Statistics Program at MSU: 1971-1976

Summary

     This period corresponds to the term of MSU president Carl W. McIntosh (1970–1977). The 1970 MSU Fall enrollment was 8187. The total college academic year expenses for an undergraduate student in 1970 were $1,440 ($9,100 in 2017 dollars) for a Montana resident and $2,110 ($13,300 in 2017 dollars) for a non-resident. Montana had 14 state-supported institutions of higher education, more per capita than any other state in the union. Shortly after McIntosh arrived, the state imposed budget reductions on the University. Under the new Montana Constitution of 1972, the university was supervised by a mandated Board of Regents (BOR) and an administrative officer, the Commissioner of Higher Education (CHE). In 1975, the BOR put a freeze on University hiring. The CHE extracted a million dollars from MSU's reserve fund and transferred the dollars to the University of Montana and the CHE office. The fallout from that controversial action led McIntosh to resign on June 30, 1977, and the CHE to resign the next year. The 1977 legislature provided generous support to MSU for the next biennium.
      Montana Courts ruled in 1976 that MSU had discriminated against women faculty members and imposed strict rules for fair personnel management. MSU was selected by the University of Washington to be Montana's participating institution in a multi-state cooperative medical education program. By 1977, MSU enrollment had grown to 9800, the largest school in the state. The physical size of the campus increased also. New buildings included: Sales Stadium, Marga Hosaeus Center, Creative Arts Complex, Wilson Hall, Sherrick Hall, and Leon Johnson Hall.
      At the National level, the Viet Nam War ended with the fall of Saigon in 1975, legislation created the Environmental Protection Agency (EPA, 1970), an expanded Clean Air Act (1970), the Clean Water Act (1972), the Student Loan Marketing Association (Sallie Mae, 1973), and a Civil Rights Act (1974). The U.S. Constitution was amended in 1971 to lower the voting age to 18 from 21. The U.S. celebrated its bicentennial in 1976.
      President Nixon, who had been re-elected in 1972, resigned in 1974 because of the Watergate scandal; VP Gerald Ford became the President. The country experienced a recession for the two years of 1973-1974. The energy crisis began in 1973 and lasted through the decade.
      In the world of statistics beyond MSU, this era produced the proportional hazard model and regression modeling of censored failure time data, an increased interest in cross-validation, Akaike's Information Criterion (AIC), theory and methods for generalized linear models, the bootstrap method, and box-and-whiskers diagrams.
      In September, 1970, Dr. Martin Hamilton, joined McFeely, Tiahrt, and Lund on the statistics faculty. In 1972, Dr. Robert Engle became department head, replacing Louis Barrett. In 1976, probabilist Dr. Warren Esty joined the department and provided unique skills of importance to the Statistics Program.
      During this era, the graduate program in statistics conferred the MS degree to 13 students and the PhD degree to 6 students. The statisticians added undergraduate courses including an Experimental Design course, an Advanced Sampling course to replace Genetic Statistics, and a lower division Bayesian course called Decision Making. The Elementary Statistics course was given the new number 216 in 1972, replacing the number 116 that it held for the previous 10 years. Initially a 3-quarter sequence on Probability and Stochastic Processes was offered, but before long it was replaced with a 1-quarter Probability course. New Graduate level courses included Regression Methods in Curve Fitting, a three-quarter sequence on the Design and Analysis of Experiments, and Advanced Sampling. Ten new courses were added that covered mentored learning or supplementary education because those previously unlisted teaching/learning activities needed course numbers to gain recognition in the BOR and MSU funding formulas.
      By 1976, all statistics course numbers were preceded by the rubric STAT rather than MATH and the Statistics Program and its courses appeared as a separate category within the department's section in the MSU course catalog. The statisticians were running an active Statistical Laboratory to provide such services as statistical analysis, data processing, consulting, and sample survey support.
      Ken Tiahrt spearheaded the effort to create the Montana Chapter of the American Statistical Association. It was formally founded in 1976 and consistently has held one or two state-wide Chapter meetings each year since.
      The Sigma 7 computer was providing main frame computer services. Statisticians used the SPSS and BMDP statistical packages or their own Fortran programs for batch submissions. But the Sigma 7 did allow limited interactive computations. Dick Lund wrote a computer package called MSUSTAT for interactive statistical calculations. By 1976, MSUSTAT was being used by students for statistics assignments and by researchers for analyzing small data sets.

