• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

The State of Internet-Related Research in Communications, Marketing, and Advertising: 1994-2003

Posted on: Wednesday, 11 October 2006, 03:00 CDT

By Cho, Chang-Hoan; Khang, HyoungKoo

ABSTRACT:

This study analyzes trends, patterns, and rigorousness in research studies about the Internet through a content analysis of published Internet-related papers in 15 major journals in communications, marketing, and advertising between 1994 and 2003. Five hundred thirty seven papers of 4,050 total papers in the 15 journals (13.3%) were about the Internet. The results of this study demonstrate a definite pattern of increase in Internet research, the existence of a wide dispersion of individual and institutional contributors, a need for a more explicit theoretical framework, and a need for better methodological rigors in Internet research. In terms of topical coverage, this study concludes that Internet research deals with diverse research topics, but has not quite reached the last phase of Wimmer and Dominick's (2002) media developmental model (i.e., how we can improve the Internet).

It is estimated that, as of September 2003, over 606 million people were using the Internet worldwide (NUA 2003). The number of Internet users is still growing, and a significant part of people's daily life is devoted to the Internet, comprising 25 minutes per day spent on-line among average U.S. Internet users (Nielsen/NetRatings 2003). A recent survey conducted by the Online Publishers Association reveals that 45% of survey participants (18 to 54 years old) chose the Internet as their first choice when asked which medium they would choose to use if they could pick only one (MediaDailyNews 2004). Because of this growing Internet usage among the general public, the Internet has already been exerting a substantial impact on our society in general, as well as on important disciplines and practices such as communications, marketing, and advertising. Less is known about the Internet than the established media, however, and researchers have already realized the importance of investigating Internet-related phenomena from various perspectives. Indeed, the Internet, as a new communication, marketing, and advertising medium, has spurred many research topics and opportunities in our fields.

The Internet was originally developed for military purposes (constructing a highly decentralized, redundant network system for protection in case of national emergencies such as war or natural disasters) under the name "ARPANET" (Miller 1996). However, the Internet is now widely used in many aspects of our daily lives (e.g., for information searches, communication, and entertainment) and is highly commercialized (with the use of e-commerce). In the realm of communication, the Internet has several unique characteristics that distinguish it from traditional mass media, such as interactivity, multimedia techniques, easy 24-hour access to the most updated information, global coverage, and so forth (Pavlik 1998). These unique characteristics and the explosive growth of Internet usage have led many communication researchers to begin to research this medium's impact on the overall communication process. The same holds true for marketing and advertising research about the Internet. Since the commercialization of the Internet as a marketing and advertising medium in 1994, that is, with the advent of the first advertising contract between Hotwired and AT&T and the first e- commerce in AOL (Zeff and Aronson 1999), many scholars and professionals have recognized the importance of the Internet as a new innovative marketing and advertising medium. In fact, Internet advertising revenue has increased significantly during the last decade: $55 million in 1995 to $9.4 billion in 2004 (eMarketer 2004). In accordance with this increase in Internet advertising expenditures, advertising scholars have already started to research the impact of the Internet on the overall marketing and advertising process by examining the unique opportunities that the Internet provides in marketing communication practices, such as increased consumer targeting, easy tracking of consumer behavior, highly customized marketing messages, two-way interactions between marketers and consumers, and interactive customer services (Mohammed et al. 2001).

Based on the increasing significance and proliferation of the Internet in communication, marketing, and advertising, it is important to explore the state of research about the Internet, to assess the development of Internet-related research and grasp its future directions in our fields. Like other studies of research trends (e.g., Kim and Weaver 2002; Wartella and Reeves 1985), the current study aims to help scholars understand future research directions to improve our knowledge about the Internet. In particular, this study will analyze the development of Internet- related research based on Wimmer and Dominick's (2000) four developmental phases of media research. The first phase covers issues about the medium itself, such as definitions, unique characteristics, regulation, economic issues, and technical issues. The second phase deals with uses of the medium, such as user characteristics, user perception, user attitudes, and usage patterns. The third phase involves effects of the medium, such as its impact on our culture, society, everyday life, the political system, and companies. Finally, the last phase is medium improvement, which encompasses future directions, advanced applications, technical development, building theories and models, and so forth. Based on this developmental model, this study examines which phases Internet research is undergoing, as well as which it has tapped into and/or has reached. The study also provides information about what we should do to improve our knowledge about the Internet. This endeavor is based on analysis of past research patterns and trends through content analysis of published papers in major journals.

Advertising is interdisciplinary in its nature; that is, it is a hybrid, or applied field, built by the combined interests of communication and marketing. Advertising is commonly defined as a paid, mediated "communication" form to "market" products by an identified sponsor (American Marketing Association 1960); that is, it is a "marketing communication" element. This is why various marketing and communication theories have been applied to advertising research (e.g., attitude, learning, decision making, persuasion, and/or information theories), even though advertising has its unique concerns, such as "skepticism,""repetition,""message coordination," and "clutter" (Nan and Faber 2004, p. 7). Accordingly, advertising is researched and taught, not only by independent advertising faculty, but also by communication, journalism, and business faculties (Soley and Reid 1983). Acknowledging this interdisciplinary nature of advertising, the current study focuses on three fields (communications, marketing, and advertising), and provides a detailed outline of Internet research for each field in terms of topical, theoretical, and methodological endeavors.

PREVIOUS EFFORTS AND RESEARCH PURPOSES

Previous studies of research trends have primarily followed three paths. The first involves an analysis of the origins of research articles or productivity, such as individual authors, departmental affiliations, academic ranks, or institutions credited (e.g., Clark 1985; Cole and Bowers 1973; Musambira 2000; Schweitzer 1988; Wheatley and Wilson 1987). The second stream of trend research, though less frequently employed, examines research topics or content of published research articles (e.g., Edwards and La Ferle 2003; Russell and Martin 1976; Yale and Gilly 1988). The third stream of trend research has focused on a specific topical area of published research articles analyzing the trends in research methods, units of analysis, sample, and conceptual foundation. The areas of focus include international service advertising (Stafford 2005), international advertising (Zou 2005), and international marketing (Aulakh and Kotabe 1993). Table 1 provides a detailed summary of the three types of extant trend research.

