How we did this
Pew Research Center has intensively studied the impact of the coronavirus outbreak, the public health dimensions of the pandemic and Americans’ views about some of the privacy implications of data collection during the outbreak. This report focuses on key parts of the contact tracing process that states are trying to implement to identify, trace the contacts of and isolate those with COVID-19 in order to disrupt how the coronavirus is transmitted.
To explore this, we surveyed 10,211 U.S. adults from July 13 to 19, 2020. Everyone who took part is a member of the Center’s American Trends Panel (ATP), an online survey panel that is recruited through national, random sampling of residential addresses. This way nearly all U.S. adults have a chance of selection. The survey is weighted to be representative of the U.S. adult population by gender, race, ethnicity, partisan affiliation, education and other categories. Read more about the ATP’s methodology.
Here are the questions used for this report, along with responses, and its methodology.
As states mount large-scale contact tracing efforts to identify and isolate those who have contracted COVID-19, a Pew Research Center survey conducted in July finds that Americans have a variety of views that could complicate the ongoing efforts of public health authorities battling the outbreak.
On the one hand, majorities of Americans say they would be at least somewhat comfortable or likely to engage with some parts of contact tracing programs – long-established public health processes that try to limit the spread of potentially deadly infectious diseases such as AIDS, Ebola and now COVID-19 by disrupting the chain of transmission.
For example, this survey finds that 58% of U.S. adults say they would be very or somewhat likely to speak with a public health official who contacted them by phone or text message to speak with them about the coronavirus outbreak. Roughly three-quarters (77%) report they would be at least somewhat comfortable sharing information with a public health official about the places they have recently visited. A smaller share – 49% – say they would be similarly comfortable sharing location data from their cellphone. And fully 93% of adults say they definitely or probably would quarantine themselves for at least 14 days if they were told they should do so by a public health official because they had the coronavirus.
Yet the survey also shows that portions of Americans could be hard to reach and relatively uncomfortable engaging with public health officials as part of the contact tracing process related to the coronavirus outbreak. For instance, 41% of those who were asked about their views on speaking with a public health official who might contact them about the coronavirus outbreak via phone or text say they would be not at all or not too likely to do so. A similar share (40%) of those who were asked about speaking with a public official who showed up at their residence to talk about COVID-19 say the same thing.1
Many factors could influence Americans’ participation in contact tracing and quarantine programs. This survey asked U.S. adults how they might behave in three major aspects of contact tracing in the context of the coronavirus outbreak: the likelihood that someone would speak with a public health official (i.e., a contact tracer) who contacted them about the coronavirus; the degree of comfort that someone would have in sharing information like the names of people with whom they have been in physical contact and the places they have recently visited, or data from their cellphone that tracked their locations; and their willingness to quarantine for 14 days if they were advised to do so by a public health official. We refer to these three steps as “speak,” “share” and “quarantine” throughout this report.
Overall, taking account of the public’s wariness with some parts of the contact tracing process, the survey shows that 48% of U.S. adults say they would be comfortable or likely to engage with all three key steps – speaking, sharing and quarantining.
If public health officials could break through people’s wariness of speaking with a public health official by phone, the results also show that another 21% might be willing. These are the individuals who say they would be very or somewhat comfortable sharing information and definitely or probably would quarantine, but also say they would be less likely to speak with a public health official by phone or text message in the first place.
Determining who is comfortable or likely to engage with contact tracing programs
This survey looked at Americans’ views of three key steps of contact tracing related to the coronavirus outbreak that would be initiated by a public health official. Two different concepts related to people’s potential willingness were measured: likelihood of speaking with a public health official by phone or text message or quarantining, and comfort with sharing information. Together, we use these questions to try to tap into people’s “willingness” and “openness” to key steps of contact tracing, compared with their “wariness.”
Of course, people’s actual cooperation with public health officials dealing with the coronavirus depends on many factors. Some individuals might be willing but uncomfortable. Others might express one view and then act differently upon picking up the phone. Still others might not exactly know what they think or might behave differently based on how the pandemic is affecting their community at the time.
