This is done by carefully controlling and supervising the performance of the company’s management. Moreover, corporate governance also plays a significant role in preventing the collusion of auditors with those who have relations and interests in the company such as the management and investors. This result agrees with the findings reached by (Bistrova et al 2012); (Bonetti et al 2012); (Meeampol et al 2013); (Basilico, 2013); and (Abdul-Hamid et al, 2016).
The following outcomes are derived from the above mentioned result:
There is a significant impact of the number of the board of directors’ members on earnings quality. According to the researcher, this is due to the business nature of the board of directors which is considered one of the main mechanism governing the performance of managers CEO through controlling their performance to reduce the undesired behavior in a manner that enables the development of the company’s strategies which aim at preventing earnings management. As well, the number of the board of directors’ members is also seen an important factor in the effectiveness of the board of directors because of the diversity of its members’ expertise and their ability to better deal with the environment. Thus, this explains the important role of the number of the board of directors’ members in achieving earnings quality. This outcome is found consistent with the findings of (Hashmi et al, 2016).
There is no significant impact of the number of meetings held by the board of directors on earnings quality. According to the researcher, this is due to the fact that the efficiency of the board of directors in implementing governance does not come from the number of rout in meetings held by the board, but rather from the efficiency of its members by having the appropriate expertise as well as their ability to deal with all problems effectively. This outcome contradicts the findings of (Zgarni et al, 2014).
There is no significant impact of the number of the audit committee members on earnings quality. According to the researcher, this is due to the fact that role of the audit committees in ensuring the quality of financial reports and achieving confidence in accounting information is the result of their supervision of internal and external audits as well as their resistance of pressures and interventions by management on the audit process. Basically, the audit committee enjoys independence and it has the necessary powers. Therefore, the efficiency of the work of the audit committee is not related to the number of its members as much as it is related to its integrity and independence from management; thus, explaining the fact that there is no significant of impact the number of the audit committee members on earnings quality, In addition to the reference to the condition that the formation of the audit committees in the Jordanian public shareholding companies is relatively recent and the conditions for the formation of these committees have not yet become entrenched, especially in terms of conditions that provide the element of independence and the financial and professional expertise of its members. This outcome contradicts the findings of the study of (Hwang, Lin, 2010) due to differences in environment and time.
There is no significant impact of the number of meetings held by the audit committee on earnings quality. According to the researcher, this goes to the fact that the efficiency of the audit committee does not come from the number of meetings held by the committee; rather, it stems from the efficiency of its members through their possession of appropriate expertise, in addition to having necessary independence and powers. This outcome contradicts the findings of (Hwang, Lin, 2010) due to differences in environment and time.
With regard to the impact of corporate governance, the researcher derived the following mathematical model as a tool to predict its impact on the quality of profits of public shareholding companies listed on the Amman Stock Exchange as follows:
EQ= 0.249 – 0.010*NBD + 0.001* NMBD – 0.006* NAC – 0.004* NMAC + ¥
Where:
EQ: Earnings Quality and the measure of total accruals.
NBD: Number of the board of directors members.
NMBD: Number of meetings held by the board of directors.
NAC: Number of the audit committee members.
NMAC: Number of meetings held by the audit committee.
¥: Margin of error.

On the other hand, and to emphasis the relationship between public shareholding company’s commitment to governance and the quality of its’ earnings in each economic sector, the researcher analyzed the data in appendices 1 and 2, and the results where as below:

Table (7): The Relationship between Corporate Governance and Earnings Quality on each Economic Sector.
Sector No. of Companies Governance Earnings Quality
Commitment % Rating EQ % Rating
Financial 13 13 100% 1 12 92% 1
Industrial 37 34 92% 2 24 65% 2
Services 10 5 50% 3 3 30% 3
Total 60 52 87% 39 70%

Based on the above data, using appendices 1 and 2, the researcher have found that:
1. When the three economic sectors are taken as a whole unit, 52 companies, or 87% of the sample apply CG principles, and this reflects on the nature of their earnings quality as only 39 companies, or 70% of the sample achieved good quality of earnings.
2. When we take each sector individually, the compliance with CG principles vary, and its effect on earnings quality varies as well. As for the Financial sector, at achieved the first place in complying 100 % with CG principles. The second was the industrial sector which complied by 92%, while the services sector came third and committed by 50%. Same applies for earnings quality, whereas the financial sector came first and scored 92%, following it came the industrial sector with a 67% and finally, 30% of earnings quality was found in the third and final sector; the services sector.
3. Comparing the level of CG practices compliance of multiple companies in different sectors, and its’ implication on earnings quality, a positive relationship is found between both. And thus, the three economic sectors are sorted descending based on the level of compliance and the degree of earnings quality.
The First was the Financial sector, the second was the Industrial and the third and last was the Services sector.
4. The researcher justifies the variance between the sectors in their commitment to comply with CG principles and practices, with their earnings quality is due to:
Firstly: the first variable, corporate governance, was measured used two main components, the number of board of directors members and the number of their meetings, and the number of the audit committee members and the number of their meetings.
Other significant factors like independence of the board of directors, and family relationships between owners and the board of directors, as well as independence and financial background of the audit committee members were left out due to lack of disclosure for such variables, which caused a problem in collecting data related to them. Noting that the services sector did not meet the minimal requirements of corporate governance practices disclosure.
The second variable was earnings quality, which was predicted using Richardson model, and mainly relays on (TACC), which makes its’ level of predictability and accuracy very limited. Remarking that (TACC) probability is usually concerned with the managements’ level of earnings’ management, which negatively affects earnings quality. ?
Recommendations
In light of the results found by the study, the researcher recommends the following:-
Urging competent parties as well as boards of directors in public shareholding companies to pay more attention to the development of more legislation pertaining to earnings quality through reducing creative accounting practices. This can be done by setting dissuasive penalties in cases of manipulation and misrepresentation of financial statements in a manner that limits the use of illegal techniques, in addition to raising awareness amongst companies on the importance of earnings quality in attracting investors and gaining their trust.
Playing a more effective role by governmental supervisory bodies and organizations to raise the implementation level of governance rules especially with regards to the number of the board of directors’ members and in accordance with the provisions and laws thereto. This will be also significant in establishing a good reputation for the companies listed on Amman Stock Exchange (ASE); thereby, attracting more foreign investments to this market which suffers of economic recession.
Achieving the independence of the majority of the board of directors’ members from any executive functions through the implementation of governance rules issued by the ASE for the year 2017, in addition to holding seminars for raising awareness of the companies’ boards of directors on the importance of the independence of these members and the significance of the role they play in detecting cases of manipulation which may take place in the company.
Increasing the number of the board of directors’ members in public shareholding companies listed on ASE as this is found by the study to have a positive effect on increasing earnings quality in companies.
Raising awareness amongst financial statements’ users, especially investors, regarding the importance of corporate governance in companies and its role in protecting their investment interests, in addition to urging them to invest in the companies which apply governance rules.
Developing and adopting a model for measuring earnings quality such as the model presented by (Ricardson et al 2003) in order to be used as an accurate indicator for investors and users to reflect the quality and credibility of profits of public shareholding companies announced in the annual reports. This will be actually important in directing the investments of such investors towards the ideal place.
Conducting more studies and researches which take into consideration many other aspects that have not been covered in this study including liquidity risks as well as systematic and unsystematic risks of the company. Furthermore, these studies should also include either a bigger sample or that they should cover more years so as to get more comprehensive results regarding earnings quality in companies.
Encourage companies to used the developed earnings quality model in predicting earnings quality in future studies.

