Generalization: Conceptions in the Social Sciences

Thomas D Cook*

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingChapter

1 Scopus citations


A major goal in the behavioral and social sciences is to offer conclusions that are valid for observed particulars extensive to abstract entities. Four different approaches to generalization are considered in this article. The first describes random sampling with its advantages and limitations. The second considers purposive selection in contrast to randomization. The third covers the possibility of generalizing causal statements and its connection to meta-analysis. And the last refers to extrapolation from a sample to nonsampled entities. This article concludes by arguing that random sampling, though limited, is the best formal method for generalization.

Original languageEnglish (US)
Title of host publicationInternational Encyclopedia of the Social & Behavioral Sciences: Second Edition
PublisherElsevier Inc.
Number of pages6
ISBN (Electronic)9780080970875
ISBN (Print)9780080970868
StatePublished - Mar 26 2015


  • Abstract constructs
  • Causal connections
  • Construct validation
  • Extrapolation
  • Meta-analysis
  • Purposive samples
  • Random sampling

ASJC Scopus subject areas

  • Social Sciences(all)


Dive into the research topics of 'Generalization: Conceptions in the Social Sciences'. Together they form a unique fingerprint.

Cite this