Real ID campaign raises questions about privacy concerns

Abstract

This study used an advertising campaign focused on the Real ID card to explore the relationship between privacy concerns and information sharing through adoption of the new form of identification. Online surveys were conducted to assess the impact of campaign ads promoting the Real ID, and to determine whether privacy concerns were a barrier to its adoption. At post-test, an ordinal regression analysis of data from registered voters in Montana (n = 302) showed a positive association between the time spent viewing the ads on social media and intent to obtain the Real ID, but no relationship between privacy concerns and Real ID intentions. To date, the states and territories have issued approximately 110 million Real ID-compliant driver’s licenses and identification cards, representing 40 percent of all driver’s licenses and identification card holders. Montana, one of the states originally most resistant to the 2005 security act, lags only slightly behind with 20 percent of drivers and ID card-holders having Real IDs. This research shows that privacy concerns about a new technology or identification system, such as the Real ID, can erode over time. Whether these changes are due to mounting privacy apathy triggered by the public’s perceived lack of control, or a willingness to sacrifice some privacy for national security in the face of rising terrorist threats needs to be researched further.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12198-023-00258-0.

Introduction

Are privacy concerns truly a barrier to adoption of the Real ID? The rollout of the Real ID has proven to be quite a mess in at least half a dozen states, including the most populous state of California (Povich 2019). Technical glitches, delays and miscommunication have undermined the Real ID implementation, causing the federal government to issue multiple extensions for the secure driver’s license needed to travel by air or enter government restricted areas after October 2021.

A handful of states, including Montana, initially delayed implementation of the 2005 Real ID Act – passed in the wake of the Sept. 11, 2001 terrorist attacks – by saying they would not comply due to privacy concerns and objections to federal mandates. The Department of Homeland Security granted 15 states more time to comply – postponing the original deadline of 2008 many times – but is now refusing further extensions. As of May 2023, only driver’s licenses that meet the Real ID security standards can be used to board commercial airline fights or enter secure federal buildings, such as military bases. The Act requires licenses to be marked with a special insignia, indicating that holders have gone through extra steps to prove their identity.

Both public resistance to, and confusion about, the new requirements have placed stress on state motor vehicle agencies, causing some states to hire more personnel and spend extra money to comply by the deadline. To ward off anticipated difficulty, the Montana MVD collaborated with a state university to develop an audience-tailored campaign, featuring historical figures of Merriweather Lewis and Captain Clark, to indicate that the Real ID is consistent with the traditional values of exploration, individualism, and national safety (www.mtrealid.gov 2020).

Given the political climate surrounding the Real ID in 2008, when Sen. Jon Tester (D-MT) showed his objections by shooting the bill with a rifle on the state senate floor, officials were leery to actively promote compliance (The Cato Institute 2008). Instead, they opted for an informational campaign that sought to clarify the identification requirements of Real IDs, and dispel rumors without triggering a feared backlash. The resulting ads with their mild tone, while not shifting public perceptions on a broad scale, correlated with increased likelihood to adopt.

Background

Campaigns to promote public safety are typically hard to do well; but they are even more challenging when rhetoric on privacy is used to encourage popular resistance. Individuals do not like to modify their lifestyle behaviors in response to government mandates (Garrett 2020; Maio et al. 2007). Mixed feelings and beliefs make any public health or safety policy based on individual choice complex. People often may know what they should do, but choose to act differently, based on convenience, comfort, political and social factors, or concerns about privacy (Adhikari and Panda 2018). Does the popular resistance to actions to promote health and safety stem from genuine concerns about the invasion of privacy, or from beliefs that are cultivated by political and social interest groups?

Because the promotion of public health and safety hinge on individual decision-making, it is important to understand the environmental influences that affect these choices. Maio et al. (2007, pp. 100–101) write, “The two key policy implications here are that generic behavior change campaigns may have different effects on individuals from different socioeconomic and cultural backgrounds and that it may be important to identify ‘predatory’ environmental influences on individual choices.” For example, price differences may act as environmental deterrents to healthy food choices among low-income households. Or, television food advertising may get in the way of parents’ efforts to persuade their children to make healthy choices. Such influences may include actions by business and political leaders to downplay risks (e.g., tobacco smoking and vaping) by invoking communal norms such as individual freedom and privacy. These authors propose, and it is likely prudent in the case of public safety campaigns, that competing factors be identified when promoting a behavior change message. Without consideration of organized belief factions, health promotion campaigns (e.g., regulations on fast food advertising and portion sizes) have met numerous barriers (Pluviano et al. 2019). By contrast, cultural factors associated with young adulthood were necessary for the success of designated driver campaigns (Winsten and DeJong 2001). Hence, promotion of the Real ID required a careful understanding of the political or psychological forces operating to resist the stepped up information sharing required for secure driver’s licenses.

