Different Ways Of Researching Education Essay

Published:

CHAPTER IV. RESEARCH METHODOLOGY - in this section you need to refer to different research approaches as written by different academic authors before justifying which approach bests suits your research design.

Early in this dissertation, the author has addressed in the "Literature review" part the knowledge framework of credit risk in general together with method of limitation and prevention of credit risk in banking sector. The paper also identifies necessity which leads to the the empirical research of credit risk of Vietnamese commercial banks. One of two principal contributions of this dissertation is the two objectives: to clarify the reasons of credit risk at commercial banks in Vietnam, and to propose some solutions to manage effectively credit risk in order to limit to minimum possible harm, contributing for development objectives of banking sector towards economic integration regionally and internationally. This chapter shall discuss the details of research strategy and means of data collection.

Lady using a tablet
Lady using a tablet

Professional

Essay Writers

Lady Using Tablet

Get your grade
or your money back

using our Essay Writing Service!

Essay Writing Service

A thesis is all the time, in one way or another, a research conducted by a student who, in this case, becomes a researcher. A research is what "people undertake to find out things in a systematic way, thereby increasing their knowledge" (Saunders, Lewis & Thornhill 2009, 5). This is a very brief and easy to understand definition. If the readers want a more comprehensive way ofc defining, this sentence can be more satisfactory "A focused and systematic enquiry that goes beyond generally available knowledge to acquire specialized and detailed information, providing a basis for analysis and elucidatory comment on the topic of enquiry" (White 2003, 21). Both of these are good explanations and emphasize important characteristics of a research.

Figure 7. Features of a Research (Adapted from White 2003, 21 and Saunders, Lewis & Thornhill 2009, 6)

In this thesis, the research is believed to contain those significant features.

Focused and systematic: the investigated subject of this thesis has been narrowed down to only one small bank. Moreover, this thesis is conducted in a structured way step by step: critical literature review, relevant information collection based on literature, organized information grouping for analysis purpose, conclusions 41

based on analysis and already existing knowledge. Also, the writer understands certain limitations that this research carries.

Beyond existing knowledge: Credit risk management in the researched bank has been given quite a lot of attention but still, the author thinks that it should be re-evaluated and developed. The researcher work on this thesis with a hope to help the bank understand more thoroughly how they are operating and give valuable comments and suggestions for their sake.

Analysis and conclusions: The analysis is primarily based on theoretical knowledge and interpretation of the data collected during the research.

A research always begins with defining the research topic and problem. The problem guides the research into the right direction. For this thesis, the problem of "implementation of credit risk management in a Vietnamese commercial joint-stock bank" has been clearly identified since the beginning. A well-designed research plan is the next step, in which the researcher decides what kind of data he is going to use, which methods he will apply to collect data and how he will collect them (data collection instruments). In the following sections, the readers will have a closer look into the data and research methods particularly employed in this written work.

3.1. Research Data

Basically, any research supporting evidence can fall into 2 categories: primary or secondary data. As presented earlier in the Introduction, both secondary and primary data have been utilized for this thesis.

3.1.1. Secondary Data

Secondary data are facts and figures collected by someone other than the researcher himself. These data can be used for purposes different from the researcher‟s. (Ghauri & Gronhaug 2010, 90) For instance, a student doing research on economic growth factors can take a line graph showing GDP data in the 2000-2010 period from the national statistics bureau.

Secondary data, in fact, usually help the researcher a lot in the beginning phase of their study, especially when the research problem is not familiar. The literature in 42. the theoretical background section is the very first example of secondary data. Secondary data can be obtained through numerous sources. Some divide the sources into publication (books, journal articles, etc.) and electronic (websites, emails, anything from the internet). Others like Saunders, Lewis and Thornhill (2009) gives these categories: documentary (books, reports, newspapers, transcripts, voice recordings, video recordings, etc.), survey-based (any data collected using survey strategy), and multiple source (documentary combined with survey-based combined) secondary data. Another way of classification is: internal and external sources (Ghauri & Gronhaug 2010, 97).

