Media Framing of NSA Data Collection: Effects on Attitudes Alexis Dufour
Student #: 10583580 adufour1214@gmail.com alexis.dufour@student.uva.nl Thesis Supervisor: Dr. Joost van Spanje
June 27, 2014
Master’s Thesis University of Amsterdam Graduate School of Communication Master’s programme Communication Science
Abstract
The recent international wiretapping scandal leaked by the mass media has brought increased scrutiny on the National Security Agency’s secretive data collection. Several researchers have studied the mass media’s ability to portray an issue in a certain way. Many claim that these media portrayals can have significant effects on how individuals perceive these issues. Media frames with the common theme of data collection by use of surveillance from the NSA were randomly assigned to 382 adult participants. To measure the possible differences in effects of media framing, the NSA’s data collection was portrayed from two different perspectives. The media frames were portrayed as either the NSA as a (1) a protector of national security or as (2) an invader of personal privacy. Results indicated that these media frames had differences in positive/negative effects on adult attitudes regarding the NSA’s invasion of personal privacy, protection of national security, and the overall fairness of their surveillance policies; domestically and internationally.
Media Framing of NSA Data Collection: Effects on Attitudes
The public became aware of the National Security Agency’s wiretapping programs in December of 2005 after a news article in the New York Times exposed that the agency was monitoring international phone calls and e-mails (Norvell, 2009, p. 230). The exposure of the wiretapping incident by the mass media regarding NSA surveillance on the German
Chancellor shows that the agency is not afraid to cross boundaries when collecting data. The agency is highly relevant to communication studies as they are the global leader in
communications surveillance.
The NSA has held significant autonomy over citizens and officials whether it is monitoring their phone calls or their online activity. In 2007, the agency developed a data collection system, PRISM, which gathers an enormous amount of data. The system has the capacity to reach roughly 75% of all U.S. Internet traffic in the hunt for foreign intelligence, including a wide array of communications by foreigners and Americans (Gorman, Valentino-Devries, 2013). This is highly controversial as an agency with such power undermines a fundamental right of citizens, which is the right to personal privacy. A Washington court ruled the NSA’s bulk collection program was likely to prove unconstitutional and was “almost Orwellian” in scale (Roberts, 2013).
Governmental wiretapping or any other form of governmental electronic
eavesdropping into a conversation between two or more individuals who are communicating via telephone constitutes a search within the meaning of the Fourth Amendment (Norvell, 2009, p. 239). The Fourth Amendment states that "the right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no Warrants shall issue, but upon probable cause, supported by Oath or
affirmation, and particularly describing the place to be searched, and the persons or things to be seized” (Norvell, 2009, p. 233).
By examining the influence of media framing on attitudes regarding the NSA, differences in effects that the mass media can have on attitudes are explored. Media framing is a very popular topic in communications research by studying the media and its effects. A media frame is a central organizing idea or story line that provides meaning to an unfolding strip of events, the frame suggests what the controversy is about, the essence of the issue (Gamson, Modigliani, 1987, p. 143). To frame is to select some aspects of a perceived reality and make them more salient in a communicating text, in such a way as to promote a
particular problem definition, causal interpretation, moral evaluation, and/or treatment recommendation (Entman, 1993, p. 52). The entire study of mass communication is based on the premise that the media have significant effects (McQuail, 1994, p. 327).
This thesis examined the attitudes of 382 adults by distribution of an experimental survey. The media frames referred to the National Security Agency’s data collection as a (1) a protector of national security or as (2) an invader of personal privacy. The two randomly distributed media frames had the common theme of NSA data collection, which is gathered by the use of surveillance over citizens and officials. One media frame portrayed data collection as lawful for the protection of national security while the other portrayed data collection as unlawful and an invasion of personal privacy.
Constitutional rights are addressed in both media frames as the invasion of personal privacy media frame states that NSA data collection is a direct violation of the Fourth
Amendment. The protection of national security media frame states that NSA data collection does not violate Constitutional rights. Adults evaluated their attitudes on the following issues after viewing one of two media frames: The NSA’s protection of national security, the NSA’s invasion of personal privacy, the fairness of NSA surveillance policies, and NSA surveillance domestically and internationally.
Some may claim that data collection from the NSA is not needed to protect the U.S. from foreign threats and is an invasion of their personal privacy. As the technical capabilities of governments to collect such data increases, and therefore, the level of data being collected increases, privacy concerns become a key issue that must be addressed (Belanger, Hiller, 2006, p. 56). Collection of data about individuals has always invoked issues of privacy and online technology increases privacy concerns as it allows for faster, easier storage of more data (Belanger, Hiller, 2006, p. 54). Some lawmakers and civil libertarians say that, given the volumes of data the NSA is examining, privacy protections are insufficient (Gorman,
Valentino-Devries, 2013). This brings us to Research Question 1: How does media framing affect citizens’ attitudes toward the NSA’s invasion of personal privacy?
Some may claim that data collection from the NSA is needed to protect the U.S. from foreign threats despite personal privacy concerns. A judge stated that privacy protections enshrined in the Fourth Amendment of the US Constitution needed to be balanced against a government need to maintain a database of records to prevent future terrorist attacks
declaring that bulk collection of telephone data does not violate the Constitution (Roberts, 2013). A previous study claimed data collection would not necessarily violate the Fourth Amendment because the NSA is aimed at protecting national security (Norvell, 2009, p. 233), which is what the protection of national security media frame emphasizes. This raises
Research Question 2: How does media framing affect citizens’ attitudes toward the NSA’s protection of national security?
