Dont forget to think Middle range theory as

Don't forget to think! Middle range theory as causal analysis Ray Pawson ESRC Research Methods Festival 2008

The basic thesis Causal inferences are made on the basis of and protected by sound research technique. Good design and analysis, such as found in RCTs and multivariate analysis are the foundation stones of our ability to begin to make causal claims. Our ability to detect causal powers lies in the process of theory testing and refinement. It is only by harnessing research designs that enhance the precision and refine the scope of theories that social science can begin to make valid, reliable and useful causal inferences.

Successionist causal logic (variable based) X Y O 1 X O 2 O 3 W O 4 X U V U Y Z V X Y

Generative causal logic (mechanism based) Context (C) Mechanism (M) Outcome Pattern (O)

Problems with variable analysis – an early sighting ‘I usually assign students in a theory class the following task: Choose any relation between two or more variables which you are interested in; invent at least three theories, not known to be false, which might explain these relations; choosing appropriate indicators, derive at least three empirical consequences from each theory, such that the factual consequences distinguish among theories. This I take to be the model of social theorizing as a practical scientific activity. A student who has difficulty thinking of at least three sensible explanations for any correlation that he is interested in should probably choose another profession. ’ Stinchcombe (1968: 13)

Correlations always yield to multiple explanations N potential explanations, smuggled in ad hoc and turning on concepts that cannot be expressed as variables X Y

Partial / Multiple Correlations always yield to multiple explanations . 26 Each path has n potential explanations, usually smuggled in ad hoc. The estimates will change with the inclusion of further variable, differently measured on different school programmes achievement. 25 . 08 mentoring. 22 quality of parental relationship . 25 self worth grades . 22 -18 -. 09 . 26 value of school skipping school -. 11 -. 26

Generative explanation in practice How Merton turned Stouffer’s Causal Arrow into an Generative Mechanism Giant survey of draftees into the US army WW II. Several pronounced ‘causal links’ including: (Marital Status Attitude Draft) 41% of married inductees claim thy should not have been drafted – 10% for single men. On which basis Stouffer et al declare: ‘Comparing himself with his unmarried associates in the Army, he could feel that induction demanded greater sacrifices from him than from them; and comparing himself with his married civilian friends, he could feel that he had been called on for sacrifices which they were escaping altogether. ’

Theories that might usefully be tested to strengthen the causal inference If ‘relative deprivation’ really is the cause we need to investigate: 1. Experience – that they feel ‘torn’ (find it hard to partake in family life whilst soldiering). 2. Norms – they know ‘how it works’ (norms whereby recruitment boards are usually softer on the married). 3. Knowledgeability – they know the ‘score’ (comparative rate of induction between singles and married). 4. Proximity – resentment increases with ‘immediacy’ (being surrounded by single in barracks and drills) N. B. Testing each conjecture would call on different strategy (mixed method) BUT N. N. B. theory leads method (mixed method alone is just description).

True causal explanations have memory and pedigree ‘An army private bucking for promotion may only in a narrow and theoretically superficial sense be regarded as engaging in behaviour different from that of an immigrant assimilating the values of a native group, or of a lower-middle-class individual conforming to his conception of upper-middleclass patterns of behaviour, or of a boy in a slum area orienting himself to the values of a settlement house worker rather than the values of the street corner gang, or of a Bennington student abandoning the conservative beliefs of her parents to adopt the more liberal ideas of her college associates, or of a lower-class Catholic departing from the pattern of his in-group by casting a Republican vote, or of a eighteenth century French aristocrat aligning himself with a revolutionary group of the time … The combination of elements may differ, thus giving rise to overtly distinctive forms of behaviour, but these may nevertheless be only different expressions of similar processes under different conditions. They may all represent cases of individuals

Another benefit of theory-testing approach The precious property of abstraction Abstraction and formalisation of a generative middlerange theory The ‘demi-regs’ of study one. The ‘demi-regs’ of study two.

Why the generative account is to be preferred • Ontologically sound – variables/attributes don’t ‘cause’ anything. • Makes tacit explicit – other approaches smuggle in ad hoc generative explanation. • Deals with constrained choice – intervening variables are not mechanisms, moderators are not contexts. • Provides explanations rather than makeshift descriptions • Deals with Lieberson’s X 14 problem – closure by theoretical saturation and adjudication. • Provides the basis for transferable and cumulative explanation.

Memo to ESRC • Data never speak for themselves • Data analysis should not be confused with explanation • Data agglomeration and technical refinement produce more complex descriptions • Descriptions are ten-for-a-penny • Don’t forget to ask researchers to think!