Privacy concerns have interfered with healthy consumer decision-making around behaviors from seat belt use to the prevention of terrorism. The debate over how to protect individual privacy has pitted privacy against the free marketplace and the common good, construing our economic or public health as in competition with our right to privacy (Schwartz 2000; Angst 2009). By framing the available solutions as a choice between top-down state control versus bottom-up consumer choices, the conversation about how to regulate privacy in cyberspace has levied the invisible hand of the market against the heavy hand of the state, leading most consumers to favor self-regulation and consumer choice (Schwartz 2000). Researchers have written about cost–benefit decision-making practices with regard to information sharing. Anderson and Agarwal (2008) suggest that people will sacrifice some degree of privacy if the rewards or incentives are deemed to outweigh the risks, while Derlega et al. (1993) propose that people will share information only if they believe it will be used in a way that does not harm them in the future. Angst states, “What is intriguing is that extant research does not explicitly draw the connection to societal good but instead focuses on concern for privacy and the potential benefits and costs that may accrue to the individual” (p. 173).

As technology has invaded our lives, concerns about privacy have grown (Angst 2009). Whether expressed as the resigned grumbles of individuals, protests of advocacy groups or eloquence of politicians, privacy has been seen as a casualty of progress. Traditional social norms that historically governed the flow of information in social contexts, such as education, health and safety, have been upended (Nissenbaum 2009). As anxiety about threats to privacy have mounted, so too has the debate about how it should be protected. Many have sought to protect privacy by protecting people’s right to control information about themselves, i.e., restricting the flow of information. In contrast, Nissenbaum (2009) argues that people care more about ensuring that information is used appropriately, rather than restricting it altogether. Schwartz (2000) argues that this debate sets up a flawed paradigm for privacy, one that hinges on the right to control one’s personal information, which he calls “privacy control.” This popular view of privacy as personal property ignores the limits on an individual’s ability to determine what happens to their personal data – especially in cyberspace; and the pitfalls of viewing personal data as a controllable possession. The view of privacy control also ignores the context in which privacy operates, recognized by theories like the Communication Privacy Management (CPM) Theory. Baruh and Cemalcilar (2014, p. 166) state, “The level of privacy that an individual may enjoy is not determined in a social vacuum. First, it has been argued that individuals can enjoy privacy only to the extent that others have a ‘similar minimum level of privacy.’ Second, disclosure of private information creates a codependency between the discloser and the recipient, making both responsible for the protection of information.” In this view, the level of privacy that any individual enjoys will be contingent on the choices that others make.

In addition to the idealistic view that private data can, in fact, be controlled, is the opinion that people, when given the chance, will make healthy choices about their own personal data. Angst states, “Naturally [opt-in choices for electronic medical records] leads to the question of whether the general consumer is capable (intellectually and technologically) of making such decisions and if their choices could negatively affect their care” (2009, p. 173). The idea that people will make rational choices to protect their own privacy is flawed; just like the notion that people will make rational choices when it comes to public health and safety. Behavioral economics has long shown individual inertia towards default terms of industry privacy agreements; the tendency to simply accept online privacy statements can limit free choice when it comes to information sharing. In other words, we always click “yes” (Solove 2012). People’s fear of the hassle or dislike of a message may prompt people to avoid it and mindlessly acquiesce (Witte 1995; Schwartz 2000).

The state, therefore, may have an important role to play in ensuring appropriate use of information to protect consumers’ right to privacy, and in promoting decision-making by individuals to promote public health and safety. Schwartz (2000, pp. 815–860) argues, “the State has a special role in two areas: (1) creating and maintaining conditions for a functioning privacy market, and (2) developing privacy norms that prevent access to personal information that would cause too great a rate of preference falsification in society.” The state, in Schwartz’s view, can help prevent “norm entrepreneurs” from being excessively meddlesome, by either zealously expanding the areas of social life regulated by norms or by inducing excessive compliance. In the case of Real ID, the state may no longer be needed to promote acceptance, since opposition to this new system seems to have eroded over time. State involvement may be necessary, however, to ensure the appropriate use of personal information shared online. Since the Real ID is only one step towards public safety, surely state involvement will be necessary to promote future behaviors that require individual decision-making, and that may be viewed as threats to privacy.