The media frames in this experiment featured the central story line of NSA data collection by use of surveillance; the controversy of invasion of personal privacy or protection of national security brought focus to these issues. Both underlying issues, however, led to attitude formation regarding the fairness of NSA surveillance policies. In a 2000 National Consumers League survey, respondents ranked personal privacy above health
care, education, crime and taxes as concerns (Paul, 2001), and some may question whether tolerance levels for NSA surveillance have declined. This raises Research Question 3: How does media framing affect citizens’ attitudes toward the fairness of NSA surveillance policies?
The NSA is focused on collecting foreign intelligence, but the streams of data it monitors include both foreign and domestic communications (Gorman, Valentino-Devries, 2013). Officials say the system's broad reach makes it more likely that purely domestic communications will be incidentally intercepted and collected in the hunt for foreign ones. (Gorman, Valentino-Devries, 2013). In some cases, the written content of emails sent between citizens within the U.S. and also filters domestic phone calls made with Internet technology (Gorman, Valentino-Devries, 2013). This brings us to Research Question 4: How does media framing affect citizens’ attitudes toward NSA domestic surveillance?
The phone tapping scandal involving secretive surveillance on the German Chancellor brought skepticism to the NSA regarding proper disclosure of their surveillance. The
President or Executive Branch officers are not less likely to conduct unreasonable searches just because one party happens to reside in another country (Norvell, 2009, p. 239). The NSA's filtering, carried out with telecom companies, is designed to look for communications that either originate or end abroad, or are entirely foreign but happen to be passing through the U.S. (Gorman, Valentino-Devries, 2013). This brings us to Research Question 5: How does media framing affect citizens’ attitudes toward NSA international surveillance? Media Frames (Independent Variable) Effects
Conceptualizations of framing developed by Entman (1993), Levin, Schneider, and Gaeth (1998), Druckman (2001), Kahneman and Tversky (1979, 1981), and Nelson, Clawson and Oxley (1997) examined media frames as the independent variable. According to Entman (1993), media framing can have an impact on attitudes and opinions while directing our
attention to the details of just how a communicated text exerts its power (p. 56). The cause and effect relationship of the positive/negative media framing of NSA data collection was expected to have a differential effect on attitude formation. Studies measuring media framing effects have not yet acknowledged effects on attitudes regarding the NSA.
According to Levin, Schneider, and Gaeth (1998), equivalency framing effects occur when frames cast the same critical information in either a positive or a negative light, which causes individuals to have different preferences (p. 150). The type of media framing used in the NSA portrayals is closely related to equivalency framing. Both media frames were very similar in content, but were framed in either a positive or negative light. The positive media frame was the protection of national security frame as it portrayed NSA data collection as lawful to the Fourth Amendment and necessary for the protection of national security. The negative media frame was the invasion of personal privacy frame as it portrayed NSA data collection as a violation of the Fourth Amendment of citizens and an invasion of personal privacy.
According to Druckman (2001), emphasis framing effects work by causing
individuals to focus on certain aspects or characterizations of an issue or problem instead of others (p. 230). Furthermore, this type of framing effect shows that by emphasizing a subset of potentially relevant considerations, a speaker can lead individuals to focus on these considerations when constructing their opinions (Druckman, 2001, p. 230). Much like equivalency framing effects, the media frames used in the NSA portrayals are related to emphasis framing effects. Participants were expected to have more positive attitudes toward the NSA’s data collection when reading the protection of national security media frame because it focused on the issue of protecting national security against foreign threats.
when reading the invasion of personal privacy media frame because it focused on the issue of violating Fourth Amendment rights by invading the personal privacy of citizens.
According to Kahneman and Tversky (1979, 1981), valence framing effects cast the same critical information in either a positive or a negative light explained by prospect theory. Prospect theory (Kahneman, Tversky, 1979) frames information encoded as positive or negative and this encoding determines which portion of a psychophysical value function would contribute to the perception of the worth of the information. This perspective has led to numerous insights into decision making behavior, and the approach forms a strong
foundation for understanding a wide variety of phenomena (Kahneman, Tversky, 1979). The media frames used in the NSA portrayals are related to valence framing effects as the
different media frames were expected to contribute to the participants’ perception of the information.
According to Nelson, Clawson, and Oxley (1997), framing is the process in which a communication source, such as a news organization, defines and constructs a political issue or public controversy (p. 567). Furthermore, frames shape individual understanding or opinion concerning an issue by stressing specific elements or features of the broader controversy, reducing a usually complex issue down to one or two central aspects (Nelson, Clawson, Oxley, 1997, p. 568). Media framing revolved around the public controversy of NSA data collection and it was expected that the positive/negative media frames would form different perceptions by stressing or non-stressing violations of Fourth Amendment rights. Frames influence opinions by stressing specific values, facts, and other considerations endowing them with greater apparent relevance to the issue than they might appear to have under an alternative frame (Nelson et al., 1997, p. 569).