Regarding the Internet, two papers have addressed the status of Internet-related research in the communication field, Tomasello (2001) and Kim and Weaver (2002). Tomasello (2001) examined five leading communication journals with regard to frequency, topic, and archetype of published Internet-based research studies between 1994 and 1999. Kim and Weaver (2002) surveyed trends in topics, methods, and theories employed in the published communication research about the Internet between 1996 and 2000. Even though both studies have provided some valuable insights into research trends of Internet studies, they were limited to the communication field and were relatively constrained in their scope of analysis. The present study addresses the need for a more updated and comprehensive research trend study with an expanded scope of analysis (e.g., more elaborate analysis of authorship, methods, concepts, units of analysis, and theories) that covers more fields (marketing, advertising,and communications) and compares them to each other. In short, the current study poses the following general research questions:

RQ1: What are the frequency, prop\ortion, and occurrence patterns of published Internet-related research articles in major communication, marketing, and advertising journals?

RQ2: What is the authorial, topical, theoretical, and methodological status of Internet-related research in major communication, marketing, and advertising journals?

More specifically, the current study analyzes past studies about the Internet for the purpose of examining:

1. how the major journals in each discipline are keeping pace with the Internet, as well as their responsiveness to this new, innovative medium;

2. which institutions/departments/individuals contribute the most to Internet-related research in general, and in each discipline;

3. how Internet-related research is developing and evolving into the future;

4. which study topics are prevalent and which are under represented in general, in each discipline and in each journal;

5. which theoretical and methodological approaches have been applied to Internet research, in general, in each discipline and in each journal; and

6. what we should do as researchers to improve our knowledge about the Internet in each discipline.

METHOD

Sample

We limited our analysis to published research articles in 15 major journals in communications, marketing, and advertising. We regarded two types of journals as major journals: leading journals and journals specifically devoted to the Internet and new technology in each field. The criteria for selecting these major journals were twofold. First, leading journals in each field were selected because they are leaders in terms of readership and quality, as they are the most influential to scholars in each discipline by enhancing their knowledge of the field. second, journals specifically devoted to the Internet and new technology could not be ignored for the obvious reason that their focus is geared toward Internet-related research topics, providing more research opportunities to Internet-focused scholars and advancing their knowledge, especially in the topical areas.

In the field of communications, five leading journals (Communication Research, Human Communication Research, Journal of Broadcasting and Electronic Media, Journal of Communication, and Journalism and Mass Communication Quarterly) and two Internet- specific journals (Journal of Computer-Mediated Communication and New Media and Society) were selected. The five leading communication journals were chosen based on the criteria used by Tomasello (2001): leading journals were identified by previous research and indexed in Social Sciences Abstracts. Journal of Computer-Mediated Communication was selected because it has the longest history of publishing Internet-related research articles, and New Media and Society was included because it published the most recent Internet- trend study (i.e., Kim and Weaver 2002). In the field of advertising, we included three leading journals identified by previous research (Barry 1990; Henthorne, LaTour, and Loraas 1998; Soley and Reid 1983; Yale and Gilly 1988): Journal of Advertising, Journal of Advertising Research, and Journal of Current Issues and Research in Advertising. Journal of Interactive Advertising was included because it focuses on Internet advertising. In the field of marketing, Journal of Marketing, Journal of Marketing Research, and Journal of Consumer Research were selected as leading journals. These journals are ranked as the top three in the marketing field by Theoharakis and Hirst's "Worldwide, North American and European Journal Rankings," which appeared in a 2002 article and was based on "a weighted index of familiarity and average rank position" (p. 391). These three marketing journals were also ranked as the top three, based on faculty perception of "importance/prestige" and "popularity/familiarity" (Huit, Neese, and Bashaw 1997, p. 40). Other previous trend studies also used these three journals (e.g., Yale and Gilly 1988; Zou 2005). Journal of Interactive Marketing was included because it has the longest history of publishing Internet marketing research. Appendix 1 provides a brief overview of selected journals in terms of circulation, acceptance rate, topical areas, and year launched.

Article Extraction

This study defines Internet-related research as any research studies that examine the Internet as a communication, marketing, and/ or advertising medium influencing our society, culture, economy, commerce, and political system. This broad definition includes the Internet in general and other Internet segments such as the World Wide Web, e-mail, listserv, on-line chatting, virtual reality, bulletin board, instant messaging, FTP, newsgroup, and intranet. It also includes Internet access and infrastructure issues such as ISP (Internet service provider), on-line services (e.g., AOL), broadband, wireless, and so forth. Even though the World Wide Web became available to the public in 1992 (Greenlaw and Hepp 1999), the year 1994, right after the release of the first user-friendly Web browser (Mosaic) to the public in April 2003, seems to be the starting point of the Internet's popularity among the general public (Hofstetter 1998; Tomasello 2001). This becomes more apparent when we look at the historical number of Web sites (50 Web sites as of July 1992, 150 in July 1993, and 3,000 in July 1994) and observe that the percentage of U.S. public schools connected to the Internet increased from almost 0% in 1993 to 35% in 1994 (Gromov 1995). In addition, 1994 is when the first advertising contract between hotwired and AT&T was initiated and the first e-commerce on AOL occurred (Zeff and Aronson 1999). This is why we selected 1994 as the starting year for the current study. Following the example of previous research (e.g., Barry 1990; Henthorne, LaTour, and Loraas 1998), we used full-length published research articles as a unit of analysis for the current study, excluding editorials, book reviews, and other materials.

Measures

Each article was coded for year of publication, name of journal, name and affiliation of authors, research topics, theoretical applications, research questions/hypotheses, research methods, sampling methods, sample size, response rates, unit of analysis, and statistical analyses.

Journal and Year

The name of journal and year of publication for each article were coded because this information helps us understand which journals published more Internet-related research articles (the prevalence of Internet-related research) and how much focus was assigned to the Internet over time (year trend).

Authors and Institutions

Each article was also coded for names and affiliations of authors (organization and department at the time of publication). This information provides us with the knowledge of which universities, organizations, departments, and/or individuals have been productive in Internet-related research. Even though quantity of publication by organizations or individuals may not indicate the quality of research, organizations or individuals, it might help students and scholars with regard to employment, selection of schools for degrees, networking, a starting point for literature reviews, and so forth. It could also help us understand the degree of variety or uniformity among contributors to Internet-related research areas.

Research Topics

The research topics refer to the main subjects that the authors aimed to examine in their articles. The coding schemes for the research topics were inductively developed as the analysis proceeded. We started from very detailed classification pieces and grouped them into broader categories. Using this process, we developed 19 research topic categories:

1. Regulation and policy (privacy issues, equal access, etc.)

2. Internet usage, perception, and attitude toward the Internet (user characteristics, diffusion, motives, credibility, infrastructure, broadband, etc.)

3. Economic issues (economic effects, economic improvement, etc.)

4. Social/political issues (community activities, political campaigns, etc.)

5. Cultural issues (cross-cultural perspectives)

6. Technical issues (3D, rich media, virtual reality, etc.)

7. Web site design, content, and perception (Web site effectiveness, Web site evaluation, attitude toward the site, etc.)

