Implementation of Knowledge Management in Thai public sector

Chapter 3
Research Methodology
3.1 Introduction
3.2 Research design
3.3 Research philosophy
3.4 Research approach
3.5 Data collection methods
3.6 Survey
3.7 Sampling
3.8 Data analysis
3.9 Research Limitations
3.10 Ethical considerations
Chapter 3
Research Methodology

3.1 Introduction
Methodology helps a researcher to conduct his study in a structured manner. It helps him to collect
only those data which are relevant to his study. The success of a research project greatly depends
in the ability of the researcher to use appropriate methodology. Methodology should be determined
based on the nature of the topic under study. A typical research methodology consists of many
elements such as research design, research philosophy, research approach, data collection methods,
data analysis, ethical considerations, limitations of the study etc. For convenience, these
elements are included under different subheadings below.
3.2 Research design
A design or plan required for the construction of a building or home. A designer decides how the
living rooms, dining rooms, kitchen, drawing rooms, site out etc are constructed and placed within
a home. Same way, a design is necessary for the successful completion of a research study. It helps
the researcher to identify what topics should be investigated, what should be the aim and purpose
of the research, how the data should be collected and analyzed, how conclusions can be drawn out of
the research etc. In short, research design is a framework that helps the researcher to conduct his
study from beginning to the end in a successful manner. A carefully constructed research design
would enable the researcher to draw inferences and identify relationships among the variables from
the collected data (Miller and Salkind, 2002).
Research design should be decided based on the nature of the problems addressed in the study. If
the research problem has not been clearly defined at the beginning of the study, an exploratory
research design can help the researcher. Exploratory research design is used in the current study
since the research problem is not defined clearly at the beginning. For example, it is difficult to
understand the nature of the current research questions such as; what are the challenges in the
externalization of tacit knowledge for the proper implementation of KM in the public sector
undertakings in Thailand or how knowledge management can be implemented successfully in the public
sector undertakings of Thailand. Exploratory research design would help the researcher to review
the previous research studies and identify the key problems and concepts involved in his study
(Jupp, 2006). That is why a comprehensive literature review has been conducted as part of this
study to know the various dimensions of the research problems of this study. Saunders et al (2012)
have constructed a research design or research onion which is helpful in conducting an exploratory
research like this.

Research Onion (Saunders et al., 2012)
It is evident from the above illustration that philosophies, approaches, strategies, choices, time
horizons, techniques and procedures are the major elements of a research design. Positivism,
realism, pragmatism and interpretivism are the major research philosophies whereas the research
approach can be inductive or deductive. On the other hand research strategies include, survey,
interviews, case study, action research etc whereas choices involve, mono method, mixed method and
multi-methods. Moreover, various techniques and procedures such as induction, deduction, abduction,
etc are used for the data analysis of a study. These elements are described in detail below

3.3 Research philosophy
Positivism, pragmatism, interpretivism and naturalism are some of the most commonly used research
philosophies. These research philosophies are used based on the nature of the research problem
investigated. For example, some research problems require 100% accuracy in the results. In such
cases, positivism can help researchers as this research philosophy is based on scientific methods
of investigations. In some other cases, the researchers may come across with many unexpected data
which require interpretation. Scientific methods may not be able to interpret such data. In such
cases, researchers make use of naturalism research philosophy which is based on probabilities and
possibilities. Positivism works on quantitative tools, techniques that give more importance to
measuring and counting, whereas naturalism prefers qualitative tools such as observation,
questioning, and description (Sagepub, n.d.).
The present study makes use of a missed research philosophy: a mixture of positivism and
naturalism. Positivists are of the view that the true source of knowledge is science and anything
without the backing of scientific theories and principles may not be true all the time. They
believe that a theory is reliable only when it is able to provide the same result on different
occasions of testing (Gabriel, 1990). Positivism works on the principle that a research study
should be free from the influence of researcher’s beliefs and attitudes (Hallebone and Priest,
2009).
While positivism works on facts, naturalism works on probabilities. Naturalism gives more
importance to analysis, judgements, beliefs and probabilities (The University of Utah, 2009).
Naturalists are of the view that there are many things which can be explained with the help of
scientific methods. For example, it is impossible to measure the quantity of knowledge possessed by
an employee. No scientific methods can measure the magnitude of knowledge possessed by a person.
Same way, it is impossible to know whether an experienced employee has transferred all his
knowledge to the fresh employees. In other words, it is impossible to measure the magnitude of
tacit knowledge converted to explicit knowledge. In such cases, it is logical to use probabilities
and possibilities rather than scientific facts. In other words, naturalism will be helpful while
interpreting such data. The purpose of naturalist research approach is not to have answers to put
through a computer to achieve percentages or statistics, but to get an insight of the meaning of
what people are saying, in different contexts (Gabriel, 1990).
Both quantitative and qualitative data are important for this study and hence it is impossible to
avoid either of the two research philosophies mentioned above. For example, the performances of a
public sector organization in Thailand before and after the implementation of knowledge management
can be measured with the help of statistics or numbers. Same way, the reasons of the success or
failure of KM in public enterprises in Thailand can be assessed with the help of theories,
principles or qualitative data. In short, the use of naturalism and positivism is unavoidable for
the current study.

