Privacy on the Internet: A Community Structure Approach



Nationwide Newspaper Coverage of Privacy on the Internet:

A Community Structure Approach

Sean Collins

Adam Ehrenworth

Robert Hunsicker

John C. Pollock, Ph.D.*

Paper presented at the annual conference of the New Jersey Communication Association, March 31, 2001, Monmouth University, Long Branch, NJ

*John C. Pollock, Ph.D., is an associate professor in the Communication Studies Dept., The College of New Jersey, Ewing, NJ 08628; tel. 609-771-2338; e-mail: pollock@tcnj.edu. Sean Collins (smirfk@), Adam Ehrenworth (ehrenwo2@tcnj.edu), and Robert Hunsicker are all undergraduates at The College of New Jersey.

ABSTRACT

Nationwide Newspaper Coverage of Privacy on the Internet:

A Community Structure Approach

The Internet has had an enormous impact on the lives of computer users around the world in the past decade. In addition to creating an “information superhighway,” it has also opened up a new form of business known as e-commerce. Although e-commerce has simplified individual’s lives with its convenience and accessibility, it has also sparked a heated debate over whether a person’s information being transferred over the Internet is a private item. Proponents of regulation that would protect Internet users’ believe that by governing the transmission of this data individuals would be safeguarded against such computer “evils” as electronic mass-mailings (spamming), telemarketing, credit card theft, and viruses. On the other hand, opponents trust that the e-commerce companies can regulate themselves and that the information sent over the World Wide Web should be accessible to all users. Despite the fact that both sides of this issue have evidence to support their view, the debate continues on a political as well as an economic level.

In this study, primary research was conducted to investigate how newspapers from different cities in the United States reported on Internet Privacy. The study compared hypotheses on different city characteristics and newspaper coverage of Internet Privacy using the "community structure approach," tested in earlier versions in Minnesota by Tichenor, Donohue and Olien (1973, 1980) and elaborated in nationwide studies by Pollock and others (1977, 1978,1994-2000), suggesting that certain demographic structures of a community are systematically linked to newspaper reporting on critical issues.

A national cross-section sample of 21 newspapers was selected from the DIALOG newspaper database, and a sample of 25 articles above 350 words in each newspaper on Internet Privacy were drawn from January 27th of 1996 to November 7th of 2000. The resulting total of 495 articles was analyzed using content analysis. A single score, the Janis-Fadner Coefficient of Imbalance, was calculated to combine attention given to each article as well as reporting direction (favorable, balanced/neutral, or unfavorable). Pearson correlations were used to link different city characteristics to coverage of Internet privacy regulation. The coefficients of imbalance (ranging from .211 to -.137) demonstrated a clear pattern of variation. Through Pearson correlation results, two city characteristics had a significant relationship to newspaper coverage: a measure of “vulnerability" – percent below the poverty line (r= .377; p = .046); and a measure of "privilege" -- percent in technical/professional occupations (r = -.372; p = .048). Factor analysis of city characteristics revealed that "vulnerability", "violated buffer", and "life cycle position" are the three clusters of significant or nearly significant results in the Pearson correlations. Vulnerability and “Life Cycle Position” were further emphasized as noteworthy after regression analysis of the factors found that the characteristics of “families with children under 18” and “percent below the poverty line” in a city, accounted for nearly 30 percent of the variance in coverage of Internet Privacy regulation.

Introduction

The Internet has seen enormous growth since it was originally introduced to the public for use at home in the late 1980’s. As a result, many businesses found it extremely lucrative to move into this new medium as a means of attracting new customers and improving the accessibility of their products and/or services, thus creating one of the Internet’s primary uses: e-commerce. However, as this e-commerce rapidly increased, the question of user privacy came to the forefront. Internet privacy involves the access and availability of personal information over the World Wide Web. This information can include home addresses, phone numbers, credit card data, e-mail addresses, social security numbers, etc. Improper use of this information, if it falls into the wrong hands, breeds fear in the minds of computer users worldwide. This anxiety is the result of a myriad of computer “evils” such as electronic mass-mailings (spamming), telemarketing, credit card theft, and viruses. These abuses of privacy on the Internet are growing daily, becoming more of a threat with each passing moment.

Different opinions exist about ways to prevent these types of problems from occurring. Some believe that these problems can be remedied through the addition of new rules. These regulations would help govern the use and transmission of information over the Net, primarily centering on consumer privacy, the main focus concerning “cookies”. Cookies are defined as small pieces of information sent by a web server to store on an individual computer to be accessed at a later time. It is useful for storing specific types of information such as: logins and passwords, web site preferences, and purchasing information on a commercial site. Access to these bits of data can have staggering repercussions if they fall into the possession of pernicious people.

