Determining Leisure Preference, Personality Or Demographics?

ABSTRACT - The aim of this study was to determine whether demographics or personality (measured by the Myers-Briggs Type Indicator) provide a better explanation of leisure preference. Regression analysis indicated that some attributes of leisure are better explained by personality (Planning, Follow through, Variety, People and Pace of activities), others (Household tasks, Team sports and Moderniy) by demographics, and still others (Risk) equally by the two. Neither personality nor demographics were appropriate for determining the level of involvement in activities. In all cases except the Planning scale, using a combination of demographics and personality led to a statistically significant improvement in explanatory power.

Citation:

Robyn McGuiggan (1999) ,"Determining Leisure Preference, Personality Or Demographics?", in E - European Advances in Consumer Research Volume 4, eds. Bernard Dubois, Tina M. Lowrey, and L. J. Shrum, Marc Vanhuele, Provo, UT : Association for Consumer Research, Pages: 224-231.

European Advances in Consumer Research Volume 4, 1999      Pages 224-231

DETERMINING LEISURE PREFERENCE, PERSONALITY OR DEMOGRAPHICS?

Robyn McGuiggan, University of Technology, Sydney, Australia

[This paper is based on work completed for the doctoral dissertation of Robyn L. McGuiggan entitled "The Relationship between Personality, as measured by the Myers-Briggs Type Indicator, and Leisure Preferences", Graduate School of Management, Macquarie University, Sydney, Australia, 1996.]

ABSTRACT -

The aim of this study was to determine whether demographics or personality (measured by the Myers-Briggs Type Indicator) provide a better explanation of leisure preference. Regression analysis indicated that some attributes of leisure are better explained by personality (Planning, Follow through, Variety, People and Pace of activities), others (Household tasks, Team sports and Moderniy) by demographics, and still others (Risk) equally by the two. Neither personality nor demographics were appropriate for determining the level of involvement in activities. In all cases except the Planning scale, using a combination of demographics and personality led to a statistically significant improvement in explanatory power.

INTRODUCTION

Although numerous authors have proposed various models of consumer choice, by far the most prominent models in the literature today rely on understanding the decision-making process through an information processing or problem-solving approach using logical flow models of bounded rationality to explain consumer behavior (Holbrook and Hirschman 1982). Much of the research underlying the development of these models has focused on the purchase of durable products and the tangible benefits they provide. More recently, a number of researchers have suggested that these models view consumption in too narrow a fashion, not taking into account the choice of products for more hedonic reasons. Consumers may purchase products in anticipation of having fun, to fulfil fantasies, or simply for the emotions or feelings the product will generate (Woods 1981; Holbrook et al 1984; Shimp 1993). To take account of these types of choices as well as impulse purchasing and variety seeking behavior, the traditional information processing model has been broadened into what Hirschman & Holbrook (1982) and Mowen (1988) refer to as the "experiential view" of consumer choice. In fact, in a recent paper by Bernd Schmitt (1999) he argues that this "experiential view" of consumer decision making may now be a better representation of consumer choice than the more traditional models for all types of products.

In their paper comparing the information processing and experiential views of consumer behavior, Hirschman & Holbrook (1982, p. 136) suggest that although personality has shown relatively poor performance in predicting consumer behavior, the "investigation of experiential consumption appears to offer considerable scope for the revival of personality" research. They acknowledge that all individual differences are important in determining choice, but suggest that personality may be a better predictor of choice than demographics for experiential products. Literally hundreds of studies have been completed trying to determine the influence of personality on a variety of aspects of consumer decision making across an enormous variety of product categories. However, as Kassarjian & Sheffet (1991), p.291) conclude, the results are equivocal with the amount of variance explained being 5B10% in the majority of studies with higher percentages being reported in specific cases (eg. Brody & Cunningham 1968). The results in the leisure literature are no better with most of the correlation coefficients falling in the range 0.1B0.3 with an occasional coefficient reaching 0.4 or 0.5 in studies using attributes rather than specific activities (McGuiggan 1996, p.50). However the disappointing results of previous research do not lead reviewers of either the consumer behaviour or the leisure literature, such as Kassarjian (1971), Engel et al (1973), Wells & Beard (1973), Eysenck et al (1982), Kircaldy (1985), Foxall & Goldsmith (1988), Furnham (1990) or Kassarjian & Sheffet (1991), to dismiss personality as a possible determinant of purchasing behaviour. In fact the opposite is true. All reviewers support the need for further research, but point out that for future studies to provide useful information on the relationship, researchers must take account of and avoid the pitfalls of previous studies. Barash 1997, p.4) goes so far as to say that marketplace behavior is far more likely to be influenced by "personality, temperament, character, values, ethos, mythos, and specific individual circumstances than by any combination of accumulated demographic data".