Detailed Chronicle of the Statistics Program: 1971-1976

1970-1972 MSU catalog
This is the first catalog to omit names of instructors from course descriptions. The course list indicates that four new undergraduate courses were added to the statistics program.

In 1970,Marty Hamilton joined the faculty as an Assistant Professor of Statistics.

Dr. Martin (Marty) Alva Hamilton (MSU, 1970–2002) received his Ph.D. in Statistics in 1968 from Stanford University. His dissertation “The Robbins-Monro Stochastic Approximation Approach to a Discrimination Problem” was directed by Prof. Rupert Griel Miller, Jr. Prior to arriving at MSU, Marty served 2+ years as a Commissioned Officer in the United States Public Health Service on assignment to the Biometry Branch of the National Cancer Institute in Bethesda, MD. (Marty is a co-author of this history, the one sometimes denoted by MAH.) With his arrival, the statistics faculty became four (McFeely, Tiahrt, Lund, and Hamilton) among the 23 tenure-track faculty in the department.

I (MAH) taught the 407-8-9 from the Hogg and Craig text Introduction to Mathematical Statistics. I was also assigned to teach the new 420-1-2 sequence, which was probably created for Mode. I think I used William Feller’s An Introduction to Probability Theory and Its Applications, Vol. 1, 3rd Edition followed by Emanuel Parzen’s Stochastic Processes.

Graduate Degrees Awarded: 1970 - 1972

The MS degree was awarded in 1972 to the statistics student Robert Chew.

The Ph.D. degree was awarded in 1970 to statistics students Richard Leo Schwaller, and in 1971 to Roy Neal Byrd.

• Dr. Schwaller’s dissertation  “A Method of Constrained Randomization for ...  Factorials” was supervised by Dr. Kenneth J. Tiahrt.³
• Dr. Byrd’s dissertation  “A Multivariate Runs Statistic” also was supervised by Dr. Kenneth J. Tiahrt.⁶

1972-1974 MSU catalog

Robert Engle, Dept. Head (1972-77)      1972 photo

The Math Department tenure-track faculty remained as before, although it additionally listed 15 instructors of which 3 were statisticians. On July 1, 1972, the mathematician Dr. Robert Engle (Head, 1972-77) became department head, replacing Louis Barrett.

In the statistics course offerings, Math 116 - Elementary Statistics was renumbered to become Math 216. The courses Math 428 - Genetic statistics and Math 429 - Sampling theory were replaced with Math 428 - Methodology and theory for sampling and Math 429 - Advanced methodology and theory for sampling; i.e., the Genetic Statistics course was replaced by an Advanced Sampling course. Five new courses were offered:

Although not mentioned in its description, Math 120 Decision Making was a Bayesian introduction to probability and quantitative reasoning. Initially, lectures followed the book Making Decisions by D. V. Lindley (1971), New York: John Wiley & Sons, vii + 189pp.

I (REL) taught the 527 Regression Methods course following the book Applied Regression Analysis by N. R. Draper and H. Smith (1966) New York: John Wiley & Sons, ix + 407pp. We think that the 577-8-9 Design of Experiments course relied on The Design and Analysis of Experiments – 2nd edition by O. Kempthorne (1967) New York: John Wiley & Sons and Experimental Designs – 2nd edition by W. G. Cochran and G. M. Cox (1957) New York: John Wiley & Sons, as well as Ken Tiahrt’s notes.

In 1972, I (MAH) accepted a 10% joint appointment to the WAMI (now WWAMI) medical education program to teach biostatistics and epidemiology to first year medical students. WAMI is a regional medical education program based at the University of Washington and serving Alaska, Montana, and Idaho, three states that don't have free-standing medical schools. The acronym WAMI was created from the names of the four participating states. Ultimately Wyoming joined and the acronym became WWAMI. At that time, MSU's WAMI students were admitted into the University of Washington School of Medicine, but received their basic medical instruction at MSU. Medical students were taught the same curriculum as at the University of Washington. A rigorous evaluation of this unique experiment in regional medical education required the instructors across the WAMI locations to follow the same syllabus and administer a common final exam. Success required coordination and extra effort, but the inter-campus interactions proved instructive and helpful.

During this era, MSU statisticians benefited from a new computer and statistical software for interactive computations:

Graduate Degrees Awarded: 1972 - 1974

The MS degree was awarded in 1973 to the statistics student Susan Marie Hinkins.