In contrast, others argue that information on the Internet should not be restricted. For them, the current state of the Internet is acceptable. Access to information, whether personal or commercial, is not subject to government control. They consider it an individual’s right to use the Internet as a tool for gathering information for personal and commercial use. They simply see personal information as a part of the greater whole. To some extent, these people are less concerned about the ramifications of improper retrieval of sensitive material then they are about universal access.

It is highly probable that the Internet privacy issue will receive varying newspaper coverage from city to city based upon demographics and computer usage. The demographic areas likely to be linked to differences in coverage include: income levels, age, education level, and occupation. The computer usage can refer to types of tasks that the Internet is used for, as well as others. Differences in age demographics linked to coverage variations are likely because nearly 80% of children using the Internet will divulge personal information without first consulting with their parents or another adult (Teinowitz, 1998, p. 53). The categories of education level and income could also be associated to variations in newspaper coverage. The more educated the citizens in a city, the more likely newspaper coverage will support Internet privacy regulation, since those who are well educated are more likely to be aware of the Internet’s uses as well as its risks.

Similar presumptions can be made about an individual’s income; the more money and possessions someone has, the greater their concern for privacy, because they simply have more to lose. As a result, newspaper coverage in cities with high-income levels may reveal variations that reflect the relative proportions of those who are “well off.” As far as computer usage is concerned, it is fairly predictable where the variations in newspaper coverage are most likely to occur. The more time someone spends online, the more comfortable that person is with the “system”, therefore he or she could tend to be less threatened by the idea of personal information being accessible to others.

In summary, Internet privacy is a significant issue because the entire nation as a whole is rapidly becoming increasingly computer dependent. With this dependency in mind, the idea that information is available, and at times easily accessible, fosters a two-sided debate over whether regulation is necessary to control the internet’s misuse. Some are threatened by the prospect of invasion of their lives at an information level. They are worried that their personal information is accessible to millions. However, other segments of the population are less concerned with these issues. They simply see the Internet as a valuable tool for exploring or researching different ideas, and in order for this tool to function properly; they believe regulation has to be kept to a minimum.

A Review of Communication Literature

As an important contemporary information highway, the Internet is missing essential features of reliability, functionality, confidentiality, and integrity, and it is also threatened by various security attacks (Fischer-Hubner, 1998). If the proper safeguards and programs are not in place, many risks are associated with providing a site with personal information. These risks have caused a split in the computer community, spawning a heated debate among its users.

The main focus seems to be whether or not regulation is necessary for the Internet to truly be considered a safe medium. It seems evident that this issue has reached a pinnacle of mass media coverage. However, upon investigation of various communication journals, it becomes readily apparent that there is limited concern for the issue of Internet privacy in the scholarly communication field. While numerous articles focus on the Internet itself, few tend to include the issue of personal privacy linked to Internet use. A search of the keywords “internet and privacy”, using the online index EbscoHost, and scanning various communication journals, yielded only two results. Entering the same search criteria using the Communication Institute for Online Scholarship’s Communication Abstracts database, only 5 articles were located. These five articles included topics ranging from Internet privacy in a “global information society” (Fischer-Hubner, 1998) to the “embedding of identity issues in a more commercial Internet”, but none linked Internet privacy and media.

Broadening the search parameters on Communication Abstracts to include all articles with the word “Internet” was useful in gauging how prevalent the topic actually is. This strategy returned a total of 204 articles, however the majority failed to cover any aspect of Internet privacy or the security risks of online data transfer. For example, an article from the Journal of Broadcasting and Electronic Media discusses factors that affect Internet use (Papacharissi, Zizi, and Rubin, 2000), yet it fails to concentrate on the privacy issue. In addition to this, even fewer focus on the Internet and links to newspaper coverage; although there are a select number that make these associations.

One such article was a survey performed by Stempel, Hargrove, and Bernt (2000) dealing with levels of Internet use and its association with a decline in television audiences and newspaper circulation. This article looks for a correlation between Internet usage and more traditional media; however, it does not mention the element of Internet privacy.

A particular article that does in fact deal with Internet privacy (Lyon, 1998) explores different aspects of Internet surveillance such as: employee monitoring, policing and security, and marketing. It focuses on some technologies associated with Internet privacy including “Cookies” and “Spiders”. While centering on privacy issues on the Net, this article lacks information regarding the relation of privacy to media coverage of this issue. Another journal entry that was uncovered, from Communication Abstracts spoke of an article by N.W. Allard (1999) discussing the framework for national and international policies relating to privacy on the Internet. Once again this is an example of an article that relates to Internet privacy but does not tie in the aspect of media coverage. The journal publication that most closely resembles the proposed research comes from Journalism & Mass Communication Quarterly (Johnson, Braima, Sothirajah, 1999). This article studies “the extent to which heavy users of the Internet and other non-traditional media differ from heavy users of traditional media in their knowledge of the issue stances of Bill Clinton and Bob Dole” in the 1996 Presidential Campaign (p. 99). Although this article touches upon the issue of the Internet and its effect on other media coverage, it is missing the component of personal privacy on the Net. Based on this research, a noticeable pattern in the communication field emerges: Internet privacy is an important and critical issue. However, there is still inadequate interest among communication scholars regarding Internet privacy and its links to media coverage.

