How public opinion polls define and circumscribe online privacy by Kim Bartel SheehanThe advent of new communications technologies and the integration of such technologies into individuals’ lives have resulted in major changes to society. Responding to such privacy concerns is of key interest to legislators, policy–makers, and business leaders as these groups seek to balance consumer privacy needs with the realities of this new society. These groups, and others, use public opinion polls and surveys to measure the current climate of opinion among citizens. This study examines the language of 43 opinion polls and surveys dealing with privacy and the Internet to understand how these polls define and assess online privacy. Results suggest that polls treat the complex construction of privacy in an overly simplistic way. Additionally, pollsters present many poll questions in a way that may lead survey respondents to express stronger negative feelings about privacy than really exist.
Contents
Introduction
The role of privacy and privacy polls in society
Results of analysis
Assessment of privacy polls
Conclusions
Introduction
The advent of new communications technologies and the integration of such technologies into individuals’ lives have resulted in major changes to society. While such technologies allow individuals to become more closely connected with family, friends and colleagues, they may also affect individuals’ concerns about their personal privacy. Responding to such privacy concerns is of key interest to legislators, policy makers, and business leaders as these groups seek to balance consumer privacy needs with the realities of this new society.
These groups, and others, use public opinion polls and surveys to measure the current climate of opinion among citizens. According to Gandy (2003), "references to public opinion have been used to frame the public as concerned, differentiated and, most recently, as willing to negotiate their privacy demands." This study will further explore this suggestion by examining the language of 43 opinion polls and surveys dealing with privacy and the Internet to understand how these polls define and assess online privacy.
The role of privacy and privacy polls in society
Online privacy is of interest to many different stakeholders in the Internet society. The interactive nature of the Internet allows for a two–way communication between a variety of pairs of individuals: for example, politicians and constituents, teachers and students, doctors and patients, and marketers and consumers. A by–product of such exchanges can be an electronic 'paper trail' of information about both parties. This information can be used for a number of different purposes: to craft Web pages that are easy to navigate, to provide information in an easy way; to develop new products; and to craft informative and/or persuasive online messages.
Many online users are aware of the myriad ways online entities collect and use data, but many are not. As a result, online privacy has become a hot topic of debate.
Public opinion polls often serve as catalysts for such debates. Polls survey large groups of people and report information through syndicated releases covered by news media on a regular basis. National samples in the United States range from 1,500 to 3,000 interviews, and in general the quota method of selecting a sample is used (Sheatsley, 2000). Using the quota method, the sample is representative of proportions in the population about specific characteristics seen as important to the pollsters, such as someone’s age, gender, and ethnicity.
In addition to alerting people about how fellow Americans feel about important issues of the day, polls are used by business and industry to measure effects of marketing and other business decisions, to track and assess a business’ public image, and to ascertain consumer opinions about what is important (Sheatsley, 2000). Sociologists and political scientists use poll data to track social trends and analyze how individuals make social decisions. Political scientists and candidate pollsters track opinion on candidates and issues. Elected officials use the data to analyze and assess public policy and determine the types of laws that may be necessary to implement in our society. They also use the polls for guidance in their efforts to bring public opinion in line with policies they support (Gandy, 2003).
Many believe polling is fundamental to a democracy, as polls provide a way for citizens to have a say in how they are governed (Kovach, 1990). Numerous meta–analyses suggest a strong and resilient link between public and policy (Stimson, et al., 1995). Stimson et al. suggest opinion change was congruent with policy change for about two–thirds of issues covered in surveys. This suggests that increases in levels of concern to a topic lead eventually to public policy changes.
In a study of public opinion polls and actual policy outcomes on over 500 issues, Monroe found policy outcomes to be "consistent with the preferences of public majorities on 55 percent of the cases" [ 1]. He also found opinion and policy consistency was greater on issues with high public salience: during the period Monroe studied (1981–1993), opinion/policy consistency was highest for foreign policy and energy and environment issues. Specifically regarding relations with the Soviet Union during this Cold War period, opinion/policy consistency was 81 percent.
Therefore, poll results create expectations, frame political discourse, and in the absence of strong and sustained reporting on the facts underlying an issue, polls can and do shape and create opinion.