1974-1976 MSU catalog
The Department faculty consisted of 22 tenure-track professors, including four statisticians (McFeely, Tiahrt, Lund, Hamilton) who were the statistics faculty. Among the 14 instructors, 4 were statisticians.

In the statistics course offerings, the probability/stochastic processes sequence 420-1-2 was deleted. To provide a stand alone probability course, Math 407 – Probability was separately listed. Mathematical Statistics was the subject of the sequence 408-9. The book we were using at the time, Introduction to Mathematical Statistics by R. V. Hogg and A. T. Craig (1970) Englewood Cliffs, NJ: Prentice-Hall, was suited to that partitioning because its first few chapters were dedicated to probability from the basic axioms through the central limit theorem. A new graduate course on sampling was added:

Graduate Degrees Awarded: 1974 - 1976

MS in Mathematics degrees were awarded in 1975 to the statistics students James Cass, Penny Fangmeier, and M. Gail Linnell.

PhD degrees were awarded in 1974 to the statistics students James Leonard Hansen and Reider Sverre Peterson.

• Dr. Hanson ’s dissertation  “One at a Time Plans for 2^p Factor Sequencing Designs” was directed by Dr. Kenneth J. Tiahrt;¹⁸ 
• Dr. Peterson’s dissertation  “Ratio Estimation in Randomized Response Designs” was directed by Dr. Kenneth J. Tiahrt.²¹ 

 MSU Statistics Program Hosts the Visit of ISU Professor T. A. Bancroft―1975
     by Dick Lund

Funds became available for the Department to invite a renowned figure in mathematics or statistics to visit us. In fall 1974, Ken an ever-consummate promoter of the stature of our statistics program, brought to me a query, "Do you think we could get Dr. Bancroft to come out?" Within our further conversation, I replied, "Yes, maybe; I can try. What should I tell him?"

Bancroft (1907-1986) had retired from his administrative duties in 1972, but remained on the Iowa State University (ISU) faculty in a teaching position. He had just completed a tour in Japan that fall, visiting five universities, and presenting "Role of a complete Statistical Center in a large university."  He served as a consultant and on advisory boards for numerous U.S. and foreign universities.

I wrote a letter to Bancroft, telling him about our current program and noting recognition of some important issues. This brought his willingness for a visit on February 2-4, 1975. He offered two formal presentations and he met with us statisticians, our students, and several administrators. Bancroft responded by letter after returning home from his visit, stating "I firmly believe that Montana State has already made a good beginning in the development of a viable statistical program."

T. A. Bancroft lecture, MSU 1975

We statisticians were very pleased. He offered good advice and brought our needs to the MSU administration. While he made no mention of the vision Ostle (and likely he) surely held when initiating the statistics program at MSC back in 1952, nevertheless he clearly concluded that we were fulfilling Ostle's intentions.

Bancroft's two letters to Lund  add further information about this "milestone" in the history of the development of the Statistics Program at MSU. They show too the management style of a kind and gracious leader of the ISU Department of Statistics and Director of the Statistical Laboratory. [The following links take you to the letters, to return to this page use your devices "back" button or gesture: Letter of Jan. 23, 1975, Letter of Feb. 10, 1975] The "Iowa State – Chapingo, Mexico Project" referenced in the first letter is explained in the Narrative at the end of my Vita in About the Authors on the Title Page.

Tiahrt, Lund, Pres. McIntosh, &  T. A. Bancroft, MSU 1975

 

About T. A. Bancroft and His Statistical Leadership

Theodore A. Bancroft was born in Columbus, Mississippi, but with an early education in Alabama. He attended Howard College in Birmingham and received an A.B. degree in Mathematics from the U. of Florida in 1927. He taught in Alabama public high schools a few years and moved from there to the U. of Michigan, receiving an M.A. in Mathematics in 1934. His time at ISU began as a graduate assistant in 1941. He completed a PhD, with guidance by William G. Cochran, in the World War II year of 1943. His dissertation was titled "Tests of Significance Considered as an Aid in Statistical Methodology." His interest in testing and combining tests from across research studies remained through his career.

W. G. Cochran is the coauthor with Snedecor of the renowned textbook Statistical Methods, 6th edition (Iowa State U. Press) used in many of my (Lund's) statistical methods courses at MSU. Cochran followed the tradition forward of R.A. Fisher from his time at Rothamsted (Agricultural) Experiment Station in the mid-thirties, and that carried into Bancroft's views for statistics.

