Special Report: COVID-19

Special Report:
Understanding the Impact of COVID-19 on Public Attitudes & Behaviors

Over the last few weeks, the spread of COVID-19 has affected society in innumerable ways, and news pertaining to the virus has permeated the media. As public and private institutions alike respond to the pandemic, how is the public responding? To answer this question, we conducted a study of consumer behaviors and attitudes over time as they relate to the virus.

To gauge shifts in attitudes over time, the study consisted of two surveys which were distributed one week apart. Behavioral shifts were measured over a longer period. In all, the study sought to understand:

  • Consumer confidence in their personal safety & preparedness
  • Consumer confidence in public institutions
  • Effects on reported shopping & travel behaviors
  • Effects on measurable online browsing behaviors

Below, we share highlights from the study. Additional charts are available for download via the form on the right. Methodology is available at the bottom of this page.







Attitudinal Response to COVID-19

Timing

Attitudinal responses to the COVID-19 outbreak were collected in three phases. Phase 1 responses were collected between February 29th, and March 2nd, 2020. For context, this was shortly after a series of notable public events, including the first White House press conference on the situation, the first significant stock market drop, and after the United States Senate had requested $8.5B USD in aid.

Phase 2 responses were collected between March 7th and March 9th, 2020. For context, this was shortly after global cases surpassed 100,000 and United States cases surpassed 400.

Phase 3 responses were collected between March 14th and March 16th, 2020. For context, this was shortly after the United States government declared a national emergency and imposed travel bans to Europe.

General Confidence & Safety

Week over week, the public’s general feeling of safety from the COVID-19 outbreak was relatively static, though there were nominal shifts toward a slight increase in confidence. Specifically, in the first edition of the survey, 52% of people agreed that they felt safe from the virus, which increased slightly to 55% just one week later.

Likewise, disagreement with the statement “I feel safe from the coronavirus” fell from 29% to 26% between the February 29th and March 7th editions of the survey.

Between the second and third week, however, the public’s safety concerns seemed to worsen, with only 50% agreeing that they felt safe from the virus. Disagreement with the above statement also increased to 31% between the March 7th and March 14th editions of the survey.

Sense of Safety from COVID-19 by Region
N = 23,581. DISQO Audience survey, March 14, 2020 through March 16, 2020
Q: I feel safe from the coronavirus. West Midwest South Northeast
Strongly agree 6.9% 6.5% 8.0% 5.5%
Agree 17.8% 18.3% 18.3% 14.1%
Somewhat agree 24.6% 27.3% 25.6% 24.2%
Neutral 18.6% 19.1% 18.7% 19.6%
Somewhat disagree 16.8% 15.8% 14.7% 18.0%
Disagree 8.9% 7.6% 8.3% 10.8%
Strongly Disagree 6.4% 5.4% 6.5% 7.8%

Regionally, those in the Midwest were most likely to express that they feel safe from COVID-19 (52%), though this sentiment fell notably in the region from the week prior (when it was at 60%). Meanwhile, those in the Northeast were the least likely to express feelings of safety (44%), a figure that also similarly fell from the week prior (when it was 51%).

While the sense of safety fell week over week across all regions, the rate of decline in the South was less than half that of the other regions. Indeed, positive responses to this question fell 3.1% in the South, notably less than the 6.1% to 7.5% decline seen elsewhere.

As might be evidenced by empty store shelves, people reported a significant increase in feelings of preparedness in March relative to late February.

Specifically, in the first edition of the survey, 42% of people reported feeling prepared in the event COVID-19 spread to their city. By March 7th, this figure rose to 53%, and by March 14th, it rose to 61%.

Despite increases in confidence of preparedness at the individual level, the same was not seen in respondents’ confidence in public institutions.

In total, there has been a nominal 3% shift in agreement with the statement “I feel confident that the government is prepared to handle the coronavirus.”

