Measurement has a fraught history, and it is useful to examine the ways in which measurement can be perverted by conscious and unconscious bias.
This reading activity asks students to look at the history of racist conclusions about brain size and determine the source of bias (measurement bias, selection bias, etc).
Read the reading from Cranioklepty, by Colin Dickey, and then answer the questions that follow.
Morton’s findings, published in his Crania Americana (1839) and Crania Aegyptiaca (1844), were enormously influential in appearing to demonstrate, by means of his system, that there was a clear hierarchy in brain size between different peoples. At the top of his scheme was the European, followed by the American Indian, and then the African, just one short step above the ape. Morton’s findings seemed to show, through the cold, objective truth of math and statistics, that the European brain was conclusively larger than that of other ethnic groups.
But there are lies, damned lies, and then there are statistics. In his book The Mismeasure of Man, Stephen Jay Gould cited the numerous methodological errors that Morton made in his calculations. For one, he failed to account for differences in sex and body size when calculating brain volume. He tended to include small-bodied Incas in his American Indian sample so as to bring down that average but excluded small-bodied Hindus from his Caucasian sample so as to keep that number higher. His a priori assumptions repeatedly led him to false conclusions— demonstrably false from his own data. On top of this, Morton made elementary computational and methodological errors, all of which coincidentally favored his preexisting beliefs and assumptions.
And yet, Gould concluded, “through all this juggling, I detect no sign of fraud or conscious manipulation. Morton made no attempt to cover his tracks and I must presume that he was unaware he had left them. He explained all his procedures and published all his raw data. All I can discern is an a priori conviction about racial ranking so powerful that it directed his tabulations along preestablished lines.” This in and of itself might not be so lamentable, Gould noted, if Morton hadn’t been “widely hailed as the objectivist of his age, the man who would rescue American science from the mire of unsupported speculation.”
1. In what ways were the skulls chosen by Morton not representative?
2. What did Morton not account for in his measurements?
3. What sort of bias was Morton prone to?
4. What does this story tell us about bias in measurement?
Students may want to consult Types of Bias in Statistics to answer questions three and four.
This is part of a larger series on Statistical Literacy.