You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
a random sample of its users to these surveys, which were
26
+
voluntary. Users age 18 or older were eligible to complete the surveys, and
27
+
their survey responses are held by CMU. No individual survey responses are
28
+
shared back to Youtube.
29
+
30
+
This survey was a pared-down version of the
31
+
[COVID-19 Trends and Impact Survey (CTIS)](../../symptom-survey/),
32
+
collecting data only about COVID-19 symptoms. CTIS is much longer-running
33
+
and more detailed, also collecting belief and behavior data. CTIS also reports
34
+
demographic-corrected versions of some metrics. See our
35
+
[surveys page](https://delphi.cmu.edu/covid19/ctis/) for more detail
36
+
about how CTIS works.
37
+
38
+
The two surveys report some of the same metrics. While nominally the same,
39
+
note that values from the same dates differ between the two surveys for
40
+
[unknown reasons](#limitations).
41
+
42
+
As of late April 2020, the number of Youtube survey responses we received each
43
+
day was 4-7 thousand. This was not enough coverage to report at finer
44
+
geographic levels, so this indicator only reports at the state level. The
45
+
survey ran from April 21, 2020 to June 17, 2020, collecting about 159
46
+
thousand responses in the United States in that time.
47
+
48
+
We produce [influenza-like and COVID-like illness indicators](#ili-and-cli-indicators)
49
+
based on the survey data.
50
+
51
+
## Table of Contents
52
+
{: .no_toc .text-delta}
53
+
54
+
1. TOC
55
+
{:toc}
56
+
57
+
## Survey Text and Questions
58
+
59
+
The survey contains the following 5 questions:
60
+
61
+
1. In the past 24 hours, have you or anyone in your household experienced any of the following:
62
+
- (a) Fever (100 °F or higher)
63
+
- (b) Sore throat
64
+
- (c) Cough
65
+
- (d) Shortness of breath
66
+
- (e) Difficulty breathing
67
+
2. How many people in your household (including yourself) are sick (fever, along with at least one other symptom from the above list)?
68
+
3. How many people are there in your household (including yourself)?
69
+
4. What is your current ZIP code?
70
+
5. How many additional people in your local community that you know personally are sick (fever, along with at least one other symptom from the above list)?
71
+
72
+
73
+
## ILI and CLI Indicators
74
+
75
+
We define COVID-like illness (fever, along with cough, or shortness of breath,
76
+
or difficulty breathing) or influenza-like illness (fever, along with cough or
77
+
sore throat) for use in forecasting and modeling. Using this survey data, we
78
+
estimate the percentage of people (age 18 or older) who have a COVID-like
79
+
illness, or influenza-like illness, in a given location, on a given day.
80
+
81
+
| Signals | Description |
82
+
| --- | --- |
83
+
|`raw_cli` and `smoothed_cli`| Estimated percentage of people with COVID-like illness <br/> **Earliest date available:** 2020-04-21 |
84
+
|`raw_ili` and `smoothed_ili`| Estimated percentage of people with influenza-like illness <br/> **Earliest date available:** 2020-04-21 |
85
+
86
+
Influenza-like illness or ILI is a standard indicator, and is defined by the CDC
87
+
as: fever along with sore throat or cough. From the list of symptoms from Q1 on
88
+
our survey, this means a and (b or c).
89
+
90
+
COVID-like illness or CLI is not a standard indicator. Through our discussions
91
+
with the CDC, we chose to define it as: fever along with cough or shortness of
92
+
breath or difficulty breathing. From the list of symptoms from Q1 on
93
+
our survey, this means a and (c or d or e).
94
+
95
+
Symptoms alone are not sufficient to diagnose influenza or coronavirus
96
+
infections, and so these ILI and CLI indicators are *not* expected to be
97
+
unbiased estimates of the true rate of influenza or coronavirus infections.
98
+
These symptoms can be caused by many other conditions, and many true infections
99
+
can be asymptomatic. Instead, we expect these indicators to be useful for
100
+
comparison across the United States and across time, to determine where symptoms
101
+
appear to be increasing.
102
+
103
+
104
+
## Estimation
105
+
106
+
### Estimating Percent ILI and CLI
107
+
108
+
Estimates are calculated using the
109
+
[same method as CTIS](./fb-survey#estimating-percent-ili-and-cli).
110
+
However, the Youtube survey does not do weighting.
111
+
112
+
### Smoothing
113
+
114
+
The smoothed versions of all `youtube-survey` signals (with `smoothed` prefix) are
115
+
calculated using seven day pooling. For example, the estimate reported for June
116
+
7 in a specific geographical area is formed by
117
+
collecting all surveys completed between June 1 and 7 (inclusive) and using that
118
+
data in the estimation procedures described above. Because the smoothed signals combine
119
+
information across multiple days, they have larger sample sizes and hence are
120
+
available for more locations than the raw signals.
