Emerging Lifestyles in China and Consequences for Perception of Advertising,
Buying Behaviour and Preferences for Consumption: An Exploratory Study
Paper submitted to Advertising Division, AEJMC.
Ran WEI Ph.D.
Department of Journalism & Communication
The Chinese University of Hong Kong
Shatin, N.T., Hong Kong
Tel: (852) 2609-7691
Fax: (852) 2603-5007
E-mail: [log in to unmask]
Emerging Lifestyles in China and Consequences for Perception of Advertising,
Buying Behaviour and Preferences for Consumption: An Exploratory Study
Abstract
This study segments consumers in China into five groups based on six
empirically-tested lifestyles: Traditionalists, Status-quo, Modern,
Transitioners and Generation Xers. Marked by old age, poor education and
poverty, Traditionalists lead an old-fashioned life and resist to change.
Demographically similar to Traditionalists, the Status-quo segment, however, has
not reached a stage where life goes around established routines. Ill-educated
with low income, Transitioners are much younger and open to change. The Modern
segment is the most affluent, well-educated, pursing a fashionable and
materialistic life. Generation Xers, born after the "Cultural Revolution," are
best-educated; they show disrespect for routines and tradition and worry little
about money.
1. Introduction
Western countries have experienced a proliferation of various lifestyles in the
past few decades (Johansson, 1994; Sobel, 1983). Lifestyle theorists argue that
when a society accumulates sufficient capital to generate enough leisure time
for many people, alternative standards of value and lifestyle become a feature
of society (Bell, 1976; Blumer, 1969; Lewis, 1973; Johansson & Miegel, 1992;
Zablocki & Kanter, 1976). Given the rapid-improving living standard and the
return to "class values" (O'Neil, 1997) in reformed China, will lifestyles of a
variety emerge? If so, what kind of formations will they be? And what are the
consequences of emerging lifestyles for marketing and advertising?
This study represents an attempt to establish and elaborate the possible
formations of lifestyle in Chinese consumers. Then, it will explore the linkages
between lifestyle segments and consumer Behaviour, including the perception of
advertising. Specifically, this study raises two central research questions:(1)
What are the possible formations of lifestyle in contemporary China? And (2) Are
there any significant differences among lifestyle segments in perceiving
advertising, buying behaviour and preferences for consumption of things Chinese
vis- -vis Western products?
The emergence of different lifestyles in reformed China is closely intertwined
with the country's momentous economic reform since the early 1990s. More
importantly, the creation of lifestyles through differentiation of classes and
status groups will attest to a rising consumer society in China. At the same
time, the growth of different lifestyle groups provides a rare opportunity to
witness and study the process of modernization and individualization in Chinese
consumers. Thus, research on consumer lifestyles in a Chinese context bears
theoretical importance.
In fact, studies on the making of various lifestyles in China are overdue, for
such studies can provide a valuable approach to segment this new and huge market
of 1.2 billion. Target marketing strategies can be tested based on the
differentiation of lifestyle segments. Products can be better positioned to meet
the needs of different segments and groups. Moreover, the identification and
classification of consumers according to their lifestyles will help make
advertising in China more effective. Different creative strategies can be
developed based on the understanding of various consumer segments.
2. Literature Review and Conceptual Framework
The concept of lifestyle originates from market research and consumer behaviour
(Mitchell, 1983; Pitts & Woodside, 1984; Plummer, 1974; Wells, 1974). The study
of consumer lifestyles is called "psychographics" which is concerned with ways
to identify the broad trends (e.g., trait aspects of consumers and lifestyle
characteristics) that influence how consumers live, work and play (Anderson &
Golden, 1984; Moven, 1990; Settle & Alreck, 1989; Williams, 1981) .
Defined briefly, lifestyle is about "how people live, how they spend their
money, and how they allocate their time" (Cosmas, 1982; Hawkins, Best & Coney,
1983) as well as "patterned activities" of consumers (Pingree & Hawkins, 1994).
Zablocki and Kanter (1976) further define lifestyle in terms of "shared
preferences or tastes, where the people sharing a lifestyle as collectivity that
otherwise lacks social and cultural identity." One of the most influential
studies in lifestyle research is Arnold Mitchell's book The Nine American
Lifestyles (1983). Based on Maslow's (1968) hierarchy of needs, Mitchell
distinguishes between three categories of lifestyles: need-driven,
outer-oriented and inner-oriented.
The psychographic research on consumer lifestyles assumes that individuals'
traits and characteristics, namely attitudes, beliefs, habits and behaviour, are
enduring and complex. That is, we need to identify patterned activities in which
individuals are grouped
in multi-dimensional spaces. As Pingree and Hawkins (1994) put it, attitudes,
values, behaviour and demographics of consumers need to be measured as they
"co-occur" in groups of interrelated variables.
The "Activities, Interests, and Opinions" (AIOs) inventory items (Cosmas, 1982;
Plummer, 1974; Uusitalo, 1979; Valette-Florence & Jolibert, 1990; Williams,
1981) have been most widely used in constructing an integrated theory of
lifestyle by focusing on consumers' values, views, attitudes and activities.
