Research methodology for the social sciences

Methodology is the study of methods - in other words, methodology is the science of understanding what method of conducting research is best applicable in a certain situation.

This writeup is concerned only with the social sciences: Science that cannot explicitly be expressed in numbers.

If you, for example, were to find out how many cows there are inWales, interviewing people on the streets of Tokyo is probably a bad idea. Calling and talking to all the farmers or traveling around and counting them yourself, would probably be better ideas.

Methodology, then, is the science of evaluating the different methods you have at your disposition. Methodology also allows you to ascertain that the things you are trying to research correspond with the results you are looking for, and it evaluates the validity of the methods used.

An example can be the "cut the legs off a frog, and it turns deaf" experiment: Imagine that you have a highly trained frog. When you tell it to jump, it jumps. Now, cut all the legs off the frog. Tell it to jump. It will not budge. Hence, the frog has gone deaf. Of course, everybody understands that this is a joke and makes no sense. You would be surprised, however, how often people draw the wrong conclusions based on their (otherwise valid) research material.


When doing research, the methodology should be one of the first things to think about. If you want to find out a particular thing, what is the easiest way to find out about it? Is it also the best way? Is it provable? Is it repeatable? Are there other ways to research this particular question, and if yes, do those methods bring the same answer? If no, why not?

Objectivity in methodology and research

Scientific methodology usually strives towards objectivity, by straining towards the ideal of nonbiased questions and information. Supporters of the theory of objectivity say that if you have objective research data, then you also have irrefutable evidence which can be the groundwork for debate.

When writing essays, for example, one would often do research by reading published works by renown authors. Your essay gains credibility through the words borrowed from those authors, and automatically accepts those words as being "the truth" (usually, you would quote things that prove your points). However, if the works you base your essays on turn out to be false, misleading, or erronous, your essay becomes worthless.

Those who oppose the theory of objectivity claim that Objectivity is impossible to achieve, because all research starts with a goal; Something that must be found out. They say that if there is no objectivity, this means that so-called "objective" scientific research leads to a false sense of irrefutable evidence.

Quatity vs Quality.

When we are talking about research, we often talk about quantity and / or quantifiable research data. This is research which you can use in calculations and statistics: How many people were born on May 21, 1981? How often does the average person go to the bathroom? What percentage of people who buy the newspaper The Sun are smokers? Quantifiable research data does not necessarily become better with the number of subjects or cases, but a wide spread within your target group is important. If you try to find out how many people in the UK drink alcohol every weekend, there is no use in only asking pensioners or only asking students, because the results will be skewed. Quantitative research is great for inanimate objects and statistics. Also, it tends to limit the researcher to preconceived options, and does not allow for emotions.

In qualitative research data, we are looking for something differently altogether - we are looking for the reasons why something happens. In the examples above: Let's say that 7 people in Liverpool were born on May 21, 1981, a number far lower than average. You would then speak to the parents of those 7, to find out why the children were born on May 21nd. Were they born too early? Too late? Did the parents specifically decide to conceive a child in August, so the child would be born in May? What doctor supervised the childbirth, and did he / she do anything in particular so the childbirth did not happen on May 21? All these questions cannot be answered through numbers in statistics.

The way of gathering qualitative research input is through interviews or open questionnaires. ("why do you buy the Sun?" rather than "how often do you buy the Sun").

The problem with qualitative research is that the research is not always provable and never repeatable. Also, the researcher often becomes more involved in the process. Neither of which necessarily makes it less valid, but it does make an important factor in how serious the research is perceived.

Large items of social science research are often backed by both qualitative and quantitative research. If the results are used combined in the same study, a new set of results appear through triangulation.

Schools of thought in methodology

There are several schools of thought in methodology, when applied to the social sciences:


The posivistic approach to methodology is that people react similarly, when put in different situations. Positivism tries to explain things through the use of "universal laws of behaviour" by using "numerically defined and quantifiable measures"

You can say what you will about positivism, but personally, I believe it can never work. People react differently to the same situation. Cases in point:

1) You are sitting in your sofa at home, watching television. Suddenly you hear a loud bang on your front door. What would you do? Can you think of anybody you know who would react differently?

2) You are standing in a field, picking flowers. Suddenly a cow comes up to you and starts eating grass right beside you. What would you do? Can you think of anybody you know who would react differently?

3) You are standing in the street late at night, at a cashpoint (or ATM). You are about to put your card into the machine, when two shady-looking fellows appear at the end of the street, and start walking in your general direction. What would you do? Can you think of anybody you know who would react differently?

I cannot predict what you would do in the different scenarios, but I am very sure that if you think about it, you can find somebody who would react differently from the way you react.

However, positivism is quite commonly used: It is often used in science and medicine, where external forces are used. Also, positivism has the distinct advantage of offering you quantifiable research data in fields where you would normally only get qualitative data. In the example with the cow, above, for example, instead of asking "What would you do?", we could give a subject the options a) run away, b) move away from the cow, c) just continue what you were doing or d) move closer to the cow. Although this allows you some nuance in your research, it is not precise; Some people might want to attack the cow, or ride on its back. But if you do not specifically ask those questions, people are not going to answer that. If you do ask those questions, chances are that people will fill in "attack the cow", just because it is a funny thing to do.


Empiricism (nothing to do with empirical, which is the gathering of evidence, I believe) is the belief that there are facts which we can gather in the social sciences, independently of how people interpret them.


The Realism approach to studying social sciences argues that knowledge (under which emotional and upbringing) has an effect on the way people perceive and act on matters in life. This automatically means that research dealing with people on a deeper level (beyond what you can actually count, in other words) cannot be expressed in numbers. Two of the people who follow the school of realism are Karl Marx and Sigmund Freud.

Social researchers should investigate the underlying mechanisms which make social action possible in the first place (Marx)

Our consciousness is determined by our subconciousness, and "culture is reproduced through a repressive structuring of unconscious passions" (Freud)


Interpretivist research disregards the external and observable to a large degree, and will instead deal with the internal factors. It is often used in social sciences, and is the most effective way to get qualitative research; Human emotion and motivation. See also "Qualitative research", further up in this writeup.

Interpretivism concentrates on the individual, and assumes that all people are different.


The last model of methodology is called Postmodernism. It rejects the possibility of science in the social world altogether, and sees both qualitative and quantitative models as invalid. Instead, postmodern research is purely descriptive, and attempts to "deconstruct surface appearances to reveal the hidden structure"

If you are a postmodern researcher, please /msg me, cause I don't understand a fucking thing of what you are trying to do.

but the E2 community knows its stuff.. From CtF:Postmodernism is far more complex and at the same time not nearly as complex as you've got here. Unfortunately it's not a methodology so much as a movement, and therefore doesn't have a coherent methodology. Some of it, no doubt, is what you say; but not mostly. Louis Althusser is not so much like Michel Foucault, for example. I can only comment on Michel Foucault, Gilles Deleuze, and Félix Guattari: Their methods are to dissect "universal" categories like class; they reject most aggregates as generalizations. Instead, they focus on specific interpersonal power relations (between, say, me and you, between this manager and this employee, between this confessor and this penitent, between this doctor and this patient, etc etc) in order to understand what factors go into the determination of each's goals and participation in whatever aggregate groups *they* might associate with.

Further reading:

Clough and Nutbrown (2002) A Student’s Guide to Methodology London: Sage
Gunter, B Media Research Methods London: Sage
May, T (2002) Social Research London: OUP