Statistical Models and Causal Inference: A Dialogue with the Social Sciences (review)

2010 ◽  
Vol 36 (4) ◽  
pp. 537-538
Author(s):  
Daniel J. Dutton
Keyword(s):  
Social Sciences ◽  
Causal Inference ◽  
Statistical Models ◽  
The Social

“All Models Are Wrong”

PMLA ◽  
2017 ◽  
Vol 132 (3) ◽  
pp. 668-673 ◽  
Author(s):  
Richard Jean So
Keyword(s):  
Social Sciences ◽  
Statistical Models ◽  
Test Score ◽  
Quantitative Model ◽  
Introductory Statistics ◽  
Social Scientists ◽  
Test Taking ◽  
The Social ◽  
Math Test ◽  
Statistical Graphic

Several years ago, the first thing i learned in my introductory statistics class was the following declaration, which the instructor had written in capital letters on the blackboard: “all models are wrong.” Models are statistical, graphic, or physical objects, and their primary quality is that they can be manipulated. Scientists and social scientists use them to think about the social or natural worlds and to represent those worlds in a simplified manner. Statistical models, which dominate the social sciences, particularly in economics, are typically equations with response and predictor variables. Specifically, a researcher seeks to understand some social phenomenon, such as the relation between students' scores on a math test and how many hours the students spent preparing for the exam. To predict or describe this relation, the researcher constructs a quantitative model with quantitative inputs (the number of hours each student spent studying) and outputs (each student's test score). The researcher hopes that the number of hours a student spent preparing for the exam will correlate with the student's score. If it does, this quantified relation can help describe the overall dynamics of test taking.

scholarly journals Subjective Causality and Counterfactuals in the Social Sciences: Toward an Ethnographic Causality?

2019 ◽  
pp. 004912411985237
Author(s):  
Peter Abell ◽  
Ofer Engel
Keyword(s):  
Social Sciences ◽  
Causal Inference ◽  
Statistical Inference ◽  
Observational Studies ◽  
Causal Explanation ◽  
Statistical Concept ◽  
Ethnographic Studies ◽  
The Social ◽  
Subjective Evidence ◽  
Subsequent Comparison

The article explores the role that subjective evidence of causality and associated counterfactuals and counterpotentials might play in the social sciences where comparative cases are scarce. This scarcity rules out statistical inference based upon frequencies and usually invites in-depth ethnographic studies. Thus, if causality is to be preserved in such situations, a conception of ethnographic causal inference is required. Ethnographic causality inverts the standard statistical concept of causal explanation in observational studies, whereby comparison and generalization, across a sample of cases, are both necessary prerequisites for any causal inference. Ethnographic causality allows, in contrast, for causal explanation prior to any subsequent comparison or generalization.

Studying Mechanisms To Strengthen Causal Inferences In Quantitative Research

2009 ◽  
Author(s):  
Peter Hedström
Keyword(s):  
Causal Inference ◽  
Political Science ◽  
Statistical Inference ◽  
Statistical Models ◽  
Quantitative Research ◽  
Social Outcomes ◽  
Causal Inferences ◽  
The Social

This article emphasizes various ways by which the study of mechanisms can make quantitative research more useful for causal inference. It concentrates on three aspects of the role of mechanisms in causal and statistical inference: how an understanding of the mechanisms at work can improve statistical inference by guiding the specification of the statistical models to be estimated; how mechanisms can strengthen causal inferences by improving our understanding of why individuals do what they do; and how mechanism-based models can strengthen causal inferences by showing why, acting as they do, individuals bring about the social outcomes they do. There has been a surge of interest in mechanism-based explanations, in political science as well as in sociology. Most of this work has been vital and valuable in that it has sought to clarify the distinctiveness of the approach and to apply it empirically.

A Note on the Scope of Developmental Behaviour Genetics

1995 ◽  
Vol 18 (4) ◽  
pp. 749-760 ◽  
Author(s):  
Conor V. Dolan ◽  
Peter C.M. Molenaar
Keyword(s):  
Social Sciences ◽  
Individual Differences ◽  
Human Development ◽  
Statistical Models ◽  
Gene Environment ◽  
The Social ◽  
The Subject ◽  
Behaviour Genetics ◽  
Behavioural Ontogeny ◽  
Human Ontogeny

The behaviour genetic decomposition of individual differences has been presented as being irrelevant to the study of human behavioural ontogeny. This introduces two problems. First, the analysis of systematic differences constitutes the basis for most statistical models used in the social sciences. If, generally speaking, this type of analysis is uninformative regarding development, how then can one empirically investigate human development? Second, behaviour genetic analyses are the only way to arrive at meaningful statements regarding the contributions of heredity and environment to human development. If results thus obtained are irrelevant, it is impossible to say anything on the subject of heredity, environment, and human ontogeny that is both meaningful and informative. It is argued that developmental behaviour genetics should not be viewed as a theory of development, but rather as a method of testing certain well-defined hypotheses regarding the contributions of genetic and environmental influences to human development. Individual differences assessed at any point in time reflect developmental processes prior to that time-gene-environment models are in a very basic sense inherently developmental... (Loehlin, 1975, p.41). Obviously the finding of innate differences in behaviour does not illuminate the development of that behaviour in any way (Johnston, 1988, p. 623).

scholarly journals Ontology, neural networks, and the social sciences

Synthese ◽  
2020 ◽  
Author(s):  
David Strohmaier
Keyword(s):  
Social Sciences ◽  
Neural Networks ◽  
Statistical Models ◽  
Internal Representation ◽  
Representation Learning ◽  
Machine Learning Techniques ◽  
Social Scientists ◽  
Social Phenomena ◽  
Ontological Commitments ◽  
The Social

AbstractThe ontology of social objects and facts remains a field of continued controversy. This situation complicates the life of social scientists who seek to make predictive models of social phenomena. For the purposes of modelling a social phenomenon, we would like to avoid having to make any controversial ontological commitments. The overwhelming majority of models in the social sciences, including statistical models, are built upon ontological assumptions that can be questioned. Recently, however, artificial neural networks (ANNs) have made their way into the social sciences, raising the question whether they can avoid controversial ontological assumptions. ANNs are largely distinguished from other statistical and machine learning techniques by being a representation-learning technique. That is, researchers can let the neural networks select which features of the data to use for internal representation instead of imposing their preconceptions. On this basis, I argue that neural networks can avoid ontological assumptions to a greater degree than common statistical models in the social sciences. I then go on, however, to establish that ANNs are not ontologically innocent either. The use of ANNs in the social sciences introduces ontological assumptions typically in at least two ways, via the input and via the architecture.

Sign in / Sign up

Export Citation Format

Share Document