The aim then of this study was to determine whether this is in fct true for the choice of leisure activities in general, a decision involving a "strong experiential component" (Mowen & Minor 1998, p. 9). However the objective was not to prove that only one variable need be considered when investigating leisure choice, but rather to establish the relative importance of personality and demographics in that choice. Therefore the importance of personality with regard to leisure choice was studied in relation to the importance of the most common demographic variables cited in the leisure literature.

LEISURE CHOICE

Most previous research assumes that personality and demographics have a direct influence on leisure activity choice. However, research to date has provided little support for this assertion (Kelly 1978b, 1983; McGuiggan 1996). There seems to be two issues that need to be addressed here. Firstly, is leisure activity (for example tennis, watching television) the correct unit of measure, and secondly can variables such as personality or demographics be expected to influence choice directly or do they exert their effect indirectly?

Leisure activity

Is leisure activity a simple and unambiguous natural unit of analysis? A number of problems can be identified with this unit of analysis. Unlike many other product categories, it is not possible to produce an exhaustive list of activities that can be classified as leisure, since all activities have the potential to be leisure and all activities have the potential to be non-leisure (for example competition golf for the professional golfer). Secondly, Kelly (1983, p. 159) draws attention to the fact that activity labels do not take account of the diverse types of activity and interaction which may occur in a single activity setting. For example, if a person indicates they have been #swimming’ what does this tell us about their actual activitiesBhave they swum laps, played with the children, socialized with friends, or read a book by the pool? Moreover as Neulinger (1974, p. 35) points out, if a list is produced, a decision as to what constitutes a unit of activity has to be made. Number of hours may be an appropriate measure for involvement in reading or watching television, but may not be a good measure for bungy jumping or skydiving. Even if the unit of analysis is established, self reported participation in leisure activities might be biased, respondents tending to over estimate leisure activity participation (Chase & Cheek 1979; Hultsman et al 1989). Furthermore, using leisure activities as the unit of analysis limits the generalizability of the research. Leisure activities change over time. For example bungy jumping, grass skiing and inline skating were unheard of in the 1970s, as were electronic games.

It would appear that far from leisure activity being a #simple and unambiguous natural unit of analysis’; it seems to be fraught with insurmountable problems. A number of researchers have dealt with this problem by using groups of leisure activities as the unit of analysis, for example social activities, outdoor activities, sports and hobbies. However a review of the literature does not indicate any agreement as to the way activities should be grouped. A further argument against using activity groupings is provided by Kelly (1978b) who points out that dealing with classifications of activities will invariably obscure significant differences among particular activities. Duncan’s (1978) work showing that Q and R factor analyses of leisure activities provide very different groupings of activities suggests that it is not the activity itself that is important to the individual but the underlying attributes of the activity as perceived by the individual. This suggests that predicting #generic behavior’, or attribute preferences, should be less difficult than predicting specific forms of behavior. Furthermore, Lancaster (1966) would argue that it is the characteristics/atributes of goods that provide utility to the consumer. It is the utility of the individual attributes that determines the preference for particular products/services. People evaluate alternatives on the basis of their attributes rather than compare products as a whole (Gensch & Javalgi 1987).

It would appear then that the solution is not to profile a person’s actual participation in leisure activities, but rather the attributes or meaning of those activities. The same conclusion has been reached by a number of previous authors such as Coleman (1976), Kelly (1978a), Kabanoff (1980), Crandall & Slivken (1980), Iso-Ahola (1980) and Bergier (1981). In support of this approach, Kabanoff (1981, p. 383) argues that task attributes have consistently been used to predict job satisfaction, and therefore he sees no reason why they should not also predict leisure satisfaction. A further argument for focusing on leisure attributes and meanings is provided by Balderjahn’s (1988) causal model of ecologically conscious consumer behavior in which he suggests that greater correlation should be found between personality and leisure attributes than between personality and participation in specific activities. Foxall’s (1984, pp. 115-116) contention that to measure accurately the association between two variables both must be conceptualized and measured at the same level of generality/specificity, adds a further argument for using attributes rather than activities. Personality measures apply to global views of behavior and not to the specific. Attributes are not activity specific; they can be used to describe any leisure activity.

