James W. Grice, Ph.D. Professor Ph.D., University of New Mexico, 1995 B.S., Wright State University, 1987
Department of Psychology
e-mail: james.grice]att[okstate.edu.
|
I am a personality/quantitative psychologist by training with specific
interests in multivariate statistical methods and personality assessment. I have
published and presented a number of papers on factor scores (see link below) and
have completed several major computer programs. The first,
Idiogrid, is for administering, manipulating, and analyzing repertory grids
and other forms of self-report data. The second,
Observation Oriented Modeling (OOM), accompanies a book that has been
published by Academic Press (Elsevier). OOM provides a compelling alternative to
the Pearsonian-Fisherian variable-based statistics that have dominated
psychology for over 70 years. The following is an overview of this new approach:
Recovering our Common Sense: Psychology as an Observation
Oriented Science (Delivered at SWPA, Dallas, TX, April, 2010) Since the early 1900s psychological research has been
dominated by statistical methods that are overly abstract and often ill-suited
for the types of questions most psychologists wish to ask. In this presentation
Observation Oriented Modeling will be introduced as a radical alternative to
these traditional methods of data analysis. Practically speaking, Observation
Oriented Modeling challenges researchers to develop integrated models that
explain patterns of observations rather than to estimate abstract population
parameters. The focus of research is thus shifted away from aggregate
statistics, such as means, variances, and correlations, and is instead directed
toward assessing the accuracy of judgments based on the observations in hand.
This shift brings the persons in a psychological study to the forefront of the
analysis and conclusions, while completely eschewing such confusing concepts as
Type I, Type II, Type III errors, statistical power, and the p-value.
Philosophically, this new approach is more consistent with the common sense
realism of Aristotle and Thomas Aquinas than with the idealism of René
Descartes; and the end result is an approach to data conceptualization and
analysis that is demanding and rigorous, but also straightforward and intuitive.
Following the publication of OOM I hope to never conduct and publish another
stand-alone t-test, ANOVA, multiple regression, structural equation model, between-persons
factor analysis, etc. for the remainder of my academic career. The reasons for
my departure from traditional experimental statistics can be found on the OOM
website, and they are reasons offered by some of the brightest minds in
psychology over the past 70 years. I only wish I were bright enough to have
understood their arguments many years ago when I was younger! We psychologists
have let our methods become our metaphysics which has prevented the accretion of
any genuine scientific knowledge. The bottom line is that psychology must get
beyond positivism, a philosophical view that has run its course but which has
spread its roots deep into psychology by way of research methods and
statistics. We need to re-orient ourselves to Nature through the moderate
realism of the pagan and the saint (viz., Aristotle and Thomas Aquinas).
|
|
Last update: 17-August-2018