Research Methodology

3.1 Introduction
The purpose of this chapter is to provide the reader with an understanding of the methodology and relevant research approaches adopted in our research. In this chapter, we explain the research philosophy, approaches and strategies, and why the methodology has been adopted, at the same time, the constraints associated with data collection and the limitations to the work will also be discussed.

The research aim for this dissertation is to investigate the current human resource management practices of small and medium-sized enterprises (SMEs) in China. Obtaining effective data and information is of vital concern to build an accurate picture of the issue being studied. To a large extent, methodology determines the outcomes of any research. Therefore, it is crucial to choose appropriate research methods and conduct them effectively in order to answer the research question and meet the research objectives well.

3.2 Discussion of Methodology Theory
3.2.1留學生dissertation網 Research Philosophy
The first question that any researcher should raise before conducting a real research project is what research philosophy you will adopt, this is very fundamental step and generally speaking, there are three views about the research philosophy that dominate the literature: positivism, interpretivism and realism (Saunders et al., 2003).

The key idea of positivism is that the social world exists external, and its properties should be measured through objective methods, rather than being inferred subjectively through sensation, reflection or intuition (Smith et al, 1991). If your research philosophy is positivism, you will assume the role of an objective analyst, make detached interpretations about data collected in a value-free manner and emphasize on a highly structured methodology to facilitate replication (Gill and Johnson, 1997) and quantifiable observations that lead themselves to statistical analysis.

By contrast with positivism, interpretivism is often associated with the term social constructionism which is critical of positivism and argues that rich insights into this complex world are needed. The role of the interpretivist is to seek to understand the subjective reality of those that they study in order to be able to make sense of and understand their motives, actions and intention (Saunders et al., 2003). In other words, social constructionism offers that reality is subjective and it is socially constructed and given meaning by people (Seddighi, 2005).

Finally, realism recognizes the importance of understanding people’s socially constructed interpretations and meanings, or subjective reality, within the context of seeking to understand broader social forces, structures or processes that influence, and perhaps constrain, the nature of people’s views and behaviors (Saunders et al., 2003).#p#分頁標題#e#

Based on the differences of three philosophies, this research is of an exploratory nature which is a kind of social constructionism. As Cooper and Schindler (1998, p. 131) state that, “exploratory studies tend toward loose structure with objective of discovering future research tasks”. Besides, MacDaniel and Gates (1999, p.63) claim that “exploratory research is usually small-scale research undertaken to define the exact nature of the problem and gain a better understanding of the environment within which the problem occurred”. All of these fit well with the objectives and other conditions of my research project. Therefore, we adopt the interpretivism philosophy of an exploratory nature in this study.

3.2.2 Research Approach
Inductive and deductive approaches
There are two basic research approaches available when conducting business research including deductive and inductive methods (Saunders et al., 2000). Their characteristics are described in the following table:
Table 3.1 Comparison of Deductive and Inductive Research Approach
Induction emphasizes Deduction emphasizes
-gaining an understanding of the meanings humans attach to events -scientific principles
-a close understanding of the research context -moving from theory to data

-the collection of qualitative data -the need to explain causal relationships between variables
-a more flexible structure to permit changes of research emphasis as the research progresses -the collection of quantitative data
researcher independence of what is being researched
-a realization that the researcher is part of the research process the application of controls to ensure validity of data
-less concern with the need to generalize -the operationalisation of concepts to ensure clarity of definition
-a highly structured approach
-the necessity to select samples of sufficient size in order to generalize conclusions
(Saunders et al., 2003)
The deductive approach is usually regarded as an effective way to test theories, where people develop a theory and a hypothesis (or hypotheses) and design a research strategy to test the theory (Saunders et al., 2000). The focus is on whether or not the suggested theory fits, and is appropriate for the organization (Saunders et al., 2000; Seddighi, 2000). The deduction to research owes much to what we would think of a scientific research, in which you develop a theory and hypothesis and design a research strategy to test the hypothesis (Saunders et al., 2003).On the other hand, inductive approach is usually used to build theories, in practice, people need to collect data and develop theory as a result of data analysis. Nevertheless, the two methodologies are not mutually exclusive, in reality; they are often used jointly in business and management studies (Seddighi, 2000). However, it is important to point out that the choices of research approach should base on the research question and research objectives.#p#分頁標題#e#

Therefore, there is no question that the inductive approach is appropriate for this particular research. I will use the inductive approach, from data to theory, by collecting the data first and then developing new hypothesis as a result of the data analysis. In reality, the research focuses on not only understanding why something happens in the business sector but also describing what happens. As the inductive approach usually only tell people why something happens rather than describe what happens (Saunder et al., 2000), it would have advantage to combine the two approaches in this particular research according to the nature and purpose of this research.