"Interests" items concern the consumers' preferences for job, recreation,
fashion and food among other things; "opinions" questions measure views and
feelings on things such as local, world, economic and social affairs; and
"activities" measures are concerned with consumption behaviour of individuals:
what do they buy, what do they do; how they spend their time. Demographic
information, the most basic and fundamental characteristics of consumers, are
often included in identifying consumer segments or target markets. In this way,
a target group or market segment, which is similar enough to one another on a
set of AIOs criteria but different enough from other groups or clusters of
people, can be identified. Market segments have high value for developing
advertising and marketing strategy (Fletcher & Bowers, 1991).
This study intends to expand previous lifestyle research in a Chinese context,
in which consumers have increasing freedom in choices coupled with increasing
disposable income. The AIOs framework of analysis will guide the present study
to explore patterns of lifestyle in Chinese consumers.
3. Methodology
Sample and sampling procedures. The data for this study were based on three
parallel consumer surveys conducted in three most important markets in China:
Beijing, Shanghai and Guangzhou in 1995. The surveys relied on random samples
using multi-stage stratified cluster sampling procedures. That is, city
districts were drawn at the first stage, followed by residential committees.
Households were drawn at the final stage. Such demographic variables as age and
sex were used as control in finalizing the samples.
The size of the three samples from Beijing, Shanghai and Guangzhou was
comparable, targeting urban residents aged between 12 and 60. Respondents
sampled from Beijing were interviewed by trained interviewers face-to-face.
Among the 1,024 questionnaires distributed in Beijing, 1,005 were collected with
a response rate of 98.1%. The surveys in Shanghai and Guangzhou were
self-administered. A total of 1,010 questionnaires were sent out in Shanghai,
998 were completed, producing a response rate of 98.8%. The total of
successfully completed questionnaires in Guangzhou was 950 out of 1,010. The
response rate was also high at 94.1%. Altogether, the final sample size was
2,953. At the 95% confidence interval, the sum of sampling errors was plus or
minus 4% (for more details, refer to IMI Consumer Surveys 1995, 1996).
Profiles of the sample. The sample for the present study was almost evenly
distributed in the three metro cities. One third came from Beijing (34.0%). The
remaining two thirds were from Shanghai (33.8%) and Guangzhou (32.2%). The
sample was 52.7% male and 47.3% female, ranging in age from 12 to 59 (M=34.5,
SD=12.15). Thirty-five percent of respondents were married, and 65% were not
married. In terms of education, no more than 5% finished primary school or
less; and 11.6% completed higher education. The majority finished at high school
(43%). Blue-collar workers were 26.1%; professionals were 18.5%, and government
employees were 7%. Service workers, joint-venture employees,
teachers/researchers, self-employed, unemployed and students made up the
remaining 48.3%. For monthly average income, the distribution was widely ranged
from zero to a maximum of $8,000 (Renminbi Yuan)(M=$504). About 36.4% reported
an average income less than $500, 42.6% between $501 and $1,000, and 20.9% more
than $1,000. More than 40% spent less than $300 on daily living every month ,
31.1% spent between $301 and $500, and 26.2% spent more than $500 (see Table 1).
[Table 1 about here.]
Measures of lifestyles, consumption and attitude toward advertising. A wide
range of inventories and questionnaires were checked in constructing the
questionnaire for the 1995 IMI Consumer Surveys. The sources included the US
"AIOs" inventories (i.g., Cosmas, 1982; Plummer, 1974; Wells, 1974) and those
market surveys completed in Taiwan. Specifically, the AIOs inventories covered
buying behaviour, shopping habits, entertainment sought and participation in
social and community events. Items totaled 52. The interests items included
fashion, food, money, and job. The opinion items were those about respondents
themselves, on social issues, on products and brands, and on the future among
other things. The total was 132 items on a 5-point Likert scale, including those
gauging respondents' perception of advertising.
4. Analyses and Results
To answer the two research questions, this study used a three-step procedure
for statistical analysis. In the first step, a Principle Components Factor
Analysis with Varimax Rotation was applied to determine the possible lifestyle
dimensions. The second step clustered the respondents on the basis of the factor
scores obtained from the earlier step. The purpose of cluster analysis was to
sort them into segments that shared a similar lifestyle. In the final step,
discriminant analysis was performed to seek profiling of the cluster centers
which represented different lifestyle segments.
(1) Factor analysis of lifestyle indicators
Out of a total of 30 items, a six-factor solution emerged (eigenvalues greater
than 1.0). It had 24 items with a total variance explained of 44.3%. Six items
with a communality less than .30 were eliminated. Table 2 shows the emerging
lifestyle dimensions with factor loadings.
[Table 2 about here.]
The first factor consisted of six items that were concerned with pursuing
fashions as well as a trendy and stylistic living. This factor has been labeled
"Fashionables and Sophisticates." It is the strongest factor of the solution,
explaining 13.5% of variance. The second factor, which was labeled "Restrained
and Routines," was made up of three items that dealt with what role schedules,
plans and diet played in respondents' life. This factor captured the extent to
which schedules and routines were central. The second factor accounted for 10.2%
of variance explained.
