A Temporal and Lifestyle Typology to Model Consumers' Smoking Behavior

Jacob Hornik, Tel Aviv University
ABSTRACT - This paper delineates and empirically examines consumers' cigarette smoking behavior. Fundamental elements from psychographics are integrated with demographic variables and time preference concepts to form the basis for a smoking typology. The data provide evidence to the effect that smoking behavior should be viewed as a decision-making process involving economic-like choices along certain psychographic determinants.
[ to cite ]:
Jacob Hornik (1989) ,"A Temporal and Lifestyle Typology to Model Consumers' Smoking Behavior", in NA - Advances in Consumer Research Volume 16, eds. Thomas K. Srull, Provo, UT : Association for Consumer Research, Pages: 44-50.

Advances in Consumer Research Volume 16, 1989      Pages 44-50


Jacob Hornik, Tel Aviv University


This paper delineates and empirically examines consumers' cigarette smoking behavior. Fundamental elements from psychographics are integrated with demographic variables and time preference concepts to form the basis for a smoking typology. The data provide evidence to the effect that smoking behavior should be viewed as a decision-making process involving economic-like choices along certain psychographic determinants.


According to the U.S. Surgeon General, cigarette smoking is "the chief, single avoidable cause of death in our society, and the most important public health issue of our time" (in Feinleib 1979). Smoking accounts for 25 percent of all cancer deaths in the U.S. It is estimated that 85 percent of lung cancer cases in countries like France and Israel are due to cigarette smoking.

Smoking is a complex phenomenon which stems from a variety of behavioral, personality and social factors (Lichtenstein 1982). Some of the studies that have been undertaken to reveal the causes of smoking and smoking intensity examine demographic characteristics (e.g. McAlister et al. 1984), personality traits (e.g. Loken 1982), and various economic factors (e.g. Gafni and Torrance 1984). These studies have yielded sometimes contradictory findings concerning both the decision to smoke and the determinants of smoking intensity (Barer et al. 1984).

Most health experts agree that lifestyle factors are major determinants of health behavior. They also agree on the importance of education and promotion campaigns in modifying lifestyle in health-enhancing directions. Despite increasing epidemiological support for the interrelationship between lifestyle and health, and despite the seriousness of the smoking problem and the many anti-smoking campaigns, consumer researchers have generally made scant contribution to this social marketing issue (Fox and Kotler 1981). Except for a few descriptive papers, recent literature reviews (e.g., Fuchs 1986) reveal that the consumer behavior literature has had little to offer in the way of guidance to public policy and marketing decisions involving the my aid aspects of preventive campaigns.

The present study was designed to extend consumer research on the cigarette smoking problem in two ways: first, by empirically examining the influence of psychographic, economic (time preference) and demographic characteristics on smoking behavior; and second, by offering a practical typology as an aid to the understanding of smoking behavior. The advantage of such a typology is in the identification of different types of smokers, including those that can and cannot be influenced by anti-smoking campaigns.


From an economic perspective many human health behaviors have one common characteristic they involve trade-offs between current costs and future benefits. The cost may be the loss of immediate pleasure from a cigarette. The expected benefits typically take the form of a reduction in the probability of morbidity and mortality from one or more diseases (lung cancer, chronic bronchitis, emphysema, or heart disease) sometime in the future.

On the other hand, smoking is also the response of some people to stress and to peer pressure (Corty 1983; Loken 1982). Once these dangerous responses are transformed from short-term coping reactions into lifestyle or psychographic patterns they become formidable risk factors.

The concepts of "time preference" and "psychographics" are well-known and distinct concepts in the study of consumers' preferences for products. Each of these concepts has also been applied separately in the general health-care field. In this section we review and apply these concepts to smoking behavior.

Time Preference

The concept of time preference relates to the timing of an outcome. The rate of time preference is the marginal rate of substitution between consumption in any two periods. Consider a specific amount of h of the outcome which can be received now or received later. There are three possibilities (see figure 13.

The acceptance of a current cost for a future benefit constitutes an investment. If individuals differ in their willingness or ability to undertake investment it means that they have different time preferences (Stigler and Becker 1977). Positive time preference exists if a gain is preferred earlier rather than later. This is consistent with the concept in economic theory which suggests that individuals may value future gains less than present gains, implying that future gains should be discounted (Olson & Bailey 1981). Similarly, a future health gain has to be discounted to determine its equivalent present value.

Fuchs (1982) showed that variations in individual time preferences help to explain variations in health behavior among individuals. Such differences might, therefore, help to explain also variations in smoking behavior.