There are, however, authors that have doubts regarding media framing effects. According to Entman (1989) and Graber (1988), because salience is a product of the
interaction of texts and receivers, the presence of frames in the text, as detected by
researchers, does not guarantee their influence in audience thinking. Topics involving issues of strongly held attitudes or high personal involvement are less susceptible to attribute framing effects. According to Marteau (1989), he found no framing effects across a wide variety of problems involving abortion decisions. In some instances, the framed information apparently receives little or no weight in the judgment process, so, consequently, the framing effect is negligible (Levin et al., 1998, p. 160).
This thesis examined whether the media framing of an issue had subsequent effects on adults, specifically the difference in influence that each media frame had on attitudes toward the NSA. This was done by comparing two groups of participants based on the media frame they were exposed to and observing the difference in effects on attitudes. Citizens in the experiment focused on the assigned issue in their given media frame and this had an effect on their attitudes regarding the NSA’s protection of national security, invasion of personal privacy, as well as the fairness of their surveillance policies; domestically and internationally. Based on the above media framing theories, the following hypotheses were introduced. Hypotheses
Attribute framing is when some characteristic of an object or event serves as the focus of the framing manipulation (Levin et al., 1998, p. 150). In attribute framing, a single
attribute of an object or event is framed (labeled) positively or negatively, and the effect on item evaluation is assessed by comparing attractiveness ratings for the object or event in each framing condition (Levin et al., 1998, p. 181). The object under examination is the NSA’s data collection through the use of surveillance and is framed in a positive or negative light using two issues, protection of national security and invasion of personal privacy.
Attitudes regarding the amount of personal privacy the NSA allows citizens and officials was the first issue measured by this experiment. In attribute framing, attributes are
judged more favorably when labeled in positive terms rather than negative terms (Levin et al., 1998, p. 181). The protection of national security media frame was expected to induce
positive attitudes toward privacy concerns while the invasion of personal privacy media frame was expected to have the opposite effect.
Hypothesis 1: Participants assigned to the NSA protection of national security media frame are more likely to perceive that the NSA offers a proper balance between personal privacy and national security than participants assigned to the NSA invasion of personal privacy media frame.
Attitudes regarding the amount of national security protection the NSA exerts over citizens and officials was the second issue measured by this experiment. The protection of national security media frame was expected to induce positive attitudes toward increased national security protection. The invasion of personal privacy media frame was expected to induce negative attitudes toward increased national security protection. Because attribute framing is restricted to the simplest case, it allows the most straightforward test of the influence of positive and negative framing (Levin et al., 1998: 158).
Hypothesis 2: Participants assigned to the NSA protection of national security media frame are more likely to believe that the NSA offers too little national security protection than participants assigned to the NSA invasion of personal privacy media frame.
Attitudes regarding the fairness of NSA surveillance policies was the third issue measured by this experiment. The protection of national security media frame was expected to induce positive attitudes toward the fairness of NSA surveillance policies while the
invasion of personal privacy media frame was expected to have the opposite effect. Attribute framing represents perhaps the simplest case of framing, making it especially useful for gaining a basic understanding of how descriptive valence influences information processing (Levin et al., 1998, p. 158).
Hypothesis 3: Participants assigned to the NSA protection of national security media frame are more likely to believe that NSA surveillance policies are fair than participants assigned to the NSA invasion of personal privacy media frame.
Attitudes regarding NSA surveillance over its own citizens and officials was the fourth issue measured by this experiment. The protection of national security media frame was expected to induce positive acceptance toward NSA domestic surveillance while the invasion of personal privacy media frame was expected to have the opposite effect. The two media frames used in this experiment were directly related to equivalency framing, where frames cast the same critical information in either a positive or a negative light, which causes individuals to have different preferences (Levin et al., 1998, p. 150). The protection of national security media frame casted NSA domestic surveillance in a positive light while the invasion of personal privacy media frame casted it in a negative light.
Hypothesis 4: Participants assigned to the NSA protection of national security media frame are more likely to agree with the idea that NSA domestic surveillance over its own citizens and officials is justified for national security than participants assigned to the NSA invasion of personal privacy media frame.
Attitudes regarding NSA surveillance over citizens and officials internationally was the fifth issue measured by this experiment. The protection of national security media frame was expected to induce positive acceptance toward NSA international surveillance while the invasion of personal privacy media frame was expected to induce negative reactions. A closer examination of these different kinds of framing effects can reveal critical variables involved in representing, processing, and responding to information (Levin et al., 1998, p. 150).
Hypothesis 5: Participants assigned to the NSA protection of national security media frame are more likely to agree with the idea that NSA international surveillance over citizens
and officials is justified for national security than participants assigned to the NSA invasion of personal privacy media frame.
Method Participants
The participants were 382 adults over the age of 18 (199 women, 183 men). A convenience sample was used by targeting participants from the following sampling frames: Canadian undergraduate/graduate students and professionals, American high
school/undergraduate/graduate students and professionals, European undergraduate/graduate students and professionals, and a small percentage of other professionals working in other areas of the world including: Asia, South America, Africa, and Australia (these participants accounted for just over 10% of the data). A convenience sample is one that is simply available to the researcher by virtue of its accessibility (Bryman, 2012, p. 201).
