Usage
  • 12 views
  • 5 downloads

Review: The Mind's Arrows: Bayes Nets and Graphical Causal Models in Psychology by Clark Glymour

  • Author(s) / Creator(s)
  • Introduction: Amongst people working in statistics, computer science, and philosophy, Bayes nets are a well- known tool to model causal structures. Besides other things this approach provides ways for obtaining causal relationships out of statistical data. The idea is that existing (conditional) dependency relations between random variables, which are reflected in statistical data, are consistent only with some causal structures between these variables (given certain axiomatic assumptions about causality, e.g., that the indirect causes of a variable are statistically screened off by its direct causes). In this manner, statistical data about a population yield inferences to (some of) the causal structure existing within that population. Alternatively, it can be evaluated under which conditions interesting causal inferences can be made at all (sometimes only if a good deal of the causally relevant variables are actually measured). The mathematical features of Bayesian networks have been already explored by Spirtes, Glymour, and Scheines (1993) and Pearl (2000), and algorithms have been designed to get causal structure out of statistical data. Now Clark Glymour offers a book dedicated to the application of this mathematical- computational framework to methodological problems in psychology.

  • Date created
    2003
  • Subjects / Keywords
  • Type of Item
    Review
  • DOI
    https://doi.org/10.7939/R3VT1H465
  • License
    © 2003 I. Brigandt et al. This version of this article is open access and can be downloaded and shared. The original author(s) and source must be cited.
  • Language
  • Citation for previous publication
    • Brigandt, I. (2003). Review: The Mind's Arrows: Bayes Nets and Graphical Causal Models in Psychology by Clark Glymour. Erkenntnis, 59(1), 136-140. http://www.jstor.org/stable/20013216
  • Link to related item
    http://www.jstor.org/stable/20013216