References:
Abdul-Hamid, M.A., Jatiningrum, C., Popoola, O.M.J. (2016) The impact of disclosure quality on corporate governance and earnings management: evidence from companies in Indonesia. International Journal of Economics and Financial Issues, 6 (S4).
Abu Ali, S., Al-Debi’e, M., and Abu Nassar, M. (2011) The Impact of earnings quality on the cost of equity based on the international financial reporting standards. Jordan Journal of Business Adminstration, 7 (1).
Alareeni, B., Aljuaidi, O. (2014) The modified Jones and Yoon models in detecting earnings management in Palestine Exchange (PEX). International Journal of Innovation and Applied Studies, 9(1), pp. 1472-1484.
Babbie, E., Wagner III, W. E., ; Zaino, J. (2018). Adventures in social research: Data analysis using IBM® SPSS® Statistics. Sage Publications.
Basilico, Elisabetta (2013). The quality of earnings, governance and future stock returns in Europe. An empirical study (unpublished doctoral dissertation), St. Gallen University, St.Gallen: Switzerland.
Bistrova, J., Lace, N. (2012) Quality of corporate governance system and quality of reported earnings: evidence from CEE comapnies. Jordan of Economics and Management, 17 (1).
Black, B. and Bhafat, S. (1998). The uncertain relationship between board composition and firm performance (unpublished masters thesis), Columbia Law School, Columbia: USA.
Bonetti, P., Magnan, M. Parbonetti, A (2012). The influence of country and firm-level governance on financial reporting quality: revisiting the evidence (unpublished doctoral dissertation), University of Padova, Padova: Italy.
Chipper, Katherine. (1989) Commentary on Earnings Management. Accounting Horizons (December): pp. 91–102.
Companies Control Department, (2003). Jordanian Corporate Governance Code. Amman: Jordan.
Dechow, P. M., ; Dichev, I. D. (2002). The quality of accruals and earnings: The role of accrual estimation errors. The accounting review, 77(s-1), pp. 35-59.
Dibra, R. (2016) The role of corporate governance failure in the banking sector. European Scientific Journal, 12 (34), 7.
European Corporate Governance Institute, (2012). The history of corporate governance. University of Cambridge.
Ferlauto, K., McGowan, S., Rankin, M., Stanton, P., and Tilling, M. (2012). Contemporary issues in accounting, (1st ed.) Australia: John Wiley and Sons Australia, Ltd.
Fiori, Roberta (2008). The evolution of corporate governance systems: an analysis based on cross border M;A deals (unpublished masters thesis), Bocconi University, Milan: Italy.
Hamdan, Allam (2012) The factors affecting the quality of earnings: evidence from Jordanian industrial companies. Islamic University Journal for Managerial and Economic Studies, 20 (1).
Hashmi, S.H., Younis, S.A., and Khalid, G.K. (2016) Impact of corporate governance measures on earnings quality: evidence from Pakistan. Research Journal of Finance and Accounting, 7 (3).
Hayduk, L., Cummings, G., Boadu, K., Pazderka-Robinson, H., ; Boulianne, S. (2007). Testing! testing! one, two, three–Testing the theory in structural equation models!. Personality and Individual Differences, 42(5), pp. 841-850.
Hollemans, K.G.A.C. (2011). Earnings quality: accounting conservatism in public and private firms in the Netherlands (unpublished masters thesis), Tilburg University, Tilburg: Netherlands.
Islam, A., Ali, R., Ahmad, Z. (2011) Is modified Jones model effective in detecting earnings management? Evidence from a developing country. International Journal of Economics and Finance,3 (2).
Ismail, Adibah (2011). Eranings quality, family influence and corporate governance: Empirical evidence from Malaysia (unpublished masters thesis), Victoria University, Wellington: New Zealand.
Kamiya, A., Murayama, S., Kamiya, H., Yamashiro, T., Oshiro, Y., ; Tanaka, N. (2014). Kurtosis and skewness assessments of solid lung nodule density histograms: differentiating malignant from benign nodules on CT. Japanese journal of radiology, 32(1), pp. 14-21.
Kwamboka, Nyakoe (2010). The relationship between corporate governance and risk management practices among commercial banks in Kenya (unpublished masters thesis), University of Nairobi, Nairobi: Kenya.
Lin, J.W., Hwang, M.I. (2010) Audit quality, corporate governance, and earnings management: A meta-analysis. International Journal of Auditing 14(1), pp. 57-77.
Luo, B., Groenke, K., Takors, R., Wandrey, C., ; Oldiges, M. (2007). Simultaneous determination of multiple intracellular metabolites in glycolysis, pentose phosphate pathway and tricarboxylic acid cycle by liquid chromatography–mass spectrometry. Journal of chromatography A, 1147(2), pp. 153-164.
Mahmood, Z., Kouser, R., Ali, W., Ahmad, Z., and Salman, T. (2018) Does corporate governance affect sustainability disclosure? A mixed methods study. Sustainability Journal, 10 (207), 20.
Man, C., Wong, B. (2013) Corporate governance and earnings management: a survey of literature. The Journal of Applied Business Research, 2 (29), 28.
Management, Knowledge and Learning International Conference (2013). The relationship between corporate governance and earnings quality: a case study of listed comaonies in the stock exchange of Thailand (SET), Zadar: Kroatia.
Matar, M. (2009) The Impact of Accounting Systems and Professional Standards in Enhancing Effectiveness and Efficiency of Corporate Governance Systems of Jordanian Public Shareholding Companies. Dirasat Journal of Administrative Sciences, 36(2), pp. 459-476.
Matar, M., Da’as, A., and Nour, A.A. (2014) Earnings management methods and their impact on the reliability of the published financial statements of Jordanian public shareholding companies. Irbid Journal for Research and Studies, 19(1), pp. 137-173.
OECD corporate governance committee, (2011). OECD Principles of Corporate Governance, Paris: France.
Patricia M. Dechow and Ilia D. Dichev (2002) The Quality of Accruals and Earnings: The Role of Accrual Estimation Errors. The Accounting Review: Supplement 2002, Vol. 77, No. s-1, pp. 35-59.
Paul M. Healy and James M. Wahlen (1999) A Review of the Earnings Management Literature and Its Implications for Standard Setting. Accounting Horizons: December 1999, Vol. 13, No. 4, pp. 365-383
Rahman, M. M., Moniruzzaman, M., & Sharif, M. J. (2013). Techniques, motives and controls of earnings management. International Journal of Information Technology and Business Management, 11(1), pp. 22-34.
Sarun, Anuar (2016). Corporate governance, earnings quality and firm value: evidence from Malaysia, (unpublished doctoral dissertation), Victoria University, Footscray: Australia.
Schreiber-Gregory, D. N., & Jackson, H. M. (2017). Multicollinearity: what is it, why should we care, and how can it be controlled? Pp. 1-12.
Suntory and Toyota International Centres For Economics and Related Disciplines (2008). Earnings Quality, Accruals and Subjective Goodwill Accounting, The Suntory Centre, London: UK.
WeiL, R. L. (2009). Quality of Earnings and Earnings management. AICPA monograph, available at http://www. aicpa.org/download/audcommctr/Audit_Committee_Quality_of_Earnings. pdf,(February 2009).
Zgarni, I., Halioui, K., Zehri, F. (2014) Do the Characteristics of Board of Directors Constrain Real Earnings Management in Emerging Markets? – Evidence from the Tunisian Context. The IUP Journal of Accounting Research & Audit Practices, 8 (1), pp. 46-61.