The third factor had four items concerning the strong desire for making efforts
to increase social mobility with accomplishments. It was named "Socially Mobiles
and Life Expansionists." The fourth factor consisted of five
content-with-the-status-quo items. This particular factor captured the tendency
of being easily satisfied with life, the predominance of stability and strong
resistance to change. It has been called "Content with Status quo." The third
and fourth factors explained 6.1% and 5.3% of variance respectively. The fifth
factor was titled "Social Actives and Influencers" for it included four items
that indicated respondents' active participation in social events and the amount
of perceived influence on others in society. The sixth factor was formed by two
items that highlighted the concern for cash flows and the obsession with making
money. It was thus labeled "Money Conscious." The last two factors accounted for
4.7% and 4.5% of variance explained respectively.
In summary, the factor analysis elaborated the emerging different lifestyles in
China (the sample size decreased because of missing cases, but it was still
large enough, N=1,595, to allow valid factor analysis). Although the indicators
are not completely identical, the lifestyles obtained in this study bear
resembles to some of the lifestyle categories or typologies developed by
Mitchell (1983) and Cosmas (1982). For instance, "The Restrained and Routines"
and "Money Conscious" lifestyles are similar to the need-driven categories of
"Survivors" and Sustainers" in the Mitchell study. "Fashionables and
Sophisticates," "Socially Mobiles and Life Expansionists" have similarities to
those "Mobiles," "Sophisticates" and "Life Expansionists" lifestyle groups in
the Cosmas study.
(2) Cluster analysis of lifestyle dimensions
Cluster analysis was performed next to group respondents into lifestyle
segments. In this way, homogenous groups or clusters of respondents who were
relatively homogenous within a cluster center while heterogeneous to other
cluster centers could be identified. Specifically, factors scores on the six
lifestyles were entered into a cluster analysis. Because of the large number of
respondents, it was not possible to apply Hierarchical Clusters. Instead,
K-Means cluster analysis (McRae, 1971) was used. Cluster solutions for four,
five and six clusters were compared in selecting the "best" cluster solution. In
doing so, the considerations were based on two criteria: (1) how well did each
factor cluster the sample; (2) how interpretable was the cluster solution.
In the initial four-cluster solution, scores on the cluster centers from Factor
2 "Restrained and Routines" and Factor 6 "Money Conscious" were low. The
clustering of lifestyle segments thus mainly came from the other four
lifestyles. The six-cluster solution improved the four-cluster solution. All of
the six factors scored high on the six cluster centers. However, Factor 1
"Fashionables and Sophisticates" scored high on more than three cluster centers,
making it difficult to interpret the solution. Therefor, the five-cluster
solution was accepted. It had four comparable-sized clusters. Each of the six
factors had shown some clustering power, and the solution was much more
interpretable. The cluster scores for the five cluster centers appear in Table
3.
[Table 3 about here.]
The first cluster consisted of 246 respondents who scored high on "Restrained &
Routines," "Socially Mobiles & Life Expansionists," and "Fashionables &
Sophisticates." They scored low on "Social Actives & Influencers" and "Content
with Status-quo." These respondents were labeled "Traditionalists" because they
highly agree that life goes around restraints and routines. They also seemed to
disapprove of climbing up the social ladder and experiencing a fashionable and
sophisticated lifestyle. The second cluster was make up of 295 respondents whose
scores were high on "Social Actives & Influencers" and "Fashionables &
Sophisticates," low the other four lifestyles. This group clearly indicates a
segment in its own right. They were called "Status-quo" people, for they tend to
be socially inactive, not taking-charge, and with little influence on others. A
fancy and distinctive lifestyle is not desirable to them either.
The third cluster had 340 respondents who gave relatively high rankings to
"Fashionables & Sophisticates," "Money Conscious," and "Socially Mobiles & Life
Expansionists." This group bore little similarity to the Traditionalists and
Status-quo segments. Thus, it was named "Modern." They desire to experience
fashions and pursue a fancy and distinctive lifestyle. At the same time, they
are obsessed with making money to expand life upward. The fourth cluster
consisted of 382 respondents, the largest group. These 382 respondents scored
highest on the "Content with the status quo" lifestyle. Their scores on other
lifestyles were very low. This group was called "Transitioners" because they
demonstrate the discontent with the status quo. The desire for change
characterised this segment. The last cluster was made up of 334 respondents
whose scores were highest on "Restrained & Routines" and "Money Conscious."
Scores of this group on other lifestyles were lower, particularly on "Socially
Mobiles & Life Expansionists." "Generation X" was the label for this group,
because they are not oriented toward restraints and routines, and not conscious
about money. Subsequent analyses reveal that this is the youngest segment and is
the most well-educated comparing to other four segments (see Table 4).
