Causality and causal modelling in the social sciences
•
0 likes•150 views
This document discusses causality and causal modelling in the social sciences. It addresses five philosophical questions about causality and five scientific problems in establishing causation. Causality is a complex concept that cannot be reduced to a single definition, and different causal methods use various concepts like probabilistic causality, counterfactuals, and concomitant variation. Causal models are used to detect relationships between variables, explain phenomena, and inform interventions. Epistemologically, causal discovery relies on reasoning about variations, differences, and changes to identify potential causes.
More Related Content
What's hot
What's hot (20)
Similar to Causality and causal modelling in the social sciences
Similar to Causality and causal modelling in the social sciences (20)
More from University of Amsterdam and University College London
More from University of Amsterdam and University College London (20)
Recently uploaded
Recently uploaded (20)
Causality and causal modelling in the social sciences
- 1. Causality and Causal Modelling in the Social Science Federica Russo Philosophy | Humanities | Amsterdam russofederica.wordpress.com | @federicarusso © Federica Russo
- 2. Staff for a New University @rethinkuva Rethink UvA 2
- 3. Overview Background: Causality – Philosophical Theory and Scientific Practice Causal assessment 5 philosophical questions; 5 scientific problems Methodology of causality Quantitative models Epistemology of causality Variational reasoning 3
- 4. PHILOSOPHICAL THEORY AND SCIENTIFIC PRACTICE 4
- 5. Causal assessment ‘What Causes What’ Different things we may want to establish: what’s the cause of a patient’s illness who is (legally / morally) responsible for some state of affairs what are the causes of unemployment what causes marriage dissolution or migration behaviour what causes dysfunction in an organisation which pathways explain some cellular behaviour … 5
- 6. Goals of causal analysis Knowledge-oriented Understand and explain a phenomenon of interest Action-oriented Predict, intervene on, control a phenomenon of interest Design / model / debug a system / environment 6
- 7. Do causes need to be causes? Consider: Smoking and cancer are associated. Should I quit smoking? Smoking causes cancer. Should I quit smoking? Causes trigger actions. Mere beliefs can’t, nor mere associations. 7 Source: http://xkcd.com/552/
- 8. Scientific practice first CitS / PSP / PI Philosophical questions about causation (and other topics) are motivated by methodological and practical problems in real science Start from scientific practice to bottom up philosophy Partly descriptive and partly normative No conceptual ‘straightjacket’ 8
- 9. 5 philosophical questions Metaphysics What is causality? What kind of things are causes and effects? Semantics What does it mean that C causes E? Epistemology What notions guide causal reasoning? How can we use C to explain E? Methodology How to establish whether C causes E? Or how much of C causes E? Use What to do once we know that C causes E? 5 scientific problems Inference Does C cause E? To what extent? Prediction What to expect if C does (not) cause E? Explanation How does C cause or prevent E? Control What factors to hold fixed to study the relation between C and E? Reasoning What considerations enter in establishing whether / how / to what extent C causes E? 9
- 11. How many concepts? Many! Causality Polysemic, thick concept Causal verbs Pulling, pushing, binding, … Causal methods Tracking what varies with what Understanding what produces what, and how, and when Different sources of evidence Evidence of difference making, of production … 11
- 12. 12 Causal pluralism: Causality cannot be reduced to one single concept but has to be analysed using several concepts
- 13. Inference, Prediction, Explanation, Control, Reasoning Causal Mosaic Metaphysics, Semantics, Epistemology, Methodology, Use Necessary and sufficient Levels Evidence Probabilis tic causality Counterfa ctuals Manipulat ion Invariance Exogeneit y Simpson’s Paradox Process Mechanis m Informati on Dispositio ns Regularity Variation Action Inference Validity Truth 13
- 15. Causal models Structural equation models, covariance structure models, contingency tables, multilevel models, regression models, Bayesian networks, potential outcome models, quasi-experimental models, spatial models … Quantitative models: statistical models 15
- 16. ‘Statistical’ causes Gather a large number of observations, organise them in variables E.g. socio-biological characteristics (exposure) and cancer rates (disease) Study the (in)dependencies between variables, robustness and stability of correlations Establish stable patterns of (in)dependencies to identify risk factors and possible interventions 16
- 17. What does a causal model do? A causal model: models the properties of a (social) system detects (causal) relations between the properties of the system explains the functioning of the system through its causes 17
- 18. 18 54 4 13 34 12 2 X1 Economic development X2 Social development X3 Sanitary infrastructures X4 Use of sanitary infrastructures X5 Age structure Y Mortality 4 5 54 3 34 4 3 1 13 3 2 1 12 2 1 4 4 2 2 X X X X X X X X X Y
- 19. Structural equations and explanation Y=X+ X, Y : explanatory and response variables Xs explain Y Xs are relevant causal factors in the causal mechanism 19
- 21. Causal discovery is reasoning about variations. To establish causes we need difference. 21
- 22. ‘Statistical’ variations “Gather data about socio-economic status, occupation, diet, smoking behaviour and see how steadily they are associated with cancer” Study how variations in exposure are related to variations in disease How different levels of exposure change the probability of disease Statistical reasoning: search for those factors explaining the variance of the outcome 22
- 23. FOUNDATIONS 23
- 24. Variations in Mill Agreement: comparing different instances in which the phenomenon occurs. Difference: comparing instances in which the phenomenon does occur with similar instances in which it does not. Residues: subducting from any given phenomenon all the portions which can be assigned to known causes, the remainder will be the effect of the antecedents which had been overlooked or of which the effect was as yet an un-known quantity. Concomitant Variation: in presence of permanent causes or indestructible natural agents that are impossible either to exclude or to isolate, we can neither hinder them from being present nor contrive that they shall be present alone. Comparison between concomitant variations will enable us to detect the causes. Mill (1843), System of Logic The experimental method is based on the Baconian rule of varying the circumstances The Four Methods are all based on the evaluation of variations 24
- 25. Variations in Durkheim Durkheim (1897), Le suicide A study into the variability of suicide rate A search for the causes making suicide rate vary Durkheim (1885), Les règles de la méthode sociologique The method of concomitant variations makes sociology scientific 25
- 26. Learning ‘ordinary’ causes Humean regularity Instances of smoke follow instances of fire Can’t establish logical, necessary link Create expectation, project causal belief onto the future Epistemology of causal modelling seems to be at variance with the Humean account See next 26
- 27. Learning ‘scientific’ causes Causal discovery (experiments, statistics) Search for differences Explaining differences Variation, difference, comes first 27
- 28. Regularity too Statistical regularity Causal methodology needs regularity as a constraint on variations, differences Scientific causes are ‘generic’ Population-level, repeatable Hence we need regularity to establish generic level 28
- 29. TO SUM UP AND CONCLUDE 29
- 30. Inference, Prediction, Explanation, Control, Reasoning Causal Mosaic Metaphysics, Semantics, Epistemology, Methodology, Use Necessary and sufficient Levels Evidence Probabilis tic causality Counterfa ctuals Manipulat ion Invariance Exogeneit y Simpson’s Paradox Process Mechanism Informati on Dispositio ns Regularity Variation Action Inference Validity Truth 30
- 31. Philosophical theory meets scientific practice Scientific practice first Then, be precise about your question, target specific scientific challenges Causal pluralism, in the form of ‘causal mosaic’, is a sophisticated philosophical view Methodology of causality is rich and diverse Choose the method best adapted to your problem Epistemology of causality is about how we find out about causes Variation guides causal reasoning in various forms 31