In places in this report that examine how individuals might behave at all three stages of the contact tracing process, we used data from the same random subset of respondents who were asked about 1) the likelihood they would speak with a public health official who contacted them by phone or text message to speak with them about the coronavirus outbreak; 2) their comfort sharing information about the people with whom they have been in contact and where they have been (either via names of places they’ve recently visited or location data from their cellphone); and 3) and whether they would act on advice to quarantine for 14 days if advised to do so by a public health official because they had the coronavirus. In this report, we refer to these by the shorthand “speak,” “share” and “quarantine.” (While the full sample of individuals was asked about sharing information, a random subset was asked the relevant “speaking” and “quarantine”-related questions.)
We then looked at the various combinations of responses to those questions. Some say they would be comfortable or likely to engage across the board – that they would be likely to speak, comfortable sharing and would quarantine. Others report they would be likely or comfortable in some ways, but not across the board. A small share tell us would not likely speak or be comfortable sharing anything, and would not quarantine. We use fairly expansive criteria in creating a measure of those comfortable or likely to engage. Individuals must respond that they are “very” or ”somewhat” likely to speak with a public health official and “very” or “somewhat” comfortable sharing relevant information and “definitely” or “probably” would quarantine.
A full grid of all the possible combinations of respondent answers can be found below this box. For a complete description of our calculation method, see the section of the report entitled “In all, 48% of adults say they would be comfortable or likely to engage with all three key steps of the contact tracing process.”
This study also examines several other factors that might affect people’s level of engagement and comfort with the contact tracing process. First, it looks at Americans’ openness to answering phone calls from unknown numbers and finds some evidence that even starting the contact tracing process might not be easy in many cases. Just 19% of Americans say they generally answer their cellphones when an unknown phone number calls. Some 67% say they don’t answer but would check a voicemail if one is left. And 14% say they generally don’t answer and would ignore a voicemail.
At the same time, shares of Americans say people pretend to be someone else in order to steal other people’s personal information with some frequency. Some nine-in-ten Americans think it is often (49%) or sometimes (42%) the case that people do this.
Next, the survey probes another dimension of Americans’ views on quarantining specifically, by asking how difficult it would be for people to isolate themselves for 14 days because they had the coronavirus – regardless of whether or not they would actually do so. About three-in-ten Americans (32%) say it would be very or somewhat difficult to quarantine. Among those who would find it at least somewhat difficult, 40% say that having too many other obligations is a major reason for this difficulty, and about the same share (39%) say being unable to miss work would be a major reason.
Additionally, the July survey looks at people’s views about what happens to their personal information once it is in the hands of others. Half of Americans say they are not at all or not too confident that the federal government will keep their personal records safe from hackers or unauthorized users.
For some, those concerns also apply to public health organizations. About four-in-ten Americans (41%) say they are not at all or not too confident that public health organizations will keep their personal records safe. Republicans and those who lean to the Republican Party are more likely to say this than Democrats and Democratic leaners (51% vs. 31%).2
The survey finds that people’s phone-answering habits as well as their views about their personal records and how prevalent they think scams are relate to whether they would be comfortable or likely to engage with contact tracing steps. Those who usually ignore both calls and voicemails from unknown numbers, those who think people often pretend to be someone else in order to steal others’ personal information, and those who are less confident that public health organizations will keep these records safe are less likely to say they would be comfortable or likely to engage with the contact tracing process. For instance, 70% of those who are very confident that public health organizations will keep their personal records safe from hackers or unauthorized users say they would be comfortable or likely to engage with all three steps (likely to speak, comfortable sharing, would quarantine), compared with 21% of those who are not at all confident about the security of this data in the hands of public health organizations.
Moreover, those who had seen or heard a great deal or some about contact tracing and what it entails at the time the survey was fielded are more likely to be “comfortable or likely to engage” with all steps of the process than those with less awareness.
There are also differences among some demographic and partisan groups relating to their stated comfort or likelihood of engaging with the contact tracing process. For example, Democrats are more inclined than Republicans to say they are comfortable or likely to engage with each step in the contact tracing and quarantine process. Some 36% of Republicans say they would be comfortable or likely to engage with contact tracing and quarantine protocols according to our “speak, share, quarantine” definition of engagement, compared with six-in-ten Democrats.3
Breaking down the individual steps, 68% of Democrats say they would be very or somewhat likely to speak with a public health official if they were contacted by phone or text about the coronavirus outbreak, compared with 49% of Republicans, for example. And 88% of Republicans report they would definitely or probably quarantine, compared with 97% of Democrats – with about six-in-ten Republicans (59%) saying they would definitely do so, compared with 85% of Democrats.