Analysis of variance (ANOVA) was performed between the five segments to explore
their differences in demographic characteristics. The results, as shown in Table
4, indicate a significant difference in age (p<.001). Generation X is the
youngest, while Transitioners, Modern and Status quo consumers are increasingly
older. Traditionalists turn out to be the oldest. The level of education among
the five segments is highest for Generation Xers, followed by the Modern
segment. The other three segments are more or less equally lower in education
attainment (p<.06). In terms of monthly average income and expenditure, it seems
that the Modern segment is the most affluent, while the Traditionalist segment
is the least affluent. Not surprisingly, the Modern group shows the highest
spending and buying power (p<.01) among the five segments. The buying power of
Generation X is the second highest, also above the mean. The Status-quo segment
shows the lowest spending, well below the mean.
[Table 4 about here.]
In sum, demographically, the Traditionalists are marked by old age, poor
education, and poverty. Thus, they tend to be old-fashioned, and resistant to
change psychographically. The Status-quo people share marked demographic
characteristics with the Traditionalists. However, they differ psychographically
from the Traditionalists in that they are younger, and have not reached a stage
where life goes around traditions and old-fashioned routines. Transitioners
differ little in education attainment and income from the Traditionalist and
Status-quo segments. They rather differ significantly in psychographic
directions from them, because they are much younger, and eager to change the
status quo. This segment is highly ambitious. Consumers belong to this segment
strive to get ahead through self-improvement. Since this is a group in
transition, Transitioners may stay in this segment for just a short period of
time.
The Modern segment moves in another different psychographic direction. Being
the most affluent, well-educated, and probably having a professional job, they
are interested in leading a visible and materialistic life. This segment may be
the core of the emerging middle-class in China. "Generation X," on the other
hand, consists of those who were born after the "Cultural Revolution." Some of
them are the single child in the family. Though many of them are still
completing their education, their education level has been the highest among the
five segments. Psychographically, Generation Xers are marked by disrespect for
routines and tradition. They show little worry about money. Finally, they are
more likely to be hedonistic. To a large extent, these demographic differences
validify the segmentation of surveyed respondents into different lifestyles
through cluster analysis.
(3) Multivariate discriminant analysis of lifestyle segments
In the last step, discriminant analyses were employed to establish profiles of
the five lifestyle segments. The criterion variables included perception of
advertising, buying behaviour, and preferences for consumption of Western
products vis- -vis things Chinese (the ratio between sample size and the number
of variables for discriminant analysis is well about 20:1; thus discriminant
analysis was justified, see Stevens, 1996). The expectation is that respondents
belong to different segments should also differ in those aspects in addition to
the marked demographic differences. Discriminant analysis would seem appropriate
to analyze the potential differences in viewing advertising, buying and
consumption patterns simultaneously. Exploring multivariate patterns of group
differences among segments is superior to testing group differences on a single
variable at a time (Huberty, 1984, 1994; Stevens, 1996).
Perception of advertising. Table 5a presents the result of the first
discriminant analysis of the five lifestyle segments on their perception of
advertising. A total of seven attitudinal variables on advertising were entered
as predictors following initial single-variable ANOVA tests. A stepwise
procedure of discriminant analysis was used to eliminate those variables without
satisfactory discriminating power. This procedure locates the variable that best
discriminates among the segments, finds the next best discriminator, and so on,
making the five lifestyle segments as distinctive as possible. Four out of the
seven discriminant variables showed sufficient discriminating power.
[Table 5a about here.]
On the four possible functions that could discriminate among the five segments,
only two functions are statistically significant (Wilks's lambda=.97 and .99,
p<.001). Table 5a describes each function with standardized canonical
coefficients (similar to beta weights in multiple regression analysis). It seems
that the first significant function is primarily a function of perception of
trust in advertising and evaluation of advertising content, to a lesser extent,
of trust in advertised brands. High values of Function 1 are associated with a
negative view of having trust in advertising and criticism of advertising
content, but holding a positive view toward advertised brands. Thus, it can be
described as "practical" (negative about advertising in general but positive
about benefits of advertising). It explains the largest proportion of the total
variance (59.5%).
High values of the second significant function, on the other hand, are
associated with both positive views of advertising in general and advertised
brands in particular. Thus, it is likened to "positive," accounting for 24.7% of
variance. High values of Function 3 are positively associated with "no need to
advertise," and negatively with "ads are filled with recycled ideas." This
particular function, though not significant, can still be characterized as
"negative."
The mean for each lifestyle segment (group centroids) on each function is shown
in Table 5b. The first significant function (practical) discriminates between
the Transitioners and the other four segments. The implication is that the
Status-quo, Traditionalist and Modern segments hold a practical view of
advertising, while the view of Transitioners, to a less extent, Generation Xers,
is less practical. Function 2 (positive) discriminates mainly the Modern,
Status-quo and Generation X segments. Such a result indicates that respondents
pursuing a Modern lifestyle and those belonging to Generation X have a
favourable attitude toward advertising. In comparison, the Status-quo segment
shows the least favourable attitude about advertising. Although not
statistically significant, Function 3 (negative) suggests more differences in
perceiving advertising among the five segments. Generation Xers seem to be the
least negative about advertising, while the Traditionalist segment has the most
negative perception of advertising. Such interpretations, however, should be
read with caution.