8. Audience measurement issues (traffic, reach/frequency, etc.)

9. Internet marketing issues (the Internet as a marketing tool, e- marketing, on-line promotion, etc.)

10. On-line shopping and e-commerce (characteristics of on-line shoppers, etc.)

11. Internet advertising issues (the Internet as an advertising tool, attitude toward Internet ads in general, Web ad credibility, etc.)

12. Internet communication issues (the Internet as a mass medium, on-line news, interpersonal communication, on-line journalism, etc.)

13. Interactivity (interactive features, vividness, etc.)

14. Internet advertising effectiveness (Web ad in general, banner effectiveness, effectiveness of pop-up ads, effectiveness of e-mail ads, etc.)

15. Internet research issues (methodology, sampling, response rates, off-line versus on-line, etc.)

16. The Internet in comparison with other media (credibility, attitudes, etc.)

17. Education and instructional application (on-line classes, collaborative learning, etc.)

18. Historical and philosophical issues (feminism on the Web, historical trends, etc.)

19. Other (search functions, customization, browsing behaviors, future of Web sites, effects of Web addresses, etc.)

If an article covered more than one topic, it was coded into multiple topic categories. For example, the article "Interactive Disaster Communication on the Internet: A Content Analysis of Sixty- Four Disaster Relief Home Pages" (Paul 2001), was coded into three topical categories: Web site content, on-line news, and interactivity. Fifty one percent of the coded articles contained multiple topics. The analysis of research topics provides us with information about which topics are prevalent, which a\re underrepresented, and which require further study, so we can clarify the direction of future research.

Theoretical Application/Hypotheses/Research Purpose

We coded each article in terms of theoretical framework (having testable hypotheses or generalizations based on theory or deriving theory from systematic observations). Adopting definitions of theory from Hunt (1983) and Babbie (2001), we identified a theory as a systematic explanation or lawlike generalization that is observable or empirically testable. Based on this definition, all articles were classified into two categories: explicit theoretical framework versus descriptive with no theoretical framework. For an article with an explicit theoretical framework, we also coded which theory was used, applied, and/or developed to explain Internet phenomena. This information will help us understand the overall theoretical depth of Internet-related research (whether Internet-related research lacks theoretical application or not), and which existing or new theories were applied to, or developed for, the new communication technology. We also coded each article in terms of the statement and content of research questions and hypotheses, which helped us understand which study type among exploratory, descriptive, and explanatory research has been dominating Internet- related research.

Research Method/Data Collection

The research method (e.g., content analysis, experiment, survey, interview, focus group, etc.) for each published article was coded for the purpose of understanding which research methods are most widely employed to examine Internet-related phenomena. This information also helps us understand research perspectives (qualitative versus quantitative) for Internet-related research, and how they are compared to other research areas in communications, marketing, and advertising. We also coded each article in terms of data collection method: on-line (e.g., e-mail, on-line survey, and on-line experiment) and offline (e.g., paper-and-pencil, phone interview, and face-to-face interview). This information provides a cursory sense about the question of external validity (natural setting), that is, whether Internet-related research uses the Internet (on-line) to examine Internet (on-line) phenomena.

Sampling Method/Sample Size/Response Rate

Each article was analyzed in terms of the sampling method used in the study: probability (e.g., simple random, systematic random sample, stratified, and cluster) versus nonprobability (e.g., convenience, judgment, quota, and snowball) sampling methods. We also coded sample size, response rates, and sample population (e.g., student, general population, experts, etc.) for each Internet- related article. This information on various sampling issues helps us grasp the rigor of sampling approaches for Internet-related research (external validity).

Unit of Analysis/Statistical Analysis/Reliability Test

Each article was also coded for the unit of analysis employed in the study (e.g., Web sites, banner ads, pop-up ads, games, key word search, e-mails, etc.). This provides information about which unit of analysis is most frequently or rarely used, thus helping us to understand which component of the Internet is over- or underrepresented in Internet-related research. Statistical analysis methods (e.g., frequency, mean, correlation, multiple regression, MANOVA [multivariate analysis of variance], SEM [structural equation modeling], etc.) were also coded for each research article to assess the statistical rigor of Internet-related research.

RESULTS

Perreault and Leigh's Ir (index of reliability)1 was employed as a measure of reliability in this study. This is considered to be a better reliability measure than percentage agreement, Scott's π, or Cohen's κ.2 First, two coders independently analyzed two randomly selected journals for the inclusion of each article as Internet-related research, and the resultant Perreault and Leigh's Ir was .95. Out of 537 extracted articles as Internet-related studies, 50 randomly selected articles were coded independently by the two coders for each coding category. The intercoder reliability estimate of each coding category was satisfactory, with all above the acceptable point (higher than .80) (Perreault and Leigh 1989). The Perreault and Leigh's Ir was 1.0 for year of publication, name of journal, and name of authors; .99 for institutional affiliations; .98 for departmental affiliations; .86 for research topics; .88 for theoretical applications; .94 for research methods; .95 for sampling methods; .95 for sample size; .89 for response rates; .84 for unit of analysis; and .97 for statistical analysis. Once we reached satisfactory reliability estimates for all coding categories, only one coder analyzed the remaining 487 articles.3

Journal and Year

A total of 4,050 full-length articles were published in the 15 journals over the period of the study (1994 to 2003). Five hundred thirty-seven articles out of 4,050 were extracted as Internet- related research (133%). Table 2 shows frequencies of Internet- related articles by each field, each journal, and each year. For a specific field, 18.3% (369 out of 2,015 articles) of communication journal articles were about the Internet, 4.9% (57 out of 1,172) for marketing journals, and 12.9% (111 out of 863) for advertising journals. For the 11 leading journals, 5.82% of all published articles were about the Internet. Among them, Journalism and Mass Communication Quarterly had the highest proportion of Internet- related research articles (14.0% of total articles in the journal), followed by Journal of Advertising Research (12.1%), Journal of Broadcasting and Electronic Media (8.6%), and Journal of Current Issues and Research in Advertising (7.5%). For Internet-specific journals, Journal of Computer-Mediated Communication had the highest proportion of Internet-related research articles (97.1%), followed by Journal of Interactive Advertising (87.5%), New Media and Society (49.5 %), and Journal of Interactive Marketing (32.6%). Trends in the publication of Internet-related articles across the 15 journals revealed an obvious pattern of increase in the proportion of Internet-related research articles during the ten-year period: .5% in 1994,4.3% in 1995, 8.3% in 1996, 10.2% in 1997, 12.7% in 1999, 20.1% in 2001, and 23.3% in 2003. This difference in the percentage of Internet-related articles for each year was statistically significant (χ^sup 2^ = 192.49, df = 9, p < .01). For the specific fields, we found a similar pattern of increase (see Figure 1).

