- Email: [email protected]
Level of Internet use by Chinese businesses: A preliminary study
Available online at www.sciencedirect.com
Electronic Commerce Research and Applications 6 (2007) 453–461 www.elsevier.com/locate/ecra
Level of Internet use by Chinese businesses: A preliminary study Xibao Zhang a
a,*
, Cynthia Moussi
b
Department of International Economics and Trade, Qingdao University, 7 East Hongkong Road, Qingdao 266071, China b Department of Accounting and Information Systems, University of Melbourne, Parkville 3010, Australia Received 1 September 2005; received in revised form 13 November 2006; accepted 8 February 2007 Available online 14 February 2007
Abstract A preliminary study of the level of Internet use by Chinese firms was carried out. Data were collected by a questionnaire survey of a convenience sample of Chinese managers. The major finding is that firm size plays a role only in influencing a firm’s decision regarding whether or not to have Internet access. Within the Internet-using sub-sample, firm size is not significantly related to the overall level of Internet use, or to the use of individual Internet features. In addition, the data reveal that there are a high proportion of Chinese businesses that have Internet access. Chinese Internet using firms use an average of three features, with e-mail and buying and selling online being the two most widely used features. The results are also compared to similar studies in other countries. The limitations of this study and future research are discussed. Ó 2007 Elsevier B.V. All rights reserved. Keywords: E-commerce; Level of Internet use; Chinese businesses; Internet use index; Internet development in China
1. Introduction In line with the worldwide trend, Internet use in China has experienced rapid growth in recent years. On the research front, there have been a number of studies by international scholars on Internet diffusion in China (e.g., [1–3]). However, these studies are of a general nature, covering the general diffusion and use of the Internet in China. Press et al. [3] and Fan [4], for example, both used the Mosaic Group’s Global Internet Diffusion Framework [5], which includes six dimensions (pervasiveness, geographic dispersion, sectoral absorption, connectivity infrastructure, organizational infrastructure, and sophistication of use), in analyzing China’s Internet development. In addition, since both studies used survey data collected by China Internet Network Information Center (or CNNIC, http://www.cnnic.net.cn), their analyses undoubtedly are limited to the general situation of Internet diffusion and
use in China, with little or no discussion of Internet use by Chinese businesses. Inside China, CNNIC [6] publishes a series of semiannual reports on China’s Internet diffusion and use, which is primarily based on online surveys of Internet users. In addition, Guo and Bu [7] also studied Internet use and its influences in five major Chinese cities. However, these studies primarily focused on individual Internet users, and as a result the issue of Internet use by Chinese businesses has not been explored. Therefore, it is the purpose of this paper to study the level of Internet use by Chinese businesses. This paper is part one of a two-part project aimed at investigating Chinese firms’ Internet use and related issues. Part two of the project examines respondents’ perceptions of the benefits of and impediments to e-commerce uptake, the results of which will be published in a second paper. 2. Level of Internet use
*
Corresponding author. Tel.: +86 532 8595 3733; fax: +86 532 8595 0018. E-mail address: [email protected] (X. Zhang). 1567-4223/$ - see front matter Ó 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.elerap.2007.02.007
There have been different definitions of electronic commerce by various scholars, international organizations, and commercial entities. (For a synopsis of such
454
X. Zhang, C. Moussi / Electronic Commerce Research and Applications 6 (2007) 453–461
definitions, please see [8,9].) In this research we use the definition by Zwass [10], i.e., it is ‘‘the sharing of business information, maintaining business relationships, and conducting business transactions by means of telecommunications networks’’, including both inter-organizational relationships and transactions and intra-organizational processes that support these activities. This is a more inclusive definition in that, first, it includes both inter-organizational and intra-organizational activities, and second, it includes not only online transactions but also other activities, such as information sharing and maintaining business relationships, that will eventually lead to online and offline transactions. There are two questions concerning a business’ use of the Internet: first, does it use the Internet at all? And second, if the answer to the first question is yes, to what extent does it use the Internet? These are the two questions that this paper attempts to answer in the context of Chinese firms. Businesses can use the Internet in many different ways and to serve myriad purposes, and different scholars use different terms and concepts to refer to these uses. Abell and Lim [11], for example, found in their empirical study the top eight uses of the Internet by New Zealand firms to be communication within or outside the company, getting information from suppliers, providing customer information, R&D and sharing of software and information, to be seen at the forefront of technology, market and product research, sending orders to suppliers, and receiving orders from customers. Moussi and Davey [8], in their study of Internet use by Victorian manufacturing firms, proposed a five-feature set that measures a business’ level of Internet use to conduct e-commerce, which includes e-mail, homepage, obtaining research information, gathering customer information, and buying and selling online. On the theoretical side, Poon and Swatman [12] proposed seven business activities by small firms on the Internet: consumer prospecting and advertising, involvement in specialized groups for knowledge and intelligence exchange, research and development ideas/opportunities, efficient communications, preparation for the global marketplace, on-demand linkage with customers and suppliers, and geographic reach/general accessibility. Cronin [13] pointed out three areas of Internet use by businesses: customer relations, dealing with customers, and internal company operations. Zwass [14] identified an exhaustive list of potential uses in his 5Cs aspects–opportunities conceptual framework, which categorizes Internet uses into five domains: commerce, collaboration, communication, connection, and computation. Such diversity in both the empirical measures and conceptual constructs of Internet use by businesses reflects the overwhelming impact of the Internet in transforming business activities and processes. For the sake of empirical research, however, there needs to be a simple yet comprehensive set of Internet use measures to gauge their uptake of e-commerce. For this reason the feature-based set of
Internet use measures proposed by Moussi and Davey [8] is used in this study, because it essentially covers all aspects of electronic commerce, and is relatively easy to use in conducting a survey. Regarding the first question, factors need to be identified that influence a firm’s decision to use the Internet for electronic commerce. It has been suggested by early IS scholars that firm size is an important determinant of business IS adoption, and the level of IS adoption is directly related to firm size [16–18]. Firm size, according to Ling [15], is also one of the factors that influence business use of the Internet. In this preliminary study, therefore, only firm size is tested for its influence on Internet use because it can be easily and objectively measured. Empirically, however, there has been somewhat contradictory evidence regarding the influence of firm size on Internet use. The Internet-using sample of Abell and Lim [11] consisted of more smaller firms (those with 49 employees or fewer) than larger ones, suggesting that there may be an inverse relationship between firm size and Internet use. Results from Moussi and Davey’s study [8], however, indicate that there seems to be a direct relationship between firm size and Internet use, i.e., the larger the firm, the more likely it uses the Internet in conducting business. (It should be noted that no statistical tests were performed on the relationship between firm size and Internet use in either Moussi and Davey’s study [8] or Abell and Lim’s [11].) The discrepancy can be explained by the fact that, first, New Zealand firms are predominantly small ones, and second, in the sample of Abell and Lim [11], the majority of smaller firms are those with an IT focus (45% IT focus vs. 28% non-IT focus), while IT firms constitute only a minority in the larger firm group (8% IT focus vs. 22% non-IT focus). The direct relationship between firm size and Internet use as suggested by Moussi and Davey [8] is intuitive in that a smaller firm is more likely to lack the necessary financial resources, knowledge, or personnel to use the Internet. We expect that such a relationship should also hold true for Chinese firms. Therefore: H1: For Chinese businesses, Internet use is directly related to firm size. In order to answer the second question, both the use of each of the five features and the overall level of Internet use need to be explored. Again for the same reason as discussed above, firm size is tested as a potential determinant of such uses. In their study, Moussi and Davey [8] devised an Internet use index, which is an un-weighted sum of all the features of the Internet used by a firm: Use Index ¼
5 X
Features:
n¼1
The value of the above use index ranges from 0 (no use) to 5 (all five features used), and as a result it provides a meaningful measure for analyzing the overall level of