List of Appendices
Appendix No. Content Page
Appendix 1 Corporate Governance Data 66
Appendix 2 Earnings Quality Data 74
Appendix 3 Outputs of Statistical Analysis 79
Appendix 4 Components of the equation of earnings quality data 84

Appendix (1): Corporate governance data
Financial Sector Year # of BOD Members # of Audit Committee Members # of BOD Meetings # of Audit Committee Meetings Corporate Governance
????? ??????? ??????? 2014 9 3 6 4 Applied
2015 9 3 6 5
2016 14 3 6 6
Mean 10.67 3.00 6.00 5.00
Result Applied Applied Applied Applied
????? ??????? ??????? 2014 7 3 18 19 Applied
2015 11 5 19 16
2016 11 4 16 11
Mean 9.67 4.00 17.67 15.33
Result Applied Applied Applied Applied
??? ??????? ??????? ???????? 2014 13 6 7 5 Applied
2015 13 6 7 5
2016 13 6 7 5
Mean 13.00 6.00 7.00 5.00
Result Applied Applied Applied Applied
??? ????????? ?????? ??????? 2014 11 3 6 6 Applied
2015 11 3 6 6
2016 13 3 8 5
Mean 11.67 3.00 6.67 5.67
Result Applied Applied Applied Applied
??? ??????? 2014 11 3 6 4 Applied
2015 13 6 7 7
2016 12 6 7 10
Mean 12.00 5.00 6.67 7.00
Result Applied Applied Applied Applied
??? ??????? ??????? ????????/?????? 2014 8 5 7 5 Applied
2015 11 5 6 4
2016 11 5 7 4
Mean 10.00 5.00 6.67 4.33
Result Applied Applied Applied Applied
????? ?????????? 2014 13 4 9 7 Applied
2015 11 4 7 7
2016 11 3 7 6
Mean 11.67 3.67 7.67 6.67
Result Applied Applied Applied Applied
??? ????? ??????? 2014 11 3 8 20 Applied
2015 13 3 7 14
2016 14 3 8 12
Mean 12.67 3.00 7.67 15.33
Result Applied Applied Applied Applied
??? ?????? ????? – ?????? 2014 13 3 6 4 Applied
2015 11 5 8 5
2016 12 6 8 5
Mean 12.00 4.67 7.33 4.67
Result Applied Applied Applied Applied
??? ??????? ???? 2014 13 3 6 5 Applied
2015 15 5 7 5
2016 13 3 6 7
Mean 13.67 3.67 6.33 5.67
Result Applied Applied Applied Applied
??? ?????? 2014 12 4 8 9 Applied
2015 12 4 8 8
2016 12 4 9 9
Mean 12.00 4.00 8.33 8.67
Result Applied Applied Applied Applied
????? ?????? ??????? 2014 13 3 10 6 Applied
2015 13 4 10 12
2016 13 3 12 12
Mean 13.00 3.33 10.67 10.00
Result Applied Applied Applied Applied
????? ?????? 2014 10 4 6 6 Applied
2015 11 4 6 6
2016 12 4 6 5
Mean 11.00 4.00 6.00 5.67
Result Applied Applied Applied Applied
Industrial Sector Year # of BOD Members # of Audit Committee Members # of BOD Meetings # of Audit Committee Meetings Corporate Governance
???????? ???????? ???????? / ??????? 2014 6 3 2 2 Applied
2015 5 3 6 6
2016 5 3 6 6
Mean 5.33 3.00 4.67 4.67
Result Applied Applied Applied Applied
???????? ??????? ????????? 2014 9 3 6 6 Applied
2015 10 3 6 6
2016 9 3 6 6
Mean 9.33 3.00 6.00 6.00
Result Applied Applied Applied Applied
???????? ????????? ???????? 2014 7 0 7 0 Not Applied
2015 7 0 6 0
2016 7 0 6 0
Mean 7.00 0.00 6.33 0.00
Result Applied Not Applied Applied Not Applied
??????? ?????? ???????? 2014 12 3 6 4 Applied
2015 12 3 6 4
2016 12 3 10 4
Mean 12.00 3.00 7.33 4.00
Result Applied Applied Applied Applied
??????? ???????? ???????? 2014 8 3 9 4 Applied
2015 5 3 7 4
2016 5 3 6 4
Mean 6.00 3.00 7.33 4.00
Result Applied Applied Applied Applied
????????? ???????? ???????? 2014 5 3 6 4 Applied
2015 5 3 6 4
2016 5 3 6 4
Mean 5.00 3.00 6.00 4.00
Result Applied Applied Applied Applied
??????? ?????? ???????? ???????? ???????? 2014 9 3 9 4 Applied
2015 9 3 9 4
2016 9 3 9 4
Mean 9.00 3.00 9.00 4.00
Result Applied Applied Applied Applied
??????? ?????? ??????? ???????? ?????????? 2014 11 3 6 4 Applied
2015 11 3 6 4
2016 11 3 6 4
Mean 11.00 3.00 6.00 4.00
Result Applied Applied Applied Applied
??????? ???????? ?????????? 2014 7 3 6 4 Applied
2015 7 3 6 4
2016 7 3 6 4
Mean 7.00 3.00 6.00 4.00
Result Applied Applied Applied Applied
????? ???????? ??????? 2014 9 3 5 4 Applied
2015 7 3 6 4
2016 7 3 8 4
Mean 7.67 3.00 6.33 4.00
Result Applied Applied Applied Applied
???????? ?????? ???????? 2014 7 3 0 0 Applied
2015 7 3 4 3
2016 7 3 4 3
Mean 7.00 3.00 2.67 2.00
Result Applied Applied Applied Applied
???????? ???????? ??????? / ?????? 2014 10 3 6 4 Applied
2015 9 3 6 4
2016 14 3 6 4
Mean 11.00 3.00 6.00 4.00
Result Applied Applied Applied Applied
??????? ?????? ???????? ???????? 2014 7 3 6 4 Applied
2015 7 3 6 4
2016 8 3 6 4
Mean 7.33 3.00 6.00 4.00
Result Applied Applied Applied Applied
????? ???????? ????????? 2014 10 3 6 4 Applied
2015 9 3 7 4
2016 9 3 4 4
Mean 9.33 3.00 5.67 4.00
Result Applied Applied Applied Applied
???? ???????? ????????? 2014 9 3 6 4 Applied
2015 9 3 6 5
2016 9 3 6 4
Mean 9.00 3.00 6.00 4.33
Result Applied Applied Applied Applied
??? ???? ??????? 2014 4 3 6 4 Applied
2015 5 3 6 4
2016 6 3 6 4