To further explore literature on media coverage of Internet privacy, several journals were examined across various fields including: business, marketing, law, technology, and government. As expected, countless articles dealt with the specific topic of Internet privacy. A second search on EbscoHost, reading beyond communications journals, with the keyword “internet”, returned over 5000 hits. A more refined search, “Internet and privacy”, displayed 62 hits. Another search on the Wilson Web Database, searching the spectrum of peer-reviewed literature, using the key words “internet and privacy”, yielded over 100 articles. While the majority of these pieces did not take into account the aspect of media coverage of the issue, the prevalence of this topic confirms the need for Internet privacy to be addressed more extensively.

From a business/marketing standpoint, this issue seems clearly pertinent. A case study in the Journal of Interactive Marketing concerns how the Federal Trade Commission (FTC) alleged GeoCities had “engaged in unfair or deceptive practices under Section 5A of the Federal Trade Commission Act” by disclosing customers’ personal information to advertisers and others (Roberts, 2000, p. 70). In a related article, the famous book retailer “revamped its privacy policy, eliminating users’ ability to block the ‘e-tailer’ from sharing data about their purchases and browsing patterns with others.” (Borrus, 2000, p. 54).

In the area of government, it seems as if there is as much weight given to the issue of Internet privacy as there is in other fields. Presently, the FTC is the governmental body responsible for attempting to regulate the Internet. The articles in question mainly focus on the legislation and regulations in progress for the Internet today. One such article, from Congressional Quarterly Weekly, (Ota, 2000) discusses the FTC’s proposal for legislation that would require online firms to protect the security of any online information they gathered. Another article dealing with these regulations comes from The Economist, noting that many individual states have already passed or in the process of passing privacy legislation, alarming many Internet companies “who fear they will have to deal with different standards set in each different state” (May 2000, p. 65).

In the field of technology, for both the Internet specifically and for computers, there were abundant articles on Internet privacy. They ranged from guidelines established by the Clinton administration (IEE Spectrum, 1999) to the development of security methods on the web (Rubin & Geer, 1998). For example, a journal article in the IEEE Technology & Society Magazine discussed the significance of a specific kind of technology developed for the sole purpose of Internet privacy known as the “key escrow encryption system” (Davis, 2000). A majority of the articles uncovered in the technological journals explained the risks and possible consequences directly associated with personal privacy on the Net. Two such articles were discovered in the Communications of the ACM. Both gave in-depth explanations of how disclosing personal information affects every type of Internet user, from students in school (Weinstein & Neumann, 2000) to online shoppers (McGinity, 2000). These articles reinforce the importance of Internet privacy in today’s technological age.

After careful examination of journals across a wide spectrum of subjects, starting with communication and branching outward into other fields, it is evident that more research is needed on this issue, more specifically, research concerning newspaper coverage of Internet privacy and its link to city demographics. In particular, a community structure approach for this issue would be beneficial not only to the communication field but also to any field affected by Internet privacy.

Community Structure Approach: A National Focus

The community structure approach is defined as “a form of quantitative content analysis that focuses on the ways in which key characteristics of communities (such as cities) are related to the content coverage of newspapers in those communities” (Frey, Botan, and Kreps, 2000, p. 238). In using this approach to analyze the topic of Internet privacy, this study attempts to look at how society affects newspaper coverage. The community structure approach suggests that variations in city characteristics can be systematically linked to variations in newspaper coverage of prominent issues. (For a description of the community structure approach and its uses, see Frey, Botan & Kreps, 2000, pp. 238-239.) Since Internet privacy has been widely discussed issue, it is reasonable to assume that newspapers will have ample coverage of the topic. In turn, the differences in coverage can be viewed from an analytical perspective, focusing on how these differences are influenced by variations city characteristics and demographics.

The community structure approach builds on the work of Tichenor, Donohue, and Olien (1973, 1980), Demers (1996a, 1996b) in Minnesota and expanded by Pollock and others throughout the nation (1977, 1978, 1995-2000). Today, this approach incorporates many propositions that have surfaced from previous studies concerning city characteristics and the effects on newspaper coverage of political and social change (Pollock and Yulis, 1999). One study conducted by Pollock and Killeen analyzing the Clarence Thomas and Anita Hill hearings found that “the higher the percentage of city residents who are well educated (at least having a college degree), or work in professional occupations, the more favorable the reporting on Hill” (Pollock & Killeen, 1995). Also, another study based on the community structure approach compared city characteristics and newspaper coverage of Magic Johnson’s HIV positive announcement (Awrachow, 1994), it was discovered that the higher the percentages of individuals in a city with college degrees, in professional occupations, or enrolled in health and fitness clubs in that city, the less favorable the coverage was towards Magic Johnson. These results emphasize the importance of the community structure approach in studying media coverage of important national events.

