A Plan Structure and Strategy Finance Essay

This chapter presents the procedures to investigate the relationship between the different variables in this study. These procedures comprised of the research design, research strategy, research population, sampling procedures and the development of an instrument for this research.

3.1 - Research Design.

A research in general can be considered as a systematic process to find answers for certain issues. A Research design is perceived as a road map for the researchers (Davis, 1996). A research design is identified as; A plan, structure and strategy of investigation so conceived as to obtain answers to research questions or problems' (Kerlinger, 1986). In other words, a research design is a comprehensive plan on how a research is to be accomplished, how the variables are operationalized, how the data from the proposed sample are collected and subsequently, how the results are derived by analyzing the data (Thyer, 1993).

3.2 - Nature of Research.

Generally, research can be classified into three main categories according to the nature and purpose of the research.

3.2.1 - Exploratory Research.

This type of research is being conducted when there is insufficient knowledge about certain issues to investigate new concepts or phenomena (Sekaran, 2003). In exploratory research, qualitative methods are more often being used than quantitative methods (Hair, Money, Page & Samouel, 2007). The researcher usually uses one of four approaches to solicit information namely secondary data analysis, pilot studies, case studies and experience survey to obtain initial ideas about a situation (Zikmund, 2003).

3.2.2 - Descriptive Research.

This design aims to describe the major attributes that answer questions about current situation (Sekaran, 2003). The descriptive research can be classified based on the time horizon to collect the data namely cross-sectional or longitudinal. As the name indicates, the cross-sectional approach is to collect data only once or at a certain point of time, while longitudinal approach is to describe the situation whereby data is being gathered over time (Hair et al., 2007). Data for descriptive research are collected by means of interview, observation, or questionnaire (Gay & Airasian, 2003). The statistical analyses used for the descriptive research are computing the central tendency, variances and correlations (Leedy&Ormrod, 2001).

3.2.3 - Hypotheses Testing.

This design employed a testable scheme to examine the relationships between the different variables (Saunders, Lewis & Thornhill, 2007). The main objective of this research design is to explain the variance on the dependent variable or to predict the outcome of this relationship between independent and dependent variables (Sekaran,2003). In this study, the main objective is to investigate the relationship between variables, and can be considered as a descriptive and hypotheses testing study in nature. Using the descriptive statistics to determine the main attributes of the population can provide a better understanding on the nature of the population. Hypotheses testing are also used to investigate the relationships between the independent and dependent variables to determine any differences between them.

3.3 - Choice of Research Design.

Zikmund (2003) claimed that the objectives, availability of information and costs for conducting the research are the main factors affecting the choice of research design. The main objective of this study is to investigate the relationship between accounting information and corporate governance dimensions and investment decision. the study had adopted the survey strategy. The data obtained from survey is used to examine the relationships between the dependent and independent variables (Davis, 1996). furthermore, when the respondents are bank managers, accounting staff department and highly educated, survey approach is the most suitable technique (Cooper & Schindler, 2003). This further supports the reason for choosing survey approach for this study. Other advantage of survey strategy is the large amount of data that can be gathered from the respondents and the results can be generalized to the population at large.

3.4 - Survey Strategies.

This study has adopted the survey strategy. However, several approaches can be used to collect data for the survey.

3.4.1 - Personal Interview.

Personal interview or face-to-face interview is a two-way conversation between the interviewers and the respondents in order to gather information (Cooper & Schindler, 2003). This technique can be carried out in several ways such as structured interview, which is conducted as the interviewer knows the questions to be asked. Alternately, unstructured interviews can be conducted whereby the interviewer has no ready questions (Sekaran, 2003). Personal interview has its advantages and disadvantages. The advantages are the researcher has the opportunity to get direct feedback from the respondents, or provides encouragement to get information from respondents (Zikmund, 2003).

3.4.2 Telephone Interview.

Telephone interview is the gathering of data from the respondents by telephone (Zikmund, 2003). The telephone interview technique can overcome some of the disadvantages encountered by face-to-face interview such as geographical barriers. This technique enables the interviewer to reach respondents efficiently in terms of time and cost (Saunders et al. 2007). Furthermore, combining telephone services with computer, source as Computer-Assisted Telephone Interviewing (CATI), allows the interviewer to record the respondent's answers directly to the computer that can save time and money (Zikmund, 2003). The main advantage of telephone interview method as it allows respondents to answer private or personal questions because there is no face-to-face contact. The speed of data which can be gathered from this technique is another advantage as it can reach the respondents who are located at a wide geographical locations but have access to telecommunication network such as mobile or fixed line phone services (Zikmund, 2003). On the other hand, the disadvantage of this technique is that respondents can end the interview without any caution (Sekaran, 2003). Other disadvantage is the lack of visual interaction because the interviewer cannot see the respondent's expression while telephone interviews are being conducted (Zikmund, 2003). In addition, the duration of interview can be another disadvantage because the respondents may feel bored when the duration of the interview is too long. The disadvantages are the high cost to conduct interviews, difficulties to reach the respondents who are scattered in a wide geographic area, unwillingness of some respondents to express their attitudes directly to others and, the need to provide training to interviewers (Cooper & Schindler, 2003).