In regards specifically to privacy, it appears policy change often occurs when the public’s attention has been focused on issues in response to critical events. Often, media coverage can amplify the public’s concern over such (Gandy, 2003). For example, the Video Privacy Protection Act, passed into law in the United States in 1998, protects the privacy of video rentals. The quick passage of this act was due in part to the publicity generated by publication of U.S. Supreme Court nominee Robert Bork’s video rentals (Regan, 1995). More recently, the passage of the USA PATRIOT Act was likely facilitated by the media’s coverage of terrorist threats in post–September 11th society, although it is important to note that this law is seen as threatening, rather than protecting, personal privacy (King, 2001). While media attention may not always take the lead in such events, and while a causal direction cannot be clearly distinguished, it is evident that members of the U.S. Congress pay attention to issues that the public supports (Gandy, 2003).
In 1990, James E. Katz and Annette R. Tassone analyzed public opinion trends regarding privacy and information technology. Polls taken prior to the explosive growth of the Internet indicated Americans thought privacy was important, were increasingly concerned about personal privacy, and anticipated that they would have less privacy in the future. Over a fifteen year period, from 1975–1990, reported concern with privacy stayed at a high level, and increasingly Americans believed participating in consumer society required a loss of privacy.
Katz and Tassone’s work identifies two of the biggest problems with the use of polls to tap into public opinions regarding online privacy. First, these polls define privacy in a simplistic way while privacy has been seen as being a complex issue (Sheehan, 1998). Defining exactly what privacy is, though, is difficult in the online world where context can change with a click of a mouse button. Many debates over online privacy tend to focus on information privacy, which has been defined as a state or condition of limited access to individuals (Schoeman, 1992). In other words, privacy protects individuals from any overreaching control of others. This definition suggests the contextual nature of privacy, since privacy as a state or condition suggests privacy is innately dynamic and can change according to environmental and personal dimensions, as well as societal forces.
The second problem is the framing of the questions in the polls connect privacy to negative emotions. Accurately measuring public opinion is critical given the range of groups using polling data for decision–making. How public opinion is measured, the way the questions are asked and the way results are reported may have an influence over groups that some critics find problematic. Two separate polls conducted at about the same time in the year 2000 examined whether the government should pass laws on Internet privacy. The results of the two polls differed depending on the way the poll questions were worded. The first poll, conducted by Harris in association with Business Week magazine, gave respondents three choices about government involvement and asked which would be best at this stage of the Internet’s development. The choice reading "government should pass laws now for how personal information can be collected and used on the Internet" was selected by 57 percent of respondents, while choices with less government involvement (not take action now or recommend privacy standards) were selected by far fewer people (15 percent and 21 percent, respectively). The second poll, sponsored by the Democratic Leadership Council (2000), asked respondents which statement was closer to their view on the best way to protect privacy of personal medical and financial records. Almost two–thirds selected the viewpoint saying "give individuals more personal control over who sees their records" while only 29 percent said "pass strong federal restrictions." Thus, it is possible legislative decisions on privacy could vary, based on which poll was assessed by policy–makers.
Much has changed since Katz and Tessone’s original study of privacy and new technology. Specifically, the growth of the Internet allows new forms of interactive communications between disparate individuals who might not have had the opportunity to communicate before. In order to determine how privacy is characterized in public opinion polls, 43 polls — by reputable sources which conducted public opinion polls or surveys over the past ten years — were analyzed (see Table 1). These sources included polling centers such as Harris and Gallup, news media such as ABC and Fox, academic organizations like the Pew Center, and telecommunications companies such as AT&T. Together, these research efforts resulted in 98 questions addressing the issue of online privacy.
Table 1
Summary of Privacy Polls: 1996–2003.
Polling group Year polls done Total questions 1stAmendment Center 2002 2 ABC 2000, 2001 3 ACT 2001 1 ASNE 2001 1 Atlanta Journal Constitution 2002 2 AT&T 1999 14 BW/Harris 2000 4 Center for Democracy and Technology 1997 12 CIO 2001 1 CMOR 2001 1 Consumers Union 2002 1 Council for Excellence in Government 2002, 2003 2 Direct Marketing 1996 1 Forrester 1999, 2001 2 Fox 2000 7 Gallup 2000, 2001 5 GVU 1997, 1998 3 Harris 1994, 1999, 2000, 2001, 2002, 2003 15 IPPSR 2002 1 Markle 2001 1 National Consumer League 1999 New York Times 2002 1 PANCS 2002 1 PC World 2001, 2003 3 Pew 2000, 2001 9 UCLA 2000 1 USA Today 2000 1 Yankee 2001 1 Zogby 2002 1
Initially, two coders analyzed these questions based on two assessments: breadth and direction. Breadth has to do with the scope or generality of the issue being studied and focuses on the question being asked: is the concept provided for assessment specific or generalized? In this case, how the polls defined online privacy were compared to previous definitions of online privacy to examine how well the polls represented current thinking on the nature of privacy.