The community structure perspective also has a signifcant focus on the “antecedents” of newspaper content, a topic that has, in the past, received relatively sparse attention in the scholarly communication studies or journalism literatures (Riffe, Fico & Lacy, 1998, pp. 8-10). Newspapers offer an interesting and significant medium because newspaper reporting is not always objective, often reflecting a particular narrative "frame." (Pollock, Dudzak, et al, 2000). Instead, newspapers serve at least two different purposes. They are, first and foremost, a money oriented business, requiring the marketing and selling of a product for profit (Pollock, Awrachow, and Kuntz, 1994). In addition, newspapers serve as community organizations that announce, distribute, negotiate, and discuss community concerns over specific issues (Pollock and Killeen, 1995; Pollock, Dudzak, et al, 2000, Pollock, Awrachow and Kuntz, 1994; Tichenor, Donohue, and Olien, 1980).

Scholarly literature discussing the community structure approach has suggested that mainstream mass media, such as television, radio and newspapers, can be a means of social control for powerful institutions and value systems (Demers & Viswanath, 1999, p. 419). However, political and social interest groups, that have exercised substantial power through their public involvement, have played important roles in impacting the outcomes of highly controversial issues. These influential “stakeholders” include such groups as the elderly, blue-collar workers, and minorities (Demers and Viswanath, 1999, p. 419). By recognizing the many factors that influence change on critical issues and by incorporating the community structure approach, this study maps the relation between local city characteristics and media coverage of Internet privacy regulation.

By understanding the community structure approach it becomes clear as to the impact society has on media coverage. This unprecedented study will incorporate the community structure approach and attempt to show a correlation between city characteristics and the type of newspaper coverage aimed at the topic of Internet privacy and the need for government regulation.

Hypotheses

Previous research using the community structure approach was reviewed to explore hypotheses. The following twelve hypotheses concerning Internet privacy can be organized into five distinct clusters: buffer, violated buffer, access, lifecycle positions, vulnerability, and stakeholders.

Buffer

The “Buffer" hypothesis, developed by Pollock and colleagues, is described in Investigating Communication, the communication methods textbook co-authored by Frey, Botan and Kreps as follows: "The 'buffer hypothesis' expects that, the higher the proportion of privileged groups in a community, the greater the favorable media coverage of those making human rights claims.” The Internet is and will continue to become more heavily accessed by the world as a whole. More specifically, the information superhighway is being introduced to people at a much earlier age. Today, more than ever before, students in grammar, middle, and secondary schools are all subjected to computers, as well as the Internet, as part of the school curriculum. In fact some 95% of public schools are wired currently to the Internet, as compared to only 35% in 1994 (Symonds, 2000). The human rights component of this issue comes from new exposure to Internet access and the capabilities of free information gathering.

It can be assumed that those with sixteen or more years of education are likely to support rights claims and to oppose acts that violate human rights. This is consistent with prior community structure studies determining that the higher the education level in a city (in the United States), the more likely major newspapers were to oppose China's bid for the 2000 Olympics, based on human their human rights record (Pollock, Kreuer & Ouano, 1997). This also concerns support for the “rights” of someone many regarded as harshly treated: Anita Hill during the Clarence Thomas Senate hearings (Pollock & Killeen, 1995). To the extent that Internet privacy is related to an individual’s right for confidentiality of personal information, cities with higher privilege distribution are predicted to correlate with more favorable coverage of Internet regulations.

The Internet has also found its way in the corporate world. Virtually every company today is in some way dependent on the Internet for communication, sales, advertising, etc. Those involved in professional and technical careers, for the most part, are competent with the Internet and use it on a daily basis. As a result, these individuals are more comfortable with the Internet and how it operates. Therefore, they are less threatened by some of the risks posed by online transmission of personal information. Those in more “privileged” position understand how those who are less educated may not be privy to the tricks or tactics that Internet companies use to gather personal information. They associate being favorable to government regulation of Internet privacy as a safeguard for the less informed, and as a protection of their rights. As a result, it is reasonable to assume the following:

H1 The higher the percentage of people with four years of college education in a city, the more favorable the coverage of Internet privacy regulation. (Lifestyle Market Analyst 2000)

H2 HhhThe higher the percentage of people with professional or technical occupations, the more favorable the coverage of Internet privacy regulation. (Lifestyle Market Analyst 2000)