Leisure choice

The second issue that needs to be addressed is that of choice. Choices are made, but within the constraints that order our lives. Many researchers in the consumer behavior and leisure area have identified constraints on our actual choice. These may be external to the individual such as culturally determined roles, laws and regulations, availability/accessibility of appropriate facilities, or internal such as physical ability, health, and competing obligations such as family or work. Whether constraints are internal or external, perceived or real, surveys such as the General Recreation Surveys administered by the Alberta government in 1988 and 1992, in which approximately fifty percent of respondents reported the desire to start a new recreational or leisure activity but were unable to because of various constraints (Jackson & Witt, 1994), indicate that actual choice is likely to be difficult to predict. Although we may prefer to engage in a particular activity, intervening variables may lead to a totally different activity being chosen or a compromise being made. Therefore rather than personality and demographics influencing leisure choice directly, it is proposed that they exert their influence indirectly through establishing leisure preferences. Some support for expecting personality to influence leisure choice indirectly through preference is provided by Webster & Wakshlag’s (1983) model of television program choice. Although they present no empirical data to support their theory, they propose that "specific program preference is a cause of program choice" (p. 432) and that psychological variables, e.g. needs, influence actual choice through establishing preferences.

MEASURING PERSONALITY

One of the major criticisms leveled at past research is the choice of personality instrument. Literally hundreds of personality questionnaires are available commercially. However many of the previous studies on personality and choice in the consumer behavior literature and the leisure literature have been criticized by authors such as Kassarjian (1971), Engel et al (1973), Wells & Beard (1973), Wilde (1977), Iso-Ahola (1980) Eysenck et al (1982), Kircaldy (1985), Foxall & Goldsmith (1988), Furnham (1990; 192) and Kassarjian & Sheffet (1991) for the inappropriate selection and use of personality instrument. The criticisms of these authors point to the necessity of choosing a theoretically based instrument that has been adequately tested in terms of reliability and validity. There must also be a theoretically sound justification for expecting a relationship between the personality trait/s to be measured and the leisure activity.

The Myers-Briggs Type Indicator (MBTI) has a strong theoretical framework being based on Myers’ theory of personality which she developed from her interpretation of Carl Jung’s theory of psychological types and her own observations (Myers & McCaully 1992). The MBTI describes a person’s personality on four dichotomous dimensions indicating a person’s preference for source of psychological energy (extraversion vs. introversion), perception (sensing vs. intuition), making judgements (thinking vs. feeling), and orientation to the outer world (judging vs. perceiving). The MBTI questionnaire is a forced-choice, self-report inventory, virtually self-administering and designed for use with normal subjects. The questions consist of behavioral preferences and a number of preferred self-descriptive adjectives. Each individual question is designed to elicit a preference for one of the four dimensions. Countless papers have been written reviewing the reliability and validity data on the MBTI. Generally these support the view that the four MBTI scales have construct validity and measure important dimensions of personality which approximate those in Jung’s typology (Steele & Kelly 1976; Coan 1978; Levy & Padilla 1982; Geer et al 1984; Tzeng et al 1984; Sipps et al 1985; Thompson & Borrello 1986; Wiggins 1989; Murray 1990). In addition the MBTI questionnaire is readily available, simple to administer and score and is the most widely used personality questionnaire in America for non-psychiatric populations (Murray 1990).

A number of arguments can be advanced for expecting a relationship to exist between MBTI type and leisure choice. Firstly, Jung’s and Myers’ theories imply that a relationship should exist since both are type theories, which advocate that people of similar personality can be expected to react in a similar way to many situations in life. On this basis a number of authors, including Kroeger & Thuesen (1988) and Provost (1990) have hypothesized on the existence of the relationship. In fact many MBTI advocates suggest that this relationship may be stronger than that between personality and work. Secondly, extensive data are available on the relationship between MBTI type and occupational preference or choice. If choice of both work and leisure are influenced by a common third variable such as personality, as suggested by Iso-Ahola (1980) and others, the data already collected on the MBTI showing a strong relationship between MBTI type and career choice would suggest that a relationship between type and leisure choice should also exist. Thirdly, the sensing-intuitive and thinking-feeling functions measured by the MBTI refer to the way people gather information and arrive at decisions. This parallels consumer behavior theory in terms of information search and decision making. In fact the MBTI has been used in organizations to estimate decision styles of executives (Moore 1987). So why wouldn’t the same be true for decisions on product or leisure choice? Although the MBTI has been used only to a very limited extent in consumer research, Shank & Langmeyer (1994, p. 162) concluded from their study that "in fact, the Myers-Briggs Type Indicator would seem to be the ideal personality inventory for marketers".

