AUC Academic Conference 'From Virtual to Reality' The University of
Queensland 1996AUC Academic Conference 'From Virtual to Reality' The University of
Queensland 1996
Back to Home Page ...... by Presenter ...... by Paper Title ...... by Keyword
Paper Title:
The World Wide Web
as a Research ToolPresenter / Author:
Dr. Peter M. Stevens, Victoria University of Wellington
Keywords: Research, Internet
Faculty area: Commerce
This paper needs to be viewed at full screen size. Please click on the 'size button' at the top right of this window in the title bar, or just drag the browsers window larger using the 'scale window' icon at the bottom right of this window.
Abstract:
The Internet, and particularly the World Wide Web (WWW), is attracting significant interest in the business community. An online survey conducted to research the use of the Internet by businesses in New Zealand is described. A survey 'engine' built from off-the-shelf software applications and running on standard personal computers is described which offers significant advantages over traditional 'paper and post' surveying techniques. Online surveying was shown to markedly decrease the cost of conducting large-scale surveys, and eliminated much of the drudgery of data collection and processing. Some of the more salient results of the survey are presented, indicating that the majority of New Zealand businesses with a WWW presence believe that their sites have been successful and have added value to their business.
Introduction
The Internet, long the preserve of the academic and government research institutions, is a phenomenon now commanding significant interest in the business community. Traffic volume on the World Wide Web (WWW) grew by 350,000% in 1994 (Sterne, 1995), and growth in Internet connections in New Zealand has been increasing by 10-15% per month consistently throughout 1995 and 1996 (Figure 1). While the WWW has generated considerable interest in the popular and information technology (IT) industry press, to date there has been a paucity of theoretical and empirical investigation in to the impact of the Internet on the marketing activities of organisations.
Figure 1
Total growth of full-time internet protocol (IP) connections and connections attributable to commercial (.co.nz) organisations in the New Zealand Domain Name Service (DNS) namespace December 1993 - June 1996. Logarithmic scale.
This paper will discuss how various Internet tools have been successfully used to research the use of the Internet by businesses in New Zealand. Growth in the use of the Internet, especially the (WWW), has been exponential and is currently increasing at 10-15% per month (Figure 1). Emphasis will be placed on simple, yet powerful techniques to remove much of the drudgery of data collection and processing by smart use of electronic mail (e-mail), hypertext mark up language (HTML) forms, AppleEvents, AppleScript and databases. Finally, some of the more interesting and surprising results of the research will be presented, examining issues of electronic commerce and marketing communications.
The Internet in New Zealand
New Zealand is a small country of 3.5m people with a reputation for the rapid uptake of new technologies. New Zealand has a highly developed infrastructure for telecommunications (90+% digital) and a largely deregulated telecommunications market. While the precursor to the Internet (ARPAnet) had its beginnings in the 1960's in the United States, the first coordinated link to the Internet (the 'NZGate') was only established in 1989 via a 9600 baud analogue cable link to Hawaii. Since that time, access to the Internet has grown dramatically; as of June 1996 at least four international links to the Internet exist and 2488 organisations are independently registered in the New Zealand portion of the Domain Name Service (DNS) namespace (Figure 1). The primary driver for the observed growth in DNS registrations is the establishment of WWW sites by commercial organisations (Figure 2).
Figure 2
Growth of web sites in the New Zealand Domain Name Service (DNS) namespace October 1995- June 1996. Overlay graph illustrates the percentage of those sites that are from the commercial branch (.co.nz) of the DNS.
A comprehensive programme studying the adoption of Internet-based technologies into New Zealand businesses is on-going in the Marketing Group of Victoria University of Wellington. To collect baseline data for this programme, the most appropriate instrument was a survey of the commercial entities with WWW sites. A cost-effective and efficient solution was sought to integrate and automate the mechanics of: (i) conducting an extensive survey, (ii) personalised communication with survey participants for reminders and 'thank you' notes, (iii) communication of survey results to participants and (iv) publication of the results to a wider audience.
