Quality & Quantity 32: 181–200, 1998.
© 1998 Kluwer Academic Publishers. Printed in the Netherlands.
Rediscovering and Confronting Critical
Ambiguities in the Determination of Causality
Department of Sociology, University of Calgary, Calgary, Alberta T2N 1N4, Canada
Abstract. The demonstration of non-spuriousness is both critical in making causal statements and
extremely difﬁcult. Unfortunately this issue is often summarily dealt with in methodological treat-
ments in which spuriousness is reduced to instances of “common cause”. It is argued here, among
other things, that extraneous variables need not be prior to the independent nor necessarily causally
related to either independent or dependent variable. This being so, spuriousness is a far more common
theoretical problem than is often currently acknowledged, and we might do well today to listen again
to the advice of an earlier generation of sociologists whose work on the topic deserves more careful
attention than it appears to have received.
Key words: causality, spuriousness.
The topic of causality occupies a central place in most methods treatments. At
the same time, many sociologists are ambivalent about using the term, particularly
in nonexperimental situations, since causal relations are difﬁcult to demonstrate.
Whatever the context, it is commonly agreed upon that there are three (and per-
haps four) basic criteria for the establishment of a causal relationship.
association, time order,
non- spuriousness and rationale. While there appears to
be no real argument concerning the importance of the last of these, in some treat-
ments it appears to enter through the back door, as it were, in the establishment of
relationships as non-spurious.
Of these criteria, there is little question but that the demonstration of non-
spuriousness is the most difﬁcult for students and practitioners alike, and yet it
is critical, since the charge of spuriousness amounts to claiming that though a bi-
variate relationship displays association and the appropriate time order, and though
the investigator may have a plausible rationale for its interpretation as causal, it is
nonetheless the case that the inference or interpretation of causality is incorrect
and the variables in question are only accidentally correlated. Since a great deal
of our enterprise involves the discovery of the ’causes’ of variation in human be-
haviour, such a ﬁnding is of critical importance to the development of theory and
to our substantive knowledge in any subﬁeld of the discipline. Scientiﬁc claims
normally gain credibility precisely as they withstand plausible rival hypotheses.
These competing hypotheses identify what Denzin (1989: 20-6) calls rival causal
factors “that may have distorted, or in fact, caused the observed relationship”. Yet,