X. Zhang, C. Moussi / Electronic Commerce Research and Applications 6 (2007) 453–461
Internet use by businesses. In addition, their data seem to indicate that there is possibly a direct relationship between firm size and the use index. For each individual feature, their data also appear to show that, with the exception of buying and selling online, as firm size increases, so does the use of that feature. Again the reasoning for both observations is similar to that regarding the first question, namely, generally speaking, smaller firms, even if they use the Internet, do not have the necessary financial resources, expertise, or personnel to fully take advantage of the potentials of the Internet. Such relationships are also expected to hold true for Chinese firms: H2: For Chinese businesses, the overall level of Internet use is directly associated with firm size. H3: For Chinese businesses, the level of use of each Internet feature, with the exception of buying and selling online, is directly associated with firm size. Regarding buying and selling online, Moussi and Davey [8] also observed that there are a higher percentage of firms using this feature among those with 50 employees or fewer than among those with 51 employees or more. This is an interesting observation, and the only plausible explanation may be that buying and selling online represents the highest and most sophisticated use of the Internet, and requires a total overhaul of the internal processes to support such an endeavor. Such an overhaul is possibly easier for smaller businesses simply because of their smaller size. Therefore: H4: For Chinese businesses, the level of use of buying and selling online is inversely related to firm size. 3. Redesigning the use index Clearly each of the five features of the Internet represents a different level of sophistication of Internet use. They can be arranged in ascending order of sophistication as: e-mail, homepage, obtaining research information, gathering customer information, and buying and selling online. E-mail is the least sophisticated feature in that all it requires is an Internet connection and a computer. Homepage comes next in this respect in that the most basic homepage – a static homepage – does not require the firm to update its information content frequently, and it can even be hosted by Web-hosting services. Obtaining research information represents a higher sophistication level than homepage in that the firm uses the Internet to obtain information regarding products, services, customers, prices, market trends, strategic partners, etc. Therefore it requires the firm to use the Internet more actively and more extensively. A firm can also use the Internet to acquire and store customer profiles, their needs, their buying pattern, etc., and automate the retrieval and processing of such data. This is where gathering customer information comes in. Buy doing so the firm learns customers’ needs and deliver products and services to best satisfy such needs, thereby building and maintaining an ongoing relationship. It is a
455
step higher on the sophistication scale than obtaining research information, since this requires the organization to at least overhaul its customer support and marketing processes. Finally, buying and selling online represents the most sophisticated use of the Internet in that this requires the firm to set up and operate an interactive Website to handle online transactions, which is rather costly, and re-engineer virtually all of its business processes. Therefore, a different weight is assigned to each of the five features as follows: e-mail 1/3, homepage 2/3, obtaining research information 3/3, gathering customer information 4/3, and buying and selling online 5/3. The weighted Internet use index can be expressed as UIw ¼
5 X
W iF i;
i¼1
where UIw is the weighted Internet use index, Wi the ith feature’s weight, and Fi the ith feature’s value. Both the un-weighted Internet use index developed by Moussi and Davey [8] and the weighted Internet use index developed in this study were used to test the relevant hypotheses. To standardize symbols we hereby use UIu to denote the un-weighted use index devised by Moussi and Davey [8]. 4. Methodology To test the hypotheses, a questionnaire survey was conducted. The questionnaire devised by Moussi and Davey [8] was used in this study.1 The English questionnaire was first translated into Chinese. Then backtranslation was carried out by a second person to ensure the validity of the Chinese version of the questionnaire. The questionnaire consists of four questions. The first question asks respondents to indicate firm size, which is broken into five categories on an ordinal scale of 1–5: 1 (employing 1–5 people), 2 (6–20 employees), 3 (21–50 employees), 4 (51–200 employees), and 5 (over 200 employees). The second question asks respondents whether or not their firms have an Internet connection. The third question is only reserved for respondents whose firms have Internet access. It is divided into two parts. Part one is intended to collect respondents’ level of Internet use, which includes the five features discussed above, i.e., e-mail, homepage, obtaining research information, gathering customer information, and buying and selling online, of which a respondent can either check one or a combination, depending on his or her firm’s actual Internet use. These five features are each treated as dichotomous variables, the answer to which is either ‘yes (1)’ or ‘no (0)’. Part two of the third question and the fourth question collect respondents’ perceptions of the benefits of, and impediments to, e-commerce uptake, the data of which will be analyzed in a second paper. 1 Since this paper is part one of a two-part research project, the questionnaire also contains questions for the second part.
456
X. Zhang, C. Moussi / Electronic Commerce Research and Applications 6 (2007) 453–461
The translated questionnaire was then administered to a convenience sample of students at an executive MBA class and an executive training class at the International College of Business, Qingdao University, Qingdao, China. Students were asked to participate in the survey, and those who agreed were further instructed on how to complete the questionnaire. Of the over 200 students in the two classes, 141 agreed to participate, from whom 83 usable responses were colleted. The collected data were then entered into a Microsoft Excel file and analyzed with SPSS. Due to the difficulties in collecting research data in China, convenience samples are sometimes used in China-related research, especially in exploratory or descriptive studies (e.g., [19–21]). Therefore, a convenience sample was used for this preliminary study. The respondents were not proportionately sampled, but rather were developed via the first author’s personal contacts. Given its obvious limitations, however, such a sampling method generally results in high response rates, as is the case with this study. 5. Results and discussion 5.1. Characteristics of the sample In terms of demographic information, only firm size was collected. However, the general characteristics of these respondents, as determined by the first author’s conversations with the personnel in charge of the executive MBA and executive training programs at the International College of Business, Qingdao University, are as follows. They are in management positions in their respective businesses, with the majority being middle-level managers. The businesses represented by these respondents are based in the City of Qingdao and surrounding areas, virtually all sectors
and firm sizes are represented, with manufacturing firms making up the majority, which is consistent with the general economic structure of non-agricultural sectors in China. However, very small firms generally do not send their managers to such programs. As shown in Table 1, in terms of firm size, the sample is skewed toward large firms, with those employing over 200 people constituting 68% of the sample. As far as Internet use is concerned, 85% of the sample has an Internet connection. This is a rather high proportion as compared to 74% for Victorian manufacturing firms [8], 75% for small New Zealand firms [11], and 48% for small Dutch firms [22]. In addition, if the Firm Size 2 group is excluded (see discussion below), it can be seen that Internet use does appear to increase with firm size, which agrees with the findings by Moussi [23] (Table 1a), although such a relationship needs to be statistically tested for its significance. The high proportion of Internet using firms in this sample probably has to do with the sample firms’ geographic location. The City of Qingdao is located on the Eastern Coast of China, which is the most developed region in the country. Most Chinese Internet users, domain names, and Websites are located in this region [6], hence the high proportion of Internet using firms in the sample. The Firm Size 2 group (those with 6–20 employees) was excluded in the subsequent statistical testing because it has only three cases, all of which have an Internet connection. These three cases can be considered ‘outliers’ in that such a small number in this sample does not necessarily represent the actual situation in the population at large. 5.2. Descriptive statistics As shown in Table 2, overall e-mail is most used feature of the Internet, with 70% of the respondents in the sub-
Table 1 Characteristics of the Chinese sample Firm size
Yes
No
Frequency
Percentage
1 2 3 4 5
Internet connection 0 3 5 12 51
0.00 100.00 71.43 75.00 89.47
Subtotal
71
85.54
Subtotal
Frequency
Percentage
Frequency
Percentage
0 0 2 4 6
0.00 0.00 28.57 25.00 10.53
0 3 7 16 57
0.00 3.61 8.43 19.28 68.67
12
14.46
83
100.00
Table 1a Internet use by Victorian manufacturing firms: broken down by firm size Firm size Internet connection
Yes No
Source of data: adapted from [23], p. 105.