Confidence in Government Response to COVID-19 by Age Group
N = 23,581. DISQO Audience survey, March 14, 2020 through March 16, 2020
Q: I feel confident that the government is prepared to handle the coronavirus. 18 – 24 25 – 34 35 – 44 45 – 54 55 – 64 65 – 74 75 +
Strongly agree 5.8% 5.5% 5.9% 6.7% 8.0% 7.4% 6.3%
Agree 11.8% 10.6% 12.0% 13.9% 14.5% 16.2% 21.8%
Somewhat agree 14.9% 16.3% 17.3% 19.1% 18.8% 18.5% 18.7%
Neutral 19.4% 18.5% 17.3% 16.0% 13.6% 11.4% 9.4%
Somewhat disagree 19.9% 19.2% 18.3% 17.1% 16.8% 17.5% 18.3%
Disagree 14.1% 13.4% 13.1% 12.6% 12.7% 12.8% 13.3%
Strongly Disagree 14.1% 16.5% 16.2% 14.6% 15.5% 16.2% 12.1%

When assessing confidence in government by age group, those over 55 years of age were the most likely to agree with the statement “I feel confident that the government is prepared to handle the coronavirus”, and those 75 years of age and older were the most likely to have some degree of confidence (47%).

Meanwhile, those 25 to 34 years of age were the least likely to have confidence in the government’s response to the crisis, with only 32.4% confirming a positive agreement to the statement. Rather, 49.1% of this group disagreed.

Confidence in Government Response to COVID-19 by Region
N = 23,581. DISQO Audience survey, March 14, 2020 through March 16, 2020
Q: I feel confident that the government is prepared to handle the coronavirus. West Midwest South Northeast
Strongly agree 5.3% 6.4% 7.7% 5.7%
Agree 11.4% 14.3% 15.1% 11.0%
Somewhat agree 15.9% 17.9% 19.0% 16.9%
Neutral 16.2% 15.2% 15.7% 15.7%
Somewhat disagree 17.9% 18.8% 16.8% 19.4%
Disagree 15.5% 12.9% 11.3% 14.2%
Strongly Disagree 17.7% 14.5% 14.3% 17.0%

Regionally, people in the South (41.9%) and the Midwest (38.6%) were most likely to express confidence in the government’s preparedness to handle the COVID-19 outbreak.

Conversely, those in the West (51.2%) and the Northeast (50.6%) were most likely to express a lack of confidence in the government. Given that these regions had the largest outbreaks (in the states of Washington, California, and New York), these outcomes are to be expected.

Impact on Travel

As COVID-19 has spread, so travel plans have been affected. When first polled, 12% of individuals reported that the pandemic (then classified as an epidemic) had affected upcoming travel plans. Within one week, that figure had risen to 17%, marking a nearly 50% relative increase over the week prior.

Now, with significant additional government imposed travel restrictions now in effect, that figure has risen to 33%, nearly doubling from one week ago.

Impact on Shopping

In response to the spread of COVID-19, shopping habits have also reportedly changed.

At the time of the first survey, 38% of people reported that they had made some sort of purchase in response to the situation.

Shopping for cleaning supplies was the most common, with 33% of total respondents and 87% of those who made any type COVID-19 related purchase reporting they stocked up in this category.

Shopping for cleaning supplies rose notably between week 1 and 2, though plateaued. Nevertheless, this category remains the most popular among those shopping in response to COVID-19, with 84% of the people who have reported making a COVID-19 related purchase (now up to 47% of individuals) buying within this category.

Notably, albeit expectedly, purchases for protective wear (masks & gloves) fell slightly between the first two weeks. Although motivations for individual self-reported shopping behaviors were not asked in this study, it might arguably be attributed to both a decrease in the availability of these products and the mixed public messages given regarding their efficacy.

While the number of people reporting that they’ve made a purchase in response to COVID-19 has not changed notably between the second (46%) and third week (47%) of the study, shopping habits have changed.

One week ago, two thirds (67%) of people attested that their shopping habits had not fundamentally changed.

However, just one week later, that figure fell to 45%. Now, a majority of people (55%) report that their shopping habits have changed due to COVID-19.

In early March, 41% of people reported that they had bought cleaning supplies, but only 13% claimed to have “stocked up” in this category. By mid March, the number of people stocking up on cleaning supplies rose to 18%, suggesting a more significant shift in consumer shopping behavior.

Significant shifts in behavior were noted in other areas of consumer shopping behaviors, as well. Most notably, the number of people stocking up on non-perishable foods rose sharply week-over-week, from 13% on March 9th to 29% on March 16th. This is in line with measured search behaviors for key grocery items (covered later in this report), as well as widely observed sparse store shelves throughout the country.