[Table 5b about here.]
Summary. The directional differences in perceiving advertising of each
lifestyle segment can be summarised as follows based on a discriminant plane:
The Modern segment is the least critical of advertising, for it scores the
highest on the "positive" function and second highest on the "practical"
function. Generation X is similar to the Modern segment in holding a positive
view of advertising, but it tends to be a bit critical of advertising content.
The Status-quo and Traditionalist segments share similar characteristics
between them, but show differences to other three segments. They do not
particularly dislike advertising content, but they disapprove of advertising in
general. It is interesting to note that the Transitioner segment differs from
all other four segments in perceiving advertising. This segment is unfavourable
about advertising in general, and sees the least benefit of advertising.
Therefore, it reveals its own view. Such results indicate that views on
advertising are not necessarily positive across Chinese consumers as Pollay, Tse
and Wang (1990) suggested, but a function of various lifestyles.
Buying behaviour and preferences for consumption. Following similar procedures,
the five segments were analyzed for differences in buying behaviour and
preferences for consumption. A battery of 21 shopping and consumption measures,
which survived the ANOVA test, were entered for discriminant analysis. Those
measures concerned various buying activities, such as bargain hunting, shopping
for quality and fancy products as well as preferences for Western lifestyle and
products vis- -vis things Chinese. Table 6a reports the result (nine variables
that made no contribution to the discriminant functions in subsequent analysis
were eliminated).
[Table 6a about here.]
Although five segments produced four functions, only Function 1 and Function 2
are significant (Wilks's lambda=.81 and .95, p<.001). The first significant
function seems to represent free spending and a preference for Western
lifestyle. The high values of Function 1 are correlated with buying fancy and
stylistic products, frequent visits to night markets, and picky shopping
behaviour. At the same time, aspiration for Western lifestyle is a factor of
significant importance in this function. It can be described as "free-spending
and Western-oriented," explaining the largest proportion of the total variance
(49.9%). High values of the second significant function, on the other hand, are
correlated with preference for Chinese food and shying away from Western
fast-food such as McDonld's. Moreover, concerns for saving and careful shopping
behaviour are correlated to this function. Thus, Function 2 is characterized as
"thrifty and Chinese-oriented." It accounts for 39.1% of variance.
Table 6b shows the mean for each lifestyle segment (group centroids) on each
function. The first function (free-spending and Western-oriented) mostly
discriminates between the Modern and Transitioner segments and the remaining
three segments. The interpretation is that the Modern segment is high on
spending freely and desiring for Western lifestyle. The Transitioners are the
least in free spending with the least preference for a lifestyle modeled after
the West. Function 2 (thrifty and Chinese-oriented) discriminates mainly the
Transitioner, Generation X and Traditionalist segments. Such a result
demonstrates that respondents pursuing an old-fashioned and traditional
lifestyle care about saving and prefer Chinese products and culture. In
contrast, Transitioners and Generation Xers show less concern for saving in
shopping and less preference for things Chinese.
[Table 6b about here.]
Summary. Characteristics of each consumer segments with distinctive lifestyles
can be summarised in a discriminant plane. The key discrimination for the Modern
consumers are "big spenders" on finer things in life to reflect their status and
success, and the strongest desire for Western lifestyle. At the same time, they
care about getting money's worth and have no objection for Chinese products.
Generation X is the second high on the "free spending and Western-oriented"
function, coupled with little concern for saving as well as avoiding Chinese
products. The Traditionalist and status-quo segments share similar
characteristics: both avoid spending freely in shopping and desire no Western
lifestyle. Moreover, they share the saving desire and favour things Chinese. The
Transitioner segment is dissimilar to the other four segments. Transitioners
tend to avoid speeding freely, but saving is not a primary consideration. At the
same time, they favour neither Western lifestyle or things Chinese. They thus
reflect a typical lifestyle that is in transition.
5. Conclusions and Discussion
The findings of this study strongly support a conclusion that various
lifestyles have emerged in reformed China. Chinese consumers, at least those
from metro areas, have grown into different lifestyle segments. This study
empirically established and elaborated five major lifestyle segments based on
measures of AIOs indicators: Traditionalists, Status-quo, Modern, Transitioners
and Generation Xers. Each segment is homogeneous within itself, but
heterogeneous comparing to others.
Furthermore, this study found some important differences among the five
segments with directional psychographic differences concerning advertising,
buying behaviour and preferences for consumption. Traditionalist and Status-quo
consumers disapprove of advertising in general, desire for saving and prefer
things Chinese. The Modern and Generation X segments view advertising more
positively than any other segment, spend freely and favour Western lifestyle.
Transitioners differ from the others, reflecting a transitional lifestyle. Thus,
the emergence and growth of various lifestyles in contemporary China do have
far-reaching consequences for how advertising is viewed, consumption patterns
and preferences for Western vs. Chinese products.