3.4.3 - Internet Survey.

Internet survey is conducted by using the internet network. The main advantage of this approach is the capability to reach respondents in a wide geographic area. Furthermore, the direct recording for the respondent's answers into the system will save time and money (Saunders et al. 2007). In addition to time and money saving, other advantages of internet survey are its interactive, attractive web site design and confidentiality of the respondents can enhance the response rate (Cooper & Schindler, 2003). On the other hand, the disadvantage of internet survey is the high costs of developing questionnaires on the web site that required the service of skilled programmers or IT analysts (Cooper & Schindler, 2003).

3.4.4 - Mail Survey.

Mail survey is conducted by sending the questionnaire to the respondents using mail service which is flexible and cost effective (Davis, 1996). Other advantage of mail survey is respondent's confidentiality which can be maintained. The disadvantages of this technique are the low response rate; the researcher also cannot see the respondents which makes it difficult for him to explain some complex issues. It is also difficult to include long or complex questions in the questionnaire (Cooper & Schindler, 2003; Sekaran, 2003).

3.5 - Choice of Survey Method.

Zikmund (2003) reveled that there is no best method as each one has its own strengths and weaknesses. While Cooper & Schindler (2003) believed that it is not difficult to choose the best method if the researcher can evaluate between the advantages and disadvantages for each method. However, Kumar (2005) suggested that choice of survey method depends on three major issues namely the nature of the investigation, the geographical distribution of the respondents and the respondent's characteristics. In terms of geographical distribution, the respondents for this research are located in a wide geographical area, hence, mail survey is deemed suitable. Furthermore, the targeted population consists of bank managers, accounting stuff department, who are well educated. Thus, it is assumed that they can understand and respond to the questionnaire. Based on the criteria suggested by Kumar (2005), mail survey is perceived as a better choice. However, the main disadvantage of mail survey is the low response rate. Therefore, there is a need to take steps to increase the response rate in such approach. One of these steps is to develop a good questionnaire design which can help to enhance the response rate (Sekaran, 2003).

3.6 - Population of Study.

According to Davis (1996) the population is the study unit or element which information is gathered to achieve the study purpose. The population is a collection of elements which the study is interested to examine. The target population for this study is managers, accounting staff department, from the banking sector in Libya. The unit of analysis for this study consists of bank managers and accounting staff department. The banking sector plays a pivotal role in the economic growth and contributes significantly to other economic sectors in Libya (CBL, 2005). Furthermore, the banking and financial service sector in Libya is one of the most willing adopter of new IT which are adopted to produce high quality information (Gabbard & Park, 1996).

3.7 - Sampling Frame.

A sampling frame is a comprehensive list of elements representing the population which a sample can be illustrated. According to Hair, Samouel, Babin & Money (2003), a sampling frame should meet the following criteria: 1- Frame should comprise of a list of elements representative of the study population. 2- Frame should be comprehensive. 3- Frame should be up-to-date and. 4- Frame should not include any duplicated elements. In this study, the target population is the bank managers and accounting staff who work at banks branches from 15 anchor banks in Libya. For the purpose of this study a decision was taken to include all the bank branches in the sample which means that the population is also the sample. The reasons for taking such decision are: 1- The population for this study is not too large. Gay & Diehl (1992) revealed that large sample size will be better for any study especially for generalization purposes. 2- The appropriate sample size for most of research ought to be larger than 30 and less than 500 (Sekaran, 2003). 3- The sample size should be sufficiently large for the purpose of conducting a specific data analysis such as multivariate analysis.

3.8 - Data Collection Procedure.

The main focus of this study is to examine the relationship between accounting information and corporate governance and investment decision in general. Therefore, unit of analysis for this study comprised of the bank managers, accounting staff department in Libya as they were the investment decision between these banks and investors. To achieve the objectives of this research, a questionnaire was developed to collect data from the respondents to provide answers to the research questions. Mail survey strategy was employed for this study. As stringent regulations prohibit Libyan bank employees to respond to any questionnaires unless with the consent of the top management. Therefore, a formal request was sent to seek permission from them. Permission was obtained from these banks and the questionnaires were sent to the public relation office at the head office which subsequently redirected them to the respective branches.

