The IBM 805 Test Scoring Machine
By streamlining a costly, time-intensive process, the 805 set a new standard in testing procedures for large-scale evaluations
A smiling woman sits at the console of the 805 scoring machine, black and white

In 1931, Reynold B. Johnson was working from his basement in Ironwood, Michigan, when he made an auspicious breakthrough. Just a few years prior, the Scholastic Aptitude Test, or SAT, had debuted in US high schools as a means of qualifying students for university admissions. Outside of academia, similar tests — in multiple choice or true-or-false formats — were on the rise in professional settings. And so Johnson, a high school physics teacher, was working on a tool to help expedite the tedious scoring process.

His invention achieved that and more. An electrical machine, it used tiny circuits that responded to the electrical conductivity of the lead to detect pencil marks on answer sheets — a technique that would become known as “mark sense” technology. The machine then compared the answers with a key, which had been installed in the machine, to tabulate a score.

While Johnson’s efforts led to an impressive breakthrough, he had no means of producing the machine at scale. So when he caught wind of a similar, dormant test-scoring project at IBM, he submitted his designs to a consultant who was involved and rekindled the company’s interest. What started with some basement tinkering soon became the IBM 805 Test Scoring Machine, one of the company’s most foundational technologies at the time, making possible a new era of test taking and scoring, and launching Johnson into a celebrated career as a lifelong IBM researcher.


A meeting of minds

Before the 805, test scoring — especially on a large scale — demanded substantial amounts of time and money. It was clearly a problem that needed solving. While Johnson chipped away at his designs, Ben Wood, a professor of education and research at Columbia University and an IBM consultant, had independently begun work on his own scoring machine with the blessing of IBM founder, and stalwart education advocate, Thomas J. Watson.

After two years of experimentation, the IBM project came to a halt. Wood, like Johnson, had based his design on the conductivity in the graphite pencil led. The methodology was sound, but the results were erratic. When the pencil marks were recorded with less intensity — and therefore, conducted less electricity — Wood’s device became less responsive. So the lighter the mark, the less accurate the score would be.

In 1934, Johnson heard of Wood’s research and submitted a description of his own design. Excited by what he saw, Wood called IBM. “Dr. Wood saw the possibilities in the test scoring machine, but had a tough time selling it to management until he took it to Thomas J. Watson. [Sr.],” Johnson explained in a 1971 interview with the IBM publication THINK. “Mr. Watson immediately grasped the concept and its commercial possibilities.”

In the fall of 1934, on the strength of Wood’s recommendation, Johnson was hired as a senior engineer at the IBM Endicott Engineering Laboratory and immediately went to work in developing the first production model of what would become the IBM 805 Test Scoring Machine.

Dr. Wood saw the possibilities in the test scoring machine, but had a tough time selling it to management until he took it to Thomas J. Watson [Sr.] Reynold Johnson, co-creator of the 805 Scoring Test Machine THINK magazine
The new era of scoring

Unlike Wood’s earlier design, Johnson’s machine skirted the pencil-mark problem by introducing high-resistor units into the electric circuits, raising the total resistance to the point where any variations in the pencil marks no longer mattered. Test takers needed to record their answers on specialized “mark sense” cards, accommodating 750 responses to five-choice questions, three-choice questions, and true-or-false questions.

The machine itself resembled a desk, behind which the scorer could stand — and later sit — to feed pages through a slot. Inside the 805, a contact plate with 750 small, circular electrical circuits, corresponding directly to the 750 answer positions on the cards, read the test. Guided by a scoring key, answers were marked as right or wrong.

The 805 machine accelerated the typical scoring process for a 150-question test by a factor of five or even 10, which lifted a significant burden from many institutions. In fact, the only true limitation on the 805’s output was the rate at which a scorer could feed tests into the machine. And, according to one company report, if users followed procedures correctly, absolute accuracy could also be guaranteed in more than 99% of results.

Objective testing had entered a new era, with the help of IBM’s mark sense technology. Throughout the coming decades, the Educational Testing Service (ETS), which administers the SAT and several other large examinations — including by municipal services — pioneered the use of the 805 and its mark sense technology. During World War II, the 805 was even tasked with the scoring of tests used in the placement of recruits.

The only true limitation on the 805’s output was the rate at which a scorer could feed tests into the machine
IBM 805 5-10x faster in the typical scoring process for a 150-question test 99% guarantee of absolute accuracy in results 
A mark sense legacy

By the 1960s, the IBM 805’s electrical conductivity method had been replaced by optical mark recognition (OMR) systems for surveys and tests. Meanwhile, “mark sense,” the trade name used by IBM for electrographic forms and systems, had developed into a generic term for any technology that processed written marks through both OMR and electrographic technology. By January of 1963, the IBM 805 Test Scoring Machine was officially withdrawn from production.

By then, the machine had cemented Johnson’s legacy. His achievements with the IBM 805 as well as his subsequent breakthroughs in large-scale testing machines firmly established the former teacher as one of IBM’s most innovative engineers — with 90 patents to his name. After several years at IBM’s Endicott laboratories, Johnson moved to San Jose in 1952 to manage IBM’s first West Coast laboratory. There, he led a team in the development of the world’s first commercial magnetic hard-disk drive, a technology that would revolutionize data storage — and would earn Johnson the distinction of “father of the disk drive.” The successes of Johnson and his team also helped define Silicon Valley as the center of technology development in the United States.

Johnson was named an IBM Fellow in 1965 and awarded the 1971 Machine Design Award of the American Society of Mechanical Engineers. The award cited his “many outstanding contributions to the educational and data processing fields through his numerous ingenious inventions and innovations dating back to the development of the first electric test-scoring machine.”

Related stories Reynold Johnson

A former high school science teacher became the father of the hard disk drive

The importance of education

IBM has pledged technology, expertise and funds to students and teachers around the world

The importance of education

IBM has pledged tech, expertise and funds to students and teachers worldwide

Computer science

IBM fostered a highly consequential field to find new utility for its machines and to train the next generations of technical talent