Summary methodology

Our project consists of 6 parts, which are:
- Introduction
- Literature review /lí tơ ơ chơ/
-methodology
- Fieldwork and findings
- Analysis and discussion
- Conclusion
And we're halfway there, we're in the middle, the very important part
of the project - the methodology.
Once you have decided on your general approach to answering your question,
you can think about the ‘scaffold’ or ‘design frame’ within which you will plan
and execute your research.
This chapter considers:
• some general issues in designing a piece of research;
• the main ‘frames’ to guide the way that you carry out research:
● action research
● case study
● comparative research

, ● ethnography
● evaluation /ɪˈvæl.ju.eɪt/
● experiment
● longitudinal/ˌlɑːn.dʒəˈtuː.dɪ.nəl/ , cross-sectional studies and survey
● no design frame.
And what is design? What do you know about design?
Design is about plan and structure, the whole programme /ˈproʊ.ɡræm/ of
your research, from purposes /ˈpɝː.pəsi/ to execution/ eks sơ cíu sần/ ,
constitutes/ˈkɑːn.stə.tuːt/ the design.
One particular part of the sequence of decisions you make will be about what I
am here calling the design frame.
And I call it like that because it constitutes the most important element in the
way that your research is structured: it’s like a chassis that supports your
research. The design frame provides the superstructure for your research –
connecting purposes with questions with the ways in which data can be
collected – though it does not prescribe how the data /ˈdeɪ.tə/ will be
collected.
I will be concentrating on seven types of design frame here. There are others,
but these seven are the most common structures used in small research
projects. They are:
• action research
• case study
• comparative research
• ethnography
• evaluation
• experiment
• longitudinal, cross-sectional studies and survey.
And there's something I need to tell you. It is important to reiterate that these
frames are not designs for research in themselves. The design itself is the plan
for research that you adopt from the beginning of your project.
The design frames are scaffolds within which to structure what you do. But it is
important also to say that these design frames are not in any way similar as
structures, nor are they mutually exclusive . They can exist in combination. So,
for example, action research could take shape in most of the other forms, or a
case study could include a survey. This will become clearer as we look at the
frames themselves.

,The idea of research design per se are social research that are expected to be
as
similar as possible to natural scientific research. It would be mainly
experimental and it would come complete with very specific instructions on
procedures, methods and apparatus, with the idea that anyone could come
along
after you and repeat the experiment you were doing.
In short and easy to understand:


The design frame is like a scaffold that holds your research in shape and
helps to structure it. Many different kinds of design (or ‘scaffold’) are
possible. The design frame you choose will be the one that best helps you to
answer your research question.


The design, it implies, should not be set in stone, ready to be replicated exactly
by the next researcher. Given that this is the case, some have spoken about
emergent /ɪˈmɜr·dʒənt/ design– in other words, letting the design ‘happen’ as
you find out more about the situation in which you are interested. This idea of
emergent design is an important one for social research in the interpretative
/ɪnˈtɝː.prə.tɪv/
̬ tradition, and though the word ‘design’ is still used, it really
turns the idea of ‘design’ on its head, since something that ‘emerges’ cannot
be ‘designed’. so, We should perhaps look for a new word for the process.
The trouble with using ‘design’ is that it implies all of the traditional features of
experimental design that I have talked about (e.g. specification /ˌspes.ə.fə
ˈkeɪ.ʃən/ of sample, apparatus /ˌæp.əˈreɪ.təs/ and so on), and these features
carry with them other expectations. There are, for example, expectations
about sample size (the bigger the better), reliability /rɪˌlaɪ.əˈbɪl.ə.t̬i/ (you have
to be sure of getting the same result if you do the same again) and validity/və
ˈlɪd.ə.ti/
̬ (you have to be sure that you are finding out what you set out to
find). But these are not the ground rules for interpretative research.
---->So, the word ‘design’ should be interpreted with caution in social research.
In certain kinds of research it will be more fixed; in others less so. Expectations
about it in one kind of research will not always apply in another.

Some general issues in design

-The word ‘design’ is something of a misnomer /ˌmɪsˈnoʊ.mɚ/ and leads
people to have cast-iron expectations about the structure of research, ignoring

, the tenets /ˈten.ɪt/ of different types of research – the presuppositions that
ground it.
-There can be no expectation that the ground rules and methods of one kind of
research will be appropriate in another.

Sampling
khái niệm the notion of sampling really belongs in experimental research and
research that seeks relationships among variables.
It means you are taking your group or groups for your research from a
manageable sample which is representative of a larger population.
The sample is truly representative of this wider population, the findings of
your well-designed research can then be generalised to the population.

example tự suy nghĩ ra : I will give an example to make it easier for you to
understand. Let's say America's young population is 20 million, and you'll only
do 5 million of those 20 million, and in the end you come to the general
conclusion, you don't have to experiment on all of them. 20 million people.
and to see if i took this example correctly go to the next section


The sample may not be representative: there may be selection bias of one
kind or another. There are many ways in which one can ensure /ɪnˈʃʊr/ that the
sample is representative of this wider population. One is by taking a random
sample. This is just what it says on the tin: it is a sample that is taken – by a
random process, in the same way that names could be picked from a hat.
However, it is not good enough for the sample just to be randomly taken.
Example: If you took as a random sample of university students the first dozen
people you found in the bar on Wednesday night, your sample would be
susceptible /səˈsep.tə.bəl/ to various sources of distortion – how do you know
that bar-dwellers/ˈdwel.ɚ/ are representative of the student population
generally? It may be that those who go to the bar are less likely to go to the
library, or that they tend to represent one ethnic or religious group more than
another. That is why this kind of sample (the first dozen in the bar) is called a
convenience sample, and why a convenience sample has many problems
associated with it if you are expecting to generalise/ˈdʒen.ər.əl.aɪz/ from it. To
take a true random sample, you would have to ensure that you were drawing a