Authors and Institutions

The total number of author appearances for the 537 Internet- related articles was 1,045, or an average of 1.95 authors per article. During the time frame of our study, 92.2% of all articles published in the 15 journals were submitted by academics. For ranking, we used total appearances of an organization/department/ individual. As seen in Table 3, contribution by organization/ university combining all 15 journals revealed that Michigan State University (54) appeared the most, followed by the University of Wisconsin-Madison (23), the University of Southern California (22), and the University of Florida (21). For the 11 leading journals, Michigan State University and the University of Florida were the top two (18 each). For communication journals, Michigan State University (34) appeared the most. For marketing journals, Columbia University (6) and the University of Florida (6) appeared the most. For advertising journals, Michigan State University appeared the most (19), followed by the University of Minnesota (12), the University of Missouri (9), the University of Florida (7), the University of Georgia (7), the University of Southern California (7), and the University of Texas at Austin (7).

Table 3 also shows comparative author rankings in terms of article numbers. Overall, the most frequent contributor across the three fields was Kim Sheehan, followed by Sally McMillan, Shyam Sundar, Ananda Mitra, Shelly Rodgers, and so forth. If we limit our analysis to the 11 leading journals, the most frequent contributor across the three fields was Shyam Sundar, followed by Thomas Johnson, Joseph Walther, Chang-Hoan Cho, Steven Edwards, and so forth. For communication journals, Shyam Sundar appeared the most frequently. For marketing journals, Asim Ansari, Ann Schlosser, Kim Sheehan, and Bruce Weinberg appeared the most frequently. For advertising journals, Shelly Rodgers appeared the most frequently, followed by Qimei Chen and Kim Sheehan.

Research Topics

As shown in Table 4, across all 15 journals, "Internet usage, perception, and attitude toward the Internet" was the topic most often researched. This was followed by "Internet communication issues,""Web site design, content, and perception issues,""Internet marketing issues,""Internet ad effectiveness,""social or political issues,""regulation or policy," and "Internet research issues." For communication journals, the topic most often studied was "Internet usage, perception, and attitude toward the Internet," while it was "Internet as a marketing tool and e-business" for marketing journals. For advertising journals, the top three topics most often studied were "Internet advertising effectiveness,""Web site design, content, and perception issues," and "Internet usage, perception, and attitude toward the Internet." Table 4 also shows the topic most often studied for each journal. During the first five years (1994-1998), combining all journals, the top four topics most often studied were "the Internet as a mass medium,""the Internet as a marketing medium,""regulation or policy," and "Internet usage, perception, and attitude toward the Internet." During the second five years (1999-2003), they were "Internet usage, perception, and attitude toward the Internet,""Web site design, content, and perception issues,""the Intern\et as a mass medium," and "social and political issues."

Theoretical Application

As shown in Table 5, 78 out of 537 Internet-related studies (14.5%) used an explicit theoretical framework, that is, they presented testable hypotheses or generalizations based on theory or they derived theory from systematic observations. For each field, articles with an explicit theoretical framework comprised 12.7% for communications, 14.0% for marketing, and 20.7% for advertising. Articles with an explicit theoretical framework comprised 17.8% for leading journals and 12.5% for Internet-specific journals. Over the ten-year period, during the first five years (1994-1998), 8.7% of articles had an explicit theoretical framework, and during the second five years (1999-2003), 16.3% of them did. Combining all 15 journals, theories mentioned or applied most often were uses and gratifications (18), followed by diffusion or adoption (13), agenda setting (6), third-person effect (3), dependency (1), and functional theory (1). The frequency pattern of theory mentioned/used was similar across the three fields.

Hypotheses and Research Purpose

As seen in Table 5, combining all journals, 146 out of a total of 537 Internet-related studies (27.2%) generated research hypotheses, that is, they predicted and tested relations between two or more variables. For each field, articles with research hypotheses were 20.1% for communications, 38.6% for marketing, and 45.0% for advertising. Articles with research hypotheses comprised 44.2% for leading journals and 16.4% for Internet-specific journals. Over the ten-year period, during the first five years (1994-1998), 11.8% of articles had research hypotheses, and during the second five years (1999-2003), 32.0% of them did. For the three purposes of research- exploration, description, and explanation (Babbie 2001, p. 91)- exploratory research, which clarifies and defines the nature of a problem of which researchers were previously unaware, dominated Internet-related research (63.9%). This was followed by descriptive research for the purpose of describing characteristics of a phenomenon (30.5%), and explanatory research to predict something based on cause-effect investigation (5.6%). For communication journals, the pattern was 71.5% for exploratory research, 24.1% for descriptive research, and 4.3% for explanatory research. For marketing journals, it was 56.1% for exploratory research, 35.1% for descriptive research, and 8.8% for explanatory research. For advertising journals, it was 42.3% for exploratory research, 49.5% for descriptive research, and 8.1% for explanatory research. The pattern was 45.7%, 46.6%, and 7.7%, respectively, for leading journals, and 75.4%, 20.4%, and 4.3%, respectively, for Internet- specific journals. It was 82.7%, 14.2%, and 3.1%, respectively, during the first five years (1994-1998), and 58.0%, 35.6%, and 6.3%, respectively, for the second five years (1999-2003).

Research Method/Data Collection

Table 7 provides frequencies and proportions of research methods. Combining all 15 journals, critique or essay (28.1%) was the research method most often employed. This was followed by survey (21.6%), experiment (16.4%), content analysis (14.1%), secondary data analysis (5.8%), in-depth interview (5.1%), and legal (3%). For communication journals, the top three research methods most often employed were critique/essay (30.8%), survey (19.3%), and content analysis (16%). For marketing journals, the top three methods were experiment (31%), critique/essay (25.9%), and survey (22.4%). For advertising journals, they were experiment (29.9%), survey (29-1%), and critique/essay (19.7%). For leading journals, the top research method was survey (29.4%), while it was critique/essay (39-6%) for Internet-specific journals. Critique/essay (43.4%) was the top method during the first five years (1994-1998), whereas survey (23.6%) was the top method during the second five years (1999- 2003).