All students were targeted through social media networks and e-mail, in which participants were attracted by digital advertisements on Facebook and Twitter, as well as e-mail lists of graduate students’ part of the University of Amsterdam’s Graduate School of Communication. All professionals were targeted with digital advertisements via LinkedIn. Ethical standards for the recruitment of subjects were met according to guidelines as test subjects were healthy, competent adults (aged 18 or older), who participated in a research project voluntarily through digital advertisements and were not disproportionately rewarded for doing so (Ethics Committee, 2012, p. 9).
Participation was completely voluntary and subjects were free to leave at any time if they felt uncomfortable according to guidelines for recruitment of test subjects (Ethics Committee, 2012, p. 10). Participants were not given an informed consent form as guidelines for recruitment of test subjects states; An exception to the informed consent procedure can be made in the case of research that involves a questionnaire being offered without the
researcher meeting the test subject, such as when a questionnaire is distributed by post and completed at home, or when a questionnaire is offered via a website (Ethics Committee, 2012, p. 11). The website that offered the questionnaire was Qualtrics.com.
Total anonymity for participants followed guidelines for recruitment of test subjects as the data gathered in the course of the research was not passed on to third parties (neither published nor disclosed in conversations or mutual consultation) in such a way that allowed the results or other findings to be traced back to individual test subjects (Ethics Committee, p. 13). Recruitment of test subjects fell within descriptions of standards research as standard data collection was done within publicly accessible media (Ethics Committee, 2012, p. 16). Procedure
The experiment tested whether media frames have different effects on adult attitudes regarding NSA surveillance by randomly assigning one of two media frames to participants as (1) the NSA as a protector of national security or (2) the NSA as an invader of personal privacy (see table 1). The manipulation included each participant randomly assigned to one of two media frames with different content, but very similar in context and structure.
The first media frame was the NSA as a protector of national security and was distributed to 191 respondents. Of the 191 participants, 185 responded to the five questions with a response rate of 96.9%. The second frame was the NSA as an invader of personal privacy and was distributed to the other 191 respondents. Of the 191 participants, only 139 responded to the five questions with a response rate of 72.7%. The content of media frames randomly assigned to participants is shown below.
TABLE 1. Content of Media Frames Theme: NSA surveillance
The National Security Agency as a protector of national security
Theme: NSA surveillance
The National Security Agency as an invader of personal privacy
There is no better way the National Security Agency can help protect the U.S. from threats than with the surveillance program that collect billions of domestic phone and internet records. To keep the nation secure, the NSA must continue to monitor and collect intelligence through data collection. A legal battle over the scope of NSA
surveillance ended with a court opinion declaring that bulk collection of telephone data does not violate the constitution.
The National Security Agency surveillance program that keeps tabs on billions of domestic phone and internet records in the U.S. is illegal and should be shut down considering no threats have been discovered through the massive data collection. The panel argues the NSA's tracking program implicates constitutional concerns under the First and Fourth Amendments, raises serious threats to privacy and civil liberties as a policy matter, and has shown only limited value.
Attitude (Dependent Variable) Effects
Immediately after reading one of the two assigned media frames, participants finished the survey by answering five separate questions based on their attitudes of the NSA. These included the NSA’s protection of national security, the NSA’s invasion of personal privacy, and the fairness of NSA surveillance policies; domestically and internationally using a 7 point scale.
The first question asked: “Which do you feel best describes your attitude regarding the NSA’s surveillance as an invader of personal privacy?” The 7 point scale was anchored by “not enough privacy” and “too much privacy,” to indicate their attitude regarding NSA invasion of personal privacy. The second question asked: “Which do you feel best describes your attitude regarding the NSA’s surveillance as a protector of national security?” The 7 point scale was anchored by “not enough protection” and “too much protection,” to indicate their attitude regarding NSA protection over national security. The third question asked: “Which do you feel best describes your attitude regarding the NSA’s surveillance policies?” The 7 point scale was anchored by “not at all fair” and “extremely fair,” to indicate their
The fourth question asked to what extent did they agree or disagree with the following statement: “The NSA is justified in collecting data on its own citizens and officials only if they have a reason to believe that they are a threat to national security.” The 7 point scale was anchored by “strongly disagree” and “strongly agree,” to indicate their attitude regarding the NSA’s domestic surveillance. The fifth question asked to what extent did they agree or disagree with the following statement: “The NSA is justified in collecting data on citizens and officials across international borders only if they have a reason to believe that they are a threat to national security.” The 7 point scale was anchored by “strongly disagree” and
“strongly agree,” to indicate their attitude regarding the NSA’s international surveillance. The participants finished are grouped by the media frame to which they were assigned (see table 2).
TABLE 2. Attitudes of NSA by media frame group (Scale 1-7)
Group 1 Group 2 Protector of national security media frame Invader of national security media frame H1 The NSA as an invader
of personal privacy
N 186 136
Mean 2.74 2.38
1.495 1.620
H2 The NSA as a protector of national security N 189 137 Mean 4.67 5.15 1.597 1.736 H3
The fairness of the NSA’s surveillance policies N 189 135 Mean 2.98 2.44 1.632 1.499 H4 Acceptance of NSA domestic surveillance N 188 137 Mean 4.84 4.34 1.898 2.027 H5 Acceptance of NSA International surveillance N 189 137 Mean 4.68 3.96
Results
It was expected that participants assigned to the NSA protection of national security media frame (group 1) were more likely to believe that the NSA offers a proper balance between personal privacy and national security compared to those assigned to the NSA invasion of personal privacy media frame (group 2) (H1). Independent-samples t-tests t = 2.058, MD = .360, p = .040 (two-tailed) showed that group 1 was more likely to believe that the NSA offers a proper balance between personal privacy and national security (N = 186, M = 2.74, SD = 1.495), which was significantly different in comparison to participants of group 2 (N = 136, M = 2.38, SD = 1.620), so this hypothesis is accepted. The size of the effects showed that participants in group 1 were more than 1/3 more likely to have more positive attitudes toward the NSA’s balance of personal privacy and national security compared to participants in group 2.
