Social Research Glossary

Social Research Glossary

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z Home

Citation reference: Harvey, L., 2012-24, Social Research Glossary, Quality Research International, http://www.qualityresearchinternational.com/socialresearch/

This is a dynamic glossary and the author would welcome any e-mail suggestions for additions or amendments. Page updated 8 January, 2024 , © Lee Harvey 2012–2024.

_________________________________________________________________

Cause

core definition

A cause in social research is the identified construct (variable, event) that brings about a subsequent outcome (usually referred to as the effect).

explanatory context

An example of a supposed causal relationship is that an earthquake off the coast of Japan caused a sunami that caused the deaths of very many people and the destruction of property which caused horrendous suffereing for the Japnaese people and caused a significant problem for the Japanese economy. The problem with this causal chain is where does it start and end? Can the the problem in the Japanese economy be said to have been caused by a shift in the earth's crust? The problem of reductionism in causal relations is particularly acute in social research where so many varibales are likley to intervene in any given event.

The concept of causality has been the subject of much debate and disagreement. However, it has at its centre the idea of 'producing', that is, if X is a cause of Y we mean that a change in X produces a change in Y and not merely that a change in X is followed by or associated with a change in Y.

A cause (as a noun) is thus the event that precipitates a succeeding event (usually called the effect).

[The notion of a 'cause' in the sense of a worthy idea or organisation worth supporting, such as 'Helping homeless people is a good cause' or 'Oxfam is a good cause' is not considerd here].

Essential to most notions of causality is the idea of constant conjunction. This is that whenever a cause of any given phenomenon occurs then its effect always follows.

However, constant conjunction alone does not imply a causal mechanism. Indeed, constant conjuction was a term for perfect positive correlation used by eighteenth century philosophers who did not want to imply a causal mechanism.

In practice, in social research, the idea of association is taken as a pragmatic indicator of causality. However, although the idea of association or constant conjunction may be part of the definition of causality, it is not sufficient to distinguish a causal relationship. Causation is not something that is directly observable in the real world.

However, some argue that it is possible to support a causal statement by observing variations in the dependent variable associated with different values of the independent variable (covariation): observing in some way that it is not possible to change the value of the independent variable by changing the dependent variable (casual direction): and by observing that there are not other variables in the environment which might cause changes in the dependent variable which changes at the same time as the independent variable changes (non-spuriousness). Such an approach is the basis of the quantitative social research tradition.

Thus quantitative social researchers looking for indications of causal relations use the following criteria to suggest the presence of causal relationships:

a. association between X (or Xs) and Y

b. time priority (X occurs before Y)

c. non-spuriousness (other variables do not explain away the relationship between X & Y)

d. theoretical soundness (the causal relationship makes theoretical sense)

It has been suggested that because of the problems of relating the notion of causality to the real world, one should confine the notion to hypothetical models or simplifications of reality. A causal model starts with a finite number of specific variables and then attempts to relate them in a meaningful way and to set causal relationships between them. The adequacy of the model is tested against the data.

It should be noted that causality is a working assumption of the scientist but is not something which is directly observable. For this reason some social scientists believe it borders on the mystical and that it should not therefore be used as a guiding principle in social science reseach. They have, in other words, a different methodological committment.

analytical review

Psychology Glossary (undated) states:

Cause and Effect: Cause and effect is the panacea for researchers. It is a term you've likely heard many times already, or will hear many times as you study psychology. Psychologists look to identify the "effect" one variable has on other variables...does one variable "cause" other variables to change. Establishing cause and effect is not easy and requires researchers to conduct studies that not only follow the scientific method, but also classify as "true experiments" (studies that include random assignment of participants to groups and manipulation of an independent variable). For example, can a psychologist truly say that smoking causes cancer? Can we conduct a true experiment on the effects of smoking...can you randomly assign participants to smoking and non-smoking groups, have them smoke or not smoke for a period of time and then measure the growth of cancer? You could, but not ethically. As a result, you can't establish a cause and effect relationship...you can establish that one variable (smoking) influences the other (cancer) using a correlational approach, but you really can't establish cause and effect.

Colorado State University (1993–2013) offers the following definitions:

Causal Model: A model which represents a causal relationship between two variables.
Causal Relationship: The relationship established that shows that an independent variable, and nothing else, causes a change in a dependent variable. Establishes, also, how much of a change is shown in the dependent variable.
Causality: The relation between cause and effect.

Elwell's Glossary of Sociology (undated) defines causation as:

A 'cause and effect' relationship exists wherever a change in one variable (the independent variable) induces change in another (the dependent variable). Causal factors in sociology include individual motivation as well as many external influences on human behavior that often go unrecognized.

New World Encyclopedia contributors (2017):

Causality is one of the central notions in our conception of the world. We think of the things and events we experience as connected, and causal relations between them is perhaps the most important connection. Thoughts of causation are central to how we think about our own actions, thoughts, responsibilities and relationships. Yet, however common the notion of causation is in our thoughts, it proves to be very mysterious once we focus on it in an attempt to gain a better understanding of it.

As a number of philosophers (including Nicolas Malebranche and David Hume) emphasized, it appears that we do not have any direct experience of causation. We see events follow one another, and often conclude that one caused the other, but it is not as though we perceive some sort of force passing from one to the other. Even in the case of our own actions, we are strikingly unable to explain how a decision to ask a question brings about motion in our bodies.

Even setting such epistemological problems aside, we might wonder what it is we are saying when we say that one thing causes another. We seem to be saying more than that one thing follows the other, but what is that something more? It appears to involve some thought about other events of a similar type (for instance, in saying that a thrown rock broke a window, we appear to commit ourselves to the belief that there is some sort of general connection between throwing rocks and breaking windows), but it is not clear what that involvement amounts to.

associated issues

related areas