Assume a two-period world and let r denote the discount rate. Suppose utility in each period depends upon consumption of products (P). Utility in the first period is also a function of some activity S1 (cigarette smoking) which affects health (H) and therefore utility in period two.


U1 = U1(P1,S1)

U2 = U2(P2,H2) where H2 = H(S1)

A wealth-compensated increase in r will alter the allocation of wealth between P1 and P2. But if the marginal utility of S2 depends on the quantity of P1 (and the marginal utility of H2 depends on the quantity of P2), the changes in r will also affect S1 (and H2). If P1 and S1 (and P2 and H2) are substitutes, an increase in r will lead to an increase in S1 and a decrease in H2. On the other hand, a decrease in r will lead to a decrease in S1 and an increase in H2. Thus, the more the individual is willing to forego present gratification (that is, the lower is his subjective discount rate), the more health (life expectancy) he will accumulate and the higher will be his late absolute health.

It is postulated here that an individual chooses an amount of smoking/non-smoking behavior which will maximize the present value of his lifetime health and this will depend on his subjective rate of time preference.

Measurement of Time Preference

In recent years there have been several attempts to measure time preference through household survey techniques. The objectives of the investigators have varied greatly, but the general approach has been similar: the respondent is confronted with a hypothetical situation involving different sums of money at different points in time and is asked to express a preference which will implicitly reveal a rate of time discount. Fuchs (1982), for example, used a series of questions asking the respondent to choose between a sum of money now and a larger sum at a specific point in the future, e.g., "Would you choose 51,500 now or 54,000 in five years?" The amount and the time period varied, as did the interest rate implicit in each question. This dichotomous choice type of question was used because it was shown by Fuchs to be simpler for the respondents than the open-ended or cascade-type questions used in other studies.

In another investigation, Maital and Maital (1978) reviewed some of the economic and psychological literature on time preference and reported the results of a survey of 515 Israeli adults. They studied the role of time preference in the intergenerational transmission of income inequality. They asked one cascade question involving a choice between a sum of money now and higher sums one year from now. A similar question in which gift certificates for a week's shopping at a supermarket were substituted for money was asked in an attempt to measure the real as opposed to the nominal implicit rate of interest. The results of their investigation led them to conclude that both the rate of time discount and the immediate value of behavior are inverse measures of a person s tendency to delay gratification.

Consumer Psychographics

Psychographic data analysis has become a full-fledged activity in consumer behavior research (see Wells and Mehrotra 1977 for a review). The basic premise of research on health psychographics is that the more we know about people's lifestyle, the more effectively we can communicate with them and educate them for a healthier way of life.

Psychographics as a construct permits one to think about the consumer as a total person and how consumption patterns, health matters and various activities fit into his or her daily living pattern (Hornik and Schlinger 1982). In its broadest context, the concept of consumer psychographics has been defined by Wells (1975) as a quantitative research tool intended to place consumers on psychological - as distinct from demographic - dimensions (p.197). This all-encompassing definition accurately reflects the current practice of psychographic research, which has included diverse categories of variables, such as Activities, Interests, and Opinions (often denoted as an AIO battery), personality traits, lifestyle measures, and attitude measures.

In order to assess the contribution of consumer psychographics to the understanding of the consumption of cigarettes, two major guidelines were considered:

1. The psychographic elements should be based on current relevant theories and findings of consumer behavior. Of particular interest is the theoretical descriptive work on healthy/unhealthy lifestyles (Barer et al., 1984; Regional Office for Health Education, 1986).

2. Regardless of the breadth of the psychographic definition, a comprehensive explanation of smoking/non-smoking behavior requires examination of the relationship between psychographic and other possible determinants of behavior, such as temporal orientation and demographic characteristics. Operationally, one should, therefore, examine the incremental explanatory power of psychographics rather than viewing psychographics as a sufficient determinant of smoking behavior.


As suggested earlier, the primary objective of this study was to offer a conceptual framework relating time preference, psychographic and demographic profiles to smoking behavior. Based on the theoretical discussion it is suggested that smoking behavior is a function of the value individuals ascribe to the present and the future, their activities, interests and opinions, and their demographic characteristics. This model is rooted in neoclassical economic theory but also derives from more recent marketing concepts of consumer psychographics.

Data Base

Data were drawn from a health survey conducted during May-June 1986 in Israel. Personal interviews were conducted with 612 individuals residing in all parts of the country. Respondents were selected randomly from voting registries. The sample was restricted to adults aged 20 or more, and interviewers were instructed to obtain an approximately equal distribution of male and female respondents.