Intriguing research questions abound in this domain, not the least of which is the ethical question of whether privacy-advocacy groups are truly acting in the interest of the general public or a select group of “privacy fundamentalists” (Harris Interactive and Westin 2002). There appears to be a delicate balance between acting in self-interest (some might argue that self-preservation is a more accurate depiction) or acting in the interest of society as a whole (Angst, p. 175).

Pivot on privacy attitudes

Concerns for the protection of personal privacy have repeatedly been raised as an objection to the Real ID. When the 2020 deadline for federal Real ID compliance was looming, Montana was one of the last states to implement the Real ID Act. State politicians in 21 states strongly resisted and publicly voiced their objections to the Act for several years, creating an uncertain environment around how the public would adopt the upcoming changes. In a 2008 U.S. Senate hearing, Senator John Tester (D-MT) voiced concern that rather than protecting Americans, the new IDs would foster identity theft and lead to the adoption of a mandatory national identity card (United States Senate 2008, p. 7). In other states, anti-Real ID statutes included phrases describing the Act as an invasion of privacy, and resolutions specifically referred to fears of government surveillance or identity theft that might result from the shared database system the act requires. In an effort to understand the key drivers of such resistance nationally, Regan and Deering (2009) found that “…states with stronger privacy orientations were also more likely to pass bills or resolutions in opposition” (Regan and Deering, p. 477).

A textual analysis of news reports and academic literature on the topic indicates that opposition was also related to costs and inefficiencies (Reagan and Deering 2009). Senator Tester expressed concern that the states would be forced to take costly measures such as building databases, redesigning licenses, and retraining DMV employees with no hope for reimbursement from the federal government (United States Senate 2008, p. 7). However, our state-wide data (n = 329) collected in Montana in 2018 did not reveal much resistance to Real ID adoption based on cost, inefficiencies or privacy. In an ordinal regression of variables of interest, Privacy (p < 0.10) (measured by agreement with the statement, “The Real ID is an invasion of privacy.”) was negatively related to the dependent variable, Timing of Intent to Obtain (“How soon do you plan to get a Real ID?”). The stronger a person valued privacy, the later they intended to adopt the new ID. However, the odds ratio for privacy was less than the other two variables of interest – those who embraced the law, and those who already feel safe and have no opposition to the law – indicating that privacy, while important to many people, is not an especially salient driver for adoption of the Real ID.

The public mood in relation to privacy has been shown to fluctuate. After 9/11, public acceptance of government surveillance increased (Westin 2003). “The magnitude and shock of the terrorist attacks, the likely continuation of terrorist actions within the U.S. ‘homeland,’ and the dangerous new types of weapons and techniques involved have promoted national security interests into an urgency unmatched in American experience” (Westin p. 449). Based on a belief that the federal government is responsible for defending citizens from terrorist attacks, public support for increased investigative powers, including the adoption of a national citizen identification system, and financial monitoring, have soared (Harris Interactive and Westin 2001). At the same time, however, the public expressed strong concerns about the way these powers might actually be used by U.S. law enforcement, such as surveillance without justification, checking of communication by innocent people, and surveillance for crimes other than terrorism (Harris Interactive and Westin 2001).

By September, 2002, support for some surveillance and law enforcement measures continued while support for others declined (Harris Interactive and Westin 2002). Westin (2003) predicted that future attitudes would hinge on the presence of successful terrorist attacks and/or published accounts of how the government is using its new powers, whether it is operating with care or abuse. Indeed, by the time the COVID-19 pandemic rolled around, citizens were, once again, leery of government overreach (Park et al. 2020). It is unclear, however, whether the opposition to government mandated behaviors – be it mask-wearing or Real ID adoption—was grounded in genuine concerns about privacy, or dislike of being told what to do.

Inoculation messages

In order to promote public acceptance of the Real ID and other safety or health measures, increased tolerance of government surveillance or investigative powers needs to continue. However, people do not always react to health and safety threats in a rational manner (Ivanova et al. 2016). “For some, the uncertainty and subsequent fear arising from [violent acts or threats] creates a ‘sense of powerlessness,’ causing them to feel ‘they are at the mercy of dangerous forces beyond their control.’ This emotional reaction can instigate seemingly irrational behavior” (Ivanova et al. p. 382). In a climate of fear, political factions or belief groups can subvert popular anxieties to turn against logical safety measures (Romano 2020).