H3 The higher the percentage of families with an annual income greater then $100,000, the more favorable the coverage of Internet privacy regulation. (Lifestyle Market Analyst 2000)

Violated Buffer

The 'violated buffer hypothesis', developed by Pollock and colleagues, expects that, the higher the proportion of privileged groups in a community, the less favorable the reporting on either biological threats or threats to an established way of life." (Frey, Botan & Kreps, 2000, p. 329) Taking into account indicators of privilege such as percentage of the population with professional/technical occupational status, the proportion of city residents with median incomes of $100,000 or more, it is possible to uncover a relationship between these city characteristics and newspaper coverage of Internet privacy. The privileged, in the case of Internet privacy, consider government of control of Internet privacy a potential threat to their personal information. While the government contends to be securing their personal information those who are more educated may perceive the government as having too much control in this area. Consequently, this supports a “violated buffer” hypothesis because privileged groups are at a greater risk, having more to lose, if personal information falls into unwanted hands. Therefore, it is believed that high proportions of professionals, as well as families with incomes over $100,000 in a city, will be linked to relatively negative coverage of Internet privacy regulation. Research was conducted on media coverage of Magic Johnson’s HIV announcement, in which his announcement created a new perceived vulnerability for Americans with high economic status (Pollock, Awarchow & Kuntz, 1994).

Research concerning nationwide newspaper coverage and the Internet confirmed a similar "violated buffer" hypothesis, using education level as a measure of privilege (Pollock & Montero 1998). It is expected that the educationally privileged have the same relationship to coverage of Internet privacy as the economically and professionally privileged.

Internet-based merchandising has emerged in the last few years and continues to gain popularity. The concept of being able to order essentially any item from practically anywhere, from the comfort of your home, has revolutionized the way people purchase products. This new technology brings with it, various risks concerning the access to a customer’s personal information (i.e. credit card data, social security numbers, address, phone numbers, etc.) Related to the violated buffer hypothesis, worries of Internet consumers surround the threat of improper use and distribution of this information. Those who can afford not only a computer, but also access to the World Wide Web through an Internet provider can be classified as middle to upper class. These Internet consumers are less timid in sending out personal information for online transactions. Therefore, cities with high levels of families with incomes of a higher range are likely to have a greater number of computer users, more specifically Internet users. Based upon all of these elements it can be predicted that:

H4 The higher the percentage of families with incomes greater than $100,000 in a city, the less favorable the coverage of Internet privacy regulation. (Lifestyle Market Analyst 2000)

H5 The higher the percentage of people with four years of college education in a city, the less favorable the coverage of Internet privacy regulation. (Lifestyle Market Analyst 2000)

H6 The higher the median income in a city, the less favorable the coverage of Internet privacy regulation. (Lifestyle Market Analyst 2000)

H7 HhhThe higher the percentage of people with professional or technical occupations, the less favorable the coverage of Internet privacy regulation. (Lifestyle Market Analyst 2000)

Access to Media

Cable TV

Previous nationwide newspaper studies of critical events have found a significant correlation between a community’s level of access to media and reporting accommodating social or political change. Studies of nationwide newspaper coverage of Ryan White, the teenage hemophiliac with HIV/AIDS, the legalization of same-sex marriage and legalization of physician-assisted suicide have all demonstrated a significant relationship between reporting accommodating social change and the number of AM and FM radio stations in a city (Pollock, O’Neill, et. al. 1995; Pollock & Dantas 1998; Pollock & Yulis, 1999).

Television in particular, as media source, provides its viewers with an array of information and viewpoints, especially on topics like new technology. More specifically, cable stations offer a wide range of technology information because of increasingly high number of these types of stations. Atkin and LaRose (1991) noted that over sixty percent of cable stations have at least one community access channel. In addition, they assert that such community access channels attract one in six regular cable subscribers (Atkin and LaRose 1991). Another study conducted by Baldwin, Barrett and Bates (1992) reported that cable subscribers are more inclined to view and rely on their cable news programs rather then their local news. Based on this plurality of viewpoints given under cable media it would reinforce the values within the community and strengthen social norms making people more aware of the risks associated with the transfer of personal information via the Internet. With this in mind the assumption can be made that:

H8 The higher percentage of households with cable stations in a city, the more favorable the coverage of Internet privacy. (Lifestyle Market Analyst 1999)

AM/FM Radio

According, to the Federal Communications Commission talk radio is the predominant format of AM stations. A study by Owen (1997), analyzed talk radio and audience evaluations of the presidency, noting that individuals who actively listen to the radio tend to exhibit negative perceptions of President Clinton. Similarly, Pfau et al. (1998) assessed that talk radio depicts political and government institutions, such as the armed forces, negatively. As a result, listeners demonstrated lower confidence in these government organizations. Previous research studying the relationship between AM radio and newspaper coverage, such as nationwide newspaper coverage of post-Columbine gun control legislation and the Patient’s Bill of Rights, have found a significant correlation between the number of AM stations and reporting resisting social and political change on these critical issues (Pollock, Shellenberger & Fagerty 2000; Pollock & Castillo 2000).