Survey Methodology
The use of the Internet to conduct online surveys is not new (e.g. Pitkow & Recker, 1995; Stevens & Greenwood, in press). However, such surveys are usually constructed using UNIX-based web servers, PERL scripts and 'not-so-friendly' software tools and databases. Thus to conduct the present survey, a generic 'engine' was created using off-the-shelf software components, integrating a web server (StarNine WebSTAR), database (Claris FileMaker Pro 3.0), e-mail client (QualComm Eudora Lite 1.5) and web browser (Netscape Navigator 2.0). Integration was provided by (i) a Common Gateway Interface (ACGI) script written in AppleScript for the processing of incoming data from the web server and entering it into the database, (ii) an AppleEvent- driven AppleScript applet under the control of FileMaker for the automatic generation and dispatch of personalised e-mail messages from Eudora to survey participants, and (iii) an AppleEvent- driven link between FileMaker and Navigator for the automated look-up of appropriate web site Uniform Resource Locator (URL) addresses. The process is illustrated graphically in Fig. 3.
Figure 3
Survey 'engine'. See text for details.
Sample
From December 1993 a software 'robot' has 'walked' the New Zealand DNS namespace (ie. that area of the Internet that has connections that end with .nz) every month, recording the top level electronic addresses (domains) of all organisations connected to TuiaNet (the New Zealand Internet) or that have e-mail access to it. This technique identifies the majority of the entities (usually organisations) connected to the Internet. The software robot cross-references whether the entity has established a WWW site. Within the New Zealand DNS, organisations are grouped into different 'branches' or categories reflecting their role or affiliation. Electronic labels appropriate to these branches are designated as part of an e-mail or WWW address. Thus academic organisations have the designation .ac, Company (.co), Crown Research Institute (.cri), General (.gen), Government (.govt), Iwi (.iwi), Organisation (.org), Military (.mil), Network (.net) and School (.school). For this study, only those organisations with an overt commercial focus and registered in New Zealand were desired; requests for participation were thus only sent to those entities with WWW sites and a domain name ending in .co.nz. Note that companies operating in New Zealand but with, for example, a US registration (such as ibm.com) were not surveyed.
Survey Design
All sites were direct e-mailed with a request for participation between 18 December 1995 25 January 1996. This direct solicitation outlined the purpose of the survey, and requested the participation of the most senior individual within the company responsible for marketing decisions. Respondents were requested to complete the survey online at a web site Uniform Resource Locator (URL) address that was provided. Printed versions of the survey were posted or faxed upon request. If no response was received, e-mail reminders were dispatched automatically 14 days following first contact. When respondents 'visited' the survey home page a brief introduction was provided, and then the participant was invited to either go directly to the online survey form, or to read further background information regarding the research programme. Relevant published research papers were also available in WWW format for those wishing to delve further into academic discussion of issues of online branding and marketing. The survey was structured to make maximal use of such HTML elements as pop-up lists, check boxes and selector ('radio') buttons for the convenience of the respondent, features impossible to offer in traditional paper-based surveys. An additional benefit of the use of such elements was the survey form was considerably shorter in length than would be the case in paper format - less intimidating for the respondent!
Respondents were asked to answer 24 questions that were grouped into four main categories (i) organisational data (name, number of employees, industry sector), (ii) business case for web site (prime mover, creator, budget, maintenance), (iii) web site data (address of web site, creation date, categorisation, complexity, information provided, restrictions on access, speed of connection from server to Internet backbone), (iv) success of project (success/failure, criteria for measurement of success), (v) respondent data (position in company, age, qualifications, primary computing platform).
Data Collection
After completion of the survey the respondent clicked on the 'Send my answers now!' button at the foot of the survey and the data was automatically entered via the AppleScript ACGI into the database. Feedback to the respondent was provided via a HTML 'thank you' form generated by WebSTAR, and a personalised e-mail. The e-mail was generated by an AppleScript applet controlling Eudora, and triggered by incoming data into the database. The entire survey engine ran on a Apple PowerMacintosh 7500/100. For much of the time the survey was 'live' this machine was dedicated to the task, however, for two weeks the computer was running all the survey tasks in the background whilst being used for normal office tasks such as word processing and electronic mail.
Communication & Publication of Results
Upon completion of the survey, the data was analysed and a HTML page with the summary results was attached to the survey home page. All respondents were automatically e-mailed with this URL, and thus had access to the data in advance of formal publication.
Selected Survey Results
Full presentation and discussion of the results will be presented elsewhere (Stevens, in prep.) and is available online (http://www.vuw.ac.nz/marketingsurvey/) however, some of the more salient aspects of the data will be discussed below.