3.9 - Questionnaire Design.

The development of the study instrument followed the general guides proposed by Guy & Diehl (1992) whereby the questionnaire design must be attractive, concise and easy to answer. Zikmund (2003) stressed that an instrument should meet two requirements. Firstly, relevancy as the questionnaire used to collect the information to meet the research objectives. Secondly, accuracy as the questionnaire used to collect information characterized by a high degree of reliability and validity. The following guidelines proposed by Hair et al., (2003) were taken in account while phrasing the questionnaire items. 1- Using simple words and common language. 2- Using brief and direct questions. 3- Avoiding vagueness and be clear about what you asking for 4- Avoiding leading questions to desirable answers and; 5- Avoiding double-barreled questions consist of two or more concerns. For the purpose of this study the questionnaire was structured into five main sections following the funnel approach. The simple and general questions are asked before the difficult ones. For the purpose of this study the questionnaire was structured into five main sections following the funnel approach. The simple and general questions are asked before the difficult ones. 1- Section A: This section solicits information about the respondent's demographic variables which include gender, age, level of education, work experience. 2- Section B: The questions in this section are aimed at identifying the dimensions of accounting information (reliability) and what is the impact in investment decisions. 3- Section C: This section concentrates on the information dimensions needed in the intelligence, (relevance) and impact in investment decision. 4- Section D: This section for study the relation between corporate governance (disclosure) and impact in investment decision. 5- Section E: This section was developed and explain corporate governance (transparency) impact in investment decision.

3.10 - Measurement and Operationalization of Variables.

This study aimed to examine the relationship between information quality and decision making as the main two concepts. The literature review discussed the two concepts with regard to their definitions and dimensions. For information, as revealed in the literature is a subjective concept and it was difficult to find agreement between the authors about its dimensions. To overcome this problem, a literature review was conducted to determine the most important dimensions used in the previous literature. The literature revealed that Reliability, Relevance, Disclosure, and Transparency were suitable dimensions to operationalize the concept of information needed. Consequently, information was represented by these four dimensions. For scaling purpose, a five-point scale ranging from 1 "Strongly Disagree" to 5 "Strongly Agree" was employed. Table 3.1 presents the operational definitions for each dimension of information quality with its measurement items in addition to the sources of these items from were adopted.

Information

Operational Definition and Questions

Resource

Reliability The extent to which information is correct and reliable. Correct information. Reliable information. Information is completely error-free. Information exactly matches the actual values. Winter-man,1998 Najjar 2002 Miller,2005 Slone,2006 Ahmed,2007 Relevance The extent to which information is applicable and helpful for the task at hand. Useful to our work Relevant to our work Appropriate for our work Applicable to our work Wang and Strong (1996) Lee et al. (2002) Najjar (2002) Kahn et al. (2002) Slone (2006) Disclosure The extent to which information is correct and right disclosure. Easily Disclosure information. Easily Disclosure information. Easily obtainable information. Quickly Disclosure information when needed. Nabil Ibrahim,2012 Laika,2007 Shawawra,2008 Wen,2007 Almjhli,2009 Transparency The extent to which information is correct and right Transparency. Easily Transparency information. Easily Transparency information. Easily obtainable information. Quickly Transparency information when needed. Aljadi,2010 Almjhli,2009 Hermalin&Weisbach2007 Drabek,2001 Ameri & Douai,2006

Table 3.1: Information and Dimensions Measurements.

The first dimension of accounting information is reliability . This dimension is operationalized to determine the extent of information as correct and reliable (Winter-man,1998) reliability was operationalized using three items which have been validated by other researchers (e.g. Najjar 2002, Miller,2005, Slone,2006, Ahmed,2007). The second dimension of accounting information is relevance .This dimension is operationalized to determine the extent of information as correct and reliable (Wang and Strong, 1996). reliability was operationalized using three items which have been validated by other researchers (e.g. Bovee, 2004; Kahn et al., 2002; Lee et al., 2002; Najjar, 2002; Slone, 2006). The third dimension was of corporate governance is disclosure . This dimension is operationalized to determine the extent disclos of information as correct and reliable (Nabil Ibrahim,2012). disclosure was operationalized which have been validated by other researchers (e.g. Laika,2007, Shawawra,2008, Wen,2007,Almjhli,2009). The fourth dimension was of corporate governance is transparency .This dimension is operationalized to determine the extent transparency of information as correct and reliable (Aljadi,2010). transparency was operationalized which have been validated by other researchers (e.g. Almjhli,2009, Hermalin&Weisbach2007, Drabek,2001, Ameri & Douai,2006).