Although early community structure research suggests that AM and FM radio were linked to coverage accommodating political and social change, recent analyses suggest a divergence of interests. The format of FM radio, which is viewed by a far younger and more liberal audience, is still consistently linked to a desire for social change. AM radio stations, however, have been increasing connected to news coverage that impedes political and social change. With talk radio as the prominent format of AM stations, it is easy to understand that these audiences will adopt a more conservative viewpoint with regards to any government issue, such as control of over the Internet. In recent studies this divergence between AM and FM stations has become very apparent, in which AM radio stations were linked to unfavorable coverage of the privatization of social security reform, the Patients’ Bill of Rights and post-Columbine gun control legislation (Pollock, Tanner & Delbene 2000; Pollock Castillo & Griffiths 2000; Pollock, Shellenberger & Fagerty 2000). With this pattern of AM and FM stations taken into consideration it is likely:

H9 The higher the number of AM radio stations in a city, the less favorable the coverage of Internet Privacy regulation (Gale’s Directory).

H10 The higher the number of FM radio stations in a city, the more favorable the coverage of the Internet Privacy regulation (Gale’s Directory).

Lifecycle Position

Individuals that have established a certain position in life have different values, beliefs, and priorities. For example, families with children have alternate responsibilities and concerns as compared to those without children. These include providing financial, emotional, and psychological support. The most important facet of this is the protection of their children from outside sources. It is natural for parents to be concerned about the well being of their offspring. This is especially true when dealing with information exchange over the Internet. Children are far more willing to divulge sensitive or personal information online (on consumer website, in chat rooms, etc.). Parents likely are protective of this and would welcome legislation that would attempt to protect the privacy of their own personal information.

H11 The higher the percentage of families with children in a city, the more favorable the coverage of Internet Privacy regulation. (Lifestyle Market Analyst 2000)

H12 The higher the percentage of single parent families in a city the more favorable the coverage of Internet Privacy regulation. (Lifestyle Market Analyst 2000)

Another factor included in this cluster is the elderly, more specifically those over the age of 75. A study of nationwide newspaper coverage of Dr. Jack Kevorkian between 1990 and 1993 found that the greater the proportion of individuals over the age of 75, the more supportive the news coverage of Dr. Kevorkian’s efforts to publicize the physical concerns of older citizens (Pollock, Cannaughton, et. al., 1996). Those in this age category are likely to have less Internet savvy, and more concerned about their loss of privacy on the Net. One study's showed a correlation between this age group and less favorable newspaper reporting on legalization of physician-assisted euthanasia. This city characteristic was more powerfully linked than any other demographic to reporting on the physician-assisted suicide topic. (Pollock & Yulis, 1999).

H13 The higher the percentage of people over the age of 75 in a city the more favorable the coverage of Internet Privacy regulation. (Lifestyle Market Analyst 2000)

Vulnerability

It is reasonable to expect that those in a community who are below the poverty level or unemployed will have positive views towards legislation on Internet privacy. Nationwide community structure approach studies exploring coverage of the U.S. Supreme Court decision on abortion, as well as coverage of the Crown Heights (Brooklyn, NY) ethnic conflicts revealed that the higher the percentage below the poverty level, the more favorable the coverage of both of these critical issues. (Pollock, Robinson & Murray, 1978; Pollock & Robinson, 1977; Pollock & Whitney, 1997). It is probable that a positive association exists between the proportion of those in a city who are unemployed or below the poverty level and favorable coverage concerning the Internet Privacy.

H14 The greater the proportion of those in a city below the poverty level, the more favorable the coverage of Internet Privacy regulation will be (County and City Extra)

H15 The greater the proportion of those unemployed in a city, the more favorable the coverage of the Internet Privacy regulation will be (2000 County and City Extra)

Stakeholders

In regards to stakeholders in a critical issue, Pollock and Dantas (1998) revealed a relationship between the number of the organizations marketing to the gay population in cities and relatively favorable coverage of the legalization of same-sex marriage. This relates to privacy on the Internet, because most people who own computers are aware of the advantages and disadvantages, the risks and dangers, of today’s computer technology. Computers sold back as far as the mid-nineties have come preinstalled with modems and trials to online subscription services, allowing people direct access to the Internet (Pollock & Montero, 1998). With this technology, the door is opened for personal information to be “broadcast” across the World-Wide-Web. Therefore, those entering this information, for the most part, can be assumed to have some Internet capabilities. As such, the majority of PC users have a vested interest in the protection of Internet Privacy.