Mean 5.00 3.00 6.00 4.00
Result Applied Applied Applied Applied
???????? ?????? ?????? ??????? ????????? 2014 7 3 6 4 Applied
2015 7 3 6 4
2016 5 3 6 4
Mean 6.33 3.00 6.00 4.00
Result Applied Applied Applied Applied
??????? ???????? 2014 4 3 12 4 Applied
2015 4 3 10 4
2016 5 3 8 4
Mean 4.33 3.00 10.00 4.00
Result Applied Applied Applied Applied
??????????? ?????? 2014 8 3 6 4 Applied
2015 7 3 6 4
2016 7 3 6 4
Mean 7.33 3.00 6.00 4.00
Result Applied Applied Applied Applied
?????? ???????? ???????? ??????? ???????? 2014 5 3 6 4 Applied
2015 8 3 6 4
2016 8 3 6 4
Mean 7.00 3.00 6.00 4.00
Result Applied Applied Applied Applied
???????? ??????? ?????? ???????? 2014 8 3 7 4 Applied
2015 8 3 7 4
2016 10 3 6 4
Mean 8.67 3.00 6.67 4.00
Result Applied Applied Applied Applied
??????? ??????? 2014 6 0 6 0 Not Applied
2015 6 0 5 0
2016 6 0 5 0
Mean 6.00 0.00 5.33 0.00
Result Applied Not Applied Applied Not Applied
??????? ??????? ??????? ??????? ?????????? 2014 7 3 6 4 Applied
2015 9 3 7 4
2016 10 3 8 4
Mean 8.67 3.00 7.00 4.00
Result Applied Applied Applied Applied
??? ?????? 2014 5 0 6 0 Not Applied
2015 5 0 10 0
2016 5 0 8 0
Mean 5.00 0.00 8.00 0.00
Result Applied Not Applied Applied Not Applied
????? ?????? ???????? ???????? 2014 7 3 8 4 Applied
2015 7 3 9 4
2016 7 3 10 4
Mean 7.00 3.00 9.00 4.00
Result Applied Applied Applied Applied
?????? ???????? ???????? 2014 5 3 9 4 Applied
2015 5 3 6 4
2016 5 3 9 4
Mean 5.00 3.00 8.00 4.00
Result Applied Applied Applied Applied
?????? ??????? 2014 9 3 8 4 Applied
2015 9 3 7 4
2016 9 3 9 4
Mean 9.00 3.00 8.00 4.00
Result Applied Applied Applied Applied
??????? ?????? ?????????/???? 2014 6 3 7 6 Applied
2015 6 3 8 5
2016 6 3 8 6
Mean 6.00 3.00 7.67 5.67
Result Applied Applied Applied Applied
??????? ?????? ????? 2014 5 0 6 0 Applied
2015 5 0 6 0
2016 5 0 6 0
Mean 5.00 0.00 6.00 0.00
Result Applied Applied Applied Applied
????? ???????? ???????? 2014 11 3 6 4 Applied
2015 10 3 6 4
2016 9 3 6 4
Mean 10.00 3.00 6.00 4.00
Result Applied Applied Applied Applied
????? ??????? ???????? 2014 7 7 6 6 Applied
2015 7 7 6 6
2016 7 7 6 6
Mean 7.00 7.00 6.00 6.00
Result Applied Applied Applied Applied
??????? ??????? 2014 12 3 6 7 Applied
2015 13 3 6 7
2016 19 3 6 8
Mean 14.67 3.00 6.00 7.33
Result Applied Applied Applied Applied
???? ?????? 2014 13 3 6 3 Applied
2015 9 6 6 3
2016 7 6 6 3
Mean 9.67 5.00 6.00 3.00
Result Applied Applied Applied Applied
??????? ??????? ????????? 2014 9 0 6 0 Applied
2015 9 0 8 0
2016 9 0 7 0
Mean 9.00 0.00 7.00 0.00
Result Applied Applied Applied Applied
??????????? ????????? ????????? 2014 5 3 8 4 Applied
2015 6 3 6 4
2016 6 3 7 4
Mean 5.67 3.00 7.00 4.00
Result Applied Applied Applied Applied
???? ?????????? 2014 8 3 6 4 Applied
2015 7 3 6 4
2016 7 3 6 4
Mean 7.33 3.00 6.00 4.00
Result Applied Applied Applied Applied
??????? ?????? ????? ??????? ?????????? ?? ????? ?????? 2014 11 3 6 4 Applied
2015 11 3 6 4
2016 11 3 6 5
Mean 11.00 3.00 6.00 4.33
Result Applied Applied Applied Applied
Service Sector Year # of BOD Members # of Audit Committee Members # of BOD Meetings # of Audit Committee Meetings Corporate Governance
??????? ????? ???????? 2014 9 3 19 8 Applied
2015 10 3 13 10
2016 9 3 13 8
Mean 9.33 3.00 15.00 8.67
Result Applied Applied Applied Applied
?????? ??????? ??????? ??????? 2014 7 3 6 4 Applied
2015 7 3 6 4
2016 7 3 6 4
Mean 7.00 3.00 6.00 4.00
Result Applied Applied Applied Applied
????????? ???????? ???????? 2014 7 3 6 0 Not Applied
2015 7 3 6 0
2016 7 3 6 0
Mean 7.00 3.00 6.00 0.00
Result Applied Applied Applied Not Applied
???????? ??????? ???????????? 2014 7 3 7 4 Applied
2015 7 3 8 4
2016 7 3 7 4
Mean 7.00 3.00 7.33 4.00
Result Applied Applied Applied Applied
????? ??????? ?????????? 2014 9 0 5 0 Not Applied
2015 7 0 5 0
2016 8 0 6 0
Mean 8.00 0.00 5.33 0.00
Result Applied Not Applied Applied Not Applied
?????? ????? ??????? 2014 7 0 8 0 Not Applied
2015 7 0 7 0
2016 8 0 9 0
Mean 7.33 0.00 8.00 0.00
Result Applied Not Applied Applied Not Applied
?????? ???????????? 2014 8 0 5 0 Not Applied
2015 9 0 4 0
2016 8 0 6 0
Mean 8.33 0.00 5.00 0.00
Result Applied Not Applied Applied Not Applied
???? ?????? 2014 5 0 6 0 Not Applied
2015 5 0 6 0
2016 5 0 6 0
Mean 5.00 0.00 6.00 0.00
Result Applied Not Applied Applied Not Applied
????? ??????? ????????? ???????? 2014 7 3 8 4 Applied
2015 7 3 7 4
2016 8 3 13 4
Mean 7.33 3.00 9.33 4.00
Result Applied Applied Applied Applied
??????? ??????? ?????????? 2014 11 3 12 4 Applied
2015 11 3 12 4
2016 11 3 13 4
Mean 11.00 3.00 12.33 4.00
Result Applied Applied Applied Applied