Table 5 shows a comparison of qualitative versus quantitative research methods. Combining all 15 journals, 46.6% of all Internet- related articles used qualitative research methods, while 53.4% used quantitative research methods. The pattern was 53.1% versus 46.9% for communications, 42.1% versus 57.9% for marketing, and 27% versus 73% for advertising. By journal type, it was 21.2% versus 78.8% for leading journals and 62.6% versus 37.4% for Internet-specific journals. By time frame, it was 65.4% versus 34.6% during the first five years (1994-1998), and 40.7% versus 59.3% during the second five years (1999-2003). Journal of Consumer Research and Human Communication Research had the highest proportion of quantitative research (100% and 90.9%, respectively).

As shown in Table 6, a comparison of on-line versus off-line data collection methods shows that combining all 15 journals, 41.1% of all Internet-related articles involving data collection used on- line techniques, whereas 58.9% used off-line data collection methods. The pattern was 41.4% versus 68.6% for communications, 50% versus 50% for marketing, and 53.3% versus 46.7% for advertising. There was no significant difference between leading and Internet- specific journals, or between the first and second five years.

Sampling Method/Sample Size/Response Rate/Unit of Analysis

Table 6 shows a comparison of probability versus nonprobability sampling methods. Combining all 15 journals, 23.9% of all Internet- related articles involving sampling used probability research methods, while 76.1% used nonprobability sampling methods. There was no significant difference among the three fields-neither between leading and Internet-specific journals, nor between the first and second five years for the ten-year time period. The report of response rates in articles shows that combining all 15 journals, 33% of all quantitative Internet-related articles reported response rates. There was no significant difference among the three fields, neither between leading and Internet-specific journals, nor between the first and second five years for the ten-year time period. For sample size, the mean of the sample size for all Internet-related articles was 674.7 (M = 612.3 for communications, M = 756.2 for marketing, and M = 775.8 for advertising). In terms of sample size, there was no significant difference between leading journals and Internet-specific journals (M = 730.5 versus 604.2), or between the first and second five years (M = 791.1 versus 656.3) for the ten- year time period. For sampling population of Internet-related articles, students and general population were used to a similar extent (45.9% versus 54.1%). The result was similar by journal type and by field. It is interesting to note that we had more research studies using a general population sample during the first five years (74.1%) of the ten-year time period, as compared with the second five years (51.1%). Combining all 15 journals, Web site was most often employed as the unit of analysis (40%), followed by news story (34.7%), banners (8%), Web advertising in general (5.3%), e- mails (2.7%), and auction (1.3%). For communications, news story was the unit of analysis most frequently used (46.4%). For marketing and advertising, it was banners (50%) and Web sites (41.7%), respectively.

Statistical Analysis/Reliability Test

As shown in Table 7, the statistical analysis method most frequently used in Internet-related research was frequency (31.6%), followed by ANOVA (analysis of variance) (15.3%), multiple regression (12.5%), mean description (9.2%), correlation (8.6%), factor analysis (6.3%), and MANOVA (4.6%). The pattern was similar by field, time frame, and journal type; that is, frequency was most prevalent, followed by ANOVA and multiple regression or factor analysis.

DISCUSSION

The current study is the most updated (December 2003) research trend study regarding the Internet. Compared with previous trend studies that focused on only one or two of three types of trends (productivity, topical, or methodological trends), this study undertook a more comprehensive and detailed analysis covering all three types of trends. Furthermore, recognizing that advertising is interdisciplinary in its nature-a merged field of communication and marketing-we covered and compared three research fields (communications, marketing, and advertising). This distinguishes the present study from previous trend studies focusing on only one field and lacking interdisciplinary comparison. The main purpose of the study was to provide a detailed picture of research trends, patterns, and rigorousness in Internet-related studies in the fields of communications, marketing, and advertising, and to provide valuable insights for the direction of future research aimed at improving our knowledge about Internet phenomena.

This study found that 13.3% of research articles were about the Internet. For advertising, 9.2% of leading advertising journal articles covered Internet advertising issues, which is similar to the ratio of Edwards and La Ferle's (2003) trend study for the proceedings of the American Academy of Advertising (AAA) (9.6%). These percentages are larger than that of Internet advertising, as 2.5% of total advertising expenditures was devoted to Internet advertising in 2003 (ClickZ 2004). Since the Internet is a relatively new medium, it is understandable that more advertising research efforts are devoted to it, as compared with its actual advertising expenditure proportion. For Internet research over time (year trend), this study demonstrates a definite pattern of increase in terms of the number of Internet-related research studies in the fields of communications, marketing, and advertising. This implies that the Internet is exerting a significant influence on our fields in general, and that scholars are successfully keeping pace with the increased usage and impact of this new medium, which is consistent with increasing Inter\net advertising dollars during the last decade: $55 million in 1995, $900 million in 1997, $4.6 billion in 1999, $7.1 billion in 2001, $7.3 billion in 2003, and $9.4 billion in 2004 (eMarketer 2004). One particularly interesting finding is that the number of articles is increasing with a diminishing rate, similar to a convex response curve and a diffusion curve. The Internet is considered to still be in its growth stage, and we believe that the number of Internet-related research studies should continue to increase to keep pace with the medium's evolution, since we still know less about the Internet than we do about the established media. The Internet is still growing in terms of usage, impact, and applications. The Internet is not only influenced by other media, but helps shape the future of other existing media and gives birth to new media (e.g., PDA [personal digital assistant], information kiosks, interactive television, etc.).

Using the barometer of scholar performance suggested by previous research, such as the quality and reputation of the journals in which articles are published and the quantity of articles (Huit, Neese, and Bashaw 1997; Mason 1995), this study identified major contributors to Internet-related research. The results of the current study show that there is a wide dispersion of individual and institutional contributors; they are not highly concentrated in, or exclusively dominated by, certain institutions or individuals. This is a positive sign for dynamic disciplines and perspectives in Internet-related research. There were, however, several noticeable institutions that deserve recognition. Michigan State University, the University of Wisconsin-Madison, the University of Southern California, and the University of Florida were major institutional contributors.4 Kim Sheehan, Sally McMillan, and Shyam Sudan were the noteworthy individual contributors. Scholars, especially tenure- accruing junior faculty, might use the rankings to set their individual research goals.

We noticed that some of the universities that the current study ranked high in productivity for Internet advertising research were not ranked high in general advertising research productivity as identified by previous trend studies. For example, Michigan State University is ranked first in Internet advertising research, while being ranked 33 in general advertising research by Henthorne, LaTour, and Loraas (1998); the University of Florida is ranked fourth in Internet advertising research, but did not even rank in the top 50 universities in general advertising research in a study by Barry (1990), or in the top 20 universities ranked by Soley and Reid (1983); the University of Minnesota and the University of MissouriColumbia were ranked second and third, respectively, in Internet advertising research, but were not listed in Soley and Reid's (1983) top 20 universities for general advertising research productivity. We should recognize, however, that such differences in research productivity rankings might also be due to a different time frame used for each trend study.