Americans’ stated comfort with or likelihood of engaging with key parts of the contact tracing process varies across other demographic groups as well. In particular, adults with higher incomes are more likely than those with middle or lower incomes to say they would be likely to speak with a public health official and would be comfortable sharing information. For example, 85% of those with relatively high incomes say they would be very or somewhat comfortable sharing the places they have recently visited with a public health official, compared with 73% of those with relatively low incomes who say that. And a similar pattern follows by levels of formal educational attainment.
These are some of the key findings from a Pew Research Center survey of 10,211 U.S. adults conducted July 13-19, 2020, using the Center’s American Trends Panel. They add to the body of research the Center has conducted since March about the health, economic and political effects of COVID-19. These findings also relate to survey results from September showing 49% of Americans would definitely or probably not get a coronavirus vaccine if one were available today. Additionally, this material expands on insights about people’s wariness about the likely impact of privacy and location-monitoring apps on smartphones in limiting the spread of COVID-19 as tech-focused solutions continue to be developed.
The remainder of this report is structured to map the way data collection and analysis might unfold in a contact tracing process. It starts by discussing public awareness of contact tracing, then proceeds to examine people’s behavior around calls from unknown numbers (as calls from a public health agency might arrive) and related perceptions of how secure their personal information might be. It then examines people’s views about three key steps in the contact tracing process (speak, share, quarantine) and, finally, turns to people’s views about how personal, health-related data is protected and used once it is gathered. Extensive tables about how different demographic and other groups responded to all these questions are available in the Appendix to this report.
70% of Americans say they had at least some awareness of contact tracing as of July
People’s views about contact tracing and related issues like privacy and data security could have a critical impact on the success of contact tracing programs and the trajectory of the pandemic. Public health modeling of the spread of COVID-19 has shown that successful contact tracing can depress the number of infections dramatically if those who are infected are quarantined and contacts and places visited are identified.
As public health officials and organizations work to combat misinformation about contact tracing, they also must try to communicate effectively with the public about its goals and purpose. The July survey found that while the majority of Americans said they had heard about this process as of the survey’s fielding, another share, though a minority, said they had not heard much or anything. Overall, 70% of Americans had seen or heard at least some about contact tracing as of mid-July, with 29% having heard a great deal about the process. At the same time, 29% reported hearing nothing at all or not too much about this process.
Majorities across groups had at least some awareness of contact tracing at the time, but not all Americans had seen or heard equal amounts about the process. About four-in-ten Black adults (39%) and Hispanic adults (43%) said they had not heard much or anything at all about contact tracing at the time, while roughly three-in-ten Asian American adults (31%) and about a quarter of White adults (24%) said the same.4
Younger Americans and women were also more likely to have less awareness about contact tracing. Some 40% of 18- to 29-year-olds said they had heard not too much or nothing at all as of the survey’s fielding, compared with 29% of 30- to 49-year-olds, 26% of 50- to 64-year-olds and 24% of those 65 and older. Roughly three-in-ten men (27%) and women (31%) reported having not seen or heard much, if anything, about this process.
Those who have less formal education and relatively low incomes were also less likely to have heard much about the contact tracing process. Four-in-ten adults with a high school diploma or less formal education had not heard anything at all or had not heard too much as of July, compared with about three-in-ten adults with some college (29%) and 16% of adults with a bachelor’s degree or higher. Awareness of contact tracing similarly increases with income level.
Partisan differences in the amount of information heard or seen about contact tracing were present, but were small. About three-in-ten Republicans (31%) said they had heard nothing at all or not much about the topic, compared with 26% of Democrats.
