Skip to content

[CTIS] Add note that current SEs are unweighted #908

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 3 commits into from
May 31, 2022
Merged

[CTIS] Add note that current SEs are unweighted #908

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
4100347
note equal-weight SEs
nmdefries May 16, 2022
97a174e
fix weighting link for contingency
nmdefries May 16, 2022
19199e8
align with existing notation
nmdefries May 27, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
9 changes: 8 additions & 1 deletion docs/api/covidcast-signals/fb-survey.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -726,6 +726,13 @@ of zero, and in simulations improves the quality of the standard error
estimates. See the [Appendix](#appendix) for further motivation for these
estimators.

<div style="background-color:#FCC; padding: 10px 30px;"><strong>Note:</strong>
Currently the standard errors are calculated as though all survey weights are
equal, that is \(w^\text{part}_i = 1\) for all respondents. The result is that
reported standard errors are artificially narrow for weighted estimates. This
will be corrected in a future update to the API.
</div>

The pseudo-observation is not used in $$\hat{p}$$ and $$\hat{q}$$ themselves, to
avoid potentially large amounts of estimation bias, as $$p$$ and $$q$$ are
expected to be small.
Expand All @@ -750,7 +757,7 @@ knowing someone in their community who is sick. In this subsection we will
describe how survey weights are used to construct weighted estimates for these
indicators, using community CLI as an example.

In a given aggregation unit (for example, daily-county), let $$U_i$$ the
In a given aggregation unit (for example, daily-county), let $$U_i$$ denote the
indicator that the survey respondent knows someone in their community with CLI,
including their household, for survey $$i$$, out of $$m$$ surveys collected.
Also let $$w_i$$ be the weight that accompanies survey $$i$$, normalized to sum
Expand Down
2 changes: 1 addition & 1 deletion docs/symptom-survey/contingency-tables.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -173,7 +173,7 @@ for download, and specifies the survey questions on which they are based. See
the <a href="coding.html">survey instrument codebook</a> for the full text of
all questions.</div>

The files contain [weighted estimates](../api/covidcast-signals/fb-survey.md#survey-weighting)
The files contain [weighted estimates](../api/covidcast-signals/fb-survey.md#survey-weighting-and-estimation)
of the percent of respondents who fulfill one or several criteria. Estimates are
broken out by state, age, gender, race, ethnicity, occupation, and health
conditions.
Expand Down