It was expected that participants assigned to the NSA protection of national security media frame were more likely to believe that the NSA offers too little national security protection compared to those assigned to the NSA invasion of personal privacy media frame (H2). Independent-samples t-tests t = -2.589, MD = -.481, p = .010 (two-tailed) showed that group 1 was more likely to believe that the NSA offers too little national security protection (N = 189, M = 4.67, SD = 1.597), which was significantly different in comparison to
participants of group 2 (N = 137, M = 5.15, SD = 1.736), so this hypothesis is accepted. The size of the effects showed that participants in group 1 were almost ½ less likely to believe that the NSA offers too much national security protection compared to participants in group 2.
It was expected that participants assigned to the NSA protection of national security media frame were more likely to believe that NSA surveillance policies are fair compared to those assigned to the NSA invasion of personal privacy media frame (H3).
Independent-samples t-tests t = 3.076, MD = .547, p = .002 (two-tailed) showed that group 1 was more likely to believe that the NSA surveillance policies are fair (N = 189, M = 2.98, SD = 1.632), which was significantly different in comparison to participants of group 2 (N = 135, M = 2.44, SD = 1.499), so this hypothesis is accepted. The size of the effects showed that participants in group 1 were more than ½ more likely to believe that NSA surveillance policies were fair compared to participants in group 2.
It was expected that participants assigned to the NSA protection of national security media frame were more likely to agree with the idea that NSA domestic surveillance over its own citizens and officials is justified for national security compared to those assigned to the NSA invasion of personal privacy media frame (H4). Independent-samples t-tests t = 2.276, MD = .499, p = .024 (two-tailed) showed that group 1 was more likely to agree with the idea that NSA free domestic surveillance over its own citizens and officials is justified (N = 188, M = 4.84, SD = 1.898), which was significantly different in comparison to participants of group 2 (N = 137, M = 4.34, SD = 2.027), so this hypothesis is accepted. The size of the effects showed that participants in group 1 were about ½ more likely to agree with the idea that NSA domestic surveillance over its own citizens and officials is justified compared to participants in group 2.
It was expected that participants assigned to the NSA protection of national security media frame were more likely to agree with the idea that NSA international surveillance over citizens and officials is justified for national security compared to those assigned to the NSA invasion of personal privacy media frame (H5). Independent-samples t-tests t = 3.207, MD = .721, p < .001 (two-tailed) showed that group 1 was more likely to agree with the idea that NSA international surveillance over citizens and officials is justified (N = 189, M = 4.68, SD = 1.884), which was significantly different in comparison to participants of group 2 (N = 137, M = 3.96, SD = 2.159), so this hypothesis is accepted. The results of this research question
indicate that it had the largest difference in size of effects. Participants in group 1 were almost ¾ more likely to agree with the idea that NSA international surveillance over citizens and officials is justified for national security compared to participants in group 2. All
significance values were according to equal variances assumed with a 95% confidence interval.
Conclusion
The results of this experiment showed that media framing had a clear difference in effects on attitudes regarding the NSA’s invasion of personal privacy, the NSA’s protection of national security, and the fairness of the NSA surveillance policies; domestically and internationally. By constructing media frames using issues that were especially relevant to society (protection of national security or invasion of personal privacy); the two frames positively or negatively emphasized which considerations dominated attitudes of participants. The findings are linked to the theory of attribute framing (Levin et al. 1998, p. 166). The stimulus label was NSA surveillance (data collection) and the issue-based descriptions of the target stimulus (media frame) enabled participants in this experiment to form attitudes regarding the NSA.
The findings offer many implications for links to other media framing theories. Prospect theory (Kahneman, Tversky, 1979) frames information encoded as positive or negative and this encoding determines which portion of a psychophysical value function would contribute to the perception of the worth of the information. The findings validate the prospect theory as participant attitudes were significantly different based on the
positive/negative framing they received.
The invasion of personal privacy media frame influenced responses by suppressing support for the NSA’s balance of national security protection and the fairness of the NSA’s surveillance policies. Participants assigned to this media frame had more negative attitudes
toward NSA data collection domestically and internationally. This media frame also influenced participants by promoting more negative attitudes toward the NSA’s balance between personal privacy and national security. It can be concluded that the portrayal of a violation of Constitutional rights had an effect on group 2 participant perceptions as they showed less support for NSA data collection compared to group 1.
Research Question 1: The experiment found that media framing affected citizens’ attitudes toward the NSA's invasion of personal privacy.
The protection of national security media frame influenced responses by increasing support for the NSA’s balance of national security protection and the fairness of the NSA’s surveillance policies. Participants assigned to this media frame had more positive attitudes toward NSA data collection domestically and internationally. This media frame also influenced participants by promoting more positive attitudes toward the NSA’s balance between personal privacy and national security. It can be concluded that protection of national security concerns had an effect on group 1 participant perceptions as they showed more support for NSA data collection compared to group 2.