1
2
3
4
5
Subtotal
29% 71%
55% 45%
76% 24%
89% 11%
92% 8%
74% 26%
X. Zhang, C. Moussi / Electronic Commerce Research and Applications 6 (2007) 453–461
sample reporting its use. Buying and selling online also has a strong showing, being used by 60% of the respondents, which is somewhat un-expected, since it represents the most sophisticated level of Internet use, as discussed above. Subsequent checks by the first author with some respondents reveal that, in addition to conducting transactions via a firm’s own Website, they had taken this feature to include both e-marketplaces, where transactions are initiated but not necessarily completed online and which is popular among Chinese firms, and suppliers using buyers’ online e-commerce platforms to consummate transactions. The other three features, homepage, obtaining research information, and gathering customer information, have similar lower levels of use, with about half of the respondents reporting their respective use. Similar patterns exist for Victorian manufacturing firms. As shown in Table 2a, e-mail is also the most popular feature, with 93% of the respondents reporting its use, as compared with a lower 70% for Chinese firms. The level of use of homepage, obtaining research information, and gathering customer information is close to that of Chinese firms, with the first two features at roughly 10% points higher than Chinese firms. What is interesting is that a much lower percentage of Victorian manufacturing firms use the buying and selling online feature when compared to Chinese firms. Again this can be possibly explained by Chinese respondents’ inclusion of e-marketplaces and buyers’ e-commerce platforms in the use of this feature. It should also be pointed out that, small New Zealand and Small Dutch firms also exhibit similar usage patterns, with e-mail or communication within or outside the company being the most used (Tables 2b and 2c). Because of the differences of data collection, a direct comparison between the Chinese and Dutch or New Zealand results is not possible. How-
457
ever, as Tables 2b and 2c show, the top uses of the Internet are similar among the four countries. On the individual-feature level, there seems to be a curvilinear relationship between firm size and the use of each feature except homepage in the Firm Size 3–5 range of the Chinese Internet-using sub-sample, with the percentage of users first going up from Firm Size 3–4, and then going down from Firm Size 4–5 for e-mail, obtaining research information, and gathering customer information, while buying and selling online exhibits an opposite relationship. With the exception of e-mail, however, such a seeming curvilinear relationship does not exist with Victorian manufacturing firms (Table 2a). For these firms, a linear relationship seems to exist between firm size and feature use as far as obtaining research information, gathering customer information, and buying and selling online are concerned, with the first two features exhibiting a seeming direct relationship, and the third a seeming inverse relationship. On the overall-use level, the Chinese Internet-using sub-sample reports an average use of three features, as indicated by the median un-weighted use index (UIu) in Table 3. (Since both indices are ordinal in nature, the mean statistic is not appropriate in describing central tendency. Nonetheless, it is included in Table 3 for comparison.) However, as the median of the weighted use index (UIw) shows, the overall use is skewed toward the less sophisticated features, which is corroborated by e-mail being reported by the highest proportion of respondents among the five features (Table 2). In this aspect, the overall level of Internet use by Chinese firms is similar to that by Victorian manufacturing firms, whose average level of use is also 3 (Table 3a), and skewed toward less sophisticated use (Table 2a).
Table 2 Descriptive statistics of the Chinese Internet using sub-sample: use of the five features Firm size
N
E-mail
Homepage
Frequency
Percentage
Frequency
2 3 4 5
3 5 12 51
3 3 10 35
100.00 60.00 83.33 68.63
0 2 6 30
Total
71
50
70.42
38
Obtaining research info
Gathering customer info.
Buying and selling online
Percentage
Frequency
Percentage
Frequency
Percentage
Frequency
Percentage
0.00 40.00 50.00 58.82
1 1 8 29
33.33 20.00 66.67 56.86
1 2 8 25
33.33 40.00 66.67 49.02
0 3 7 33
0.00 60.00 58.33 64.71
53.52
39
54.93
36
50.70
43
60.56
Table 2a Use of the five Internet features by the Internet using sub-sample of Victorian manufacturing firms Firm size
E-mail (%)
Homepage (%)
Obtaining research info. (%)
Gathering customer info. (%)
Buying and selling online (%)
1 2 3 4 5 All
48 96 100 92 86 93
48 42 70 56 68 59
48 55 59 69 73 64
0 45 42 47 68 49
48 27 34 6 9 19
Source of data: adapted from [23], p. 112.
458
X. Zhang, C. Moussi / Electronic Commerce Research and Applications 6 (2007) 453–461 Table 3a Overall level of Internet use (UIu) by Victorian manufacturing firms
Table 2b Top Internet use by New Zealand firms Use
Percentage
Firm size
N
Mean
Median
Range
Communication within or outside the company Getting information from suppliers Providing customer information R&D and sharing of software and information To be seen at the forefront of technology Market and product research Sending orders to suppliers Receiving orders from customers
90 66 52 52 47 41 35 29
1 2 3 4 5 All
2 22 29 34 22 109
2.00 2.65 3.03 2.67 3.04 2.84
1.00 3.00 3.00 3.00 3.00 3.00
1.00–3.00 1.00–5.00 1.00–5.00 1.00–4.00 1.00–4.00 1.00–5.00
Source of data: adapted from [23], p. 109.
Source of data: adapted from [11]. Table 4 Correlation coefficients between firm size and Internet use for the Chinese sample
Table 2c Ways of using the Internet by small Dutch firms Use
Percentage
External communication (e-mail) Searching for Web page addresses Randomly looking for information Obtaining information from suppliers Offering information to consumers Contact with governmental agencies Internal communication R&D and sharing of information and software To be seen at the forefront of modern technology Sending purchase orders to suppliers Product and market research Receiving orders from customers Voice/video conferencing Placing job vacancies
91 79 76 57 32 28 25 19 18 18 18 12 1 0
Source of data: adapted from [22], p. 567.
In addition, there also seems to be a nonlinear relationship between firm size and overall level of Internet use. For UIu, the median increases from Firm Size 2–4, then levels off. For UIw, the median also increases from Firm Size 2–4, but then drops from Firm Size 4–5. The medians peaking at Firm Size 4 can be explained by the fact that this group uses at least two features, while the minimum number of features used is one for the other three groups. For Victorian manufacturing firms, however, such a seeming curvilinear relationship does not exist (Table 3a). To determine the relationship between UIu and UIw, Kendall’s tau_b was computed using SPSS, yielding a correlation coefficient of 0.862, which is significant at the 0.01 level. This indicates that the two indices are highly correlated. Table 3 Descriptive statistics of the Chinese Internet using sub-sample: UIu and UIw Firm size
N
2 3 4 5 Total
UIu
UIw
Mean
Median
Range
Mean
Median
Range
3 5 12 51
1.67 2.20 3.25 2.98
1.00 2.00 3.00 3.00
1.00–3.00 1.00–5.00 2.00–5.00 1.00–5.00
1.11 2.20 3.14 2.92
0.33 1.67 3.00 2.67
0.33–2.67 0.67–5.00 1.00–5.00 0.33–5.00
71
2.92
3.00
1.00–5.00
2.83
2.67
0.33–5.00
Kendall’s tau_b Spearman’s q *
N
Correlation coefficient
Sig.
80 80
0.193* 0.199*
0.039 0.039
Significant at the 0.05 level.
5.3. Hypothesis testing H1: For Chinese businesses, Internet use is directly related to firm size. To test this hypothesis, Internet use is treated as a dichotomous variable, its value being either ‘yes (coded as ‘1’)’ or ‘no (coded as ‘0’)’. Bivariate correlation analysis was run on the data excluding the three cases in the Firm Size 2 category. As Table 4 shows, for firm sizes 3–5 there is a statistically significant direct relationship between firm size and Internet use. Thus Hypothesis 1 is supported. The result corroborates the non-statistically tested direct relationship between firm size and Internet use for Victorian manufacturing firms [8,23], while at the same time contradicting the non-statistically tested inverse relationship for small New Zealand firms [11].2 H2: For Chinese businesses, the overall level of Internet use is directly associated with firm size. In order to test this hypothesis, separate bivariate correlation analysis was run between firm size and the unweighted use index, between firm size and the weighted use index, on the 68 Internet-using cases excluding those in the Firm Size 2 category. As Table 5 shows, neither index is significantly correlated with firm size. Thus Hypothesis 2 is rejected. Since the descriptive statistics in Table 3 suggest a possible nonlinear relationship between firm size and the use indices, bivariate correlation was run separately on the Firm Size 3–4 and the Firm Size 4–5 sub-samples. Results 2 Statistical tests were not performed in studies of either small New Zealand firms [11], or Victorian manufacturing firms [8,23]. The study of small Dutch firms did not investigate the relationship between firm size and Internet use, even though statistical tests were performed [22]. Therefore, it is not possible to compare statistical test results with any of the three studies.
X. Zhang, C. Moussi / Electronic Commerce Research and Applications 6 (2007) 453–461
459
Table 5 Correlation coefficients between firm size and use indices for the Chinese Internet using sub-sample
Kendall’s tau_b Spearman’s q
N
UIu
UIw
Correlation coefficient
Sig.