TABLE 1

BROAD HEADINGS SUGGESTED BY THE LITERATURE-AND UTILIZED IN DEVELOPING THE LEISURE QUESTIONS

The MBTI has been used in a limited number of studies in the leisure area focusing on determining the personality type of people engaged in very specific activities. Although the results do not show overwhelming support for a relationship existing between MBTI type and leisure activity, they do indicate that a relationship may exist. Gontang et al (1977) found that marathon runners were more likely to have a preference for introversion and judging, while Clitsome & Kostrubala (1977) found them to have a preference for introversion but lso to have a preference for sensing. In contrast, Franzoi (1985) found cross-country runners overwhelmingly to have a preference for perception with some indication that a preference for extraversion, intuition or feeling may encourage participation in this sport. In addition, Schurr et al (1988) found that sensing types were more likely to attend male basket ball games and Morehouse et al (1990) showed that extroverts exhibit a greater desire for activity in general. Sensing types appear to watch more television than intuitive types and program preference can also be related to MBTI type (Nolan & Patterson 1990). Although these findings are somewhat inconsistent and the relationships found are not terribly strong, they do provide additional reason to suspect that a relationship between MBTI type and leisure preference may exist.

METHOD

To avoid the criticism levelled at past research that there was no theoretical basis to support the relationship between the personality instrument chosen and the consumer behavior measured, the MBTI literature was used to develop the leisure attributes and meanings to be studied. The MBTI literature was scanned to determine any predictions that had been made regarding the relationship between personality type and leisure preferences. Other MBTI type characteristics that might be applicable to leisure attribute preference were also noted. For example it could be expected that people who like to plan their work would also prefer to plan their leisure activities. One hundred and twenty forced choice questions were developed under 25 broad headings suggested by the literature (see Table 1). These groupings were later used in the analysis to aid in the formation of leisure attribute scales. Since the MBTI questions are presented as forced choice answers, the leisure questions were developed to replicate this format. The intention was to make questionnaire completion easier for the respondent and to minimize completion time. (For a full description of the development of the actual questions see McGuiggan (1996)).

A sample of 103 undergraduate business students was used to pre-test the questions. After skewed (non-discriminating) questions were deleted and highly correlating questions either deleted or reworded, 101 leisure questions remained. Thus the final questionnaire contained these 101 questions, the 94 scoring questions from the MBTI Form G and a number of demographic questions (gender, age, marital status, whether children live with them, highest level of education, usual occupation, country of origin, and household income).

A purposive sample was sought with the aim of achieving a large sample made up of a relatively even distribution of the 16 MBTI types. Undergraduate marketing research students at UTS were required to collect data for this study as part of their course requirement. Each student needed to have six questionnaires completed by appropriate respondents. Instructions on respondent suitability were provided in order to attempt to maximize the variation in respondent personality type. 782 useable questionnaires were obtained. However after establishing the MBTI type of the respondents, it was found that 12 of the MBTI types contained fewer than 30 respondents. Therefore further respondents were sought to boost the numbers in these 12 groups. A list of possible participants of #known’ MBTI type in each of the 12 groups sought was provided by the Director of The Institute for Type Development, Sydney. Each of these people was mailed a package, which included a copy of the questionnaire, a reply paid envelope and a covering letter on Institute for Type Development letterhead signed by the Director, requesting their help with the project. A further 126 cases were added to the data bank in this way, giving a total sample size of 908 cases.

TABLE 2

RELIABILITY OF THE LEISURE SCALES-CRONBACH'S ALPHA

Men comprised 49% of the sample, and the mean age as 35-39 years (range 18-65+ years). Seventy two percent were born in Australia and 50% had obtained at least a trade qualification. Of the 70% who were working, 58% worked full-time and 12% part-time contributing to an average household income of $65,000 per year. In comparison with the general Australian population the sample has a higher percentage of respondents who work and is skewed towards the upper quartile of household income brackets. This is not unexpected considering the method of sample selection. However the sample showed a good spread of occupations contributing to a broad range of household incomes ($20,00B$150,000+ per year).