To conclude, this study provides empirical support for the advent of a consumer
society in post-Mao China. China's ambitious modernisation drive has led to the
rise of different classes and status groups in consumers. More importantly, the
results of this study shed some light on how to approach Chinese consumers
beyond demographics. Western products and fashions should target those consumers
pursuing a Modern lifestyle and Generation Xers, to a less extent,
Transitioners, who are willing to spend, favourable toward Western lifestyle and
more receptive to advertising.
This study has successfully established and elaborated six differentiated
lifestyles. However, they are by no means exhaustive (the number of missing
cases may somewhat affect the findings). Furthermore, the sample used for
analysis was drawn from the metro areas of China, the generalisability of the
results remain to be tested. Future research, therefore, can expand the present
study by attempting a nationwide sample and further refine the lifestyle
categories in a Chinese context.
References
Anderson, T. & Golden, L. (1984) Lifestyles and psychographics: A critical
review
and recommendation. In Advances in Consumer Research XI (Ed.) Kinnear, T.,
pp.
405-411. Ann Arbor, Mich: Association for Consumer Research.
Bell, D. (1976) The Coming of Post-Industrial Society. New York: Basic Books.
Blumer, H. (1969) Fashion: From class differentiation to collective selection.
The
Sociological Quarterly, 3, 275-291.
Cosmas, S. (1982) Life styles and consumption patterns. Journal of Consumer
Research
, 8, 453-455.
Fletcher, A. & Bowers, T. (1991) Fundamentals of Advertising Research, 4th
Edition.
CA: Wadsworty.
Hawkins, D., Best, R. & Coney, K. (1983) Consumer Behaviour: Implications for
Marketing Strategy. TX: Business Publications.
Huberty, C. (1984) Issues in the use and interpretation of discriminant
analysis.
Psychological Bulletin, 156-171.
Huberty, C. (1994) Applied Discriminant Analysis. New York: Wiley.
IMI Consumer Surveys 1995 in Being, Shanghai & Guangzhou. (1996). Beijing: China
Finance and Economy Press.
Johansson, T. & Miegel, F. (1992) Do the Right Thing: Lifestyle and Identity in
Contemporary Youth Culture. Stockholm: Almqvist & Wiksell International.
Johansson, T. (1994). Later modernity, consumer culture and lifestyles: Toward a
cognitive-affective theory. In Media Effects and Beyond: Culture,
Socialisation and
Lifestyles (Ed.) Rosengren, K., pp. 265-294. London: Routledge.
Lewis, R. (1973) The New Service Society. London: Longman.
Maslow, A. (1968) Toward a Psychology of Being. New York: Van Nostrand Co.
McRae, D.J. (1971) Micka, A fortman IV iterative K-Means cluster analysis
program.
Behavioural Sciences, 16, 423-424.
Mitchell, A. (1983) The Nine American Lifestyles: Who We Are and Where We Are
Going
. New York: Warner Books.
Moven, J. (1990). Consumer Behaviour, 2nd Edition. New York: Macmillan.
O'Neill, M. (1997) Returning to class values. South China Morning Post, January
24,
24.
Pingree, S. & Hawkins, R. (1994) Looking for patterns in Lifestyles Behaviors.
In
Media Effects and Beyond: Culture, Socialisation and Lifestyles (Ed.)
Rosengren, K.
pp. 76-96. London: Routledge.
Pitts, R.E. & Woodside, A.G. (1984) Personal Values and Consumer Psychology.
Toronto: Lexington Books.
Plummer, J. (1974) The concept and application of life style segmentation.
Journal
of Marketing, January, 33-37.
Pollay, R., Tse, D. & Wang, Z. (1990) Advertising, propaganda, and value change
in
economic development: The new cultural revolution in China and attitudes
toward
advertising. Journal of Business Research, 20, 83-95.
Settle, R. & Alreck, P. (1989) Why They Buy: American Consumers Inside and Out.
New
York: John Wiley & Sons.
Sobel, M. (1983) Lifestyle expenditures in contemporary America. American
Behavioural Scientist, 26-4, 521-533.
Settle, R. & Alreck, P. (1989) Why They Buy: American Consumers Inside and Out.
New
York: John Wiley & Sons.
Stevens, J. (1996) Applied Multivariate Statistics for the Social Sciences, 3rd
Edition. New Jersey: LEA.
Uusitalo, M. (1979) Consumption Style and Way of Life: An Empirical
Identification
and Explanation of Consumption Style Dimensions. Helsinki: The Helsinki
Press.
Valette-Florence, P. & Jolibert, A. (1990) Social values, A.I.O., and
consumption
patterns. Journal of Business Research, 20, 109-122.
Wells, W.D. (1974) Life style and psychographics: Definitions, uses and
problems.
In Life Style and Psychographics (Ed.) Wells, W.D, pp. 354-363, American
Marketing
Association.
Zablocki, B. & Kanter, R. (1976) The differentiation of lifestyles. Annual
Review
of Sociology, 2, 269-298.
