The Birth of Psychometrics: Cambridge 1886 to 1889

The Old Cavendish Laboratory, Cambridge, C. 1900It is a little known fact that psychometrics as a science began in Cambridge between 1886 and 1889. The first laboratory dedicated to the subject was set up within the Cavendish Physics Laboratory in the University of Cambridge by James McKeen Cattell in 1887. In Cattell's own words "In the Cavendish Laboratory was also set up apparatus for research and this was the beginning of the first British laboratory of psychology" (Cattell, Early Psychological Laboratories,1928). Cattell, an American, completed his PhD, entitled ‘Psychometric Investigations’ (Cattell, 1886), with Wundt at Leipzig. During his period in Leipzig, Cattell had been in frequent correspondence with Francis Galton at his Anthropometric Laboratory in London (Galton, 1887) and he quickly saw the potential for synergising Wundt’s psychophysics with Galton’s mathematical approach to the examination of individual differences. On leaving Leipzig, and after a brief visit to America, Cattell returned to Europe to take up an appointment in the University of Cambridge in October 1886 as ‘Fellow Commoner’ at St John’s College and lecturer in the University.

Anthropometrics at Cambridge 1885 - 1886

Before the arrival of Cattell, Galton had presented the Rede Lecture at Cambridge in June 1884 on the topic “The nature, principles and objects of the quantitative estimate of some of the less commonly and less easily measured of the human faculties”, and presented the University with several instruments similar to those he had used in his South Kensington Anthropometric Laboratory.  John Venn (most well know for the ‘Venn diagram’), then Lecturer in Moral Science, Fellow of Gonville and Caius College, Cambridge, and Secretary of the Anthropological Institute of which Galton was President, endeavoured to carry out a study of Cambridge undergraduates using Galton’s techniques. He reported:

“Some difficulty was experienced at first in choosing a suitable room in which measurements could be carried out, as the University had but little available space, and unless some room could be found the position of which would bring the subject prominently under the notice of the students, no very extensive results could be hoped for. At first the Committee Room of the Union Society was put at our disposal, but this was not very long available, as some demur was made by the authorities there to the use of the room by undergraduates who were not members of the Society. After a time the library of the Philosophical Society, situated at the centre of the new museums and lecture rooms, was secured, and the measurements were taken there by Mr White, the Librarian of the Society.”  (Venn, 1889)

The instruments used for this study measured keenness of eyesight, strength (of pull and squeeze), head size and shape, breathing capacity, height and weight. In addition, 'intellectual characteristics' were assessed by college tutors, who divided the students into groups as follows: 'first-class man', 'remaining honours men' and 'poll-men'. Venn’s conclusion from this study of 1,450 Cambridge students was that “there does not seem to be the slightest difference between one class of our students and another”. That is, first class mens' heads were no bigger than the heads of lesser mortals (physically at least), neither did they differ in their other physical attributes. These results anticipated those of Wissler’s famous New York study of Columbia students (Wissler, 1901) by more than 10 years.

Cattell's Psychometric Laboratory 1887 - 1889

While Venn was carrying out his anthropometric studies, James Ward, a Fellow of Trinity College (and later to become Professor of Moral Philosophy and Logic), had attempted to gain Senate approval for the establishment of a Psychology Laboratory at Cambridge. However, this was turned down by Senate on the grounds that such a laboratory “would insult religion by putting the human soul in a pair of scales” (Sokal, 1971). On Cattell’s arrival, he and Ward discussed the possibility of finding a suitable site for the psychometric testing apparatus he had brought with him from Leipzig.  Several departments and faculties were approached. Cattell stated in a letter to his father in September 1887 (Sokal, 1981)

“I have been busied this afternoon trying to find a place for a psychological laboratory. All the buildings are very crowded. Some of the colleges are rich but the university itself is poor, and finds it expensive to house laboratories and museums which have grown rapidly during the past few years. I expect, however, we will be able to get something. I dine with Mr Ward tomorrow to talk it over.”

Two days later, he had his laboratory.

“Yesterday I saw the Professor of Physics (J. J. Thompson, also a Fellow of Trinity College), and have made arrangements to start a laboratory in the building for physics. I shall set up apparatus to make original research, and look after any who wish to study the subject.”

Thus, the world’s first laboratory devoted to psychometric study was established within the Cavendish Laboratory, one of the most significant buildings in the history of modern science . The Cavendish Laboratory was later to house 21 Nobel Prize winners, including Earnest Rutherford, awarded for the splitting of the atom, and James Watson and Francis Crick for the discovery of the double helix structure of DNA.

Much of the apparatus was Cattell’s own, designed for his PhD work and manufactured in Germany using the same facilities as Wundt. Cattell also had access to equipment from Galton’s Anthropometric Laboratory in South Kensington, and had some built by Horace Darwin’s ‘Cambridge Scientific Instrument Company’. Horace Darwin was Charles Darwin’s son. Although Cattell received no financial support from the University for his laboratory, he was allowed by the Physics Laboratory’s Director, J. J. Thompson (who was later to become well-known for the discovery of the electron), to use some of the laboratory’s own equipment and technical services, and was provided with a ‘boy in wait’. Henry Sidgwick, the Professor of Moral Philosophy also donated £200 (Sokal, 1972). Cattell's ‘mental tests’ included those of reaction time, least noticeable difference, colour naming, accuracy, judgement, memory and attention. As well as demonstrating the use of the various pieces of apparatus within an experimental approach to psychology, Cattell held classes on mental testing for students from a variety of disciplines. Today Cattell’s rooms in the Old Cavendish Laboratory in Free School Lane in central Cambridge are occupied by the Department of Social and Developmental Psychology.