Direction characterizes the nature of the response to the question, such as whether the respondent is for or against a concept (such as statements about practices that are an invasion of privacy).
Once the poll questions were evaluated, the researcher grouped the questions into clusters to explore how the poll questions describe privacy. The poll questions are segmented into four groups. The first group of questions attempt to broadly define privacy, most often in terms of aspects of control. The second group of questions examines individuals’ privacy sensitivity to different types of data items selected. The third group of questions investigates online activities online users may view as problematic in terms of individuals’ privacy, and the fourth group of questions evaluate protection options.
The results of analysis
Broad definitions: privacy as control
The first group of questions presents broad definitions of privacy for individuals to respond to in terms of privacy in the online context ( Table 2).
While most of these questions defined privacy in terms of some type of control, nine questions presented the problem as simply privacy on the Internet. This question is quite broad in breadth, and directionally this question’s wording tended to be measured in terms of a level of concern, with polls reporting a majority of Americans were concerned. Of the remaining questions, eight defined privacy as control over the collection of information and six defined privacy as control over who has access to information. In most of these cases, respondents viewed their own personal control over the information as very important. Poll wording also indicated respondents saw a lack of control as risky, and were uncomfortable with this lack of control. Four questions defined privacy in terms of information that can be stolen and directionally queried respondents about how worried or how much at risk they felt for such an activity. This type of question set up a possible negative bias for the response, that is, it framed the question in a negative sense (information could be stolen) rather than a positive sense (information is safe). Finally, one question asked about the right of privacy online and queried individuals as to the degree to which this was essential online ( Table 2).
Table 2
Poll Results: Definitions of Privacy; Privacy as Control.
Question Number of
questionsMeasurement Response range Privacy on the Internet 9 Levels of concern 30–50% very concerned; 79–83% concerned Privacy as control over collection of information 8 Degree of importance 69–74% view as extremely important Privacy as control over who gets information 6 Degree of importance, risk, comfort level 75–84% extremely important; 50% uncomfortable over who gets information; 75% see as a risk. Privacy: information that can be stolen 4 Level of risk or worry 43–70% worried/extremely worried; 69% see as risk Right of Privacy 1 Degree to which is essential 81% say essential
Polls and information sensitivity
The second category of questions examined the type of information collected, and included 23 questions (see Table 3). Nine questions asked about the collection of financial records and credit card information, directionally assessing level of concern, with a majority of Americans concerned about such information. Five questions asked about the collection of information found in a standard phone directory (such as name, address and phone number), with polls reporting about half of Americans being very concerned. Three questions asked about health records, reporting a relatively high level of concern. Americans appear most concerned about the collection of Social Security numbers, measured in two questions with having high levels of concern of the collection. Two questions did not ask about a specific category of information, yet queried the public as to the potential for abuse of personal information online. These questions were the only ones framed in an obviously negative way in this group.
Table 3
Poll Results: Infomration Sensitivity and Types of Information Collected.
Question Number of
questionsMeasurement Response range Financial records/credit card information 9 Level of concern 64–84% concerned/very concerned Directory info home/phone 5 Level of concern 51–54% very concerned Health records 3 Level of concern 47–65% concerned/very concerned Personal information 2 Potential for abuse 22–24% concerned SSN 2 Level of concern 75% very concerned Age 1 Level of comfort in providing 31% uncomfortable E–mail address 1 Level of comfort in providing 24% uncomfortable
Activities as threats to privacy
A total of 25 questions asked about specific activities that could lead to threats on individuals’ privacy online (see Table 4). Most of these questions dealt with online commercial entities collecting information from individuals online. Eight of these questions dealt with Web sites tracking information, with responses falling into questions about a level of comfort or a type of violation. Perceptions of information sharing among Web sites was also measured in a variety of ways, assessing levels of concerns, levels of comfort, and whether such an activity was an invasion of privacy. The activity of making a purchase online was presented as a threat to privacy in most polls, with individuals asked to agree or disagree with the statement as well as respond to whether it was a risk. One question asked about cookies, which were also presented as a threat to privacy.
Seven questions asked about government access or monitoring of information, and the majority of these questions asked whether respondents supported or opposed such activities. While the number of polls about online privacy have been decreasing over the years, it is interesting to note that the few polls occurring after September 11 tend to focus on government collection and usage of information rather than marketers’ activities.
Table 4
Poll Results: Activities as Privacy Threats.