Lady using a tablet
Lady using a tablet

Comprehensive

Writing Services

Lady Using Tablet

Plagiarism-free
Always on Time

Marked to Standard

Order Now

Secondary data are particularly useful to this thesis. A teacher‟s materials at school triggered an interest in the author to write something in risk management. Intense review of books, articles, academic journals (publications) or electronic sources related to the topic during one month has helped to narrowed down the research problem to credit risk management in a commercial bank. Accidentally, the writer went through an article about risk management framework improvements in Vietnamese small commercial banks and would really love to study more. In Vietnam, disclosure of information in banking business is an extremely sensitive issue. Primarily the researcher has to base her analysis on information provided by the bank in published annual reports or news and existing regulation documents. Without secondary data, this thesis would never become a reality.

3.1.2. Primary Data

Primary data, in contrast with secondary data, are originally collected by the researcher with the aim of directly supporting the research topic at hand (Ghauri & Gronhaug 2010, 90). Primary data are superior to secondary ones in a way that they are chosen and collected so that they completely fit the purpose of the research. For instance, you need only GDP figures for your study. However the sources where you take the figures from contain also family income information, tax revenues, etc. Sometimes, all of these are presented in one table and one chart and the researcher has to take out the GDP data only and draw the chart by himself. Nevertheless, it is undeniable that gathering primary evidence takes 43

the theoretical background section is the very first example of secondary data. Secondary data can be obtained through numerous sources. Some divide the sources into publication (books, journal articles, etc.) and electronic (websites, emails, anything from the internet). Others like Saunders, Lewis and Thornhill (2009) gives these categories: documentary (books, reports, newspapers, transcripts, voice recordings, video recordings, etc.), survey-based (any data collected using survey strategy), and multiple source (documentary combined with survey-based combined) secondary data. Another way of classification is: internal and external sources (Ghauri & Gronhaug 2010, 97).

Secondary data are particularly useful to this thesis. A teacher‟s materials at school triggered an interest in the author to write something in risk management. Intense review of books, articles, academic journals (publications) or electronic sources related to the topic during one month has helped to narrowed down the research problem to credit risk management in a commercial bank. Accidentally, the writer went through an article about risk management framework improvements in Vietnamese small commercial banks and would really love to study more. In Vietnam, disclosure of information in banking business is an extremely sensitive issue. Primarily the researcher has to base her analysis on information provided by the bank in published annual reports or news and existing regulation documents. Without secondary data, this thesis would never become a reality.

3.1.2. Primary Data

Primary data, in contrast with secondary data, are originally collected by the researcher with the aim of directly supporting the research topic at hand (Ghauri & Gronhaug 2010, 90). Primary data are superior to secondary ones in a way that they are chosen and collected so that they completely fit the purpose of the research. For instance, you need only GDP figures for your study. However the sources where you take the figures from contain also family income information, tax revenues, etc. Sometimes, all of these are presented in one table and one chart and the researcher has to take out the GDP data only and draw the chart by himself. Nevertheless, it is undeniable that gathering primary evidence takes 43

time, frequently costs more and depends a great deal on the willingness, honesty and ability of the respondents. (Ghauri & Gronhaug 2010, 99-100)

However tough gathering the data is, primary data are still encouraged in a student‟s final thesis because it is a way for the student to practice and apply what he has learnt in reality. Being capable of conducting a research is indispensable in academic learning.

Besides secondary data, primary data are also used in this thesis. The main objective of this study is to evaluate actual execution of a credit risk management system. The staffs‟ activities are just done, not written in any documents that can be found in the bank‟s record files. Moreover, the assessment presented in the thesis emphasizes on actual practices but an official research on the transaction office‟s credit risk management operation has never been conducted anywhere. Hence, primary data are truly essential to this study. There are several options for collecting primary data, which will be discussed in the next part. Also, which options are employed will be presented.