3.10.1 - Validity and Reliability of the Measurement Instrument.

Assessment of the validity and reliability of the items were conducted before the questionnaires were distributed to the respondents. This was to ensure the items were suitable for use in this study. The validity of the instrument can be classified into two main categories namely content validity and construct validity. The content validity is the conformity of the instrument whether it measures what it is proposed to measure. This can be achieved through adopted items which were used in previous research (Saunders et al., 2007). To assess the content validity, Hair et al.(2007) suggested seeking opinion from individuals such as academics who are experts in their respective area. Individuals from the population can also be chosen to obtain the feedback on the questionnaire items. On the other hand, construct validity is concerned with the theoretical and hypothetical development of the relationships between the variables (Pallant, 2007). According to Hair et al. (2007), construct validity can be verify using two approaches namely convergent validity which is to examine whether the construct of the study related positively with other measures of this construct. The other approach is discriminate validity which is to examine whether correlations exist between the study constructs. and other different constructs. For the purpose of this study, the definitions of the main variables were carefully reviewed from related literature as suggested by Saunders et al. (2007). A pretest was conducted by seeking feedback from experts, academicians, students and bank managers. Based on their comments, items amendments were carried out to ensure the familiarity, wordings and the clearness of the questionnaire items. The second criterion for assessing the measurement scale is the reliability of measurement. As validity is related to accuracy, the reliability on the other hand, is related to consistency (Hair et al., 2007). Reliability is perceived as the degree by which similar results can be obtained when repeating the same course of action under different circumstances (Crowther & Lancaster, 2009). Two approaches can be used to assess the reliability of the measurement namely test retest which is appraised by administrating the questionnaire to the sample in different circumstances and comparing the differences of their correlations. The second measure of reliability is to examine the internal consistency between items using Cronbach's coefficient alpha (Pallant, 2007). Cronbach's coefficient alpha value is the most widely used statistics to determine the reliability of the measurement (Crowther & Lancaster, 2009; Hair et al., 2007; Leedy & Ormrod, 2001; Pallant, 2007; Saunders et al., 2007; Sekaran, 2003). The value of Cronbach's coefficient alpha ranges from 0 to 1. The acceptance of this value depends on the nature and the research objectives. Commonly accepted values is around 0.7 and the value can be reduced to 0.5 for the exploratory research (Hair et al., 2007).The strength of relations for Cronbach's coefficient alpha value is summarized in Table 3.2. Alpha Coefficient Range. Strength of Association.

<0.6

poor

0.6 to <0.7

Moderate

0.7 to < 0.8

good

0.8 to < 0.9

Very good

> 0.9

Excellent Source: Hair et al. (2007). Table 3.2: Alpha Coefficient Ranges and Strength.

3.10.2 - Pilot Test.

A pilot test was conducted before the questionnaires were distributed to the target respondents. The major objective of the pilot test was to assess the goodness of the measurement in terms of validity and reliability. To achieve this objective a total of 30 questionnaires represented 15 percent of the sample were sent to banks as a subsample from the study target population to obtain their feedback. According to Cooper & Schindler (2003), the range from 25 to 100 is a suitable size for pilot test in general. After three weeks from the date of sending the questionnaire to the respondents, a total of 10 questionnaires were returned and this represents a return rate of 50%. This made the response rate for the pilot test to 50 percent which is sufficient for such research (Sekaran, 2003). The 15 questionnaires were subjected to analysis procedures to get the feedback about the reliability of the information . Analysis was conducted on accounting information dimensions and the corporate governance effectiveness as these were the two major concepts in this study. For accounting information , there were initially 22 items representing 2 different dimensions of accounting information . The results indicated misinterpretation to a reverse item about the "reliability and relevance "information is not sufficiently current for our work. This items was subsequently deleted from the questionnaire. The items related to information quality dimensions were reduced to 20 items instead of 22 items. For corporate governance , there were initially 22 items representing 2 different dimensions of corporate governance . The results indicated misinterpretation to a reverse item about the "disclosure, transparency "information is not sufficiently current for our work. This items was subsequently deleted from the questionnaire. The items related to corporate governance dimensions were reduced to 20 items instead of 22 items. After deleting the four items, the questionnaires were subjected to data analysis using the SPSS. The results were being more satisfactory after modifications.