This study also evidences that many of the people listed in Table 3 were not professors or scholars identified by previous research as major individual contributors in the advertising field (e.g., in the studies by Barry [1990], or Henthorne, LaTour, and Loraas [1998]). This implies that the Internet has definitely been providing more research opportunities to a more diverse set of faculty members. Even Ph.D. students frequently appeared as contributors to Internet- related research. Overall, the results of authorial trends indicate that our fields are continuously evolving and that leadership is passing on to new generations of scholars, at least in the area of Internet research. We also compared individual authors in terms of their interdisciplinary efforts (publishing articles across journals in three fields). Kim Sheehan and Hairong Li were the only two scholars who published across journals in three fields. There were several scholars who published across two fields: advertising and communication journals (e.g., Sally McMillan), advertising and marketing journals (e.g., Goutam Chakraborty), and communication and marketing journals (e.g., Donna Hoffman). This implies that despite the interdisciplinary nature of our fields, most scholars tend to publish their articles in journals devoted to one area.

The findings of topical analyses indicate that Internet-focused scholars have diverse interests in a variety of topics covering communication, marketing, and advertising, as well as social, economic, cultural, regulatory, and technological issues, and so forth. Among such topics, advertising scholars have put more emphasis on "Internet advertising effectiveness,""Web site design, content, and perception,""Internet usage, perception, and attitudes,""the Internet as a marketing tool," and "comparison with other media." Although we cannot directly compare it would be interesting to see how these topics are analogous to the topics identified by previous advertising trend studies. The common topical areas identified in this study (e.g., attitude toward the Web site, attitude toward the ad, Web site design, content and message, etc.) are equivalent to common advertising research topics identified in Edwards and La Ferle (2003), Muncy (1991), and Yale and Gilly (1988), such as attitude toward the ad, ad content, copy, execution of message, creative aspects of advertising, ad effects, and ad practice.

Based on Wimmer and Dominick's (2000) developmental model of media research, we can conclude that Internetrelated research has already tapped into the third phase (effects of the Internet), while still undergoing the first phase (issues for the Internet itself such as function, regulation, economic issues, technical issues, etc.) and the second phase (uses of the Internet such as Internet user characteristics, user perception, user attitudes, usage pattern, etc.). However, we have not quite reached the last phase of development (how we can improve the Internet), for we are lacking important research efforts such as studies on future directions, advanced applications, and technological development (e.g., interactive television and wireless), as well as building theories and models. We therefore conclude that these topics dealing with the improvement of the Internet require more study in the future.

The findings of this study also reveal that only 14.5% of all Internet-related articles explicitly mentioned or used any theory in their studies. For Internet research in advertising, the percentage (20.7%) was similar to that of advertising research on China, that is, 19% in Sin, Ho, and So (2000). However, it was significantly lower than that of advertising media planning research: 78.9% in Pasadeos et al. (1997). The corresponding percentage (14.0%) for Internet research in marketing was significantly lower than that of international marketing research, that is, 46% (Aulakh and Kotabe 1993). Similarly, for Internet communication research, the percentage (12.7%) was relatively lower than that of general communication research, that is, 27.6% in Cooper, Potter, and Dupagne (1994). This denotes that Internet-related research has not yet achieved an equivalent level of theoretical rigorousness, as compared to other communication, marketing, and advertising research areas.

The comparison of use of an explicit theoretical framework over the ten-year period, however, revealed that research studies during the second half (1999-2003) were more theoretically solid than those from the first half (1994-1998). This implies that Internet-related research is becoming more theoretically sound, which is a positive sign for our fields. Nevertheless, this does not mean that we are satisfied with the current theoretical rigorousness of Internet- related research. If we look at the kinds of theories used or mentioned in Internet-related research, we can easily see that there has been no new theory developed to explain Internet phenomena. Even more serious to consider is that only a small fraction of existing theories have been applied to Internet research (uses and gratifications, diffusion theory, agenda setting, etc.). We not only need to apply more existing concepts and theories to comprehend general Internet phenomena, but should also strive to develop new concepts and theories for understanding new aspects of the Internet that might not be completely explained by the existing knowledge structure. Similarly, the current study indicates that more explanatory research studies are needed for Internet-related research in communication, marketing, and advertising (only 5.6% of all Internet-related studies was explanatory research), even though the ratio did increase over the ten-year period (3.1% for the first five years and 6.3% for the second five years). This is consistent with our previous! argument that there is a need for more studies related to the fourth research phase (e.g., developing theories and models).

In terms of research perspectives of overall Internet-related research, this study reveals that quantitative research (e.g., survey and experiment) is slightly more prevalent than qualitative research (e.g., critique/essay, in-depth interview, ethnography, etc.), with 53.4% of the studies using quantitative research and 46.6% using qualitative research. This finding is consistent with those of previous studies (e.g., Cooper, Potter, and Dupagne 1994; Perloff 1976). For advertising and marketing, qualitative Internet research was significantly exceeded by quantitative research, and the gap was higher for advertising (27% versus 73%). This ratio is similar to that of general advertising/marketing research. For example, Yale and Gilly (1988) showed that 31.1% of 907 research articles in six advertising/marketing journals used qualitative research. Staffor\d (2005) also found less qualitative research (26%) than quantitative research (74%) in her trend study of international services advertising. The picture was different for Internetrelated research in communications, however; that is, there was slightly more qualitative research (53.1%) than quantitative research (46.9%). This finding is consistent with the results of Kim and Weaver's (2002) trend study of communication research about the Internet, although it was slightly less skewed than their finding (72.9% for qualitative research and 26.7% for quantitative research). Overall, we can conclude that Internet-related communication research has a better balance between qualitative and quantitative perspectives than does Internet-related advertising and marketing research. Indeed, researchers conducting advertising and marketing studies about the Internet would do well to strive for a better balance between the two approaches (i.e., by employing a more qualitative viewpoint). Comparison of the two perspectives over the ten-year period reveals that we seem to be proceeding toward more balance in this regard (65% quantitative versus 35% qualitative for the first five years, and 41% quantitative versus 59% qualitative for the second five years).