Majority of Americans do not generally answer their cellphone when unknown numbers pop up, but many of them listen to voicemails
The contact tracing process generally begins when someone who is diagnosed with an infectious disease is asked about the places they have been and the people with whom they have been in contact. Contact tracers then try to track down others who might have come in contact with the infected person. And while some contact tracing efforts rely on text messages or in-person visits, the outreach often begins with a phone call.
In light of general public wariness about calls from unknown numbers, the Center asked in this survey about people’s general practice when an unknown number calls their cellphone – as would often be the case when a contact tracer was trying to get in touch.
The results indicate that the first step in the contract tracing process may be challenged by reluctance to answer unknown calls. Only 19% of Americans report they generally would answer the phone to see who it is; another 14% would not answer and ignore a voicemail if one is left. The majority of U.S. adults (67%) would not answer the phone but would check a voicemail if left.
Small differences are present across demographic groups in terms of general call-answering behavior. Men and younger adults are more likely to answer a call from an unknown number than their counterparts. Black adults (22%) and Hispanic adults (25%) are somewhat more likely to answer the call than White adults (17%), while a fifth of Asian Americans would answer the call from an unknown number. And Black and Hispanic adults are also somewhat more likely to not answer and ignore a voicemail (19% and 17%, respectively) compared with a smaller share of White adults (12%).
Majorities of Americans think others are often or sometimes out to steal people’s information and that their information is less secure than it used to be
This survey also explored the degree to which people are worried about others pretending to be someone else in order to steal people’s personal information – an issue that concerns those running contact tracing efforts because it could affect people’s willingness to answer their phones and provide personal or sensitive information. Some groups are more concerned than others about this prospect.
Fully 90% of Americans think people pretend to be someone else in order to try to steal people’s personal information often or sometimes, with roughly half (49%) saying they think this happens often. While these overall shares are fairly consistent across demographic groups, there is more variation across groups on whether people often pretend to be someone else to steal information.
Black adults (63%) and Hispanic adults (56%) are more likely than White (45%) or Asian adults (35%) to say this type of attempted theft happens often. Additionally, those with relatively low incomes and less formal education are more likely to say people pretend to be someone else in order to steal others’ information often. Roughly six-in-ten of those who have lower incomes say so, compared with about half of those from middle-income households (47%) and about four-in-ten of those with higher incomes (37%).
Similarly, 57% of those with a high school education or less, 50% of those with some college and 36% of those with a bachelor’s degree or higher say this is often the case. There are no differences in people’s views on this when it comes to partisanship.
Beyond their perceptions about people stealing other people’s information, a majority of Americans also feel their personal information is less secure than it used to be. This survey finds that roughly two-thirds of Americans (68%) feel their personal information is less secure compared with five years ago. Some 27% say that their information is about as secure as five years ago, while only 5% say that their personal information is more secure than five years ago. These results are in line with those gathered in a 2019 Center poll on Americans’ views of privacy.
At least six-in-ten Americans across demographic groups say they feel their information is less secure. However, there are some small differences in shares who say this by race and ethnicity, age, education, income, and partisanship. White adults are more likely to say that their information is less secure today than those of other racial and ethnic groups, with about seven-in-ten White adults (71%) saying this compared with about six-in-ten of those who identify as part of other racial and ethnic groups (62% each of Black, Hispanic and Asian adults). Older respondents are also more likely than younger adults to say their information is less secure; 76% of those ages 65 and older say they feel their personal information is less secure now than it was five years ago.
Those with higher levels of formal education and relatively high incomes are also more likely to say their personal information is less secure. Those with a bachelor’s degree or higher (72%) are more likely than those with less formal education to say so. And those with higher incomes (72%) and middle incomes (69%) are more likely to say so than those with lower incomes (63%).
About four-in-ten adults say they would not be likely to speak with public health officials by phone or text message or at their home
Americans’ concerns about the safety of their personal data is an important backdrop to their views about various aspects of contact tracing programs. In addition to coping with Americans’ wariness about picking up their phones and their perceptions about the frequency of scams, contact tracers face another struggle: Some adults say they would not be likely to speak with a public health official by phone or text message or in person.