Lady using a tablet
Lady using a tablet

This Essay is

a Student's Work

Lady Using Tablet

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Examples of our work

3.2. Data Collection Methods

The options for collecting data that we mentioned above are technically called research methods. Before getting into further discussion, a distinction between research methodology and research methods should be clarified. While research methodology plays a role as the "philosophical basis" for the research - what approach is used, research methods are practical techniques adopted to gather research information. (White 2003, 20)

Basically qualitative and quantitative are the two primary alternatives for any researcher to conduct his/her research. Which method is more suitable and more effectively reflect the whole target population really depends on the research problem that he/she has. Qualitative methods are often used for exploratory purposes (hypothesis-generating) while quantitative ones are to test hypotheses. Qualitative research results in non-quantification data. Quantitative research, on the other hand, gives numerical analysis of the issues. 44

There is a growing trend in the business and management research that different methods are employed in one single research (Saunders, Lewis & Thornhill 2009, 151). When both qualitative and quantitative methods appear, it is called mixed methods research. As argued by Saunders, Lewis & Thornhill (2009, 153), mixed methods approach is advantageous in (i) serving different purposes of a study, (ii) multiplying the likelihood of unanticipated outcomes, and (iii) minimizing the „method effect‟ and making the conclusions more valid and reliable („method effect‟ always exists because each research method has its pros and cons and all the time influences the research results). In more details, the grounds for choosing mixed methods research can be:

- Triangulation: Mixed data collection methods to corroborate research findings within a study

- Facilitation: one data collection method will facilitate research using another data collection method in a study

- Complementarity: different aspects of an investigation will be studied by using multiple research methods

- Generality: a source of data will provide the main study context and quantitative analysis give sense of relative importance

- Aid interpretation: qualitative data explain relationships of quantitative variables

- Study different aspects: quantitative discovers macro aspects while qualitative discovers micro ones.

- Solving a puzzle: another data collection method is used because the initial method provide insufficient findings for analysis

(see Saunders, Lewis & Thornhill 2009, 154)

In the practical case of this thesis study, qualitative and quantitative methods are indeed employed to tackle different aspects of the research problem. Whereas the qualitative method is used to gain understanding of the actual practices inside the bank and the transaction office, the quantitative one has served a smaller purpose of figuring more about the credit staffs (their characteristics like education, skills, experience, responsibilities, etc.). Nevertheless, the outcomes of each method do 45

not separate from each other but together facilitate the research findings‟ analysis, which means the „triangulation‟ benefit is still grasped. Triangulation can also be perceived as cross-checking or cross examination of results of the two methods. If the readers have a look at the content of the interview and questionnaire in appendix 3 and 4, they can realize several questions in the interview have been repeated in the questionnaire. For instance, while the interview asks about the types of risk management policies and their elements, the questionnaire seeks for the employees‟ assessment of their own understanding of the policies by a score scale from 1-5. Later in the analysis, these findings will be corroborated to find out the employees know the bank has those policies but their true knowledge of the policies‟ details is not as good as expected. Hence, by cross examination, this thesis will generate more reliable outcomes.

Another important reason for utilizing multiple methods in this thesis is to mitigate the „method effect‟. The interview is conducted with 2 credit staffs at the same time and is face to face. To some extent, the researcher‟s comments, tone or non-verbal behaviors or one respondent‟s answers can influence the other‟s responses. To reduce these negative effects, the questionnaire is more personal and it will protect the respondents‟ privacy as well as enable their honesty in answering to the questions.

More explanation of the two collection methods will be in the following parts.