Correlation coefficient
Sig.
68 68
0.037 0.040
0.365 0.373
0.042 0.049
0.341 0.344
tion was run separately on the Firm Size 3–4 and the Firm Size 4–5 sub-samples. Results from the two sub-samples are shown in Table 8. It can be seen that there is only a statistically significant direct relationship between firm size and obtaining research information in the Firm Size 3–4 sub-sample.
from the two sub-samples are shown in Table 6. It can be seen that this hypothesis is supported only in the Firm Size 3–4 group with UIu significant at the 0.05 level and UIw significant at the 0.10 level. H3: For Chinese businesses, the level of use of each Internet feature, with the exception of buying and selling online, is directly associated with firm size. H4: For Chinese businesses, the level of use of buying and selling online is inversely related to firm size.
5.4. Further discussions It should be noted that, just as is the case with the general economic and social development in China, Internet use and e-commerce uptake by businesses are a rather dynamic process in a developing country like China. This study, of course, represents only a ‘snapshot’ of such an ongoing process. Therefore, one should bear this dynamism in mind when interpreting and using the results of this study. As shown in Table 9, Internet use has experienced explosive growth in China, as gauged by several key indicators of Internet development. It is argued that in developing countries governments play important roles in Internet development [24,25]. The rapid development of the Internet in China to a large extent can also be attributed to favorable government policies aimed at promoting Internet use by both individuals and businesses. As early as 1993, the Chinese government launched the Golden Project, with its aim being, among others, to build a national
Bivariate correlation analysis was run between firm size and each feature on the 68 Internet-using cases excluding those in the Firm Size 2 category. As Table 7 shows, there is no statistically significant correlation between firm size and each of the five Internet features. Therefore, both hypotheses are rejected. It is interesting to note that e-mail and gathering customer information are negatively correlated with firm size, even though such a relationship is not statistically significant. The statistical test results for Hypotheses 2–4 contradict the nonstatistically tested findings of a possible linear relationship between firm size and either individual features, or overall Internet use for Victorian manufacturing firms [8,23]. In order to test for the possible nonlinear relationship between firm size and individual features, bivariate correla-
Table 6 Correlation coefficients between firm size and use indices for the Chinese Internet using sub-sample: broken down by firm size Firm size 3–4 N
UIu UIw * **
17 17
Firm size 4–5
Kendall’s tau_b
Spearman’s q
N
Correlation coefficient
Sig.
Correlation coefficient
Sig.
0.392* 0.282**
0.041 0.093
0.434* 0.331**
0.041 0.097
63 63
Kendall’s tau_b
Spearman’s q
Correlation coefficient
Sig.
Correlation coefficient
Sig.
0.077 0.047
0.250 0.333
0.086 0.055
0.253 0.335
Significant at the 0.05 level. Significant at the 0.10 level.
Table 7 Correlation coefficients between firm size and individual feature use for the Chinese Internet using sub-sample N
E-mail Homepage Obtaining research info. Gathering customer info. Buying and selling online
68 68 68 68 68
Kendall’s tau_b
Spearman’s q
Correlation coefficient
Sig.
Correlation coefficient
Sig.
0.057 0.105 0.060 0.067 0.050
.315 .189 .306 .288 .337
0.059 0.108 0.062 0.068 0.051
0.317 0.191 0.308 0.290 0.339
460
X. Zhang, C. Moussi / Electronic Commerce Research and Applications 6 (2007) 453–461
Table 8 Correlation coefficients between firm size and individual feature use for the Chinese Internet using sub-sample: broken down by firm size Firm size 3–4 N
Kendall’s tau_b Correlation coefficient
E-mail Homepage Obtaining research info. Gathering customer info. Buying and selling online *
Firm size 4–5 Spearman’s q Sig.
Correlation coefficient
N Sig.
Kendall’s tau_b
Spearman’s q
Correlation coefficient
Sig.
Correlation coefficient
Sig.
17 17 17
0.251 0.091 0.426*
0.158 0.358 0.044
0.251 0.091 0.426*
0.166 0.364 0.044
63 63 63
0.128 0.070 0.078
0.157 0.291 0.269
0.128 0.070 0.078
0.159 0.293 0.271
17
0.247
0.162
0.247
0.170
63
0.139
0.137
0.139
0.139
17
0.015
0.475
0.015
0.477
63
0.052
0.341
0.052
0.343
Significant at the 0.05 level.
Table 9 Internet development in China, 1997–2006
October 1997 June 2006
Number of Internet users
Number of domain names registered
Number of .com & .com.cn domain names registered
Number of Websites
International connections bandwidth (Mbp)
620,000
4066
2131
1500
25
123,000,000
2,950,000
1,435,768
788,400
214,175
Source of data: adapted from [6].
information highway and to promote the development of information technology in China [26]. With infrastructure development rapidly moving forward, the Chinese government also promulgated the Key Informatization Program of the 10th Five-Year Plan on National Economy and Social Development, which was intended to further promote the emergence of an information society. To promote the adoption of e-commerce, the Law of Electronic Signature was passed by the Standing Committee of the 10th National People’s Congress [27]. Therefore, against this backdrop of dynamic growth of Internet diffusion in China, the pattern of using the Internet to conduct e-commerce by Chinese businesses will inevitably change. This being said, however, this study does provide some insights into the current state of Internet use and e-commerce uptake by Chinese firms, which can be used as a starting point for further studies on this topic, and as a point of comparison for future studies as well. 6. Conclusion First of all, given the fact that China is a developing country, there are a surprisingly high percentage of Internet users (85%), as previously discussed, among Chinese firms as compared to results from developed countries. In addition to the sample’s locational factor already mentioned, the time lag between this study and previous ones may also explain the difference in that, as time passes, the Internet’s role in business is gradually shifting from one of gaining an additional advantage over competitors to one of necessity, i.e., firms need to become Internet users
just to avoid being overtaken by others, rather than to get ahead of others, as may have been the case in the early days of e-commerce. Furthermore, the first author’s conversations with several respondents afterwards reveal that many Chinese firms use the Internet to access e-marketplaces such as GlobalSources.com, as mentioned previously in this paper, to prospect for international customers. With her WTO accession in 2001, China is rapidly emerging as a major exporter in the world economy, and as a result it is not surprising that more and more Chinese firms are using the Internet to tap the international market. Second, firm size plays a role only in a firm’s decision regarding whether or not to become an Internet user; once a firm decides to have Internet access, its size does not influence its overall level of Internet use, or the choice of use of individual Internet features. This suggests that the threshold of Internet use is in fact the decision to get connected to the Internet, which represents a major commitment of financial and human resources; once this threshold is crossed, however, businesses may choose to use a combination of Internet features as they see fit. As the fourth question in the questionnaire, the results of which will be analyzed in a second paper, reveals, the major impediments of Internet adoption for Chinese firms are cost and unfamiliarity with the Internet and its potential benefits to business. The policy implications of this is that, to get more Chinese businesses online, governments should provide educational programs to familiarize them with how to use the Internet to conduct business and the potential benefits thereof. In addition, both governments
X. Zhang, C. Moussi / Electronic Commerce Research and Applications 6 (2007) 453–461
and Internet service providers should attempt to find ways to offer low-cost Internet access to small firms to help them get over the threshold. 7. Limitations and future research A major limitation is the use of a convenience sample in this preliminary study. As such the results of this study must be regarded accordingly, even though convenience sampling is appropriate for a preliminary study. One should especially treat the results of this study with care in that the first two firm size groups were not represented in testing the hypotheses. Therefore, it is obvious that future research is needed to build on and extend the limited findings of this study. For example, a replication of this study with a more systematic, proportionate sampling method should be conducted to provide further, more rigorous empirical tests on the relationship between firm size and Internet use in the Chinese context. Acknowledgements The authors thank the two anonymous reviewers and Prof. Jae Kyu Lee, ECRA co-editor, for their insightful comments on earlier versions of this manuscript. We also acknowledge Miss Jessica Milton for her assistance in the preparation of the final manuscript. References [1] W. Foster, S.E. Goodman, The diffusion of the Internet in China.
