What is it about?
In a 23-page section of "The Mismeasure of Man," Stephen Jay Gould characterized the Army Beta as an example of biased scientists creating an intelligence test in order to perpetuate their discriminatory behavior. However, we compared Gould's judgments to scholarly work about the Army Beta from the time and find that most of Gould's criticisms are either exaggerated or groundless. In reality, the Army Beta was well designed by the standards of the time and measured intelligence--and still does so to an extent today.
Photo by Kowit Phothisan on Unsplash
Why is it important?
"The Mismeasure of Man" is a hugely popular science book published in 1981 (and republished in 1996) in which Stephen Jay Gould argued that intelligence testing and related fields were based on a rotten foundation of racism, sexism, classism, and other social evils. One of the targets of his opprobrium was the Army Beta, a test designed in World War I to screen illiterate men for military service. Gould argued that the test was engineered to show lower scores for immigrants and exaggerate the low mental ability of poorly educated men. We show in this article that none of this is true. Indeed, much of Gould's book is based on falsehoods, and it is not a trustworthy source of information about the history of science or the social issues surrounding the study of humans.
Read the Original
This page is a summary of: Stephen Jay Gould’s Analysis of the Army Beta Test in The Mismeasure of Man: Distortions and Misconceptions Regarding a Pioneering Mental Test, Journal of Intelligence, February 2019, MDPI AG, DOI: 10.3390/jintelligence7010006.
You can read the full text:
Raw item-level data of the replication in the article.
Army Beta test administration instructions
Instructions (with proper time limits) for administering the Army Beta.
Army Beta test stimuli
Stimuli (i.e., test forms) for administering the Army Beta.
OSF documentation, including pre-registration, open data, stimuli, photos of examinees in the replication, and data analysis files.
Kirkegaard's re-analysis of the item data
Emil O. W. Kirkegaard re-analyzed the item level data and showed that for my replication sample, the items often had low factor loadings and poorly discriminated among educated scorers. This supports my belief that the Army Beta is too easy for a modern, educated sample.
James Thompson's summary and viewpoint
James Thompson's blog post summarizing the study and giving his perspective on the article.
Quillette article about "Mismeasure"
Nontechnical article that I wrote for Quillette magazine about the larger issue of preventing bias in science and using Gould and the Army Beta as an example.
The following have contributed to this page