Cattell's return to America

In 1889, Cattell moved with his new English wife (Josephine Owen from London, who he had first met while in Leipzig) to a newly established Professorship in Psychophysics at the University of Pennsylvania, taking all his laboratory materials with him. No other psychology laboratory was to exist in Cambridge for at least ten years. However, the Cattells’ stay in Philadelphia was to be short lived. He moved to Columbia University in New York in 1891 as Professor of Psychology, where he remained until 1918 when he was dismissed for opposing the draft (his own son was a conscientious objector). Cattell sued the University and won a substantial sum and an annual pension. He was also an advocate of academic freedom for university faculty and his case was instrumental in establishing the principle of academic tenure in American colleges and universities. The settlement with Columbia enabled him in 1921 to establish the very successful test publishing company ‘The Psychological Corporation’, which is today part of Pearson Assessment.

Sir Francis Galton and James McKeen Cattell

Sir Francis Galton has often been described as ‘the father of psychometrics’, and as early as 1883 he had suggested that people of genius might also possess other psychological attributes such as unusually fine sensory discrimination. However, he appears to have been diverted from the point by other interests and have contended himself with the general impression without pursuing the matter in systematic investigation (Spearman, 1904). In 1890 Cattell published his seminal paper 'Mental Tests and Measurement', and over the next 10 years he was to apply these tests to a large number of individuals, believing this would enable "the scientific investigation of changes over time, inter-relatedness and variation under different circumstances" (Cattell, 1890). He also suggested that individuals would "find their tests interesting and useful in regard to training and mode of life".

In a footnote to Cattell's 1890 text, Galton commented rather dismissively:

"One of the most important objects of measurement is hardly if at all alluded to. The sort of estimate I have in view and which I would suggest should be something of this kind: 'mobile, eager, energetic; well shaped, successful at games requiring good eye and hand; sensitive; good at music and drawing' " (Cattell, 1890)

Cattell replied:

"It is convenient to follow Mr. Galton in combining tests of body, such as weight, size, colour of eyes, etc., with psychophysical and mental determinations ... (but) these latter alone are the subject of the present discussion." (Cattell, 1890).

Galton's conjecture was not finally refuted until the next century when Wissler (1901) published the definitive work on Cattell's accumulated data. It is noteworthy that the schism between those who, like Galton and his followers, were interested in the assessment of mental faculty from an anthropological and eugenicist point of view, and those who wished to assess mental function from an applied or experimental perspective was already present during these early days of psychometric endeavour.

Galton’s focus was neither psychology nor psychometrics. The driving force behind his work was always eugenics, his own protégé subject, and its experimental method of anthropometrics. It was to further this work that he caused to be set up the Department of Eugenics at University College London, endowing a Chair of Eugenics in that institution on his death. In fact, it is the discipline of statistics itself that was to be the main beneficiary of his endeavours; a statistics without which none of today’s measurement sciences of psychometrics, biometrical genetics, educational assessment, biometrics and econometrics would have been possible. The contribution of the Galton team to the building of statistics was enormous. Galton himself derived the standard deviation and regression; Karl Pearson, the first Professor of Eugenics, gave us the correlation coefficient, and Charles Spearman factor analysis. Between them they originated twin studies in human biometrics, and a later holder of the Galton Chair of Eugenics , R. A. Fisher, gave us analysis of variance, non-parametric statistics and maximum likelihood estimation, among other things. While not denying the importance of Galton’s statistical legacy to psychometrics, the rest of his contribution has been rather a poisoned chalice from the psychometrician's point of view. Galton’s devotion to eugenics pre-empted scientific racism with all its horrible consequences, and the enthusiasm with which its adherents took up the banner of IQ testing continues to be an embarrassment to the profession to this day.

References

Cattell, J. McK. (1886) Psychometrische Untersuchungen. Philosophische Studien, 3, 305-335; 452-492.

Cattell, J. McK. (1888) The Psychological Laboratory at Leipzig. Mind, 13, 37-51.

Cattell, J. McK. (1890) Mental Tests and Measurements. Mind, 15, 373-381.

Cattell, J. McK. (1928) Early Psychological Laboratories. Science, 67, No.1744, 543-548.

Galton, F. (1883)  Inquiries into the human faculty and its development. Macmillan, London

Galton, F. (1887) On recent designs for anthropometric instruments. Journal of the Anthropological Institute, 16, 2-8.

Sokal, M. (1971) The unpublished autobiography of James McKeen Cattell. American Psychologist. July 1971, 26(7), 626-635.

Sokal, M. (1972) Psychology at Victorian Cambridge – the unofficial laboratory of 1887-1888. Proceedings of the American Philosophical Society, 116, 145-147.

Sokal, M. (1981) An Education in Psychology: James McKeen Cattell's Journal and Letters from Germany and England, 1880-1888. MIT Press, Cambridge, MA

Spearman, C. (1904) General Intelligence objectively determined and measured. American Journal of  psychology, 15, 201-293.

Venn, J. (1889) Cambridge Anthropometry. Journal of the Anthropological Institute, 18

Wissler, C. (1901) The correlation of mental and physical tests. Psychological Review Monograph Supplements, 3(6).

© John Rust, August, 2008.