Editor’s note: Dave Fish is strategic research manager with Toyota Motor Sales, U.S.A. Inc., Torrance, Calif.

The terms “psychographics” and “lifestyle segmentation” have permeated the marketing literature and practice over the last several years. The intent of this short review is to clarify what is meant by these ill-defined and imprecise terms. The following provides a brief overview of how consumer psychologists view the components that drive human behavior.

There are roughly five intrapersonal classes of attributes, each of which varies in terms of malleability, generalizability, and predictive behavioral power. Malleability refers to the ease with which each structure is changed by outside forces (i.e., communication). Generalizability refers to the degree with which a given class of attributes will be manifested in the form of commonalties across a range of behaviors. Predictive power refers to how accurate a prediction can be made from the given class of attributes. Towards the center of the different classes are those that are most resistant to change, can be generalized across a wide range of behavior, and are the least predictive of specific behavior. Moving outward results in structures that are less resistant to change by outside forces, are less generalizability to a wide class of behaviors, but offer greater predictive validity.

Only one of the five layers is observable (behavior). Any one of these levels (or a combination of several) can be used as a basis for segmentation. These levels are also used in a variety of other research efforts pertaining to product development, marketing, and advertising. Each class of attributes is addressed separately below, starting from a person’s core (personality) and working outward.

1) Personality - Most psychologists agree that personality has a high degree of “heritiability” (i.e., it is passed along from generation to generation) and is at least partially innate. Most psychologists also agree that a person’s personality can be shaped to some degree by early life experiences. While there is long-standing controversy over the degree to which nature vs. nurture contributes to one’s personality, there is general agreement that personality is fairly fixed before a person’s early teens. Recently, there has been some consensus on the underlying factors of personality (Digman, 1990; McCrae, 1992). Called the 5PF personality factors (or the Big Five), they include: extro-/introversion, openness to new experiences, conscientiousness, narcissism, and agreeableness. Older personality schemes (e.g., Myers-Briggs, 16PF, etc.) are still employed but with less regularity. The Big Five have been, more or less, accepted by the academic community. Personality can only be reliably measured by validated psychological batteries that tend to be fairly long.

2) Values - According to Milton Rokeach, the father of values, a value is “an enduring belief that a specific mode of conduct or end-state of existence is personally or socially preferable to an opposite or converse mode of conduct or end-state of existence.” Values are formed very early in life and do not generally change in the course of one’s lifetime. Values are such concepts as freedom, happiness, and obedience. They are generally based in and influenced by one’s culture, religion, and moral foundation. Methodologically, it is recommended that they should only be measured in an “ipsative” or ranking fashion, since they have a high degree of social desirability and can only be measured in a relative sense. Values have also been derived through second-order factor analyses of attitudes. Values may manifest themselves across a wide band of behaviors but are not effective in predicting specific behavior.

3) Core attitudes - Core attitudes tend to be those that guide one’s life. These are also sometimes called zero-order attitudes. Attitudes such as a belief in democracy or monogamy are examples of zero-order attitudes. Attitudes can be influenced by heritiability, but are generally formed by life experiences and behaviors. Core attitudes are usually shaped early on in life and are difficult to change. Attitudes are typically measured with a Likert-summated rating scale or on occasion with a Guttman or Thurstone approach. All scales constructed must be validated and proved reliable through pre-testing before use.

4) Attitudes - This level refers to attitudes which are specific to a certain object. This is typically where needs-based segmentation takes place. This level tends to be more malleable than core attitudes. For example, my attitude towards four-wheel drive systems will not be as deeply held as my opinions about equal rights. For that reason, these “lighter” attitudes are easier to influence through communication. In many cases, these attitudes may not even exist at all. For example, most Toyota Corolla owners probably do not have an opinion about locking differential systems. In these cases attitudes can be significantly influenced by communication.

A special class of attitudes called behavioral intentions are used to predict future behavior. Measurement efforts should specify the target, the action, the context, and the timing of the behavior (Fishbein and Azjen, 1980). For example, in predicting consumers’ likelihood to purchase a vehicle, my questionnaire item might read: “I plan on purchasing a new LS400 in the next six months from a Lexus dealer.” The target is the new LS400, the action is purchasing it, the context is a Lexus dealer and the timing is the next six months. Behavioral intentions are the best choice for predicting specific future behavior.

5) Behaviors - The only overt manifestations that are commonly measured, behaviors are usually the outcome of the attitudes, values, and personality of a person. Behaviors may be product choices, lifestyle activities, product usage, or any other overt and measurable behavior. This is typically the level where lifestyle segmentation takes place. However, you can reasonably assume that attitudes, values, and personality usually drive a person’s lifestyle. What is known as lifestyle segmentation uses behavior to infer (predict) other associated behaviors (e.g., product preferences). This approach has met with mixed success, depending on the application. Behaviors can be used as a source of segmentation or to profile segmentation based on covert levels (attitudes). Obviously, they have the highest validity of prediction (they are reporting behaviors!). The same can not be said for the lifestyle application of this level (behaviors to predict other behaviors). Behaviors are also the most malleable class of attributes.

References and further reading

Ajzen, I. and M. Fishbein. Understanding Attitudes And Predicting Social Behavior. Englewood Cliffs, N.J.: Prentice Hall, 1980.

Digman, J. M. “Personality Structure: Emergence of the Five-factor Model.” Annual Review of Psychology (1990), 41, 417-440.

Kahle, L. and L. Chiagouris. Values, Lifestyles, and Psychographics. Mahwah, N.J.: Lawrence Erlbaum Associates, 1997.

McCrae, R.R. (Ed.). “The Five-Factor Model: Issues and Applications.” Journal of Personality (Special Issue). 60(2), June 1992.

McAdams, Dan P. “The Five-factor Model in Personality: A Critical Appraisal.” Journal of Personality (1992), 60, 329-361.

Myers, J.H. Segmentation and Positioning for Strategic Marketing Decisions. Chicago: American Marketing Association, 1996.

Oksamp, S. Attitudes and Opinions. Englewood Cliffs, N.J., Prentice Hall, 1991.

Rokeach, M. Understanding Human Values (2nd edition). New York: Free Press, 1979.

Wedel, M. and W.A. Kamakura. Market Segmentation: Conceptual and Methodological Foundations. Boston, Kluwer Academic Publishers, 1999.