Bancroft was appointed Director of the ISU Statistical Laboratory and Head of the Department of Statistics in 1950, at which time there were 12 faculty members and 5 graduate assistants. When moving away from that leadership position in 1972, there were 30 faculty members and 25 graduate assistants. But he produced not just numbers, also he fostered the growth of service teaching, statistical computing, and statistical consulting. He initiated an undergraduate major program in statistics, in addition to promoting and enlarging the graduate program.

His professional stature is demonstrated by selection as President of the American Statistical Association (1970) and President of the Eastern North American Region of the International Biometric Society (1964). He was recognized as a Fellow of several professional organizations, including the American Statistical Association, the Institute of Mathematical Statistics, and the American Association for the Advancement of Science. He was an elected member of the International Statistical Institute. Many more details are offered in: Fuller, Wayne A. and Kempthorne, Oscar, "In Memoriam: Theodore Alfonso Bancroft, 1907-1986," The American Statistician, Vol.1, No. 3, August, 1987.

Bancroft coauthored with R.L. Anderson the book Statistical Theory in Research, which I (Lund) used in one course at UW and he wrote Topics in Intermediate Statistical Methods (now in 6^th edition), that was used in one class I took from him at ISU in 1964. He was my initial advisor at the start of my PhD program at ISU, and continued to be a mentor as my program advanced over three years as a graduate student. Bancroft was my direct supervisor in the ISU/Chapingo Mexico Project for developing a Graduate Program in Statistics and Economics at Mexico's National Agricultural School, this being funded by the Ford Foundation. More specifically, the institution is the Colegio de Postgradios, at Escuela Nacional de Agricultura located at Chapingo, Mexico.

              – Dick Lund

1976-1978 MSU catalog

For the first time, the Statistics program and courses were listed as a separate category within the Department of Mathematics. The heading was:

Statistics – Department of Mathematics – Head of Department: Dr. R. D. Engle.
Associate Professors: M.A. Hamilton, R.E. Lund, F.S. McFeely, K.J. Tiahrt.

In 1976, the probabilist Warren Esty joined the Department's faculty.
Dr. Warren Wilder Esty (1976-2015) received his PhD in Mathematics from the University of Wisconsin in 1973. His dissertation “Conditioned Limit Laws in Age-Dependent Branching Processes and Some Associated Diffusions” was supervised by Prof. Peter E. Ney. Warren came to MSU following three years of academic experience at Cleveland State University. The statistics faculty requested this position because the program needed an expert in probability who could contribute to both the statistics program and the mathematical analysis program. Esty perfectly fit the need. It turned out that, because of his communication skills and breadth of knowledge, he made important contributions to the mathematics education program too.

Ken Tiahrt spearheaded the effort to create the Montana Chapter of the American Statistical Association and it was formally founded in 1976. Then, as is still the case, the Chapter represented the entire state of Montana, with hot spots of members in Bozeman, Butte, Helena, and Missoula. The MSU Statistics Program has been the primary supporter of the Chapter throughout the years.

For all statistics courses, the rubric "Stat" replaced the rubric "Math." The course numbers, names, prerequisites, and descriptions were unchanged, except that some new courses were added. The statistics option and degree program were as follows:

This 1976-1978 catalog lists two new undergraduate courses and eight new graduate courses, all of which were introduced because of a university-wide attempt to document via credits the actual teaching effort expended by its faculty. These courses pertain to mentored work or supplementary education activities that merited recognition.

By 1976, an active Statistical Laboratory was functioning. It was described in two sections of this catalog:

In the Mathematics Department section:

Elsewhere in the catalog:

Graduate Degrees Awarded: 1976 - 1978

MS degrees were awarded to several statistics students, in 1976 to James Bergum, Milton Loyer, Lynda Steele, Wendy (Whitmore) Swanson, and Jack Williams, and in 1977 to Edwin G. Landauer, James Pellegrini, and Kent Sandefer.

PhD degrees were awarded in 1976 to the statistics students Dennis Patrick Brady and Charles Edward Shaffer.

• Dr. Brady’s dissertation  “Elimination of Continuous Variates Used in Classification and Discrimination When Both Binary and Continuous Variables are Present” was supervised by Dr. Kenneth J. Tiahrt;²⁴ 
• Dr. Shaffer’s dissertation  “Estimation of Cumulative Totals” was supervised by Dr. Kenneth J. Tiahrt.²⁷ 
  

  Stat Prog at MSU during Era 7
   Next topic (Adv in Statistics) during Era 6
  Table of Contents

 Last revised: 2021-06-06