Quantitative and Qualitative Approach
There are two main types of marketing research in terms of the two, essentially different types of data that are generated by fundamentally different research approaches-quantitative and qualitative methods (Adcock et al., 1995).

Quantitative research involves the collection of information that can be expressed using a numerical measure (Brassington and Pettitt, 2003). However, it includes not only numerical data such as sales figures, market share, market size and demographic information, but also the numerical aspects of other data, often derived from primary research, such as questionnaire-based surveys and interviews (Adcoco et al., 1995). Obviously, the quantitative research usually involves large-scale surveys that enable a factual base to be formed with sufficient strength to allow statistically rigorous analysis (Brassington and Pettitt, 2003). Therefore, the success of quantitative research to a large extent depends on establishing a representative sample that is large enough to ensure that the data collected are reliable and objective. Due to time and financial constraint, obviously, it is unpractical to conduct a truly quantitative based research regarding the timetable and the requirement of the research project. That is not saying that quantitative data will not be utilized in this project. In fact, it is important to obtain the quantitative data from secondary sources in order to support the argument.

Qualitative research, on the other hand, usually involves the collection of non-numerical data that is open to interpretation, such as customer’s opinions, where there is no intention of establishing statistical validity (Brassington and Pettitt, 2003). The essence of qualitative research is that it is ‘diagnostic’; therefore, it is especially useful for investigating attitudes, motivations, beliefs and intentions. In practice, they are usually based on small-scale samples; therefore, it cannot be generalized in numerical terms (Brassington and Pettitt, 2003). Chisnall (1997) further characterized the method as impressionistic rather than conclusive, he also pointed out that the approach could provide a better understanding of certain factors that might influence buying decisions. However, it is important to point out that the results generalized from the process are often subjective. But for all its limitations, qualitative research is an effective way to reflect the complexity of the interrelationships associated with marketing activities (Chisnall, 1997).#p#分頁標題#e#

Table 3.2 The differences between quantitative and qualitative research methods
Quantitative methods Qualitative methods
•Emphasis on testing and verification
•Focus on facts and /or reasons for social events
•Logical and critical approach
•Controlled measurement
•Objective “outsider view” distant from data
•Hypothetical-deductive; Focus on hypothesis testing
•Result oriented
•Particularistic and analytical
•Generalization by population membership
•Emphasis on understanding
•Focus on understanding from respondent’s/ informant’s point of view
•Interpretation and rational approach
•Observations and measurements in natural settings
•Subjective “insider view” and closeness to data
•Explorative orientation
•Process oriented
•Holistic perspective
•Generalization by comparison of properties and contexts of individual organization
Source: Based on Reichardt and Cook (1979)

According to the objectives of my research, qualitative methods will be the main research method to provide intricate details and understanding in my project, because qualitative data is associated with such concepts and is characterized by their richness and fullness based on your opportunity to explore a subject in as real a manner as is possible (Robson, 2002). However, it is important to point out that the results generated from this process could be subjective. But besides its limitations, qualitative research is an effective way to reflect the complexity of the interrelationships (Chisnall, 1997). Nevertheless I am not saying that quantitative data will not be utilized in this project. In fact, quantitative data is very important to present the facts and backgrounds of the company and the industry as a whole. Due to time and financial constraint, I will obtain the quantitative data from secondary sources in order to support the argument.

3.3 Data Collection
This section introduces the issues about data collection, which includes data source, data collection techniques, access, sample selection, and data analysis methods and so on.

3.3.1 Data Source
There are two sources of data including secondary data and primary data. Secondary data are information collected by others for purposes that can be different from ours, while primary data are original data collected by us for the research questions at hand (Ghauri and Gronhaug, 2002). In this research, data will be collected from both the secondary source and the primary source. On the one hand, the data will be collected from public sources such as books, published journal articles, reports and websites. On the other hand, the data will be also collected from the some Chinese SMEs that are selected in the sample of this research.