?
Appendix (2): Earnings quality data
Financial Sector Year TACC Average of TACC Result
????? ??????? ??????? 2014 0.0299 0.0283 There is earnings quality
2015 0.0158
2016 0.0392
????? ??????? ??????? 2014 0.0728 0.1174 There is earnings quality
2015 0.1280
2016 0.1515
??? ??????? ??????? ???????? 2014 0.0476 0.0283 There is earnings quality
2015 0.0182
2016 0.0190
??? ????????? ?????? ??????? 2014 0.2751 0.1033 There is earnings quality
2015 0.0178
2016 0.0169
??? ??????? 2014 0.1251 0.0589 There is earnings quality
2015 0.0153
2016 0.0362
??? ??????? ??????? ????????/?????? 2014 0.1387 0.0994 There is earnings quality
2015 0.1149
2016 0.0446
????? ?????????? 2014 0.0046 0.0232 There is earnings quality
2015 0.0448
2016 0.0202
??? ????? ??????? 2014 0.0921 0.0554 There is earnings quality
2015 0.0681
2016 0.0060
??? ?????? ????? – ?????? 2014 0.2346 0.1578 There is no earnings quality
2015 0.1864
2016 0.0524
??? ??????? ???? 2014 0.0625 0.0352 There is earnings quality
2015 0.0193
2016 0.0238
??? ?????? 2014 0.0316 0.0480 There is earnings quality
2015 0.0889
2016 0.0234
????? ?????? ??????? 2014 0.2615 0.1198 There is earnings quality
2015 0.0168
2016 0.0811
????? ?????? 2014 0.0622 0.0612 There is earnings quality
2015 0.0622
2016 0.0593
Industrial Sector Year TACC Average of TACC Result
???????? ???????? ???????? / ??????? 2014 0.0792 0.1016 There is earnings quality
2015 0.1323
2016 0.0932
???????? ??????? ????????? 2014 0.1111 0.2696 There is no earnings quality
2015 0.0652
2016 0.6327
???????? ????????? ???????? 2014 0.1036 0.1631 There is no earnings quality
2015 0.3406
2016 0.0453
??????? ?????? ???????? 2014 0.1402 0.1370 There is earnings quality
2015 0.2267
2016 0.0440
??????? ???????? ???????? 2014 0.1835 0.1187 There is earnings quality
2015 0.1450
2016 0.0276
????????? ???????? ???????? 2014 0.0857 0.4650 There is no earnings quality
2015 0.2227
2016 1.0866
??????? ?????? ???????? ???????? ???????? 2014 0.3739 0.1772 There is no earnings quality
2015 0.0577
2016 0.1000
??????? ?????? ??????? ???????? ?????????? 2014 0.0268 0.0492 There is earnings quality
2015 0.0459
2016 0.0750
??????? ???????? ?????????? 2014 0.0054 0.0978 There is earnings quality
2015 0.1159
2016 0.1721
????? ???????? ??????? 2014 0.0539 0.1245 There is earnings quality
2015 0.0796
2016 0.2400
???????? ?????? ???????? 2014 0.0192 0.1301 There is earnings quality
2015 0.2514
2016 0.1198
???????? ???????? ??????? / ?????? 2014 0.1122 0.1594 There is no earnings quality
2015 0.2607
2016 0.1054
??????? ?????? ???????? ???????? 2014 0.0024 0.0301 There is earnings quality
2015 0.0425
2016 0.0454
????? ???????? ????????? 2014 0.0623 0.1630 There is no earnings quality
2015 0.2788
2016 0.1481
???? ???????? ????????? 2014 0.0849 0.0548 There is earnings quality
2015 0.0417
2016 0.0379
??? ???? ??????? 2014 0.0662 0.0649 There is earnings quality
2015 0.0501
2016 0.0784
???????? ?????? ?????? ??????? ????????? 2014 0.0675 0.0819 There is earnings quality
2015 0.0939
2016 0.0844
??????? ???????? 2014 0.1039 0.0486 There is earnings quality
2015 0.0326
2016 0.0092
??????????? ?????? 2014 0.0466 0.0950 There is earnings quality
2015 0.1710
2016 0.0673
?????? ???????? ???????? ??????? ???????? 2014 0.0840 0.3758 There is no earnings quality
2015 0.4150
2016 0.6285
???????? ??????? ?????? ???????? 2014 0.1268 0.1750 There is no earnings quality
2015 0.2598
2016 0.1385
??????? ??????? 2014 0.1185 0.1142 There is earnings quality
2015 0.0741
2016 0.1501
??????? ??????? ??????? ??????? ?????????? 2014 0.3238 0.1125 There is earnings quality
2015 0.0046
2016 0.0092
??? ?????? 2014 0.2339 0.1311 There is earnings quality
2015 0.0072
2016 0.1521
????? ?????? ???????? ???????? 2014 0.1896 0.2029 There is no earnings quality
2015 0.2697
2016 0.1494
?????? ???????? ???????? 2014 0.0360 0.0807 There is earnings quality
2015 0.0989
2016 0.1071
?????? ??????? 2014 0.1604 0.4075 There is no earnings quality
2015 0.4902
2016 0.5717
??????? ?????? ?????????/???? 2014 0.0948 0.0707 There is earnings quality
2015 0.1138
2016 0.0036
??????? ?????? ????? 2014 0.7618 0.3364 There is no earnings quality
2015 0.2006
2016 0.0466
????? ???????? ???????? 2014 0.1315 0.0803 There is earnings quality
2015 0.0316
2016 0.0777
????? ??????? ???????? 2014 0.1123 0.0906 There is earnings quality
2015 0.0505
2016 0.1090
??????? ??????? 2014 0.0799 0.0726 There is earnings quality
2015 0.0969
2016 0.0411
???? ?????? 2014 0.2223 0.1809 There is no earnings quality
2015 0.2315
2016 0.0888
??????? ??????? ????????? 2014 0.0046 0.1059 There is earnings quality
2015 0.1802
2016 0.1330
??????????? ????????? ????????? 2014 0.3090 0.3982 There is no earnings quality
2015 0.5878
2016 0.2979
???? ?????????? 2014 0.0815 0.0587 There is earnings quality
2015 0.0531
2016 0.0415
??????? ?????? ????? ??????? ?????????? ?? ????? ?????? 2014 0.0421 0.0143 There is earnings quality
2015 0.0001
2016 0.0007
Service Sector Year TACC Average of TACC Result
??????? ????? ???????? 2014 0.0357 0.0596 There is earnings quality
2015 0.0565
2016 0.0867
?????? ??????? ??????? ??????? 2014 0.0250 0.0335 There is earnings quality
2015 0.0402
2016 0.0353
????????? ???????? ???????? 2014 0.1949 0.1484 There is no earnings quality
2015 0.1140
2016 0.1365
???????? ??????? ???????????? 2014 0.4387 0.2772 There is no earnings quality
2015 0.3157
2016 0.0772
????? ??????? ?????????? 2014 0.1710 0.3365 There is no earnings quality
2015 0.7702
2016 0.0684
?????? ????? ??????? 2014 0.0171 0.0221 There is earnings quality
2015 0.0056
2016 0.0438
?????? ???????????? 2014 0.1195 0.1725 There is no earnings quality
2015 0.0173
2016 0.3808
???? ?????? 2014 0.1571 0.1968 There is no earnings quality
2015 0.2541
2016 0.1791
????? ??????? ????????? ???????? 2014 0.9061 0.3520 There is no earnings quality
2015 0.0774
2016 0.0724
??????? ??????? ?????????? 2014 0.3245 0.1808 There is no earnings quality
2015 0.1551
2016 0.0626