Question Number of
questionsMeasurement Response range Government access/monitoring 7 Support or oppose 65–74% oppose Web tracking 8 Level of comfort/violation 95% uncomfortable; 43–67% agree is violation Website sharing information 6 Level of concern/comfort/invasion of privacy 65–89% consider violation of privacy; 50–84% concerned; 92% uncomfortable Online buying is a threat to privacy 3 Level of agreement/risk 41–72% agree/extremely agree; 70% see as a risk Cookies are an invasion of privacy 1 Level of agreement 19% agree
Policy Options
A final group of 27 questions asked about how individuals’ online information should be protected (see Table 5). Three types of questions addressed government involvement. Nine questions measured a level of agreement that government should pass laws to protect online privacy, with a range of agreement from 57 to 87 percent. Six questions addressed whether current laws protect consumers, with a level of disagreement from 38–63 percent. One question asked whether respondents agreed violators of laws should be disciplined, and an overwhelming majority agreed. Three additional types of questions addressed non–legislative ways to address this problem. Four questions assessed opt–in policies, with a high level of support for such policies. Similarly, one statement assessed level of agreement to whether Web sites should disclose policies (resulting in a high level of agreement) and one statement assessed levels of agreement to a statement proposing consumer education was better than law (resulting in a high level of agreement).
Table 5
Poll Results: Policy Options.
Question Number of
questionsMeasurement Response range Government should pass laws to protect online privacy 9 Level of agreement 57–87% agree Current laws protect consumers 6 Level of agreement 38–63% disagree Violators should be disciplines 1 Level of agreement 94% agree Non–legislative options Opt–in 4 Support or oppose 78–88% support Websites should disclose policies 1 Level of agreement 93% agree Consumer education better than law 1 Level of agreement 71% agree
Assessment of privacy polls
Overall, the polls and surveys assessing online privacy appear to treat privacy as a somewhat simplistic concept. Many questions define privacy in highly ambiguous terms, and little can be taken away of any relevant meaning for policy–makers or online marketers and content providers from these simplistic statements. Even when privacy is somewhat defined, rarely is context provided: for example, when consumers are asked about government accessing or monitoring information are they asking about accessing information about possible terrorists, or about one’s self? The context would be likely to result in different answers from many individuals.
Overall, a single question does not seem appropriate to measure a complex construct like privacy. Few polls offer a range of questions that can measure such a complex construct. For example, researcher Alan Westin suggested consumers can be broken into three groups regarding privacy: fundamentalists who always chose privacy controls over consumer benefits; pragmatists, who weigh the benefits of consumer opportunities against the degree of personal information sought, and the unconcerned, who give up most privacy claims in exchange for consumer benefits (Federal Trade Commission, 1996). Other studies (Sheehan, 2002) have further supported such segmentation claims. Clearly, the polls, which often try to tap into privacy opinions using a limited number of questions, do not approach a level of depth allowing perceptions of context to be explored.
The polls dedicate significant time and space to examining the dimension of information sensitivity, with almost one–fourth of poll questions asking about privacy perceptions of the collection of different types of information. A range of information types resulted in a range of responses about the information relative to privacy. Although the polls report a range of responses, not all information is the same: directory information appeared to be less of a concern than the collection of financial information and Social Security numbers.
When questioning about online activities, polls often framed these questions in highly negative ways, suggesting to respondents that many types of online activities, instigated by both online entities and the online users themselves, were threats to or invasions of privacy. As previously discussed, this type of framing is likely to result in responses reflecting the negative nature of the question, that is, most people when presented with the idea that something is a threat will respond they believe it is a threat.
The polls also provide inconclusive evidence as to what online consumers believe can or should be done by government and industry to protect personal information, and depends on how the question is asked. Consumers’ view of solutions is also likely to differ based on an individual’s orientation toward privacy. Highly concerned individuals are probably more likely to favor broad government controls, while those with moderate or no privacy concerns may be likely to favor selective government interventions. The polls clearly find that consumers want to be in control. Finding the best way to provide them with this control is likely to affect public policy decisions in the next few years.
Three perspectives about privacy protection appear throughout the literature regarding public policy and privacy. First is the feeling that laws should safeguard rights of privacy and free speech. These laws should give individuals the right to keep personal information private, to send encrypted communications. The second is that consumers should be empowered to protect their rights. This perspective gives consumers more say in how entities collect and use their information. It also promotes filter programs that block out material. The final perspective states that policy should curb overly permissive rights to protect online communities. Specifically, regulations should stop 'fringe' groups in society from promoting porn and sending anonymous messages. Public standards should apply to a public communications medium (Public Agenda, 2003). This review of privacy polls provides inconclusive results on just how consumers would want their privacy to be protected, and is most likely due to the way in which questions about control were asked. Some polls found moderate to high levels of agreement that the government should pass laws to protect online privacy, which might lead some to suggest legislation is the only way to address privacy concerns. However, the high support for opt–in and the findings that consumer education is better than law might moderate this legislative fervor.