In this survey, Americans were asked specifically about how they would respond to outreach by public health officials trying to contact them to speak about the coronavirus outbreak. Half of respondents were asked how likely they would be to speak with a public health official on the phone or via text message about the coronavirus outbreak, and the remainder were asked how they would respond to a public health official who showed up at their residence. About four-in-ten Americans (41%) who were asked about their likelihood of speaking with a public health official if they contacted them by phone or text message say that they would be not at all or not too likely to do this, and 40% of those asked about speaking with an official who showed up at their residence say they would be not at all or not too likely to do the same.
People’s stated likelihood of speaking with an official varies somewhat by the mode of contact when examining Americans’ views by gender, age, and race and ethnicity. Women and men are about equally likely to say they would talk with a public health official via phone or text message, but women are less likely to report they would speak in person: 44% of women, compared with 36% of men, say they would be not at all or not too likely to speak with a public health official who showed up at their residence. Younger adults are less likely than their older counterparts to say they would speak with a public health official via phone or text, despite their relative willingness to generally answer the phone for unknown numbers. Some 49% of those ages 18 to 29 say they would be not at all or not too likely to speak with a public health official in this manner, compared with 34% of those 65 and older.
There are also differences in people’s reported likelihood of speaking with a public health official by income and education. For instance, 30% of Americans with a bachelor’s degree or higher say they would be not at all or not too likely to speak by phone or text message, compared with 44% of those with some college and 48% of those with a high school diploma or less; and 35% of those with at least a bachelor’s degree say they would be not at all or not too likely to speak in person, compared with 42% each of those with a high school diploma or less or some college. A similar pattern applies by income. Those with relatively low and middle incomes are less likely to say they’d be likely to speak with a public health official about the coronavirus outbreak compared with those from households with the highest incomes.
Finally, there are clear partisan differences in what Americans say about their likelihood of speaking with a public health official, and the differences are larger when the issue is phone- or text-based contact than for in-person contact. Half of Republicans say they would be not at all or not too likely to do so via phone or text, compared with 31% of Democrats. These differences persist when considering ideology. For example, 52% of conservative Republicans versus 26% of liberal Democrats say they would be not at all or not too likely to speak by phone or text, and 47% of conservative Republicans versus 30% of liberal Democrats say the same about speaking with a public health official in person if they showed up at their residence.
Some express discomfort sharing information with a public health official about where they have been, whom they were in physical contact with
In addition to asking people whether they might engage at all with a public health official during the COVID-19 outbreak, this survey explored another dimension of potential cooperation – comfort with sharing several pieces of relevant information within the contact tracing process. Before they answered this question, individuals were told: “In ‘contact tracing,’ public health officials interview people who have a confirmed COVID-19 diagnosis in order to identify everyone who has had contact with them, and who are therefore at risk of getting COVID-19.” Individuals were asked about their comfort with sharing the names of people they may have been in physical contact with, places they have recently visited and location data from their cellphone.
Roughly three-in-ten Americans say they would be not at all or not too comfortable sharing names of people with whom they have been in physical contact, and 22% express similar discomfort with disclosing the places they have recently visited.
And even as tech companies and governments have worked to research and release technology-focused solutions or supplements to contact tracing, a sizable portion of Americans continue to express discomfort with such efforts. This survey specifically asked about people’s comfort sharing location data from their cellphones and finds half of all U.S. adults say they would be not at all (29%) or not too (21%) comfortable doing this. (An April 2020 Center survey similarly found that 48% of Americans said it was unacceptable for the government to use people’s cellphones to track the location of people who had tested positive for the coronavirus, and 54% said the same for tracking people who had been in contact with a positive case.)
Majorities of Americans from both major parties express comfort sharing the names of people they might have been in physical contact with and the places they recently visited, though Republicans express less comfort than Democrats – with gaps of 15 percentage points or more in the shares of partisans who say they are not at all or not too comfortable sharing such information. Some 37% of Republicans say they would feel not at all (20%) or not too (18%) comfortable sharing names of contacts with a public health official, compared with 18% of Democrats who say the same (6% not at all and 12% not too comfortable). Some three-in-ten Republicans report they would be not at all (17%) or not too (14%) comfortable sharing places they have visited, compared with 14% of Democrats who say they would be not at all (5%) or not too (9%) comfortable doing the same. And 60% of Republicans would be not at all (40%) or not too (20%) comfortable sharing location data from their cellphone, compared with 42% of Democrats (19% and 22%, respectively).