In response to the extreme fear caused by politically motivated violence, effective messages that bolster community resilience can speak to the concerns of citizens and responders. Inoculation Theory is a social psychological/communication theory that explains how an attitude or belief can be maintained against persuasion or influence in much the same way a body can be protected against disease–for example, through pre-exposure to weakened versions of a stronger, future threat (Ivanova et al. 2016). Communication before a threat, in this view, has the ability to inoculate the public against extreme fear and doubts towards public agencies, bolstering public resilience – and possibly willingness to share information for the public good (Ivanova et al. p, 382).

Social media marketing, public health and safety

Public health and safety agencies have an opportunity to use social media to engage audiences to foster responsible choices; however, many agencies are still in the early adoption phase when it comes to social media, using it for information dissemination alone, rather than audience engagement and decision-making guidance (Thackeray et al. 2012). Engagement is a consumer’s psychological state arising from the sharing and interacting with content, a dynamic process that generates value in the online marketplace. Although progress has been made in utilizing the rapid format of social media for health information dissemination during the COVID-19 epidemic (Chan et al. 2020), much remains to be learned about how to maximize consumer engagement without compromising privacy rights. One solution for marketers has been to adopt traditional tactics of public relations (Gesualdi 2019; Tsai et al. 2011). With the rise of regulations on the use of personal data, such as the General Data Protection Regulation, marketers have focused on building public trust and promoting, at least the perception of, mutually beneficial relationships. Such strategies include the use of social media influencers (e.g., micro celebrities), content marketing, and brand journalism. Yet, these approaches raise many questions. For example, who is responsible for managing a crisis when an influencer takes actions that tarnish a brand (or, in the case of health and safety messaging, harm the message)? How much personal data is acceptable to use in targeted or tailored marketing? What do agencies and companies do with unsolicited Facebook comments that bely an intended message (Wiyeh et al. 2019)? More research is needed on how to manage personal data without sacrificing the protection of public health and safety in the age of social media. Part of this work will necessarily involve, as stated by Mosteller and Poddar (2017, p. 27), “understanding what motivates consumers to share information on social media, [and] what prevents them from sharing information online [to] provide insights into barriers to engagement.” Thus, this research focuses on how privacy concerns have affected the sharing of personal information to obtain secure driver’s licenses.

The real me campaign

The Montana MVD campaign, entitled “The Real Me,” was developed based on audience research undertaken in 2017 and 2018, about attitudes towards and knowledge of the Real ID. Results showed Montanans did not resist the new identification cards, but were confused about how and when they would be implemented. Results of our 2018 survey of Montana residents, showed that 57.3% planned to get it within the next 6 months, and 24.6% planned to get it within the next year. While confusion did not show strongly as a predictor of resistance in our survey findings, focus group research with 40 Montana residents, ages 18–55, showed a great deal of misinformation about the Real ID. When asked what they knew about the Real ID prior to the focus group meeting, respondents said, “Didn’t know a thing;” “Zero. I know nothing;” and “Well, I have been hearing things. But I thought it had already happened or wasn’t going to happen or something. I was surprised to find out that it’s just now happening in 2018.”

Thus, an informational campaign – featuring historical characters familiar to Montanans, Lewis and Clark – was designed to deliver the relevant information in a straightforward and direct manner. The branding and slogan, “The Real Me,” was intended to appeal to the sense of authenticity important to Montanans, and emphasize the role Real ID plays in national security. In the campaign ads, aired on radio, TV and social media, the characters of Lewis and Clark find themselves in situations where they would need a Real ID, such as trying to fly to avoid using their canoe or hiking. The goal was to use the Lewis and Clark characters to raise awareness about Real ID and drive the audience to visit the website (www.mtrealid.gov) to learn more about it. The ads did not address privacy concerns; rather, they attempted to gain viewers’ and listeners’ attention through humour in order to drive audience members to the state website to learn more information about the process of obtaining the new license or ID card (see Attachment A, B & C).