One hundred and seven (107) completed questionnaires were received from the 209 e-mailed requests for participation (response rate of 51.1%). One incomplete questionnaire was received and was excluded from the survey results. Respondents were largely from senior management, marketing or information systems positions in the organisations (GM, CEO or Owner Operator, 53%; Information Systems / Computer Related, 23%; Marketing Manager,10%). Organisations in the computer / IT industry were most strongly represented (51%), followed by publishing (10%), manufacturing (7.4%) and tourism (4.7%) sectors. Small companies with less than 10 employees formed the majority (66%) of the respondents, with medium (11-50 employees) and (>51) large organisations each forming 17% of the respondents. Average number of employees was 126.6, median 5.0, range 14,500. This distribution differs significantly (d.f=2, c2=229.7, P<0.0001) from that observed in New Zealand business in general (Department of Statistics Business Activity Statistics, 1994), with a lower representation of small (<10 employees) organisations, and over representation of both medium and large organisations in the survey sample (Figure 4).
Figure 4
Size of organisations responding to the survey in comparison to the Department of Statistics Business Activity Statistics (1994). Small organisations (<10 employees), medium (10-50), large (51+).
The process of building and maintaining a web site can be separated into four distinct phases: ideation (who was the prime mover behind the web site project), creation (who created the web site), budget (which budget pays for the site) and maintenance (who has responsibility for the maintenance of the web site on an on-going basis). Where responsibility for each of these tasks lies within the organisation provides an interesting perspective on how each organisation views the role of the Internet in their corporate strategy. The combined data set indicated that the prime mover behind the establishment of the web site in the majority (53.3%) was found to be the 'Top' individual (CEO or General Manager; in smaller organisations the Owner/Operator). This individual also played a large role in both the creation and maintenance of the web sites. External consultants (usually Internet service providers) were involved in creating 21.6% of the web sites. The cost of the establishment and maintenance of the web site was funded from either general overhead (42.1% of organisations) or marketing budgets (23.4%), while the work involved in maintenance was largely undertaken by the Information Systems employees (29.9%). These patterns of responsibility were found to be significantly affected by the size of company, where in larger companies (c2=29.3, d.f.=4, P<0.001) (Figure 5). Small companies (< 10 employees) typically had a more active involvement by the 'Top' and funded web site development straight from overhead. The majority of medium and large companies funded their web development from marketing (32.4%) or information systems (21.6%) budgets rather than general overhead, and used their computer/information systems employees to maintain the information in their web site.
Figure 5
The process of building and maintaining web sites. For details see text. Aggregate data shown at top. Differing patterns of responsibility based on size of organisation shown below.
Currently most New Zealand commercial web sites provide information rather than offering an opportunity to purchase a product. Using the classification scheme proposed by Hoffman & Novak (1995) 60.7% of web sites were informational or image-based "Internet Presence" sites, 24.3% served as "Online Storefronts", and a further 6.5% were primarily run as "White/Yellow Pages". The majority (57.9%) of companies considered their web site to be a success, although of these companies, only 65.4% acknowledged that they had any criteria in place for measurement of this success. Of those that indicated they had success/failure criteria, the majority (55.7%) used a combination of criteria, 21.4% the number of electronic requests for information ('hits'), 10% leads/enquiries, 4.3% the number of sales and 4.3% the number of entries in their online guest book.
At an aggregate level, opinions were evenly divided on the importance of the web site for the business, ranging from the Internet representing just another channel of communication for their customers, through to a decisive statement that the Internet represented the future of the company. Size of business was found to have a very significant (df=4, c2=20.8, p<0.001) effect on organisational response. Small companies represented were more positive than medium/large companies, with 64.2% stating that the role of the Internet "probably represents a fundamental change in the way they communicate with customers and business partners in their business" or is "the future of their company". This compares with 44.4% of medium/large companies. Industry sector was found to have a lesser, but still significant, (df=4, c2=10.5, p<0.05) effect on response, with those from the computer/information technology (IT) sectors more positive (69% either probably be or the future as defined above) than other industry sectors (43.2%). Medium/large companies and those from outside the computer/IT sectors are more likely view the Internet as "just another channel of communications to customers and business partners".