Research Question 2: The experiment found that media framing affected citizens’ attitudes toward the NSA’s protection of national security.
The positive and negative representations of the NSA’s data collection by the use of media framing were seen to have had a significant effect on the difference in attitudes toward the NSA and their surveillance policies. Findings are linked to valence framing (Kahneman and Tversky, 1979, 1981), where framing effects cast the same critical information in either a positive or a negative light explained by prospect theory. It can be concluded that participant perceptions regarding the fairness of NSA surveillance policies was more positive when viewing the protection of national security media frame and more negative when viewing the invasion of personal privacy media frame.
Research Question 3: The experiment found that media framing affected citizens’ attitudes toward the fairness of the NSA’s surveillance policies.
It can be concluded that participant perceptions of NSA domestic surveillance was more positive when viewing the protection of national security media frame and more negative when viewing the invasion of personal privacy media frame. Group 1 participants were shown that Constitutional rights were not violated in the protection of national security media frame because NSA surveillance was necessary to protect the U.S. from foreign threats. This media frame centered on the NSA’s responsibility to protect the nation from foreign threats. Group 2 participants were shown that Constitutional rights were violated and that NSA surveillance was not needed because it has shown limited effects for the protection of national security. This media frame centered on the NSA violating an individual’s
fundamental right to personal privacy. Findings are linked to emphasis framing (Druckman, 2001), where effects work through a psychological process where individuals consciously and deliberatively think about the relative importance of different considerations suggested by a frame (p. 230).
Research Question 4: The experiment found that media framing affected citizens’ attitudes toward NSA domestic surveillance.
It can be concluded that participant perceptions of NSA international surveillance was more positive when viewing the protection of national security media frame and more
negative when viewing the invasion of personal privacy media frame. Findings are linked to equivalency framing (Levin et al., 1998), where framing effects occur when frames cast the same critical information in either a positive or a negative light, which causes individuals to have different preferences (p. 150). Group 1 participants were exposed to the positive media frame (protection of national security) and group 2 participants were exposed to the negative media frame (invasion of personal privacy) and statistical differences in positive/negative
attitudes indicated that participants saw these frames for how they were supposed to be portrayed.
Research Question 5: The experiment found that media framing affected citizens’ attitudes toward NSA international surveillance.
The results conclude that media framing had a clear effect on differences in attitudes regarding the NSA. The two randomly assigned media frames had a significantly different effect on how participants felt about the NSA’s protection of national security, invasion of personal privacy, the fairness of NSA surveillance policies, and NSA surveillance
domestically and internationally. Discussion
Previous approaches to media framing have shown how frames affect perceptions of causality and responsibility for social problems such as crime and poverty (Iyengar, 1991), and the relationship between policy opinions and predictors of those opinions (Kinder & Sanders, 1990, 1996; Nelson & Kinder, 1996). This approach showed how media framing affected perceptions of policy opinions, specifically the NSA’s data collection through use of surveillance. This is an extremely controversial topic as the NSA is the global leader in communications surveillance and is supposed to be completely “secretive” in their surveillance techniques.
This experiment brought an entirely new insight into the field of media framing effects as results showed significantly different effects on attitudes toward the NSA. Studies in which frames serve as independent variables typically are more interested in the effects of framing, as the question becomes, does the framing of issues influence evaluations of issues or political actors? (Scheufele, 1999, p. 107). Results also showed that hypotheses one to five were accepted as the media framing of NSA data collection produced a difference in effects. The choice of media frame had important implications for the balance of public opinion on an
issue as media frames provided cues about how adults should integrate their beliefs into attitudes (Nelson et al., 1997, p. 577).
The results offer many practical implications for how the mass media can have a different effect on attitudes. The findings indicate that positive/negative media framing can have a significant impact on how citizens perceive an issue and how the mass media can influence public opinion. Frames make suggestions to the individual by rendering
considerations more important than others, thereby leading these considerations to be applied when forming an opinion (Lecheler, de Vreese, 2012, p. 186). The mass media should
certainly continue to frame news stories in different ways, especially if they want to influence public opinion. Furthermore, the NSA should be more careful and disclose all data collection as the mass media’s ability to uncover undisclosed surveillance can have a negative impact on how citizens view the agency.
The findings showed considerable support for all five hypotheses as both groups of participants had significantly different attitudes based on the media frame they were exposed to. The effect sizes observed were positive as the manipulation of media frames had their desired effect on participants for each research question. Studies have generally tested this link by presenting participants with either equivalency or emphasis frames, and by focusing on issue interpretations, attitudes, or opinions as dependent variables (Lecheler, de Vreese, 2012, p. 186). Participants of group 1 receiving the protection of national security media frame were more attracted to the NSA offering too much personal privacy. Participants of group 2 receiving the invasion of personal privacy media frame were more attracted to the NSA offering not enough personal privacy.
Participants of group 1 had more positive attitudes regarding the fairness of the NSA’s surveillance policies, whereas participants of group 2 had more negative attitudes. Participants of group 1 were more attracted to agree that NSA domestic surveillance over
citizens and officials was needed for protection of national security and participants of group 2 were more attracted to disagree. Participants of group 1 were more attracted to agree that NSA international surveillance over citizens and officials was needed for protection of national security and participants of group 2 were more attracted to disagree.