RESULTS

SPSS, a statistical package for the social sciences, was used to analyze the data. Data reduction of the 101 leisure questions was achieved by utilizing the original 25 leisure attribute categories from the literature. Three of these groups were dropped from the analysis since correlation analysis indicated that the questions within these groups were not closely related. The individual question scores in each of the remaining leisure categories were added together to create 22 new variables. Factor analysis was utilized to determine whether any of these 22 scales should be combined, and reliability analysis used to improve the internal consistency of the scales. The final outcome was the formation of 11 simple additive scales utilizing 79 of the 101 leisure questions. (For a full description of the data reduction process and the actual questions in each of the scales see McGuiggan (1996). These leisure scales were correlated with the 4 MBTI continuous scalesBExtraversion-Introversion, Sensing-Intuitive, Thinking-Feeling, Judging-Perceiving (created as described in the MBTI Manual (Myers & McCaully, 1992 p. 9))Band the demographics. The Length of vacation scale showed no significant relationship with any of the MBTI scales nor demographics, and was therefore excluded from further analysis. The reliability of the remaining ten scales is presented in Table 2. With the exception of the People scale, the internal reliability of these scales are all above Nunnaly’s (1967) cut off value of 0.5.

To determine the relative influence of MBTI personality type and demographics on leisure attribute preference, two sets of nested multiple regression analyses were undertaken with the dependent variables being the leisure attribute scales. The full model for both sets of analyses included the 4 MBTI continuous scales plus the demographic variablesBage, household income, education level, gender, children under 18 years living in the household, usual occupation and cultural heritage (dummy variables having been created for this variable). In the first set of regressions the 4 MBTI continuous scale scores were used as the independent variables in the nested equation while in the second the demographic variables were utilized. In each case the change in R² and F-value was examined to determine the relative contribution to exploratory power provided by the 2 sets of variables. The results are presented in Table 3.

From Table 3 it is evident that the power of the MBTI to explain the variability in the leisure scales differs greatly from a low of 6.5% for the Involvement scale to a high of 46.6% for the Planning scale. For five of the leisure scales, Planning, Variety, Risk, People and Team sports, more than 20% of the variance could be explained by the MBTI scales. Taking demographics alone, the variability explained varies from a low of 4.5% for the planning scale to a high of 23.8% for the Modernity scale. For the three scales Risk, Pace of activities and Modernity more than 20% of the variance could be explained by demographics. It would appear from this that personality, as measured by the MBTI, is in fact a better predictor of leisure preference than demographics. However from examination of the full model and he change in R² and F-value, it can be seen that the explanatory power of all the leisure scales, except that of planning, could better be explained by a combination of both personality and demographic variables.

In the case of the Planning scale, demographics add nothing to the explanatory power of the equation, in fact the Judging-Perceptive personality scale alone accounts for the full 46.6% of the variability (see Table 4). For a further four of the scalesBFollow through, Variety, People and Team sport, personality provides a much better explanation than demographics alone (although adding demographics leads to a statistically significant increase in explanatory power of between 2.9 and 7.5 percent). Personality is also a better predictor of Involvement in activities than demographics, but this scale is poorly predicted by all variables measured. At the other end of the continuum, Household tasks, the Pace of activities and the Modernity scales are best described in terms of the demographic profile of the respondent (18.3%, 21.2% and 23.8% respectively), although adding personality leads to a significant increase in explanatory power (5.5%, 3.9% and 4.7% respectively). For the Risk scale, personality and demographics when used separately have approximately equal predictive power (21.4% and 22.4% respectively), but again by using both together the explanation of variability in the scale is significantly enhanced (32.2%). Therefore, in all but the Planning scale, it appears that both demographics and personality are needed to explain leisure attribute preference. The significant standardized b values for the full equations are presented in Table 4.