Discussion
The Internet is a phenomenon commanding significant interest in the business community. While detractors charge that the Internet is over-hyped and causing significant wastage of business resources (Nulty, 1995; Stoll, 1995), the survey reported in this paper indicates that the majority of New Zealand businesses with WWW sites believe that their sites have been successful and have added value to their business. The survey revealed some interesting perspectives on the decision making process within the respondent organisations - these will be discussed elsewhere (Stevens, in prep.).
Conducting an online survey of the type reported here has many advantages over existing 'post & paper' survey techniques if the intended survey participants are on the Internet. Even though the companies surveyed here could be expected to be predisposed to using the Internet, clearly the Internet was the most convenient form of communication for them. Indeed, only six of the companies surveyed requested the survey via either facsimile or post. For the company, completing the survey online was a convenient one-step process obviating the need to enclose their reply in an envelope for return, and providing them instant feedback that their answers had been received. An additional benefit for the respondent was easy access to background information on the research programme, participating academics and other literature that was provided at the web site. Respondents could choose to go directly to the survey, or delve into supporting literature at their leisure.
For the surveyor, an online survey provides a very cost-competitive and efficient solution. Benefits can be found in four main areas:
(i) Cost: electronic dissemination of letters and survey forms costs a fraction of the total labour, consumable and postage costs of traditional paper-based surveys. In addition, an unintended consequence of online surveys is that some of the cost is born by the respondent by way of online fees or data transmission charges. In addition, the respondent is actually entering the data directly into the survey database - no data entry costs for the surveyor.
(ii) Accuracy: no keying of data is required, obviating the necessity for cross-checking and error correction.
(iii) Automation: information can be dispatched and processed automatically in near real time. This provides gains in productivity and efficiency for the surveyor, and a level of responsiveness unobtainable with traditional surveying techniques.
(iv) Scalability: factors (i) to (iii) listed above combine to make online surveying suitable for both small and large scale surveys. Ordinarily large scale surveys are a massive undertaking, requiring significant logistical and clerical resources. Consequently, such large scale surveys are usually undertaken by specialist market research organisations, or in the case of university researchers, with the aid of substantial research grants. With online surveys even large scale projects can be undertaken with standard equipment and off-the-shelf software.
All the advantages listed above accrue to anybody adopting online surveying techniques. However, in this case the use of software tools such as AppleScript and AppleEvents provided an additional level of flexibility and cost minimisation for the surveyor. The intra- and inter-application communication and control provided by these two tools enable surveyors to mix-and-match from a range of standard, off-the-shelf software programmes for different stages of the process (server, database, e-mail, browser). In particular, the English-like scripting language of AppleScript avoided having to employ computer programmers. Re-purposing the scripts for different surveys or tasks is trivial and can be done by the academic surveyor without specialist programming assistance. Such generic engines can be deployed in variety of Intranet and Internet settings, automating such tasks as conference registration, equipment booking and class assessments (e.g. Greenwood & Recker, in press).
Back to Home Page ...... by Presenter ...... by Paper Title ...... by Keyword
References
Greenwood, J. & Recker, M. (in press). Networked, asynchronous student evaluations of courses and teaching: An architecture and field studies. In Educational Telecommunications, 1996, AACE: Virginia.
Hoffman, D.L. & Novak, T.P. (1996). Marketing in Computer-Mediated Environments: Conceptual Foundations. Journal of Marketing: July.
Nulty, P. (1995). Why you can - and should- wait to get on the Internet. Fortune June 26: 96.
Pitkow, J. and Recker, M. (1995). Using the Web as a Survey Tool: Results from the Second WWW User Survey. Journal of Computer Networks and ISDN Systems, Vol. 27, no. 6.
Stevens, P.M. (in prep.). Adoption of Emerging Technologies: Business Decision Making Processes and Success Evaluation.
Stevens, P.M. & Greenwood, J. (in press). Online executive education. In: Proceedings of the European Distance Education Network Conference, 1996.
Stoll, C. (1995). Silicon Snake Oil. Bantam Doubleday Dell.
Dr. Peter M. Stevens, Senior Lecturer, Marketing Group, Victoria University of Wellington
PO Box 600, Wellington, New Zealand
e-mail: Peter.Stevens@vuw.ac.nz, Phone: +64-4-472-1000 x8245, Fax: +64-44-471-2200
Back to Home Page ...... by Presenter ...... by Paper Title ...... by Keyword