Table 1: Profile of the Sample from the Three Metro-cities in China
________________________________________________________________________________
_
Region Beijing 34.0%
Shanghai 33.8%
Guangzhou 32.2%
Age 12 to 29 33.7%
30 to 40 34.1%
41 to 59 32.3%
Gender Man 52.7%
Women 47.3%
Education Below high school 32.5%
High school 42.9%
Above high school 24.6%
Marital status Single 34.8%
Married 65.2%
Occupation Workers 26.1%
Professionals 18.5%
Government employees 7.0%
Service workers 8.7%
Joint ventures employees 7.7%
Teachers/researchers 12.7%
Self-employed 2.7%
Unemployed 11.9%
Students 3.9%
Monthly average income $500 and less 36.4%
$501 to $1,000 42.6%
$1,001 and more 20.9%
Monthly average expenditure $300 and less 42.7%
$301 to $500 31.1%
$501 and more 26.2%
________________________________________________________________________________
_
Note: N ranges from 2,953 to 2,914
Table 2: Factor Analysis of Lifestyle Indicators
________________________________________________________________________________
_
Statements Factor 1 Factor 2 Factor 3 Factor4 Factor 5 Factor 6
________________________________________________________________________________
_________
Fashionables & Sophisticates
1. I'm fashionable in the eyes of others. .69 .03 -.06 -.16 .16 -.09
2. I dress up to show off my personality. .68 .10 .00 .04 .08 .03
3. I enjoy having new and fashionable things. .65 -.10 .04 .00 .13 .04
4. I pay close attention to trends in fashion. .64 -.01 .04 .06 .04 .01
5. A fancy and distinctive living attracts me. .58 .16 .12 .23 .02 -.03
6. I enjoy stylistic dresses. .50 .05 .22 .06 -.07 .08
Restrained & Routines
1. My life centers around schedules. -.06 .73 -.00 .09 .04 -.07
2. I'm watchful for my diet. .07 .70 .07 .02 -.02 -.01
3. Planning is my habit in doing things. -.05 .55 .25 .10 .28 .10
Socially Mobiles & Life Expansionists
1. Life is meant to take on challenges and risk. .14 -.09 .65 .03 .08 .08
2. I'll take some courses to brighten my future. .07 .12 .65 -.00 -.02 -.06
3. Doing nothing will make me uncomfortable. .14 .05 .56 .09 .15 -.06
4. I have high hopes on what I can accomplish. .07 .11 .50 .01 .15 .21
Content with Status quo
1. A living space of my own will make me happy. .04 -.01 .12 .64 .06 -.04
2. I prefer to do nothing but relax during holidays. .14 .02 .10 .64 .02 -.12
3. Economic security is primarily important to me. -.02 .25 .24 .50 .05 .37
4. Jobs should not be changed easily if they are ok. -.03 .09 -.17 .49 .14 .10
5. I prefer stable and secure jobs.* -.09 .40 -.10 .42 -.11 .15
Social Actives & Influencers
1. I can mingle with strangers easily. .01 -.07 -.03 .10 .64 -.04
2. I'm active at social functions. .21 .03 .21 .02 .63 -.10
3. Hesitation is not my style in doing things. .01 .28 .10 .04 .58 .21
4. I have a lot of influences over my friends. .27 -.00 .14 .10 .46 -.01
Money Conscious
1. I'd rather work for money than resting. .09 -.16 -.04 -.12 .07 .80
2. I check my balances periodically. -.02 .24 .18 .26 -.11 .47
________________________________________________________________________________
_____
Eigenvalue 3.25 2.45 1.45 1.28 1.14 1.06
Variance explained 13.5% 10.2% 6.1% 5.3% 4.7% 4.5%
________________________________________________________________________________
_____
The scale was: 1= Strongly disagree; 2=Disagree; 3=Neutral; 4=Agree; 5=Strongly
Agree.
* Item with cross loadings. (N=1,595)
Table 3: Cluster Centers Scores on the Six Lifestyles
________________________________________________________________________________
_
Cluster 1 Cluster 2 Cluster 3 Cluster 4 Cluster 5
________________________________________________________________________________
_
Traditionalists Status-quo Modern Transitioners Generation Xers
(n=246) (n=295) (n=340) (N=382) (n=334)
________________________________________________________________________________
_
Fashionables & Sophisticates -.48 -.74 .82 .16 .38
Restrained & Routines .79 -.32 .34 -.22 -.91
Socially Mobiles & Life Expansionists -.59 .26 .58 -.16 -.18
Content with Status quo .17 .38 .29 -1.56 .51
Social Actives & Influncers .12 -1.04 .48 -.17 .33
Money Conscious -.20 .33 .63 -.00 -.82
________________________________________________________________________________
_
Table 4: Demographic Characteristics of the Five Lifestyle Segments (ANOVA)
________________________________________________________________________________
_
Age Education Income Expenditure
________________________________________________________________________________
_
Total 1.99 1.93 1.85 1.83
Traditionalists 2.16 1.89 1.80 1.78
Status-quo 2.09 1.89 1.83 1.76
Modern 1.92 1.97 1.91 2.00
Transitioners 1.89 1.89 1.82 1.78
Generation Xers 1.79 2.03 1.86 1.86
________________________________________________________________________________
_
F 11.62 2.31 1.06 4.65
Significance p<.001 p<.06 p=n.s. p<.01
________________________________________________________________________________
_
Age categories: 1=12-29; 2=30-40; 3=41-59.