Although inoculation messages to promote public security measures in the wake of 9/11 or COVID-19 may have been sensible, erosion of public opposition to the Real ID Act seems to have undermined the need for such sophisticated marketing campaigns in this instance. No good data is available to document the level of public resistance to the Act at the time of its passage (Keller et al. 2020). The fact that numerous state governments opposed the federal mandate with anti-Real ID statutes over the past 10–15 years, does not confirm the presence of popular opposition – either then or now. Such state government resistance may only have represented the opinions of politicians and special interest groups.

Hence, this research was undertaken to assess the impact of the campaign and to determine the role of privacy concerns, leading to the following research questions:

• RQ1: Do individuals exposed to the campaign have greater awareness about the Real ID Act compared to individuals not exposed to the campaign?

• RQ2: Do individuals exposed to the campaign have greater intentions to obtain Real IDs than those who did not see the ads?

• RQ3: Do individuals with stronger privacy concerns have lower intentions to get the Real ID?

Methods

An online survey was conducted in Summer 2020 to evaluate the impact of a marketing campaign, entitled “The Real Me,” undertaken by the Motor Vehicle Division to clarify Montana residents’ understanding of the Real ID Act. A random sample of 5,000 was generated from a statewide list of registered voters. Respondents were contacted online by Gravis Marketing, a market research company hired for the project, and invited to complete the survey anonymously through a link to Qualtrics Survey Software. Questions probed on intent to get a Real ID, possession of a Real ID or passport, exposure to the “Real Me” marketing campaign ads, and attitudes about the Real ID. The study was approved by the university’s Institutional Review Board, given that respondents remained anonymous and neither researcher received any personal compensation for the study; all funds provided by the MVD were used to purchase the mailing list and hire a research assistant.

The questionnaire was a modified version of an instrument that had been previously pre-tested with a sub-sample of Montana residents and administered in 2018, to gather baseline data for the development of the campaign described in the present study. Modifications to the instrument included the addition of a privacy scale and questions about campaign recall, exposure, and attitudes. A simple random sample of 5,000 Montana voters was generated from a statewide list of approximately 752,000 registered voters by a market research company hired for the project. The instrument probed on attitudes and awareness of Real ID. The privacy scale measured two dimensions of concerns about privacy [privacy as a right, and informational privacy] (Baruh and Cemalcilar 2014). Approximately 302 responses were obtained, skewed towards a more senior and educated demographic than the state population. Once it was discovered that both upper income and lower-income individuals were less likely to be enthusiastic adopters, we felt confident that the sample would show, if anything, more reluctance to adopt Real IDs than might be present in the population. The fact that age was not correlated with our attitudinal outcomes indicated that age was not a significant source of bias.

No experimental design was employed. We used regression analysis to identify predictors of the timing of intent to obtain, which became our dependent variable. Results were analyzed using SPSS software to identify the frequencies of common barriers to the ID, frequencies of early vs. late adopters, and campaign exposure.

Ordinal logistic regression was used to identify predictors of variance in the timing of intent to obtain a new ID. In this analysis, the timing of the intent to obtain was used as dependent variable, and all other variables were treated as independent variables. While a structural equation model is preferable with interval level data and more than one dependent variable, ordinal regression is appropriate when the dependent variable is an ordered category. Ordinal regression has more statistical power than multinomial regression., the latter of which is appropriate when the dependent variable is categorical, but not ordered (Issa and Kogan 2014).

Several actions were undertaken to mitigate the problem of common methods variance, which is a potential problem in self-report surveys. First, the questionnaire was constructed in a way to lessen this possibility (MacKenzie and Podsakoff 2012), separating the items of interest in the survey instrument (Podsakoff and Organ 1986), and carefully wording each item in the study (Tourangeau et al. 2000).

Results

Variables and descriptive statistics

Of the 5,000 registered voters who were survey, 302 responses were obtained. These were skewed towards a more senior and educated demographic than the state population. Responses were interpreted with this precaution in mind, by considering how the older, more affluent characteristics of the sample might influence responses. Table 1 displays the frequencies of the variables used in the analysis. A total of 302 valid responses were obtained. Of these, 2.0% (6) were 18–25; 3% (9) were 26–35; 12.9% (39) were 36–45; 18.2% (55) were 46–55; 23.5% (71) were 56–65; 29.8% (90) were 66–75; and 6.6% (20) were older than 76, and 12 (4%) were missing. The sample is skewed to an older age than the actual population of Montana which is spread more evenly between the age groups up to age 65, and then declines. The genders were fairly evenly split in our sample: 57% (172) were female, and 42.7% (129) male. Participants came from all over the state, with the largest number of respondents coming from Yellowstone County (17.9%, followed by Gallatin County (17.2%), Cascade County (12%), Lewis and Clark County (11.3%), Flathead County (9.6%), Missoula County (7.6%), and many smaller percentages in other counties and tribal nations around the state. This is a good representation of the state except that Yellowstone County is the largest county in the state.