A limitation, however, was found in research question 2. Participants of group 1 were more attracted to the NSA offering too little national security protection, whereas participants of group 2 were more attracted to the NSA offering too much national security protection. This is a problem as both media frames seem to have had a negative effect on attitudes toward the NSA’s protection of national security. Offering too little national security
protection and offering too much national security protection can both be seen as negative. Of course, there may be some that claim that you can never have too much national security protection, but this is contested, as there must always be an equal balance in a functioning democracy.
Observations of findings from participants in different areas of the world allow us to make a few generalizations. The differences in responses from American participants and European participants are discussed as they accounted for 72% of participants. About 52% of European participants indicated that the NSA gave “not enough privacy” to citizens when collecting data. This was the most extreme answer possible and was different from the mere 19% of Americans who chose this option. About 35% of Europeans indicated that the NSA had “too much protection” over citizens when collecting data. This was the most extreme answer possible and was different from the mere 11% of Americans who chose this option.
About 43% of European participants indicated that NSA surveillance policies were “not at all fair” and this was the most extreme answer possible. Only about 18% of
Americans chose this option. There was not a significant difference between U.S. and EU participants regarding NSA domestic surveillance, however, U.S. participants were generally
more accepting than those from the EU. There was, however, a large difference between responses from an international perspective. About 27% of European participants “strongly disagreed” with the idea that NSA surveillance across international borders was acceptable compared to less than 2% of Americans who chose this option.
Another limitation was generalizing results to an overall population because of the use of a convenience sample to attract participants. The problem with convenience sampling is that it is impossible to generalize the findings, because we do not know of what population this sample is representative (Bryman, 2012, p. 201). There may be a large sampling bias because a convenience sample is not representative of the entire population, thus resulting in a low level of external validity. External validity examines whether or not an observed causal relationship should be generalized to and across different measures, persons, settings, and times (Calder, Phillips, & Tybout, 1982, p. 240).
An ideal sample in this case would have been a randomized sample as it is more generalizable to the general population. Also, there was a great deal of overrepresentation in the sample. Over half of the participants were over the age of 30 and over 95% of participants held at least a bachelor’s degree. The education level of participants is not representative of the global population, however, the results allowed conclusions to be drawn based on the difference in responses. Another limitation was the difference in response rates between the two media frame groups. The protection of national security media frame had a significantly higher response rate than the invasion of personal privacy media frame. This may have been because the invasion of personal privacy media frame contained content that many people may not have agreed with, been more sensitive to, or that went against their personal ideologies.
There were some observed differences in responses from participants holding a bachelor’s degree and those holding a master’s degree. About 29% of participants holding a
bachelor’s degree indicated that the NSA gave “not enough privacy” to citizens when collecting data. This was the most extreme answer possible and was different from the 44% of participants with a master’s degree who chose this option. About 18% of participants holding a bachelor’s degree indicated that the NSA had “too much protection” to citizens when collecting data. This was the most extreme answer possible and was different from the 32% of participants with a master’s degree who chose this option. About 23% of participants holding a bachelor’s degree indicated that NSA surveillance policies were “not at all fair.” This was the most extreme answer possible and was different from the 42% of participants with a master’s degree who chose this option.
Less than 1% of participants with a bachelor’s degree indicated that they “strongly disagreed” with NSA domestic surveillance compared to the 15% of participants with
master’s degrees. A smaller difference was found with NSA surveillance across international borders as 10% of participants with bachelor’s degrees “strongly disagreed” with NSA international surveillance compared to 21% of participants with master’s degrees. Participants with a higher level of education were generally less accepting of NSA surveillance.
A significant limitation was the durability of the media framing effects, which refers to how long effects last on citizens. Framing experiments have focused on immediate effect measurement, and have thus delivered only short-term evidence of framing effects in political communication (Lecheler, de Vreese, 2011, p. 964). The conclusions drawn by a majority of studies on duration of framing effects carefully suggest that one-shot framing effects might dilute relatively quickly, and that only multiple exposure to (the same) news frames can produce lasting effects (Lecheler, de Vreese, 2011, p. 965). Unfortunately, the experiment only measured differences in media framing effects once, so it is highly likely that attitude formations will fade over time.
However, Lecheler and de Vreese also point out that a majority of framing authors argue that framing effects are applicability effects, which means that a frame renders certain belief considerations more important and are, in turn, more likely to be incorporated into subsequent judgments (p. 965). We believe that the use of experimental (survey) designs in framing effect research should be encouraged (Lecheler, de Vreese, 2011, p. 975). From this, it can be assumed that attitude effects may not produce lasting effects, but can possibly come up again when making other evaluations. Price and Tewksbury (2007) back up this claim by calling these applicability effects, where once activated, ideas and feelings retain some residual activation potential, making them more likely to be activated and used when making subsequent evaluations (p. 486).
A future avenue for research is recommended as the recent international phone tapping scandal could have been an interesting topic with the use of opposing media frames. The experiment could have framed the controversy as either an invasion of personal privacy or as a necessity for the protection of national security. An article published in February of this year by RT, a Russian media outlet, stated that the Snowden documents cited earlier this month by the German media suggested the NSA began monitoring the chancellor of
Germany starting in 2002.