461
[8] C. Moussi, B. Davey, Level of Internet use by Victorian manufacturing firms, Working Paper, School of Business Information Technology, RMIT University, Australia, 2000. [9] United States General Accounting Office, International Electronic commerce: definitions and policy implications,
Electronic Commerce Research and Applications 6 (2007) 453–461 www.elsevier.com/locate/ecra
Level of Internet use by Chinese businesses: A preliminary study Xibao Zhang a
a,*
, Cynthia Moussi
b
Department of International Economics and Trade, Qingdao University, 7 East Hongkong Road, Qingdao 266071, China b Department of Accounting and Information Systems, University of Melbourne, Parkville 3010, Australia Received 1 September 2005; received in revised form 13 November 2006; accepted 8 February 2007 Available online 14 February 2007
Abstract A preliminary study of the level of Internet use by Chinese firms was carried out. Data were collected by a questionnaire survey of a convenience sample of Chinese managers. The major finding is that firm size plays a role only in influencing a firm’s decision regarding whether or not to have Internet access. Within the Internet-using sub-sample, firm size is not significantly related to the overall level of Internet use, or to the use of individual Internet features. In addition, the data reveal that there are a high proportion of Chinese businesses that have Internet access. Chinese Internet using firms use an average of three features, with e-mail and buying and selling online being the two most widely used features. The results are also compared to similar studies in other countries. The limitations of this study and future research are discussed. Ó 2007 Elsevier B.V. All rights reserved. Keywords: E-commerce; Level of Internet use; Chinese businesses; Internet use index; Internet development in China
1. Introduction In line with the worldwide trend, Internet use in China has experienced rapid growth in recent years. On the research front, there have been a number of studies by international scholars on Internet diffusion in China (e.g., [1–3]). However, these studies are of a general nature, covering the general diffusion and use of the Internet in China. Press et al. [3] and Fan [4], for example, both used the Mosaic Group’s Global Internet Diffusion Framework [5], which includes six dimensions (pervasiveness, geographic dispersion, sectoral absorption, connectivity infrastructure, organizational infrastructure, and sophistication of use), in analyzing China’s Internet development. In addition, since both studies used survey data collected by China Internet Network Information Center (or CNNIC, http://www.cnnic.net.cn), their analyses undoubtedly are limited to the general situation of Internet diffusion and
use in China, with little or no discussion of Internet use by Chinese businesses. Inside China, CNNIC [6] publishes a series of semiannual reports on China’s Internet diffusion and use, which is primarily based on online surveys of Internet users. In addition, Guo and Bu [7] also studied Internet use and its influences in five major Chinese cities. However, these studies primarily focused on individual Internet users, and as a result the issue of Internet use by Chinese businesses has not been explored. Therefore, it is the purpose of this paper to study the level of Internet use by Chinese businesses. This paper is part one of a two-part project aimed at investigating Chinese firms’ Internet use and related issues. Part two of the project examines respondents’ perceptions of the benefits of and impediments to e-commerce uptake, the results of which will be published in a second paper. 2. Level of Internet use
*
Corresponding author. Tel.: +86 532 8595 3733; fax: +86 532 8595 0018. E-mail address: [email protected] (X. Zhang). 1567-4223/$ - see front matter Ó 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.elerap.2007.02.007
There have been different definitions of electronic commerce by various scholars, international organizations, and commercial entities. (For a synopsis of such
454
X. Zhang, C. Moussi / Electronic Commerce Research and Applications 6 (2007) 453–461
definitions, please see [8,9].) In this research we use the definition by Zwass [10], i.e., it is ‘‘the sharing of business information, maintaining business relationships, and conducting business transactions by means of telecommunications networks’’, including both inter-organizational relationships and transactions and intra-organizational processes that support these activities. This is a more inclusive definition in that, first, it includes both inter-organizational and intra-organizational activities, and second, it includes not only online transactions but also other activities, such as information sharing and maintaining business relationships, that will eventually lead to online and offline transactions. There are two questions concerning a business’ use of the Internet: first, does it use the Internet at all? And second, if the answer to the first question is yes, to what extent does it use the Internet? These are the two questions that this paper attempts to answer in the context of Chinese firms. Businesses can use the Internet in many different ways and to serve myriad purposes, and different scholars use different terms and concepts to refer to these uses. Abell and Lim [11], for example, found in their empirical study the top eight uses of the Internet by New Zealand firms to be communication within or outside the company, getting information from suppliers, providing customer information, R&D and sharing of software and information, to be seen at the forefront of technology, market and product research, sending orders to suppliers, and receiving orders from customers. Moussi and Davey [8], in their study of Internet use by Victorian manufacturing firms, proposed a five-feature set that measures a business’ level of Internet use to conduct e-commerce, which includes e-mail, homepage, obtaining research information, gathering customer information, and buying and selling online. On the theoretical side, Poon and Swatman [12] proposed seven business activities by small firms on the Internet: consumer prospecting and advertising, involvement in specialized groups for knowledge and intelligence exchange, research and development ideas/opportunities, efficient communications, preparation for the global marketplace, on-demand linkage with customers and suppliers, and geographic reach/general accessibility. Cronin [13] pointed out three areas of Internet use by businesses: customer relations, dealing with customers, and internal company operations. Zwass [14] identified an exhaustive list of potential uses in his 5Cs aspects–opportunities conceptual framework, which categorizes Internet uses into five domains: commerce, collaboration, communication, connection, and computation. Such diversity in both the empirical measures and conceptual constructs of Internet use by businesses reflects the overwhelming impact of the Internet in transforming business activities and processes. For the sake of empirical research, however, there needs to be a simple yet comprehensive set of Internet use measures to gauge their uptake of e-commerce. For this reason the feature-based set of
Internet use measures proposed by Moussi and Davey [8] is used in this study, because it essentially covers all aspects of electronic commerce, and is relatively easy to use in conducting a survey. Regarding the first question, factors need to be identified that influence a firm’s decision to use the Internet for electronic commerce. It has been suggested by early IS scholars that firm size is an important determinant of business IS adoption, and the level of IS adoption is directly related to firm size [16–18]. Firm size, according to Ling [15], is also one of the factors that influence business use of the Internet. In this preliminary study, therefore, only firm size is tested for its influence on Internet use because it can be easily and objectively measured. Empirically, however, there has been somewhat contradictory evidence regarding the influence of firm size on Internet use. The Internet-using sample of Abell and Lim [11] consisted of more smaller firms (those with 49 employees or fewer) than larger ones, suggesting that there may be an inverse relationship between firm size and Internet use. Results from Moussi and Davey’s study [8], however, indicate that there seems to be a direct relationship between firm size and Internet use, i.e., the larger the firm, the more likely it uses the Internet in conducting business. (It should be noted that no statistical tests were performed on the relationship between firm size and Internet use in either Moussi and Davey’s study [8] or Abell and Lim’s [11].) The discrepancy can be explained by the fact that, first, New Zealand firms are predominantly small ones, and second, in the sample of Abell and Lim [11], the majority of smaller firms are those with an IT focus (45% IT focus vs. 28% non-IT focus), while IT firms constitute only a minority in the larger firm group (8% IT focus vs. 22% non-IT focus). The direct relationship between firm size and Internet use as suggested by Moussi and Davey [8] is intuitive in that a smaller firm is more likely to lack the necessary financial resources, knowledge, or personnel to use the Internet. We expect that such a relationship should also hold true for Chinese firms. Therefore: H1: For Chinese businesses, Internet use is directly related to firm size. In order to answer the second question, both the use of each of the five features and the overall level of Internet use need to be explored. Again for the same reason as discussed above, firm size is tested as a potential determinant of such uses. In their study, Moussi and Davey [8] devised an Internet use index, which is an un-weighted sum of all the features of the Internet used by a firm: Use Index ¼
5 X
Features:
n¼1