3.4 Research approach
Research approach is helpful in connecting theories and principles with realities. Moreover, it is
helpful in formulating theories and principles from collected data (Burney 2008). Both inductive
and deductive approaches used in this study. The reason for the use of deductive approach is the
positivism research philosophy used whereas the reason for the use of inductive approach is the
naturalism research philosophy used for this study. In other words, positivists use only deductive
approach whereas naturalists use inductive approach while doing research (Sekaran, 2006; Easterby
et al. 2012). For example, there may be many reasons for the success or failures of conversion or
externalization of tacit knowledge into explicit knowledge. Some of these reasons can be
interpreted with the help of scientific methods. But there may be many other reasons which cannot
be measured accurately with the help of scientific methods. In such cases, it is logical to use
both inductive and deductive approaches while investigating a problem (Hyde 1998).
3.5 Data collection methods
A mixed research method was employed for the collection of data for this study. Both primary and
secondary research techniques used for the collection of primary and secondary data. Moreover,
quantitative as well as qualitative data also collected with the help of different methods for this
study. As pointed out by Saunders et al (2012), mixed research method refers to the research
conducted with the help of quantitative and qualitative research techniques.
As the name indicates, quantitative research is more interested in collecting quantitative data. It
is often referred as logical and data driven research because of the use of statistical and
numerical tools. It employs data collection methods such as questionnaires, surveys, experiments,
econometrics, and mathematical modeling. It concentrates more on answering questions starting with
how many. On the other hand, qualitative research makes use of data collection methods such as
interviews and observations. It answers questions starting with why rather than how many (Neimark,
1995). Quantitative research techniques such as surveys are capable of collecting data from a wider
population whereas qualitative research techniques such as interviews are incapable of doing so
(Easterby-Smith et al 2012). That is why this study has given more importance to quantitative
research techniques.
Quantitative research is capable of making theories and principles from the collected data related
to human behavior. Moreover, it is capable of predicting how human responds in certain situations
(Bogdan & Biklen, 1998). According to Castellan (2010), quantitative research is able to establish
various relationships among variables and statistical description. Moreover, it is capable of
establishing facts, validation, prediction and control and testing hypotheses (Castellan, 2010).
Another important utility of quantitative method is its ability to generalize the results to a
wider population (Wu and little, 2011). Moreover, it is capable of providing accurate answers to
statistical problems (Creswell, 2013).
One of the major limitations of quantitative research is its inability to provide qualitative data.
Qualitative data are necessary along with quantitative data for the success of a research study.
The conclusions made out of quantitative research cannot be applied to a wider population, if it
lacks the support of qualitative data. For example, the present study is investigating the KM
implementation problems in the public sector of Thailand. The conclusions made out of this study
cannot be applied to a wider context since the survey was administered only in the public
enterprises of Thailand. It is not necessary that the data collected from Thailand’s public sector
may be relevant to public sector companies in other countries.
At the same time, the findings from a qualitative research can be applied to a wider population
since it works on universally accepted theories and principles. Qualitative research is capable of
providing in-depth understanding about the problems (Bluhm et al. 2011). It can develop and test
many hypotheses and assumptions. The data produced by qualitative research consists of descriptions
of situations, events, people, interactions and observed behaviors etc (Sykes, 1990). Qualitative
research helps the researcher to manipulate subjective information. Moreover, it enables the
researcher to identify the trends related to the research problem by establishing greater
communication with the respondents. At the same time, the major limitations of this type of
research are: hard labor, time consuming, technical issues etc (Creswell, 2013).
Despite of the limitations and advantages of qualitative and quantitative researches, it is a fact
that these two research techniques are unavoidable for the success of a research project. It is
often said that quantitative data defines whereas qualitative data explains a problem. In other
words, quantitative data is incapable of going deep into the research problem whereas qualitative
data is capable of doing so. In other words, quantitative research attacks a research problem only
at the surface level whereas qualitative research is capable of attacking a problem from the grass
root level. That is why Bryman (1988) mentioned that qualitative method can be used to facilitate
the quantitative part of the study, or then the other way around, or then both approaches can be
given equal emphasis. As per the views of Collins et al. (2006) and Bryman (1992), the success of a
research study depends on the ability of the researcher to mix the qualitative and quantitative
research techniques judiciously.
