Using a “wisdom of the crowds” approach to gathering data, Dynata is asking people visiting their survey platform if they or anyone in their household is experiencing symptoms associated with COVID-19. Launched on March 22, Dynata is collecting approximately 350,000 daily responses reporting symptoms globally and working with partners, including researchers at Harvard Kennedy School, to analyze and visualize the data.
Our goal is to leverage our collective power to provide valuable data points to assist governments and businesses in making the decision to reopen, while also helping these organizations – and health authorities and healthcare systems – stay one step ahead by identifying potential hot spots of infection before their communities and local healthcare systems are overwhelmed. As recently featured in Forbes, we hope media and other public information sources can help raise awareness of the availability of this data so that those organizations and individuals on the front lines of this crisis can use it in their efforts.
We are doing our part by collecting and updating the largest amount of COVID-19 symptom data, and making the data open access and accessible for free. Let’s take on these new challenges we’re all facing, together. We’d love to partner with you on this project. If you are a government or public health official, healthcare or hospital executive, media or potential data partner and wish to work with us on this initiative, please contact us at info@dynata.com.
Learn how we used a "wisdom of the crowds" approach to gather this data from around the world.
Starting March 22, 2020, Dynata has asked internal survey platform visitors if they or anyone in their household was experiencing symptoms associated with COVID-19. Visitors can answer the question a maximum of once every 48 hours, reporting information on behalf of their entire household.
Best practice requires that members of “wise crowds” be diverse, independent from one another and decentralized (i.e. able to draw on personal/local knowledge). Dynata recruits from a variety of sources, without focusing on any one type of individual. These independent members do not interact with each other, as they would in a community or social media environment, and our COVID-19 question requires that people only draw from their own experience.
Sample source: A "wisdom of the crowds" methodology relies on volume and is not a representative sample. Quotas by geography or demographics, margin of error and confidence interval are not applied to the data.
Scale: 300,000 – 400,000 data points are reported within a typical 24-hour period. Fewer completes are gathered on weekend days.
Field dates: Data collection began on March 22.
Unit of data: The data represents households, not individuals.
Demographics: Where gender or age is shown, it applies to the person reporting the symptoms on behalf of their entire household. That person may or may not be the person suffering the symptoms. For example, an 18-year-old reporting symptoms may be reporting that a parent in the household has those symptoms. Income data applies to the total household, not the person answering.
Language: The question is not fielded in all languages spoken in each country. For example, the survey is in English for India and Mandarin for China.
Geography: The countries displayed represent the places where Dynata’s thousands of research clients are conducting research. Countries where researchers conduct the most research have correspondingly higher volumes of data. Note on Italy: We do not display data for Italy due to local legal restrictions.
Symptom combinations: When multiple symptoms are reported for a household, one person could be experiencing all of the symptoms, or several people could each be experiencing one symptom.
PLEASE NOTE: The information provided here is self-reported by Dynata respondents and has not been verified for accuracy or completeness.
Are you or anyone in your household exhibiting any of the following symptoms (select all that apply):
Each of the maps are interactive and can provide additional insights as you explore the data. Below are some tips as you start to interact with the maps for additional insights.
Filters:
Date The data is aggregated for a two-week period prior to the date filter that you see labeled “Date: Rolling 14 days.” We suggest starting with exploring the filters. Click on it and a calendar should pop up as seen in the image below.
Try selecting different days to see how the symptoms progress over time in each geographic region. This can be a potential indicator of COVID-19 prevalence.
Symptoms Next try looking at individual symptoms or combinations of symptoms that were reported. Click on the Symptom drop down box and explore different combinations.
Age This filter is only included at the Global View. Please note that this is NOT ALWAYS the age of the person displaying symptom, but it is the age of the person who was responding for the household
Minimum Reporting Size This will allow you to determine how much data is required before it is reported on. Lowering the minimum requirement will make more geographies appear on the map, but each of these new areas will have more errors associated as the sample size in these areas are smaller. We recommend staying between 50 and 100 on this tool to avoid reporting on areas that don’t have enough data to even look at trends. Note that as this is not a probability sample (see probability section), so we do not report margin of errors associated with the data.
Map Features:
Local View In the US View you can look at the data at a county or a postal code level. Double clicking on a state should bring up a menu that allows you to view that specific state at either a County or Postal Code level of aggregation.
To return to the overall US view just select any location and a menu will pop up to return to the US View.
Map Tools Any of the maps will allow you to zoom in or out. Just click on the plus (+) or minus (-) sign found in the top right corner.
You can also double click anywhere in the map to zoom in or shift double click to zoom out.
If you want to search for a specific geography you can also use the magnifying glass in the top right and type in what you are looking for.
The home button will return to the default map settings.
Lastly, you can zoom in on a specific location by hitting the arrow, then selecting the box with the magnifying glass. Draw the box around what you want to see and it will zoom to your selection.