Appendix (3): Components of the equation of earnings quality data

Financial Sector ????? WC ?WC NCO ?NCO FIN ?FIN TACC
????? ??????? ??????? 2014 -1,712,377,352 46,926,927 -52,151,475 23,564,743 11,107,725 7545000 0.030
2015 -1,743,980,375 31,603,023 -36,841,376 15,310,099 9,107,715 -2000010 0.016
2016 -1,641,815,082 102,165,294 -36,749,468 91,908 14,166,627 5058912 0.039
????? ??????? ??????? 2014 -845,223,646 82,936,263 4,077,977 1,893,921 0 0 0.073
2015 -1,017,408,126 172,184,479 -6,982,894 11,060,871 7,090,000 7090000 0.128
2016 -826,013,089 191,395,037 462,545 7,445,439 0 -7090000 0.152
??? ??????? ??????? ???????? 2014 -5,184,742,343 332,810,200 -92,093,928 27,087,352 23,147,561 1335561 0.048
2015 -5,064,483,118 120,259,225 -69,356,858 22,737,070 24,624,589 1477028 0.018
2016 -4,928,342,373 136,140,745 -83,720,006 14,363,148 22,795,894 -1828695 0.019
??? ????????? ?????? ??????? 2014 -1,225,412,484 468,425,103 23,095,804 5,673,208 19,925,164 7314075 0.275
2015 -1,205,050,212 20,362,272 32,473,422 9,377,618 22,021,783 2096619 0.018
2016 -1,225,886,632 20,836,419 37,013,477 4,540,055 27,312,114 5290331 0.017
??? ??????? 2014 -1,589,030,421 265,037,686 1,105,428 17,235,987 0 0 0.125
2015 -1,555,272,054 33,758,367 3,848,529 2,743,101 0 0 0.015
2016 -1,644,096,952 88,824,898 7,648,623 3,800,094 0 0 0.036
??? ??????? ??????? ????????/?????? 2014 -759,551,507 130,389,736 -20,614,949 12,277,752 9,910,585 9910585 0.139
2015 -651,069,672 108,481,835 -18,671,653 1,943,296 17,725,000 7814415 0.115
2016 -687,180,647 36,110,975 -15,722,934 2,948,719 28,360,000 10635000 0.045
????? ?????????? 2014 -493,520,569 348,814 3,603,828 3,392,022 1,378,847 -39153 0.005
2015 -461,168,059 32,352,511 5,528,240 1,924,412 5,000,000 3621153 0.045
2016 -478,929,234 17,761,175 6,474,343 946,103 5,500,000 500000 0.020
??? ????? ??????? 2014 -1,249,038,162 177,552,344 -48,734,958 6,325,682 12,168,358 6008358 0.092
2015 -1,125,336,684 123,701,478 -40,076,398 8,658,560 15,158,414 2990056 0.068
2016 -1,111,008,362 14,328,323 -52,919,294 12,842,896 0 -15158414 0.006
??? ?????? ????? – ?????? 2014 -596,625,773 210,384,715 6,548,563 4,654,523 16,109,873 -11705783 0.235
2015 -820,781,490 224,155,717 6,596,215 47,652 17,459,133 1349260 0.186
2016 -888,169,028 67,387,538 3,508,099 3,088,116 15,215,196 -2243937 0.052
??? ??????? ???? 2014 -1,685,276,031 169,787,732 -51,376,599 7,995,884 107,399,984 -30800016 0.062
2015 -1,646,201,158 39,074,872 -47,342,206 4,034,393 113,199,984 5800000 0.019
2016 -1,563,186,610 83,014,548 -31,492,579 15,849,627 73,543,068 -39656916 0.024
??? ?????? 2014 -1,476,482,867 73,251,677 -17,545,274 4,480,628 0 -8508000 0.032
2015 -1,513,402,770 36,919,903 -13,333,363 4,211,911 155,000,000 155000000 0.089
2016 -1,439,126,877 74,275,893 -14,308,431 975,068 25,000,000 -130000000 0.023
????? ?????? ??????? 2014 -1,585,471,812 124,936,101 11,200,017 509,726,288 0 -26661835 0.261
2015 -1,607,773,282 22,301,471 16,251,073 5,051,056 14,647,510 14647510 0.017
2016 -1,827,465,628 219,692,346 3,030,022 13,221,051 10,037,793 -4609717 0.081
????? ?????? 2014 -17,620,492,450 1062937250 -888,366,000 503,895,000 283,639,000 41696000 0.062
2015 -16,075,218,600 1545273850 -1,065,462,000 177,096,000 170,397,000 -113242000 0.062
2016 -14,917,405,800 1157812800 -666,149,000 399,313,000 50,450,000 -119947000 0.059
Industrial Sector ????? WC ?WC NCO ?NCO FIN ?FIN TACC
???????? ???????? ???????? / ??????? 2014 1,719,115 1,852,022 12,496,473 353,537 -255 16131 0.079
2015 4,741,339 3,022,224 12,904,719 408,246 -3,105 -2850 0.132
2016 2,610,953 2,130,386 13,263,174 358,455 318 3423 0.093
???????? ??????? ????????? 2014 -1,080,570 518,820 -3,363,833 520,448 2,506,814 -192848 0.111
2015 -847,593 232,977 -2,435,745 928,088 1,789,793 -717021 0.065
2016 -502,749 344,844 -3,296,135 860,390 3,522,098 1732305 0.633
???????? ????????? ???????? 2014 2,178,191 730,774 1,605,731 113,550 -974,374 -42878 0.104
2015 1,095,515 1,082,676 1,449,900 155,831 19,741 994115 0.341
2016 1,722,615 627,100 1,579,217 129,317 -426,828 -446569 0.045
??????? ?????? ???????? 2014 3,417,731 203,169 11,264,504 3,498,449 -2,022,179 -1446270 0.140
2015 1,954,525 1,463,206 15,120,773 3,856,269 -3,025,232 -1003053 0.227
2016 2,195,995 241,470 16,571,475 1,450,702 -5,632,977 -2607745 0.044
??????? ???????? ???????? 2014 5,576,389 1,962,335 8,604,775 2,121,269 3,120,679 -16065 0.183
2015 2,852,306 2,724,083 8,531,974 72,801 3,120,679 0 0.145
2016 2,542,839 309,467 8,499,969 32,005 3,303,859 183180 0.028
????????? ???????? ???????? 2014 7,967,779 540,084 3,165,917 244,423 -7,917,445 409144 0.086
2015 6,352,533 1,615,246 3,096,326 69,591 -6,925,536 991909 0.223
2016 -4,350 6,356,883 -5,637,331 8,733,657 5,637,331 12562867 1.087
??????? ?????? ???????? ???????? ???????? 2014 11,142,511 5,719,972 5,437,871 106,405 810,447 5253537 0.374
2015 12,018,955 876,444 4,941,988 495,883 1,083,590 273143 0.058
2016 14,050,283 2,031,328 4,423,626 518,362 1,424,544 340954 0.100
??????? ?????? ??????? ???????? ?????????? 2014 8,404,340 291,389 12,612,111 637,331 42,958 0 0.027
2015 9,356,358 952,018 11,996,954 615,157 42,958 0 0.046
2016 8,073,843 1,282,515 11,184,628 812,326 42,958 0 0.075
??????? ???????? ?????????? 2014 3,599,115 163,649 3,004,641 44,567 -909,985 -262876 0.005
2015 2,845,074 754,041 2,861,716 142,925 -788,430 121555 0.116
2016 3,374,928 529,854 2,734,280 127,436 -163,420 625010 0.172
????? ???????? ??????? 2014 23,312,833 399,651 18,125,363 1,671,608 -10,459,703 781438 0.054
2015 24,576,538 1,263,705 16,586,680 1,538,683 -9,115,494 1344209 0.080
2016 19,203,209 5,373,329 15,515,098 1,071,582 -4,256,597 4858897 0.240
???????? ?????? ???????? 2014 5,873,177 896,831 1,234,452 137,004 -2,041,200 -1189447 0.019
2015 5,042,104 831,073 884,071 350,381 -1,370,107 671093 0.251
2016 4,642,969 399,135 776,950 107,121 -1,029,481 340626 0.120
???????? ???????? ??????? / ?????? 2014 3,365,149 249,694 7,960,785 727,131 -120,687 756558 0.112
2015 513,842 2,851,307 7,305,160 655,625 0 120687 0.261
2016 1,036,941 523,099 6,597,937 707,223 0 0 0.105
??????? ?????? ???????? ???????? 2014 8,855,632 55,096 3,833,359 376,705 814,284 -482058 0.002
2015 8,352,601 503,031 3,603,553 229,806 852,044 37760 0.042
2016 7,839,864 512,737 3,302,997 300,556 784,215 -67829 0.045
????? ???????? ????????? 2014 1,990,212 108,933 959,621 187,125 1,839,922 305491 0.062
2015 1,630,098 360,114 5,200,679 4,241,058 27,687 -1812235 0.279
2016 329,489 1,300,609 5,049,162 151,517 268,712 241025 0.148
???? ???????? ????????? 2014 138,624 480,476 8,633,518 1,103,682 621,562 -213502 0.085
2015 -255,066 393,690 8,160,834 472,684 413,741 -207821 0.042
2016 103,255 358,321 7,768,462 392,372 260,411 -153330 0.038
??? ???? ??????? 2014 447,712 80,466 512,216 27,740 -34,979 -32544 0.066
2015 372,708 75,004 516,964 4,748 -63,403 -28424 0.050
2016 308,333 64,375 521,519 4,555 -56,331 7072 0.078
