What question did this study set out to answer?

Define a coherent framework for data science tasks to prevent the conflation of associative prediction with causal inference.

January 2, 2019CHANCEOpen Access

A Second Chance to Get Causal Inference Right: A Classification of Data Science Tasks

View Full Paper

Ask AI

Bookmark

Discussion

Member takes

Implication

Methodological review demonstrates a three-part classification of data science tasks in analytics, highlighting the critical role of counterfactual prediction for sound causal inference.

Key Points

Define a coherent framework for data science tasks to prevent the conflation of associative prediction with causal inference.
Conceptual classification dividing quantitative data science into description, prediction, and counterfactual prediction.
Methodological review evaluating historical approaches to causal reasoning in statistics, epidemiology, and machine learning.
Delineates data science into three distinct tasks—description, prediction, and causal inference—each demanding different assumptions and validation methods.
Emphasizes that predictive models rely solely on observational correlations, whereas causal inference fundamentally requires expert knowledge and unverifiable structural assumptions.

Cite This Study

A 2019 study studied this question.

synapsesocial.com/papers/6a812f79080e7f4ac74d6355 https://doi.org/10.1080/09332480.2019.1579578

View Full Paper

Ask AI

Bookmark

Implication

Methodological review demonstrates a three-part classification of data science tasks in analytics, highlighting the critical role of counterfactual prediction for sound causal inference.

A Second Chance to Get Causal Inference Right: A Classification of Data Science Tasks

Discussion

Member takes

Implication

Key Points

Cite This Study

A Second Chance to Get Causal Inference Right: A Classification of Data Science Tasks

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider