1943-2008

Richard A. Harshman Professor Department of Psychology University of Western Ontario London, Ontario, CANADA N6A 5C2

Richard was committed to the passionate pursuit of knowledge. Richard has been a valued member of this department since 1976 and rose through the ranks over the years, becoming a Full Professor. Richard was truly a renaissance man with interests that ranged from the effects of marijuana on cognitive abilities to analysis of individual differences in cerebral organization to the developments of three-player chess. He is, of course, most identified with creative and important developments in statistical analysis. An entry for him in Wikipedia notes that he was one of the pioneers in latent semantic analysis. He made two extremely important contributions to psychometrics, dealing first with the analysis of asymmetric square tables and second, in the analysis multiway tables. His work on PARAFAC is used in biomedical applications, chemometrics and wireless communications. One external reviewer of his work noted that “ He is one of the most influential quantitative methodologists in the past 30 years….he is a fundamental thinker, not distracted by appearances of currently fashionable approaches. He is always walking a few steps ahead of most of us”. Another noted: “His work is well known..his presence and influence looms large and he basically single handedly started the complex discussions about uniqeness several years prior to receiving his PhD.”

We, and the scientific community, not only have lost a major contributor to theory, we have also lost a truly gentle man- a man  playful with ideas, kind-hearted and generous.  The world is diminished with his passing.

Dr. Richard Allan Harshman is survived by his wife Elizabeth Hampson and was the cherished son of Allan and Louise Harshman and son-in-law of John and Margaret Hampson. Richard was also the dear brother of Susan (Greg) Liddle and dear brother-in-law of Cheryl (Ken) Haddrell, Crystal and Mark. He leaves behind many valued friends and colleagues at U.W.O.

A graduate scholorship has been set up in in memory of Dr. Harshman. Donations will be gratefully accepted for the

Richard A. Harshman Graduate Scholarship
c/o Foundation Western
University of Western Ontario
London Ontario, Canada
N6A 3K7
or by calling 519-661-3140.

 

Research Focus
	Develop new models and methods for uncovering dimensional structure in complex multivariate data e.g., n-way exploratory factor analysis and MDS, decomposition of asymmetries, new mathematical notation for array algebra
	Apply new (and standard) methods to data sets of colleagues in order to gain experience with problems that arise in real data situations, and to devise solutions to these problems, e.g., in psychometrics and chemometrics
	Explore ways of analyzing novel kinds of data and/or "difficult" data (e.g., subjective causal maps, fMRI data) to gain insights into interesting measurement issues

Graduate Students:

• Sungjin Hong, PhD (Degree granted June 2002)
  Dissertation Title: "Factor analytic methods useful for rotational and dimensional indeterminacies"
  Current Position:
  Assistant Professor, Department of Psychology
  University of Illinois
  603 E. Daniel Street
  Champaign, IL 61820
• Zhen Zhang, MSc (Degree granted June 2007)
  Thesis Title: "Testing a measurement model for an analysis method that combines aspects of Parafac with structural equation modeling"
  Currently: PhD student at the University of Western Ontario

• Recent Publications and Conference Presentations
  • Introducing new 'ways' into data analysis: Making linear models multilinear gives them important new properties. Talk presented at the York University Department of Mathematics and Statistics weekly colloquium series, York University, Toronto, Nov. 24, 2006 (invited). Abstract (1 p)         Equation list (6 pp)
  • A randomization method of obtaining valid p-values for model changes selected "post hoc". (IMPS 2006 poster)
  • Generalization of Parafac and Tucker models to canonical correlation models. (TRICAP 2006 paper)
  • Improvements to three-way DEDICOM for applications in social network analysis. (TRICAP 2006 paper)
  • Generalizing factor analysis and canonical correlation to three-way arrays increases their ability to disentangle information. (SSC 2006 poster)
  • Valid p-values for stepwise regression and other post-hoc model selection methods. (SSC 2006 poster)
  • Shifted factor analysis--Part I: Models and properties.
  • Shifted factor analysis--Part II: Algorithms.
  • Shifted factor analysis--Part III: N-way generalization and application.

• Psychometric/chemometric publications
• Individual differences/cognition/neuropsychology publications
• Conference presentations
  • 1996--present (entire list)
  
  Selected years: 1996-97;  2000-2002;  2003-2004;  2005;  2006;  2007
  • 1978, 1985-1995 (Selected psychometric presentations)

* Partial funding since 1980 has been provided by the Natural Sciences and Engineering Research Council of Canada

Conference Information

Upcoming:  

Past:  
• TRICAP (June 2003)
• Tensor Decompositions Workshop (July 2004)
• IMPS (July 2005)
• Workshop on Tensor Decompositions (August 2005)
• Statistical Society of Canada (May 2006)
• TRICAP (June 2006)
• IMPS (June 2006)
• IMPS (July 2007)
• International Conference on Data Mining (October 2007)

Full Text of Hard-to-Find and/or Basic Publications

Parafac related:  

• Harshman, R. A. (1970). Foundations of the PARAFAC procedure: Models and conditions for an "explanatory" multi-modal factor analysis. UCLA Working Papers in Phonetics, 16, 1-84. (UMI Serials in Microform, No. 10,085).
• Harshman, R. A. (1972). Determination and proof of minimum uniqueness conditions for PARAFAC1. UCLA Working Papers in Phonetics, 22, 111-117. UMI Serials in Microform, No. 10,085).
• Harshman, R. A. (1972). PARAFAC2: Mathematical and technical notes. UCLA Working Papers in Phonetics, 22, 30-44. UMI Serials in Microform, No. 10,085).
• Harshman, R. A., Ladefoged, P., & Goldstein, L. (1977). Factor analysis of tongue shapes. Journal of the Acoustical Society of America, 62, 693-707.
• Harshman, R. A., & Berenbaum, S. A. (1981). Basic concepts underlying the PARAFAC- CANDECOMP three-way factor analysis model and its application to longitudinal data. In D. H. Eichorn, J. A. Clausen, N. Haan, M. P. Honzik, & P. H. Mussen (Eds.), Present and past in middle life (pp. 435-459). New York: Academic Press.
• Harshman, R. A., & Lundy, M. E. (1984). The PARAFAC model for three-way factor analysis and multidimensional scaling. In H. G. Law, C. W. Snyder, Jr., J. Hattie, & R. P. McDonald (Eds.), Research methods for multimode data analysis (pp. 122-215). New York: Praeger.
• Harshman, R. A., & Lundy, M. E. (1984). Data preprocessing and the extended PARAFAC model. In H. G. Law, C. W. Snyder, Jr., J. Hattie, & R. P. McDonald (Eds.), Research methods for multimode data analysis (pp. 216-284). New York: Praeger.
• Harshman, R. A. (1984). "How can I know if it's 'real'?" A catalog of diagnostics for use with three-mode factor analysis and multidimensional scaling. In H. G. Law, C. W. Snyder, Jr., J. Hattie, & R. P. McDonald (Eds.), Research methods for multimode data analysis (pp. 566-591). New York: Praeger.
• Harshman, R. A., & DeSarbo, W. S. (1984). An application of PARAFAC to a small sample problem, demonstrating preprocessing, orthogonality constraints, and split-half diagnostic techniques. In H. G. Law, C. W. Snyder, Jr., J. Hattie, & R. P. McDonald (Eds.), Research methods for multimode data analysis (pp. 602-642). New York: Praeger.
• Lundy, M. E., Harshman, R. A., & Kruskal, J. B. (1989). A two-stage procedure incorporating good features of both trilinear and quadrilinear methods. In R. Coppi & S. Bolasco (Eds.), Multiway data analysis (pp. 123-130). Amsterdam: Elsevier.
• Kruskal, J. B. (1989). Rank, decomposition, and uniqueness for 3-way and N-way arrays. In R. Coppi & S. Bolasco (Eds.), Multiway data analysis (pp. 7-18). Amsterdam: Elsevier.
• Kruskal, J. B., Harshman, R. A., & Lundy, M. E. (1989). How 3-MFA can cause degenerate PARAFAC solutions, among other relationships. In R. Coppi & S. Bolasco (Eds.), Multiway data analysis (pp. 115-121). Amsterdam: Elsevier.
• Deerwester, S., Dumais, S. T., Furnas, G. W., Landauer, T. K., & Harshman, R. A. (1990). Indexing by latent semantic analysis. Journal of the American Society for Information Science, 41, 391-407.
• Harshman, R. A., & Lundy, M. E. (1994). PARAFAC: Parallel factor analysis. Computational Statistics and Data Analysis, 18, 39-72.
• Harshman, R. A., & Lundy, M. E. (1996). Uniqueness proof for a family of models sharing features of Tucker's three-mode factor analysis and PARAFAC/CANDECOMP. Psychometrika, 61, 133-154.
• Harshman, R. A., & Lundy, M. E. (1997). An "extended PARAFAC model" incorporating singly-subscripted constants: Theory and application. Poster presented at TRICAP '97, a conference on three-way methods in chemistry and psychology, Lake Chelan, Washington, May.
• Harshman, R. A. (2004). The problem and nature of degenerate solutions or decompositions of 3-way arrays. Paper presented at the American Institute of Mathematics Tensor Decomposition Workshop, Palo Alto, CA, July. (invited)
• Harshman, R. A. (2004). An annotated bibliography of articles on degenerate solutions or decompositions: Supplementary information for the above paper.
• Harshman, R. A. (2005). Tutorial: The Parafac model and decomposition and their variants. Paper presented at the Centre International de Rencontres Mathématiques (CIRM) Workshop on Tensor Decompositions and Applications, Marseilles, France, August.

N-way related:  

• Harshman, R. A. (2001). An index formalism that generalizes the capabilities of matrix notation and algebra to n-way arrays. Journal of Chemometrics, 15, 689-714.
• Harshman, R. A., & Hong, S. (2002). 'Stretch' vs 'slice' methods for representing three-way structure via matrix notation. Journal of Chemometrics, 16, 198-205.

DEDICOM related:   

• Harshman, R. A. (1978). Models for analysis of asymmetrical relationships among N objects or stimuli. Paper presented at the First Joint Meeting of the Psychometric Society and The Society for Mathematical Psychology, Hamilton, Ontario, August.
• Harshman, R. A. (1981). Alternating least squares estimation for the single domain DEDICOM model. Unpublished technical memorandum, Bell Laboratories, Murray Hill, NJ.
• Harshman, R. A., Green, P. E., Wind, Y., & Lundy, M. E. (1982). A model for the analysis of asymmetric data in marketing research. Marketing Science, 1, 205-242.
• Dawson, M. R., & Harshman, R. A. (1986). The multidimensional analysis of asymmetries in alphabetic confusion matrices: Evidence for global-to-local and local-to-global processing. Perception and Psychophysics, 40, 370-383.
• Harshman, R. A., & Lundy, M. E. (1990). Multidimensional analysis of preference structures. In A. deFontenay, M. H. Shugard, & D. S. Sibley (Eds.), Contributions to economic analysis: Vol 187. Telecommunications demand modeling: An integrated view (pp. 185-204). Amsterdam: Elsevier.
• Lundy, M. E., Harshman, R. A., Paatero, P., & Swartzman, L. C. (2003). Application of the 3-way DEDICOM model to skew-symmetric data from paired preference ratings of treatments for chronic back pain. Poster presented at TRICAP 2003, a conference on trilinear methods in chemistry and psychology, Lexington, Kentucky, June.

Miscellaneous

Ongoing internet math/psych/3-way research project

Marijuana, cognitive style, and lateralized hemispheric functions (mentioned in 12 November 2003 letter to National Post)

• http://three-mode.leidenuniv.nl/
  P. Kroonenberg's Three-Mode Company with links to a large bibliography, TMC analysis programs, three-way data sets, three-mode research conferences, contact information for an extensive list of workers in the multiway area (some with pictures), including important contributors that do not have web pages (e.g., Jos ten Berge, J. Douglas Carroll), etc.
• http://www.models.kvl.dk/users/rasmus
  R. Bro's site with multi-way analysis and chemometrics links, free Parafac and Tucker related software (in Matlab), and short music clips
• http://www.models.kvl.dk/research/theses/index.asp
  downloadable book Multi-way Analysis in the Food Industry (1998); a good introduction to multilinear models that starts simply, goes on to in-depth treatment, and presents new results
• http://www.eigenvector.com
  chemometrics research and applications company, a leading developer of three-way applications, including The PLS Toolbox which contains Matlab based software for PARAFAC, Partial Least Squares Regression (PLS), Principal Components Regression, etc.
• N. Sidiropoulos's publications; includes, for example, applications for PARAFAC and related three-way methods in electrical engineering (e.g., signal processing) and papers on mathematical analysis of uniqueness of n-way arrays
• http://www.ppsw.rug.nl/~kiers/
  Henk Kiers site, which lists many important multiway articles, e.g., on Tucker and constrained Tucker models, advanced multiway algorithms.

More information about the psychology department at UWO can be found at http://www.psychology.uwo.ca/

Last revised 16 Dec 2008