Younger adults express less comfort sharing information than older adults. For example, about a third (32%) of those ages 18 to 29 say they would be not at all or not too comfortable sharing the names of those with whom they have been in contact with a public health official, compared with just 21% of those 65 and older who feel that way. Overall, those under 50 are more likely than those 50 and older to say they are not at all or not too comfortable sharing the names of people they’ve been in contact with (30% of those 18 to 49 say this vs. 24% of those 50+) and location data from their cellphone (55% vs. 46%). Some 43% of those ages 18 to 29 report being very comfortable sharing places they’ve recently visited, compared with larger shares of other age groups saying the same (48% of those ages 30 to 49, 50% of those 50 to 64 and 57% of those 65 and older).
Those with lower incomes and less formal education also tend to be less likely to be comfortable sharing information. About three-in-ten individuals with lower (28%) and middle (29%) incomes say they would be not at all or not too comfortable sharing names of contacts, compared with 19% of those with higher incomes. Some 29% of those with a high school diploma or less express a similar lack of comfort with sharing places they’ve visited, compared with 22% of those with some college and 14% with a bachelor’s degree or more education.
Majority of Americans say they would quarantine for two weeks if told they should do that by a public health official, but some report this would be difficult
Americans report high levels of willingness to quarantine if advised to do so by a public health official because they had the coronavirus. But they offer mixed reports about how difficult it would be to do this.
Half of the Americans surveyed were asked about whether they would act on the advice of a public health official to quarantine because they had the coronavirus, while the other half were asked about the difficulty of doing so. Over nine-in-ten adults (93%) say they definitely (73%) or probably (20%) would act on advice to quarantine for 14 days if told they had the coronavirus. Just 2% say they definitely would not follow that request, and 5% report they probably would not do so.
Although the vast majority of Americans who identify with either party say that they would definitely or probably quarantine, partisan differences again emerge. Some 88% of Republicans say they would definitely or probably act on advice to quarantine, while 97% of Democrats say the same. Republicans are 26 percentage points less likely than Democrats to say they definitely would quarantine, with 59% of this group saying so compared with 85% of Democrats.
Women are also more likely than men to report they would definitely quarantine (80% vs. 65%). Additionally, Black (76%) or Hispanic adults (80%) are more likely to report this than White adults (70%). And those with at least a bachelor’s degree are more likely to report they would definitely quarantine; 79% of college-educated adults say so, compared with 71% of those with some college and 69% of those with a high school diploma or less.
The small share of Americans – 7% – who say they would definitely or probably not quarantine were then asked about possible reasons why they may not act on this advice. Some 44% of this group say a major reason they would be unlikely to quarantine is that they just don’t think it would be necessary, while 35% say the same about not being able to miss work.
Smaller shares of those who say it’s unlikely they would isolate due to COVID-19 cite having too many other obligations (23%), their concern about being isolated from others (15%) or being unable to arrange child care (11%) as major reasons.
Yet of those asked how difficult it would be to act on advice to quarantine if told they had the coronavirus, Americans express more varied views. Roughly three-in-ten adults say that it would be very (10%) or somewhat (22%) difficult to quarantine. Men are more likely than women to say so (35% vs. 29%), and younger adults are more likely to say it would be very or somewhat difficult to quarantine than those 65 and older.
Education and income again play a role when it comes to the difficulty people might have in quarantining. Some 35% of those with a high school diploma or less report it would be very or somewhat difficult to quarantine, compared with 29% of those with a bachelor’s degree. And Americans are less likely to express difficulty as their income rises: 38% of adults from households with lower incomes, 30% of those from households in the middle income category and 26% of those with higher incomes say it would be very or somewhat difficult to quarantine.
The 32% of Americans who said that they would find it very or somewhat difficult to quarantine for 14 days were similarly asked why that would be so. Some 40% of this group said a major reason would be that they have too many other obligations; 39% say the same about being be unable to miss work. Less frequently cited as major reasons were concern about being isolated from others (23%), being unable to arrange child care (22%) and just not thinking it’s necessary (21%).