Regarding on-line versus off-line data collection methods for Internet-related research, this study shows that off-line methods of data collection are more common than on-line methods in communications (68.6% versus 31.4%). The two methods were equal in prevalence for marketing (50% each), and on-line methods were slightly more common than offline methods for advertising (53.3% for on-line and 46.7% for off-line). This implies that external validity, or using the Internet (on-line) for data collection to study the Internet (a more natural setting), is still questionable for Internetrelated research. Hence, we need to approach Internet- related research using more on-line data collection techniques (e.g., e-mail, listserv, on-line experiments, etc.). Morris, Woo, and Cho (2003) sufficiently summarize the advantages and disadvantages of using on-line data collection in comparison with off-line data collection. On-line data collection methods have several benefits over off-line methods, such as quicker and more accurate gathering of survey information with minimal cost, as compared with a traditional paper-and-pencil method (Kelley-Milburn and Milburn 1995; Rosen and Petty 1995). On-line data collection does have some limitations, however. Sheehan (2002) identified skewed Internet users, difficult accessibility of e-mail addresses, possible multiple responses, and low response rates as four problems of on-line data collection. This may be why we found that research studies using online data collection were more likely to use a nonprobability sample (75%) than a probability sample (25%) compared to those using off-line data collection (57% versus 43%). There were no noticeable differences between on-line and off-line data collection methods in terms of qualitative versus quantitative research, sampling population, or report of response rate.

With regard to the sampling rigor of Internet-related research, the results suggest that we need to have more scientific sampling (probability sampling): 76.1% of overall Internetrelated research and 77.8% of Internet research in advertising used nonprobability sampling methods. This percentage is similar to Sin, Ho, and So's (2000) figure (75%). We also need to strive for more representative samples (general population samples rather than student samples) and more accountability (i.e., report of response rates). It is important to understand why such probability sampling of general Internet users is not widely used in Internet-related research. Morris, Woo, and Cho (2003) have pointed out that because no available exhaustive sampling frames list all Internet users, most on-line research studies have used nonprobability samples rather than probability samples, and student samples have been very common in Internet research. They suggest that an accurate and exhaustive list of general Internet users is an essential prerequisite for improving the sampling rigor of Internet research.

The mean sample sizes for overall Internet research and advertising-specific Internet research were 675 and 776, respectively. This indicates that the sample size in Internet research surpasses the minimum of 100 participants agreed on by many researchers (Bailey 1982). Regarding statistical methods employed in Internet research in advertising, we found that many Internet research studies in advertising have used basic descriptive statistical analyses such as frequency, mean, and correlation (40%). This ratio is similar to that of general advertising research, that is, 48% in Yale and Gilly (1988). In general, advertising research is often criticized for using basic statistical techniques (Soley and Reid 1981; Yale and Gilly 1988), and our study shows that Internet-related research in advertising is not an exception. Therefore, we need to strive to achieve enhanced statistical rigor for Internetrelated research by employing more advanced inferential statistical methods such as SEM, MANOVA, multiple regression, multiple discriminant analysis, and canonical correlation. We should notice, however, that the above conclusion (the need for more scientific sampling, more representative samples, and more advanced inferential statistics) is based on our positivistic approach to research (i.e., scientific sampling and advanced statistical methods are desirable). To identify future research needs, we primarily employed quantitative criteria (e.g., the number of articles using probability sampling methods) rather than qualitative criteria (e.g., the quality or appropriateness of specific sampling and statistical methods used in each study). A qualitative trend analysis would have provided a different picture than that found in the current quantitative analysis.

Our study was limited to 15 journals in the fields of communications, marketing, and advertising. If we had included more journals, such as Marketing Science, International Journal of Advertising, Journal of Consumer Behavior, and so forth, it might have provided a different picture of the authorial, topical, theoretical, and methodological trends and patterns. In particular, regarding authorship, there may be other major contributors not listed in our study who have published in other journals. These researchers are not fully reflected in our findings, but deserve recognition for their accomplishments. In addition, regarding contributors to the fields under consideration, we did not use other important criteria such as number of citations and usage by other researchers, or weighted ranking based on award winning and authorship order (sole versus coauthorship, first versus second, etc.). We should also note that the analysis of institutional contribution may be biased because each institution has a different faculty size. Therefore, in future research, it would be valuable to replicate the current study, using more journals and various ranking and analysis systems.

NOTES

1. Perreault and Leigh's Ir =

{[F^sub 0^/N - (1/K)] [K/(K - 1)]}^sup 5^ if F^sub 0^/N ≥ 1/ K,

0 if F^sub 0^/N < 1/K,

where F^sub 0^ = frequency of observed agreement, N is the total number of pairwise judgment, and K is the number of categories into which the responses can be coded.

2. Even though both π and κ are considered better than simple percentage agreement because they correct for chance agreement, both measures are overly conservative due to the way they estimate the agreement cell frequencies to be expected by chance, that is, they don't consider any agreement among judges on the marginal distributions (Rust and Cooil 1994), and their view of chance ignores the fact that judges can agree about the categories into which cases are more likely to fit (Brennan and Prediger 1981). In remedy, Perreault and Leigh's approach "makes it possible to compute directly a sample-based estimate of reliability and the associated standard error of estimate" (1989, p. 140) and improve the quality of judgment-based nominal scale data.

3. Even though multiple coders are desirable for content analysis, we employed only one coder since the nature of coding schemes or categories for the current study (i.e., author's name, year of publication, research topics and methods, and statistics used in the articles) is not latent, but manifest (i.e., content that is easily recognized and counted). Therefore, the coder did not need to interpret the meanings embedded in the content, but instead simply recorded the manifest categories presented in the articles.

4. We ranked institutions based on their total appearances through author affiliations, not by average productivity per scholar. Consequently, some institutions may be ranked high because they have larger faculties or programs. If we had used average productivity instead of total productivity, the results could have been different.

REFERENCES

Adams, Edward, and John V. Bodle (1995), "Research Presented at Conventions: How Well Are Women Doing?" Journalism and Mass Communication Educator, 50 (2), 14-22.

American Marketing Association (1960), Definitions: A Glossary of Marketing Terms, Chicago: American Marketing Association.

Aulakh, Preet S., and Masaaki Kotabe (1993), "An Assessment of Theoretical and Methodological Development in International Marketing: 1980-1990," Journal of International Marketing, 1 (2), 5- 28.

Babbie, Earl (2001), The Practice of Social Research, 9th ed., Belmont, CA: Wadsworth.

Bailey, Kenneth D. (1982), Methods of Social Research, 2nd ed., New York: Free Press

Barry, Thomas E. (1990), "Publication Productivity in the Three Leading U.S. Advertising Journals: Inaugural Issues Through 1988," Journal of Advertising, 19 (1), 52-60.

Blake, Ken, Joh\n V. Bodle, and Edward E. Adams (2004), "A Fifteen-Year Census of Gender and Journal Productivity," Journalism and Mass Communication Educator, 59 (2), 156-170.