The research instrument used was a pretested 78-item questionnaire divided into three major sections. The first section of the questionnaire included a battery of psychographic variables; the second - time preference measures; and the third section consisted of demographic items. The questionnaire required about forty minutes to complete. As an incentive respondents were offered either a small gift or the opportunity to enter a sweepstake.

Dependent Variables

The subjects were classified by smoking status and consumption intensity (Corty 1983; Loken 1982); cigar or pipe smokers were excluded. Included in the category of "smokers" were consumers who indicated that they had smoked cigarettes during the previous year. "Non-smokers" were those who reported no history of smoking. "Ex-smokers" were those who had smoked cigarettes previously but had not smoked during the year preceding the survey. Information about smoking intensity was also obtained from current smokers, including the number of cigarettes smoked per day and the number of years the respondent had smoked regularly.

Independent Variables

Three sets of explanatory variables to smoking behavior were used:

1. Time Preference: Fuchs' (1982) time preference measurement procedure, as recently further modified by Thomas and Ward (1984), was used. Subjects were offered eight hypothetical choices between a sum of money available now (X0) and another sum available at time t in the future (Xt). [n each choice, the subject was given two of the quantities, X0, Xt and t, and was asked to set the third quantity such that he or she would be indifferent to the two options. For example, "4000 I.S. now, or 10,000 I.S. in 5 years; ". The responses to the eight questions were averaged for each subject and an index of the rate of time preference was obtained. In addition to the implicit interest-rate series of questions, a cascade-type question with an explicit interest rate (beginning at 8% and rising to 50%) was used: "Suppose you won a money prize of 20,000 I.S. from "Lotto" and were offered the choice of taking the money or leaving it with them for a year. How much interest would they have to pay you in order for you to agree to leave the money with "Lotto" (Cascade - stop reading when choice mentioned)."

2. Psychographic Profiles: In the present study, psychographics is defined by a broad battery of items relating, for example, to cognitive style - way consumer thinks; coping style - way person meets stress; relating style way person handles interpersonal relationships; temporal style - consumer's time orientation; health style - way person avoids poor health. Respondents were asked to indicate agreement with 63 psychographic statements each relating to one of 14 pretested lifestyle scales. Agreement with individual statements was expressed at one of six levels, ranging from "definitely disagree" to ''definitely agree . The pretest of the lifestyle scales involved interviewing 120 households in a suburb of Tel Aviv. Forty five scales (factors) with 255 statements were explored. These factors account for 64.8 percent of the total variance. Out of the 45 factors 14 (of 63 statements) were retained, according to criteria of meaningfulness, relevance to health or smoking and (with one exception) representation by three or more items with loading of 0.40 or greater. These retained orthogonal factors, which account for 44.7 percent of the variance, are shown in Table 1.

Table 1 contains the names of the psychographic factors, the number of statements per factor and the associated Spearman-Brown corrected split-half coefficients; a sample statement (and its factor loading) is also provided for each scale. As expressed by the split-half coefficients and stable factor structure, the scales demonstrate moderately acceptable reliability and construct validity.

3. Demographic.Variables: The demographic variables were selected on the basis of previous work by Feinleib (1979) and McAlister et al. (1984) which suggested they might be relevant to smoking behavior. The following six demographic and socioeconomic variables were used: sex of respondent, employment status, age, occupation, education, and monthly family income.




Multiple regression analyses were conducted to assess the relationship of the three sets of explanatory variables to smoking behavior and smoking intensity, and cross-validated using a holdout sample. The cross-validated results suggested highly stable regression equations, and, therefore, for presentation purposes, the results are for the total sample. Three sets of regression analyses were run:

1. Demographic characteristics alone.

2. Demographic plus psychographic characteristics .

3. The same run as in the second stage, plus the time-preference measures.



The results of these multiple regression analyses (Table 2) suggest that the inclusion of the individual psychographic characteristics improves considerably the predictive efficiency of smoking behavior. The R2 increases from 0.19 for demographic characteristics alone to 0.33 for demographic characteristics plus psychographic data. This finding is consistent with previous research relating psychographics to various consumer behavior issues (e.g., Ahmed and Jackson 1979).

Statistically, lifestyle should account for more variance than demographics in smoking behavior because substantially more lifestyle variables are included in the regressions. However, there is no way to estimate how much more variance lifestyle should explain, because as variables are added to a regression equation the explained variance is apt to increase at a decreasing rate. Thus one cannot assume that the ratio of R2s for psychographics versus demographics should be in direct proportion to the number of measures of each.

