Monthly Archives: January 2014

How Beer Helped to Improve Statistical Analysis

Posted by Nathan on

Statistical Significance

If you’ve ever had to measure the effectiveness of a new product or process design, you know how valuable significance testing can be. But did you know how beer made that testing much easier?

Relevance and significance are two of the most important concepts in modern statistical analysis. Relevance helps us to understand the connection between the research and the problem to be solved and significance tells us if that research represents a change that is greater than can be attributed to chance. While relevance is typically determined through methods of logic, significance is a purely mathematical exercise.

In most statistical trials, managers ask for large numbers of data points in order to reach a desired confidence level. Traditionally the magic number has been 30 data points. (Actually, there is no magic to the number 30. It just seems that in most distributions the standard error reduces to about 1.0 somewhere between 20 and 30 data points.)

 But what about when such large sample sizes are impractical?

Anyone who has ever taken an introductory class in statistics has learned to use the Student’s T-test of statistical significance. The “Student’s T-test” is one of the most basic tests of small-sample significance in statistics. But many people might be surprised to learn that it really has nothing to do with students, and almost everything to do with beer.

In 1906, William S. Gosset was a bright young chemist working for the Guiness Brewery in Ireland. Gosset developed a small-sample method for measuring the deviation of means in the production of Guiness’ dark beers. This allowed for accurate tests of significance without the need for excessively large sample sizes.

His method was so successful he submitted it for publication in Biometrika, a professional journal published by his friend and professor Karl Pearson. And here, believe it or not, is one of the few times in history where a story about math gets interesting.

A popular version holds that Guiness had a policy against its employees publishing their work. So Gossett compromised with his employer and used the pen name he had published under before: Student. Some still like to contend that Gosset never let his employer in on the fact that he was publishing and submitted his work under cloak and dagger. Probably not true, but makes for intriguing reading.

It is far more likely that Gosset published under Student because Guiness did not want its competitors to learn that it was using statistical analysis as a part of quality control. This was a new concept at the time and the brewery probably wanted to keep it under wraps as long as possible. That did little to diminish the drama, however, as the new formula only added to the animosity between Professor Pearson and one of his chief rivals R.A. Fisher, with whom he continued to bicker for many years.

And it all started over a glass of beer.

So, the next time you are tasked with testing the significance of a difference in means with a small-sample data trial, it might be appropriate to raise a glass of beer to good Mr. Gosset.

Improper Design of Experiment: How 1970s Sexism Almost Killed the Computer Revolution

Posted by Nathan on

Sexism in Experiment Design

All the marketing research technologies in the world cannot overcome the disadvantages of poor research planning.

As marketers we get excited about the potential of advanced tools like neuromarketing and predictive analytics but they are rendered worthless when coupled with a poor design of experiment. The design of experiment (or DOE) helps us to structure our research questions to maximize the relevance of the results. In short: are we asking the right question and are we asking the right people? If the answer to either is no, then it really doesn’t matter what confidence level our research returns; the results are pointed in the wrong direction. As a statistics mentor of mine once said, significance without relevance is simply a way to go wrong with confidence.

In the 1970s, one company’s research team had struck gold in the fledgling personal computer industry. They had a phenomenal product and they were asking the right question. But, arguably due to a sexist view of a woman’s role in the workplace of the future, they asked the wrong people. They failed in design of experiment.

Birth of the Personal Computer Age

The personal computer, the mouse, email, word processing, the first paint program, the Ethernet and the laser printer were all created and introduced by one company in 1977. That company was Xerox. But a marketing blunder fueled in part by 1970s sexism sent each of these products out to be development by other companies.

In 1969, at Rochester, NY-based Xerox Corporation, CEO Peter McColough and his chief scientist Jacob Goldman had the foresight to steer the world’s leading copier maker directly into the computer age. They purchased a computer company in Palo Alto, CA and created a pure research facility filled with high tech equipment, bean bags, sodas and some of the brightest minds in the still young computer industry. The facility would become known as Palo Alto Research Center, or PARC, and it would develop for Xerox a modern personal computing paradise. And it would be lost.

A Stellar Nursery

When McColough and Goldman developed PARC they did everything right. They picked the right people, they made available the right resources and they gave it an almost unlimited budget for research and development. Its staff list was a veritable Who’s Who of the computing world: George Pake, Robert Taylor, Butler Lampson, Alan Kay, Adele Goldberg, Lynn Conway and many others.

