Fumiko Samejima (b. 1930)
by Ryan Barnhart, York University
Dr. Fumiko Samejima is recognized as a pioneer in modern psychometrics. Her contributions have been diverse, but her work on latent trait theory, otherwise known as item response theory (IRT), has been described as a “benchmark… in modern test theory” (Educational Measurement: Issues and Practices, 1991). She is recognized a major founder of polytomous IRT (Ackerman, 1998) and her methods have been widely applied for more than 30 years.
Fumiko Samejima was born in Tokyo, Japan, on December 25, 1930. She developed an interest in mathematics early in her school life while attending Jo-koshi Fuzoku High School, a girls’ school considered to be one of the best in Japan (F. Samejima, personal communication, April 14, 2008). After a year of college study in mathematics, she switched schools to attend Keio University, a well known institution of Western learning in Tokyo. Here she encountered a twin study which probed the nature versus nurture argument. Feeling a resonance with this paper, Samejima changed her major to psychology and applied to graduate school. Her initial encounters with psychology were rather discouraging. It appeared to her that what psychology represented was not science as she understood it. Encountering Frederic Lord’s 1952 Psychometrika monograph, she became intrigued by latent trait theory. Latent trait theory proposes that one can evaluate individuals using test items which actually measure unobservable characteristics of interest (e.g., intelligence testing and ability testing). Her discovery of psychometrics represented a natural return to mathematics. While studying under Dr. Tarow Indow, a pioneer in cognitive science (Romney, 2007), she coauthored a book entitled LIS Measurement Scale of Non-Verbal Reasoning Ability. This book would later be cited in Frederic Lord and Melvin Novick’s Statistical Theories of Mental Test Scores in 1968. She obtained her Ph.D. in psychology from Keio University in 1965.
Following graduation, Samejima worked for two research organizations before taking a one month trip to the United States. Her trip opened up many opportunities. She met Frederic Lord, Melvin Novick and importantly Norman Fredrickson of the Educational Testing Service (ETS), the world’s largest private educational testing and measurement research organization. Samejima suggests that the warm welcome she received was important in her decision to ultimately spend such an extended period of time in the United States (Wainer & Robinson, 2007). Shortly after returning to Japan, Norman Frederickson extended an invitation for her to join ETS. She accepted the offer with enthusiasm and spent a year there as a visiting research psychologist.
Reluctant to leave the United States following her time at ETS, Samejima accepted another invitation to spend a year as a research fellow at the L.L. Thurstone Psychometric Laboratory of the University of North Carolina at Chapel Hill. Here she met and influenced other psychometricians like Darrel Bock, who found her work and view of IRT to be innovative.
With the pressing concern of a three-year visiting scholar’s visa about to expire, Samejima chose to pursue work in Canada. There, she could avoid the issues with her visa while also being able to return to the United States at a later date. She settled into a position teaching test theory, statistics, and other topics at the University of New Brunswick for the next two years. Here she published one of her more frequently cited monographs in Psychometrika, “Estimation of Latent Ability Using a Response Pattern of Graded Scores” (Samejima, 1969).
Samejima retained a desire to return to the United States. She accepted an invitation to Bowling Green State University in 1970, where she spent a brief period of time. Here she published another of her more notable papers: “A General Model for Free Response Data” (Samejima, 1972). In 1973, she accepted a position at the University of Tennessee Knoxville where she remains as full professor.
Samejima’s research at the University of Tennessee Knoxville covers a wide range of topics and has garnered her a number of awards. Her ability to perceive future changes in technological equipment, most notably computing, has led to comprehensive general models which make modern data collection of synchronic measures in cognitive experiments, very useful (Educational Measurement: Issues and Practices, 1991). She also produced 35 technical reports for the Office of Naval Research, who funded a portion of her research from 1977 to 1992. The Law School Admission Council has also used Samejima’s research, and funded her from 1999 to 2001. As acknowledgement of her contributions, Samejima sat on the board of trustees for the Psychometric Society (1989-1990), received the Outstanding Technical Contribution from the National Council on Measurement in Education (1991) and served as the president of the Psychometric Society (1996-1997). She has served on the editorial board of several journals, including Applied Psychological Measurement from 1975 to the present, and she received the Outstanding Researcher Award from Behaviormetric Society (in Japan) in 2007, among others honors (F. Samejima, personal communication, April 14, 2008).
Samejima’s career and experiences as a Japanese woman in the particularly male-dominated field of statistics must be put into context. The disadvantages presented by her gender and ethnicity stood as possible points of discouragement. Her achievements stand as evidence of how she always faced challenge as opportunity. Challenge always emboldened her as a woman of non-American descent, and her research drive was influenced by this sentiment. In regards to the women’s liberation movement, she always felt the struggle for female autonomy was indicative of what was good and hopeful about the United States (Wainer & Robinson, 2007). She saw the movement as freeing for both men and women, noting not only advances in female opportunity, but also the migration of men into professions not typically considered “masculine.”
As a professor emerita, Samejima continues to teach and conduct her research at the University of Tennessee Knoxville. Her research continues to expand not only our mathematical theories, but our understanding and interpretation of measured quantities. Her models have been, and continue to be, used to shed light on many issues in cognitive psychology.
Ackerman, T. (1998). [Review of the book Handbook of Modern Item Response Theory]. Journal of Educational Measurement, 35(4), 346-351.
Educational Measurement: Issues and Practices (1991). 1991 NCME Award for outstanding technical contribution to educational measurement. Educational Measurement: Issues and Practice, 10 (3), 36.
Romney, K.A. (2007) In memorium: Tarow Indow (Aug. 22, 1923 – Sept. 22, 2007). Color Research and Application, 33(2), 92-93.
Samejima, F. (1969). Estimation of latent ability using a response pattern of graded scores. Psychometrika Monograph Supplement, 34(4, Pt.2).
Samejima, F. (1972). General model for response data. Psychometrika Monograph Supplement, 37(1, Pt.2).
Wainer, H., & Robinson, D.H. (2007). Profiles in research: Fumiko Samejima. Journal of Educational and Behavoural Statistics, 32(2), 206-222.