While demographics and lifestyle contribute to the understanding of consumer behavior, the nature of the two constructs is very different. First, demographic and lifestyle data may correlate, but also supplement, one another (Lee and Ferber 1977). Second, there are many more potential, independent psychographic dimensions than demographic ones. This is an inherent difference that reflects the richness of the psychographic construct (Kassarjian 1971) and is its raison d'etre.

The increase in explained variance obtained by adding the time preference measures to the demographic and psychographic data is fairly moderate and not as large as in other studies investigating the explanatory power of time preference in health behavior (e.g., Fuchs 1982). This indicates that the explanatory powers of the psychographic and time preference variables are not independent. Nevertheless, it is the incremental gain in explanatory power that is of interest. The time preference variables improves R2 to 0.41.

Examination of Specific Variables

Demographic Variables: The strongest variable to enter the demographic - only equation, as indicated by the standardized regression coefficients, is employment, followed by the sex of the respondent and occupation status. The highest percentage of smokers was found among the unemployed and those in administrative or managerial occupations. It was found that males smoke more than females and that income is positively correlated with cigarette smoking, but not at a significant rate. This supports previous studies which found that smoking is more a function of stress than wealth (Lichtenstein 1982).

Psychographic Variables: Adding the psychographic variables to the demographic variables improves the explained variance by more than 73 percent. Table 2 shows significant partial correlations between several psychographic factors and the smoking measures. The two factors that have the most significant relationship with smoking are "present oriented" and "active sports" (actually, nonactive sports). The higher the value of present time (and the lower the value of future time), the more positive and intensive the smoking behavior. This, in addition to the significant link between the "indulger" factor to smoking, reflects the immediate gratifiers' way of life, which places emphasis on instant satisfaction and hedonistic activities.

It was found that people who do not engage in active sports and are heavy drinkers are more inclined to be smokers. It seems that individuals who generally do not maintain a healthy way of life also tend to smoke. Also, individuals who worry about their health are more able to control their smoking. Finally, as expected, a significant relationship was found between the consumption of cigarettes and feelings of depression.

Time Preference: Although the time preference measures do not have the expected predictive power for smoking behavior, the results do confirm the expectation that cigarette smoking does increase with higher implicit interest rates. These results generally support the proposition that heavy smokers have a higher preference for immediate satisfaction whereas ex-smokers and non-smokers have a higher preference for delayed satisfaction.


In view of the widely recognized health risks accompanying cigarette smoking, prime importance is attached to understanding the factors influencing smoking behavior. Our data suggest that smoking behavior is related to time preference and to time orientation. Other things being equal, subjects with a high discount rate are likely to prefer the immediate satisfaction of smoking more than subjects with a low discount rate. Also, individuals with a present-time orientation are more likely to smoke than individuals more preoccupied with the future. In order words, high discount rates of time and high present orientation are associated with a tendency to smoke.

There seems to be a strong link between individual lifestyle and smoking. Cigarettes are consumed to meet different personal needs and gratifications. This is particularly evident in the case of the non-active, unemployed and depressed. On the other hand, individuals who are more health conscious and physically active refrain from smoking.

In general, the psychographic dimensions provided a more meaningful and valid measurement of smoking behavior than the other two sets of variables. At the same time, the temporal concept helps to understand that individuals also elaborate on economic-like choices in smoking, according to their time preference. The relatively small contribution of time preference leads, however, to the hypothesis that smoking not only contains an element of economic rational choices but is also a part of the individual's way of life. Time preference should, therefore, be viewed as complementary to psychographics in predicting smoking behavior.

Even if not immediately evident, it appears that a logical step from the current literature is to more deliberately incorporate time-preference measures into lifestyle research. The term "lifestyle" in consumer behavior is normally used to characterize a set of consumer decisions relating to the acquisition of goods and services. In reality, the decisions in question are also based on time preferences among the various activities, something that might more accurately be described as "timestyle", a term already suggested by Feldman and Hornik (1981). Since the concept of time preference has not been particularly evident in consumer behavior research, this extension seems especially timely.


"Most of the bad things that happen to people are at present beyond the reach of medicine " (Wildavsky 1977). Both marketing and public policy makers could benefit from an understanding of the significant temporal and lifestyle factors which encourage smoking. Thus efforts directed toward disease prevention and health promotion can extend beyond the traditional medical-care system to include institutional measures to promote and modify individual lifestyles.