The value of the above use index ranges from 0 (no use) to 5 (all five features used), and as a result it provides a meaningful measure for analyzing the overall level of
X. Zhang, C. Moussi / Electronic Commerce Research and Applications 6 (2007) 453–461
Internet use by businesses. In addition, their data seem to indicate that there is possibly a direct relationship between firm size and the use index. For each individual feature, their data also appear to show that, with the exception of buying and selling online, as firm size increases, so does the use of that feature. Again the reasoning for both observations is similar to that regarding the first question, namely, generally speaking, smaller firms, even if they use the Internet, do not have the necessary financial resources, expertise, or personnel to fully take advantage of the potentials of the Internet. Such relationships are also expected to hold true for Chinese firms: H2: For Chinese businesses, the overall level of Internet use is directly associated with firm size. H3: For Chinese businesses, the level of use of each Internet feature, with the exception of buying and selling online, is directly associated with firm size. Regarding buying and selling online, Moussi and Davey [8] also observed that there are a higher percentage of firms using this feature among those with 50 employees or fewer than among those with 51 employees or more. This is an interesting observation, and the only plausible explanation may be that buying and selling online represents the highest and most sophisticated use of the Internet, and requires a total overhaul of the internal processes to support such an endeavor. Such an overhaul is possibly easier for smaller businesses simply because of their smaller size. Therefore: H4: For Chinese businesses, the level of use of buying and selling online is inversely related to firm size. 3. Redesigning the use index Clearly each of the five features of the Internet represents a different level of sophistication of Internet use. They can be arranged in ascending order of sophistication as: e-mail, homepage, obtaining research information, gathering customer information, and buying and selling online. E-mail is the least sophisticated feature in that all it requires is an Internet connection and a computer. Homepage comes next in this respect in that the most basic homepage – a static homepage – does not require the firm to update its information content frequently, and it can even be hosted by Web-hosting services. Obtaining research information represents a higher sophistication level than homepage in that the firm uses the Internet to obtain information regarding products, services, customers, prices, market trends, strategic partners, etc. Therefore it requires the firm to use the Internet more actively and more extensively. A firm can also use the Internet to acquire and store customer profiles, their needs, their buying pattern, etc., and automate the retrieval and processing of such data. This is where gathering customer information comes in. Buy doing so the firm learns customers’ needs and deliver products and services to best satisfy such needs, thereby building and maintaining an ongoing relationship. It is a
455
step higher on the sophistication scale than obtaining research information, since this requires the organization to at least overhaul its customer support and marketing processes. Finally, buying and selling online represents the most sophisticated use of the Internet in that this requires the firm to set up and operate an interactive Website to handle online transactions, which is rather costly, and re-engineer virtually all of its business processes. Therefore, a different weight is assigned to each of the five features as follows: e-mail 1/3, homepage 2/3, obtaining research information 3/3, gathering customer information 4/3, and buying and selling online 5/3. The weighted Internet use index can be expressed as UIw ¼
5 X
W iF i;
i¼1
where UIw is the weighted Internet use index, Wi the ith feature’s weight, and Fi the ith feature’s value. Both the un-weighted Internet use index developed by Moussi and Davey [8] and the weighted Internet use index developed in this study were used to test the relevant hypotheses. To standardize symbols we hereby use UIu to denote the un-weighted use index devised by Moussi and Davey [8]. 4. Methodology To test the hypotheses, a questionnaire survey was conducted. The questionnaire devised by Moussi and Davey [8] was used in this study.1 The English questionnaire was first translated into Chinese. Then backtranslation was carried out by a second person to ensure the validity of the Chinese version of the questionnaire. The questionnaire consists of four questions. The first question asks respondents to indicate firm size, which is broken into five categories on an ordinal scale of 1–5: 1 (employing 1–5 people), 2 (6–20 employees), 3 (21–50 employees), 4 (51–200 employees), and 5 (over 200 employees). The second question asks respondents whether or not their firms have an Internet connection. The third question is only reserved for respondents whose firms have Internet access. It is divided into two parts. Part one is intended to collect respondents’ level of Internet use, which includes the five features discussed above, i.e., e-mail, homepage, obtaining research information, gathering customer information, and buying and selling online, of which a respondent can either check one or a combination, depending on his or her firm’s actual Internet use. These five features are each treated as dichotomous variables, the answer to which is either ‘yes (1)’ or ‘no (0)’. Part two of the third question and the fourth question collect respondents’ perceptions of the benefits of, and impediments to, e-commerce uptake, the data of which will be analyzed in a second paper. 1 Since this paper is part one of a two-part research project, the questionnaire also contains questions for the second part.
456
X. Zhang, C. Moussi / Electronic Commerce Research and Applications 6 (2007) 453–461
The translated questionnaire was then administered to a convenience sample of students at an executive MBA class and an executive training class at the International College of Business, Qingdao University, Qingdao, China. Students were asked to participate in the survey, and those who agreed were further instructed on how to complete the questionnaire. Of the over 200 students in the two classes, 141 agreed to participate, from whom 83 usable responses were colleted. The collected data were then entered into a Microsoft Excel file and analyzed with SPSS. Due to the difficulties in collecting research data in China, convenience samples are sometimes used in China-related research, especially in exploratory or descriptive studies (e.g., [19–21]). Therefore, a convenience sample was used for this preliminary study. The respondents were not proportionately sampled, but rather were developed via the first author’s personal contacts. Given its obvious limitations, however, such a sampling method generally results in high response rates, as is the case with this study. 5. Results and discussion 5.1. Characteristics of the sample In terms of demographic information, only firm size was collected. However, the general characteristics of these respondents, as determined by the first author’s conversations with the personnel in charge of the executive MBA and executive training programs at the International College of Business, Qingdao University, are as follows. They are in management positions in their respective businesses, with the majority being middle-level managers. The businesses represented by these respondents are based in the City of Qingdao and surrounding areas, virtually all sectors
and firm sizes are represented, with manufacturing firms making up the majority, which is consistent with the general economic structure of non-agricultural sectors in China. However, very small firms generally do not send their managers to such programs. As shown in Table 1, in terms of firm size, the sample is skewed toward large firms, with those employing over 200 people constituting 68% of the sample. As far as Internet use is concerned, 85% of the sample has an Internet connection. This is a rather high proportion as compared to 74% for Victorian manufacturing firms [8], 75% for small New Zealand firms [11], and 48% for small Dutch firms [22]. In addition, if the Firm Size 2 group is excluded (see discussion below), it can be seen that Internet use does appear to increase with firm size, which agrees with the findings by Moussi [23] (Table 1a), although such a relationship needs to be statistically tested for its significance. The high proportion of Internet using firms in this sample probably has to do with the sample firms’ geographic location. The City of Qingdao is located on the Eastern Coast of China, which is the most developed region in the country. Most Chinese Internet users, domain names, and Websites are located in this region [6], hence the high proportion of Internet using firms in the sample. The Firm Size 2 group (those with 6–20 employees) was excluded in the subsequent statistical testing because it has only three cases, all of which have an Internet connection. These three cases can be considered ‘outliers’ in that such a small number in this sample does not necessarily represent the actual situation in the population at large. 5.2. Descriptive statistics As shown in Table 2, overall e-mail is most used feature of the Internet, with 70% of the respondents in the sub-
Table 1 Characteristics of the Chinese sample Firm size
Yes
No
Frequency
Percentage
1 2 3 4 5
Internet connection 0 3 5 12 51
0.00 100.00 71.43 75.00 89.47
Subtotal
71
85.54
Subtotal
Frequency
Percentage
Frequency
Percentage
0 0 2 4 6
0.00 0.00 28.57 25.00 10.53
0 3 7 16 57
0.00 3.61 8.43 19.28 68.67
12
14.46
83
100.00
Table 1a Internet use by Victorian manufacturing firms: broken down by firm size Firm size Internet connection
Yes No
Source of data: adapted from [23], p. 105.
