Mark Mobius on the measurement of ‘meaning’ and why it matters

“WHAT do you mean by that?” It’s a question often asked in a conversation. When someone says “I love that building”, it is bound to elicit a response asking for more clarification. “What do you mean by that? How do you like that building? Is it architecturally beautiful? Do you like its size?” Or “I think she is beautiful.” “What do you mean by that? Do you think she is stylish, hot, strong or what?”

Words are supposed to have meanings, but a single word or phrase cannot truly encompass the totality of meaning. It is this challenge to communication that people in the advertising industry, the politicians trying to win an election, and people in any endeavour requiring communication have. The problem has been the topic of scholars for quite some time, and some have come up with intriguing answers.

When I was studying at the University of Wisconsin in 1960, one of my professors was Percy Tannenbaum. In 1957, he wrote a book with Charles Osgood and George Suci called The Measurement of Meaning, which presented a fascinating theory of what constituted meaning.

Osgood was a prominent American psychologist who made significant contributions to the field of semantics and psycholinguistics. He was born in 1916 and died in 1991, and he received his PhD from Yale University. Osgood’s research on semantics and the semantic differential had a big influence in a wide variety of fields.

My professor, Prof Tannenbaum, did a lot of work applying the semantic differential to practical applications, even in assessing political candidates’ campaigns. In his lectures, he focused on how meaning is understood by individuals. His work helped psychologists understand human cognition and behaviour.

Suci was also an eminent scholar in psychology and psychometrics. His contribution to the work was in the area of measurement techniques. His contributions to the field include his work on the development and refinement of measurement techniques, as well as his insights into the nature of meaning and its representation in human cognition. At Cornell University, he and his students studied early learning behaviour in children.

The central focus of The Measurement of Meaning was the “semantic differential”, a construct which provided a systematic method of quantifying the meaning of objects, people, concepts, paintings and anything else subject to people’s subjective evaluation.

Nobel Prize-winning psychologist Daniel Kahneman’s doctoral thesis was on the subject of the semantic differential. The idea was that consciously or unconsciously whenever we are seeing, hearing or feeling something, our mind develops a perceptive structure related to a whole series of opposite scales, such as hot/cold, good/bad, active/passive, strong/weak, and the placement of where on each of those scales the subject matter sits creates “meaning” for us.

It is a way to measure people’s subjective evaluations on any feeling, picture, action, sound, colour, building, person, and so on. The idea was that people evaluate and perceive concepts or objects along opposite dimensions such as hot/cold, strong/weak, active/passive, and that the sum total of such dimensions creates meaning. This method the psychologists created provided a quantitative framework for understanding human “meaning”.

At a practical level, applying the system involved asking respondents to rate a concept or products by using that series of bipolar scales or pairs of adjectives with opposite meanings. By analysing the respondents’ selections on those scales, the researchers can gain insights into the subjective meanings related to different products or ideas.

The semantic differential can be and has been used in a wide range of fields such as marketing, psychology, sociology and linguistics. The advertising industry quickly caught on to the idea to use it in creating a “meaning space” for products, so that advertising programmes could be developed to create the optimal image for a product or service. It acts as a useful tool for understanding how people perceive products and enables advertising agencies, for example, to modify their appeal to the consumer.

At the University of Wisconsin campus in Madison, we had fun, with the encouragement of Prof Tannenbaum, to make actual graphic three-dimensional representations of the results from scores of the various adjective opposites as represented by a line with the zero point as the axis for the neutral point between such opposites as hot/gold, big/small, and so on. A whole set of adjectives were developed for the areas of beauty, power, motion and work, among others.

The system has not only been used for benign purposes such as in advertising, but also in psychological warfare. In 1958, the US Central Intelligence Agency (CIA) funded Osgood to do a global study of 620 key words in 30 cultures using the semantic differential. The work was used by the CIA to create more effective culturally specific propaganda in order to destabilise foreign governments. For example, the semantic differential was applied to identify words that would be most effective in creating negative attitudes in Chile towards the socialist Allende government.

Factor analysis

Roget’s Thesaurus was an early effort to classify most adjectives into categories by English speakers. It was used by researchers to reduce the number of adjectives used in the semantic differential to manageable subsets.

Factor analysis by Osgood and his team resulted in the finding that three recurring attitudes most often occur that people use: evaluation (for example, good/bad), potency (strong/weak) and activity (fast/slow). Of course, to get an in-depth understanding for how a person perceives something, it was necessary to use an entire set of adjectives such as bright/dark, sweet/bitter, fair/unfair, for each of those three constructs.

Today, the semantic differential is used for website designs in addition to other Internet applications. For example, onscreen typefaces can be evaluated in terms of potency, evaluative and activity dimensions. Typefaces with high potency would score high on strength, power and force. Those with high activity would score high on energy, movement and action.

Typical scales for activity would be passive/active or loud/quiet. For the evaluation construct, the adjectives would include cheap/expensive, beautiful/ugly or bad/good. Potency opposites would include hard/soft, delicate/rugged and relaxed/stiff.

With burgeoning global Internet traffic, words and their meaning are becoming more and more critical. It is estimated that each day, more than 500 billion messages are sent over the Internet. This includes the more than 360 billion e-mails sent daily and the estimated 140 billion messages sent over WhatsApp. Then there are Facebook Messenger and other platforms, with Facebook Messenger having more than seven billion conversations daily. Of course, that only covers words. There are also images and sounds that are also communicated globally which all have their own meanings. These too, can be measured by the semantic differential. We have an exciting research path ahead in this field.

Gaining insight

In the area of investments and finance, the measurement of meaning is a critical qualitative aspect regarding the financial content of news articles, earnings call transcripts from companies, and general Internet chatter of the main websites such as YouTube, X and Instagram. If sentiment analysis indicates a negative attitude towards a certain company, that can trigger a fall in the stock price.

Another use of the measurement of meaning would be to analyse the meaning inherent in the publications by a company regarding ESG factors. By assessing a firm’s approach towards environmental, social and governance factors, investors can assess the commitment of that firm to socially important objectives such as climate change.

The economist Robert Shiller studied how popular narratives drove economic and investor behaviour. By measuring those narratives for various trends, the investor can gain insight into future market trends. So, a narrative of optimism in media content can enable the investor to predict an upward stock market movement.

Important for new product development and even new business development would be the measurement of meaning in popular culture regarding, for example, health consciousness. Such a trend would provide an opportunity for a new health-oriented business.

All of these kinds of analyses involve social media analysis, where we monitor and analyse social media dialogue. Also, content analysis of reports, speeches and other output is critical. Public surveys and polls are another tool to obtain a measure of investor sentiment. Finally, and increasingly important, machine learning algorithms are being increasingly used in business to find trends and patterns in public data.

The writer is an investment professional with over 30 years of experience and is co-founder of Mobius Capital Partners