TABLE 3

CHANGE IN R² AND F VALUE FOR NESTED REGRESSION ANALYSES

TABLE 4

FULL REGRESSION MODEL-STANDARDIZED SIGNIFICANT B VALUES

From Table 4 it would seem that people with a preference for judging are overwhelmingly more likely to plan their leisure time. Younger males are likely to have a favorable attitude towards risk taking in their leisure activities, especially if they have a preference for extraversion, intuition, thinking or perceiving. Furthermore, younger males in more professional occupations, who don’t have children and have a preference for extraversion, intuition, or perception, are more likely to want lots of variety in their leisure activities. Young males with children with a preference for extraversion or perception are more likely to want to watch or participate in team sports in their leisure time. Modern culture seems more likely to be preferred over traditional culture by young males with less formal education but who work in more professional occupations, if they have a preference for extraversion, perceiving or sensing. Young singles in less prestigious occupations with a preference for either extraversion or perception are more likely to want to spend their leisure time with other people. Younger single males in more professional occupations seem to have a preference for faster paced, competitive leisure activities, especially if they also have a preference for extraversion, thinking or perception. People who are likely to do tasks around the house in their leisure time are generally younger and single with an above average household income and have a preference for sensing, thinking or judging. Older people (perhaps of Asian or European decent) employed in more professional types of occupations with a preference for extraversion, sensing or judging are more likely to follow through with whatever leisure plans they make. Although the association is weak, males with a preference for thinking or intuition score higher on the involvement scale.

DISCUSSION

In all cases except the Planning scale, explanation was improved by use of both demographic and personality variables. The combination of personality and demographic variables explained 20% or more of the variance in eight of the ten leisure attribute scales. However, even though using the two types of variables significantly improved explanation in all bar one case, the Planning scale, the improvement varied considerably. By adding demographics to the personality only model eplanation was improved from 2.6% to 15.1%, while adding personality to the demographics only model lead to between 3.7% and 16.3% improvement in explanation. Therefore it would appear that some attributes of leisure are better explained by personality (Planning, Follow through, Variety, People and Team sport), others (Household tasks, Pace of activities and Modernity) by demographics, and still others (Risk) equally by the two. Furthermore it is apparent that neither personality nor the demographic variables used in this study are appropriate for determining the level of involvement in activities nor the preferred length of vacation. However it is also apparent that both demographics and personality both contribute to the explanation of leisure attribute preference.

A possible reason for the variability in the proficiency of both the personality and the demographic variables to explain leisure attribute preference, could be the internal reliability of the various leisure scales. As demonstrated in Lastovicka & Joachimsthaler’s (1988) paper, the coefficient of determination is as dependent on the reliability of the dependent variable as it is on the reliability of the independent variables. Therefore further research needs to be undertaken to improve the internal reliability of the leisure attribute scales, which may in turn enhance the explanatory power of the MBTI scales and the demographics. Of course other explanations for the variation in explanatory power are possible. For example, the leisure attribute may not in fact be related to personality or to demographics. On the other hand, other demographic variables not measured in this study may provide more exploratory power. Alternatively, the respondent may not see the attribute as leisure, and therefore lower correlations might be expected. Although the leisure attributes for this study were based on MBTI theory, these may not in fact be the most salient descriptors of leisure activities.

CONCLUSION

The results of this study provides support for Holbrook & Hirschman’s (1982) contention that personality should be considered an important influence in the choice of experiential products such as leisure. However, the study also indicates that not all attributes of leisure are equally likely to be influenced by personality. In some cases demographics may provide a better explanation. The overwhelming conclusion from this study is that both demographic and personality variables play an important role in the formation of preferences for experiential products such as leisure, and thus must both be measured to gain a thorough understanding of consumer preference.

REFERENCES

Balderjahn, I. (1988). Personality variables and environmental attitudes as predictors of ecologically responsible consumption patterns. Journal of Business Research, 17, 51-56.

Barash, R. (1997). The dumbing-down of America. Marketing News, 31 (22), 4.

Bergier, M. J. (1981). A conceptual model of leisure-time choice behavior. Journal of Leisure Research, 13 (2), 139-157.

Brody, R. P. & Cunningham, S.M. (1968). Personality variables and the consumer decision process. Journal of Marketing Research. 5(Feb), 50-57.

Chase, D. R., & Cheek ,N.H., Jr. (1979). Activity preferences and participation: conclusions from a factor analytic study. Journal of Leisure Research, 11 (1), 92-101.

Clitsome, T.& Kostrubala, T. (1977). A psychological study of 100 marathoners using the Myers-Briggs TypeIndicator and demographic data. Annals of the New York Academy of Science, 301, 1010-1019.

Coan, R.W. (1978). Review of the Myers-Briggs Type Indicator. In The eighth mental measurements yearbook, 970-975. Highland Park, New Jersey: Gryphon Press.

Coleman, D. V. (1976). Biographical, personality, and situational determinants of leisure time expenditure: with specific attention to competitive activities (athletics) and to more cooperative activities (music). Dissertation Abstracts International, 37 (1-B July), 520-B.