In all, 48% of adults say they would be comfortable or likely to engage with all three key steps of the contact tracing process
A successful contact tracing program is built around people’s compliance with key steps of the process. This survey asked people about their attitudes toward three of those steps that would be initiated by a public health official during the coronavirus outbreak. First, they were asked how likely they would be to speak with a public health official who contacted them to speak about the coronavirus outbreak by phone or text message. Second, they were asked about how comfortable they would be to share information about the names of people with whom they might have been in contact and about where they have been (either via names of places they’ve recently visited or via cellphone location data). Third, they were asked about whether they would act on advice from a public health official to quarantine for 14 days because they had the coronavirus. Our combination of these three steps – speak, share and quarantine – uses data from the random subset of respondents who were asked about all three of these things.5
Looking at the various combinations of responses to those questions, this survey finds that 48% of Americans would be relatively comfortable or likely to engage with all three of those steps, and 51% would be relatively resistant to at least one core aspect of the process. In making this calculation, we used a somewhat generous definition of engagement. (The percentages who actually would cooperate could be higher or lower.) The 48% referred to in most of this report as “comfortable or likely to engage” with all steps are those who say they would do all of the following in the coronavirus contact tracing process:
- Speak: Be very or somewhat likely to speak with a public health official if they were contacted by phone or text message to speak about the coronavirus outbreak.
- Share names of contacts: Be very or somewhat comfortable sharing with a public health official the names of people they might have been in physical contact with.
- Share information about where they have been: Be very or somewhat comfortable sharing with a public health official information about the places they have recently visited, OR very or somewhat comfortable sharing location data from their cellphone.6
- Quarantine: Definitely or probably would quarantine themselves for at least 14 days if they were told by a public health official they had the coronavirus.
If a more stringent definition of engagement were applied – taking only those who said they would be very likely to speak, those who are very comfortable sharing information (about whom they might have been in physical contact with and where they’d been), and those who would definitely quarantine – the share of Americans who say they would be comfortable or likely to engage is far lower. Only 16% of Americans would be considered comfortable or likely to engage with all three steps by this more stringent definition. However, for our primary analysis that follows, we use the more “lenient” version – counting one as likely or comfortable to engage with the entire process if they’ve identified in the top two most likely, comfortable or willing categories.
There are some notable differences among groups when it comes to their stated engagement with these three contact tracing activities. Republicans are less likely than Democrats to say they are comfortable or likely: 36% of Republicans would fall into this group, according to our relatively “generous” definition of stated engagement, compared with six-in-ten Democrats.
In addition, younger adults, Hispanic adults, those with lower incomes and those with less formal education are less likely to be fully comfortable or likely to engage with the process than their counterparts. About four-in-ten adults ages 18 to 29 (41%) say they would be comfortable or likely to engage with all three steps, compared with 56% of those ages 65 and older. And while White and Black adults are similarly likely to fall into this group given their responses – about half of each group expresses full comfort or likelihood engaging – Hispanic adults are less likely to say this (four-in-ten Hispanic adults say so).
Roughly six-in-ten individuals with a bachelor’s degree or higher formal education (60%) and those from households with higher incomes (61%) say they would be fully comfortable or likely to engage. Smaller shares of those with lower incomes and less education say the same.
Beyond these differences in stated comfort or likelihood by demographic and partisan breaks, there are also differences in people’s comfortable or likely engagement with contact tracing programs based on their awareness of contact tracing, their general behaviors such as willingness to answer their phones when an unknown number calls, and their general views about individuals trying to steal people’s personal information.
For example, results from this survey show that people’s comfort with or likelihood of engaging with the contact tracing and isolation process (speak, share, quarantine) rises with increased awareness of contact tracing. Some 63% of those who had seen or heard a great deal about contact tracing as of July say they would be comfortable or likely to engage with all three parts of the contact tracing effort that we explore here: that is, speaking with a public health official by phone or text message, sharing names of contacts and names of locations or location data, and quarantining. In contrast, 35% of those who have seen or heard not too much say they would do the same, and only 29% of those who’ve seen or heard nothing at all say this.
