Method

Sampling

A random sampling method was applied. Four major Chinese cities “ Beijing, Hangzhou, Guangzhou and Xian “ were selected, representing northern , eastern , southern and western China respectively. Data were collected by research assistants in local cities, using personal interview methods in key local shopping districts. These research assistants were mainland Chinese students recruited by referral and trained in the basic techniques of conducting personal interviews. Proper guidance and instruction were given before data collection started. Mall intercepts were carried out between December 1999 and February 2000 with every fifth person until a target of 200 was reached. Finally, 200 usable questionnaires each were collected from Beijing, Hangzhou and Xian. Some questionnaires from the Guangzhou group had a significant number of missing values resulting in only 196 usable cases. In the aggregate sample, there were 391 females and 402 males so the sample is evenly distributed between males and females.

40.2 per cent were aged 17 “25, 24.8 per cent 26 “35,
15.7 per cent 36 “45, 15.9 per cent 46 “55 and 3.4 per cent 56 “68. These age groups represent those who are in active work and financially independent.

Measurement

After an extensive literature review, ten items related to trust in business ethics, environmental consciousness and health consciousness were selected to measure lifestyle dimensions. The questionnaire was initially developed in English, translated into Chinese and translated back into English following standard blind procedures to ensure the accuracy of translation. Finally, a Chinese version was used for data collection. A five- point Likert type scale was used for both lifestyle and marketing mix statements on consumers' post purchase perceptions of product quality, service quality and price, ranging from (1) 'strongly agree' to (5) 'strongly disagree '.

To ensure that all the measurements have good internal consistency, a Cronbach alpha test was applied to the three derived scales: trust in business ethics (a=0.68), environmental consciousness (a=0.62) and health consciousness (a=0.55). While alphas for two out of three scales are reasonably acceptable, alpha for the health consciousness construct is slightly lower than the threshold of a=0.60. However, from an exploratory perspective, a slightly lower alpha is still acceptable (Nunnally, 1978). To confirm the existence of three dimensions, a factor analysis was conducted . A principle component analysis was used as the extraction method. The rotation converged on four iterations, using the varimax method. It can be observed from Table 4.8.1 that three factors (Eigenvalue >1) were derived, resulting in an accumulated explained variance of 58.67 per cent.

Table 4.8.1: Factor analysis of lifestyle variables

	Factor 1	Factor 2	Factor 3
Trust in Business Ethics (a=0.69)
You believe that product quality is improving	.721	117	“7.339E-02
You believe that firms are beginning to pay attention to customers' complaints	.721	4.763E	“6.228E-02
You believe that product trademarks start to follow rules and regulations	.720	159	7.097E-02
You believe that the dangers arising from using products are decreasing	.707	“9.076E-02	5.912E-02
Environmental Consciousness (a=.62)
My country should invest in environmental protection	1.900E-02	.820	3.729E-02
The possibility of an energy crisis still exists	2.112E-02	.760	.188
From the country's point of view, the one child policy is necessary	.132	.666	1.006E-02
Health Consciousness (a=.55)
Drinking is a bad hobby	3.293E-02	“3.172E-02	.865
Smoking can cause cancer	5.635E-02	.249	.768
Eigenvalue	2.081	1.805	1.394
Variance explained	23.126	20.057	15.491
Extraction Method: Principal Component Analysis.
Rotation Method: Varimax with Kaiser Normalization.
Rotation converged in 4 iterations.

Results

Data from four cities “ Guangzhou, Beijing, Hangzhou, and Xian “ were compared using a one-way ANOVA post hoc Scheffe test (see Table 4.8.2). There is no difference in consumers' health consciousness across the four cities and with all the arithmetic means below 2.02 ('1' strongly agree and '5' strongly disagree), indicating that Chinese consumers in all four cities are very health conscious.

Table 4.8.2: One-way ANOVA post hoc Scheffe Test of differences in lifestyles among Guangzhou, Beijing, Hangzhou and Xian

	Guangzhou (1)			Beijing (2)				Hangzhou (3)			Xian (4)
	SD	N	Mean	SD	N	Mean	SD	N	Mean	SD	N	Mean
Health Consciousness	186	1.946	.930	198	1.980	.714	199	2.020	.763	198	1.846	.846
Confidence in Business Ethics	185	2.635	.607	195	2.542	.545	196	2.297	.515	196	2.431	.551
Environmental Consciousness	178	2.150	.906	195	1.670	.537	197	1.702	.457	195	1.622	.531
a Scales range from (1) 'Strongly Agree' to (5) 'Strongly Disagree'

Consumers in Guangzhou have a lower confidence in business ethics than those in Hangzhou and Xian. While there is no significant difference in consumers' confidence in business ethics between Guangzhou and Beijing, consumers in the latter turn out to be less confident than those in Hangzhou.

Consumers in Guangzhou are less environmentally conscious than in the other three cities. Interestingly, there is no difference between these three cities. Nevertheless, all the respondents indicated a fairly good environmental consciousness as all the arithmetic means are below 2.15.