Their directive was simple: Use your imagination and skills to determine where this company needs to go with computers. And that they did. Within seven years they had developed—or perfected—all the components of the modern personal computer. They had created for their Xerox bosses a desktop computer called the Alto featuring the first WYSIWYG (What You See Is What You Get) display. The software package included the first word processing program. It could send email to other computers via the Ethernet, a new network protocol they developed just for that purpose. Documents could be printed out on a lightning-fast laser-driven printer. And the system could be controlled by a revolutionary pointing device they called a mouse.

Excitedly, they prepared to present their new system to the Xerox bosses.

Paradise Lost

The PARC team spent weeks preparing for their executive presentation. Believing that a live demonstration would make the most impact, they got several of the Xerox secretaries to help them train a handful of the executives for a live show. It was the 70s, and female secretaries did most of the real office work—typing, filing, communicating and scheduling. Who better to help train the male executives?

On November 10, 1977, the revolutionary system was presented to the Xerox brass at an aptly named event called Futures Day during a conference in Boca Raton, FL. The top executives watched as some of their counterparts took the stage, writing documents, sending messages back and forth to connected computers, controlling software with the space age pointing device and printing at the speed of light. But the dramatically staged presentation closed to the sound of crickets.

The executives were largely underwhelmed. Most of them were former copier salesmen who were used to rating their business success on the volume of a copier’s clicks—the copy counter, the currency of the old Xerox world that told the company how successful the machine was. The executives looked on this new hardware the way a 15th century sail maker might stare confusedly at a jet engine. They could not understand how to make use of this new world of digital connectivity and productivity.

But the people who could understand it were the executives’ wives. In that 1970s world of Xerox the salesmen’s wives were mostly former secretaries, the backbone of the office, the job description that included the tasks of typing, filing, scheduling, communications and planning that kept the office running, the very tasks this new system was designed to revolutionize. While the executives largely avoided the elaborate demonstration suites the engineers had set up to entertain post-demonstration interest, their wives flocked to it. They wanted to see the machine, the software, the new pointing device, and the laser printer. Their praise was effusive and on point. They knew this would be big stuff.

But, Xerox being a corporation of the era, listened to the silence of its executives. While top management hesitated, wondering what to do next with all this stuff, its architects quietly left to work for—or, in a few cases, found—such companies as Apple, Sun Microsystems, Microsoft and Adobe Systems. The revolution would take place elsewhere.

The Responsibility of Marketers

The failure of Xerox to exploit the fruits of PARC’s labor can’t be blamed entirely on myopic sexism; given the era, there surely were many companies with the same malady. What seems to have been missing was a voice of reason that could have helped top management and the engineers overcome their prejudices and direct their critical questions to the people who would best know the product’s potential.

As marketers, we are very rarely the innovators in our companies. But we generally are the best communicators, and we often understand better than most the complexity in the relationship between our products and our stakeholders. We have a responsibility to the innovators on our teams to help them understand the markets they are focused on, to see through the prejudices in the market place and to properly define the battle spaces where future revolutions will be taking place.

Helping talented engineers solve the right problem is a small part to play in a revolution. But it is no less critical.

________

Further Reading:

The definitive book on the Xerox PARC story is Michael Hiltzik’s Dealers of Lightning: XEROX PARC and the Dawn of the Computer Age. It is a highly entertaining read with critical lessons for anyone in the fields of engineering or marketing.

How the Brain Connects Music to Advertising

Posted by Nathan on

Music and Advertising

Originally Published May 2013

There’s little doubt that music improves the quality of advertising, or any visual presentation for that matter. But we rarely give much thought to the physiological factors that create music’s impact.

I should disclose up front that I am not a professional music theorist. My knowledge on the subject comes through the teachings of others and from some professional application. Although I’m an experienced drummer and visual artist, I can claim no original discovery in the field of music theory.

Understanding the Power of Music

The philosopher Plato, writing in his epic work The Republic, recognized that music had the ability to enter most powerfully into the innermost part of the soul and lay forcible hands upon it. He not only knew music’s power; he was somewhat wary of it. Plato, like many of the ancient philosophers of his day, was in love with rational thought and had no patience for any other type of thinking. In fact it is their prejudice handed down over hundreds of years that has led us to look down on irrational thoughts and actions. However Plato also knew that emotions could be powerful motivators of human behavior. And music seemed particularly able to evoke these emotions. But he had no idea why.

Most often when we ask anyone in the advertising business to quantify the positive effects of music, they turn first to a series of statistics showing the preference audiences have for ads and presentations that include a soundtrack. But what drives that preference? Would any music do? What’s the difference between music and noise? If music is universal, why don’t we all like the same kinds of music? Lots of questions. And, surprisingly, there are some answers.