Qualitative data for the current study were collected through secondary research. Secondary sources
such as books, journals, articles, newspapers and internet were used for the collection of
secondary data. According to Jackson (1994), secondary data is readily available and easily
accessible to all after the arrival of computers and internet. The literature review has given many
insights about the factors that are affecting KM implementation in public sector. Based on these
factors, the researcher has developed six hypothesizes. In order to test the validity of these
hypothesizes in the context of Thailand’s public sector, the researcher has included many questions
in the questionnaire (Please see the appendix for more details about the questionnaire)
3.6 Survey
An online questionnaire survey was conducted as part of this study for the collection of primary
data. As per the views of Walonick (1993) aquestionnaire survey is a mechanism which allows the
researcher to collect the views and opinions of people with the help of open or closed ended
questions. In his opinion, questionnaire survey is an estimation apparatus which is capable of
collecting meaningful data from a wider population. The survey for this study was constructed with
the help of closed ended questions only. The researcher is aware of the fact that closed ended
questions have limitations in collecting qualitative data as the respondents forced to answer the
questions objectively. On the other hand, open ended questions are capable of collecting
qualitative and descriptive data. At the same time, open ended questions are time consuming
compared to closed ended questions from the point of view of the respondents. Therefore,
respondents usually hate answering open ended questions while taking part in a survey. In other
words, the response rate in a survey will be weakened if the survey consists of too many open ended
questions. In order to increase the response rate as much as possible, the survey conducted for
this study used only closed ended questions (Please see appendix for more details of the survey
questions).
Awareness about KM, organizational resources, externalization of tacit knowledge, sharing of
knowledge, political will, and bureaucratic power were taken as the independent variables and
employee resistances, barriers, employee knowledge level employee stress level, policy changes
administrative barriers and knowledge management implementation were taken as the dependent
variables in this study. The questionnaire was prepared based on the following hypothesis:
• Awareness about KM reduces employee resistance
• Availability of organizational resources reduces barriers against the implementation of KM
• Externalization of tacit knowledge helps the implementation of KM.
• Sharing of knowledge reduces employee stress level
• Political will is necessary for the implementation of KM
• Lack of bureaucratic power is essential for the implementation of KM
LINE application helped a lot in the administration of the survey. LINE is an instant communication
application just like WhatsApp. It can be installed on electronic devices such as smartphones,
tablet computers and personal computers. LINE users normally use this application for exchanging
texts, images, video and audio. LINE application is extremely popular in Thailand. It was used for
circulating the link for surveying because the message can go directly to many targeted
respondents. The respondents targeted for this study were only the government officers. No private
employees were allowed to take part in the survey as the present study was limited its scope only
to the public sector enterprises in Thailand. Among the government officers in Thailand, there are
many groups, classified by the target users, that sending info among each other. For example, there
are several line groups, closed-friends groups, co-worker groups, professional groups, etc among
the people working in Thailand’s public sector. So circulating the link via LINE app with some
brief introduction about what this questionnaire is for, can be the most convenient way to attract
more participants towards this survey. The researcher has used the website: www.surverycan.com for
the administration of the survey. It is quiet similar to www.surveymonkey.com. The difference
between surverycan.com and surveymonkey.com is the fact that surverycan.com is constructed in Thai
language whereas www.surveymonkey.com is constructed in English language. Both these websites give
free services to its customers. Since proficiency of Thai people in English language is not so
good, the researcher decided to use the services of www.surverycan.com in order to help the
participants in answering the questions properly. The researcher might be able to collect around
500 responses from the survey.
As mentioned earlier, surveys are capable of collecting data from a wider population. It can be
administered easily with the help of online or offline channels. Moreover, it is highly cost
effective and cheaper compared to other research methods such as interviews. One of the distinct
features of surveys is its ability to reach distant population. It is possible for a researcher to
collect data from any part of the world while surveys are administered through online channels. The
data collected from surveys can be analysed for validity and credibility with the help of software
such as SPSS, MS Excel etc. At the same time surveys have the following disadvantages also:
incapable of providing descriptive data, reliability of data depends on the attitudes and
capabilities of the respondents, validity is less while administering a survey with closed ended
questions alone, chances of data error and misunderstanding the questions by the respondents etc
(Sincero, 2015)