1
2
3
4
5
Subtotal
29% 71%
55% 45%
76% 24%
89% 11%
92% 8%
74% 26%
X. Zhang, C. Moussi / Electronic Commerce Research and Applications 6 (2007) 453–461
sample reporting its use. Buying and selling online also has a strong showing, being used by 60% of the respondents, which is somewhat un-expected, since it represents the most sophisticated level of Internet use, as discussed above. Subsequent checks by the first author with some respondents reveal that, in addition to conducting transactions via a firm’s own Website, they had taken this feature to include both e-marketplaces, where transactions are initiated but not necessarily completed online and which is popular among Chinese firms, and suppliers using buyers’ online e-commerce platforms to consummate transactions. The other three features, homepage, obtaining research information, and gathering customer information, have similar lower levels of use, with about half of the respondents reporting their respective use. Similar patterns exist for Victorian manufacturing firms. As shown in Table 2a, e-mail is also the most popular feature, with 93% of the respondents reporting its use, as compared with a lower 70% for Chinese firms. The level of use of homepage, obtaining research information, and gathering customer information is close to that of Chinese firms, with the first two features at roughly 10% points higher than Chinese firms. What is interesting is that a much lower percentage of Victorian manufacturing firms use the buying and selling online feature when compared to Chinese firms. Again this can be possibly explained by Chinese respondents’ inclusion of e-marketplaces and buyers’ e-commerce platforms in the use of this feature. It should also be pointed out that, small New Zealand and Small Dutch firms also exhibit similar usage patterns, with e-mail or communication within or outside the company being the most used (Tables 2b and 2c). Because of the differences of data collection, a direct comparison between the Chinese and Dutch or New Zealand results is not possible. How-
457
ever, as Tables 2b and 2c show, the top uses of the Internet are similar among the four countries. On the individual-feature level, there seems to be a curvilinear relationship between firm size and the use of each feature except homepage in the Firm Size 3–5 range of the Chinese Internet-using sub-sample, with the percentage of users first going up from Firm Size 3–4, and then going down from Firm Size 4–5 for e-mail, obtaining research information, and gathering customer information, while buying and selling online exhibits an opposite relationship. With the exception of e-mail, however, such a seeming curvilinear relationship does not exist with Victorian manufacturing firms (Table 2a). For these firms, a linear relationship seems to exist between firm size and feature use as far as obtaining research information, gathering customer information, and buying and selling online are concerned, with the first two features exhibiting a seeming direct relationship, and the third a seeming inverse relationship. On the overall-use level, the Chinese Internet-using sub-sample reports an average use of three features, as indicated by the median un-weighted use index (UIu) in Table 3. (Since both indices are ordinal in nature, the mean statistic is not appropriate in describing central tendency. Nonetheless, it is included in Table 3 for comparison.) However, as the median of the weighted use index (UIw) shows, the overall use is skewed toward the less sophisticated features, which is corroborated by e-mail being reported by the highest proportion of respondents among the five features (Table 2). In this aspect, the overall level of Internet use by Chinese firms is similar to that by Victorian manufacturing firms, whose average level of use is also 3 (Table 3a), and skewed toward less sophisticated use (Table 2a).
Table 2 Descriptive statistics of the Chinese Internet using sub-sample: use of the five features Firm size
N
Homepage
Frequency
Percentage
Frequency
2 3 4 5
3 5 12 51
3 3 10 35
100.00 60.00 83.33 68.63
0 2 6 30
Total
71
50
70.42
38
Obtaining research info
Gathering customer info.
Buying and selling online
Percentage
Frequency
Percentage
Frequency
Percentage
Frequency
Percentage
0.00 40.00 50.00 58.82
1 1 8 29
33.33 20.00 66.67 56.86
1 2 8 25
33.33 40.00 66.67 49.02
0 3 7 33
0.00 60.00 58.33 64.71
53.52
39
54.93
36
50.70
43
60.56
Table 2a Use of the five Internet features by the Internet using sub-sample of Victorian manufacturing firms Firm size
E-mail (%)
Homepage (%)
Obtaining research info. (%)
Gathering customer info. (%)
Buying and selling online (%)
1 2 3 4 5 All
48 96 100 92 86 93
48 42 70 56 68 59
48 55 59 69 73 64
0 45 42 47 68 49
48 27 34 6 9 19
Source of data: adapted from [23], p. 112.
458
X. Zhang, C. Moussi / Electronic Commerce Research and Applications 6 (2007) 453–461 Table 3a Overall level of Internet use (UIu) by Victorian manufacturing firms
Table 2b Top Internet use by New Zealand firms Use
Percentage
Firm size
N
Mean
Median
Range
Communication within or outside the company Getting information from suppliers Providing customer information R&D and sharing of software and information To be seen at the forefront of technology Market and product research Sending orders to suppliers Receiving orders from customers
90 66 52 52 47 41 35 29
1 2 3 4 5 All
2 22 29 34 22 109
2.00 2.65 3.03 2.67 3.04 2.84
1.00 3.00 3.00 3.00 3.00 3.00
1.00–3.00 1.00–5.00 1.00–5.00 1.00–4.00 1.00–4.00 1.00–5.00
Source of data: adapted from [23], p. 109.
Source of data: adapted from [11]. Table 4 Correlation coefficients between firm size and Internet use for the Chinese sample
Table 2c Ways of using the Internet by small Dutch firms Use
Percentage
External communication (e-mail) Searching for Web page addresses Randomly looking for information Obtaining information from suppliers Offering information to consumers Contact with governmental agencies Internal communication R&D and sharing of information and software To be seen at the forefront of modern technology Sending purchase orders to suppliers Product and market research Receiving orders from customers Voice/video conferencing Placing job vacancies
91 79 76 57 32 28 25 19 18 18 18 12 1 0
Source of data: adapted from [22], p. 567.
In addition, there also seems to be a nonlinear relationship between firm size and overall level of Internet use. For UIu, the median increases from Firm Size 2–4, then levels off. For UIw, the median also increases from Firm Size 2–4, but then drops from Firm Size 4–5. The medians peaking at Firm Size 4 can be explained by the fact that this group uses at least two features, while the minimum number of features used is one for the other three groups. For Victorian manufacturing firms, however, such a seeming curvilinear relationship does not exist (Table 3a). To determine the relationship between UIu and UIw, Kendall’s tau_b was computed using SPSS, yielding a correlation coefficient of 0.862, which is significant at the 0.01 level. This indicates that the two indices are highly correlated. Table 3 Descriptive statistics of the Chinese Internet using sub-sample: UIu and UIw Firm size
N
2 3 4 5 Total
UIu
UIw
Mean
Median
Range
Mean
Median
Range
3 5 12 51
1.67 2.20 3.25 2.98
1.00 2.00 3.00 3.00
1.00–3.00 1.00–5.00 2.00–5.00 1.00–5.00
1.11 2.20 3.14 2.92
0.33 1.67 3.00 2.67
0.33–2.67 0.67–5.00 1.00–5.00 0.33–5.00
71
2.92
3.00
1.00–5.00
2.83
2.67
0.33–5.00
Kendall’s tau_b Spearman’s q *
N
Correlation coefficient
Sig.
80 80
0.193* 0.199*
0.039 0.039
Significant at the 0.05 level.
5.3. Hypothesis testing H1: For Chinese businesses, Internet use is directly related to firm size. To test this hypothesis, Internet use is treated as a dichotomous variable, its value being either ‘yes (coded as ‘1’)’ or ‘no (coded as ‘0’)’. Bivariate correlation analysis was run on the data excluding the three cases in the Firm Size 2 category. As Table 4 shows, for firm sizes 3–5 there is a statistically significant direct relationship between firm size and Internet use. Thus Hypothesis 1 is supported. The result corroborates the non-statistically tested direct relationship between firm size and Internet use for Victorian manufacturing firms [8,23], while at the same time contradicting the non-statistically tested inverse relationship for small New Zealand firms [11].2 H2: For Chinese businesses, the overall level of Internet use is directly associated with firm size. In order to test this hypothesis, separate bivariate correlation analysis was run between firm size and the unweighted use index, between firm size and the weighted use index, on the 68 Internet-using cases excluding those in the Firm Size 2 category. As Table 5 shows, neither index is significantly correlated with firm size. Thus Hypothesis 2 is rejected. Since the descriptive statistics in Table 3 suggest a possible nonlinear relationship between firm size and the use indices, bivariate correlation was run separately on the Firm Size 3–4 and the Firm Size 4–5 sub-samples. Results 2 Statistical tests were not performed in studies of either small New Zealand firms [11], or Victorian manufacturing firms [8,23]. The study of small Dutch firms did not investigate the relationship between firm size and Internet use, even though statistical tests were performed [22]. Therefore, it is not possible to compare statistical test results with any of the three studies.
X. Zhang, C. Moussi / Electronic Commerce Research and Applications 6 (2007) 453–461
459
Table 5 Correlation coefficients between firm size and use indices for the Chinese Internet using sub-sample
Kendall’s tau_b Spearman’s q
N
UIu
UIw
Correlation coefficient
Sig.
Correlation coefficient
Sig.