Education categories: 1=Below high school; 2=High school; 3=Above high school.
Monthly income categories: 1=$500 and less ; 2=$501-1,000; 3=$1,001 and more.
Monthly expenditure categories: 1=$300 and less; 2=$301-500; 3=$501 and more.
Table 5a: Summary of Discriminant Function Analysis of the Five Cluster
Centers
on Perception of Advertising
________________________________________________________________________________
_
Statements Function 1 Function 2 Function 3 Function 4
________________________________________________________________________________
_
Advertising can be trusted. -.61 (-.53) .82 (.75) .22 (.11) .13
(.38)
Advertised brands are trustworthy. .54 (.35) .63 (.46) .02
(.02) -.69 (-.75)
Good products need no advertising. .47 (.46) .06 (.03) .80 (.85) .38
(.28)
Ads are filled with recycled ideas. .58 (.47) .26 (.37) -.52 (-.58) .56
(.57)
Eigenvalue .05 .02 .01 .00
Variance explained 59.5% 24.7% 14.9% 0.1%
Canonical correlation .22 .15 .11 .03
Degrees of freedom 9 4 1 --
Wilks's lambda .97 .99 .99. --
Significance p<.001 p<.001 p=n.s. p=n.s.
________________________________________________________________________________
_
Note: Variables are ordered based on their function loadings. Standardized
discriminant function coefficients and structure coefficients (in parentheses)
are reported.
Scale: 1=Strongly disagree; 2=Disagrre; 3=Neutral; 4=Agree;= 5=Strongly Agree
Table 5b: Canonical Discriminant Function Centroids of the Five Lifestyle
Segments
________________________________________________________________________________
_
Function 1 Function 2 Function 3 Function 4
________________________________________________________________________________
_
Traditionalists .12 -.07 .12 -.03
Status-quo .19 -.19 -.11 .02
Modern .14 .21 .08 .03
Transitioners -.46 -.07 .05 .02
Generation Xers -.08 .14 -.18 -.03
________________________________________________________________________________
_
Note: Entries are standard scores.
Table 6a: Summary of Discriminant Function Analysis of the Five Cluster
Centers
on Buying Behavior and Preferences for Consumption
________________________________________________________________________________
_
Statements Function 1 Function 2 Function 3 Function 4
________________________________________________________________________________
_
I enjoy buying stylistic products. .28 (.48) -.05 (-.05) .13 (.07) .03
(.06)
I prefer to buy quality and costly products. .43 (.62) .45 (.44) -.21
(-.14) .13 (.04)
It's worthwhile to spend on vacations. .29 (.46) -.20 (-.21) .13
(.11) -.41 (-.40)
I care much about the atmosphere of stores. .28 (.48) -.12 (-.11) -.07
(-.04) .08 (.04)
I read packing information with care. .10 (.28) .20 (.30) .34
(.28) .16 (.10)
Price is my primary consideration in buying. .21 (.14) .23 (.31) -.08
(-.04) .24 (.21)
I enjoy shopping at night markets. .31 (.42) -.15 (-.27) -.07 (-.01) .11
(.16)
I charge a lot purchases to my credit cards. .05 (.14) -.23 (-.28) .39
(.49) -.57 (-.45)
The Western lifestyle is desirable to me. .23 (.34) -.19 (-.30) .03
(-.05) -.14 (-.11)
I often eat at McDonald's. .12 (.26) -.35 (-.52) .10 (.11) .63 (.52)
I enjoy Chinese operas. -.03 (-.14) .20 (.21) .80 (.81) .29 (.23)
I prefer to eat Chinese food to Western. .11 (.23) .41 (.55) .03
(.02) -.29 (-.29)
Eigenvalue .22 .18 .04 .01
Variance explained 49.9% 39.1% 9.3% 1.8%
Canonical correlation .43 .39 .20 .09
Degrees of freedom 33 20 9 --
Wilks's lambda .81 .95 .99 --
Significance p<.001 p<.001 p=n.s. p=n.s.
________________________________________________________________________________
_
Note: Variables are ordered based on their function loadings. Standardized
discriminant function coefficients and structure coefficients (in parentheses)
are reported.
Scale: 1=Strongly disagree; 2=Disagrre; 3=Neutral; 4=Agree;= 5=Strongly Agree
Table 6b: Canonical Discriminant Function Centroids of the Five Lifestyle
Segments
________________________________________________________________________________
_
Function 1 Function 2 Function 3 Function 4
________________________________________________________________________________
_
Traditionalists -.25 .42 .18 .09
Status-quo -.22 .38 -.31 -.07
Modern .82 .04 .12 -.07
Transitioners -.57 -.55 .15 -.09
Generation Xers .20 -.51 -.20 .12
________________________________________________________________________________
_
Note: Entries are standard scores.
|