Pollsters should work with business communities to develop a deeper understanding of specific consumer concerns. Polls could focus on a single issue (such as online purchasing) and ask a series of questions tapping into the complexities of online data collection, relative to the five dimensions of privacy. For example, a poll could ask if online users are more or less willing to purchase from entities they are familiar with, compared to unfamiliar entities. Additionally, a poll could ask users if they are hesitant about sharing some types of information and not about others, and compensation (such as discounts) might address concerns.
Online marketers, content providers, industry groups, government, and pollsters themselves should renew their focus on education about online activities. One study showed 40 percent of consumers read privacy statements before providing personal information to Websites and 30 percent of online consumers find Web site privacy statements easy to understand (Jupiter, 2002). Consumer campaigns can remind consumers to look at privacy policies and instruct them about what certain types of language really means to their privacy. A downloadable brochure that a range of government and industry sites could link to, coupled with a broad reaching advertising promotional campaign, could help.
Conclusions
Privacy opinion polls can be valuable monitors of public perceptions of a range of topics. For a complex issue like online privacy, though, polls are highly limited in the utility of the information they provide to society. Business communities should work with pollsters to design studies that better explain the complexities of online privacy and investigate a range of solutions to the issues. Polls do not serve society if they only simplify complex issues in order to make clear — if erroneous — points.
About the author
Kim Bartel Sheehan is Associate Professor at the School of Journalism and Communication at the University of Oregon. Her work on privacy has appeared in The Information Society, Journal of Public Policy and Marketing, the Journal of Interactive Marketing and the Journal of Advertising. She is the author of Controversies in Contemporary Advertising (Thousand Oaks, Calif.: Sage, 2004).
E–mail: ksheehan@ballmer.uoregon.edu.
Note
References
Federal Trade Commission. 1996. "Consumer Information Privacy Hearings," at http://www.ftc.gov.
Oscar H. Gandy, 2003. "Public Opinion Surveys and the Formation of Privacy Policy," Journal of Social Issues, volume 59, number 2, pp. 283–299.
Jupiter, 2002. "Seventy Percent Of U.S. Consumers Worry About Online Privacy, But Few Take Protective Action, Reports Jupiter Media Metrix," at http://www.jupiterresearch.com/.
Rachel King, 2001. "Statement of Rachel King on Anti–Terrorism Act of 2001," before the House Judiciary Committee, Washington D.C. (24 September), at http://www.aclu.org/NationalSecurity/NationalSecurity.cfm?ID=9139&c=111.
Bill Kovach, 1990. "The Impact of Public Opinion Polls," Nieman Reports, at http://www.nieman.harvard.edu/reports/99-4_00-1NR/Kovach_Impact.html.
Alan D. Monroe, 1998. "Public Opinion and Public Policy, 1980 to 1993." Public Opinion Quarterly, volume 62, pp. 6–28.
Public Agenda. 2003. "The Perspectives in Brief, 2003," at http://www.publicagenda.org/.
Priscilla M. Regan, 1995. Legislating Privacy: Technology, Social Values and Public Policy. Chapel Hill: University of North Carolina Press.
Ferdinand D. Schoeman, 1992. Privacy and Social Freedom. New York: Cambridge University Press.
Paul B. Sheatsley, 2003. "Public Opinion," Encyclopedia Americana, at http://gi.grolier.com/presidents/ea/side/pubop.html, accessed 1 May 2003.
Kim Bartel Sheehan, 2002. "Toward a Typology of Internet Users and Online Privacy Concerns." The Information Society, volume 18, pp. 21–32.
Kim Bartel Sheehan, 1998. "Antecedents and Effects of Privacy Concerns among Online Users, " PhD dissertation in Communications, University of Tennessee, Knoxville.
James A. Stimson, Michael B. MacKuen, and Robert S. Erickson, 1995. "Dynamic Representation," American Journal of Political Science, volume 89, pp. 543–565.
Editorial history
Paper received 6 May 2004; accepted 16 June 2004.
Copyright ©2004, First Monday
Copyright ©2004, Kim Bartel Sheehan
How public opinion polls define and circumscribe online privacy by Kim Bartel Sheehan
First Monday, volume 9, number 7 (July 2004),
URL: http://firstmonday.org/issues/issue9_7/sheehan/index.html