3.3.1.1 Secondary Data Collection
Sauders et al (2003) divide secondary data into three main subgroups: documentary data, survey-based data and those compiled from multiple sources. It has a lot of advantages in processing my research. Secondary data is the data which has been collected by someone else for a separate purpose, and is then used by other researcher. The value of secondary research is in what it can bring to the author as a basis for the work and individual contribution.#p#分頁標題#e#
Commonly, secondary data come from printed sources (Books, Magazines, Journals, and trade Newspapers) and in electronic sources (CD-ROM encyclopaedias, Software packages, or online services, such as the Internet.) Books are general resource to provide relevant theory support for the study. Journals are a useful resource for information on a daily basis. They could provide the latest views and development in the particular area of study. The Internet is also very helpful for the data collecting, especially professional websites
Advantages of Secondary Data Collection
The most important factor determining me to use the secondary data is that it is easy to do and helps to save both time and money. My research dissertation has to be completed within three months, which implies constraints from the budget, time, and some other factors. Usually, it is much less expensive to use secondary data than to collect the primary data. Furthermore, the verification process is more rapid and the reliability of the information and conclusion is greatly enhanced.
Disadvantages of Secondary Data Collection
Although secondary data collection is very useful for us to collect the existing information, the information collected through this method may be incomplete. Using this technique requires people to be quite clear about what they are looking for and this technique is restricted to the data that already exist. It is not a flexible way to collect data. If you are not clear about what you are looking for, you may spend much time but not collect the data you need. So this method of documentation should be used base on the assumption that you have the clear objective about what kind of data you want to collect.
And the data and information that you want to collect through this method must be the real and existing information available in the market. So this method just can help to collect the past and historic information and can not collect the in-depth information. The data sources of secondary data collection are from the public and existing materials and are not directly from the customers or other market sources.

3.3.1.2 Primary Data Collection
There are many data collection techniques to collect primary data from the research including interview, questionnaires, focus group and case study and so on. Primary data collection is every important for us to collect in-depth data and information, while secondary data collection only can provide us with the existing and skin-deep data and information.

Table 3.3 Advantages and Disadvantages of Primary Data
Advantages of Primary Data Disadvantages of Primary Data:
Can probe deeper answers Takes time
Can get detailed information on what causes problems or benefits. More opportunities for bias creep in when results are "coded"
Can also elicit more honest and more sensitive information Also the ever-present problem of all self-report measures
By using interviews, future change agents can also introduce themselves to the people in an organization, and establish both rapport and trust Accuracy#p#分頁標題#e#

(Saunders et al., 2003)
Due to the advantages of primary data, it is important for us to collection primary data through data collection techniques like interviews. Primary data collection can provide us with the in-depth data and information related to our research questions. Therefore, primary data collection is crucial to the success of my research despite of its main disadvantage that it could take a long time and cost much to collect the data and information.

3.3.2 Data Collection Methods
According to the different types of research, there are many different data collection techniques such as interviews, questionnaires, survey, observation, focus group, and documentation review and so on (Saunders et al, 2003). The different data collection techniques have their different advantages and disadvantages, and should be adopted according to the different research purposes. In this research, three data collection methods of documentation review, in-depth interviews and observation will be adopted.

3.3.2.1 Documentation Review
Among data collection techniques, documentation review is the method to quickly and accurately collect the existing information and historical information. This method is to collect data through reviewing the formal company documents and reports about production, sales and finance. (Saunders et al, pp. 104, 2003)

Advantages and Limitations
We choose this method of documentation review, because this method is easy to do and has few limits to the market environment and we need to use this method to collect some existing and historic data and information for the research. This method of documentation review has many advantages and is generally used in the case study. The information collected through documentation review is factual information and through this method, we can get the comprehensive and historical information with few biases. So this method is very useful for us to understand the real information about the company at the beginning, because the documents about the company are all announced by the company, and are real and open to all the people concentrating on the company.

Although the research method of documentation review is very useful for us to collect the existing information, this method often takes time and the information may be incomplete. Using this technique requires people to be quite clear about what they are looking for and this technique is restricted to the data that already exist. It is not a flexible way to collect data. If you are not clear about what you are looking for, you may spend much time but not collect the data you need. So this method of documentation should be used base on the assumption that you have the clear objective about what kind of data you want to collect. And the data and information that you want to collect through this method must be the real and existing information available in the market. So this method just can help to collect the past and historic information and can not collect the in-depth information. The information sources of this method are from the public and existing materials and are not from the customers or other market sources.#p#分頁標題#e#

3.3.2.2 In-depth Interviews
Generally, Observation, interviews and questionnaires are three main primary data collection methods (Sekaran, 2000). According to the purpose of this project and the restriction of time and geography, I will mainly focus on the method of interviews to collect qualitative data. The semi-structure interview has been used to collect data.

Interview is a good data collection technique to get the in-depth information and it is particularly useful for getting the story behind a participant’s experiences. This method is used when we want to fully understand the impressions or experiences of someone or want to learn about their answers to the questionnaires. (Creswell, pp.125, 1994) Interview is the process of the communication between the interviewer and interviewee. The communication includes the free talking and the discussion about the assigned questions. This data collection technique provides people with a communication way to know the information they want to know from other people or know the opinions of other people on the assigned questions. This method of interview has strong pertinence to investigate the special issue and widely used in the market research to obtain the in-depth information.