3.2.1. Qualitative Methods

Qualitative in research means descriptive, non-numerical, non-quantified way to collect and interpret data. "With a qualitative investigation the researcher observes a great deal and any results are mostly descriptive in nature rather than sets of numerical data." (White 2003, 28)

A number of qualitative techniques have been developed from time to time, representative of which are interviews, focus groups, and observations. Interview, more specifically in-depth interview (IDI), is the technique that collects information for this study. As defined by the business dictionary, an in-depth or face-to-face interview is "conducted usually on one to one basis, an IDI is designed to reveal the underlying motives of the interviewee's attitudes, behavior, 46

and perceptions." (Business Dictionary 2010) Face-to-face interview allows much interaction between the interviewer and interviewee. In this type of interview, the researcher often has a pre-determined set of questions beforehand but flexibility is a must, which means the researcher will adapt the questions to the practical situation. He can add more questions, change the questions a bit or re-order them, etc. Like other qualitative methods, the interview aims at giving a complete, detailed description, gaining understanding and providing insights into the research problem. Saunders, Lewis and Thornhill (2009, 324) concludes that an interview is the most suitable approach if (i) the number of questions is large, (ii) the questions are complex and open-ended, and (iii) the order and logic of questioning need to be varied.

Focus groups and observations obviously are inapplicable in this research because the population is so small (only three credit staffs in the unit). For this thesis, a face-to-face interview with two employees (one relationship manager and one credit assessment officer) in the credit unit of the transaction office was carried out. The researcher prepared 15 questions, which is definitely not a small number. The questions seek to dig deeply into the practices inside the bank and are very flexible (see Appendix 3). Thus, in-depth interview is the most appropriate method. The interview went quite well and lasted for about 1 hour. Basically 15 questions were fully answered. The order of the questions was changed a little to adapt to the flow of the conversation.

Qualitative techniques are extremely useful for exploratory research - a research that has an unfamiliar problem - because they help to establish hypotheses for analysis purpose (Ghauri & Gronhaug 2010, 106). However, it does not mean a descriptive and causal research cannot employ these techniques. This thesis is an example of descriptive and causal research. The research problem is fully understood and the thesis‟ aim is to describe risk management framework in the targeted bank as well as to comment on the effectiveness of the bank‟s risk management implementation. Nevertheless, qualitative research method here in this case still proves to be of great assistance in several ways: 47

- Gathering details of the subject bank‟s credit policies: how the policies look like and if they contain fundamental information.

- Discovering how credit unit in the transaction office work and whether its operation complies with the policies

- Understanding how well the transaction office is performing based on comparing historical data and figures from year to year

- Detecting the credit employees‟ understanding of and attitudes towards the bank‟s formal policies.

3.2.2. Quantitative Methods

Unlike qualitative methods in which the researcher uses non-quantification way to interpret the collected data, quantitative methods mean the assessment of the research results is the product of a series of mathematical and statistical calculations and presentation (White 2003, 24). Very often statistical analysis is used to test some hypotheses. For instance, people who smoke possess a higher threat of getting cancer, is it true? This can be evaluated if the researcher, for example, has the number of smoking and non-smoking people getting cancer in the last five years at hand and knows basic knowledge of mathematics or statistics.

A typical quantitative method is survey, which can be conducted either through an interview (like qualitative interview) or through a questionnaire. A survey only fulfills its job when the researcher carefully chooses the sample which is representative of the population to reduce bias and ensure reliability and validity. (White 2003, 49)

For this particular thesis, the population is so small because the credit unit of the bank's transaction office has only 3 staffs. Hence, sampling here is unnecessary. For the purpose of gaining understanding of the employees‟ skills and personal opinions on the bank‟s policies, a questionnaire is designed. As we all know, a questionnaire is a series of questions, frequently in an established order, that give respondents a number of fixed-response alternatives to choose for their answers (White 2003, 50). 48

A questionnaire gives the researcher more freedom to choose the way he approaches the target respondents. If for in-depth interviews it should be face-to-face, for a quantitative interview, the researcher can conduct it through telephone or mail. Questionnaire in general saves time and money. Instead of interviewing each and every person, a large number of people can fill in the questionnaire at the same time. (Saunders, Lewis & Thornhill 2009, 362-366)

