Which Americans are on the wrong side of the digital divide?

The digital divide — the gulf between citizens with ready access to modern information and communications technologies, and those without — impacts millions of Americans spanning generations, demographics, and geographies.

How many millions? No one seems to know for certain. 

The Federal Communications Commission’s 2019 Broadband Deployment Report states that 21.3 million Americans lack access to broadband internet, including wired and fixed wireless connections; a year later, the FCC slashed that number to 14.5 million. But according to independent research organization BroadbandNow, at least 42 million Americans lack broadband access, while Microsoft estimates that 157.3 million people nationwide — close to half of the overall U.S. population — do not use the internet at broadband speeds.

President Biden’s $1.2 trillion Infrastructure Investment and Jobs Act promises to narrow America’s digital gap, committing $65 billion to expanding broadband availability and 5G wireless connectivity across the U.S. But ensuring investments are made in the areas that most urgently need assistance requires accurate and meaningful data — information that is largely unavailable. Moreover, a truly comprehensive and effective solution to bridging the digital divide must address more than infrastructure and access, because infrastructure and access alone do not ensure adoption. Affordability remains a significant roadblock; adaptability (i.e., an individual’s ability to learn and navigate technologies that are unfamiliar to them) also plays a decisive role in exacerbating digital inequity. 

Civis Analytics set out to remedy this dearth of reliable data on the nation’s digital divide, using our always-on national survey infrastructure to collect insights government agencies can use to more effectively and cost-efficiently address the digital needs of the communities they serve. Over several months, we polled more than 10,000 Americans ages 18 and older to:

While we used quotas and weighting to ensure that survey respondents were representative of the national adult population, there were inherent limitations to address: namely, that people accessing the survey via tablet, mobile device, or computer already demonstrate a baseline level of internet access, meaning the people who are fully offline can’t be captured in our data — and that the data may therefore underestimate certain digital divide needs and overestimate others. With that being said, even looking only at people with a basic level of access can still yield important conclusions, especially regarding digital affordability and digital adoption. 

One other caveat: this is a national survey, and the digital divide is not a consistent problem across all parts of the U.S. Some of the findings here may tell a different story from trends identified in specific regions or municipalities. 

Read on to explore our findings, or use the table of contents below to jump ahead. 

Since February 29, 2020, have you or anyone in your household experienced a substantial loss of income due to job loss or furlough?

We found 26 percent of households have suffered financial hardship resulting from job loss or furlough since the start of the COVID-19 pandemic. Sixty-eight percent of households have not experienced a substantial loss of income related to employment issues, with the remaining 6 percent unsure.  

Pandemic-related adversity impacted Americans across all economic strata. While 26 percent of survey respondents in households reporting annual income between $50,001 and $75,000 say they or someone else in their household lost significant income due to job loss or furlough, 25 percent of households earning more than $150,000 a year have grappled with these same challenges. 

Shifting the focus to ethnicity, however, we found that COVID-related job loss or furlough affected 32 percent of non-white respondents, compared to 23 percent of white respondents. In addition, 34 percent of respondents younger than 35 were impacted, versus 23 percent of respondents 35 years of age or older. 

Would you consider yourself to be an expert in your knowledge and ability to do things online? 

Just 45 percent of survey respondents express strong confidence in their ability to access and navigate the web, Civis found. Forty-three percent say they do not consider themselves experts in digital knowledge and ability, and the remaining 12 percent are unsure. 

More men (51 percent) than women (40 percent) consider themselves online experts. In addition, more non-white Americans (55 percent) than white Americans (40 percent) consider themselves experts at doing things on the web. 

Age plays a decisive factor: 39 percent of respondents aged 35 and older say they consider themselves online experts, as opposed to 59 percent of those younger than 35. Educational attainment also plays a critical role, with 37 percent of respondents with a high school diploma or less declaring themselves online experts, compared to 48 percent of respondents with a college education and 59 percent with an advanced degree.

How important is it for you to be able to get online every day?

Internet access is fundamental for the overwhelming majority of Americans: 48 percent of respondents say going online each day is “absolutely critical,” and another 42 percent say it is “somewhat important.” The remaining 10 percent say going online is “not important at all.” 

Daily web access is most important to younger Americans. Fifty-three percent of respondents between the ages of 18 to 34, as well as 53 percent of respondents ages of 35 and 49, say going online each day is absolutely critical, compared to 46 percent of respondents in the 50-to-64 group and 36 percent of those 65 and older. 

Shifting the focus to educational attainment, we found that getting online each day is absolutely critical to 59 percent of Americans with advanced degrees and 52 percent of respondents with college degrees. Only 39 percent of respondents with high school degrees or less say it is absolutely critical to go online each day.  

Do you know how to do basic troubleshooting (like updating software, scanning for viruses and malware, or searching online for help) to fix your laptop or tablet if you need to?

Seventy percent of respondents are familiar with basic troubleshooting techniques for laptops and tablet devices: 75 percent of men are familiar, compared to 65 percent of women, Civis found. Eighteen percent of respondents are unfamiliar with tech troubleshooting basics, however, and 6 percent don’t know. (The remaining 5 percent don’t own a laptop or a tablet.) 