121
+
122
+
## Lag and Backfill
123
+
124
+
This indicator has a lag of 2 days. Reported values can be revised for one
125
+
day (corresponding to a lag of 3 days), due to how we receive survey
126
+
responses. However, these tend to be associated with minimal changes in
127
+
value.
128
+
129
+
130
+
## Limitations
131
+
132
+
When interpreting the signals above, it is important to keep in mind several
133
+
limitations of this survey data.
134
+
135
+
***Survey population.** People are eligible to participate in the survey if
136
+
they are age 18 or older, they are currently located in the USA, and they are
137
+
an active user of Youtube. The survey data does not report on children under
138
+
age 18, and the Youtube adult user population may differ from the United
139
+
States population generally in important ways. We don't adjust for any
140
+
demographic biases.
141
+
***Non-response bias.** The survey is voluntary, and people who accept the
142
+
invitation when it is presented to them on Youtube may be different from
143
+
those who do not.
144
+
***Social desirability.** Previous survey research has shown that people's
145
+
responses to surveys are often biased by what responses they believe are
146
+
socially desirable or acceptable. For example, if it there is widespread
147
+
pressure to wear masks, respondents who do *not* wear masks may feel pressured
148
+
to answer that they *do*. This survey is anonymous and online, meaning we
149
+
expect the social desirability effect to be smaller, but it may still be
150
+
present.
151
+
152
+
Whenever possible, you should compare this data to other independent sources. We
153
+
believe that while these biases may affect point estimates -- that is, they may
154
+
bias estimates on a specific day up or down -- the biases should not change
155
+
strongly over time. This means that *changes* in signals, such as increases or
156
+
decreases, are likely to represent true changes in the underlying population,
157
+
even if point estimates are biased.
158
+
159
+
### Privacy Restrictions
160
+
161
+
To protect respondent privacy, we discard any estimate that is based on fewer than 100 survey responses. For
162
+
signals reported using a 7-day average (those beginning with `smoothed_`), this
163
+
means a geographic area must have at least 100 responses in 7 days to be
164
+
reported.
165
+
166
+
This affects some items more than others. It affects some geographic areas
167
+
more than others, particularly areas with smaller populations. This affect is
168
+
less pronounced with smoothed signals, since responses are pooled across a
169
+
longer time period.
170
+
171
+
172
+
## Source and Licensing
173
+
174
+
This indicator aggregates responses from a Delphi-run survey that is hosted on the Youtube platform.
175
+
The data is licensed as [CC BY-NC](../covidcast_licensing.md#creative-commons-attribution-noncommercial).
Copy file name to clipboardExpand all lines: src/server/endpoints/covidcast_utils/db_signals.csv
+3-3
Original file line number
Diff line number
Diff line change
@@ -1575,7 +1575,7 @@ NSSP does not report county-level data for all counties with reporting EDs; some
1575
1575
The following states report no data through NSSP at the county level: CA, WA, AK, AZ, AL, CO, SD, ND, MO, AR, FL, OH, NH, CT, NJ.
1576
1576
1577
1577
South Dakota, Missouri, and territories report no data through NSSP at the state level.",Percentage,percent,other,bad,FALSE,FALSE,FALSE,FALSE,FALSE,,,,,,
1578
-
nssp,pct_ed_visits_rsv,FALSE,pct_ed_visits_rsv,FALSE,COVID Emergency Department Visits (Percent of total ED visits),TRUE,Percent of ED visits that had a discharge diagnosis code of rsv,Percent of ED visits that had a discharge diagnosis code of rsv,National Syndromic Surveillance Program,rsv,Hospitalizations,USA,"county,state,hrr,msa","hrr,msa",2022-10-01,,ongoing,,week,Week,weekly,,,All,None,hospitalized,,"Data is available for 78% of US emergency departments. California, Colorado, Missouri, Oklahoma, and Virginia have the most noticeable gaps in coverage, with many counties in those states having either no eligible EDs or having no recently reported data in NSSP. However, most states have some counties that do not contain any reporting EDs.
1578
+
nssp,pct_ed_visits_rsv,FALSE,pct_ed_visits_rsv,FALSE,RSV Emergency Department Visits (Percent of total ED visits),TRUE,Percent of ED visits that had a discharge diagnosis code of rsv,Percent of ED visits that had a discharge diagnosis code of rsv,National Syndromic Surveillance Program,rsv,Hospitalizations,USA,"county,state,hrr,msa","hrr,msa",2022-10-01,,ongoing,,week,Week,weekly,,,All,None,hospitalized,,"Data is available for 78% of US emergency departments. California, Colorado, Missouri, Oklahoma, and Virginia have the most noticeable gaps in coverage, with many counties in those states having either no eligible EDs or having no recently reported data in NSSP. However, most states have some counties that do not contain any reporting EDs.