Similarly, the number of people avoiding crowds and in-store shopping altogether both rose significantly over the last week. Specifically, crowd avoidance rose from 15% to 27%, and store avoidance rose from 3% to 7%.

Behavioral Response to COVID-19

After an initial ramp in browsing activity in January, activity fell slightly and plateaued for much of February. Notably, visitation of COVID-19 related websites spiked 23.5% as global cases reached 10,000, compared to activity 5 days prior or thereafter, as noted by the dashed teal line in the above chart.

The subsequent spike (denoted by the dashed pink line in the above chart) aligns with a rapid sequence of domestic news pertaining to the virus (to be discussed further below).

The first significant spike in COVID-19 related browsing behavior was on February 27th, 2019, when page view activity increased 240% over the preceding 7 day average. Given the dominance of the topic in the news cycle, this jump was to be expected. In the preceding 24 hours before this spike in activity, the White House held a press conference on the epidemic, and the United States Senate formally requested $8.5B USB in aid to respond to the crisis. At the same time, the DOW had dropped over 1200 points.

The next noticeable spike in browsing behaviors took place on March 2nd, which is when the total number of cases within the United States reached 100. This was comparable to the previous spike, albeit 3% higher. Compared to activity the preceding week, activity was 60% higher on this date.

Of course, March 2nd also marked the first notable jump in fatalities within the United States, and activity throughout the remainder of the work week continued to increase. Specifically, COVID-19 related page views from March 2nd through March 6th was 59% higher than the preceding work week (February 24th through 28th).

In the pursuit of understanding how the spread of COVID-19 affected other search behaviors, we also measured search volumes for related topics which did not explicitly include the name of the virus — in any variation — in the search. This included searches for things such as masks (N-95, N-99, surgical, etc), gloves (latex, surgical, vinyl, etc), cleaning supplies (bleach, disinfectants, sanitizers, etc), and official sources of information (CDC, WHO, etc).

Lift in this area was significant, though slight compared to other shifts in behaviors. The largest week over week increase in related search activity took place February 24th through March 1st, wherein related searches raised 21.2% over the week prior. The first week of March (the 2nd through the 9th) similarly had a notable increase in related search activity (17.2% higher than the week prior).

As expressed in the attitudinal component of the study, some individuals are stocking up on groceries. Indeed, according to analysis of common grocery search terms, there has been a significant lift in the number of people searching for groceries thus far in March. Specifically, the chart above details the lift for the first 12 days of March, 2020 compared to the last 12 days of February.

Several of the items are non-perishable staples (rice, pasta, water, etc), whereas others may be indicators of people planning to cook more at home (french fries, ground beef, frozen pizza, etc).

Methodology

The Audience

All participants in this study are opted-in members of the DISQO Audience. Digital activity for the behavioral component of this study was captured via a 100% opt-in first-party data set. Attitudinal responses for the survey component of the study was likewise collected directly from DISQO Audience members.

Attitudinal Study

The survey was conducted online, distributed to a nationally representative sample of DISQO Audience members within the United States via the Survey Junkie platform, which is wholly owned by DISQO. Surveys were taken on both desktop and mobile devices.

To assess how sentiment shifted over time, three waves of the study were conducted one week apart. The first wave of the study generated 24,805 responses, which were collected over a 3 day period, from February 29, 2020 through March 2, 2020. The second wave of the study generated 23,838 responses, which were likewise collected over a 3 day period, from March 7, 2020 to March 9, 2020. The third wave of the study generated 23,581 responses, which were collected over a 3 day period, from March 14, 2020 through March 16, 2020.

Behavioral Study

To assess how the spread of covid-19 affected digital behaviors, DISQO measured online activity over a 100 day period, from December 1, 2019 through March 09, 2020. Of the 220,000 DISQO Audience members selected for the study, behaviors of 217,917 were included in the final analysis.

For clarity, direct search was defined as search activity that explicitly mentioned the coronavirus, covid-19, or corona within the context of the virus. Related searches was defined as search activity for related protective gear (N95 masks, gloves, etc), cleaning supplies (bleach, sanitizer, etc), and general info (CDC, WHO, etc). Specific browsing was defined as page views of coronavirus related pages on official channels (CDC, WHO, OSHA, etc), news sites, third party trackers (Worldometers, Arcgis, etc), and others.