Advantages and Limitations
This data collection technique of interviews has many advantages. It is the most direct way for market researchers to communicate with the interviewees. Through this face-to-face way, the method of interviews can collect the full range and depth of information from the interviewees. This kind of information is very useful and may not be collected through other ways. And this method can be flexible with the different interviewees and the interviewers can determine and choose the questions based on their thoughts and objectives to discuss with the interviewees and obtain the data and information they need. Besides, it is also a good way to develop the relationship with the clients. (Creswell, pp.136, 1994)

And data collection technique also has many disadvantages and may lead to the biases of the collected information, because the use of this method may take much time, may be hard for people to analyze and compare, may be costly and the interviewer may bias the responses and opinions of the interviewees. On the one hand, the interviewees may not tell the full information they know or may tell the wrong information to the interviewers. And on the other hand, the interviewers may misunderstand the information from the interviewees and causes the bias during the process of data collection.
3.3.2.2 Participant Observation
According to Collis and Hussey (2003), the method of participant observation is where the researcher becomes a working member of the group or situation to be observed. The aim of this method is to understand the situation from the inside: from the viewpoints of the people in the situation, and the researcher shares the same experiences as the subjects and this form of research can be particularly effective in the study of small groups/small firms. Participant observation can be overt (everyone knows it is happening) or covert when the subjects being observed fro research purposes are unaware it is happening (Collis and Hussey, 2003).#p#分頁標題#e#

In this research, the author works as an internship work in Ken Bang Business Investigation Company, and this company is one of SMEs in China. This company is a good case for the author to study the research question about the current human resource management practices of SMEs in China, and this is the reason for the author to choose this company for internship. During the period of internship, the author also studies the research questions through observation in daily work of the company.

3.3.3 Sample Selection
Besides the company in which the author worked during the period of internship, the author also chose other ten small and medium sized enterprises in the sample for research. There are many SMEs in the whole China, so it is impossible for the research to include all SMEs in the sample. Because the author was working in Beijing during the period of internship, so ten SMEs in Beijing were randomly selected in the sample.

3.3.4 Access
I contact with these SMEs in the sample to make an appointment with their directors and managers for interviews. The interviews will be conducted in these companies.

Based on the method of interview to collect primary data from the directors or managers of these selected SMEs, I just simply make the sources anonymous and the information about the interviewees such as the name and position will be kept in confidentiality. This technique will not affect the data collection and the research. The true and effective data and information collected from these anonymous sources will be used to analyze the research questions.

3.4 In-depth Interview Design
It is true that the quality of the data to a large extent depends upon the structure of the interview as well as the design of each individual question (Saunders et al., 2003). Interview is the method that may take much time and lead to the bias. So we must design the appropriate questions and design the process of the interview. In order to collect the data and information we need for the analysis, I design the questions according to our research aims and objectives.

The research aim of our paper is to investigate the current human resource management practices of Chinese SMEs. Based on this main research aim and other research objectives that have been decided before the research, the research should focus on the current practices of human resource management in Chinese SMEs including the status quo of the current human resource management for Chinese SMEs, the main problems involved with Chinese SMEs in human resource management. So the questions designed for the interview must help collect the data and information focusing on these aspects relating to research aim and objectives. The collected data must help the study to realize the research objectives.

Besides the contents of the questions, the process of interview is also very important. I plan to start with some simple closed format questions focusing on some general information about human resource management of Chinese SMEs and then majority of questions with open format are created during the interview for the interviewees to express their opinions freely. This way that asking the general questions first and then discussing the in-depth questions gives both the interviewer and the interviewee more flexibility to communicate and discuss the issues. One important point for us to attention is that we should ask the questions in a clear, easy and simple way for the interviewees to understand and discuss.#p#分頁標題#e#

The designed questions for the interview are summarized as follows:

1） Do you think human resource management is important to the development of the enterprise?
2） What do you think of the current human resource management in your enterprise?
3） What are the differences between practices of human resource management in China and those in the western developed countries?
4） What do you think of the role of human resource management to SMEs in China?
5） Do you think the current human resource management in your enterprise is effective or not?
6） What are the main problems involved with the human resource management in your company?
7） As a SME in China, do you think your enterprise have to face more problems in human resource management compared with the large enterprises in China?
8） What are your suggestions for Chinese SMEs to improve their human resource management?
3.5 Data Analysis
The data collected in this research include both the quantitative data and the qualitative data, so the different methods of data analysis have to be adopted to analyze the different types of data.