Music is a complex structure of many components. Two of the most important, particularly from a physiological perspective, are rhythm and melody. Rhythm is the pulse and pattern that form the clock speed of music while melody contains the pitch profile, what we might refer to as the tune. Rhythms exist throughout nature and originate in our own bodies as heart rates and breathing rates.

Melody has roots in the natural world as well. Much like the Golden Mean and other mathematical ratios that provide a foundation for what we consider to be visual art, fixed ratios of pitch—believed to have first been identified by the philosopher Pythagoras—are perceived by the ear in a similar fashion and form the basis of how we distinguish music from noise.

Melodies are thought to mimic the pitch and frequency profiles of human emotions. Dirges and requiems have been said to imitate the wails and moans of sinners and mourners while the more upbeat works of happier composers reflect the quick, staccato exhalations of laughter. Other researchers have suggested that the similar pitches of sustained high notes on the violin and infants’ shrieks are what makes horror movie music so effective.

But to fully understand just how these aspects of music can affect the listener’s brain, it is helpful to review a little about the way incoming information is processed.

How the Brain Listens

The brain receives input from a variety of different channels—visual, auditory, olfactory, tactile, etc.—and this information is transduced into data streams that are coming in at varying data rates. This input goes through a bundle of neurons called the reticular activating system or RSA, which filters out very high data rate information (the 26 TVs and 18 conversations around you in a restaurant) and the very low data rate information (the clouds and sun moving across the sky).

The filtered sensory input then goes through the limbic system which consists of several different structures along a pathway known as the Papez circuit. The limbic system is an older part of the brain that was once thought to be the very center of emotion, and its structures have demonstrated the ability to store information that is emotionally important such as threats to our existence (negative emotions like fear) or experiences that make us exceedingly happy. These limbic structures compare incoming information to the stored copies and, if they find a match, they trigger a hypothalamic response through the autonomic nervous and endocrine systems that we perceive as a physiological change. In other words, we feel an emotional response.

All of this happens before any information is sent to the conscious neocortex. For instance, if you are in the woods and you notice something wiggle at your feet, your heart rate increases, your diaphragm contracts sharply, and you jump back with a sudden gasp before realizing it was just a piece of rope you had stepped on. The limbic system, based on visual information that matched that of a snake stored in memory, produced an autonomic response before the cortex had an opportunity to analyze what had actually happened.

Another relevant physiological function is entrainment, which is the body’s way of adapting to a new condition or environment. It can do this in a number of ways—physically, biochemically, etc.—and it’s the process the body uses for learning habits and assimilating new patterns. Imagine a band begins to play and the audience begins clapping along. A spattering of hand-smacks quickly condenses into a roomful of people clapping in unison. That’s entrainment.

Music’s Assault on the Brain—and the Body

We perceive music in a number of ways, not just through our auditory senses. The pace and pulse of the music’s rhythm produce waves of energy in the air that impact various receptor cells in many parts of the body, particularly in the skin and subcutaneous tissues. The pitch profile of the music’s melody is likewise felt in addition to being heard. All of this information across a range of senses is delivered to the RSA, through the limbic system and on to the neocortex.

At this point the music’s rhythm and melody have been perceived through a variety of senses and have begun to trigger emotional matches in the limbic system as well as conscious associations in the neocortex. The listener is feeling the effects of the positive emotional responses at the same time she is consciously remembering the band, the tune, or the happiness she may have felt the first time she heard the song.

If she makes a positive association with the music in her neocortex, her body begins to entrain itself to the rhythm and melody. She now begins to assimilate the emotional expression contained within that music.

If all this is happening during a commercial, the music has opened a series of neural pathways leading to the most trusting part of her brain just in time for the sponsor’s logo or tag line to be seen, heard and associated with the positive emotions.

The result is a positive emotional experience—part conscious and part subconscious—that will be associated with the visual presentation the music accompanies. And emotion has been shown to have a strong influence on behavior, much stronger than cognitive factors.

This is great news for advertisers, and it underscores the value music can bring to a visual presentation. Music has the ability to engage multiple sensory systems, penetrate deeply into the body, activate the emotions through the limbic system, and produce physiological effects that influence behavior.

Plato, it seems, was right all along.

___________

For Reference and Further Reading:

The Emotional Brain. LeDoux, Joseph. (1996). New York, NY: Simon and Schuster.

The Music Effect: Music Physiology and Clinical Applications. Schneck, Daniel J. and Berger, Dorita S. (2006). London, UK: Jessica Kingsley Publishers.