Yet again, both age and education exert influence on the survey results. Seventy-nine percent of respondents between the ages of 35 and 49 know device troubleshooting basics, followed by the 50–64 group at 74 percent; 18-to-34-year-old respondents are next at 65 percent, with respondents 65 and older at 63 percent. 

As for educational attainment, 82 percent of respondents with an advanced degree know basic troubleshooting techniques, 6 percentage points higher than college-educated respondents. However, the number plunges to 59 percent among respondents with a high school diploma or less.   

Income also plays a key role. Eighty percent of respondents with an annual income exceeding $100,000 are familiar with basic laptop and tablet troubleshooting, ahead of 76 percent of respondents in the $50,001-$100,000 annual income bracket. Among respondents earning $50,000 or less a year, however, troubleshooting knowledge plunges to 59 percent. 

How much do you agree with the following statement: “Going online isn’t worth risking my personal safety”?

Twenty-nine percent of respondents neither agree nor disagree that accessing the web isn’t worth putting their personal safety at risk, and 29 percent strongly disagree with the statement presented to them. Twenty-two percent somewhat disagree, however, and 14 percent somewhat agree, compared to 10 percent of respondents who strongly agree. 

While only 8 percent of white respondents strongly agree that going online isn’t worth the risk to their personal safety, the number jumps to 14 percent among non-white respondents. Seventeen percent of Black respondents agreed with the statement at 17 percent, compared to 9 percent of Asians. Meanwhile, 29 percent of white respondents strongly disagree with the statement “Going online isn’t worth risking my personal safety,” versus 23 percent of non-whites. In this instance, 29 percent of Asians strongly disagreed, compared to 20 percent of Blacks. 

In terms of age, 17 percent of respondents younger than 35 strongly agree “Going online isn’t worth risking my personal safety,” in contrast with 7 percent of respondents 35 or older. At the same time, 31 percent of Americans 35 or older strongly disagree with the statement, as opposed to 19 percent of those under 35.  

Drilling down on educational attainment yields some interesting similarities and discrepancies. Eleven percent of respondents with advanced degrees strongly agree with the statement “Going online isn’t worth risking my personal safety,” closely followed by college-educated respondents and respondents with a high school diploma or less at 10 percent each. However, 33 percent of respondents with advanced degrees strongly disagree, compared to college-educated respondents at 30 percent and respondents with a high school diploma or less at 22 percent.

If a website uses cookies, it means that the site:

Sixty-four percent of Americans surveyed by Civis correctly completed the statement above by selecting “Can track your visits and activity on the site” from among the five available responses. Fourteen percent aren’t sure, 10 percent selected “Can see the content of all the files on the device you are using,” 8 percent responded “Will automatically prompt you to update your web browser software if it is out of date” and 5 percent responded “Is likely to infect your device with a computer virus.” 

Sixty-eight percent of white respondents completed the statement by selecting “Can track your visits and activity on the site,” versus 55 percent of non-whites. There is significant variation here among non-white ethnic groups: while 72 percent of Asians selected “Can track your visits and activity on the site,” only 55 percent of Hispanics and 48 percent of Blacks chose the same response.

Age and educational attainment once again make an impact. Sixty-eight percent of respondents 35 and older selected “Can track your visits and activity on the site,” compared to 52 percent of those under 35. In addition, 75 percent of respondents with advanced degrees and 68 percent of college-educated respondents selected “Can track your visits and activity on the site,” compared to 55 percent of respondents with a high school diploma or less. 

What does it mean when a website has “https://” at the beginning of its URL, as opposed to “http://” without the “s”?

Thirty-eight percent of respondents told Civis they are “not sure” what it means when a website URL begins with “https://” Thirty percent selected the correct answer, “Information entered into the site is encrypted,” while 23 percent selected “The site has been verified as trustworthy.” Five percent selected “The site is only accessible to people in certain countries,” and the remaining 4 percent selected “The content on the site is safe for children.” 

Twenty-eight percent of white respondents completed the statement by selecting “Information entered into the site is encrypted,” versus 32 percent of non-whites. There is again notable variation among non-white ethnic groups: while 47 percent of Asians selected “Information entered into the site is encrypted,” only 30 percent of Blacks and 28 percent of Hispanics chose the same response.

The differences among age groups are less acute in this instance, however. Thirty-one percent of respondents younger than 35 selected “Information entered into the site is encrypted,” edging out respondents 35 or older at 29 percent. Forty percent of respondents 35 and up say they “don’t know” what it means when a website URL begins with “https://” versus 33 percent of those younger than 35. 

Forty-one percent of respondents with advanced degrees and 34 percent of college-educated respondents selected “Information entered into the site is encrypted,” compared to 21 percent of respondents with a high school diploma or less. Moreover, 25 percent of Americans with a high school diploma or less incorrectly selected “The site has been verified as trustworthy,” and 46 percent don’t know.