H16 The higher the number of PC users in the city, the more favorable the coverage of Internet Privacy regulation. (Lifestyle Market Analyst 2000)

The political aspect of this issue is strongly partisan. The Democrats, particularly President Clinton have primarily sided with Internet companies self-regulating their privacy policies. Republicans on the other hand tend to agree with the FTC’s assessment that “self-regulation alone [is] not enough to protect consumers’ online privacy. (Rosen, 2000, p. 3)They associate with four distinct principles when discussing the needs for privacy legislation, they are: notice, choice, access, and security. Republicans typically view legislation as the most logical remedy for Internet privacy issues. As a result, the following presumptions can be made:

H17 The higher the percentage of Democrats in the city, the less favorable the coverage of Internet Privacy regulation. (Lifestyle Market Analyst 2000)

H18 The higher the percentage of Republicans in the city, the more favorable the coverage of Internet Privacy regulation. (Lifestyle Market Analyst 2000)

Methodology

A total of twenty-one papers were examined over a five-year period ranging from January 1996 to present day coverage of Internet privacy regulation. An average of 23.5 of the longest articles from each time period was sampled. (Up to 25 of the longest articles in each paper were allowed.) All the articles that were included in the sample mentioned the words “privacy” and “Internet” or any synonym such as the World-Wide-Web, the Net, or Information Superhighway, as the subject of discussion. Any article over 350 words that dealt with Internet privacy was included in the sample. In all, 495 articles were read and analyzed. If a paper’s article count exceeded twenty-five, then a random sample of the articles was taken, using a program* developed by Paul Houle on , to achieve the desired amount.

For this sample, several newspapers were selected from each of the nation’s major regions. These newspapers were selected for their geographic dispersion and for their availability on either the DIALOG Computer Information Program. Papers such as the New York Times and Washington Post were not included in the sample because their national status, both politically and financially, may have lead these papers to express views well beyond their city boundaries and demographics.

To compare differences in coverage, each of the newspaper articles were measured and coded for two types of information. The first was an “attention” or “display” score, ranging from 0 to 16 points. This sum represented the total scores for each of the following four factors: the placement of the article, the headline column width, the article length, and the inclusion of photographs or other graphics (charts, illustrations, etc.). Newspapers with higher attention scores presented coverage of Internet Privacy more prominently than did those that received lower attention scores.

The second type of information was a “content” score that assigned a category that represented the article’s content direction on Internet Privacy regulation. Each article’s content was evaluated as being “favorable”, “unfavorable”, or “balanced neutral” towards issues concerning privacy on the Internet:

• Coverage that was considered “favorable” toward Internet Privacy supported it by reporting on the existing dangers, which currently threaten privacy on the Internet. These articles call for government legislation to be established to protect privacy on the web, as well as expressing a general fear of private information falling into the wrong hands.

• Coverage “unfavorable” toward Internet Privacy did not support the idea of government regulations being placed upon the Internet. The tone of these articles suggested that the current measures that the Internet takes to govern control of information are satisfactory, and information that is currently being transacted over the web is relatively well protected.

• Coverage that was considered “balanced” or “neutral” toward Internet Privacy gave basic information concerning the issue, emphasizing neither the positive or negative attributes of the Internet and the privacy concerns that are attached to it. These articles left it up to the reader to decide whether or not restrictions should placed upon the Internet to ensure privacy.

After each article was coded for both attention and content, the articles were sorted into the three content categories for each newspaper, and the Janis-Fadner “coefficient of balance” was calculated for each newspaper. (Janis & Fadner, 1965: 153 – 160) This resulted in a single score for each newspaper, that fell between + 1 or -1. The scores above zero indicated relatively favorable coverage of Internet privacy regulation, while those below indicated relatively unfavorable coverage. (see Table 1)

Articles using the Janis-Fadner Coefficient of Imbalance in Communication research have been accepted for publication in such journals as Comparative Politics, Society, Journalism Quarterly, Newspaper Research Journal and the edited, refereed collection Communication Yearbook. (See respectively Hurwitz, Green & Segal, 1976; Pollock & Robinson, 1977; Pollock, Murray & Robinson, 1978; Pollock, et. al., forthcoming; and Pollock & Guidette, 1980) Similarly, papers using the Janis-Fadner Coefficient of Imbalance have been accepted for presentation at professional conferences, examining topics such as coverage of political refugees as well as comparative newspaper reporting on NAFTA debates and decision; on the Crown Heights disturbances in Brooklyn, NY; and on the Anita Hill’s testimony in the Clarence Thomas hearings. (Respectively, Pollock, Shier and Slattery, 1995; Pollock and Whitney, 1997; and Pollock and Killeen, 1995)

After the collected articles were assigned their directional scores, three researchers coded a systematic sub-sample of one-third of the articles to calculate the Holsti’s Coefficient of Intercoder Reliability. The number of articles agreed upon (465), was divided by overall article count, to yield a coefficient of 93.9%.