68 68
0.037 0.040
0.365 0.373
0.042 0.049
0.341 0.344
tion was run separately on the Firm Size 3–4 and the Firm Size 4–5 sub-samples. Results from the two sub-samples are shown in Table 8. It can be seen that there is only a statistically significant direct relationship between firm size and obtaining research information in the Firm Size 3–4 sub-sample.
from the two sub-samples are shown in Table 6. It can be seen that this hypothesis is supported only in the Firm Size 3–4 group with UIu significant at the 0.05 level and UIw significant at the 0.10 level. H3: For Chinese businesses, the level of use of each Internet feature, with the exception of buying and selling online, is directly associated with firm size. H4: For Chinese businesses, the level of use of buying and selling online is inversely related to firm size.
5.4. Further discussions It should be noted that, just as is the case with the general economic and social development in China, Internet use and e-commerce uptake by businesses are a rather dynamic process in a developing country like China. This study, of course, represents only a ‘snapshot’ of such an ongoing process. Therefore, one should bear this dynamism in mind when interpreting and using the results of this study. As shown in Table 9, Internet use has experienced explosive growth in China, as gauged by several key indicators of Internet development. It is argued that in developing countries governments play important roles in Internet development [24,25]. The rapid development of the Internet in China to a large extent can also be attributed to favorable government policies aimed at promoting Internet use by both individuals and businesses. As early as 1993, the Chinese government launched the Golden Project, with its aim being, among others, to build a national
Bivariate correlation analysis was run between firm size and each feature on the 68 Internet-using cases excluding those in the Firm Size 2 category. As Table 7 shows, there is no statistically significant correlation between firm size and each of the five Internet features. Therefore, both hypotheses are rejected. It is interesting to note that e-mail and gathering customer information are negatively correlated with firm size, even though such a relationship is not statistically significant. The statistical test results for Hypotheses 2–4 contradict the nonstatistically tested findings of a possible linear relationship between firm size and either individual features, or overall Internet use for Victorian manufacturing firms [8,23]. In order to test for the possible nonlinear relationship between firm size and individual features, bivariate correla-
Table 6 Correlation coefficients between firm size and use indices for the Chinese Internet using sub-sample: broken down by firm size Firm size 3–4 N
UIu UIw * **
17 17
Firm size 4–5
Kendall’s tau_b
Spearman’s q
N
Correlation coefficient
Sig.
Correlation coefficient
Sig.
0.392* 0.282**
0.041 0.093
0.434* 0.331**
0.041 0.097
63 63
Kendall’s tau_b
Spearman’s q
Correlation coefficient
Sig.
Correlation coefficient
Sig.
0.077 0.047
0.250 0.333
0.086 0.055
0.253 0.335
Significant at the 0.05 level. Significant at the 0.10 level.
Table 7 Correlation coefficients between firm size and individual feature use for the Chinese Internet using sub-sample N
E-mail Homepage Obtaining research info. Gathering customer info. Buying and selling online
68 68 68 68 68
Kendall’s tau_b
Spearman’s q
Correlation coefficient
Sig.
Correlation coefficient
Sig.
0.057 0.105 0.060 0.067 0.050
.315 .189 .306 .288 .337
0.059 0.108 0.062 0.068 0.051
0.317 0.191 0.308 0.290 0.339
460
X. Zhang, C. Moussi / Electronic Commerce Research and Applications 6 (2007) 453–461
Table 8 Correlation coefficients between firm size and individual feature use for the Chinese Internet using sub-sample: broken down by firm size Firm size 3–4 N
Kendall’s tau_b Correlation coefficient
E-mail Homepage Obtaining research info. Gathering customer info. Buying and selling online *
Firm size 4–5 Spearman’s q Sig.
Correlation coefficient
N Sig.
Kendall’s tau_b
Spearman’s q
Correlation coefficient
Sig.
Correlation coefficient
Sig.
17 17 17
0.251 0.091 0.426*
0.158 0.358 0.044
0.251 0.091 0.426*
0.166 0.364 0.044
63 63 63
0.128 0.070 0.078
0.157 0.291 0.269
0.128 0.070 0.078
0.159 0.293 0.271
17
0.247
0.162
0.247
0.170
63
0.139
0.137
0.139
0.139
17
0.015
0.475
0.015
0.477
63
0.052
0.341
0.052
0.343
Significant at the 0.05 level.
Table 9 Internet development in China, 1997–2006
October 1997 June 2006
Number of Internet users
Number of domain names registered
Number of .com & .com.cn domain names registered
Number of Websites
International connections bandwidth (Mbp)
620,000
4066
2131
1500
25
123,000,000
2,950,000
1,435,768
788,400
214,175
Source of data: adapted from [6].
information highway and to promote the development of information technology in China [26]. With infrastructure development rapidly moving forward, the Chinese government also promulgated the Key Informatization Program of the 10th Five-Year Plan on National Economy and Social Development, which was intended to further promote the emergence of an information society. To promote the adoption of e-commerce, the Law of Electronic Signature was passed by the Standing Committee of the 10th National People’s Congress [27]. Therefore, against this backdrop of dynamic growth of Internet diffusion in China, the pattern of using the Internet to conduct e-commerce by Chinese businesses will inevitably change. This being said, however, this study does provide some insights into the current state of Internet use and e-commerce uptake by Chinese firms, which can be used as a starting point for further studies on this topic, and as a point of comparison for future studies as well. 6. Conclusion First of all, given the fact that China is a developing country, there are a surprisingly high percentage of Internet users (85%), as previously discussed, among Chinese firms as compared to results from developed countries. In addition to the sample’s locational factor already mentioned, the time lag between this study and previous ones may also explain the difference in that, as time passes, the Internet’s role in business is gradually shifting from one of gaining an additional advantage over competitors to one of necessity, i.e., firms need to become Internet users
just to avoid being overtaken by others, rather than to get ahead of others, as may have been the case in the early days of e-commerce. Furthermore, the first author’s conversations with several respondents afterwards reveal that many Chinese firms use the Internet to access e-marketplaces such as GlobalSources.com, as mentioned previously in this paper, to prospect for international customers. With her WTO accession in 2001, China is rapidly emerging as a major exporter in the world economy, and as a result it is not surprising that more and more Chinese firms are using the Internet to tap the international market. Second, firm size plays a role only in a firm’s decision regarding whether or not to become an Internet user; once a firm decides to have Internet access, its size does not influence its overall level of Internet use, or the choice of use of individual Internet features. This suggests that the threshold of Internet use is in fact the decision to get connected to the Internet, which represents a major commitment of financial and human resources; once this threshold is crossed, however, businesses may choose to use a combination of Internet features as they see fit. As the fourth question in the questionnaire, the results of which will be analyzed in a second paper, reveals, the major impediments of Internet adoption for Chinese firms are cost and unfamiliarity with the Internet and its potential benefits to business. The policy implications of this is that, to get more Chinese businesses online, governments should provide educational programs to familiarize them with how to use the Internet to conduct business and the potential benefits thereof. In addition, both governments
X. Zhang, C. Moussi / Electronic Commerce Research and Applications 6 (2007) 453–461
and Internet service providers should attempt to find ways to offer low-cost Internet access to small firms to help them get over the threshold. 7. Limitations and future research A major limitation is the use of a convenience sample in this preliminary study. As such the results of this study must be regarded accordingly, even though convenience sampling is appropriate for a preliminary study. One should especially treat the results of this study with care in that the first two firm size groups were not represented in testing the hypotheses. Therefore, it is obvious that future research is needed to build on and extend the limited findings of this study. For example, a replication of this study with a more systematic, proportionate sampling method should be conducted to provide further, more rigorous empirical tests on the relationship between firm size and Internet use in the Chinese context. Acknowledgements The authors thank the two anonymous reviewers and Prof. Jae Kyu Lee, ECRA co-editor, for their insightful comments on earlier versions of this manuscript. We also acknowledge Miss Jessica Milton for her assistance in the preparation of the final manuscript. References [1] W. Foster, S.E. Goodman, The diffusion of the Internet in China.
461
[8] C. Moussi, B. Davey, Level of Internet use by Victorian manufacturing firms, Working Paper, School of Business Information Technology, RMIT University, Australia, 2000. [9] United States General Accounting Office, International Electronic commerce: definitions and policy implications,