3.11 - Data Analysis.

Prior to the data analysis, tests for normality and outliers assessments were conducted. Five different methods of analysis namely descriptive statistics, factor analysis, test of differences, correlations and multiple regressions were conducted to provide answers to the research objectives.

3.11.1 - Descriptive Statistics.

Descriptive statistic was undertaken to provide background information of the respondents. Pallant (2007) revealed that descriptive statistic aimed to: 1- Depict the different attributes of the data. 2- Verify any violation of the principal assumptions for the statistical methods to be used in the study. 3- To address particular research questions. In this study, the descriptive statistics were undertaken using central tendency and variation statistics such as means, ranges and standard deviation. Frequencies, percentages and relevant charts were also computed for nominal scale data.

3.11.2 - Factor Analysis.

Hair et al. (2007) described factor analysis as a method used to reduce a large number of variables by combining the related variables together in a factor. In this study, factor analysis was undertaken to determine the dimensions of the two major concepts namely accounting information and corporate governance, impact investment decision. Factor analysis was carried out following the main steps suggested by Pallant (2007) which consist of: 1- Consideration of the appropriateness of the data for the factor analysis by fulfilling the required assumptions such as adequate sample size, existence of adequate correlations between the variables in the same factor, achieving linearity condition and checking for outliers. 2- Factor extraction using suitable techniques to verify the smallest number of factors. In this study the principle component analysis (PCA) was adopted since this technique was widely used by researchers (Pallant, 2007). In PCA, the main variables were grouped into smaller linear variables and analyzed all the shared variance by using a mathematical model (Tabachnick & Fidell 2007). Stevens (1996) preferred PCA as it does not include any problems like other related analysis. Furthermore, Tabachnick & Fidell (2007) considered this approach as the best choice in the case of looking for an. experimental review of the variables. For these reasons, PCA was adopted for this study. 3- Factor rotation and explanation is the last step in factor analysis conducted. In specific cases, there is a need to repeat the rotation. When there appears to have high loadings in more than one factor. After the factor analysis, reliability test was undertaken to assess the goodness of the measurement. Specifically, reliability analysis is to determine the internal consistency of the measurement items after factor analysis. The most widely measurement for the reliability of the scale is Cronbach's alpha value that ranged from 0 to 1. According to Hair et al. (2007) a value of 0.7 is an acceptable alpha value for research in general.

3.11.3 - Test for Differences.

In this study, the test for differences between variables was conducted for different objectives. Test for differences between the early and late response was conducted to ensure there was no response biased in this study. The other objective to undertake this test was to test the hypotheses which were developed to answer the research questions. For this purpose, paired sample t-tests were used to test the differences between the related variables using data collected from the same respondents under two different situations (Pallant, 2007)

3.11.4 - Correlation Analysis.

Correlation analysis is described as the assessment of the relationship between two variables (Hair et al., 2007). This study aimed to examine the relationships between different variables namely information quality dimensions as the independent variables and decision effectiveness as the dependent variables. Correlation analysis was conducted for this study between information quality variables and decision effectiveness variables for three main purposes. Firstly, the tests were conducted to determine the direction of the relationship between these variables. Secondly, the tests were conducted to determine the strength of these relationships. Thirdly, the tests were conducted to examine if there were any multicollianearity between these variables. In this study, the correlations between accounting information and corporate governance, impact in investment decision effectiveness were conducted by using Pearson correlation. Pearson correlation is used to describe the strength of the relationship between two variables.

3.11.5 - Multiple Regressions Analysis.

The final method of data analysis for this study is multiple regressions. According to Pallant (2007), multiple regression is used to examine the effect of more than one independent variables on one single dependent variable. In this study, the multiple regression was employed to predict the strongest item between accounting information and corporate governance as an independent variable namely accuracy, accessibility, completeness, relevancy and timeliness on each single item of investment decision variables as a dependent variable namely quality, commitment, satisfaction and time of decisions. The rationality of using the multiple regressions to investigate the relationship between the variables is the realistic of this analysis. According to Hair et al. (2007) in the real world most of the relations were affected by many variables at the same time not by a single variable only.

3.12 - Summary.

This chapter highlighted the methodology adopted for this study. It discussed the research design and the rationality for choosing the method used in this study. The population of the study and the target respondents were also described. This was followed by data collection procedures and the development of the instrument. Finally, data analysis techniques used to analyze data were discussed. The next chapter discusses the results from the output of data analysis.

Cite this page