The Practice of Social Research

Chapter Four.  Research Design: Causation

THE LOGIC OF NOMOTHETIC EXPLANATION
    Criteria for Nomothetic Causality
    False Criteria for Nomothetic Causality
    Necessary and Sufficient Causes

 

    This section of the chapter deals with the topic of causation: causal relationships among variables.  We begin with the three fundamental criteria for us to conclude that one variable causes another, such as saying that increased age causes greater religious involvement.

    First there must be an empirical correlation between the two variables.  If young and old people were equally religious, then we couldn't say there was a causal relationship, since there's no relationship at all.

    The second criteria has to do with time order: the cause has to happen before the effect.  In our present example, age is logically prior to religious involvement.  We can imagine that something about growing older could make someone more religious, but your level of religiosity can't change your age.

    Finally, there's no third variable that can explain away the original relationship as spurious or non-genuine.  Several examples in the chapter will illustrate this possibility.

    Having looked at the criteria for causal relationships, we'll look at some false criteria that sometimes confuse students.  For example, there does not need to be a 100% relationship between two variables for us to conclude a causal relationship occurs.  That is, contradictory cases, such as an elderly atheist or young priest, do not disprove the overall relationship.

    In this context, we'll see the difference between necessary causes, which must be present for the effect to exist, and sufficient causes, which also cause the effect to exist.