1579
1579
1580
1580
NSSP does not report county-level data for all counties with reporting EDs; some reporting EDs are only included in state-level values.
1581
1581
@@ -1596,14 +1596,14 @@ NSSP does not report county-level data for all counties with reporting EDs; some
1596
1596
The following states report no data through NSSP at the county level: CA, WA, AK, AZ, AL, CO, SD, ND, MO, AR, FL, OH, NH, CT, NJ.
1597
1597
1598
1598
South Dakota, Missouri, and territories report no data through NSSP at the state level.",Percentage,percent,other,bad,TRUE,FALSE,FALSE,FALSE,FALSE,,,,,,
1599
-
nssp,pct_ed_visits_influenza,TRUE,smoothed_pct_ed_visits_influenza,FALSE,Influenza Emergency Department Visits (Percent of total ED visits),TRUE,3-week moving average of percent of ED visits that had a discharge diagnosis code of influenza,3-week moving average of percent of ED visits that had a discharge diagnosis code of influenza,National Syndromic Surveillance Program,flu,Hospitalizations,USA,"county,state,hrr,msa","hrr,msa",2022-10-01,,ongoing,,week,Week,weekly,,,All,None,hospitalized,,"Data is available for 78% of US emergency departments. California, Colorado, Missouri, Oklahoma, and Virginia have the most noticeable gaps in coverage, with many counties in those states having either no eligible EDs or having no recently reported data in NSSP. However, most states have some counties that do not contain any reporting EDs.
1599
+
nssp,pct_ed_visits_influenza,TRUE,smoothed_pct_ed_visits_influenza,FALSE,"Influenza Emergency Department Visits (Percent of total ED visits, 3-week average)",TRUE,3-week moving average of percent of ED visits that had a discharge diagnosis code of influenza,3-week moving average of percent of ED visits that had a discharge diagnosis code of influenza,National Syndromic Surveillance Program,flu,Hospitalizations,USA,"county,state,hrr,msa","hrr,msa",2022-10-01,,ongoing,,week,Week,weekly,,,All,None,hospitalized,,"Data is available for 78% of US emergency departments. California, Colorado, Missouri, Oklahoma, and Virginia have the most noticeable gaps in coverage, with many counties in those states having either no eligible EDs or having no recently reported data in NSSP. However, most states have some counties that do not contain any reporting EDs.
1600
1600
1601
1601
NSSP does not report county-level data for all counties with reporting EDs; some reporting EDs are only included in state-level values.
1602
1602
1603
1603
The following states report no data through NSSP at the county level: CA, WA, AK, AZ, AL, CO, SD, ND, MO, AR, FL, OH, NH, CT, NJ.
1604
1604
1605
1605
South Dakota, Missouri, and territories report no data through NSSP at the state level.",Percentage,percent,other,bad,TRUE,FALSE,FALSE,FALSE,FALSE,,,,,,
1606
-
nssp,pct_ed_visits_rsv,TRUE,smoothed_pct_ed_visits_rsv,FALSE,COVID Emergency Department Visits (Percent of total ED visits),TRUE,3-week moving average of percent of ED visits that had a discharge diagnosis code of rsv,3-week moving average of percent of ED visits that had a discharge diagnosis code of rsv,National Syndromic Surveillance Program,rsv,Hospitalizations,USA,"county,state,hrr,msa","hrr,msa",2022-10-01,,ongoing,,week,Week,weekly,,,All,None,hospitalized,,"Data is available for 78% of US emergency departments. California, Colorado, Missouri, Oklahoma, and Virginia have the most noticeable gaps in coverage, with many counties in those states having either no eligible EDs or having no recently reported data in NSSP. However, most states have some counties that do not contain any reporting EDs.
1606
+
nssp,pct_ed_visits_rsv,TRUE,smoothed_pct_ed_visits_rsv,FALSE,"RSV Emergency Department Visits (Percent of total ED visits, 3-week average)",TRUE,3-week moving average of percent of ED visits that had a discharge diagnosis code of rsv,3-week moving average of percent of ED visits that had a discharge diagnosis code of rsv,National Syndromic Surveillance Program,rsv,Hospitalizations,USA,"county,state,hrr,msa","hrr,msa",2022-10-01,,ongoing,,week,Week,weekly,,,All,None,hospitalized,,"Data is available for 78% of US emergency departments. California, Colorado, Missouri, Oklahoma, and Virginia have the most noticeable gaps in coverage, with many counties in those states having either no eligible EDs or having no recently reported data in NSSP. However, most states have some counties that do not contain any reporting EDs.
1607
1607
1608
1608
NSSP does not report county-level data for all counties with reporting EDs; some reporting EDs are only included in state-level values.
0 commit comments