Abstract
Many steps are involved in completing a Bayesian statistical analysis. Some are skilled tasks requiring the expertise of a professional, others are purely mechanical. The former include such tasks as choice of model, specification of the prior distribution and interpretation of the posterior distribution; the latter include such things as the arithmetic necessary to combine the prior distribution with the data to produce the posterior distribution and to produce probability statements from that distribution. Unfortunately, it is all too often the case that the arithmetic gets in the way of the professional's decision-making responsibilities by breaking concentration and line of thought; and at times the sheer bulk of computation precludes the use of advanced techniques by the unaided researcher. What is required is a monitoring system that does all of the arithmetic and, even further, sees to it that all of the steps in the analysis are performed correctly and in their proper sequence. Also, within an instructional process, it can be very useful to have a system that helps a student learn by guiding his steps through a valid statistical analysis even if he doesn't yet fully understand what he is doing. For these and other reasons, a system of Computer-Assisted Data Analysis (CADA) was developed at the University of Iowa (Novick, 1971, 1973). Further investigation into available computer technology coupled with expansion of the theoretical base on which the original system rested resulted in the refinement and expansion of the available programs and the construction of a monitor to facilitate their use.
- 1 Abramowitz, M. and Stegun, I. A. Handbook of Mathematical Functions. U.S. Department of Commerce, NBS Applied Mathematics Series, 1964. Google Scholar
Digital Library
- 2 Borgmann, R. E. and Ghosh, S. P. Statistical distribution programs for a computer language. Research Report, IBM Watson Research Center. Yorktown Heights, New York: RC-1094, 1963.Google Scholar
- 3 Christ, D. E. The CADA Monitor. ACT Technical Bulletin No. 12. Iowa City, Iowa: The American College Testing Program, 1973.Google Scholar
- 4 Fröberg, C. E. Introduction to Numerical Analysis. Reading, Massachusetts: Addison Wesley Publishing Company, Inc., 1965.Google Scholar
- 5 Hastings, C. Approximations for Digital Computers. Princeton, New Jersey: Princeton University Press, 1955. Google Scholar
Digital Library
- 6 IBM System 136 Scientific Subroutine Package Version III (360A-CM - 03X). Programmer's Manual, Fifth Edition, Gh20-0205-4. White Plains, New York, 1968.Google Scholar
- 7 Isaacs, G. L. Interdialect translatability of the BASIC programming language. Sigcue Bulletin, 1974, 8, 11-22.Google Scholar
- 8 Isaacs, G. L., Christ, D. E., Novick, M. R., and Jackson, P. H. Tables for Bayesian Statisticians. Iowa City, Iowa: The University of Iowa, 1974.Google Scholar
- 9 Jackson, P. H. Formulae for generating highest density credibility intervals. ACT Technical Bulletin No. 20. Iowa City, Iowa: The American College Testing Program, 1974.Google Scholar
- 10 Novick, M. R. Bayesian computer-assisted data analysis. ACT Technical Bulletin No. 3. Iowa City, Iowa: The American College Testing Program, 1971.Google Scholar
- 11 Novick, M. R. High school attainment: An example of a computer-assisted Bayesian approach to data analysis. International Statistical Review, 1973, 41, 264-271.Google Scholar
Cross Ref
- 12 Novick, M. R. A course in Bayesian statistics. The American Statistician, 1975, 29, 94-97.Google Scholar
- 13 Novick, M. R. and Jackson, P. H. Statistical Methods for Educational and Psychological Research. New York: McGraw-Hill, 1974.Google Scholar
- 14 Novick, M. R. and Lindley, D. V. Assessing utilities (with applications to education). Unpublished manuscript, August 1975.Google Scholar
- 15 Stroud, A. H. and Secrest, D. Gaussian Quadrature Formulas. Englewood Cliffs, New Jersey: Prentice-Hall, 1966.Google Scholar
- 16 Sukhatme, P. V. On Fisher and Behrens' test of significance for the difference in means of two normal samples. Sankhya, 1938, 4, 39-48.Google Scholar
Index Terms
The bayesian computer-assisted data analysis (CADA) monitor
Recommendations
The bayesian computer-assisted data analysis (CADA) monitor
Proceedings of the SIGCSE-SIGCUE joint symposium on Computer science educationMany steps are involved in completing a Bayesian statistical analysis. Some are skilled tasks requiring the expertise of a professional, others are purely mechanical. The former include such tasks as choice of model, specification of the prior ...
The bayesian computer-assisted data analysis (CADA) monitor
SIGCSE '76: Proceedings of the ACM SIGCSE-SIGCUE technical symposium on Computer science and educationMany steps are involved in completing a Bayesian statistical analysis. Some are skilled tasks requiring the expertise of a professional, others are purely mechanical. The former include such tasks as choice of model, specification of the prior ...






Comments