In addition, Americans’ comfort with and likelihood of engaging with the process are also related to their general tendency to pick up the phone; for example, 54% of those who would answer a call on their cellphone from an unknown number say they would be comfortable or likely to engage with every step, compared with 33% of those who would ignore both a call and a voicemail.
Looking specifically at likelihood of speaking with a contact tracer by phone or text (the first step in the process), 68% of those who would generally answer the phone say they would be very or somewhat likely to speak with a contact tracer by phone or text, compared with 58% of those who would not answer but check a voicemail and 42% of those who would not answer and ignore a voicemail.
Finally, those who think others often pretend to be someone else to steal others’ personal information are less likely to be comfortable or likely to engage with these key contract tracing and isolation steps. Roughly four-in-ten of those who say that people often pretend to be someone else to steal others’ personal information say they are also comfortable or likely to engage with the entire process, compared with about half of those who say people sometimes do this (52%) or rarely or never do this (51%).
About six-in-ten Americans are confident public health organizations will protect their records from hackers or unauthorized users, but Republicans are less confident
In light of general privacy concerns about tech-based solutions to contact tracing, as well as new findings about general trust in public institutions, the Center survey asked Americans about their confidence in the ability of specific potential collectors and users of contact tracing data to keep their records safe.
Overall, Americans generally feel their doctors and health care providers will keep their personal records safe from hackers or unauthorized users, with about eight-in-ten Americans expressing confidence that they will do so. A majority feels the same way about public health organizations, though four-in-ten say they are not at all or not too confident that these organizations will keep their records safe. And roughly half of U.S. adults (49%) are at least somewhat confident in the ability of the federal government to keep their records safe.
Smaller shares of Americans are not at all confident that these institutions will keep their records safe, but roughly a fifth of Americans (21%) say this about the federal government. Some 15% of Americans say the same about public health organizations, and only 6% are not at all confident that their medical doctor or health care providers will keep their records safe.
While these confidence levels are fairly consistent across most groups, Republicans are more likely than Democrats to express a lack of confidence that public health organizations, their medical doctors or health care providers, and the federal government will keep their personal records safe.
These differences are particularly large for public health organizations: Republicans are 20 percentage points more likely to say that they are at not at all or not too confident that public health organizations will do this, with 51% saying this compared with 31% of Democrats. Republicans are also more likely to be at most not too confident that their doctors or providers will keep their records safe, though the gap between parties is smaller; 22% express at most not too much confidence, compared with 15% of Democrats. An even smaller but significant gap exists between Republicans and Democrats who are not at all or not too confident in the federal government’s ability to keep their records safe (52% vs. 48%).
One important consideration, then, is whether those who express less confidence that public health organizations will keep their records safe from hackers or unauthorized users are less likely to say they’d be comfortable or likely to engage with key contact tracing protocols.
This survey finds that this is indeed the case. Examining the group of individuals asked about their comfort or likely engagement with all three steps, we find that fully seven-in-ten of those who are very confident that public health organizations will protect their personal records from hackers or unauthorized users also say that they would be comfortable or likely to engage with all three steps of the process. That compares with smaller shares of those who express less confidence (56% of those who are somewhat confident, 36% of those who are not too confident and 21% of those who are not at all confident).
Three-quarters of Americans think it is acceptable for public health officials to share personal information with infectious disease researchers, yet smaller shares say this for other groups
There has been some controversy about how data tied to coronavirus infections and deaths is being collected and shared with public health officials and researchers. Moreover, researchers themselves are sometimes resistant to data sharing.
This survey shows that three-quarters think it is acceptable for public health officials to make personal information that may be collected during the coronavirus outbreak available for use by infectious disease researchers, while smaller shares say the same about pharmaceutical companies (40%), health insurance companies (40%) and state and local elected officials (37%).
Once again, Republicans and Democrats vary in their views of the acceptability of sharing information with other entities, particularly when it comes to sharing data with state and local elected officials and with researchers who study infectious diseases. Republicans are 17 percentage points more likely than Democrats to think it is very or somewhat unacceptable for information to be available for use by state and local elected officials and by researchers who study infectious diseases. Partisans are more similar in their views of personal information being made available to health insurance companies or pharmaceutical companies.