Crandall, R., & Slivken, K. (1980). Leisure attitudes and their measurement. In S. E. Iso-Ahola (Ed.), Social psychology perspective on leisure and recreation, (1st ed.) (pp. 261-284). Springfield, Illinios: Charles C Thomas.

Duncan, D. J. (1978). Leisure types: factor analyses of leisure profiles. Journal of Leisure Research, 10 (2), 113-125.

Engel, J. F., Kollat, D. T., & Blackwell, R. D. (1973). Consumer Behavior. Hinsdale, Illinios: Dryden Press.

Eysenck, H. J., Nias, D. K. B., & Cox, D. N. (1982). Sport and personality. Advances in Behavior Research & Theory, 4 (1) 1-56.

Foxall, G. (1984). The meaning of marketing and leisure: issues for research and development. European Journal of Marketing, 18 (2), 23-32.

Foxall, G. & Goldsmith, R.E. (1988). Personality and consumer research: another look. Journal of the Market Research Society. 30(2), 111-125.

Franzoi, S. (1985). Personality characteristics of the cross country hitchhiker. Adolescence. 20(79), 655-668.

Furnham, A. (1990). Personality and demographic determinants of leisure and sports preference and performance. Journal of Sport Psychology. 21(3), 218-236.

Furnham, A. (1992). Personality at work: the role of individual differences in the workplace. London: Routledge.

Geer, C.,S. Ridley, E. & Roberts, A. (1984). Jungian personality as a predictor of attendance at the black college day march. Psychological Reports. 54, 887-890.

Gensch, D.H. & Javalgi, R.G. (1987). The influence of involvement on disaggregate attribute choice models. Journal of Consumer Research, 14 (June), 71-82.

Gontang, A., Clitsome, T., & Kostrubala, T. (1977). A psychological study of 50 sub-3-hour marathoners. Annals of New York Academy of Sciences. 301, 1020-1028.

Holbrook, M.B., Chestnut, R.W., Oliva, T.A. & Greenleaf, E.A. (1984). Play as a consumption experience: the roles of emotions, performance, and personality in the enjoyment of games. Journal of Consumer Research, 11 (Sept), 728-739.

Holbrook, M.B. & Hirschman, E.C. (1982) The experiential aspects of consumption: consumer fantasies, feelings, and fun. Journal of Consumer Research, 9 (Sept), 132-140.

Hultsman, W. Z., Hultsman, J. T., & Black, D. R. (1989). Response peaks as a component of measurement error: assessment implications for self-report data in leisure research. Journal of Leisure Research, 21 (4), 310-315.

Iso-Ahola, S. E. (1980). The social psychology of leisure and recreation. Dubuque, Iowa: Wm. C. Brown Company.

Jackson, E. L., & Witt, P. A. (1994). Change and stability in leisure constraints: a comparison of two surveys conducted four years apart. Journal of Leisure Research, 26 (4), 322-336.

Kabanoff, B. (1980). A task attributes description of leisure: its relation to personality. Australian Psychologist, 15 (2), 268-269.

Kabanoff, B. (1981). Validation of a task attributes description of leisure. Australian Journal of Psychology, 33 (3), 383-391.

Kassarjian, H. H. (1971). Personality and consumer behavior: a review. Journal of Marketing Research. 8 (November), 409-418.

Kassarjian, H. H. & Sheffet,M. J. (1991). Personality in consumer behavior: an update. In Perspectives in consumer behavior,ed. H. H. Kassarjian & T. S. Robertson, 4th edition, 281-303. Engelwood Cliffs, New Jersey: Prentice Hall.

Kelly, J. R. (1978a). A revised paradigm of leisure choice. Leisure Sciences, 1 (4), 345-363.

Kelly, J. R. (1978b). Situational and social factors in leisure decisions. Pacific Sociological Review, 21 (3 July), 313-330.

Kelly, J. R. (1983). Leisure identities and interactions. London, UK: George Allen & Unwin.

Kirkcaldy, B.D. (1985). The value of traits in sport. In Individual Differences in Movement, ed. Bruce D. Kircaldy, 257-277. Lancaster, England: MPT Press Limited.

Kroeger, O., & Thuesen, J. M. (1988). Type talk: the 16 personality types that determine how we live, love, and work. New York: Bantam Doubleday Dell Publishing Group, Inc.