Single-Score Content Analysis: Calculating a Coefficient of Imbalance

Definitions

f = sum of the attention scores coded “favorable”

u = sum of the attention scores coded “unfavorable”

n = sum of the attention scores coded “neutral/balanced”

r = f + u + n

If f > u (or if the sum of the “favorable” attention scores is greater than the sum of the “unfavorable” scores), then use the following formula:

Coefficient of Favorable Imbalance

C(f) = (f2 – fu)

----------

r2

Answer lies between 0 and + 1

If f < u (or if the sum of the “favorable” attention scores is less than the sum of the “unfavorable” attention scores), then use the following formula:

Coefficient of Unfavorable Imbalance

C(u) = (fu – u2)

----------

r2

Answer lies between 0 and – 1

By using the single score available for each newspaper, it became possible to rank each newspaper based upon their Coefficients of Imbalance, ranging from most to least favorable towards Internet privacy regulation. It also became possible to associate newspaper “imbalance” rankings with their corresponding city rankings, permitting the researchers to test several hypotheses. Correlations were performed to test the relationship between distinct city characteristics, and variations in coverage of Internet privacy regulation.

Results

Newspaper scores and rankings:

The Coefficients of Imbalance ranged from .211 to -.137. Of the twenty-one newspapers studied, the majority of the coverage on Internet privacy regulation was highly balanced /neutral. Only a handful of the cities could be considered significantly favorable/unfavorable. Both the Boston Globe and the Cincinnati Post showed slightly positive coverage of the issue, while the San Francisco Chronicle had negative coverage of this topic. The following is a list of cities and their corresponding Coefficients of Imbalance, (Table 1) ranked from highest to lowest:

Table 1

Newspapers and Corresponding Coefficients of Imbalance

|Newspaper |Coefficients of Imbalance |

|Cincinnati Post |0.211 |

|Boston Globe |0.201 |

|St. Louis Post-Dispatch |0.149 |

|(Memphis) Commercial Appeal |0.101 |

|(Portland) Oregonian |0.080 |

|Detroit Free Press |0.072 |

|San Jose Mercury News |0.060 |

|Tallahassee Democrat |0.060 |

|Milwaukee Journal Sentinel |0.048 |

|Kansas City Star |0.046 |

|Pittsburgh Post-Gazette |0.041 |

|(Albany) Times-Union |0.037 |

|Denver Post |0.022 |

|Phoenix Gazette |0.017 |

|Seattle Times |0.014 |

|Charlotte Observer |0.010 |

|Fresno Bee |0.003 |

|Atlanta Journal |0.000 |

|Lexington Herald-Leader |-0.029 |

|Philadelphia Daily News |-0.039 |

|San Francisco Chronicle |-0.137 |

Twenty-one city characteristics were used as independent variables to assess the association between city structure and newspaper coverage of Internet privacy regulation. Two city characteristics, poverty level and percentage of technical/professional occupations, had a significant relationship (.05 or better) to newspaper coverage. Four additional city characteristics had a directional relationship, significant at the .10 level or better: Families with children 8 to 10 years old, families with children 11-12 years old, families with children 13-15 years old, and families with an annual income over $100,000. (See Table 2)

Table 2

Significant/Directional Pearson Correlations

|Hypothesis |Correlation |Probability |

|Poverty Level |0.377 |0.046* |

|Technical/Professional |-0.372 |0.048* |

|Families w/ Children 8-10 |0.334 |0.070 |

|Families w/ Children 11-12 |0.328 |0.073 |

|Income Over $100,000 |-0.321 |0.078 |

|Families w/ Children 13-15 |0.295 |0.097 |

|Hispanic |-0.236 |0.152 |

|Age (75 and over) |0.227 |0.161 |

|AM Radio |-0.210 |0.181 |

|Families w/ Children 5-7 |0.198 |0.195 |

|Single Parents |0.181 |0.210 |

|Crime Index |0.177 |0.221 |

|Republican |0.174 |0.225 |

|Computer Users |-0.168 |0.233 |

|Families w/ Any Children |0.161 |0.243 |

|Children 5-17 |0.155 |0.252 |

|College Educated |-0.131 |0.286 |

|Democrat |-0.102 |0.330 |

|African Americans |0.100 |0.333 |

|Households w/ Cable |-0.093 |0.344 |

|Women in the Workforce |-0.068 |0.384 |

|FM Radio |0.042 |0.429 |

|Median Income |0.013 |0.477 |

* Significant at ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download