110 Years of FPL: Remembering the Statisticians

In celebration of 110 years of research at the Forest Products Laboratory (FPL), we are revisiting blog posts that detail some of our most interesting historic people, places, and projects. Enjoy!

Statistics is commonly viewed as the collection, collation, and presentation of numerical data. FPL has long recognized that the field of statistics is critical for testing research hypotheses and making inferences to untested populations. Statistics has provided extensive and powerful tools for designing studies, analyzing data, summarizing or modeling data, and interpreting results for many research studies at FPL.

Continue reading

Throwback Thursday: Statistics Program Transcends Decades

The following blog is from the book Forest Products Laboratory 1910-2010, Celebrating A Century of Accomplishments.

This major study, preformed in the 1980s, is referred to as the “in-grade testing of structural lumber program,” and is one of the largest cooperative research programs ever undertaken by the North American wood engineering community.

It included universities, the Southern Pine Inspection Bureau, the West Coast Lumber Inspection Bureau, the Western Wood Products Association, a number of companies, and the Forest Products Laboratory (FPL).

Failure of a piece of structural lumber that was subjected to bending forces to help determine allowable design properties for structural lumber.


Physical and mechanical properties information was obtained on 33 species or species groups of visually graded structural lumber. Over an eight-year period, nearly 70,000 pieces of lumber — approximately 1,000,000 board feet — were tested to destruction in bending, tension, or compression.

The information provided the basis for more accurately estimating mechanical properties of lumber and revising allowable design properties.

Thanks to FPLs statistical design for selecting the wood samples, testing them, and analyzing the outcomes, the results from the study were useful and applicable to structural lumber in general. The data garnered from this research is now part of the National Design Standard, and is still used today.

Thowback Thursday: The Evolution of FPL Statistics Section

Statistical analysis is at the core of many scientific fields, and its use in research at the Forest Products Laboratory (FPL) is no exception.

The first statistical group, or section, at FPL was commonly known as “Computing.” Its job was to summarize test information generated by engineers and scientists. Because much of this was done by hand and simple adding machines, the work required a number of people.

Members of the Computing section, 1912.

Through the 1960s and 1970s, the section was known by a number of names, such as “Administrative & Technical Services,” “Mathematical and Computing Services,” “Biometrics Section,” “Statistics and Numerical Analysis,” “Statistical Research,” and “Statistics and Automated Data Processing.” The statistics and computing functions diverged into separate groups in 1980.

The computer group known as “Systems and Automated Data Processing” and later as “Management Systems,” took on the responsibility of implementing and supporting the national computer program, focusing on administrative processes with the Data General computer system and later IBM servers.


Diana Smith with Fisk University Sophomore Ethel Hunter using a 1960s IBM computer.

That group’s work at the lab ended in 2005 when computer support was consolidated at the Forest Service’s Washington Office.

The “Statistics Group” began in 1980 with one statistician. By 2001, the Statistics group consisted of three mathematical statisticians and two statisticians. It was turned into a research work unit (RWU), the “Statistical Methods in Wood and Fiber Research,” to better reflect the importance of statistics to FPL research.

In 2007, the Statistics and Economics RWUs were merged into the combined RWU that exists today.