???????? ?????? ?????? ??????? ????????? 2014 6,221,152 513,350 43,667,772 4,240,053 -5,658,433 -938249 0.068
2015 9,441,831 3,220,679 46,796,024 3,128,252 -6,118,525 -460092 0.094
2016 8,620,017 821,814 52,486,799 5,690,775 -6,859,336 -740811 0.084
??????? ???????? 2014 2,172,186 437,813 3,703,947 74,646 574,835 698460 0.104
2015 2,618,629 446,443 3,464,289 239,658 267,489 -307346 0.033
2016 3,099,354 480,725 3,619,154 154,865 -245,032 -512521 0.009
??????????? ?????? 2014 4,427,635 897,068 -3,871,129 142,394 10,531,054 171390 0.047
2015 1,521,214 2,906,421 -4,509,007 637,878 11,579,257 1048203 0.171
2016 2,022,600 501,386 -5,170,103 661,096 12,257,422 678165 0.067
?????? ???????? ???????? ??????? ???????? 2014 439,326 410,419 2,035,916 74,928 -1,245,317 -880537 0.084
2015 -390,557 829,883 1,919,143 116,773 -908,633 336684 0.415
2016 -1,232,943 842,386 1,833,844 85,299 -337,620 571013 0.628
???????? ??????? ?????? ???????? 2014 8,077,138 1,404,852 1,319,369 156,041 23,148 -99483 0.127
2015 5,641,731 2,435,407 952,794 366,575 116,574 93426 0.260
2016 7,133,610 1,491,879 807,888 144,906 108,862 -7712 0.138
??????? ??????? 2014 35,883,723 9,117,421 41,495,521 1,664,220 0 0 0.118
2015 39,374,501 3,490,778 38,652,811 2,842,710 0 0 0.074
2016 31,168,695 8,205,806 35,562,635 3,090,176 0 0 0.150
??????? ??????? ??????? ??????? ?????????? 2014 -129,409 7,826 -35,353,127 5,839,792 35,605,755 5839490 0.324
2015 -154,743 25,334 -32,523,782 2,829,345 32,907,826 -2697929 0.005
2016 -164,700 9,957 -27,483,723 5,040,059 27,601,676 -5306150 0.009
??? ?????? 2014 5,513,481 3,203,724 3,815,067 461,217 -2,808,031 -362471 0.234
2015 5,858,491 345,010 3,311,658 503,409 -3,560,908 -752877 0.007
2016 7,169,676 1,311,185 2,936,235 375,423 -3,117,835 443073 0.152
????? ?????? ???????? ???????? 2014 3,862,820 342,969 520,869 71,778 283,796 769671 0.190
2015 2,764,185 1,098,635 295,116 225,753 680,357 396561 0.270
2016 2,005,718 758,467 138,393 156,723 794,519 114162 0.149
?????? ???????? ???????? 2014 9,443,656 2,074,480 19,346,822 69,625 -1,936,662 -651890 0.036
2015 9,759,658 316,002 24,292,783 4,945,961 -2,313,207 -376545 0.099
2016 14,145,614 4,385,956 27,957,224 3,664,441 -4,018,973 -1705766 0.107
?????? ??????? 2014 638,216 269,275 500,678 113,552 283,940 -80161 0.160
2015 354,580 283,636 248,619 252,059 334,618 50678 0.490
2016 -80,890 435,470 213,873 34,746 90,126 -244492 0.572
??????? ?????? ?????????/???? 2014 4,068,566 825,189 5,576,206 926,756 837,536 -16866 0.095
2015 5,987,915 1,919,349 5,051,869 524,337 250,684 -586852 0.114
2016 6,516,470 528,555 5,121,889 70,020 -290,434 -541118 0.004
??????? ?????? ????? 2014 -1,717,112 1,463,110 3,274,633 2,194,399 356,750 510764 0.762
2015 -1,814,321 97,209 4,600,140 1,325,507 302,033 -54717 0.201
2016 -1,433,639 380,682 5,056,575 456,435 50 -301983 0.047
????? ???????? ???????? 2014 199,188,000 20,124,000 63,000,000 147,436,000 109,852,000 -8236000 0.132
2015 213,483,000 14,295,000 56,057,000 6,943,000 125,674,000 15822000 0.032
2016 186,066,000 27,417,000 3,405,000 52,652,000 133,880,000 8206000 0.078
????? ??????? ???????? 2014 -12,644,469 4,429,843 109,089,770 9,140,494 -14,359,683 7804128 0.112
2015 -7,145,531 5,498,938 100,755,810 8,333,960 -18,340,763 -3981080 0.051
2016 4,354,982 11,500,513 93,400,749 7,355,061 -16,757,232 1583531 0.109
??????? ??????? 2014 115,834,000 10,243,000 226,068,000 46,864,000 105,427,000 18712000 0.080
2015 96,236,000 19,598,000 176,126,000 49,942,000 134,345,000 28918000 0.097
2016 119,451,000 23,215,000 167,650,000 8,476,000 140,482,000 6137000 0.041
???? ?????? 2014 24,174,001 8,735,965 41,045,842 1,194,247 -26,208,285 6635111 0.222
2015 16,959,766 7,214,235 40,516,327 529,515 -18,239,035 7969250 0.232
2016 14,195,522 2,764,244 40,263,929 252,398 -15,320,826 2918209 0.089
??????? ??????? ????????? 2014 8,541,109 905,722 5,752,248 34,863 -4,125,540 -864437 0.005
2015 7,611,661 929,448 5,552,064 200,184 -2,606,174 1519366 0.180
2016 8,022,556 410,895 4,903,969 648,095 -1,778,494 827680 0.133
??????????? ????????? ????????? 2014 11,709,669 7,570,434 -20,971,646 1,684,239 9,543,045 2796087 0.309
2015 -437,938 12,147,607 -17,479,147 3,492,499 11,597,743 2054698 0.588
2016 4,568,802 5,006,740 -14,317,370 3,161,777 11,318,318 -279425 0.298
???? ?????????? 2014 2,089,464 267,091 974,410 187,684 925,861 1758 0.081
2015 2,468,773 379,309 830,552 143,858 702,016 -223845 0.053
2016 2,316,496 152,277 628,533 202,019 575,183 -126833 0.042
??????? ?????? ????? ??????? ?????????? ?? ????? ?????? 2014 15,514 65,630 -44,992 9,778 225,584 -4141 0.042
2015 19,292 3,778 -53,622 8,630 212,939 -12645 0.000
2016 17,978 1,314 -53,520 102 210,429 -2510 0.001
Service Sector ????? WC ?WC NCO ?NCO FIN ?FIN TACC
??????? ????? ???????? 2014 7,381,210 1,591,805 4,191,692 131,842 1,550 -750 0.036
2015 6,770,958 610,252 6,480,605 2,288,913 1,550 0 0.056
2016 7,968,160 1,197,202 10,218,694 3,738,089 1,550 0 0.087
?????? ??????? ??????? ??????? 2014 2,723,471 89,428 -305,338 63,469 850,192 62711 0.025
2015 3,053,418 329,947 -241,983 63,355 815,597 -34595 0.040
2016 3,317,226 263,808 -171,297 70,686 784,783 -30814 0.035
????????? ???????? ???????? 2014 -537,811 2,789,867 -38,280,970 3,053,931 28,446,636 718532 0.195
2015 -1,966,840 1,429,029 -41,479,478 3,198,508 28,298,787 -147849 0.114
2016 -3,631,979 1,665,139 -39,230,295 2,249,183 29,987,399 1688612 0.136
???????? ??????? ???????????? 2014 1,025,690 1,108,301 545,326 24,972 0 0 0.439
2015 537,581 488,109 523,176 22,150 0 0 0.316
2016 705,538 167,957 526,432 3,256 0 0 0.077
????? ??????? ?????????? 2014 -4,917,986 3,257,310 -32,773,672 582,482 26,747,487 2486731 0.171
2015 31,210,140 36,128,126 -7,363,171 25,410,501 -3,977,618 -30725105 0.770
2016 34,517,661 3,307,521 -7,393,695 30,524 -4,538,229 -560611 0.068
?????? ????? ??????? 2014 5,518,960 101,563 -26,992,423 382,234 25,794,683 422019 0.017
2015 5,647,971 129,011 -267,122 26,725,301 -1,364,431 -27159114 0.006
2016 10,155,096 4,507,125 -270,689 3,567 -3,203,768 -1839337 0.044
?????? ???????????? 2014 -824,968 718,561 -15,778,489 17,766,385 10,922,768 9431675 0.120
2015 -188,280 636,688 -15,828,484 49,995 10,589,751 -333017 0.017
2016 531,851 720,131 -11,416,960 4,411,524 11,414,751 825000 0.381
???? ?????? 2014 -53,004,880 14,772,465 157,173,480 26,229,113 19,000 1294500 0.157
2015 -92,887,086 39,882,206 194,156,786 36,983,306 26,250 7250 0.254
2016 -122,190,682 29,303,596 228,610,502 34,453,716 134,000 107750 0.179
????? ??????? ????????? ???????? 2014 -1,791,712 782,065 -31,588,443 13,932,956 31,613,155 13930105 0.906
2015 -579,728 1,211,984 -32,214,100 625,657 32,274,895 661740 0.077
2016 20,579,112 21,158,840 20,661,834 52,875,934 -28,277,288 -60552183 0.072
??????? ??????? ?????????? 2014 -7,910,038 5,213,365 34,740,047 9,266,977 63,775 14383 0.325
2015 -5,134,633 2,775,405 39,605,871 4,865,824 61,522 -2253 0.155
2016 -5,627,514 492,881 42,314,607 2,708,736 66,894 5372 0.063

