[MUSIC]

Hi, in this module I'm going to introduce the methodology or

methods section of a research proposal.

Every research proposal must describe the methodology that

is to be employed in the study.

And, in particular, the discussion should focus on how challenges of the sort

that we talked about in part one of this class will be addressed by the methods.

Remember that in part one we talked about a lot of the challenges

to drawing conclusions in a social science study.

Problems with omitted variables, problems with reverse causality.

And we have to convince our audience that we can actually

address these problems by whatever methodology we are going to employ.

This is also where by referring back to the literature or

background section we will highlight any differences between

the methods that we propose to employ and

the ones that have been proposed or used in other studies.

We may often provide examples of successful studies that

use the methodologies that we propose to employ.

Studies that perhaps were carried out in other contexts,

this is a way of showcasing the usefulness of the methods that we proposed to employ.

Now this basic goals for the methodology section

hold, regardless of whether the study is qualitative or quantitative.

So a few basic questions for the methodology section to address.

One is how well the methods in the propose study

improve on those in previous studies?

What are the known limitations of the methods, and

why are they not issues in the specific case at hand.

Almost any research methodology that we can imagine qualitative or

quantitative has known limitations.

And we have to explain why we think that it's safe to

not worry about them in the study that we're about to conduct.

And then we have to talk about how the methods that we are going to

use address or deal with known limitations of the data.

So when we talking about the data,

we maybe noting again we'll come back to this in the next module.

But our discussion of the data may actually identify problems with the data

certain kind of people omitted, certain kinds of events or information left out.

And here is where we have to anticipate those problems,

and talk about how the methods that we're going to employ will overcome those

problems or at least not be affected by them.

And most of the time we would like to talk in our section on

methodology about how the methods that we hope to employ will

help us make some claims about cause and effect.

As we learned about in part one there are specific methods that

source of science researches are often use to try to come with decisive or

conclusive findings of cause and effect relationships.

And we may talk about those in this section.

And finally, we have to justify

our choice of methods over other possible choices of methods.

So it's not typically sufficient in a discussion of methodology simply to

lay out exactly what we plan to do.

But we have to again explain why we

chose to use that particular approach as opposed to other possible approaches.

For any problem that we might be interested in, there are probably multiple

approaches, qualitative or quantitative, that could be used to study the problem.

And we have to convince our audience that whatever we have chosen to do

Is the best fit for the task at hand.

Now if we're talking about quantitative methods there's a few

specific things we have to keep in mind when we write our methodology section.

We have to talk about what our outcome variable is.

Again, the earlier parts of the proposal, especially the discussion of theory

may have talked in very general terms about relationships.

And the statement of the hypothesis may have suggested what the indicators or

what variables would be used to measure the concepts

that we were trying to relate to each other.

But here is where we might go into detail On the outcome variable.

So when a study of education, and health we might describe now in considerable

detail the measure that we are going to use in our, later when we get to our data.

The measure that we're going to use to describe health.

We may also here discuss the control variables that we are going to include in

our analysis if we are running what is referred to as a regression analysis.

So if you take an introductory statistics course, you'll almost certainly be exposed

to regression and you'll know that you have to choose control variables.

We talked about this in part one.

To be included on the right-hand side of your estimation.

This is where, in general terms,

you'll talk about the variables that you want to control for in your model.

You'll justify your choice of control variables.

And then, having laid out our outcome variables and our control variables,

we have to explain why they comply with basic assumptions

that are required by the statistical models that we may be implying.

Again, if you take a basic or

advanced course on statistical regression you'll know that for

the regression to work for the models to be estimated reliably,

certain assumptions have to hold about the distributions and

other characteristics of the outcome and the right hand side variables.

And this is where we have to acknowledge any problems that might be there

and essentially argue that they are not a problem.

So if the variables don't appear to comply with our restrictions or our restrictions

or assumptions, then we have to convince the audience that this is not a concern.

And finally this is where we will talk about how different possible

patterns of results might be consistent with different possible hypothesis.

So this is where we lay out, in terms of out actual indicators,

our actual measures, what relationships we expect to see in the real world,

in our data, in detail if our hypothesis are actually correct.

For qualitative research, the method searching for

the research proposal is just as important.

There are several goals of the method section

when we're talking about qualitative research.

One is to introduce the criteria for the selection of the cases.

Now remember back in literature or background section,

we talked about whether not the cases were representative.

But here is actually were you try justify the selection of those cases.

By referring back to the literature and background section, and

try to talk about how these particular cases will give us some insights

into perhaps broader or wider diversity of contexts.

So for example if we've selected a village or a set of villages or

perhaps neighborhoods in which to conduct our qualitative research.

And in the literature in the background section we've already

talked about the issue of representativeness and

we've identified distinguishing feature of these settings.

Here is where we actually try to argue that by studying these particular

settings these villages, these neighborhoods, etc.

We can actually learn something about.

Other context beyond again specific ones that we're studying.

Another important feature of the method section in a proposal for

qualitative research.

If multiple cases are to be considered

is the discussion of how the cases are similar, and how they are different.

Because in qualitative research, quite often the approach is to

compare cases, and look for similarities and differences.

Qualitative resources that involves the comparison of cases quite often seeks

cases that are as similar as possible on dimensions

not relevant to the focus of the study but differ

on the dimensions that are of interest for the purposes of testing our theory.

So, for example, if we are conducting studies of villages in order to try to

understand the implications of particular forms of family organization,

particular marriage customs.

We might want to select a variety of villages that

are broadly similar in terms of other dimensions.

Their geographic context, climate crops and so forth and try to

pick some villages that differ primarily in terms of their marriage customs.

And then through our field work, we compare these villages and

then try to elucidate the implications for the differences in marriage customs

for other features of village social organization that we're interested in.

Hopefully, that gives us some insight into the implications of marriage

customs more generally for social or economic organization.

We'll also in this section talk about what the criteria are for

comparing cases and reaching conclusions.

When it comes to qualitative research,

there is a great diversity of methodologies that scholars employ for

conducting research just as there is for quantitative research.

So here is where you need to identify the Specific type

of qualitative methodology that you plan to employ and

justify that choice of methodology over other possible choices.

So here I've tried to sketch out a list of very broad terms what the key

goals of the methodology section of a research proposal are.

Basically the most important goal is to convince the audience that

you have thought very carefully about your choice of method and

that it is the best possible method, the best fit for the particular task at hand.

1.6 Methodology

Social Science Approaches to the Study of Chinese Society Part 2

Rencontrer les enseignants