Lancaster, K.J. (1966). A new approach to consumer theory. Journal of Political Economy, 75 (April), 132-157.

Lastovicka, J.L. & Joachimsthaler, E.A. (1988). Improving the detection of personality-behavior relationships in consumer research. Journal of Consumer Research, 14 (March), 583- 587.

Levy, N. & Padilla, A.. Z (1982). A Spanish translation of the Myers-Briggs Type Indicator Form G. Psychological Reports. 51, 109-110.

McGuiggan, R. L. (1996). The relationship between personality, as measured by the Myers-Briggs Type Indicator, and leisure preferences. Ph.D. diss., Macquarie University, Australia.

Moore, T. (1987). Personality tests are back. Fortune, 115, 76-81.

Morehouse, R. E., Farley, F. & Youngquist, J. V. (1990). Type T personality and the Jungian classification system. Journal of Personality Assessment. 54(1&2), 231-235.

Mowen, J.C. (1988). Beyond consumer decision making. The Journal of Consumer Marketing, 5 (1, Winter), 15-25.

Mowen, J.C. & Minor, M. (1998). Consumer Behavior. New Jersey, USA: Prentice-Hall.

Myers, I., & McCaulley, M. H. (1992). Manual: A Guide to the Development and Use of the Myers-Briggs Type Indicator. Palo Alto, California: Consulting Psychologists Press, Inc.

Murray, J.B. (1990). Review of research on the Myers-Briggs Type Indicator. Perceptual and Motor skills. 70, 1187-1202.

Neulinger, J. (1974). The psychology of leisure: research approaches to the study of leisure. Springfield, Illinios: Charles C Thomas.

Nolan, L.L. & Patterson, S.J. (1990). The active audience: personality type as an indicator of TV program preference. Journal of Social Behavior and Personality. 5(6), 697-710.

Nunnally, J.C. (1967). Psychometric Theory. New York: McGraw-Hill Book Company.

Provost, J. A. (1990). Work, play, and Type. Palo Alto, California: Consulting Psychologists Press, Inc.

Schmitt, B. (1999). Experiential Marketing. Journal of Marketing Management. 15, 53-67.

Schurr, K. T., Wittig, A.F., Ruble, V.E., & Ellen, A.S. (1988). Demographic and personality characteristics associated with persistent, occasional, and non-attendance of university male basketball games by college students. Journal of Sports Behaviour. 11(1), 3-17.

Shank, M.D. & Langmeyer, L. (19940. Does personality influence brand image? The Journal of Psychology. 128 (2), 157-164.

Shimp, T.A. (1993). Promotions management & marketing communications, 3rd ed. Harcourt Brace & Company.

Sipps, G.J., Alexander, R.A., and Friedt, L. (1985). Item analysis of the Meyers-Briggs Type Indicator. Educational and Psychological Measurement. 45, 789-796.

Steele, R.S. and Kelly, T.J. (1976). Eysenck personality questionnaire and Jungian Myers-Briggs Type Indicator of extraversion-introversion. Journal of Consulting and Clinical Psychology. 44, 690-691.

Thompson, B. & Borrello, G.M. (1986). Second-order factor structure of the MBTI a construct validity assessment. Measurement and Evaluation in Counselling and Development. 18(4, January), 148-153.

Tzeng, O.C.S., Outcault, D., Boyer, S.L., Ware, R., & Landis, D. (1984). Item validity of the Myers-Briggs Type Indicator. Journal of Personality Assessment. 48(3), 255-256.

Webster, J. & Wakshlag, J. 91983). A theory of television program choice. Communication Research. 10(4), 430-446.

Wells, W.D. & Beard, A.D. (1973). Personality and Consumer behaviour. In Consumer behaviour: theoretical sources, ed. S. Ward & T.S. Robertson, 141-199. Englewood Cliffs, New Jersey: Prentice-Hall, Inc.

Wiggins, J.S. (1989). Review of Myers-Briggs Type Indicator. In Tenth mental measurements yearbook, ed. J. C. Conoley and J. J. Kramer, 537-538. Lincoln, New England: University of Nebraska Press.

Wilde, G. J. S. (1977). Trait description and measurement by personality questionnaires. In R. B. Cattell, & R. M. Dreger (Eds.) Handbook of modern personality theory, (pp. 69-103). Washington D.C.: Hemisphere Publishing Corporation.

Woods, W.A. (1981). Consumer Behavior, New York, North-Holland.

----------------------------------------