Table 1.

Demographic and Dependent Variable Characteristics

Variables	Categories	Frequency	Percentage
Sex	Male	129	42.7
	Female	172	57
Age	18–25	6	2.0
	26–35	9	3.0
	36–45	39	12.9
	46–55	55	18.2
	56–65	71	23.5
	66–75	90	29.8
	76-up	20	6.6
Annual Income	Under $10,000	7	2.3
	$10,000-$29,000	28	9.3
	$30,000-$49,000	47	15.6
	$50,000-$74,000	64	21.2
	$75,000-$99,000	50	16.6
	$100,000-$249,000	47	15.6
	$250,000 0r more	9	3.0
	Prefer not to say	50	16.6
Intent to Obtain Real ID	Never	52	20.1
	More than 1 Year	66	21.9
	7 Months – 1 Year	67	22.2
	2 Months – 6 Months	54	17.9
	3 Weeks – 1 Month	14	4.6
	1 Week – 2 Weeks	6	2.0
	Missing	43	14.2

Discrepancies in these frequencies is due to missing data

Incomes of respondents were also fairly well distributed. The largest percentage of respondents (21.2%) reported incomes between $50,000-$74,999, followed by $30,000-$54,999 (15.6%), $75,000-$99,999 (16.6%), $10,000-$24,999 (9.3%), and many who preferred not to say (16.6%).

Table 1 also displays frequencies for Timing of Intent to Obtain, consisting of six, ordinal variables responding to the question: “On available, how soon do you intend to get your Real ID? These are 1 = never (20.1%), 2 = more than 1 year (21.9%), 7 months – 1 year (22.2%), 2–6 months (17.9%0, 3 weeks – 1 month (4.6%), 1–2 weeks (2.0%).

Independent variables of interest

The independent variables consisted of the demographic indicators described above, plus the four items of interest; privacy, television, radio, and social media. The last three variables were operationalized by asking respondents to use a sliding scale indicating their contact with Real ID messaging in each of these media in terms of exposure in minutes, on television, radio, or social media. Concerns about privacy were measured by the privacy orientation scale established by Baruh and Cemalcilar (2014) which includes four dimensions of privacy, of which we used two (informational privacy and privacy as a right). How respondents view informational privacy was measured by asking them if they are concerned about: their personal information being be shared by others, people knowing too much about them, the consequences of people sharing identity information, and sharing information with more people than intended. Concerns about privacy as a legal construct (i.e., a “right”) was measured by asking respondents how much they agreed or disagreed with the need to strengthen privacy laws; the need for legal protection against misuse of their personal data, and whether privacy should be encoded as a fundamental right (Baruh and Cemacilar).

Table 2 displays the results of a factor analysis using Varimax Rotation. Two factors were produced with eigenvalues above 1.00 which together, explain 73.29 percent of the variance. The factors loading aligned as expected in accordance with the previously validated questions of Baruh and Cemalcilar (2014). The first, Informational Privacy, consists of a four item scale assessing respondents’ views in terms of personal information privacy (Cronbach’s Alpha 0.87), while the second privacy variable, Privacy Rights (Cronbach’s Alpha 0.87) consists of three items assessing respondents’ views of privacy as an individual, legal right. The factor analysis showed that respondents cared most about informational privacy (57% of the variance in privacy attitudes was represented by this dimension, Table 2), and less about privacy as a right (16% of the variance was captured by the second dimension). By contrast, Baruh and Cemacilar found that privacy advocates (people who scored high on the belief that privacy is a right, and were concerned about protecting the privacy of others) were the most frequent segment in their data. Their least frequent audience segment, privacy individualists, seemed most similar to our respondents who are concerned not sharing information, but don’t view privacy as a right.

Table 2.

