Have you ever seen your present reflected in an object from the past? This summer I’ve discovered glimpses of my daily life working with statistical software in words written more than 70 years ago. Bear with me because this blog post takes the scenic route to arrive at modern statistics.
I’ve been reading through a treasure trove of old love letters that were sent in 1940-41 by my grandfather to my grandmother, who saved them all of these years. This was an exciting time of changes for them both. My grandfather was a farm boy from a small town in Connecticut, who grew up without electricity. He became an engineer and, by the time of these letters, he was working on top-secret defense projects for Sperry Gyroscope in Brooklyn. My grandmother had recently completed college and had just moved from her small hometown elsewhere in Connecticut to start her first year of teaching in a one-room schoolhouse. They started dating in the summer of 1940 and the letters started that August.
Reading them took me back to a time where train travel was the norm and telephones were a luxury. The distance between Connecticut and Brooklyn seemed much greater back then than it does today! The slower times made it more difficult to see each other and maintain their relationship. Letter-writing became their main form of communication during this first year. The importance of these letters is made evident by a photograph my grandfather took of his writing desk as he wrote one of his letters. You didn’t take photos willy-nilly back then!
As time went by, they begin corresponding about marriage. My grandmother made my grandfather work to get a “yes.” I’m rooting for you, Grandpa! In the spring of 1941 he got his wish as I read in the first letter that begins “Dear Wife-to-Be”! Ultimately, they married on July 5, 1941.
Hmm. My grandfather was an engineer, but he also loved writing…maybe that’s where I get it from?
A Glimpse of the Future
My grandfather wrote of courtships and radio programs, but did not mention television. The 1940 election was a current event, and cars were a luxury. But one thing in his letters struck me as decidedly modern: the nature of his top-secret project. He couldn’t say anything specific about it; he wrote about working hard in the lab and his frequent travels to assist with demonstrations for military brass.
But I knew what he was working on from my conversations with him while he was alive: a mechanical analog computer that controlled anti-aircraft fire.
Targeting aircraft is complicated. You can’t just shoot at where the aircraft is, because it will be in a different location by the time the shell arrives. Given the higher altitudes and faster speeds of the new planes in World War II and the long ranges of the big guns, visual aiming was just too inaccurate. To increase accuracy, you needed to accurately measure a number of variables and use the data to predict the aircraft’s future location, adjust for various environmental factors, account for the curved trajectory of the shell, properly aim the gun, and calculate the flight time for the fuses. And, of course, you needed to do all this in real time!
To meet this challenge, sophisticated fire-control systems gathered data, such as wind direction and speed, air temperature, target location, and the speed and direction of the target. The analog computer performed complex calculations in real-time to point the guns so that the shell arrived at the correct point, at the correct time, and with a properly timed fuse. All of this was done with 11,000 parts that formed an intricate clockwork of gears and cams!
The idea of collecting data and using a computer to analyze it in real time to help people perform a task beyond human abilities is common today. But in 1940, it was a glimpse of things to come!
Statistical Computation Over the Decades
Just as the mechanical computers and fire-control systems of 1940 gave a glimpse of the future, so did the first version of modern statistical software over 40 years ago. Back then, it was distributed in boxes of punch cards for use on a mainframe computer. Prior to the rise of mainframes, statistics mainly had been calculated in statistical labs that employed many human calculators. These people worked with punch-card tabulators, mechanical adding machines, and slide rules to collectively perform an analysis. Calculations were slow and expensive, but without the labs even limited statistical analyses would have been impractical.
With statistical software, computers could perform the tedious statistical calculations, freeing students to concentrate on learning what really matters — the statistical concepts. The use of statistical software on mainframes set the stage for the future. As personal computers became more powerful and cheaper, more and more people gained access to computation power that far exceeded the statistics labs of the past!
Today, statistics and its benefits are everywhere. Individuals can easily run complex analyses in a matter of seconds. This power allows anyone to understand experimental data and draw the correct conclusions. There are no scientific, medical, environmental or political studies that do not rely on statistics. The quality of our lives is literally improved by statistics because quality engineers can use statistical software to produce more reliable and safer products. Statistical process control is easier than ever. And Six Sigma projects regularly include statistical techniques to monitor and improve quality.
Because statistics are so pervasive now, I’d argue that statistics should be a standard part of the high school curriculum. After all, the most difficult issues of modern times are often statistical questions. For example:
- Is global climate change a real trend?
- Is the economy improving?
Many daily life issues are also statistical questions:
- Do vitamins have health benefits?
- Does childhood vaccination cause autism?
Without statistics, you can easily get lost in the glut of data that we have!
Speaking of words from the past that illuminate the future, I think H. G. Wells’ prediction from a century ago is accurate for today:
“Statistical thinking will one day be as necessary for efficient citizenship as the ability to read and write.”