3.5.1 Analysis of Quantitative Data
Quantitative data refers to numerical and standardized data, which usually can be analyzed through using of diagrams and statistics (Saunders et al., 2000). Basically, data can be divided into categorical and quantifiable data, it is important to edit and code data before start analysis (Saunders et al., 2000). In practice, it is useful to conduct exploratory data analysis in the initial stages of data analysis, this approach highlights the importance of using diagrams to explore and understand data (Saunders et al., 2000). For instance, pie charts and percentage component bar charts can be used to explain the proportions of customers agreeing or disagreeing with the statements which are presented in the questionnaire. “Subsequent analysis will involve describing your data and exploring relationships using statistics” (Saunders et al., 2000). Finally, it is essential to re-emphasis the importance of using PC software programmes during the process of data analysis, such as SAS, APAK EXCEL etc (Greenfield 2002).

3.5.2 Analysis of Qualitative Data
It is obvious that qualitative data here refers to the data collected by semi-structured interviews and the possible secondary source of data. According to the nature of qualitative data, it is important to develop data categories and to classify them into appropriate categories before analysis, “otherwise the most that can result will be an impressionistic view of what it means” (Saunders et al., 2000, p381). It is also important to recognize the relationship between different categories of data, as well as to develop and test hypotheses with regard to the research question and objectives (Saunders et al., 2000). In reality, it is necessary to develop a provisional set of categories from the research question, research themes and initial propositions, the categories should be closely related to the research question. In short, it is critical to conduct data analysis effectively in order to answer the research question and achieve research objectives.#p#分頁標題#e#

3.6 Ethical Issues
“Research ethics refer to the appropriateness of researchers’ behavior in relation to the rights of those who become the subject of your work and are affected by the work” (Saunders et al., 2000, p142). They are likely to occur during the whole process of the research: seeking access, data collection, and data analysis and reporting (Skearan, 2000). There are some ethical issues to be considered in my research. First is privacy, which may be seen as the cornerstone of the ethical issues that confront those who undertake research. In my interviews, I will not ask the participants to fill their name and I will keep all the information I have got from the interviews as confidentiality. No matter during the design and initial access stage, the data collection stage or the analysis and reporting stage, the confidentiality and anonymity are always important. I will not attempt to apply any pressure on intended participants to grant access. Once access has been granted, I will remain within the aims of my research project that I shared and agreed with my intended participants. Sometimes, the findings may be used to make a decision that could adversely affect the collective interest of those who will be my participants, it is ethical for me to refer to this possibility even though it may reduces the level of access that I will achieve. The last problem is netiquette, which has been developed to provide a heading for a number of “rules” or guidelines about how to act ethically when using Internet. Thus, I will pay much attention to the netiquette when I distribute my on-line questionnaire via email.
Moreover, the research does not harm the safe or health of the participants in the research. Thirdly, I will keep the confidentiality of the private information about all participants in the research and also keep the confidentiality of the data collected in the research, and only use the collected data and information for the purpose of research.

3.7 Research limitations
Several limitations were encountered during the research. Firstly, the topic of this dissertation should be quite big, and we have the every reason to believe that any each section of this work should be discussed in a great length of article. From this point of view, the discussion of each section could be relative shallow. The second limitation was “time”. Due to the time constraints of the assignment, the study could not be as in-depth as desired. The third limitation was about the data collection in the research. Although we used the data collection techniques to collect primary data and secondary data from the extensive sources, there are some limitations. On the one hand, the secondary data may not perfectly suit the purpose of current research since data was collected for other purposes. Like Zikmund (1997) said “An inherent disadvantage of secondary data is that they were not designed specifically to meet the researcher’s needs” Also, secondary data may not be complete enough. On the other hand, the bias may exist in the primary data because the interviewer may misunderstand the answers from the interviewees and the interviewees may not express their real opinions and thoughts on the questions.#p#分頁標題#e#

3.8 Conclusion 留學生畢業dissertation網
Research methods are quite crucial to any business research. Choosing an appropriate method to a large extent determines the success of the research. Both inductive and deductive approaches are used in this particular project. Both quantitative and qualitative data will be collected. The qualitative data is mainly from primary sources based on the in-depth interviews that have been conducted with the directors of the chosen SMEs in the sample and the observation of the author during the period of internship in the case company, while the quantitative data is mainly obtained from secondary resources such as some existing and historic information. In short, the methodology used in this research is considerably appropriate and effective in order to analyze the research questions and meet the research objectives.