Factor Analysis

Item	Rotated Components (Varimax with Kaiser normalization)
	1	2
	Informational Privacy	Privacy as a Right
Initial Eigenvalues Total	3.94	1.19
Percent of Variance	56.33	16.96
Cronbach’s Alpha	.87	.87
Shared by others	.866	.176
People know too much	.853	.161
Consequences of sharing identity information	.717	.384
Sharing with more people than intended	.805	.283
Strengthen privacy laws	.228	.857
Need legal protection	.219	.829
Privacy a fundamental right	.238	.792

Within informational privacy, respondents reported higher concern for having their information “shared by others” (F = 0.866) than the “consequences of sharing their identity information” (F = 0.713). This contrasts with research by Derlega et al. (1993) on the calculus of privacy, whereby people were very concerned with the consequences of sharing information, and would only be willing to share their identifying information if they trust that it will not be used in a way that affects them negatively in the future.

Privacy concerns

Contrary to our expectations, concerns about privacy had no significant relationship with timing of intent to obtain a Real ID (Table 3).

Table 3.

Privacy not significant

Variable		Coeffecient	SE	Wald Chi-Square	Odds Ratio [Exp(B)]
How soon do you plan to get a Real ID?	Never	-.89	.738	1.45	.412
	More than one year	.32	.739	.19	1.38
	7 mos – 1 year	1.42	.746	3.63*	4.14
	2 weeks – 6 mos	2.99	.769	15.13***	19.86
	Immediately	4.25	.838	25.71***	362.64
Privacy as a Right		-.04	.200	.03	.96
Informational Privacy		.19	.159	1.46	1.12

Table 3 display the dependent variable, Intent to Obtain the Real ID, against the same two measures of privacy. Neither of the privacy variables are statistically significant. This indicates that we found no evidence that concerns about privacy are related to the timing of the intent to obtain the Real ID. Of course, the absence of evidence does not mean that such an association does not exist, but we did not find support for such a relationship.

Ordinal regression analysis

Most respondents (71%, 214) said they had seen the Real ID promoted in advertisements. Table 4 displays the results of an ordinal regression analysis using Timing of Intent to Obtain as the dependent variable. The result indicate a positive association between the time spent viewing the ads on Social Media and Intent to Obtain the Real ID (p < 0.1) (Garson 2014; Stat Press 2013). However, neither Television (p = 0.48), nor Radio (p = 0.43) was statistically significant. In ordinal regression the dependent variable is associated with values whereby larger values are related to higher outcomes. Effect size consists of the cumulative odds ratio, which is the probability that the “dependent variable is less than or equal to the given category, divided by … the probability that the dependent variable is more than the given category” (Garson 2014). This is in contrast to the standardized b coefficients, which are used to assess and compare effect sizes in OLS regression. In ordinal regression, while holding constant the other independent variables, a unit increase in the independent variable is associated with the multiplicative change in odds, or odds ratio, odds of the dependent variable being in a higher category (Crowson 2019). The odds ratio for Social Media is (Exp(B) in SPSS) 1.012, indicating only a slight increase in the intend to obtain the Real ID in relation minutes spent exposed to the campaign on social media. Lacking a passport (Exp(B) = 2.32, p < 0.01), or a Military ID (Exp(B) = 1.85, p < 1.0) had a much greater impact on the likelihood of respondents’ intent to obtain the Real ID.

Table 4.

Dependent variable: How soon do you plan to obtain a Real ID?

Variable		Coeffecient	SE	Wald Chi-Square	Odds Ratio [Exp(B)]
Sex (female)	.37	.23	2.69*	1.45
Under $10,000	-.91	.82	1.26	.40
$10,000-$29,000	-.05	.47	.01	.95
$30,000-$49,000	.01	.40	.00	1.005
$50,000-$74,000	.17	.37	.21	1.19
$75,000-$99,000	-.22	.39	.32	.80
$100,000-$249,000	.39	.41	.91	1.48
$250,000 0r more	2.00	.90	4.91**	7.36
No Passport	.84	.27	9.57***	2.32
No Military ID	.62	.36	2.87*	1.85
Television	-.00	.00	.48	1.00
Radio	-.00	.00	.43	1.00
Social Media	.01	.00	3.49*	1.01

^*** ≤ p < .01, ** = p < .05, * ≤ p < .10

(Garson 2014; Stat Press 2013)

Conclusion

In conclusion, while the data are not definitive due to the non-matched samples from baseline to post-test, it appears that exposure to the Lewis and Clark advertising campaign was positively associated with intent to get a Real ID. People who were exposed to the ads on social media were more likely to have a Real ID, or intend to get one in the near future, regardless of their opinion of the ads. Since neither TV nor radio ads were significantly related to Real ID intent, we can conclude that the social media ads were more effective.