?
Appendix (4): Outputs of statistical analysis

Explore
Tests of Normality
Kolmogorov-Smirnova Shapiro-Wilk
Statistic df Sig. Statistic df Sig.
Standardized Residual .067 161 .072 .974 161 .004
a. Lilliefors Significance Correction
Standardized Residual

Regression
Variables Entered/Removeda
Model Variables Entered Variables Removed Method
1 number of meetings held by the audit committee , number of the board of directors members , number of meetings held by the board of directors , number of the audit committee membersb . Enter
a. Dependent Variable: earnings quality
b. All requested variables entered.
Model Summary
Model R R Square Adjusted R Square Std. Error of the Estimate
1 .235a .055 .034 .16246
a. Predictors: (Constant), number of meetings held by the audit committee , number of the board of directors members , number of meetings held by the board of directors , number of the audit committee members
ANOVAa
Model Sum of Squares df Mean Square F Sig.
1 Regression .270 4 .067 2.553 .041b
Residual 4.619 175 .026
Total 4.889 179
a. Dependent Variable: earnings quality

Model Unstandardized Coefficients Standardized Coefficients t Sig. Collinearity Statistics
B Std. Error Beta Tolerance VIF
1 (Constant) .249 .057 4.383 .000
number of the board of directors members -.010 .005 -.168 -2.013 .046 .775 1.290
number of meetings held by the board of directors .001 .006 .020 .233 .816 .716 1.396
number of the audit committee members -.006 .010 -.054 -.597 .552 .658 1.520
number of meetings held by the audit committee -.004 .006 -.070 -.649 .517 .462 2.165
a. Dependent Variable: earnings quality

Correlations

Correlations
earnings quality number of the board of directors members number of meetings held by the board of directors number of the audit committee members number of meetings held by the audit committee
earnings quality Pearson Correlation 1 -.216** -.043 -.151* -.165*
Sig. (2-tailed) .004 .568 .043 .027
N 180 180 180 180 180
number of the board of directors members Pearson Correlation -.216** 1 .128 .362** .445**
Sig. (2-tailed) .004 .086 .000 .000
N 180 180 180 180 180
number of meetings held by the board of directors Pearson Correlation -.043 .128 1 .123 .496**
Sig. (2-tailed) .568 .086 .099 .000
N 180 180 180 180 180
number of the audit committee members Pearson Correlation -.151* .362** .123 1 .548**
Sig. (2-tailed) .043 .000 .099 .000
N 180 180 180 180 180
number of meetings held by the audit committee Pearson Correlation -.165* .445** .496** .548** 1
Sig. (2-tailed) .027 .000 .000 .000
N 180 180 180 180 180
**. Correlation is significant at the 0.01 level (2-tailed).
*. Correlation is significant at the 0.05 level (2-tailed).

Descriptives

Descriptive Statistics
N Minimum Maximum Mean Std. Deviation Skewness Kurtosis
Statistic Statistic Statistic Statistic Statistic Statistic Std. Error Statistic Std. Error
number of the board of directors members 180 4.00 19.00 8.6333 2.73994 .539 .181 -.029 .360
number of meetings held by the board of directors 180 .00 19.00 7.1000 2.45427 .334 .181 2.425 .360
number of the audit committee members 180 .00 7.00 2.8722 1.48373 -.193 .181 1.256 .360
number of meetings held by the audit committee 180 .00 20.00 4.3167 3.14896 .744 .181 1.461 .360
Valid N (listwise) 180

Descriptive Statistics
N Minimum Maximum Mean Std. Deviation
v2014 60 .00 .91 .1445 .16222
v2015 60 .00 .77 .1404 .14932
v2016 60 .00 1.09 .1279 .18465
allyears 180 .00 1.09 .1376 .16526
Valid N (listwise) 60
Descriptive Statistics
N Skewness Kurtosis
Statistic Statistic Std. Error Statistic Std. Error
allyears 180 1.081 .181 3.293 .360
Valid N (listwise) 180

Post Author: admin