Brennan, Robert L., and Dale J. Prediger (1981), "Coefficient Kappa: Some Uses, Misuses, and Alternatives," Educational and Psychological Measurement, 41 (Winter), 687-699.

Clark, Gary L. (1985), "Productivity Ratings of Institutions Based on Publication in Eight Marketing Journals: 1983-84," Journal of Marketing Education, 1 (Fall), 12-23.

________, and Nissim Hanna (1986), "An Analysis of the Source of Articles in the Journal of Marketing Education Since Its Founding in 1979," Journal of Marketing Education, 8 (Summer), 71-74.

ClickZ (2004), "U.S. Online Ad Growth Underway," available at www.clickz.com/stats/sectors/advertising/article. php/5941_2234931 (accessed May 27, 2005).

Cole, Richard R., and Thomas A. Bowers (1973), "Research Article Productivity of U.S. Journalism Faculties," Journalism Quarterly, 50 (Summer), 246-254.

Cooper, Roger, James W. Potter, and Michael Dupagne (1994), "A Status Report on Methods Used in Mass Communication Research," Journalism Educator, 48 (4), 54-61.

Edwards, Steven M., and Carrie La Ferle (2003), "Trends of the American Academy of Advertising Conference Proceedings: 1983-2002," in Proceedings of the 2003 Conference of the American Academy of Advertising, Les Carlson, ed., Clemson, SC: American Academy of Advertising, 210-220.

eMarketer (2004), "US Online and Total Media Advertising Spending," available at www.emarketer.com (accessed November 10, 2004).

Greenlaw, Raymond, and Ellen Hepp (1999), In-line/On-line: Fundamentals of the Internet and the World Wide Web, Boston: WCB/ McGraw-Hill.

Gromov, Gregory R. (1995), "History of Internet and WWW: The Roads and Crossroads of Internet History," available at www.netvalley.com/intvalstat.html (accessed May 28, 2005).

Henry, William R., and Earl E. Burch (1974), "Institutional Contributions to Scholarly Journals of Business," Journal of Business, 47 (1), 56-66.

Henthorne, Tony L., Michael S. LaTour, and Tina Loraas (1998), "Publication Productivity in the Three Leading U.S. Advertising Journals: 1989 Through 1996," Journal of Advertising, 21 (2), 53- 63.

Hickson, Mark, Joshua Turner, and Jean Bodon (2003), "Research Productivity in Communication: An Analysis, 1996-2001," Communication Research Reports, 20 (4), 308-319.

Hofstetter, Fred T. (1998), Internet Literacy, New York: McGraw- Hill.

Huit, G. Tomas M., William T. Neese, and R. Edward Bashaw (1997), "Faculty Perceptions of Marketing Journals, "Journal of Marketing Education, 19 (Spring), 37-52.

Hunt, Shelby D. (1983), Marketing Theory: The Philosophy of Marketing Science, Homewood, IL: Irwin.

Hyman, Michael R., Richard Tansey, and James W. Clark (1994), "Research on Advertising Ethics: Past, Present, and Future," Journal of Advertising, 23 (3), 5-15.

Iowa Guide (2002), "Scholarly Journals in Mass Communication and Related Fields," available at http://fmp2.its.uiowa. edu/iowaguide/ FMPro?-db=iowaguide_.fp5&-lay=plain&format=search.html&-view (accessed May 25, 2005).

Kelley-Milburn, Deborah, and Michael A. Milburn (1995), "Cyberpsych: Resources for Psychologists on the Internet," Psychological Science, 6 (July), 203-212.

Kim, Sung T., and David Weaver (2002), "Communication Research About the Internet: A Thematic Meta-Analysis," New Media and Society, 4 (4), 518-539.

Mason, J. Barry (1995), "Marketing Education in the 1990's: A Dean's Retrospective and Prospective View," Marketing Education Review, 5 (Spring), 1-8.

MediaDailyNews (2004), "Internet Users Would Rather Give Up TV Than Web" (September 22), available at www .mediapost.com/(accessed November 10, 2004).

Miller, Steven E. (1996), Civilizing Cyberspace: Policy, Power, and the Information Superhighway, New York: ACM Press.

Mohammed, Rafi A., Robert J. Fisher, Bernard J. Jaworski, and Aileen M. Cahill (2001), Internet Marketing: Building Advantage in a Networked Economy, Boston: McGraw-Hill/Irwin.

Moore, Laurence J., and Bernard W. Taylor III (1980), "A Study of Institutional Publications in Business-Related Academic Journals, 1972-1978," Quarterly Review of Economics and Business, 20 (Spring), 87-97.

Morris, Jon D., Chongmoo Woo, and Chang-Hoan Cho (2003), "Internet Measures of Advertising Effects: A Global Issue, "Journal of Current Issues and Research in Advertising, 25 (Spring), 25-43.

Muncy, James A. (1991), "The Journal of Advertising: A Twenty Year Appraisal, "Journal of Advertising, 20(4), 1-11.

Musambira, George W. (2000), "Top Convention Paper Productivity in the U.S.: Analysis of National Communication Association (NCA) and International Communication Association (ICA) Awards, 1994- 1998," Communication Education, 49 (3), 284-296.

Nan, Xialin, and Ronald J. Faber (2004), "Advertising Theory: Reconceptualizing the Building Blocks," Marketing Theory, 4 (1/2), 7- 30.

Nielsen/NetRatings (2003), "Ten Out of Twelve Global Markets See Time Spent Online Increase over Past Six Months," available at www.nielsen-netratings.com (accessed November 10, 2004).

NUA (2003), "How Many Online," available at www.nua.com/surveys/ how_many _online/(accessed August 20, 2004).

Pasadeos, Yorgo (1985), "A Bibliometric Study of Advertising Citations, "Journal of Advertising, 14(4), 52-59.

________, Arnold Barban, Huiuk Yi, and Bong-Hyun Kim (1997), "A 30-Year Assessment of the Media Planning Literature," Journal of Advertising, 19 (1), 23-36.

Paul, Mary Jae (2001), "Interactive Disaster Communication on the Internet: A Content Analysis of Sixty-Four Disaster Relief Home Pages," Journalism and Mass Communication Quarterly, 78 (4), 739- 744.

Pavlik, John V. (1998), New Media Technology: Cultural and Commercial Perspectives, 2nd ed., Boston: Allyn and Bacon.

Perloff, Richard M. (1976), "Journalism Research: A Twenty-Year Perspective," Journalism Quarterly, 53 (1), 123-126.

Perreault, William

Source: Journal of Advertising

More News in this Category