The authors conclude that the distribution methods and channel selection of radio and social media seemed to have provided a widely received campaign and should be repeated. The authors recommend a similar approach for future national resilience campaigns, with a focus on clarity of information by continuing to link the recommended action (in this case, the Real ID) to historical traditions familiar to the target audience. Of the tested attributes – neither concerns about privacy, perceptions of national security, nor confusion – stood out as a significant barrier to getting the new identification. Nor did any demographics.

The lessons learned through this study can be used to inform the current federal failure to lead the nation to public safety in fighting the Coronavirus pandemic. It is clear that the U.S. public and many lawmakers have a strong anti-federalist sentiment and are resistant to top-down federal measures, especially when such mandates interfere with individual lifestyle behaviour choices. A recent content analysis of discussion boards on news reports of case counts on Facebook found significant resistance to compliance with a public mask mandate (64% of relevant comments), with the most common reasons being beliefs that masks are not effective in preventing infection and suspicion that the seriousness of the pandemic is being overblown by the media and/or government officials (Keller et al. 2021). Of particular concern was the prevalence of comments suggesting a conspiracy to mislead the public and enact government control over the citizenry. Concerns about government intrusion on privacy rights are frequently touted as a reason to oppose vaccination, social distancing or mask-wearing.

The fact that privacy concerns were not related to Real ID attitudes in this study raises some important questions, not just about the Real ID, but about acceptance of government mandates in general. People say they really care about privacy as a right (e.g., rhetoric of individual rights) and the right to protect their personal information – but they do not routinely behave that way (e.g., indiscriminately sharing information online). Baruh and Cemalcilar (2014) found that privacy advocates – despite their strong concerns – were less likely than other segments to use privacy settings when using Facebook. Despite high levels of privacy concerns, individuals often choose to share their personal information for the purposes of social networking (Fatima et al. 2019).

In this paper, we have discussed how privacy concerns can fluctuate over time; represent a range of sometimes contradictory social concerns and practices; be influenced by predatory factors; and occupy multiple forms. Privacy is shown to have multiple dimensions by this and prior work by Baruh and Cemalcilar (2014). People who are strong privacy advocates (people who see privacy as a right) are also concerned about the privacy of others, and are less likely to use social networking to satisfy their social curiosity. By contrast privacy individualists (people concerned with informational privacy) do not seem to care about others’ privacy, and have been shown to practice voyeuristic behaviour on social media (Baruh and Cemalcilar; Baruh 2010).

Do people use privacy, then, as a reason not to comply with COVID prevention, vaccines and other government mandates? Does talking about privacy give people a false sense of security? Is privacy a galvanizing term, thrown about to catapult political opposition around the cause of the moment? With the arrival of the vaccine, privacy has again become a front and center issue. Do people have a right to know whether or not you have been vaccinated? Should a business be able to exclude customers because they cannot produce evidence of immunity to a contagious virus?

Our findings indicate that, when confronting a national threat that requires lifestyle behaviour change on a societal level, the government may have a role in shaping norms about privacy and regulating the market of information. In the case of national emergencies of terrorism, pandemics or disasters, societies will need to determine when sacrifices of personal freedom (or privacy) will be warranted for the end goals of safety and peace (Angst 2009). To parlay this conversation, messages of urgency need to strike a balanced, unbiased tone; a tone that simultaneously communicates caution and invokes the values of independence and freedom. “If the threat of terrorism can ultimately prompt the U.S. public to put down the sword of privacy and adopt Real IDs, then the threat of COVID-19 [and future global viruses] should be able to inspire us to envision a society free of this virus as consistent with our goals of freedom and public safety” (Keller et al. 2021, p. 5641).

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Sarah N. Keller, Timothy Wilkinson and A. J. Otjen. The first draft of the manuscript was written by Sarah N. Keller and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding

This work was supported by a grant from the Montana Department of Justice, Motor Vehicle Division report.

Data availability

Data and materials are not available.

Declarations

Competing interests

There are no relevant financial or non-financial competing interests.

Conflict of interest

The authors have no conflict of interest to report.

All research with human subjects was approved by the Institutional Review Board of Montana State University Billings. Informed consent was obtained from all participants, and no personally identifying information was obtained. All participation was voluntary.