In the investigation for this thesis, the questionnaire is handed out to the respondents by hand so the threat of getting low response rate is totally eliminated. The respondents are three credit staffs that I have mentioned above, two of who also participated in the interview earlier. The questionnaire consists of 12 questions in a structured order. The first four questions aim at getting to know about the credit person. The other eight examines his/her perspective of the relationship between their job and credit risk management as well as his/her role in making any changes to the bank policies. This questionnaire actually facilitates understanding of credit staff‟s skills and know-how, which plays an important role in determining the quality of credit risk management practices. (See Appendix 4)

3.3. Validity and Reliability

Validity and reliability of a research is a key determinant of the true value of this research in the practical working life. While reliability is concerned with the result consistency (Proctor 2005, 208; Saunders, Lewis & Thornhill 2009, 156), validity is about the „honest‟ nature of the research conclusion and applicability (Ghauri & Gronhaug 2010, 65). Certain obstacles, either subjective or objective, may hinder a research study‟s reliability or validity.

The four „threats‟ to reliability are:

- participant error: the research conducted in different times may generate different results.

- participant bias: the participant (or respondent) may not be honest because of some fears

- observer error: researchers have different ways to carry out the research

- observer bias: researcher A interprets the findings differently from researcher B

49

(Saunders, Lewis & Thornhill 2009, 156-157)

Hindrances to validity can be history (external events occurring at the same time of the research may have an impact on the response), testing (the research in a way or another affect the respondents), maturation (sudden changes during the research period), mortality (participants drop out of the studies) and selection bias (subjects are not chosen randomly). (Saunders, Lewis & Thornhill 2009, 157; Ghauri & Gronhaug 2010, 66)

Reflecting those impediments on this research, the thesis is proved to be fairly reliable and valid. On the side of the research participants, they are interviewed at the time when there are no customers so that they feel more relaxed at answering the questions. Also, the interviewees are voluntarily willing to participate in the study so the participant bias should be eliminated. On the researcher‟s side, the content and the analysis of the interview‟s and questionnaire‟s results are based closely on the theories presented in the theoretical and research methodology part. All the recorded findings are fair and truthful. The researcher believes that other observers will have the same conclusions after conducting the research on this transaction office at the same time.

History, maturation, mortality and selection bias effects were not existent in the research. The investigated transaction office has only three credit staffs and all of them have taken part in the study. The testing effect, however, does exist because this research investigates directly the activities of the research participants. But its negative impact has been minimized by the researcher when she ensures the interviewees that this is anonymous and secret information.

Besides, the benefits of using both qualitative and quantitative methods (argued earlier) can also reinforce the validity and reliability. Furthermore, the research analysis takes into consideration not only findings from the primary data but a lot of secondary data have also been gathered and interpreted. The secondary data (annual reports, policies, business results) are officially published by well-known sources and cannot be manipulated by the researcher or the respondents.

To summarize, most of the validity and reliability impediments have been solved and therefore, this research is believed to remain valid and reliable. 50

3.4. Limitations

Limitations are what impede the perfection of this thesis study. First and foremost, due to the data secrecy and overprotection culture in the Vietnamese banking business, the author is not allowed to reveal the bank‟s and the transaction office‟s names in the thesis. They are always called the subject bank or the investigated transaction office. This poses some difficulty for the readers, especially the ones that are interested to know more about the bank.

Second, the performance statistics are limited. As will be discussed in the empirical study, Vietnamese banking sector is still very simple and non-transparent. The research can obtain only three-year financial data of the bank because the staff explained that they used another system before 2007 and it was impossible to take them (!). Due to the newness of the transaction office, business results of only two recent years have been acquired.

Finally, although the thesis mainly concerns with the transaction office, an interview with someone at the management level (e.g. from risk management department) would better facilitate the research findings and assessment.