3.7 Sampling
Random sampling was employed while the survey was administered. It is a sampling technique in which
each member of the subset has an equal probability of being chosen (Investopedia, 2015). Random
sampling helps the researcher to take avoid prejudices and biases while selecting the samples. It
helps the researcher to select samples from different age groups, different genders and different
cultures. For example, the thinking patterns of Thailand female employees about the KM
implementation need not be the same as that of Thai male employees. Moreover, it is not necessary
that the views and opinions of fresh employees about the implementation of KM may be the same as
that of the experienced ones. While using random sampling methods, the researcher will be able to
collect data from all groups, subgroups and communities working in Thailand’s public sector
enterprises.
3.8 Data analysis
Data analysis is the most important part of any research study. Even if a researcher was able to
collect ample data, improper data analysis techniques may prevent him from making meaningful
conclusions. Data analysis can be compared to the processes in a manufacturing unit. In a
manufacturing unit, the raw materials undergo various processes before the production of final
product. The processes should be appropriate in order to get the desired output. Same way, raw data
undergo many data analysis techniques before attaining the shape of results, findings or
conclusions. The data analysis techniques should be appropriate in order to get the right results.
Since the deductive and inductive approaches used for this study, it is inevitable to use deduction
and induction as the data analysis methods of this study. Deduction is the process of finding data
to explain a theory whereas induction is the process of making a theory based on the collected
data. Since theorizing is involved in both the processes, Partington (2002) mentioned that “there
is no point in differentiating induction and deduction” (Partington, 2002, p.155). These processes
can be illustrated as follows:
It is evident from the above algorithms that deduction gives 100% right results all the time
whereas same thing cannot be said about induction and abduction. In other words, facts are the core
of deduction method, whereas probabilities are the core of induction and abduction methods.
Deduction and induction can be represented schematically as follows;
Theory observations/findings (Deduction)
Observations/findings Theory/Induction
(Bryman and Bell 2007, p.13).
Since deduction works on scientific methods and principles, the results obtained from deduction
methods of data analysis will be 100% right all the time. At the same time, induction works on
probabilities and hence the results obtained through this type of data analysis may not be 100%
right all the time. Deduction and induction can be represented schematically as follows
If X => Y, and Y => Z, then X => Z (deduction)
If X => Y, and X => Z, then Y => Z (induction) (Apgaylard, 2013)
Data collected from the survey will be analyzed with the help of MS Excel software. The major
advantage of using MS Excel for data analysis is its ability to represent data analysis
graphically. MS Excel is capable of making various tables, graphs and other illustrations from the
collected data.
3.9 Research Limitations
Lack of qualitative data is the major limitation of this study. No interviews will be conducted as
part of this study. Interviews could have provided more qualitative data and such data could have
been used to substantiate the arguments from the quantitative data. However, lack of funds and time
forced the researcher to avoid interviews.

3.10 Ethical considerations
Privacy is an issue of concern in all countries and it is so in Thailand also. The researcher has
decided not to disclose the names and personal details of the survey participants without written
consent from them. In order to avoid plagiarism problem, the researcher has cited all the
information taken from external sources in Harvard format in this paper.
TO HAVE YOUR ASSIGNMENT DONE AT A LOW PRICE FOR A HIGH QUALITY PAPER, PLACE THIS ORDER OR A SIMILAR
ORDER WITH US NOW.