A question that can be raised is whether each of the five research questions were similar. The first two research questions were similar as they both evaluated amounts of personal privacy or national security the NSA gives citizens. These are similar in that they both provide evaluations on the balance of personal privacy and national security the NSA maintains, which is extremely important. The third research question was different from all the others as citizens evaluated the fairness of the NSA’s surveillance policies. No other research question did this as the fourth and fifth research questions were similar to each other in that they both measured the level of acceptance citizens have for NSA surveillance. The
only difference here was that one question measured acceptance of domestic surveillance while the other measured acceptance of surveillance across international borders. The way the questions were posed allows this thesis to have five separate conclusions regarding citizen attitudes toward the NSA.
References
Belanger, F., Hiller, S. (2006). A framework for e-government: privacy implications. Business Process Management Journal, 48-60. doi: 10.1108/14637150610643751 Bryman, A. (2012). Social Research Methods (2nd ed). Oxford University Press.
Calder, B. J., Lynn W. P., & Tybout, A. (1982). The Concept of External Validity. Journal of Consumer Research, 240-244. http://www.jstor.org/stable/2488620
Davis, D. W., Silver, B. D. (2004). Civil liberties vs. security: public opinion in the context of the terrorist attacks on America. American Journal of Political Science, 28–46. doi:
10.1111/j.0092-5853.2004.00054.x
Druckman, J. N. (2001). The Implications of Framing Effects for Citizen Competence. Political Behavior, 225-256. http://www.jstor.org/stable/1558384
Entman, R. M. (1989). Democracy without citizens: Media and the decay of American politics. New York: Oxford University Press.
Entman, R. M. (1993). Framing: Towards clarification of a fractured paradigm. Journal of Communication, 51-58. doi: 10.1111/j.1460-2466.1993.tb01304.x
Erikson, R. S., Tedin, K. L. (2003). American Public Opinion. New York: Longman. 6th ed. Ethics Committee. (January 2012). Ethical Review Procedure for Research at the Department of Communication Science, University of Amsterdam, 1-32. Retrieved from
http://blackboard.uva.nl/bbcswebdav/pid-4804861-dt
contentrid5939305_1/courses/2100.77644T001Y.S2.1.20132014/Procedure_for_ethical_revi ew2.pdf.
(2014 February). Former German chancellor surprised that NSA continued to spy on Merkel. RT. Accessed June 17, 2014. Retrieved from http://rt.com/usa/chancellor-schroeder-nsa-surprised-091/.
Gamson, W., Modigliani, G. (1987). The Changing Culture of Affirmative Action. 137-177 in Research in Political Sociology.
Graber, D. A. (1988). Processing the news: How people tame the information tide. (2nd ed.). New York: Longman.
Gorman, S., Valentino-Devries, J. (2013 August). New details show broader NSA surveillance reach. The Wall Street Journal. Accessed June 20, 2014. Retrieved from http://online.wsj.com/news/articles/SB10001424127887324108204579022874091732470. Iyengar, S. (1991). Is anyone responsible? Chicago: University of Chicago Press.
Kahneman, D., Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica, 263-291. doi: 10.2307/1914185
Kahneman, D., Tversky, A. (1982). The psychology of preferences. Scientific American, 160–173. http://psycnet.apa.org/psycinfo/1982-22772-001
Lecheler, S., de Vreese, H. (2011). Getting Real: The Duration of Framing Effects. Journalism of Communication, 959-983. doi: 10.1111/j.1460-2466.2011.01580.x Lecheler, S., de Vreese, H. (2012). News Framing and Public Opinion: A Mediation Analysis of Framing Effects on Political Attitudes. Journalism and Mass Communication Quarterly, 185-204. doi: 10.1177/1077699011430064
Levin, Schneider, & Gaeth. (1998). All Frames Are Not Created Equal: A Typology and Critical Analysis of Framing Effects. Academic Press, 149–188. doi:
10.1006/obhd.1998.2804
McQuail, D. (1994). Mass Communication Theory: An introduction by Denis McQuail. SAGE Publications.
Nelson, T. E. Clawson, R. A. & Oxley, Z. M. (1997). Media Framing of a Civil Liberties Conflict and its Effect on Tolerance. The American Political Science Review, 567-583. http://www.jstor.org/stable/2952075
Nelson, T. E. & Kinder, D. R. (1996). Issue framing and group-centrism in American public opinion. Journal of Politics, 1055-1078. http://www.jstor.org/stable/2960149
Norvell, B. C. (2009). The Constitution and the NSA warrantless wiretapping program: A fourth amendment violation? Yale Journal of Law and Technology, 230-259.
http://digitalcommons.law.yale.edu/yjolt/vol11/iss1/8
Paul, P. (2001). Mixed signals. American Demographics, 44-9.
Price, V., Tewksbury, D., & Powers, E. (1997). Switching trains of thought: The impact of news frames on reader’s cognitive responses. Communication Research, 481-506. doi: 10.1177/009365097024005002
Roberts, D. (2013, December 27). NSA mass collection of phone data is legal, federal judge rules. The Guardian. Retrieved from http://www.theguardian.com/world/2013/dec/27/judge-rules-nsa-phone-data-collection-legal.
Scheufele, D. A. (1999). Framing as a theory of media effects. Journal of Communication, 103-122. doi: 10.1111/j.1460-2466.1999.tb02784.x