If consumers can, ex ante, be segmented into smoking status groups based on their lifestyle and causal epidemiological evidence, then evidence linking various lifestyles to health-care costs could be used to construct differential health insurance premiums. Alternatively, prevention programs could incorporate strategies to increase belief in the lifestyle-smoking-illness link and the efficacy of a change in lifestyle involving reduction of cigarette consumption.


Ahmed, Sactrudin A., and Douglas N. Jackson (1979), "Psychographics for Social Policy Decisions: Welfare Assistance", Journal of Consumer Research, S(4), pp. 229-39.

Anderson, T.W. and L.L. Golden (1984), "Lifestyle and Psychograpics: A Critical Review and Recommendation", in T. Kinnear (ed.), Advances in Consumer Research, 11, 405-11.

Barer, Morris L., Robert G. Evans, Greg L. Stopddart, Roberta Labelle. and Jane Fulton (1984), "Lifestyles, Linkages and Liabilities", QESP Research Report, No. 112, McMaster University: Program for Quantitative Studies in Economics and Population.

Corty, E. (1983), "The Test-Retest Reliability of Measures Used in Cigarette Smoking Research", Addictive Behavior, 8(3), 315-18.

Feinleib, M. (1979), "A Comparison of Blood Pressure, Total Cholesterol and Cigarette Smoking in Parents in 1950 and Their Children in 1970", American Journal of Epidemiology, 110(3), 291303.

Feldman, Laurence and Jacob Hornik (1981), "The Use of Time: An Integrated Conceptual Model", Journal of Consumer Research, 7, 407419.

Fox, Karen F. and Philip Kotler (1981), "Reducing Cigarette Smoking: An Opportunity for Social Marketing?", Journal of Health Care Marketing, Winter, 8 - 17.

Fuchs, Victor R. (1982), "Time Preference and Health: An Exploratory Study", in Victor R. Fuchs (ed.), Economic Aspects of Health, The University of Chicago Press, 93 - 120.

Gafni, Amiram and George W. Torrance (1984), "Risk Attitude and Time Preference in Health", Management Science, 30(4), 440-51.

Hornik, Jacob and Mary Jane Schlinger (1982), "Dimensions of Health Maintenance Activities and Opinions", in Jerry C. Olson (cd), Advances in Consumer Research, 7, 627-32.

Kassarjian, Harold H. (1971), "Personality and Consumer Behavior", Journal of Marketing Research, 8, 409- 18.

Lee, Lucy C. and Robert Ferber (1977), 'The Use of Time as a Determinant of Family Market Behavior", Journal of Business Research, 5, 75-91.

Lichtenstein, E. (1982), "The Smoking Problem: A Behavioral Perspective", Journal of Consulting and Clinical Psychology, 50, 804-19.

Loken, Barabara (1982), "Heavy Smokers', Light Smokers, and Nonsmokers' Beliefs About Cigarette Smoking", Journal of Applied Psychology, 67(5), 616-22.

Maital, Shlomo and Sharona Maital (1978), "Time Preference, Delay of Gratification, and the Intergenerational Transmission of Economic Inequality: A Behavioral Theory of Income Distribution", in Orley Ashenfelter and Wallace Quates (eds.), Essays in Labor Market Analysis, New York, John Wiley, 179-99.

McAlister, A.C., J.A. Krosnick and M.A. Milburn (1984), "Cases of Adolescent Cigarette Smoking: Tests of a Structural Equation Model", Social Psychology Quarterly, 47, 24-33.

McKeown, T. (1976), The Role of Medicine, London; Nuffield Provincial Hospitals Trust.

Olson, Mancur and Martin J. Bailey (1981), "Positive Time Preference", Journal of Political Economy, 89(1), 1-25.

Regional Office for Health Education, Europe (1983), "Lifestyles and Health", Social Science and Medicine, 22(2), 117-24.

Stigler, George J. and Gary S. Becker (1977), "De Gustibus Non ESL Disputondum", American Economic Review, 67, 76-90.

Thomas, Ewart A. and Wanda E. Ward (1984), "Time Orientation, Time Preference and Spending Behavior", Research Report No. 1296A, Stanford University, Department of Psychology.

Wells, William (1975), "Psychographics: A Critical Review", Journal of Marketing Research, 12, 196213.

Wells, William and S. Mehrotra (1977), "Psychographics and Buyer Behavior" in Arch Woodside (ed.), Consumer and Industrial Buying Behavior, New York: North-Holland, 49-66.

Wildavsky, A. (1977), "Doing Better and Feeling Worse: The Political Pathology of Health Policy", Daedalus, 106. 105-23.